1559 2863 28 1 1

VOLUME 28
NUMBER 1
1-104
OPERATIVE DENTISTRY, Inc.
EDITORIAL
Where Have All the Mentors Gone?
MA Cochran . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1
CLINICAL RESEARCH
Technique Sensitivity in Bonding to Vital, Acid-Etched Dentin
M Ferrari FR Tay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3
OPERATIVE DENTISTRY
JANUARY/FEBRUARY 2003
LABORATORY RESEARCH
Voids and Porosities in Class I Micropreparations Filled with Various Resin Composites
NJM Opdam JJM Roeters T de Boer D Pesschier E Bronkhorst . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9
Effects of In-Office Bleaching Products on Surface Finish of Tooth-Colored Restorations
P Wattanapayungkul AUJ Yap . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15
Effects of Regional Enamel and Prism Orientation on Resin Bonding
Y Shimada J Tagami . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .20
The Effect of a Resin Coating on the Interfacial Adaptation of Composite Inlays
PR Jayasooriya PNR Pereira T Nikaido MF Burrow J Tagami . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .28
Effects of Finishing/Polishing Techniques on Microleakage of Resin-Modified Glass Ionomer Cement Restorations
AUJ Yap WY Yap EJC Yeo JWS Tan DSB Ong . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .36
Effects of Instrumentation Time on Microleakage of Resin-Modified Glass Ionomer Cements

AUJ Yap EJC Yeo WY Yap DSB Ong JWS Tan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .47
Surface Geometry of Three Packable and One Hybrid Composite After Finishing
M Jung S Voit J Klimek . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .53
Relationship Between Nanoleakage and Microtensile Bond Strength at the Resin-Dentin Interface
S Guzmn-Armstrong SR Armstrong F Qian . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .60
Effects of Er:YAG and Nd:YAP Laser Irradiation on the Surface Roughness and Free Surface Energy of Enamel
and Dentin: An In Vitro StudyV Armengol O Laboux P Weiss A Jean H Hamel . . . . . . . . . . . . . . . . . . . . . .67
Effects of Cavity Size on Apoptosis-Induction During Pulp Wound Healing
C Kitamura Y Ogawa T Morotomi M Terashita . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .75
The Effect of Placement of Glass Fibers and Aramid Fibers on the Fracture Resistance of Provisional
Restorative MaterialsG Saygili SM Sahmali F Demirel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .80
LITERATURE REVIEW
Tooth-Colored Post Systems: A Review
AJE Qualtrough F Mannocci . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .86
volume 28 number 1 pages 1-104
The Effect of Flowable Resin Composite on Microleakage in Class V Cavities

AR Yazici M Baseren B Dayanga . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .42
INVITED PAPER
Minimal Intervention Dentistry: Rationale of Cavity Design
GJ Mount . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .92
AWARDS
AAGFO Clinician of the Year Award . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
DEPARTMENTS
INSTRUCTIONS TO CONTRIBUTORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104

10-9385
Operative Dentistry
Indiana University School of Dentistry, Rm S411
1121 West Michigan Street
Indianapolis, IN 46202-5186 USA
Periodicals
january-february 2003
Classifieds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
Announcements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
Operative Dentistry Home Page . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
Corporate Sponsorship . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
january/february 2003 volume 28 number 1 1-104

(ISSN 0361-7734)
VOLUME 28
NUMBER 1
Editorial Board
Aim and Scope

Operative Dentistry publishes articles that advance the practice of
operative dentistry. The scope of the journal includes conservation
and restoration of teeth; the scientific foundation of operative dental
therapy; dental materials; dental education; and the social, political,
and economic aspects of dental practice. Review papers, book reviews,
letters, and classified ads for faculty positions are also published.
Subscriptions: Fax (317) 852-3162

Operative Dentistry (ISSN 0361-7734) is published bimonthly by Operative Dentistry, Indiana University School of
Dentistry, Room S411, 1121 West Michigan Street,
Indianapolis, IN 46202-5186. Periodicals postage paid at
Indianapolis, IN, and additional mailing offices. Postmaster:
Send address changes to: Operative Dentistry, Indiana University School of Dentistry, Room S411, 1121 West Michigan
Street, Indianapolis, IN 46202-5186.
Current pricing for individual, institutional and dental student subscriptions (both USA and all other countries) can be found at our website: www.jopdent.org, or by contacting our subscription manager:
Fax 317/852-3162
E-mail: [email protected]
Information on single copies, back issues and reprints is also available. Make remittances payable (in US dollars only) to Operative
Dentistry and send to the above address. Credit card payment (Visa,
MasterCard) is also accepted by providing card type, card number,
expiration date, and name as it appears on the card.
Contributions
Contributors should study the instructions for their guidance printed
in this journal and should follow them carefully.
Permission
For permission to reproduce material from Operative Dentistry please
apply to Operative Dentistry at the above address.
The views expressed in Operative Dentistry do not necessarily represent those of the Academies or of the Editors.
Editorial Office
Operative Dentistry
Indiana University School of Dentistry, Room S411
1121 West Michigan Street, Indianapolis, IN 46202-5186
Telephone: (317) 278-4800, Fax: (317) 278-4900
URL: http://www.jopdent.org/
Editorial Staff
Editor: Michael A Cochran
Editorial Assistant/Subscription Manager: Joan Matis
Editorial Associate: Karen E Wilczewski
Associate Editors: Bruce A Matis, Edward J DeSchepper
and Richard B McCoy
Managing Editor: Timothy J Carlson
Assistant Managing Editors: Joel M Wagoner
and Ronald K Harris
Kinley K Adams
Maxwell H Anderson
Daniel J Armstrong
Steven R Armstrong
Tar-Chee Aw
Wayne W Barkmeier
Douglas M Barnes
Gardner Bassett
Mark W Beatty
Lars Bjrndal
Lawrence W Blank
Paul K Blaser
Murray R Bouschlicher
William W Brackett
James C Broome
K Birgitta Brown
William Browning
Paul A Brunton
Michael Burrow
Fred J Certosimo
Daniel CN Chan
David G Charlton
Gordon J Christensen
Kwok-hung Chung
N Blaine Cook
David Covey
Gerald E Denehy
Joseph B Dennison
Kim E Diefenderfer
E Steven Duke
Willaim J Dunn
Frederick C Eichmiller
Sigfus T Eliasson
Omar M El-Mowafy
John W Farah
Dennis J Fasbinder
Jack L Ferracane
Mark Fitzgerald
Kevin B Frazier
James C Gold
Carlos Gonzalez-Cabezas
Valeria V Gordan
Kevin M Gureckis
Mark S Hagge
Carl W Haveman
Van B Haywood
Charles B Hermesch
Harald O Heymann
Thomas J Hilton
Richard J Hoard
Barry W Holleron
Ronald C House
Poonam Jain
Gordon K Jones
Barry Katz
Robert C Keene
William P Kelsey, III
Edwina A M Kidd
George T Knight
Kelly R Kofford
Harold R Laswell
2003 Operative Dentistry, Inc. Printed in USA
1-104
Mark A Latta
James S Lindemuth
Melvin R Lund
Dorothy McComb
Jonathan C Meiers
Georg Meyer
Ivar A Mjr
Michael P Molvar
B Keith Moore
Graham J Mount
David F Murchison
Ann-Marie Neme
Jennifer Neo
Jacques Nr
John W Osborne
Michael W Parker
Craig J Passon
Tilly Peters
Anne Peutzfeldt
Frank E Pink
T R Pitt Ford
Jeffrey A Platt
L Virginia Powell
James C Ragain
John W Reinhardt
Eduardo Reston
Philip J Rinaudo
Andr Ritter
J William Robbins
Frank T Robertello
Howard W Roberts
Boyd E Robinson
Clyde L Roggenkamp
William Rose
Jean-Francois Roulet
Frederick A Rueggeberg
Henry A St Germain, Jr
David C Sarrett
John W Shaner
Gregory E Smith
W Dan Sneed
Ivan Stangel
James M Strother
James B Summitt
Edward J Swift, Jr
William H Tate
Franklin R Tay
Choi Gait Toh
Peter T Triolo, Jr
Karen Troendle
Richard D Tucker
Marcos Vargas
Douglas Verhoef
Warren C Wagner
Joel M Wagoner
Charles W Wakefield
Steve W Wallace
Timothy F Watson
Nairn H F Wilson
Peter Yaman
Adrian U J Yap
Andrea G F Zandona
These Dental Manufacturers have joined Operative Dentistry in our commitment to

publish quality dental literature in a timely manner. We thank them for their support.
Please view the Corporate Sponsors Page at our website (http://www.jopdent.org)

for direct links to these companies.
Operative Dentistry, 2003, 28, 1
Editorial
Where Have
All the Mentors Gone?
n the early days of dentistry, before the establishment of dental schools and colleges, training was
primarily done on an apprenticeship basis. As
with any apprenticeship, the novice would work for
an experienced master craftsman in return for
instruction and hands-on training. This was usually a one-on-one or very small group situation, and
the masters role was that of mentor, which Websters
New World Dictionary defines as a wise adviser,
teacher or coach. The mentor would demonstrate a
procedure for the apprentice, then watch carefully
while the apprentice performed the technique, giving
advice and criticism and evaluating the outcome.
This would be repeated again and again until the student could demonstrate a thorough knowledge and
the ability to apply that knowledge. Even today, individual or small group instruction, particularly in
teaching manual dexterity skills, is recognized as one
of the best methods for imparting knowledge.
Unfortunately, the sheer size of dental school classes
and the limited curricular time and number of faculty make true mentoring of every student extremely
difficult, if not impossible. Because of this, a large
percentage of our profession must rely on various
types of continuing education courses to further their
knowledge and improve their skills. The obvious
question is if mentoring is an appropriate way to
impart clinical skills, what role does it play in the
modern continuing education process?
mentor (lecturer) is essentially removed from the students (listeners) and has no way of knowing if the
information delivered has been understood or how it
will be applied. There is no opportunity to observe the
students clinical skills and offer constructive criticism. Conversely, attendees at this type of course are
often impressed with the beautiful cases they are
shown and then disappointed when the promoted
material or technique does not give them the same
result in their own office, and there is no follow-up to
explain why. The natural tendency is to simply blame
the new material and not recognize a possible lack of
understanding or ability, so that the product is
pushed to the back of a shelf or disposed of and the
experience becomes a waste of valuable time.
Dental continuing education is a vast arena of different types of programs. The majority are primarily
lecture-format presentations that vary from an hour
or two to several days in length. Practitioners often
attend multiple sessions that encompass a variety of
subjects during annual meetings of various professional organizations. The emphasis of this type of
continuing education is usually on newer materials
and equipment with some visual examples of clinical
applications and, occasionally, some preliminary data
on longevity and success. While this is certainly a
very important part of continuing education, the
The best and most proven source of clinical continuing education is the operating study club. By this, I
dont mean a social group of colleagues who meet once
a month for dinner and drinks and have a guest lecturer make a one-hour presentation. Im referring to
a group of dedicated practitioners who meet on a regular basis and actually perform various types of
restorative treatment on patients under the guidance
of a knowledgeable and clinically proficient mentor.
This is true hands-on education, where every session provides experience and the opportunity to make
More clinically oriented continuing education is

offered as hands-on courses that provide laboratory
and possibly clinical instruction. The best of these
usually involve several days of attendance, encompassing lecture, laboratory and/or clinical demonstration, and finally, performance activity by the students
under the observation, guidance and critique of
instructors. These courses at least offer the opportunity to actually work with the products and receive
advice and correction of errors. Unfortunately, this is
still a one-time experience, often with no provision
for continuing input and instruction related to individual clinical problems that arise or long-term development and improvement of necessary manual skills.
2
mistakes and receive constructive criticism as well as
advice and demonstration on how to recognize and
correct problems as they arise. Ongoing sessions
ensure that the clinical relevance and impact on performance and longevity of the skills, techniques and
materials used will become evident. The goal of these
groups is to constantly improve the level of treatment
for their patients by evaluating materials, equipment
and techniques and learning new clinical skills and
refining them to their highest level.
There are two basic requirements for an operating
study club. Members who are willing to devote the
necessary time, effort and expense as well as accept
the scrutiny and criticism of their work, and a mentor who has the knowledge, skills and dedication to
earn the respect of the group and provide the instruction.
Herein lies the problem. Where are the mentors
today? Who are the successors to the Blacks,
Hollenbacks and Markleys of yesterday? All of us can
name individuals who have made a tremendous
impact on the profession through their unselfish
mentoring (I wont insert any names myself, because
it would take too much space and I would certainly
miss some), and recite stories of their regular journeys across several states and to other countries to
supervise different study clubs and promote excellence. Unfortunately, the world changes and motivations shift. The face of dentistry is in constant flux as
the market is inundated with new products, while
Operative Dentistry
others disappear at the same rate. Continuing education has become big business for some and a reputation builder for others. This does not imply that they
are not skilled or knowledgeable, merely that their
time is spent in an attempt to reach as large an audience as possible. This certainly has merit and provides useful information to busy practitioners, but it
does not really lead to improved clinical skills.
What the profession needs are more study clubs and
more mentors. The Tucker Study Clubs have set up
an excellent model of training apprentices whom
they send to mentor new study groups. While there is
a natural hesitation (feeling of inadequacy) in assuming a mentors role, a good mentor recognizes when a
student is ready to become a teacher and needs to
encourage this step. Most good mentors continue to
function as students in their original study group
since continued learning can only improve their own
teaching. For all of you who have enjoyed the experience of study club activity, I strongly encourage you
to help expand the availability of this type of continuing education by starting new groups and encouraging your colleagues to participate. Considering all the
changes and problems in dental education today, we
sorely need to not only keep mentoring alive, but to
increase its impact on the profession.
Michael A Cochran
Editor
Operative Dentistry, 2003, 28, 3-8
Clinical Research
Technique Sensitivity
in Bonding to Vital,
Acid-Etched Dentin
M Ferrari FR Tay
Clinical Relevance
Technique sensitivity, previously reported in vitro using a moist bonding technique on

acid-etched dentin, is applicable in vivo when bonding to vital dentin, in the small
number of samples examined in this study.
SUMMARY
Just as vital dentin is moist after removing the
smear layer, avoiding collapse of the collagen
matrix after acid-etching requires in vivo validation. This study hypothesizes that there is no difference between moist bonding performed in
vitro or in vivo, and that excessive drying or wetting of vital acid-etched dentin produces inferior
results. Resin-dentin interfaces bonded with a
moist bonding technique (control), either in vitro
or in vivo with Excite DSC (Vivadent), were
examined with and without tracer penetration
using transmission electron microscopy.
Specimens bonded in vivo under excessively dry
and wet conditions were also examined. The patterns of silver deposition were similar within the
adhesive and hybrid layers created in vitro or in
M Ferrari, MD, DDS, PhD, dean of School of Dental Medicine,
chair and professor of Restorative and Dental Materials
Department, Policlinico Le Scotte, University of Siena, Italy
*Franklin R Tay, BDSc (Hons), FADM, PhD, honorary assistant
professor, Departments of Pediatric Dentistry & Orthodontics,
and Conservative Dentistry, The University of Hong Kong, The
Prince Philip Dental Hospital, Hong Kong
*Reprint request: Faculty of Dentistry, 34 Hospital Road, Hong
Kong SAR, China; e-mail: [email protected]
vivo. No hybrid layer was observed in vivo after

excessive drying. Excessive wetting in vivo
resulted in more extensive nanoleakage and
water tree formation along resin-dentin interfaces.
INTRODUCTION
The moist bonding technique (Kanca, 1992) was introduced a decade ago for optimizing bonding of acetoneand ethanol-based dentin adhesives to acid-etched
dentin. Moist bonding prevents collapse of demineralized collagen fibrils during resin-infiltration
(Gwinnett, 1994). Interfibrillar spaces about 20 nm
wide when fully extended must remain open to enable
diffusion of resin monomers into the demineralized
intertubular dentin (Eick & others, 1997; Wang &
Spencer, 2002). Desiccation of acid-etched dentin
diminishes the surface wettability of the bonding substrate (Rosales & others, 1999), increases the capacity
for hydrogen bonding among amino acid residues of
collagen fibrils (Miles & Ghelashvili, 1999; Pashley &
others, 2001) and results in incomplete resin-infiltration within the hybrid layer (Tay & others, 1996;
Nakajima & others, 2000). Manufacturers of total-etch
dentin adhesives almost exclusively recommend using
moist bonding or aqueous rewetting agents on air-
Operative Dentistry
dried, acid-etched dentin (Perdigo & Frankenberger,

2001; Pashley & others, 2001; Saeki & others, 2001).
Despite numerous in vitro studies on the efficacy of
the moist bonding technique (Nakabayashi &
Hiranuma, 2000; Hashimoto & others, 2002), the
necessity to avoid collapse of the collagen matrix after
acid-etching of vital human dentin has not been substantiated in vivo. As vital dentin is inherently moist,
with transudation of dentinal fluid after smear layer
removal (Itthagarun & Tay, 2000), the presence of this
intrinsic source of moisture (Pereria & others, 1999;
Moll & Haller, 2000) may be all that is required to
maintain the demineralized collagen matrix in its
hydrated state, or even to cause regional disruption of
resin-dentin bonds in some adhesive systems (Tay,
Gwinnett & Wei, 1998; Pereria & others, 1999). Thus,
the objective of this study was to examine the ultrastructure and extent of tracer penetration in resindentin interfaces created in deep, vital acid-etched
dentin under different degrees of hydration of the demineralized collagen matrices using transmission electron microscopy (TEM). The null hypotheses tested
were: a) there is no difference between moist bonding
performed in vitro or in vivo, and b) there is no difference in the in vivo results produced with excessive drying or wetting of vital acid-etched when compared with
the manufacturers recommended technique for moist
bonding.
METHODS AND MATERIALS
Sixteen teeth were used in this study. In vitro and in
vivo specimens were restored by different operators.
Four extracted anterior teeth were collected after the
patients informed consent had been obtained under a
protocol reviewed and approved by the institutional
review board from the University of Hong Kong. These
teeth were assigned to the in vitro control group and
restored by one operator (FRT). Twelve vital, periodontally-compromised, caries-free, unrestored anterior
teeth scheduled for extraction, were selected. Informed
consent of the subjects was obtained under an in vivo
protocol reviewed and approved by an ethical committee from the University of Siena. These teeth were
assigned to the three in vivo groups and restored by a
second operator (MF).
Experimental Design
Tooth preparation was performed on the labial surface
of each tooth with unbevelled cavosurface margins
located occlusally in enamel and cervically beyond the
cementodentinal junction. Cavities were etched with a
37% phosphoric acid gel for 20 seconds and rinsed for
20 seconds. Bonding was performed with Excite DSC
(Vivadent, Schaan, Liechtenstein), an ethanol-based,
two-step, dual-curable, single-bottle adhesive. The selfcuring mode was activated by dipping the catalyst-
impregnated microbrush tip into the light-cured adhesive and mixing for five seconds. One layer of the activated adhesive was applied with gentle agitation for 10
seconds, leaving a glossy adhesive film over the entire
cavity surface after the solvent evaporated. The experimental groups represented different conditions of the
acid-etched dentin prior to adhesive application:
1. In vitro control group (moist bonding). Acidetched dentin of each extracted tooth was blotdried with a piece of lint-free gauze immediately
before placing the adhesive. Evaporation of the
adhesive solvent was achieved by gentle air drying for five seconds with a triple syringe located
10 cm away from the dentin surface, followed by
more aggressive air drying for another five seconds.
2. In vivo control group (moist bonding). Taking into
account the higher intraoral relative humidity in
the absence of rubber dam placement, and the
intrinsic wetness of deep vital dentin after
removal of the smear layer (Itthagarun & Tay,
2000), the wet, vital acid-etched dentin from each
periodontally-compromised tooth was gently air
dried for one second at a distance of 10 cm from
the bonding surface to remove gross excess of
water, leaving the etched dentin visibly moist.
After applying the adhesive, evaporation of the
adhesive solvent followed what had been
described for the previous group.
3. In vivo excessively dry group. The wet acid-etched
dentin was air dried for five seconds to collapse
the demineralized collagen matrix prior to adhesive application. Drying was assessed clinically by
observing frostiness over the adjacent acid-etched
enamel. Evaporation of the adhesive solvent was
performed in the same way as the other groups.
4. In vivo excessively wet group. After rinsing, the
acid-etched dentin was not air dried, and a visible
excess of surface moisture was present during
adhesive application. Evaporation of the adhesive
solvent was performed in the same way as the
other groups.
For each group, the adhesive was light activated for
20 seconds, according to the manufacturers instructions. Each bonded cavity was restored incrementally
with Tetric Flow (Vivadent), a light-cured flowable
resin composite, to facilitate ultramicrotomy. Each 2
mm composite increment was light activated for 40
seconds. The composites were finished using tungsten
carbide finishing burs (Axis Dental Corp, Irving, TX
75038, USA), followed by diamond-impregnated polishing disks (Pogo, Dentsply Caulk, Milford, ME
19963, USA) to obtain a high luster. The teeth from the
three in vivo groups were extracted one week after
placing the restorations. Extracted teeth from all the
Ferrari & Tay: Bonding to Vital Dentin
four groups were stored in distilled water

at 22C for three hours before laboratory
processing.
Laboratory Specimen Preparation
A 0.9 mm slab of restored mid-coronal
dentin that contained the bonded interface
that was to be examined was sectioned
mesial-distally from each tooth using a
slow-speed saw (Isomet, Buehler Ltd,
Lake Bluff, IL 60049, USA) under water
lubrication. Half of the slabs from each group
were completely demineralized in 0.1 M
buffered ethylenediamine tetraacetic acid
(pH=7.0) for morphological examination of
the resin- dentin interfaces. The remainder
of the undemineralized slabs were
immersed in a tracer solution for examining the extent of tracer penetration within
the bonded interfaces.
An aqueous solution of 50 wt% ammoniacal silver nitrate [pH=9.5] was used as a
TEM tracer (Tay, Pashley & Yoshiyama,
2002). It was prepared by dissolving 50 g
of silver nitrate crystals (Sigma Chemical
Co, St Louis, MO 63178, USA) in 25 ml of
distilled water. Concentrated (28%)
ammonium hydroxide (Sigma) was used
to titrate the black solution until it
became clear as ammonium ions complexed the silver into diamminesilver (I)
ions ([Ag(NH3)2]+). This solution was
diluted to 50 ml with distilled water to
achieve a 50 wt% solution. Specimens
were coated with fast-setting nail varnish
applied 1 mm from the bonded interfaces.
Without allowing the slabs to become
dehydrated, they were immersed in the
tracer solution for 24 hours. The silverimpregnated slabs were rinsed thoroughly in distilled water and placed into photodeveloping solution for eight hours
under a fluorescent light to reduce the
diamminesilver (I) ions into metallic silver grains within potential voids along
the bonded interfaces.
Demineralized and undemineralized
epoxy resin-embedded 70-90 nm thick sections were prepared according to the TEM
protocol of Tay, Moulding & Pashley
(1999). Demineralized sections were double-stained with uranyl acetate and lead
citrate. Silver-impregnated, undemineralized sections were not additionally
stained. They were examined using a
transmission electron microscope (Philips
Figure 1A.
Figure 1B.
In vitro control group bonded using a moist bonding technique with blot-drying of the acidetched dentin. Figure 1A. Morphology of resin-dentin interface in stained, demineralized section. Collagen fibrils along the surface of the 6 m-thick hybrid layer (H) exhibited a shag carpet-like appearance (pointer). Filler clusters (arrowhead) were sparsely distributed within the
adhesive layer (A). D: intertubular dentin; T: resin tag within a dentinal tubule. Arrows: lateral
branches of the dentinal tubules. Figure 1B. Extent of tracer penetration in unstained, undemineralized section. Demineralized, resin infiltrated collagen fibrils within the unstained hybrid
layer (H, between open arrows) appeared electron-lucent. Two patterns of silver deposition
could be identified within the hybrid and adhesive layers (A). The reticular pattern existed as
fine, interconnecting silver deposits (nanoleakage) within the interfibrillar spaces of the hybrid
layer (pointer) and water trees (arrow) within the adhesive layer. The latter originated from,
and were perpendicular to, the surface of the hybrid layer. The spotted pattern (open arrowheads) consisted of isolated silver grains that were randomly distributed within both the hybrid
and adhesive layers. D: intertubular dentin.
Figure 2A.
Figure 2B.
In vivo control group bonded using a moist bonding technique with air drying of the acid-etched
dentin for one second. Figure 2A. Stained, demineralized section. The thickness of the hybrid
layer was more variable and ranged from 2-5 m thick. The surface of the hybrid layer (H) was
smooth. Lateral branches (arrow) were rarely observed within the hybrid layer. C: resin composite; A: adhesive; T: resin tag; D: intertubular dentin. Figure 2B. Extent of tracer penetration
in the unstained, undemineralized section. Despite the difference in morphology of the hybrid
layer between the in vivo and in vitro bonded specimens, there was no difference in the silver
staining patterns observed in the hybrid layer (H; between open arrows) and adhesive layer
(A) in both control groups. Pointer: reticular pattern of nanoleakage within hybrid layer, Arrow:
Water tree extending from the surface of the hybrid layer into the adhesive layer; Open arrowheads: isolated silver grains within the hybrid and adhesive layers; D: intertubular dentin.
Operative Dentistry
EM208S, Philips, Eindhoven,

Netherlands) operating at 80 kV.
The
RESULTS
Hybrid layers created with moist bonding
in vitro using the blot-drying method had
a looser texture with the presence of tufted
surface collagen fibrils and the frequent
observation of lateral branches of the
dentinal tubules (Figure 1A). Conversely,
those created with moist bonding in vivo
using air drying exhibited a smooth surface
morphology, the absence of lateral branches
and more variability in thickness (Figure 2A).
Despite these morphological differences,
the patterns of silver penetration within
the bonded interfaces were similar in both
control groups. Two patterns of silver deposition were recognized in both the hybrid
layer and adhesive layer (Figures 1B and
2B). The reticular pattern consisted of fine
interconnecting strands of silver deposits
and was manifested as nanoleakage (Sano
& others, 1995) within hybrid layers and
as water trees (Stepp & others, 1996;
Tay & Pashley, 2002) within adhesive layers. The spotted pattern appeared as individual, unconnected silver grains in both
the hybrid and adhesive layers.
Figure 3B.
Figure 3A.
In vivo, excessively dry group with acid-etched dentin air dried for five seconds. Figure 3A.
Stained, demineralized section. No hybrid layer could be identified except for an electron-dense
crust (pointer) on the surface of the acid-etched dentin. C: resin composite; A: adhesive layer;
Te: empty dentinal tubule; D: intertubular dentin. Figure 3B. Unstained, undemineralized section.
Re-expansion of the uninfiltrated, collapsed, acid-etched dentin during immersion in ammoniacal
silver nitrate resulted in almost complete obliteration of the demineralized collagen matrix (CM;
between open arrows) in most of the specimens examined. Additional isolated spots of silver grains
(open arrowhead) could be identified within the adhesive layer (A).T: resin tag; D: intertubular dentin.
Except for the presence of a thin, highly

electron-dense, stained surface crust, no
hybrid layer could be identified from vital
dentin sections that were bonded under
excessively dry conditions (Figure 3A).
Bonded specimens that were immersed in
aqueous ammoniacal silver nitrate
revealed almost complete obliteration of the
Figure 4A.
Figure 4B.
re-expanded, previously collapsed dem- In vivo, excessively wet group with the acid-etched dentin visibly over-wetted during adhesive
ineralized collagen matrix by silver application. Figure 4A. Stained, demineralized section. The hybrid layer (H) was looser in
appearance and individual collagen fibrils could be identified along the surface (pointer). A:
deposits (Figure 3B).
adhesive layer; T: resin tags; D: intertubular dentin. Figure 4B. Unstained, undemineralized sec-
The morphological appearance of hybrid tion. Similar to the in vivo control group, both the spotted (open arrowheads) and the reticular
layers in vital acid-etched dentin speci- patterns of silver deposits were observed within the adhesive (A) and hybrid layers (H; between
mens bonded under excessively wet condi- open arrows). However, the reticular patterns were considerably denser in the hybrid layer
tions (Figure 4A) was similar to the in (pointer) and water tree formation was more extensive (arrows) and penetrated deeply into the
vitro control group. However, there was a adhesive layer, almost reaching the composite-adhesive interface (not shown). T: resin tag; D:
intertubular dentin.
more profuse manifestation of the reticucomplementary information on the efficacy of resin
lar pattern of silver deposits in both the hybrid and
infiltration within interfaces that were examined with
adhesive layers (Figure 4B). Water treeing (Ross,
stained, demineralized sections. While the authors con1998) within the adhesive layer could be observed withcede that using additional chemical-analytical techin the bulk of the adhesive, extending almost to the
niques (Van Meerbeek & others, 2000) would be highly
adhesive-composite interface (not shown).
desirable in improving their understanding of the status of resin infiltration within hybrid layers (Spencer &
DISCUSSION
others, 2000) and phase separations of resin compoUsing an adjunctive tracer penetration technique with
nents within the adhesive interfaces (Eliades,
unstained, undemineralized TEM sections provides
Vougiouklakis & Palaghias, 2001), some of the ultra-
Ferrari & Tay: Bonding to Vital Dentin

structural features demonstrated in this study could
only be observed with the resolution of transmission
electron microscopy. Thus, within the limits of the
small sample size in this study, the authors accept the
first null hypothesis that there is no difference between
moist bonding performed in vitro or in vivo, and reject
the second null hypothesis that there is no difference
when excessive drying or wetting of vital acid-etching
was performed when compared with the manufacturers recommended technique for moist bonding.
A slightly different moist bonding technique was used
in the in vitro and in vivo procedures due to the anticipated increase in intraoral relative humidity and the
possibility of transudation of dentinal fluid in vital
dentin after removal of the smear layer. The shag carpet-like appearance of stained collagen fibrils in the in
vitro specimens was the result of continuous agitation
during adhesive application (Inoue & others, 2000). Airdrying for one second in vivo resulted in partial collapse
of the demineralized collagen matrices. However, since
an ethanol-based adhesive invariably contains a small
amount of water, this extrinsic water, together with the
increase in intrinsic moisture caused by removal of the
smear layer, could have resulted in rehydration of the
partially collapsed collagen matrix during the 10 seconds of adhesive application (Van Meerbeek & others,
1998).
The extent of silver penetration within adhesive
interfaces was the least among the three in vivo groups
when the manufacturers recommended technique for
moist bonding was followed. The spotted pattern of silver deposits could only be seen when basic ammoniacal
silver nitrate was used. These silver grains increase in
size with specimen aging and were absent when specimens were immersed in conventional acidic silver
nitrate (Tay, Pashley & Yoshiyama, 2002). They probably represent microdomains within the adhesive and
hybrid layers where an acid-base reaction occurs
between resin monomers with acidic functional groups
and the basic diamminesilver (I) ions. The reticular
pattern of silver deposits within the hybrid layer in
both control groups was similar to the nanoleakage previously reported using conventional acidic silver nitrate
(Sano & others, 1995). As incomplete resin infiltration
caused by a diffusion gradient of resin components usually occurs along the base of hybrid layers (Spencer &
others, 2000); these randomly distributed interconnecting silver deposits more likely represent areas where
water was incompletely removed from the interfibrillar
spaces. Silver deposition within the adhesive layers
had only been reported recently (Li, Burrow & Tyas,
2001). The occurrence of the reticular pattern of
nanoleakage outside the hybrid layer cannot be attributed to incomplete resin infiltration and is likely to be
related to water that was trapped within the adhesive
layers during polymerization.
Similar to nanoleakage, the term water tree was

first in introduced in Japan by Miyashita (1969). It is a
well-known phenomenon that is responsible for the
water-induced deterioration of polymer insulation of
electrical cables after aging. Water trees in polyethylene- coated cables are submicroscopic, self-propagating,
water-filled tracks that are formed electrochemically by
the oxidation of the hydrophobic polymer into more
hydrophilic moieties, with the condensation of moisture
from the hydrophobic polymer into the hydrophilic, electrooxidized regions (Dissado & Fothergill, 1992). The
increase in water conductivity results in self-propagation along these tracks and the growth of a microscopic,
tree-like pattern of water channels. In the context of
dentin bonding, an electrochemical process is not
required for water tree formation, as both hydrophilic
resin monomers and water are present simultaneously
in resin-dentin interfaces. Water trees along resindentin interfaces may act as stress raisers during functional stresses, or as channels that facilitate resin leaching during the degradation of resin-dentin bonds. This
has to be investigated in future studies. There is initial
evidence to show that water treeing can occur along
the periphery of optimally-polymerized, non-solvented
hydrophilic adhesive resins by water sorption and they
propagate internally with increased time of water storage (Yiu & others, 2002).
CONCLUSIONS
Within the limits of the small sample size in this study,
the technique sensitivity previously reported in vitro is
also applicable in vivo. Excessive air drying of vital,
acid-etched dentin resulted in a minimal hybrid layer
along resin-dentin interfaces, confirming previous in
vitro results (Nakajima & others, 2000). The complete
obliteration of the demineralized collagen matrix with
silver deposits could be explained by the re-expansion of
the collapsed collagen matrix that was devoid of resin
infiltration after immersion for 24 hours (Saeki & others, 2001) in aqueous silver nitrate. In this study, the
authors did not observe blister formation (Tay & others,
1998) along resin dentin interfaces when bonding to
excessively wet dentin in vivo. This could be explained
by the use of an ethanol-based adhesive that has a
milder potential in displacing water along the bonded
interfaces compared to an acetone-based adhesive.
However, it is notable that the density of the reticular
silver deposits manifested as nanoleakage in hybrid layers and water trees in adhesive layers increased under
the condition of bonding to excessively wet, vital, acidetched dentin. It is unlikely that water can be completely removed from these interfaces due to the relatively
high Hoys solubility parameter for hydrogen bonding
(h) of the hydrophilic resin monomers and the ethanol
solvent employed (Pashley & others, 2001). This may
eventually result in an increase in the rate of degradation of resin-dentin bonds created in vital dentin.
8
ACKNOWLEDGEMENT
Operative Dentistry
The adhesives and composites used in this study were generously

sponsored by Vivadent.
Pereira PN, Okuda M, Sano H, Yoshikawa T, Burrow MF &

Tagami J (1999) Effect of intrinsic wetness and regional difference on dentin bond strength Dental Materials 15(1) 46-53.
(Received 3 April 2002)
Perdigo J & Frankenberger R (2001) Effect of solvent and

rewetting time on dentin adhesion Quintessence International
32(5) 385-390.
References
Dissado LA & Fothergill JC (1992) Electrical degradation and
breakdown in polymers Chapter 4 Water treeing degradation
IEE Material and Devices Series 9 London Peregrinus 75-116.
Eick JD, Gwinnett AJ, Pashley DH & Robinson SJ (1997)
Current concepts on adhesion to dentin Critical Reviews in
Oral Biology & Medicine 8(3) 306-335.
Eliades G, Vougiouklakis G & Palaghias G (2001) Heterogeneous
distribution of single-bottle adhesive monomers in the resindentin interdiffusion zone Dental Materials 17(4) 277-283.
Rosales JI, Marshall GW, Marshall SJ, Watanabe LG, Toledano

M, Cabrerizo MA & Osorio R (1999) Acid-etching and hydration influence on dentin roughness and wettability Journal of
Dental Research 78(9) 1554-1559.
Ross R (1998) Inception and propagation mechanisms of water
treeing The Institute of Electrical and Electronic Engineers
Transactions on Dielectrics and Electrical Insulation 5 660680.
Saeki K, Marshall SJ, Gansky SA & Marshall GW (2001)
Etching characteristics of dentin: Effect of ferric chloride in citric acid Journal of Oral Rehabilitation 28(4) 301-308.
Gwinnett AJ (1994) Chemically conditioned dentin: A comparison of conventional and environmental scanning electron
microscopy findings Dental Materials 10(3) 150-155.
Sano H, Yoshiyama M, Ebisu S, Burrow MF, Takatsu T, Ciucchi

B, Carvalho R & Pashley DH (1995) Comparative SEM and
TEM observations of nanoleakage within the hybrid layer
Operative Dentistry 20(4) 160-167.
Hashimoto M, Ohno H, Kaga M, Sano H, Endo K & Oguchi H

(2002) Fractured surface characterization: Wet versus dry
bonding Dental Materials 18(2) 95-102.
Spencer P, Wang Y, Walker MP, Wieliczka DM & Swafford JR

(2000) Interfacial chemistry of the dentin/adhesive bond
Journal of Dental Research 79(7) 1458-1463.
Inoue S, Van Meerbeek B, Vargas M, Yoshida Y, Lambrechts P &

Vanherle G (2000) Adhesion mechanism of self-etching adhesives in Tagami J, Toledano M & Prati C editors Proceedings of
Conference on Advanced Adhesive Dentistry. Third
International Kuraray Symposium December 3-4, 1999,
Granada, Spain. Cirimido, Italy: Grafiche Erredue 131-148.
Stepp D, King JA, Worrall J, Thompson A & Cooper DE (1996)

High-resolution study of water trees grown in silver nitrate
solution The Institute of Electrical and Electronic Engineers
Transactions on Dielectrics and Electrical Insulation 3 392398.
Itthagarun A & Tay FR (2000) Self-contamination of deep dentin

by dentin fluid American Journal of Dentistry 13(4) 195-200.
Kanca JA 3rd (1992) Resin bonding to wet substrate. 1. Bonding
to dentin Quintessence International 23(1) 39-41.
Li HP, Burrow MF & Tyas MJ (2001) The effect of long-term storage on nanoleakage Operative Dentistry 26(6) 609-616.
Miyashita T (1969) Deterioration of water-immersed polyethylene coating wire by treeing Proceedings of the 1969 IEEENEMA Electrical Insulation Conference Boston September
1969 131-135.
Miles CA & Ghelashvili M (1999) Polymer-in-a-box mechanism
for the thermal stabilization of collagen molecules in fibers
Biophysics Journal 76(6) 3243-3252.
Moll K & Haller B (2000) Effect of intrinsic and extrinsic moisture on bond strength to dentine Journal of Oral
Rehabilitation 27(2) 150-165.
Tay FR, Gwinnett AJ, Pang KM & Wei SHY (1996) Resin permeation into acid-conditioned, moist, and dry dentin: A paradigm using water-free adhesive primers Journal of Dental
Research 75(4) 1034-1044.
Tay FR, Gwinnett JA & Wei SH (1998) Relation between water
content in acetone/alcohol-based primer and interfacial ultrastructure Journal of Dentistry 26(2) 147-156.
Tay FR, Moulding KM & Pashley DH (1999) Distribution of
nanofillers from a simplified-step adhesive in acid-conditioned
dentin Journal of Adhesive Dentistry 1(2) 103-117.
Tay FR & Pashley DH (2002) Water treeinga potential mechanism for degradation of dentin adhesives American Journal of
Dentistry (in press).
Tay FR, Pashley DH & Yoshiyama M (2002) Two modes of
nanoleakage expression in single-step adhesives Journal of
Nakabayashi N & Hiranuma K (2000) Effect of etchant variation

on wet and dry dentin bonding primed with 4-META/acetone
Dental Materials 16(4) 274-279.
Van Meerbeek B, Vargas M, Inoue S, Yoshida Y, Perdigo J,

Lambrechts P & Vanherle G (2000) Microscopy investigations.
Techniques, results, limitations American Journal of Dentistry
13(Spec No) 3D-18D.
Nakajima M, Kanemura N, Pereira PN, Tagami J & Pashley DH

(2000) Comparative microtensile bond strength and SEM
analysis of bonding to wet and dry dentin American Journal of
Dentistry 13(6) 324-328.
Van Meerbeek B, Yoshida Y, Lambrechts P, Vanherle G, Duke

ES, Eick JD & Robinson SJ (1998) A TEM study of two waterbased adhesive systems bonded to dry and wet dentin Journal
of Dental Research 77(1) 50-59.
Pashley DH, Agee KA, Nakajima M, Tay FR, Carvalho RM,

Terada RS, Harmon FJ, Lee WK & Rueggeberg FA (2001)
Solvent-induced dimensional changes in EDTA-demineralized
dentin matrix Journal of Biomedical Materials Research 56(2)
273-281.
Wang Y & Spencer P (2002) Quantifying adhesive penetration in

adhesive/dentin interface using confocal Raman microspectroscopy Journal of Biomedical Materials Research 59(1) 46-55.
Yiu CKY, Tay FR, Pashley DH, King NM, Suh BI & Itthagarun
(2000) A Effect of resin hydrophilicity on tracer penetration
a preliminary study American Journal of Dentistry (in press).
Laboratory Research
Voids and Porosities

in Class I Micropreparations
Filled with
Various Resin Composites
NJM Opdam JJM Roeters T de Boer
D Pesschier E Bronkhorst
Clinical Relevance
Small occlusal preparations are difficult to restore free of voids. The best results are
achieved with a technique that uses a first layer of uncured flowable composite followed
by a second layer of a medium-viscous composite injected into the cavity.
SUMMARY
In this in vitro study, voids inside a minimal
occlusal restoration using different consistencies
of resin composite and various application techniques were investigated.
One hundred and fifty-two simulated, minimally
invasive preparations, including a prepared fissure and an excavated carious dentin lesion,
were ground in perspex blocks. After applying an
acrylic primer (Artglass Connector, Kulzer), the
*Niek JM Opdam, DDS, PhD, assistant professor, University of
Nijmegen, The Netherlands
Joost JM Roeters, DDS, PhD, associate professor, University of
Tim de Boer, dental student, University of Nijmegen, The
Netherlands
Dennis Pesschier, dental student, University of Nijmegen, The
Netherlands
Ewald Bronkhorst, MSc, PhD, assistant professor, University of
*Reprint request: University of Nijmegen, Department of
Cariology and Endodontology, PO Box 9101, NL 6500 HB
Nijmegen, The Netherlands; e-mail: [email protected]
preparations were restored with the adhesive

PhotoBond (Kuraray) and one of three resin
composites: a packable composite (Prodigy
Condensable-Kerr), a syringable composite
(Clearfil Photo Posterior, Kuraray) and a flowable composite (Revolution, Kerr). The restorations were inserted according to eight protocols
(n=19). In three groups, the composite was placed
in bulk. In another three groups, a layer of flowable composite was placed first, then cured, followed by a second layer of one of the three composites. In two groups, the first layer of flowable
composite was left uncured before a second layer
of a packable or syringable composite was inserted. The perspex blocks were sectioned and
inspected for the presence of voids. Statistical
analysis was conducted using Fischers exact
tests at p<0.05.
The results showed that restoring minimal
preparations in the absence of porosities and
voids was very difficult to achieve. Placing a
layer of flowable composite that was left
uncured, directly followed by injecting a medium-viscous composite, was the technique that
resulted in the most homogeneous restoration.
10
Operative Dentistry
INTRODUCTION
Adhesive techniques enable the clinician to save sound

tooth tissue when caries is treated in such a way that
only the decayed tooth substance is removed. Since the
restorative material can strengthen weakened tooth
structures after preparation (Ausiello & others, 1999),
undermined enamel can be left in place. Therefore,
treatment of occlusal caries may result in an occlusal
opening of fissures that have a small dimension and
local extension into the dentin at locations where
decayed tissue has to be removed. These minimal
preparation techniques may include using air-abrasion
for preparation of occlusal caries (Goldstein & Parkins,
1994). After preparation, the clinician can restore the
irregularly-shaped cavities with an adhesive technique
using resin composite. Preparations with a narrow
occlusal opening are more difficult to fill than larger
cavities (Opdam & others, 2002a). Choosing a low or
medium viscous composite helps to obtain a more completely filled cavity compared to high viscous composites (Opdam & others 1996; Opdam & others, 2002).
Flowable composites with a low viscosity are also recommended as the material of choice for restoring small,
narrow cavities, especially when a needle-shaped application tip that provides good access to the cavity is used
(Peters & McClean, 2001b). However, due to low filler
content, flowable composites have inferior physical
properties, such as higher polymerization shrinkage
and lower strength compared to normal hybrid composites (Bayne & others, 1998). Applying a lining of flowable composite under a resin composite restoration can
result in less voids inside the restoration (Chuang, Liu
& Jin, 2001b). However, no research is available to support claims of a superior adaptation for flowable composites when these materials are used as a restorative
material for Class I restorations. On the other hand,
some studies mention the risk of entrapment of air bubbles in flowable composites or sealants (Lekka,
Papagiannoulis & Eliades, 1991; Chuang & others,
2001a). Also, the effect on the adaptation and the presence of voids in small micro-preparations, when a combination of a flowable composite with a syringable or
packable composite is used, is unknown.
Figure 1. Occlusal view from the experimental preparation.
This study compared the adaptation and homogeneity

of different types of composite in small occlusal preparations.
To enable standardization of shape and size, preparations were made in eight blocks of clear polymethylmethacrylate. The dimensions of these blocks were
200 x 18 x 18 mm. Preparations were made according
to a standardized protocol. First, a cylindrical cavity
was ground 3 mm in diameter and 4 mm in depth with
an industrial bur. To simulate an excavation procedure in dentin at the bottom of the cylindrical cavity,
an extension was prepared using a stainless steel
round bur (001-018, Meisinger, Dsseldorf, Germany).
The bur was placed on the bottom of the cavity and
used perpendicular to the ascending cavity wall to
simulate an excavation procedure. The shaft of the
bur acted during this procedure as a stopgap against
further extension of the cavity. The cavity was further
extended using a small fissure bur (HM132-008,
Meisinger, Dsseldorf, Germany) that simulated a
narrow fissure 3 mm in depth, 3 mm in length and 1
mm in width. Figures 1 and 2 show the final preparation simulating a minimal or micro-preparation,
including a small fissure opened with a small bur or
by air-abrasion, and a local deeper dentin cavity
resulting from local excavation of carious dentin. All
preparations were made in line through the center of
the perspex blocks, so that after sectioning the bar, all
restorations were available for inspection.
In order to bond to the perspex blocks and provide
the operator with a surface that allows proper wetting
by the composite, Artglass Connector (Kulzer,
Wehrheim, Germany) was applied on the perspex surface followed by a layer of adhesive resin (Photo Bond,
Kuraray, Osaka, Japan). Both materials were applied
and cured according to the manufacturers instructions.
The cavities were filled with one of three composites
varying in consistency or with a combination of materials.
Figure 2. Section of the experimental preparation design showing the

preparation part and the fissure part.
11
Opdam & Others: Adaptation of Composite in Micropreparations

The following composites were used in the study:
- A highly viscous or packable composite supplied in
pre-loaded tips; (Prodigy Condensable, Kerr,
Romulus, MI 48174, USA)
- A medium-viscous or syringeable composite (Clearfil
Photo Posterior, Kuraray, Osaka, Japan) inserted in
tips by the operator (Centrix, Hawe Neos, Bioggio,
Switzerland).
- A low viscous or flowable composite (Revolution,
Kerr) injected by a needle tip into the cavity.
These composites were placed in bulk or in combinations. In two groups, a layer of flowable composite was
injected on the cavity floor. This layer was left uncured
and, subsequently, a second layer of a composite with
higher viscosity was applied into the cavity. This resulted in the flowable composite being extruded from the
cavity by the more viscous material. This technique
was called the snowplough-technique. After removing
the excess material, the restoration was cured for 40
seconds from the occlusal surface. In another three
groups, the initial layer of flowable composite was
cured before one of the three composites was applied.
Finally, the composites were applied according to one
of the following protocols:
Group 1. The cavities were filled in bulk with the
packable composite and cured for 40 seconds.
Group 2. According to Group 1, however, a syringable
composite was used.
Group 3. According to Group 1, however, a flowable
composite was used.
Group 4. Placement of a flowable composite in the
cavity leaving it uncured before injecting the syringable
composite (snowplough technique). The restoration was
cured for 40 seconds.
Group 5. Same as Group 4, but instead of a syringable
composite, a packable composite was used.
Group 6: The initial layer of flowable composite was
applied on the cavity floor and cured for 20 seconds.
Subsequently, the restoration was completed with a
second layer of flowable composite that was also cured
for 20 seconds.
Group 7: The initial layer of flowable composite was applied on the cavity floor and
cured for 20 seconds. Subsequently, the
restoration was completed by injecting the
syringable composite that was cured for 20
seconds.
Group 8: The initial layer of flowable composite was applied on the cavity floor and
cured for 20 seconds. Then, the restoration
was completed with the packable composite,
which was packed with a hand-instrument and cured

for 20 seconds.
Each group consisted of 19 restorations. All restorations were placed by one operator, a dental student
close to graduation. All curing procedures were performed from the occlusal surface using a Kulzer
Translux CL polymerization unit (Kulzer, Wehrheim,
Germany, output 700mW/cm2).
The perspex blocks with the restorations were stored
for a minimum of 24 hours, then placed in a milling
machine (Friedrich Deckel FP1, Mnchen, Germany)
and sectioned through the middle of the restorations.
The restorations were evaluated by two independent
examiners using a light microscope (Carl Zeiss, Jena,
Germany) at magnification 10x. Both the section with
the preparation and excavation and the prepared fissure were inspected for voids and porosities.
Score 0 = voids or porosities absent, score 1 = voids or
porosities present.
In the case of disagreement between the two examiners, the samples were examined again until a mutual
agreement was obtained.
Differences between the groups were calculated for
their statistical significance using Fischers exact test
at p=0.05. The p-value was corrected for multiple group
testing.
RESULTS
Table 1 summarizes the results of the presence of voids
inside the restorations. Based on the results, it is obvious that a restoration completely free of porosities and
voids was hard to achieve. Only in Group 4, eight
restorations were scored as free of voids. In Group 2,
three restorations, and in Group 7, only one restoration
was totally free of voids. All the other restorations had
some porosities or voids, either in the fissure or in the
preparation part of the restoration.
The best results were achieved in Group 4, where an
uncured flowable was combined with a syringable composite. In the preparation portion of the sections, this
technique was significantly better compared to all other
techniques (p<0.05) except the syringable composite
Table 1: Presence of Voids in the Total Restoration

Group
No Voids (n)
Voids Present (n)
uncured flowable+syringable
Application Mode
11
syringable bulk
16
cured flowable + syringable
18
flowable bulk
19
uncured flowable +packable
19
cured flowable + packable
19
flowable in 2 layers
19
packable
19
12
Operative Dentistry
Table 2: Presence of Voids in the Preparation Part

Group
Application Mode
No Voids (n)
Voids Present (n)
12
syringable bulk
11
14
flowable bulk
18
18
19
19
packable
19
*p>0.05 (Fishers exact test)
Table 3: Presence of Voids in the Fissure Part

Group
Application Mode
No Voids (n)
Voids Present (n)
14
syringable bulk
14
14
packable
15
16
18
18
flowable bulk
19
*p>0.05 (Fishers exact test)
injected in bulk (Group 2) and the cured flowable combined with the syringable (Group 7). These findings are
shown in Table 2. In the fissure part of the sections,
Group 4 was significantly better than all the other techniques (p<0.05) as documented in Table 3.
DISCUSSION
This study compared the homogeneity of different types
of composites in small occlusal preparations. For that
purpose, a flowable, syringable and packable composite
was used to restore standardized cavities. Only a few
studies regarding consistencies of dental composites are
available (Opdam & others, 1996b; Tyas & others,
1998), and the results of those studies are quickly outdated due to the ongoing introduction of newer products
that replace older versions. The choice of syringable
material was based on the results of a 1996 study
(Opdam, & others, 1996b). The other materials were
selected based on the information supplied by the manufacturers, which indicated that their materials were
flowable or packable.
From other studies on homogeneity of restorations, it
is known that injection of a syringable composite
results in a better restoration with less voids compared
to a packing technique with a highly viscous composite
(Opdam & others, 1996a; Opdam & others, 2002).
Furthermore, it has been shown that small Class I
preparations are more difficult to restore adequately
when compared to larger cavities (Opdam & others,
2002). In this study, small occlusal preparations were
made, as this type of preparation is the

most tooth-preserving technique for the
treatment of primary caries lesions (Peters
& McClean, 2001a). These preparations
can be ground using a small diamond stone
or alternately by using the air-abrasion
technique. In all cases, this will result in
narrow cavities of different shapes and
form. The cavity design used in this study
can be considered to be clinically representative of a small, occlusal dentin lesion
that can be treated by an opening in the
enamel followed by an enamel-undermining excavation procedure along the enamel-dentin junction. An adjacent, discolored
fissure was ground to remove discolored
caries tissue at the outline. As adhesion to
enamel and dentin was not the subject of
the study, it was possible to make the standardized preparations in perspex blocks.
Applying a bonding agent ensured that a
clinical representative surface of the cavity
was present when applying the resin composite.
The results of this study show that it is

almost impossible to restore a small, narrow preparation without including voids. The fact that
only one operator placed all the restorations may have
significance. In a study regarding the presence of voids
inside Class II restorations, an experienced operator
achieved better results (Chuang & others, 2001a).
However, in another study with six operators, including
a dental student, where Class I restorations with various types of composite were placed, it was concluded
that the homogeneity was not influenced by the operators. In that study, preparations were larger than in the
current study. Nevertheless, only a small number of
sections of restorations were free of porosities, while it
was concluded that small preparations were more difficult to restore free of voids than larger preparations
(Opdam & others, 2002).
Therefore, it can be assumed that a micropreparation
is even more difficult to fill properly, which is in accordance with the results of this study. It appears almost
impossible to avoid any porosities inside the restoration. Light-curing resin composites taken directly taken
from the syringe already contain porosities 0.05 to 1.4%
by volume (Fano, Ortalli & Pozela, 1995). In the 1995
study, the highest amount of porosities was found in a
highly viscous resin composite. Furthermore, injecting
the light-curing materials out of a needle resulted in
less porosities with the exception of stiff composites,
which showed an increase in porosities. A study by
Jrgensen & Hisamitsu (1983) recorded an increase of
porosites in composites that were injected due to the
composite sticking to the inside of the tip. However, by
13
Opdam & Others: Adaptation of Composite in Micropreparations

coating the inside of the tip with a thin layer of resin
before inserting the composite, the number of porosities
was largely reduced. In this study, this procedure was
followed when filling the tips with the syringable composite.
The clinical relevance of the current study is not clear.
Large porosities at the tooth-restoration interface may
result in gross microleakage and subsequent failure of
the restoration due to caries. In a study by Estafan,
Estafan & Leinfelder, 2000, Class I restorations made
with a condensable composite without a first layer of
flowable composite exhibited voids at the restorationpreparation interface, while the voids were reduced
when combined with a flowable composite.
Furthermore, large porosities and voids will have a
weakening effect on the whole restoration, since an
interface between improper adepted materials may act
as a site where failure can start (Huysmans & others,
1996). Some clinical studies report the presence of
porosities detected on radiographs (Kreulen & others,
1995). However, clinical failures due to such porosities
are seldom reported, and the clinical relevance of a
small porosity inside a minimal invasive restoration, as
made in the current study, seems to be minimal.
Nevertheless, voids and porosities appear to have a
negative effect on the physical properties of the material and should be avoided as much as possible.
This study showed that the technique of placing an
initial thin layer of flowable composite, left uncured,
followed by injecting a syringable composite, is best
suited to fill a micropreparation completely. With this
so-called snowplough-technique, the flowable composite helps to wet the cavity walls and will be pressed out
of the cavity when injecting the more viscous syringable
composite. One should expect that this technique would
also be effective in combination with a packable composite. Nevertheless, this study demonstrated that this
was not the case. It is possible that the stiff, packable
composite contains irregularities that result in entrapment of air at the interface with the flowable composite. Furthermore, it is possible that the porosities were
already present inside the composite material (Fano &
others, 1995).
Another explanation could relate to the size of the tip
in which the composites were placed. The end of the
Hawe tip can be placed in the preparation and, therefore, the composite can be injected while in direct contact with the flowable composite. The packable composite is delivered in a tip with a large diameter to facilitate injection of the stiff material. The end of such a tip
would not fit into the opening of the cavity. When injecting the composite, the opening of the tip is positioned
above the occlusal surface and the composite does not
fill the cavity from the cavity floor and may trap air.
Other studies, looking in greater detail at the voids,
found a reduction in the number of voids if a flowable

composite is placed in advance of a packable composite
(Chuang & others, 2001b). A flowable composite
appeared to be beneficial in reducing microleakage of
restorations made with stiff, packable composites
(Tung, Estafan & Scherer, 2000) but failed to demonstrate the same positive effect in combination with a
normal microhybrid composite (Jain & Belcher, 2000).
Though the reduction of microleakage by combining a
flowable with a packable composite is evident,
microleakage was still greater than when a normal
microhybrid composite was used without a flowable lining (Leevailoj & others, 2001).
Using a flowable composite may also have some disadvantages. The volumetric curing-shrinkage of flowable composites is considerably higher and most of the
physical properties are less than from normal hybrid
composites (Bayne & others, 1998). In case the initial
layer of flowable composite is not cured before inserting
the stiffer composite, most of the flowable composite
will be pressed out of the cavity, avoiding thick layers of
the flowable composite at the tooth-restoration interface in locations loaded by occlusion and articulation. It
is often suggested that micropreparations are best
restored using a flowable composite (Peters & McClean,
2001b). Advantages of these materials are their low viscosity, making it possible to let the material flow into a
narrow cavity, and the application with a needle tip,
which facilitates placement deep into the fissure.
However, in this study, restorations placed with only
flowable composite, either in bulk or in an incremental
technique using the needle tip, resulted in the presence
of voids in almost all cases, indicating that these materials are not as easy to manipulate as they appear.
Chuang & others (2001a) already concluded that an
experienced operator was more successful in placing a
flowable composite without entrapping air. Others stated that flowable composites are difficult to manipulate
and air is easily entrapped on removal of the syringe
(Leevailoj & others, 2001). Wibowo & Stockton (2001)
found it easy to let a flowable composite adapt to the
deepest part of an approximal box but noted that after
sectioning, incorporating porosities inside the restoration were frequent. Incorporating air bubbles is not
only a problem of flowable composites but is also seen
in other low, viscous materials such as pit and fissure
sealants (Lekka & others, 1991). Flowable composites
are also recommended as fissure sealants but are not
always an improvement, as sealing with a flowable
composite requires the use of an adhesive resin.
CONCLUSIONS
It can be concluded from this study that narrow
occlusal preparations are difficult to restore completely
free of voids. Flowable composites are sometimes recommended for this purpose but they showed disap-
14
pointing results in this study when used as the only
restorative material. The most homogeneous restorations were obtained by combining an initial layer of
uncured flowable composite with a second layer of
medium-viscous composite injected into the cavity.
Packable composites exhibited poor results even in
combination with an uncured flowable composite.
(Received 29 January 2002)
References
Ausiello P, Davidson CL, Cascone P, de Gee AJ & Rengo S (1999)
Debonding of adhesively restored deep Class II MOD restorations after functional loading American Journal of Dentistry
12(2) 84-88.
Operative Dentistry
Jain P & Belcher M (2000) Microleakage of Class II resin-based
composite restorations with flowable composite in the proximal box American Journal of Dentistry 13(5) 235-238.
Jrgensen KD & Hisamitsu H (1983) Porosity in microfil
restorative composites cured by visible light Scandinavian
Kreulen CM, Van Amerongen WE, Akerboom HBM &
Borgmeijer PJ (1995) Two-year results with box-only resin
composite restorations Journal of Dentistry for Children 62(6)
395-400.
Leevailoj C, Cochran MA, Matis BA, Moore BK & Platt JA
(2001) Microleakage of posterior packable resin composites
with and without flowable liners Operative Dentistry 26(3)
302-307.
Lekka M, Papagiannoulis L & Eliades G (1991) Porosity of pit and
fissure sealants Journal of Oral Rehabilitation 18(3) 213-220.
Bayne SC, Thompson JY, Swift EC Jr, Stamatiades P &

Wikerson MA (1998) A characterization of first-generation
flowable composites Journal of the American Dental
Association 129(5) 567-576.
Opdam NJM, Roeters FJM, Peters MCRB, Burgersdijk RC &

Teunis M (1996a) Cavity wall adaptation and voids in adhesive Class I resin composite restorations Dental Materials
12(4) 230-235.
Chuang SF, Liu JK, Chao CC, Liao FP & Chen YH-H (2001a)
Effects of flowable composite lining and operator experience on
microleakage and internal voids in Class II composite restorations Journal of Prosthetic Dentistry 85 177-183.
Opdam NJM, Roeters FJM, Peters MCRB, Burgersdijk RC &

Kuys R (1996b) Consistency of resin composites for posterior
use Dental Materials 12 350-354.
Chuang SF, Liu JK & Jin YT (2001b) Microleakage and internal

voids in Class II composite restorations with flowable composite linings Operative Dentistry 26(2) 193-200.
Opdam NJM, Roeters FJM, Joosten M & Veeke O (2002)

Porosities and voids in Class I restorations by six operators
using a packable or syringable composite Dental Materials
18(1) 58-63.
Estafan D, Estafan A & Leinfelder KF (2000) Cavity wall adaptation of resin-based composites lined with flowable composites American Journal of Dentistry 13(4) 192-194
Peters MC & McClean ME (2001a) Minimally invasive operative

care. I. Minimal intervention and concepts for minimally invasive cavity preparations Journal of Adhesive Dentistry 3(1) 7-16.
Fano V, Ortalli I & Pozela K (1995) Porosity in composite resins

Biomaterials 16(17) 1291-1295.
Peters MC & McClean ME (2001b) Minimally invasive operative

care. II. Contemporary techniques and materials: An overview
Journal of Adhesive Dentistry 3(1) 17-31.
Goldstein RE & Parkins FM (1994) Air-abrasive technology: Its

new role in restorative dentistry Journal of the American
Dental Association 125(5) 551-557.
Huysmans MCDNJM, Vander Varst PGT, Lautenschlager EP &
Monaghan P (1996) The influence of simulated clinical
handling on the flexural and compressive strength of posterior composite restorative materials Dental Materials 12(2)
116-120.
Tung FF, Estafan D & Scherer W (2000) Microleakage of a condensable resin composite: An in vitro investigation
Quintessence International 31(6) 430-434.
Tyas MJ, Jones DW & Rizkalla AS (1998) The evaluation of
resin composite consistency Dental Materials 14 424-428.
Wibowo G & Stockton L (2001) Microleakage of Class II composite
restorations American Journal of Dentistry 14(3) 177-185.
Effects of In-Office
Bleaching Products on
Surface Finish of
Tooth-Colored Restorations
P Wattanapayungkul AUJ Yap
Clinical Relevance
In-office bleaching systems that employ strong oxidizing agents are not detrimental to
the surface finish of tooth-colored restorative materials.
SUMMARY
A number of high power in-office bleaching
products have recently been re-introduced into
the market. The use of such strong oxidizing
agents has raised questions as to possible
adverse effects on tooth structure and restorative materials. This study evaluated the effects of
35% carbamide peroxide (Opalescence Quick)
and 35% hydrogen peroxide (Opalescence Xtra)
on the surface finish of four tooth-colored
restorative materials (Spectrum TPH, Dyract AP,
Reactmer and Fuji II LC). Twenty-seven matrixfinished specimens of each material were fabricated, stored in distilled water at 37C for seven
days and randomly divided into three groups.
Specimens in Group 1 were stored in distilled
water at 37C (control). Specimens in Groups 2
and 3 were treated with 35% carbamide peroxide
and 35% hydrogen peroxide, respectively. A total
*Pranee Wattanapayungkul, DDS, MSD, teaching fellow,
Department of Restorative, Faculty of Dentistry, National
University of Singapore
Adrian UJ Yap, BDS, MSc, PhD, FAMS, FADM, FRSH, associate
professor, Department of Restorative Dentistry, Faculty of
Dentistry, National University of Singapore
*Reprint request: Department of Restorative, Faculty of
Dentistry, National University of Singapore, 5 Lower Kent
Ridge Road, Singapore 119074, Republic of Singapore; e-mail:
[email protected]
of three 30-minute bleaching sessions were conducted at one-week intervals. Storage medium
during the hiatus period was distilled water at
37C. Surface roughness measurements were carried out using profilometry after each bleaching
session. Data was analyzed using ANOVA/
Scheffes test at a 0.05 significance level. No significant difference in surface roughness was
observed between the bleached and the control
groups for all materials. In-office bleaching products are not detrimental to the surface finish of
composites, compomers, giomers and resin-modified glass ionomer cements.
INTRODUCTION
Over the last decade, home vital tooth bleaching has
attracted the interest of patients and dentists due to its
high success rates, ease of use and media publicity. This
procedure utilizes low concentrations of hydrogen peroxide (3% to 7%) or carbamide peroxide (10% to 20%).
Recently, new in-office bleaching products that utilize
high concentrations of hydrogen peroxide or carbamide
peroxide have been re-introduced. The latter procedure,
which involves 30% to 35% carbamide peroxide or
hydrogen peroxide, is totally under the dentists control
and has the potential for bleaching quickly in situations in which it is effective. High concentrations of
hydrogen peroxide have been reported to cause surface
roughening of teeth and etching-like patterns (Flaitz &
16
Operative Dentistry
Hicks, 1996; Shannon & others, 1993; Zalkind & others,

1996).
84095, USA) and 35% hydrogen peroxide (Opalescence

Xtra, Ultradent).
The effects of such strong oxidizing agents on the

physico-mechanical properties of restorative materials
have, however, not been widely studied. Surface roughness of restorations is one clinically important physical
property that warrants investigation. The surface finish of restorations influences aesthetics and oral health,
as the presence of irregularities may influence appearance, plaque retention, surface discoloration and gingival irritation (Weitman & Eames, 1975; Dunkin &
Chambers, 1983; Chan, Fuller & Hormati, 1980;
Shintani & others 1985). Studies have shown that
using home bleaching systems increases the surface
roughness of some composite restoratives (Cooley &
Burger, 1991; Bailey & Swift, 1992). Mor & others
(1998) found that 10% carbamide peroxide and 10%
hydrogen peroxide caused a significant increase in surface adherence of S mutans and S sobrinus, while a
decrease in adherence of Actimimyce viscosus was
found after treatment with 10% hydrogen peroxide.
Little is known about the effects of in-office bleaching
systems on the surface finish of composites, compomers, giomers or PRG composites and resin-modified
glass ionomer cements. Giomer is a new category of
hybrid aesthetic restorative material that employs the
use of pre-reacted glass ionomer (PRG) technology (Yap
& Mok, 2002). Unlike compomers, the fluoroalumino
silicate glass is reacted with polyacrylic acid prior to
inclusion into the resin matrix. The bonding and handling is similar to compomers. The manufacturers
claims include fluoride release and recharge, smooth
surface finish, excellent aesthetics and clinical stability.
The restorative materials were placed in the rectangular recesses (4 mm long x 3 mm wide x 2 mm deep) of
customized acrylic molds and covered with acetate
matrix strips (Hawe-Neos Dental, Bioggio,
Switzerland) to achieve the smoothest surface finish
(Bauer & Caputo, 1983; Pratten & Johnson, 1988; Yap,
Lye & Sau, 1997) and to avoid problems of operatorinduced variables during finishing and polishing. A
glass slide was placed over the molds and pressure was
applied to extrude excess material. The restoratives
were light polymerized according to manufacturers
cure times with a Poly LUX II light cure unit (KaVo
Dental, Warthausen, Germany). Mean intensity of the
light source (597 10 mW/cm2) was determined with a
radiometer (CureRite, EFOS INC, Ontario, Canada)
prior to starting the experiment. Cure times were as follows: Spectrum20 seconds; Dyract40 seconds;
Reactmer30 seconds and Fuji II LC20 seconds. The
specimens were stored in distilled water at 37C for
seven days and randomly divided into three groups.
Specimens in Group 1 were not exposed to any bleaching systems and served as the control group. Group 2
specimens were bleached with 35% carbamide peroxide
(Opalescence Quick) for 30 minutes without any light
activation or reapplication of bleaching gel. Group 3
specimens were bleached with 35% hydrogen peroxide
(Opalescence Xtra) for 15 minutes with 20 seconds light
activation. After 15 minutes, the gel was washed away,
fresh gel was reapplied and the aforementioned treatment was repeated. The combination of the two cycles
resulted in a total bleaching time of 30 minutes (Table
1). After bleaching, the specimens were washed and
surface roughness measurements (Ra) were taken at
the center of the specimens using a profilometer
(Surftest SV-400; Mitutoyo, Kanagawa, Japan). The
average surface roughness, Ra values is the arithmetric
average height of roughness component irregularities
from the mean line measured within the sampling
length. Smaller Ra values indicate smoother surfaces.
Four sampling lengths of 0.25 mm were used, giving a
total evaluation length of 1 mm. The specimens were
bleached for another two sessions at one-week intervals. Storage medium for all groups during the hiatus
period was distilled water at 37C. All statistical analysis was carried out at significance level 0.05. Multiple
This study evaluated the effects of in-office bleaching

systems on the surface finish of different tooth-colored
restorative materials. The surface roughness of the different materials was also compared.
Four tooth-colored restorative materials and two commercial bleaching agents were selected for this study.
The restorative materials included a composite resin
(Spectrum Dentsply/Detrey, Konstanz, Germany), a
compomer (Dyract AP, Dentsply/Detrey, Konstanz,
Germany), a giomer (Reactmer, Shofu Inc, Kyoto,
Japan) and a resin modified glass ionomer (Fuji II LC
Capsule,
GC
Corporation,
Table 1: Summary of Treatment Groups
Tokyo,
Japan).
The
bleaching
Groups
Bleaching Agents
agents were 35%
No treatment with
Group 1
carbamide perox(Control)
bleaching agents
ide (Opalescence
Group 2
35% Carbamide Peroxide
(Opalescence Quick)
Quick, Ultradent
Products, Inc, UT
Group 3
35% Hydrogen Peroxide
(Opalescence Xtra)
Treatment Time
Light Activation
Reapplication of Gel
Not Applicable
Not Applicable
Not Applicable
30 minutes
No
No
30 minutes
Yes
Every 15 minutes
17
Wattanapayungkul & Yap: Effects of In-Office Bleaching Products on Tooth-Colored Restorations

Table 2: Mean Ra Values [10-2] of Four Materials After the Various Bleaching Sessions (standard deviations in parenthesis)
Materials
Spectrum TPH
Dyract AP
Reactmer
Fuji II LC
Group 1 Group 2 Group 3 Group 1 Group 2 Group 3 Group 1 Group 2 Group 3 Group 1 Group 2 Group 3
Session 1
5.00
(1.58)
5.22
(2.22)
4.78
(1.09)
5.57
(1.58)
6.00
(1.94)
6.11
(2.09)
8.44
(1.88)
7.22
(1.64)
8.89
(3.86)
10.89
(1.83)
10.11
(2.26)
10.89
(2.09)
Session 2
4.56
(0.5)
5.00
(0.70)
5.00
(1.50)
7.78
(2.54)
6.44
(1.81)
7.00
(2.17)
7.56
(2.92)
9.44
(4.44)
10.33
(3.00)
9.33
(2.45)
11.44
(2.04)
8.89
(1.27)
Session 3
4.78
(1.09)
5.00
(1.41)
4.56
(0.76)
8.22
(1.79)
6.22
(1.48)
6.44
(1.67)
8.78
(2.81)
8.37
(2.24)
9.33
(3.76)
9.11
(1.36)
9.67
(1.12)
8.33
(1.41)
Note: At all treatment sessions, there is no significant difference between Group 1, 2 and 3 for all materials.
Table 3: Comparison of Ra Values Between Materials at the Various Treatment

Sessions
Differences
Session 1
Session 2
Session 3
Group 1
Spectrum, Dyract < Reactmer < Fuji II LC
Group 2
Spectrum, Dyract, Reactmer < Fuji II LC
Group 3
Spectrum < Reactmer, Fuji II LC

Dyract < Fuji II LC
Group 1
Spectrum < Dyract, Fuji II LC
Group 2
Spectrum, Dyract < Reactmer, Fuji II LC
Group 3
Group 1
Spectrum < Dyract, Reactmer, Fuji II LC
Group 2

Dyract < Fuji II LC
Group 3

Dyract < Reactmer
*Results of one-way ANOVA/Scheffe test at significance level 0.05.

< indicates statistically significant difference.
analysis of variance (ANOVA) was used to determine

the interaction among various variables. One-way
ANOVA and Scheffes post-hoc test were used to establish the effects of bleaching systems on individual materials and to compare the surface roughness of the various materials after bleaching.
RESULTS
The mean Ra values of four materials after the various
bleaching sessions are shown in Table 2, while Table 3
shows the results of statistical analysis comparing
materials.
Multiple analysis of variance showed no significant
interaction between materials, treatment groups and
sessions. At all treatment sessions, no significant difference in surface roughness was observed between the
control and the bleached groups for all materials. The
use of in-office bleaching systems was therefore not
detrimental to the surface finish of the tooth-colored
restorative materials evaluated. Significant differences
in surface roughness were, however, observed between
materials. Differences between materials varied somewhat depending on the treatment session. For all treatment sessions and groups, Spectrum was significantly
smoother than Fuji II LC. After the third

treatment session, Spectrum was significantly smoother than Dyract, Reactmer
and Fuji II LC for the control group. No
significant difference was observed
among the latter three materials. For
the bleached groups, Spectrum was only
significantly smoother than Reactmer
and Fuji II LC. No significant difference
in surface roughness was observed
between Spectrum and Dyract. Ra values obtained with Dyract were significantly lower than Fuji II LC for Group 2
and Reactmer for Group 3.
DISCUSSION
Vital tooth bleaching using high concentrations of hydrogen peroxide was

described as early as the 1900s (Henderson, 1910;
Fisher, 1911; Ames, 1937). The procedures were both
complicated and time-consuming; furthermore, gingival
irritation was relatively frequent. The new in-office
bleaching products being marketed also utilize high concentrations of carbamide peroxide or hydrogen peroxide.
The delivery systems are, however, more friendly and
the consistency more workable. The concentration of
hydrogen peroxide and the pH of bleaching products is
important to clinicians as they may have adverse effects
on both tooth structure and restorations. Price, Sedarous
& Hiltz (2000) measured the pH of 26 tooth-whitening
products available in the market. They found that home
bleaching products have a pH range from 5.66 to 7.35.
The pH range of in-office bleaching systems was lower
and ranged from 3.67 to 6.53. Among the systems evaluated, Opalescence Xtra had the lowest mean pH (3.67)
and Opalescence Quick had the highest mean (6.53).
These two bleaching systems were thus selected for the
current study. With the exception of conventional glass
ionomer cements, the materials selected represent the
entire continuum of tooth-colored restorative materials
currently available.
18
Surface alterations to resin composites and glass
ionomer cements after exposure to bleaching agents have
been reported (Bailey & Swift, 1992; Lee, Grimuado &
Shen, 1999; Kilimitzoglou & Wolff, 2000; Turker &
Biskin, 2000). The products used in these studies were
at-home systems and over-the-counter bleaching products. Roughening was suggested to result from loss of
matrix rather than filler particles (Bailey & Swift, 1992).
Other studies (Burgess & others, 1991; Souyias,
Hoelscher & Neme, 2000) have, however, demonstrated
no significant increase in surface roughness. The apparent discrepancies may be explained, in part, by the differences in experimental methodologies and bleaching
agents used. While some researchers have adopted clinically relevant protocols, others have employed continuous exposure of restorative materials to bleaching agents
over stipulated time periods. The frequency of change of
bleaching agents may also contribute to the disparity in
results.
The contact time between bleaching products and
teeth/restorations for home vital bleaching is much
longer than that for in-office vital bleaching. In this
study, three sessions of 30-minute bleaching treatment
with one-week intervals were employed to simulate clinical conditions. At all treatment sessions, no significant
difference in surface roughness was observed between
the control and bleached groups for all materials. Using
Opalescence Quick and Xtra are, therefore, not detrimental to the surface finish of the composite, compomer,
giomer and resin-modified glass ionomer cements evaluated clinically. It is important to note that results may be
material dependent, as some restorative materials are
pH sensitive. For example, the surface finish of smart
composites, such as Ariston pHc (Vivadent, Schann,
Liechtenstein) that use a low oral pH to increase fluoride
release (Combe & Douglas, 1998), may be affected by the
low pH of some hydrogen peroxide-based bleaching systems. For all treatment sessions and groups, the composite (Spectrum) was significantly smoother than the
resin-modified glass ionomer cement (Fuji II LC). This
can be explained by the differences in microstructure.
The mean particle size of Spectrum is under 1 m, while
that of Fuji II LC is 4.8 m. Treatment with strong oxidizing agents appeared to stabilize the surface of the
compomer evaluated. For the control group, a gradual
increase in roughness was observed for Dyract specimens. The Ra values of bleached Dyract specimens, however, remained relatively stable over the experimental
period. The aforementioned accounts of the significantly
smoother surface of Dyract as compared to
Reactmer/Fuji II LC for the bleached groups after three
weeks storage in water at 37C. Compomers are known
to uptake water and expand (Yap & others, 2000). Water
uptake is necessary for establishing an acid-based reaction and fluoride release (Yap, Khor & Foo, 1999). Water
uptake may result in stress corrosion and complete or
Operative Dentistry
partial debonding of fillers leading to increased surface
roughness (Sderholm, 1983). The exact mechanism for
the stabilization effect of in-office bleaching agents is not
known and warrants further investigation.
Although in-office bleaching systems are not detrimental to the surface finish of tooth-colored restoratives, care
should still be taken when bleaching teeth with restorations. Hydrogen peroxide was found to have higher levels of penetration into the pulp chamber in restored teeth
compared to sound teeth (Gokay, Tuncbilek, & Ertan,
2000). The mechanical properties and, durability of
tooth-colored restoratives may also be affected by inoffice bleaching agents. Dentists should, therefore, limit
treatment time to as short as possible since extended
bleaching treatment with such high concentrations of
peroxide along with low pH may cause some alterations
to both tooth structure and restorations.
CONCLUSIONS
1. The use of in-office bleaching systems that
employ strong oxidizing agents is not detrimental to the surface finish of composite, compomer,
giomer and resin-modified glass ionomer
cements evaluated.
2. The surface finish of the composite Spectrum
was significantly better than the resin-modified
glass ionomer regardless of bleaching treatment.
(Received 5 February 2002)
References
Ames JW (1937) Removing stains from mottled enamel The
Dental Cosmos 52 701-702.
Bailey SJ & Swift EJ (1992) Effects of home bleaching products
on composite resin Quintessence International 23(7) 489-494.
Bauer JG & Caputo AA (1983) The surface of composite resin
finished with instruments and matrices Journal of Prosthetic
Dentistry 50(3) 351-357.
Burgess JO, Dutton FB, Fray SE, Luce TW, Cronin R Jr &
Summit JB (1991) Roughness of six restorative materials
stored in bleaching Journal of Dental Research 70 Abstract
#952 p 385.
Chan KC, Fuller JL & Hormati AA (1980) The ability of foods to
stain two composite resins Journal of Prosthetic Dentistry
43(5) 542-545.
Combe EC & Douglas WH (1998) The future of dental materials
Dental Update 25 411-417.
Cooley RL & Burger KM (1991) Effect of carbamide peroxide on
composite resins Quintessence International 22(10) 817-821.
Dunkin RT & Chambers DW (1983) Gingival response to Class
V composite resin restorations Journal of the American
Wattanapayungkul & Yap: Effects of In-Office Bleaching Products on Tooth-Colored Restorations

Fisher G (1911) The bleaching of discolored teeth with H2O2
The Dental Cosmos 53 246- 247.
Flaitz CM & Hicks MJ (1996) Effects of carbamide peroxide
whitening agents on enamel surfaces and caries-like lesion
formation: A SEM and polarized light microscopic in vitro
study Journal of Dentistry for Children 63(4) 249-256.
Gokay O, Tuncbilek M & Ertan R (2000) Penetration of the pulp
chamber by carbamide peroxide bleaching agents on teeth
restored with a composite resin Journal of Oral
19
Shintani H, Satou J, Satou N, Hayashihara H & Inoue T (1985)

Effects of various finishing methods on staining and accumulation of Streptococcus mutans HS-6 on composite resins
Sderholm KJM (1983) Leaking of fillers in dental composites
Souyias J, Hoelscher DC & Neme AL (2000) Effect of bleaching
on posterior composite materials Journal of Dental Research
79 Abstract #1077 p 278.
Henderson HJ (1910) Bleaching teeth with pyrozone and sodium dioxide The Dental Cosmos 52 701-702.
Turker SB & Biskin T (2000) Effect of bleaching agents on the

surface properties of different esthetic restorative materials
Journal of Dental Research 79 Abstract #2356 p 438.
Kilimitzoglou D & Wolff MS (2000) The surface roughness of a

microfil and hybrid composite after exposure to carbamide
peroxide Journal of Dental Research 79 Abstract #1070 p 277.
Weitman RT & Eames WB (1975) Plaque accumulation on composite surfaces after various finishing procedures Journal of
the American Dental Association 91(1) 101-106.
Lee J, Grimuado NJ & Shen C (1999) Bleaching effects on resin

reinforced light-cured glass ionomer Journal of Dental
Research 78 Abstract #1700 p 277.
Yap AUJ, Lye KW & Sau CW (1997) Surface characteristics of

tooth-colored restoratives polished utilizing different polishing systems Operative Dentistry 22(6) 260-265.
Mor C, Steinberg D, Dogan H & Rotstein I (1998) Bacterial

adherence to bleached surfaces of composite resin in vitro
Oral Surgery Oral Medicine Oral Pathology Oral Radiology
Endodontics 86(5) 582-586.
Yap AUJ, Khor E & Foo SH (1999) Fluoride release and antibacterial properties of new-generation tooth-colored restoratives Operative Dentistry 24(5) 297-305.
Pratten DH & Johnson GH (1988) An evaluation of finishing

instruments for an anterior and a posterior composite
Journal of Prosthetic Dentistry 60(2) 154-8.
Price RBT, Sedarous M & Hiltz GS (2000) The pH of toothwhitening products Journal of the Canadian Dental
Shannon H, Spencer P, Gross K & Tira D (1993)
Characterization of enamel exposed to 10% carbamide peroxide bleaching agents Quintessence International 24(1) 39-44.
Yap AUJ, Wang HB, Siow KS & Gan LM (2000) Polymerization

shrinkage of visible-light-cured composites Operative
Dentistry 25(2) 98-103.
Yap AUJ & Mok BYY (2002) Surface finish of a new hybrid aesthetic restorative material Operative Dentistry 27(2) 161-166.
Zalkind M, Arwaz R, Goldman A & Rotstein I (1996) Surface
morphology changes in human enamel, dentin and cementum
following bleaching: A scanning electron microscopy study
Endodontics and Dental Traumatology 12(2) 82-88.
Effects of Regional Enamel

and Prism Orientation
on Resin Bonding
Y Shimada J Tagami
Clinical Relevance
Within the limitation of this in vitro study, the bond strengths of the self-etching primer
system and the one-bottle adhesive system were influenced by the anisotropic structure
of enamel. The effect of the self-etching primer system was less influenced than that of
the one-bottle adhesive system.
SUMMARY
Human enamel, with its prismatic, rod-like
apatitic morphology, is an anisotropic material.
Because of this structural anisotropy, variation
in enamel bonding sites might influence the
bonding ability of current adhesive systems. This
study investigated the effects of regional enamel
and the direction of enamel sectioning on the
bonding ability of two commercially available
resin adhesives: a self-etching primer system
(Clearfil SE Bond) and a one-bottle adhesive system intended for use with a total-etch wet bonding technique (Single Bond). Two regions of
enamel, cuspal and mid-coronal enamel, were
chosen, then sectioned in three different directions, horizontally, axially and tangentially.
Slices of the sectioned enamel were then bonded
with each adhesive system and submitted to a
micro-shear bond test. The results of a microshear bond testing showed that the bonding of a
*Yasushi Shimada, DDS, PhD, instructor, Tokyo Medical and
Dental University, Tokyo, Japan
Junji Tagami, DDS, PhD, professor and chair, Tokyo Medical and
*Reprint request: Cariology and Operative Dentistry,
Department of Restorative Sciences, Graduate School, Tokyo
Medical and Dental University, 1-5-45 Yushima, Bunkyo-ku,
Tokyo 113-8549, Japan; e-mail: [email protected]
one-bottle adhesive system (Single Bond) to

enamel was high at the surface perpendicular to
the enamel prisms (40 MPa to 51 MPa) and low at
the surface parallel to the enamel prisms (24 MPa
to 27 MPa). In the case of a self-etching primer
system (Clearfil SE Bond), 35 MPa to 45 MPa
bond strengths were obtained from all surfaces.
The bond strengths of the two adhesive systems
were significantly influenced by the anisotropic
structure of enamel (p<0.05). However, the effect
of a self-etching primer system was less influenced by the orientation of the prismatic structure of enamel than that of a one-bottle adhesive
system (p<0.05).
SEM and CLSM microphotographs showed that
the self-etching primer effectively modified the
smear layer without being excessively destructive to the enamel.
INTRODUCTION
Human enamel mainly consists of apatitic calcium
phosphate arranged in a highly ordered prismatic
array (Lees & Rollins, 1972; Rasmussen & others,
1976; Munechika & others, 1984; Carvalho & others,
2000). Because of this microstructural anisotropy, variation in enamel bonding sites might influence the
bonding strength of direct restorative systems to this
substrate (Munechika & others, 1984; Shimada & oth-
Shimada & Tagami: Effects of Regional Enamel and Prism Orientation on Resin Bonding
ers, 1999a). However, the effects of regional variations
in tooth structure, such as the orientation of the enamel prisms, and the variability introduced by tooth sectioning (cavity preparation) on dental adhesion, are
still not completely understood.
Today, two kinds of adhesive systems are commercially available; they are the so-called one-bottle and
self-etching adhesives (Nakabayashi, 1984; Wang,
Nikaido & Nakabayashi, 1991; Tay & others, 1996;
Watanabe, Nakabayashi & Pashley, 1994). This study
investigated how regional enamel microstructural
variation and the effects of enamel sectioning (orientation of enamel prisms) influence the bonding ability of
a self-etching primer system and an acid-etch, one-bottle adhesive system. In addition, surfaces with the
bonding removed after the shear bond test for failure
mode, the adhesive interface between the enamel and
resin and the conditioned enamel surface without any
bonding were studied morphologically using scanning
electron microscopy (SEM) or confocal laser scanning
microscopy (CLSM). CLSM has been widely used in
biology for non-invasive and non-destructive imaging
in vivo of many organ tissues and has found numerous
applications in dental research (Watson, 1989;
Shimada & others, 1999b). Specimens examined by
CLSM do not require any special preparation and are
not subjected to the distortions caused by dehydration
that results from procedures such as SEM.
Two bonding systems were evaluated: Clearfil SE
Bond system, which includes a self-etching primer and
a bonding agent (Kuraray Co, Osaka, Japan), and
Single Bond system, which uses an etching gel with
35wt% phosphoric acid and a one-bottle adhesive (3M,
St Paul, MN 55144, USA). All the materials were handled according to manufacturers instructions.
Micro-shear bond tests were used to measure the
bonding in this study (Shimada & others, 1999a).
21
Tooth Slices
One hundred and ninety-two human molars were
stored under refrigeration in a preservative solution
containing 0.2 g sodium azide 100 mL distilled water
before sectioning. The cuspal and mid-coronal enamel
regions chosen as substrates for micro-shear bond test
were sectioned in one of three ways, horizontally, axially or tangentially (Figure 1a). Horizontal slices were
cut perpendicular to the tooth axis, axial slices were
cut parallel to the tooth axis and tangential slices were
cut parallel to the dentino-enamel junction. Because
each of the two regions of enamel was directionally sectioned three ways, a total of six kinds of slices were
examined. Thirty-two slices each, approximately 1.0
mm thick, were obtained from every six kinds of enamel
(192 slices total) by cutting with a slow rotating diamond blade (Struers Minitom, Struers, Copenhagen,
Denmark) under a flow of water. The enamel surfaces
were then resurfaced with wet 280-grit SiC paper until
a depth half way between the dentin enamel junction
and outer surface was obtained.
Micro-Shear Bond Test
Twenty enamel slices from the 32 slices, each corresponding to the six kinds of enamel slices, were randomly chosen (120 slices total) and further divided into
two groups (10 slices each) according to the adhesive
system used. Each slice underwent one of the following
treatments:
Group 1 (Single Bond system): Etched with Single
Bond etchant for 15 seconds, thoroughly rinsed for 15
seconds and blot-dried with absorbent paper for
removal of excess water, leaving a moist surface. Two
coats of Single Bond Adhesive were consecutively
applied, air thinned and light cured for 20 seconds.
Group 2 (Clearfil SE Bond system): Treated with
Clearfil SE Bond Primer for 20 seconds and dried.
Clearfil SE Bond was applied, air thinned and light
cured for 10 seconds.
Prior to the irradiation of bonding resin, an iris cut
from microbore Tygon tubing (R-3603, Norton
Performance Plastic Co, Cleveland, OH, USA) with an
internal diameter and a height of approximately 0.8
mm and 0.5 mm, respectively, was mounted on the
enamel to restrict the bonding area. A hybrid restorative resin composite, shade A3 (Clearfil AP-X, Kuraray
Co, Osaka, Japan) was placed into the cylinder and a
clear celluloid sheet was placed over the resin and gently pressed flat and irradiated for 40 seconds. In this
manner, very small cylinders of resin, approximately
0.8 mm in diameter and 0.5 mm in height, were bonded to the surface. The specimens were stored at room
temperature (23C) for one hour prior to removing the
tygon tubing. The specimens were then stored in water
at 37C for 24 hours.
Figure 1a: Tooth regions and sectioning directions.
22
Operative Dentistry
After the specimens were cooled to room temperature, their shear bond strength was measured by
micro-shear testing (Figure 1b). The tooth slice with
the resin cylinders was adhered to the testing device
(Bencor-Multi-T, Danville Engineering Co, San
Ramon, CA 94583, USA) with a cyanoacrylate adhesive (Zapit, Dental Ventures of America, Corona, CA
92882, USA), which, in turn, was placed in a universal
testing machine (Ez-test-500N, Shimazu Co, Kyoto,

Japan) for shear bond testing. A thin wire (diameter
0.20 mm) was looped around the resin cylinder, making contact with half of the cylinder base, and it was
held flush against the resin/enamel interface. A shear
force was applied to each specimen at a crosshead
speed of 1.0 mm/min until failure occurred.
The adhesive systems (SE Bond and Single Bond),
enamel regions (cuspal and mid-coronal regions) and
direction of sectioning (horizontally, axially and tangentially) were the three factors analyzed using threeway analysis of variance (ANOVA). After three-way
ANOVA, one-way ANOVA and multiple comparisons
were made using Fishers PLSD test. Statistical significance was defined as p<0.05.
All the debonded enamel surfaces after the shear
bond test were examined under an optical microscope
at 30x magnification and SEM (JXA840, JEOL Ltd,
Tokyo, Japan) so that the mode of failure and orientation of enamel prisms could be identified.
Morphological Study Using SEM and CLSM
Six additional slices, each corresponding to one of the
six kinds of enamel slices (36 slices total), were divided
into two subgroups (three slices each) and conditioned
or primed (but not bonded) in the same manner as the
shear bond test specimens. The slices treated with the
self-etching primer were given an additional 60-second
acetone rinse to remove any crystals or other residue
remaining from the primer. The surfaces were sputtercoated with gold and observed in a SEM.
Also, the remaining six slices that corresponded to

the six kinds of enamel slices (36 slices total) were used
for interfacial observations and bonded in the same
Figure 1b: Micro-shear bond test.
manner as that employed for the shear bond test. The
bonded specimens were cut in
Table 1: Micro-Shear Bond Strength (MPa) and Mode of Failure on Enamel
half, then ground and polished
using wet silicon carbide papers
Area
Direction
Bond Strength
Failure Mode
and diamond pastes of decreasMean (SD)
N
A*
C**
AC***
ing particle size down to 0.25 m.
Single Bond
One of the polished surfaces was
Cuspal
Horizontal
51.7 (5.2)
10
10
0
0
sputter-coated with gold and
Axial
24.9 (3.5)
10
0
10
0
observed under SEM. The other
Tangential
40.1 (3.6)
10
10
0
0
half of the polished pair surface
Mid-coronal
Horizontal
27.3 (5.3)
10
0
7
3
was viewed under CLSM
Axial
26.4 (3.9)
10
0
10
0
(1LM21H/W,
Lasertec
Co,
Tangential
42.7 (8.4)
10
8
0
2
Yokohama, Japan).
Clearfil SE Bond
Cuspal
Mid-coronal
Horizontal
45.5 (5.2)
10
10
RESULTS
Axial
36.6 (4.9)
10
10
Micro-Shear Bond Test
Tangential
41.9 (5.4)
10
10
Horizontal
39.6 (3.9)
10
The mean shear bond strength

and standard deviations in MPa
and mode of failure are shown in
Table 1. The results of the statistical analysis are shown in
Axial
35.7 (5.2)
10
10
Tangential
42.9 (7.5)
10
10
**100% adhesive failure between enamel or hybrid enamel layer and overlying adhesive resin; **100% cohesive failure in enamel;
***Mixed failure with adhesive failure (A) and cohesive failure in enamel (C)
Table 2: Results of Statistical Analysis
ANOVA and Interactions
Sum of Squares
F Value
p Value
1. SE Bond * Single Bond
713.310
713.310
24.876
<.0001
2. Cuspal * Mid-Coronal
568.624
568.624
19.830
<.0001
0.160
1*2
Mean Squares
171.607
171.607
5.985
3085.660
1542.830
53.804
<.0001
1*3
465.327
232.663
8.114
.0005
2*3
1855.050
927.525
32.346
<.0001
637.756
318.873
11.120
<.0001
3. Horizontal * Tangential * Axial
1*2*3
Adhesive system
Enamel region
SE Bond > Single Bond
Cuspal > Mid-coronal
Sectioning direction
Horizontal & Tangential > Axial
> indicates statistically significant difference (Fishers PLSD test, p<0.05)
Single Bond vs SE Bond
SE Bond
Cuspal
Mid-Coronal
Horizontal
Axial
Tangential
NS
NS
NS
NS
NS
Single Bond
Cuspal
Mid-Coronal
Horizontal
Axial
Tangential
NS
NS
NS
NS
S indicates statistical significance. NS indicates no statistical significance (Fishers PLSD test, p<0.05)
Table 3: Statistical Analysis of Single Bond

Enamel Region
Cuspal > Mid-Coronal
23
versus
mid-coronal,
p<0.05). In the case of
directions of sectioning,
only the axial slices
showed
significantly
lower bond strength values (p<0.05).
The enamel bonding of
one-bottle adhesive system (Single Bond) was
significantly influenced
by the direction of sectioning
(One
way
ANOVA,
df=5,
54,
F=2.386, p=9.47 x 10-18).
Even though the mean
shear bond strength
values also tended to be
lower at the axial slice
(Fishers PLSD test,
p<0.05), the effect of the
self-etching primer system (Clearfil SE Bond)
was less influenced by
the direction of sectioning than it was in the
test with Single Bond
(One way ANOVA,
df=5,
54,
F=2.386,
p=0.00076).
Morphological Study Using SEM and

CLSM
The SEM part of this study showed that the

axially sectioned cuspal and mid-coronal
regions and the horizontally sectioned midSingle Bond
Mid-Coronal
Cuspal
coronal regions mostly contained longitudinally cut enamel prisms, whereas the tanT
A
H
T
A
gentially sectioned cuspal and mid-coronal
Cuspal
Horizontal
S
S
S
S
S
regions and the horizontally sectioned cusAxial
S
NS
NS
S
pal region mostly contained crosscut enamTangential
NS
S
S
el prisms. SEM photomicrographs of enamMid-Coronal
Horizontal
S
NS
el surfaces etched with phosphoric acid gel
Axial
S
or the self-etching primer are shown in
*S indicates statistical significance. NS indicates no statistical significance (Fishers PLSD test, p<0.05)
Figure 2. The self-etching primer showed a
less distinct pattern than phosphoric acid
Tables 2, 3 and 4. The ANOVA indicated that there
etchant. SEM and CLSM images of adhesive interfaces
were statistically significant interactions between the
between the enamel and the resin are shown in Figure
adhesive systems and the directions of sectioning
3. When the parallel prismatic zone was etched by
(p=0.0005), as well as the enamel regions and the
phosphoric acid gel, not only the apatite crystals but
directions of sectioning (p<0.0001). However, no signifalso the subsurface enamel prisms separated from the
icant interaction was observed between the adhesive
deeper part of the enamel (Figure 3b). In the case of
systems and the enamel regions (p=0.160). Fishers
the self-etching primer, the formation of etched enamPLSD test indicated that significant differences existel tags was not as evident compared to that of the
ed between the adhesive systems (SE Bond versus
phosphoric acid etching (Figures 3c, d).
Single Bond, p<0.05) and the enamel regions (cuspal
Sectioning Direction
Horizontal & Tangential > Axial
> indicates statistically significant difference (Fishers PLSD test, p<0.05)
24
Operative Dentistry
Table 4: Statistical Analysis of SE Bond
DISCUSSION
Microfractures of longitudinally cut enamel

prisms have been described (Lees & Rollins,
1972; Rasmussen & others, 1976).
> indicates statistically significant difference.
NS indicates no statistically significant differences (Fishers PLSD test, p<0.05).
Munechika & others (1984) reported higher
tensile bond strengths for crosscut etched
enamel than for longitudinally cut enamel.
SE Bond
Mid-Coronal
Cuspal
In this study, similar results were obtained
T
A
H
T
A
with the total-etch wet-bonding treatment
Cuspal
Horizontal
NS
S
S
NS
S
with phosphoric acid. An examination of
Axial
S
NS
NS
S
bond failure at the enamel surface indicated
Tangential
NS
S
NS
that enamel fracture generally occurred,
Mid-Coronal
Horizontal
NS
NS
especially at the parallel prismatic zone,
Axial
S
while adhesive failure between enamel or
*S indicates statistical significance. NS indicates no statistical significance (Fishers PLSD test, p<0.05)
hybrid enamel and overlying adhesive resin
generally occurred at the crosscut prismatic
Figure 2: SEM Images of Conditioned Tooth Surfaces
zone. Even though it has been reported that this mode
of failure is due to the non-uniform stress distribution
generated in the shear test arrangement (Van Noort &
others, 1991; DeHoff, Anusavice & Wang, 1995), the
bonding to the parallel-prismatic zone seems to contribute to the strength of enamel and the inherent
strength of the enamel might be the weak link
(Munechika & others, 1984; Shimada & others, 1999a;
Carvalho & others, 2000). This is because enamel failure did not occur as much in the area of crosscut prismatic zone and because these surfaces produced significantly higher bond strengths (Fishers PLSD, p<0.05).
On the other hand, the fracture toughness of crosscut
enamel prisms seems to be higher than that of the
bonding resin or the bonding between the bonding
Figure 2a: Horizontally sectioned cuspal enamel etched by 35% phosresin and enamel (Munechika & others, 1984; Shimada
phoric acid gel for 15 seconds. Enamel surface is roughened and micro& others, 1999a; Carvalho & others, 2000).
Enamel Region
Cuspal NS Mid-Coronal (p=0.0574)
Sectioning Direction
Horizontally & Tangentially > Axially
irregularity based on apatite crystals is apparent.
Figure 2b: Axially sectioned cuspal enamel etched by 35% phosphoric

acid gel for 15 seconds. Unsupported apatite crystals were observed at
the etched surface (arrow).
Figure 2c: Horizontally sectioned cuspal enamel conditioned by Clearfil

SE Bond primer for 20 seconds. The smear layer was removed, revealing slight porosity at the enamel surface.
25
Figure 3: SEM and CLSM Images of Adhesive Interface
Figure 3a: SEM image of Single Bond to the crosscut prismatic zone.
Bonding resin flows and interlocks to the interprismatic and intercrystalline spaces. The etched zone is approximately 10 m deep.
Figure 3b: CLSM image of Single Bond to the longitudinally sectioned

prismatic zone. Separation of superficial enamel prism was observed
(arrow).
Figure 3c: SEM image of the Clearfil SE Bond to the longitudinally sectioned prismatic zone. Superficial apatite crystals were partially separated (arrow).
Figure 3d: CLSM image of Clearfil SE Bond to the longitudinally sectioned prismatic zone.
Resin composites shrink as they polymerize, and contraction stresses grow within the resin (Davidson, de
Gee & Feilzer, 1984). The lower bond strength of Single
Bond obtained from the parallel prismatic zone may be
attributed to the intrinsic weakness of the anisotropic
substrate. It is highly likely that polymerization shrinkage stress produced during the light curing of adhesive
caused the enamel to crack, especially when an unfilled
adhesive such as Single Bond system was used (Figure
3b). On the contrary, the adhesive of Clearfil SE Bond
is a filled adhesive that might cause less shrinkage
stress. The inclusion of filler particles in the composition of the adhesive may result in a stress-breaking
behavior (Davidson & Abdalla, 1994). A variety of factors
may influence the cuing stress value and the development of stress with time. Further study regarding the
effect of curing stress on enamel is needed (Aarnts,
Akinmade & Feilzer, 1999).
The depth of enamel demineralization caused by
phosphoric acid etching has been reported to be approx-
imately 5 to 50 m and the depth of the porous zone to

be 5 to 20 m (Silverstone & others, 1975; Shinchi,
Soma & Nakabayashi, 2000). In this study, the authors
observed an interprismatic space approximately 10 m
deep caused by phosphoric acid (Figure 3a). The phosphoric acid etching of 10 m depth might also be
aggressive for the parallel-prisms site, to the point
where it could cause the shearing off of the subsurface
enamel prisms or apatite crystals.
In the case of the self-etching primer system, the
effect on enamel bonding was less influenced by prism
orientation and enamel (Table 1). Compared to the
results obtained with the one-bottle adhesive system,
the bond strengths of this system were comparable at
the crosscut enamel surfaces (tangentially sectioned
cuspal and mid-coronal regions and the horizontally
sectioned cuspal region) and significantly higher at the
parallel-prisms site (axially sectioned cuspal and midcoronal regions and the horizontally sectioned mid-coronal regions; Fishers PLSD, p<0.05). The bonding to
26
enamel is reportedly achieved by micromechanical
adhesion resulting from the diffusion of resin
monomers into the pretreated enamel and polymerization, therein, creating a hybrid layer in enamel
(Nakabayashi, 1984; Kanemura, Sano & Tagami, 1999;
Shinchi & others, 2000). The depth of enamel demineralization and the penetration depth of the monomer
seem to be identical, since the processes run parallel to
each other (Watanabe & others, 1994). The self-etching
primer may provide optimal demineralization of enamel
not only for the crosscut prismatic zone but also for the
parallel zone (Watanabe & others, 1994; Kanemura &
others, 1999; Shimada & others, 1999a; Hannig,
Reinhardt & Bott, 1999).
Previous studies of the properties of enamel show different values of stiffness for occlusal, cuspal and side
enamel (Stanford & others, 1960; Ng & others, 1989). In
this study, no significant differences in bond strengths
existed except in the case of the horizontal sections. The
discrepancy observed with the horizontal sections also
seemed to be dependent on the prism orientation, as
horizontal cuspal slices involved crosscut prisms and
resulted in higher bond strengths, while the mid-coronal site created almost parallel prisms that yielded
lower bond strength (Table 1). In particular, phosphoric
acid treatment produced a high bond strength value (51
MPa) at the horizontally sectioned cuspal enamel.
The formation of the Hunter-Schreger bands was
reported to result from a change in the direction of the
enamel prisms between successive groups of prisms
(Hirota, 1982). Close to the central axis of the cuspal
enamel, the enamel prisms were arranged in a tight
spiral, which is oppositely directed with adjacent
groups around the central axis of the tooth (Hirota,
1982). The reason for the high bond strength of horizontally sectioned cuspal enamel is probably the complexity of prism orientation plus the high density of
enamel prisms (Hirota, 1982; Shimada & others, 1999a).
Enamel crack or gap might occur even with an enamel
margin cavity because of the low bond strengths
obtained in the parallel-prismatic zone. If lower bond
strengths develop in the parallel prismatic zone in cavities with high c-factors, then the stress that develops
during polymerization may produce debonding that is
localized to these regions (Brnnstrm & Nyborg, 1973;
Davidson & others, 1984). Marginal integrity of composite resin restorations has been demonstrated to
improve when enamel margins are beveled (Hinoura,
Setcos & Phillips, 1988; Munechika & others, 1984;
Shimada & others, 1999). Thus, enamel margins should
be beveled, especially when phosphoric acid is used as
an enamel conditioner for parallel prismatic walls.
Although it seems that the self-etching primer system
could reduce the crack of enamel margins, the longterm stability of the bonds needs to be evaluated.
Further study is needed, especially under occlusal func-
Operative Dentistry
tion that might fatigue the bonding. It has also been
reported that an intact, unground enamel surface,
where the margin of the restoration is placed, is prismless and hypermineralized, which reduces the effect of
bonding of a self-etching primer system (Kanemura &
others, 1999; Pashley & Tay, 2001). The effect of cavity
depth on a direct restorative system should be studied
with respect to microgap formation, as well.
CONCLUSIONS
Within the limitations of this in vitro study, the following conclusions were drawn:
1.The bond strengths of the two adhesive systems
were influenced by the anisotropic structure of
enamel (p<0.05).
2.When phosphoric acid etching was applied to the
parallel prismatic enamel, the enamel surface
appeared to be over-etched, resulting in lower
bonding.
3.The self-etching primer system produced higher
bond strength to parallel prismatic enamel compared to the one-bottle adhesive system (p<0.05).
Acknowledgements
The authors thank 3M ESPE and Kuraray Co for supplying the
materials for this study. This work was supported by Tokyo
Medical and Dental University.
References
Aarnts MP, Akinmade A & Feilzer AJ (1999) Effects of photoinitiator concentration on conversion, strength and stress development Journal of Dental Research 78 Abstract #1026 p 234.
Brnnstrm M & Nyborg H (1973) Cavity treatment with a
microbicidal fluoride solution: Growth of bacteria and effect on
the pulp Journal of Prosthetic Dentistry 30(3) 303-310.
Carvalho RM, Santiago SL, Fernandes CA, Suh BI & Pashley DH
(2000) Effects of prism orientation on tensile strength of enamel Journal of Adhesive Dentistry 2(4) 251-257.
Davidson CL & Abdalla AI (1994) Effect of occlusal load cycling
on the marginal integrity of adhesive Class V restorations
American Journal of Dentistry 7(2) 111-114.
Davidson CL, de Gee AJ & Feilzer A (1984) The competition
between the composite-dentin bond strength and the polymerization contraction stress Journal of Dental Research 63(12)
1396-1399.
DeHoff PH, Anusavice KJ & Wang Z (1995) Three-dimensional
finite element analysis of the shear bond test Dental Materials
11(2) 126-131.
Hannig M, Reinhardt KJ & Bott B (1999) Self-etching primer vs
phosphoric acid: An alternative concept for composite-to-enamel
bonding Operative Dentistry 24(3) 172-180.
27
Hinoura K, Setcos JC & Phillips RW (1988) Cavity design and

placement techniques for Class 2 composites Operative
Dentistry 13(1) 12-19.
Shimada Y, Kondo Y, Inokoshi Y, Tagami J & Antonucci JM

(1999b) Demineralizing effect of dental cements on human
dentin Quintessence International 30(4) 267-273.
Hirota F (1982) Prism arrangement in human cusp enamel

deduced by X-ray diffraction Archives of Oral Biology 27(11)
931-937.
Shinchi MJ, Soma K & Nakabayashi N (2000) The effect of phosphoric acid concentration on resin tag length and bond
strength of a photo-cured resin to acid-etched enamel Dental
Materials 16(5) 324-329.
Kanemura N, Sano H & Tagami J (1999) Tensile bond strength

to and SEM evaluation of ground and intact enamel surfaces
Journal of Dentistry 27(7) 523-530.
Lees S & Rollins FR Jr (1972) Anisotropy in hard dental tissues
Journal of Biomechanics 5(6) 557-566.
Munechika T, Suzuki K, Nishiyama M, Ohashi M & Horie K
(1984) A comparison of the tensile bond strengths of composite
resins to longitudinal and transverse sections of enamel
prisms in human teeth Journal of Dental Research 63(8) 10791082.
Nakabayashi N (1984) Biocompatibility and promotion of adhesion to tooth substrates CRC Critical Review Biocompatibility
1 25-52.
Ng SY, Payne PA, Cartledge NA & Ferguson MW (1989)
Determination of ultrasonic velocity in human enamel and
dentine Archives of Oral Biology 34(5) 341-345.
Pashley DH & Tay FR (2001) Aggressiveness of contemporary
self-etching adhesives. Part II: Etching effects on unground
enamel Dental Materials 17(5) 430-444.
Rasmussen ST, Patchin RE, Scott DB & Heuer AH (1976)
Fracture properties of human enamel and dentin Journal of
Shimada Y, Antonucci JM, Schumacher GE, McDouough WG &
Tagami J (1999a) Effects of regional tooth structure and sectioning orientation on micro-shear bond strength. Tagami J,
Toledano M & Prati C editors Advanced Adhesive Dentistry 3rd
International Kuraray Symposium Kuraray Co Ltd P 91-103.
Silverstone LM, Saxton CA, Dogon JL & Fejerskov O (1975)

Variation in the pattern of acid-etching of human dental
enamel examined by scanning electron microscopy Caries
Research 9(5) 373-387.
Stanford JW, Weigel KV, Paffenbarger GC & Sweeney WT (1960)
Compressive properties of hard tooth tissues and some
restorative materials Journal of the American Dental
Association 60 746-756.
Tay FR, Gwinnett AJ, Pang KM & Wei SH (1996) Resin permeation into acid-conditioned, moist, and dry dentin: A paradigm
using water-free adhesive primers Journal of Dental Research
75(4) 1034-1044.
Van Noort R, Cardew GE, Howard IC & Noroozi S (1991) The
effect of local interfacial geometry on the measurement of the
tensile bond strength to dentin Journal of Dental Research
70(5) 889-893.
Wang T, Nikaido T & Nakabayashi N (1991) Photocure bonding
agent containing phosphoric methacrylate Dental Materials
7(1) 59-62.
Watanabe I, Nakabayashi N & Pashley DH (1994) Bonding to
ground dentin by a Phenyl-P self-etching primer Journal of
Watson TF (1989) A confocal optical microscope study of the morphology of the tooth/restoration interface using Scotchbond 2
dentin adhesive Journal of Dental Research 68(6) 1124-1131.
The Effect of a
Resin Coating
on the Interfacial Adaptation
of Composite Inlays
PR Jayasooriya PNR Pereira T Nikaido
MF Burrow J Tagami
Clinical Relevance
The application of a resin coating consisting of a dentin bonding system and a low
viscosity microfilled resin was shown to improve the interfacial adaptation of composite
inlays when using a dual-cured resin cement.
SUMMARY
Michael F Burrow, MDS, PhD, associate professor, School of

Dental Science, University of Melbourne, Melbourne, Australia
(Protect Liner F) was applied to the cavities of

half of the prepared teeth, while the remaining
teeth served as non-coated control specimens. All
the teeth were restored with composite inlays
(Estenia) fabricated by the indirect method and
cemented with a dual-cured resin cement
(Panavia F). After finishing the margins with
superfine burs, the bonded inlays were thermocycled between 4C and 55C for 400 cycles.
Specimens were sectioned with a diamond saw
and the tooth-restoration interfaces were
observed with a confocal laser scanning microscope. The data were analyzed with two-way
ANOVA and Fishers PLSD test (p<0.05). The percentage length of gap formation at the dentinrestoration interface of the resin coated teeth
(7.1 3.5) was significantly less than that of the
non-coated teeth (85.7 6.7) (p<0.05). The concept
of coating the prepared cavity with an adhesive
system and a low viscosity microfilled resin
resulted in observing fewer gaps at the internal
dentin-restoration interface compared with the
non-coated specimens.
Junji Tagami, DDS, PhD, professor and chairman, Tokyo Medical

and Dental University, Tokyo, Japan
INTRODUCTION
The relatively low bond strengths of resin

cements to dentin may result in poor interfacial
adaptation of composite inlays. This study determined whether the interfacial adaptation of composite inlays could be improved by applying an
adhesive system and a low viscosity microfilled
resin to the prepared cavity walls before making
an impression. Ten MOD cavities were prepared
on extracted human premolars with gingival
margins located above and below the
cementoenamel junction. A resin coat consisting of a self-etching primer system (Clearfil SE
Bond) and a low viscosity microfilled resin
*Primali R Jayasooriya, BDS, graduate student, Tokyo Medical
and Dental University, Tokyo, Japan
Patricia NR Pereira, DDS, PhD, assistant professor, University
of North Carolina-School of Dentistry, Chapel Hill, NC
Toru Nikaido, DDS, PhD, senior lecturer, Tokyo Medical and
*Reprint request: Cariology and Operative Dentistry,

Department of Restorative Sciences, Graduate School, Tokyo
Medical and Dental University, 1-5-45, Yushima, Bunkyo-ku,
Tokyo 113-8549, Japan; e-mail: [email protected]
Posterior composite restorations have risen in popularity

as a result of the development of improved resin composites, bonding systems and operating techniques. In
29
Jayasooriya & Others: Interfacial Adaptation of Composite Inlays

posterior restorations, direct placement composites are
the preferred treatment over indirect composite
restorations because they require minimal intervention
and cavity preparation (Tyas & others, 2000). However,
a major limitation of direct composites is the inability
to control polymerization shrinkage and depth of cure
(Versluis & others, 1996). This, in turn, may lead to
marginal gap formation and microleakage resulting in
secondary caries (Fontana & others, 1996) and subsequent failure of the restoration. In posterior proximal
restorations with enamel gingival margins, different
incremental filling techniques (Versluis & others, 1996)
and inserts (Ferderlin, Thonemann & Schmalz, 2000)
provide some means to control polymerization shrinkage. However, these methods may be inadequate for
teeth that require larger restorations (Cheung, 1990).
To overcome the polymerization shrinkage of posterior composite restorations, the indirect fabrication of a
composite inlay and cementation with a resin cement
has been advocated (Reeves & others, 1992). However,
current dual-cured resin cements do not bond as
strongly to the tooth as do resin adhesives that are
designed for direct composite restorations (Burrow &
others, 1996). Moreover, the inability of the resin cement
to attain high bond strengths shortly after insertion of
the restoration (Kitasako & others, 1995) and contamination from temporary filling materials, saliva and
blood may lead to reduced bond strengths (Xie, Powers
& McGuckin, 1993; Nikaido & others, 1998; Kaneshima
& others, 2000) and it may adversely affect the longevity of the restorations. A relatively weak bond may lead
to gap formation, producing post-operative sensitivity
(Christensen, 2000) that results in premature failure of
the indirect composite restorations.
By applying a resin coating that consists of a dentin

bonding system and a low viscosity microfilled resin
immediately following cavity preparation prior to making an impression, the prepared tooth surface is sealed
to protect the pulp from mechanical trauma, thermal
stimuli and bacterial invasion during impression making, provisional restoration fabrication and final
cementation (Satoh & others, 1994). Moreover, the
same method has been shown to improve the early
bond strength of resin cement to dentin (De Goes & others, 2000).
This study evaluated gap formation at the internal
toothrestoration interface of MOD cavities bonded
using a preliminary resin-coat by means of a confocal
laser scanning microscope (CLSM) to determine
whether a resin-coating technique could reduce gap
formation at the interface of the indirect composite
inlay (CI). The null hypothesis tested was that applying
a resin coating immediately following cavity preparation does not affect the interfacial adaptation of CI.
The materials used for this study and their restorative
procedures are presented in Table 1. Recently extracted,
non-carious, human maxillary premolars stored in
water at 4C were used. Ten mesio-occluso-distal (MOD)
cavities with slightly rounded internal line angles were
prepared using a superfine bur (SF 145, Shofu Co,
Kyoto, Japan) mounted on a high-speed handpiece
under water coolant (Figure 1). Dimensions of the cavity
preparation were approximately 3 mm wide bucco-lingually, 3 mm deep (occlusal cavity), while the depth of
the proximal boxes depended on the crown length of
each tooth. The mesial and distal margins of each cavity
were located 1 mm above and below the cemento-enamel
Table 1: Material Used and Summary of Restorative Procedure

Materials
Batch #
Composition
Directions
Dentin bonding system

Clearfil SE Bond
003A
Primer: MDP, HEMA, water, photo-initiator

Bond: Multi-functional methacrylates,
microfillers, photo-initiator
Primer: 20 seconds apply, dry

Bond: apply, dry, 20 seconds polymerize
Low viscosity micro-filled resin

Protect Liner F
003
BIS-GMA, TEGMA, micro-fillers, photoinitiator
Mix with opaquer, apply, 20 seconds

polymerize
Indirect resin composite

Estenia
0019B
Hydrphobic methacrylates, 72wt%

micro-fillers, 16wt% superfine fillers
Apply separator on stone cast,

prepare inlay, three minutes polymerize
in alpha-light, 15-minute heat cure
Resin cement
Panavia F
0075B
ED primer: Primer A-MDP, HEMA, 5-NMSA,

chemical initiator
Resin coated teeth+ all inlays: apply

mixture of ED primer+ porcelain activator
60 seconds, dry
Primer B- 5-NMSA, water
Non-coated teeth: apply ED primer

60 seconds, dry
A paste-Quatz glass, micro-filler, MDP,

methacrylate, photoinitiator. B pasteBarium glass, NaF, methacrylates, chemical
initiator
Mix cement pastes, seat inlays, remove

excess, polymerize 60 seconds
All materials were manufactured by Kuraray Co, Tokyo, Japan.

MDP- 10 metacylxydecyle dihydrogen phosphate, HEMA- 2 hydroxy ethyl methacrylate
5-NMSA- N methacryloyl 5 amino salicylic acid, NaF-sodium fluoride, TEGMA- tri ethylene glycol methacrylate
30
Operative Dentistry
Figure 1: Schematic illustration of the specimens.
Figure 2: Acquired tooth-restoration interface (10x). The total length of

gap formation was evaluated on this image.
junction, which means an enamel and dentin margin

was created for each restoration, respectively. The prepared teeth were randomly divided into two groups,
that is, the experimental group (n=5) that received the
resin -coating and the control group (n=5) without the
resin coating.
the inlays, then dried. Equal amounts of two pastes of

the dual-cured resin cement (Panavia F, Kuraray
Medical Co, Tokyo, Japan) were mixed and placed in
the cavities and the inlays were seated and polymerized
for 60 seconds in total, 20 seconds each from occlusal,
buccal and palatal directions. The teeth were placed in
water at 37C for 24 hours, then the margins were finished with superfine burs (V16ff, GC Co, Tokyo, Japan).
The teeth were thermocycled between 4C and 55C for
400 thermal cycles (each cycle containing a 55 seconds
dwell time and five seconds transfer time). The teeth
were kept well hydrated at 37C in a humidifier
throughout the restorative procedure to prevent desiccation.
Immediately following cavity preparation, the teeth

were stabilized by embedding the roots in utility wax.
For the specimens in the resin-coat group, the cavity
surface was prepared using a self-etching primer bonding
system (Clearfil SE Bond, Kuraray Medical, Tokyo,
Japan) following the manufacturers instructions. The
specimens were then coated with a low viscosity microfilled resin (Protect Liner F, Kuraray Medical Co,
Tokyo, Japan) and were polymerized for 20 seconds
(New light-VLII, GC Co, Tokyo, Japan). To distinguish
the two layers of resin coating under the CLSM, a
small amount (5% wt) of opaque resin (Clearfil Opaquer,
Kuraray Medical Co, Tokyo, Japan) was added to the
low viscosity microfilled resin. An agar (Aromaloid, GC
Co, Tokyo, Japan) and or irreversible hydrocolloid
(Aromafine DF II, GC Co, Tokyo, Japan) combination
impression (Takano, Nikaido & Tagami, 2001) was
made of each cavity. The impressions were cast in a
Type III stone (Zo Gypsum, GC Co, Tokyo, Japan). The
next day, the composite inlays (Estenia, Kuraray Medical
Co, Tokyo, Japan) were fabricated on casts according to
the manufacturers instructions. Trial insertion prior to
cementation was performed to ensure a good fit for each
inlay. For the cementation procedure, the enamel and
dentin cementing surfaces of non-coated teeth were
primed with ED primer for 60 seconds and dried. The
cavities for the resin-coated teeth were etched with 37%
phosphoric acid gel (K-etchant, Kuraray Medical Co,
Tokyo, Japan) for 5-to-10 seconds, rinsed and dried in
order to remove debris. A mixture of resin cement priming agent and a silane-coupling agent (Porcelain activator, Kuraray Medical Co, Tokyo, Japan) was applied for
60 seconds to the etched surface and fitting surfaces of
The teeth were sectioned mesio-distally along the long

axis using a slow rotating diamond saw (Isomet,
Buehler Co, Lake Bluff, IL 60044, USA) and polished to
a high gloss with diamond pastes (grit 6, 3, 1, 1/4 m;
DP paste, Struers Co, Denmark). Gaps at the toothrestoration interfaces were observed directly using a
CLSM (Laser Tech, Tokyo, Japan) at 10x magnification
and hard copy images were printed directly from the
CLSM, forming a montage of the cavity-tooth interfacial region (Figure 2). The length of gap formation at
the following three segments of the interface: gingival
enamel and axial dentin on mesial wall (Enamel
margin wall), occlusal dentin (Occlusal wall) and gingival dentin and axial dentin on distal wall (Dentin
margin wall) were measured from the hard copy
images. The percentage of length of gap formation was
calculated by dividing the total length of gap by the
total length of each segment of the interface (Belli &
others, 2001). Areas with gap formation were also
observed under high magnification and recorded. The
data were analyzed by two-way ANOVA at 5% level of
significance. The factors analyzed were region of cavity
floor and with and without resin coating. The percentage of length of gap formation for each region with and
31

without coating was then analyzed by one-way ANOVA
and Fishers PLSD test at the confidence level of 95%.
significantly lower than both the enamel and dentin

margin walls.
The gap width and maximum thickness of the bonding agent, low viscosity microfilled resin and resin
cement were also measured for each segment. The data
were analyzed with one-way ANOVA and Fishers
PLSD test (p<0.05).
CLSM photomicrographs of the interfaces with and

without resin coating are shown in Figures 3 and 4,
respectively. For the conventional method, no gaps
were observed at the enamel margin (Figure 3a), while
gap formation was observed at the dentin margin,
parts of the proximal box surfaces (Figure 3b) and the
whole occlusal wall. For the resin-coated teeth, good
interfacial adaptation with significantly fewer gaps
was observed at both the enamel and dentin margins
(Figures 4a,b). However, gaps were present at the
occluso-mesio and occluso-distal angles, where a thin
layer of SE Bond and Protect Liner F were observed
(Figure 4c).
RESULTS
The percentage of length of gap formation is summarized in Table 2. Two-way ANOVA revealed that the percentages of gap formation were influenced by resin coating (F=31.3; p<0.0001) and the region of the cavity floor
(F=8.9; p<0.05). Extensive gap formation was observed
in the three segments in the non-coated group, while
the gap formation was significantly reduced at each
DISCUSSION
segment in the resin-coated group (p<0.05). The percentage of length of gap formation at the occlusal wall
To date, CLSM has been used to evaluate caries
was significantly higher than the enamel margin wall
(Fontana & others, 1996), collagen shrinkage (Nakaoki
and the dentin margin wall in both the non-coated and
& others, 2000) and toothrestoration interfaces
resin coated groups (p<0.05). However, there was no dif(Watson & Wilmot, 1992; Griffiths, Watson & Sherriff,
ference in the gap formation
between the enamel and dentin
Table 2: The Length of Gap Formation (%) of Composite InlayMOD Cavity Interface
margin walls (p>0.05). The maxiEnamel Margin Wall
Occlusal Wall
Dentin Margin Walls
mum gap width ranged from 10With Resin
4.3 (4.0)
13.3 (7.4)
4.5 (4.1)
12 m for both groups, and no
Coating
statistically significant differ100.0 (0.0)
77.8 (10.1)
Without
73.9 (8.3)
ences in the gap width between
Resin
groups or sites (p>0.05) were
Coating
observed (Table 3). The minimum
*n=10, Mean (SD)
gap width was 7 m.
The same superscript letters among figures represent no statistically significant difference (p>0.0001)
Maximum thickness of the
bonding agent, low viscosity
microfilled resin, the resin
cement for the three segments
and the statistical outcomes are
shown in Table 4. The maximum
thickness of resin cement for both
methods ranged from 150 to 300
m. Moreover, the thickness of
resin cement at the dentin margin wall was significantly less
than both the enamel margin
walls and the occlusal walls in
both groups. In the resin-coated
group, the maximum thickness
of Clearfil SE Bond and Protect
Liner F were 145 and 93 m at
the enamel margin walls, 62 and
58 m at the occlusal walls and
85 and 152 m at the dentin
margin walls, respectively. The
maximum thickness of both
Clearfil SE Bond and Protect
Liner F at the occlusal wall was
Table 3: The Maximum Gap Width for Composite Inlay MOD Cavity Interface (m)
Enamel Margin Wall
Occlusal Wall
With Resin
Coating
10.5 (2.5)a
11.5 (2.3)a
Dentin Margin Wall

11.0 (2.4)a
Without-Resin
Coating
10.0 (2.4)a
11.6 (2.4)a
11.5 (2.3)a
N=10, mean (SD)

The same superscript letter among figures represents no statistically significant difference (p>0.05).
Table 4: The Maximum Thickness of the Materials Used in the Resin Coating and Resin
Cement
Enamel Margin Wall
Occlusal Wall
Dentin Margin Wall
With Resin
Coating
1. SE Bond
2. Protect Liner F
3. Panavia F
145.0 (85.4)a
62.7 (60.3)b
134.7 (67.3)a
93.3 (66.5)
58.3 (54.8)
85.5 (41.1)c
241.1 (65.1)
243.4 (87.2)
152.9 (50.7)e
96.8 (52.3)f
266.9 (81.0)f
194.4 (51.5)g
Without Resin
Coating
1. Panavia F
n=10, mean (SD)

The same superscript letters among materials within each group represent no statistically significant difference (p>0.05.)
32
Operative Dentistry
Figure 3: Photomicrographs to show the tooth-restoration interface of a non-coated tooth (50x).
Figure 3a: Enamel-restoration interface. No gap formation is visible

between enamel (E) and resin cement (C).
Figure 3b: Dentin-restoration interface. Note the presence of gap (arrow)

between dentin (D) and resin cement (C).
Figure 4: Photomicrograph to show the tooth-restoration interface of a resin-coated tooth (50x).
Figure 4a: Enamel-restoration interface. Note the absence of gaps

between enamel (E), resin coating (R) and resin cement (C) at the interfaces.
Figure 4b: Dentin-restoration interface. Note the absence of gaps

between dentin (D), resin coating (R) and resin cement (C) at the interfaces.
1999). Though studies have documented the use of

CLSM simultaneously with the fluorescent-labeled
dyes, only limited information is available in the literature regarding the use of CLSM to identify gaps without fluorescent-labeled dyes (Belli & others, 2001). The
main advantage of CLSM is that it requires simple
preparation of specimens without subjecting them to
desiccation and high vacuum. Therefore, the destruction and artifacts that occur as a result of these procedures are reduced. In addition, the specimens can be
observed under ambient conditions.
Figure 4c: Disto-occlusal angle. Also, note the presence of gap (star) at
disto-occluso proximal line angle and the thin resin coating.
A resin-coating technique has recently been developed in which both a hybrid layer and a tight sealing
film are produced on the dentin surface with a dentin
adhesive system and a low-viscosity microfilled resin
(Nikaido & others, 1992; Otsuki, Yamada & Inokoshi,
1993). It enables coverage and protection of the pre-

pared dentin immediately after cavity preparation to
provide high bond strength of resin cement (De Goes &
others, 2000) and good adaptation of composite inlays
(Otsuki & others, 1993). Therefore, this technique has
the potential to minimize pulp irritation and post-operative sensitivity (Satoh & others, 1994). The selection of
appropriate materials for impressions and provisional
restorations after using the resin coating technique has
been reported and is important for the success of the
final restoration (Takano, Nikaido & Tagami, 2001;
Takada & others, 1995).
In this study, the two-step self-etching primer system,
Clearfil SE Bond, and the low-viscosity microfilled
resin, Protect Liner F, were used for the resin coating.
An acidic monomer in the primer, such as MDP in
Clearfil SE Bond, dissolves the smear layer and demineralizes the underlying dentin, resulting in mild surface etching. Good bonding performance with a selfetching primer system for a direct resin composite has
already been shown in a previous laboratory study
(Nikaido & others, 2002). The additional application of
the low-viscosity microfilled resin can protect and promote polymerization of the underlying adhesive, resulting in an increase in bond strength (Jayasooriya & others, 2001).
Significantly more gaps in the resin-dentin interface
were observed without resin coating than with resin
coating for each site (Table 2). The early bond strengths
of resin cement, Panavia F, to dentin without resin
coating was reported to be significantly lower than with
resin coating, using the combination of Clearfil Liner
Bond 2V and Protect Liner F (De Goes & others, 2000).
Moreover, considering the shape of the MOD cavity, a
relatively high configuration factor (Feilzer, de Gee &
Davidson, 1989) may contribute to high polymerization
stress during cementation with the conventional
method, while resin coating may act as a stress
absorber (Kemp-Scholte & Davidson, 1990) and reduce
polymerization shrinkage stress in the resin coated
group. These factors may explain the significant differences in gap formation with and without resin coating.
A previous study (Dietschi & Herzfeld, 1998) reported
extensive gap formation, similar to the current study, at
the resin-dentin interface of composite inlays with the
conventional method. However, relatively few gaps
were observed at the enamel margins in both groups
(Figures 3a and 4a). Good marginal adaptation at
enamel was also observed by Reeves & others (1992)
and is supported by Kitasako & others (1995), who
reported high bond strengths of resin cement to enamel.
The site within the cavity also influenced the gap formation. For occlusal walls, significantly greater gap formation was observed compared to the other segments
in both the coated and non-coated groups. For the coated group, the maximum thickness of the adhesive and
the low-viscosity microfilled resin at the occlusal site
33
was significantly thinner than that at the enamel and

dentin margin walls (Table 3). The relatively thin layers
were created due to the gentle air blowing of the bonding agent prior to polymerization. The thinnest layer of
the bonding agent was observed at the mesial and distal occluso-proximal line angles, where gap formation
was often observed (Figure 4c). Zheng & others (2001)
reported that reducing the thickness of the adhesive
layer decreased the bond strength to dentin when the
self-etching primer system, Clearfil Liner Bond 2 V,
was used. Therefore, care should be taken not to create
very thin layers of bonding agent so as to provide good
interfacial adaptation. The maximum thickness of the
resin cement at different walls ranged from 150-to-300
m. Sorensen & Munksgaard (1995) reported cement
film thickness in the range of 50-to-300 m for ceramic
inlays. Therefore, the thickness of the resin cement
observed in this study is consistent with other studies.
However, the thickness of the resin cement along the
dentin margin wall was significantly less than both the
enamel and occlusal walls in both groups. The reason is
unclear; however, the dentin margin wall was longer
than other walls and tended to taper out to the margin
(Figure 2), hence, providing a slip-joint effect that
allowed for better flow of the cement and, thus, better
seating.
Considering the shape of the MOD inlay, air entrapment might occur at the occlusal wall during cementation. The presence of air at the interface may inhibit
polymerization of the resin cement, which could cause
gap formation especially in the non-coated group.
Furthermore, the presence of air by itself may also
appear as a gap. However, it was not possible to differentiate air entrapment from gap formation that occurs
as a result of polymerization shrinkage when using
CLSM and it can be considered a limitation of this
study. Nevertheless, any space formed can be deleterious, as it permits the restoration to deform under the
occlusal load and may result in premature failure of the
composite inlays.
In addition, regional variability of dentin (Pereira &
others, 1999; Bouillaguet & others, 2001), the depth of
the cavity (Burrow & others, 1994), the burs selected
for cavity preparation (Ogata & others, 2001) and the
distance from a curing unit (Hansen & Asmussen,
1997) may also contribute to the differences observed.
Applying a resin coating on the prepared cavities
was shown to reduce the number of gaps at the interface. However, it was not possible to completely seal all
internal cavity surfaces of the restoration. Therefore,
further improvement of techniques and materials is
necessary to achieve gap-free interfacial adaptation of a
restoration. Moreover, the oral environment may also
adversely influence the bonding performance of the
materials used due to contamination from blood and
saliva (Xie & others, 1993). Therefore, in a clinical situ-
34
Operative Dentistry
ation, a rubber dam should be used to isolate the prepared tooth and prevent contamination during the
restorative procedure. However, as performance of the
material in a clinical situation was not considered,
there is a need for a clinical trail that involves the resin
coating technique to further confirm the validity of this
procedure.
CONCLUSIONS
Within the limitations of this study, the following conclusions were drawn:
1. Using a resin coating reduced the percentage
length of gap formation between the interface of
composite inlays and the preparation cavity surface.
2. Site within the cavity was shown to influence
gap formation at the interface of the composite
inlays, as the highest percentage of gap formation was observed at the occlusal walls in both
groups.
3. A confocal laser scanning microscope is useful in
observing gap formation at internal cavity surfaces of a restoration.
References
Belli S, Inokoshi S, Ozer F, Pereira PN, Ogata M & Tagami J
(2001) The effect of additional enamel etching and a flowable
composite to the interfacial integrity of Class II adhesive composite restorations Operative Dentistry 26(1) 70-75.
Bouillaguet S, Ciucchi B, Jacoby T, Wataha JC & Pashley D
(2001) Bonding characteristics to dentin walls of Class II cavities, in vitro Dental Materials 17(4) 316-321.
Burrow MF, Nikaido T, Satoh M & Tagami J (1996) Early bonding of resin cements to dentinEffect of bonding environment
Burrow MF, Takakura H, Nakajima M, Inai N, Tagami J &
Takatsu T (1994) The influence of age and depth of dentin on
bonding Dental Materials 10(4) 241-246.
Cheung GS (1990) Reducing marginal leakage of posterior composite resin restorations: A review of clinical techniques
Christensen GJ (2000) Resin cements and post-operative sensitivity Journal of the American Dental Association 131(8) 11971199.
De Goes MF, Nikaido T, Pereira PNR & Tagami J (2000) Early
bond strengths of dual-cured resin cement to resin-coated
dentin Journal of Dental Research 79(Special Issue) Abstract
#2477 p 453.
Dietschi D & Herzfeld D (1998) In vitro evaluation of marginal
and internal adaptation Class II resin composite restorations
after thermal and occlusal stressing European Journal of Oral
Science 106(6) 1033-1042.
Federlin M, Thonemann B & Schmalz G (2000)

Insertsmegafillers in composite restorations: A literature
review Clinical Oral Investigations 4(1) 1-8.
Feilzer AJ, de Gee AJ & Davidson CL (1989) Increased wall to
wall curing contraction in thin bonded resin layers Journal of
Fontana M, Dunipace AJ, Gregory RL, Noblitt TW, Li Y, Park
KK & Stookey GK (1996) An in vitro microbiological model for
studying secondary caries formation Caries Research 30(2)
112-118.
Griffiths BM, Watson TF & Sherriff M (1999) The influence of
dentine bonding systems and their handling characteristics on
the morphology and micropermeability of the dentine adhesive
interface Journal of Dentistry 27(1) 63-71.
Hansen EK & Asmussen E (1997) Visible-light curing units:
Correlation between depth of cure and distance between exit
window and resin surface Acta Odontologica Scandinavia
55(3) 162-166.
Jayasooriya PR, Pereira PNR, Nikaido T & Tagami J (2001)
Micro-tensile bond strengths of resin cement to dentin using
resin-coating technique Journal of Dental Research
80(Special Issue) Abstract #1818 p 754.
Kaneshima T, Yatani H, Kasai T, Watanabe EK & Yamashita A
(2000) The influence of blood contamination on bond strengths
between dentin and an adhesive resin cement Operative
Dentistry 25(3) 195-201.
Kemp-Scholte CM & Davidson CL (1990) Marginal integrity
related to bond strength and strain capacity of composite resin
restorative systems Journal of Prosthetic Dentistry 64(6) 658664.
Kitasako Y, Burrow MF, Nikaido T, Harada N, Inokoshi S,
Yamada T & Takatsu T (1995) Shear and tensile bond testing
for resin cement evaluation Dental Materials 11(5) 298-304.
Nakaoki Y, Nikaido T, Pereira PN, Inokoshi S & Tagami J (2000)
Dimensional changes of demineralized dentin treated with
HEMA primers Dental Materials 16(6) 441-446.
Nikaido T, Takada T, Burrow MF & Tagami (1992) The early
bond strength of dual cured resin cement to enamel and dentin
Journal of Japanese Dental Materials 11 910-915.
Nikaido T, Takada T, Sasafuchi Y, Takano Y, Satoh M & Tagami
J (1998) Clinical factors influencing dentin bonding
Proceedings: Modern Trends in Adhesive Dentistry 59-67.
Nikaido T, Kunzelmann KH, Chen H, Ogata M, Harada N,
Yamaguchi S, Cox CF, Hickel R & Tagami J (2002) Evaluation
of thermal cycling and mechanical loading on bond strength of
a self-etching primer system to dentin Dental Materials 18(3)
269-275.
Ogata M, Harada N, Yamaguchi S, Nakajima M & Tagami J
(2001) Effect of different burs on dentin bond strengths of selfetching primer bonding systems Operative Dentistry 26(4)
375-382.
Otsuki M, Yamada T & Inokoshi S (1993) Establishment of a
composite resin inlay technique. Part 7. Use of low viscous
resin Japanese Journal of Conservative Dentistry 36 13241330.
Pereira PN, Okuda M, Sano H, Yoshikawa T, Burrow MF &
Tagami J (1999) Effect of intrinsic wetness and regional difference on dentin bond strength Dental Materials 15(1) 46-53.
35
Reeves GW, Lentz DL, OHara JW, McDaniel MD & Tobert WE

(1992) Comparison of marginal adaptation between direct and
indirect composites Operative Dentistry 17(6) 210-214.
Tyas MJ, Anusavice KJ, Frencken JE & Mount GJ (2000)

Minimal intervention dentistrya review FDI commission
project 1-97 International Dental Journal 50(1) 1-12 Review.
Satoh M, Inai N, Nikaido T, Tagami J, Inokoshi S, Yamada T &

Takatsu T (1994) How to use Liner Bond System as a dentin
and pulp protector in indirect restorations Japanese Journal of
Adhesive Dentistry 12(1) 41-47.
Versluis A, Douglas WH, Cross M & Sakaguchi RL (1996) Does

an incremental filling technique reduce the polymerization
shrinkage stresses? Journal of Dental Research 75(3) 871-878.
Sorensen JA & Munksgaard EC (1995) Interfacial gaps of resin

cemented ceramic inlays European Journal of Oral Science
103(2(Part I)) 116-120.
Takada T, Nikaido T, Koh Y, Inokoshi S & Takatsu T (1995)
Effect of temporary filling materials on adhesion of resin
cement to dentin Japanese Journal of Conservative Dentistry
38 422-427.
Takano Y, Nikaido T & Tagami J (2001) Visual and SEM observation of resin coated dentine after taking impression
Japanese Journal of Adhesive Dentistry 19(2) 117-124.
Watson TF & Wilmot DM (1992) A confocal microscopic evaluation of the interface between Syntac adhesive and tooth tissue
Xie J, Powers JM & Mc Guckin RS (1993) In vitro bond strength
of two adhesives to enamel and dentin under normal and contaminated conditions Dental Materials 9(5) 295-299.
Zheng L, Pereira PN, Nakajima M, Sano H & Tagami J (2001)
Relationship between adhesive thickness and microtensile
bond strength Operative Dentistry 26(1) 97-104.
Effects of
Finishing/Polishing Techniques
on Microleakage of
Resin-Modified
Glass Ionomer Cement Restorations
AUJ Yap WY Yap EJC Yeo
JWS Tan DSB Ong
Clinical Relevance
The effect of finishing/polishing techniques on microleakage of resin-modified glass

ionomer cements are tissue and material dependent. Wet finishing/polishing techniques
that employ the use of one or two-step rubber abrasives at speeds between 10,000 and
12,000 rpm generally resulted in less leakage.
SUMMARY
This study investigated the effect of finishing/polishing techniques on the microleakage of resinmodified glass ionomer restorations. Class V
preparations were made on the buccal and lingual/palatal surfaces of freshly extracted teeth.
The cavities on each tooth were restored with
Fuji II LC (FT [GC]) and Photac-Fil Quick (PF
[3M-ESPE]) according to manufacturers instructions. Immediately after light-polymerization,
gross finishing was done with eight-fluted tungsten carbide burs. The teeth were then randomly
divided into four groups and finishing/polishing
*Adrian UJ Yap, BDS, MSc, PhD, FAMS, FADM, FRSH, associate professor, Department of Restorative Dentistry, Faculty of
WY Yap, student, Faculty of Dentistry, National University of
Singapore
Egwin JC Yeo, student, Faculty of Dentistry, National University
of Singapore
Jane WS Tan, student, Faculty of Dentistry, National University
of Singapore
Debbie SB Ong, student, Faculty of Dentistry, National
*Reprint request: 5 Lower Kent Ridge Road, Singapore 119074,
Republic of Singapore; e-mail: [email protected]
was done with one of the following systems: (a)

Robot Carbides (RC); (b) Super-Snap system (SS);
(c) OneGloss (OG) and (d) CompoSite Polishers
(CS). The sample size for each material-finishing/polishing system combination was eight. After
finishing/polishing, the teeth were stored in distilled water at 37C for one week. The root apices
were then sealed with acrylic and two coats of
varnish was applied 1 mm beyond the restoration
margins. The teeth were subsequently subjected
to dye penetration testing (0.5% basic fuchsin),
sectioned and scored. Data was analyzed using
Kruskal-Wallis and Mann-Whitney U tests at a significance level of 0.05. Results of statistical analysis were as follows: Enamel margins: PFOG<SS;
FT-OG<RC; Dentin margins: PFno significant difference; FTOG & CS<RC. Regardless of the finishing/polishing technique, leakage at dentin
margins was significantly greater than at enamel
margins for FT. For PF, no significant difference
in leakage scores was observed between dentin
and enamel with the exception of finishing/polishing with OG. FT restorations had significantly
less enamel and dentin leakage than PF restorations when treated with OG. The effect of finishing/polishing techniques on microleakage was
both tissue and material dependent.
Yap & Others: Effects of Finishing on Microleakage of Resin-Modified Glass Ionomer Cements
INTRODUCTION
Glass ionomer cements are comprised of a basic glass
and an acidic polymer that are set by an acid-base reaction when mixed. These cements are popular as restorative materials due to their numerous desirable properties, including fluoride release, adhesion to dentin and
enamel, similar thermal expansion to dentin and low
solubility in oral fluids when set. Resin-modified glass
ionomer cements (RMGICs) were introduced to help
overcome the problems of moisture sensitivity and low
early mechanical strength associated with conventional
glass ionomer cements, while maintaining their clinical
advantages (Sidhu & Watson, 1995). Setting characteristics are also improved and finishing/polishing of resinmodified glass ionomer restorations can be carried out
almost immediately after photo curing (Mount, 1993).
Microleakage can be defined as the passage of bacteria, fluids, molecules or ions between a cavity wall and
37
the restorative material (Kidd, 1976). Clinically, it may

lead to staining, postoperative sensitivity and/or recurrent caries. RMGICs, as with all other restorative materials, exhibit some degree of microleakage (Yap, Tan &
Teh, 2000; de Magalhaes, Serra & Rodrigues, 1999; Yap,
Lim & Neo, 1995). Variations in finishing/polishing
techniques have been shown to affect the ability of composite restorations to resist leakage (Owens, Halter &
Brown, 1998; Brackett, Gilpatrick & Gunnin, 1997;
Barkmeier & Cooley, 1992; Yu & others, 1990). The
effect of finishing/polishing techniques on microleakage
of RMGICs has, however, not been widely investigated
(Yap & others, 2000; Wilder & others, 2000). In addition,
most manufacturers have been ambiguous regarding
their recommendations for finishing/polishing of their
RMGICs. Vague statements like finish under water
spray using standard techniques are frequently used.
This study evaluated the effect of different finishing/polishing systems on the microleakage of RMGICs. The relative ability of the RMGICs to seal enamel and dentin
margins after the various finishing/polishing procedures was also compared.
The resin-modified glass ionomer cements investigated

and their technical profiles are shown in Table 1. Thirtytwo freshly extracted, non-carious premolars were
selected for this study. The teeth were disinfected with
2% formaline-saline, cleaned and stored in distilled
water at 4C until use. Wedge-shaped Class V preparations were made on the buccal and lingual/palatal surfaces of each tooth. The cavity dimensions were approximately 4 mm mesio-distally and 3 mm occluso-gingivally. An internal line angle of 90 was maintained such
that length of both the occlusal and gingival walls was
about 2 mm (Figure 1). The occlusal cavosurface margin
was located in enamel, while the gingival cavosurface
Figure 1: Grading scale used to evaluate the degree of dye penetration
margin was located in dentin. The cavities on each tooth
at the tooth-restoration interface.
were restored with capsulated Fuji II LC (GC) and
0 : no evidence of dye penetration
Photac-Fil Quick (3M-ESPE, St Paul, MN 55144, USA).
1 : dye penetration up to one-third cavity wall
Cavities that were to be restored with Fuji II LC were
2 : dye penetration more than one-third but less than two-thirds cavity
wall
first treated with Cavity Conditioner (GC) for 10 sec3 : dye penetration more than two-thirds to full cavity wall
onds, while cavities to be restored with Photac-Fil were
treated
with
Ketac Conditioner
Table 1: Technical Profiles of the Resin-Modified Glass Ionomer Cements Evaluated
(3M-ESPE) for 10
Material
Manufacturer
Powder
Liquid
Filler Particle Size (m)
seconds. The cavities were then
Fuji II LC
GC Corp,
Aluminosilicate
Polyacrylic acid, distilled
4.5
Tokyo, Japan
glass, pigments
water, HEMA (17%).
washed for 30
Lot #9912202
dimethacrylate monomer,
seconds and gencamphoroquinone
tly air dried. The
Photac-Fil
ESPE Dental AG,
Calcium aluminium
HEMA (40%),
7.0
resin-modified
Quick
Seefeld, Germany
fluorosilicate glass,
difunctional monomers,
glass
ionomer
Lot #0065231
freeze-dried
water, camphoroquinone
cements
were
copolymers of
polyacrylic and maleic
mixed according
acids, tartaric acid,
to manufacturactivator, pigments
38
Operative Dentistry
(SS); (c) Onegloss (OG) and (d)

Product
Finishing/
Usage
Handpiece Speed
Manufacturer
CompoSite
Polishing System
Polishers (CS).
Robot Carbide
Details of the finSH134F
Graded carbide
Wet, 12 strokes
300,000 rpm
Shofu Inc,
ishing/polishing
SH134UF
Wet, 12 strokes
300,000 rpm
Kyoto, Japan
sequences
are
Super-Snap
reflected in Table
Coarse
Graded abrasive
Dry, 6 strokes
12,000 rpm
Shofu Inc,
2. The restored
Medium
discs
Dry, 6 strokes
12,000 rpm
Kyoto, Japan
Fine
Dry, 6 strokes
12,000 rpm
teeth were then
Extra fine
Dry, 6 strokes
12,000 rpm
stored in distilled
OneGloss
water at 37C for
One-step rubber
Wet,12 heavy strokes
10,000 rpm
Shofu Inc,
one week. After
abrasive
Wet,12 heavy strokes
10,000 rpm
Kyoto, Japan
the
one-week
CompoSite
storage period,
Polishers
the apical third of
CompoSite
Two-step rubber
Wet, 12 strokes
12,000 rpm
Shofu Inc,
the roots of each
CompoSite Fine
abrasive
Dry, 12 strokes
12,000 rpm
Kyoto, Japan
tooth were sealed
with acrylic resin.
Two layers of nail varnish were painted, staying
Table 3: Distribution of Dye Penetration Scores for Photac-Fil Quick
clear of the restoration margins by 1 mm. The teeth
Enamel
Dentin
were then soaked in 0.5% basic fuchsin dye at 37C
Finishing/Polishing System
0 1 2 3
0 1 2 3
for 24 hours. The nail varnish and excess dye were
Shofu Robot
2 0 0 6
0 0 0 8
removed after dye penetration testing. The teeth
Super-Snap
0 0 0 8
0 0 0 8
were then sectioned bucco-lingually/palatally with a
Microslice 2 precision saw (Cambridge Instrument
OneGloss
4 0 0 4
0 0 0 8
Ltd, Cambridge, England). The degree of dye peneCompoSite
3 0 0 5
1 0 0 7
tration was then graded at 10x magnification with a
stereomicroscope (Nikon SE, Nikon, Tokyo, Japan)
Table 4: Distribution of Dye Penetration Scores for Fuji II LC
using the following scale (Figure 1):
Table 2: Finishing/Polishing Systems and Sequences
Enamel
Dentin
0: No evidence of dye penetration.
Finishing/Polishing System
0 1 2 3
0 1 2 3
1: Dye penetration up to one-third cavity wall.
Shofu Robot
4 0 0 4
0 0 0 8
Super-Snap
6 0 1 1
0 0 3 5
OneGloss
8 0 0 0
1 0 3 4
CompoSite
5 1 2 0
1 0 3 4
2: Dye penetration more than one-third but less

than two-thirds cavity wall.
3: Dye penetration more than two-thirds to full
cavity wall.
ers instructions and injected into the cavities.

Transparent preformed cervical matrixes (Hawe-Neos
Dental, Bioggio, Switzerland) were placed over the filled
cavities and pressure was applied to extrude excess
material. The cements were then light polymerized for
20 seconds using a curing light (Spectrum; Dentsply Inc,
Milford, DE 19963, USA) with an output intensity 420
mW/cm2, as assessed with a curing radiometer (Cure
Rite, EFOS Inc, Ontario, Canada).
Data was subjected to non-parametric statistical

analysis at a significance level of 0.05.
Immediately after light-polymerization, the cervical

matrixes were removed and gross finishing was done
with 8-flute tungsten carbide burs (Robot Carbide
SH134, Shofu, Kyoto, Japan). Gross finishing was performed in one direction under water spray using a highspeed handpiece at 300,000 rpm. The burs were
replaced after gross finishing of every eight restorations.
The restored teeth were then randomly divided into four
groups and finished/polished with the following systems: (a) Robot Carbides (RC); (b) Super-Snap system
Tables 3 and Table 4, respectively, show the distribution

of dye penetration scores for Photac-Fil Quick and Fuji
II LC. Results of statistical analyses are shown in
Tables 5 to 7. The effect of the finishing/polishing system on microleakage was both material and tissue
dependent. At the enamel margins of Photac-Fil
restorations, finishing/polishing with OneGloss resulted in significantly less leakage than with Super-Snap.
No significant difference in dentin leakage was
observed among the various finishing/polishing sys-
The Kruskal-Wallis test was used to determine significant differences between finishing/polishing techniques, and the Mann-Whitney U-test test was used to
evaluate inter-technique, material and tissue differences.
RESULTS
Table 5: Comparison of Microleakage Scores Among the Different
Finishing/Polishing Systems
Tissue
Photcal-Fil Quick
Enamel
OneGloss < Super-Snap
Fuji II LC
OneGloss < Robot Carbide
Dentin
NS
OneGloss, CompoSite <

Robot Carbide
< indicates statistical significance and NS indicates results no statistical significance (Results of KruskallWallis and Mann-Whitney U tests at p<0.05)
Table 6: Comparison of Leakage Score Among Materials

Finishing/Polishing
System
Enamel
Robot Carbide
NS
Dentin
39
imparted to resin-modified glass ionomer

cements occurs when they are allowed to set
against a cellulose strip or matrix (Hoelscher &
others, 1998; Yap, Lye & Sau, 1997). However,
despite the careful placement of matrixes,
removal of excess material or re-contouring of
restorations is often necessary clinically. The
finishing/polishing systems evaluated included
carbide burs (Robot Carbide), graded abrasive
disks (Super-Snap), and one-step (OneGloss) and
two-step (CompoSite Polishers) rubber abrasives.
The effect of finishing/polishing systems on

microleakage was both material and tissue
Super-Snap
Photac-Fil > Fuji II LC
NS
dependent. At the enamel margins of PhotacFil restorations, finishing/polishing with
OneGloss
Super-Snap resulted in grade 3 leakage scores
CompoSite
NS
NS
for all specimens. Super-Snap, being a totally
> indicates statistical significance and NS indicates results no statistical significance (Results of MannWhitney U tests at p<0.05)
dry finishing/polishing system, could generate
substantial heat. In view of the significant difference in thermal coefficients of expansion
Table 7: Comparison of Leakage Score Between Enamel and Dentin
between RMGICs and teeth (Sidhu & Watson,
Margins
1995), finishing/polishing with Super-Snap
Finishing/Polishing
Phototac-Fil Quick
Fuji II LC
could result in stress at the restoration-tooth
System
interface, creating microgaps that allow
Robot Carbide
NS
Dentin > Enamel
microleakage to occur. In addition, contraction
Super-Snap
NS
Dentin > Enamel
under desiccating conditions could also lead to
OneGloss
Dentin > Enamel
Dentin > Enamel
increased microleakage (Bouschlicher, Vargas
CompoSite
NS
Dentin > Enamel
& Denehy, 1996; Wilson & Paddon, 1993). The
> indicates statistical significance and NS indicates results no statistical significance (Results of Mannlack of statistical significance in dentin leakage
Whitney U tests at p<0.05)
between finishing/polishing systems is not due
to the resistance of Photac-Fil to finishing/poltems. At the enamel margins of Fuji II LC restorations,
ishing procedures but to the generally poor dentin seal
finishing/polishing with OneGloss resulted in signifiof Photac-Fil. Grade 3 leakage was observed for almost
cantly less leakage than with Robot Carbide.
all specimens regardless of the finishing/polishing sysFinishing/polishing with OneGloss and CompoSite
tem employed. Although the enamel seal of Fuji II LC
resulted in significantly less dentin leakage than with
restorations was not susceptible to the effect of dry finRobot Carbide (Table 5). Fuji II LC restorations had
ishing/polishing (Super-Snap), it was susceptible to the
significantly less enamel and dentin leakage than
detrimental effect of ultra high-speed finishing/polishPhotac-Fil restorations after finishing/polishing with
ing. Finishing/polishing with Robot Carbide (300,000
OneGloss. Fuji II LC restorations also had significantrpm) resulted in significantly greater enamel leakage
ly less enamel leakage than Photac-Fil restorations
than treatment with OneGloss. A similar trend was
when treated with Super-Snap. Regardless of the finobserved at the dentin margins. Finishing/polishing
ishing/polishing technique, leakage at dentin margins
with OneGloss and CompoSite resulted in significantly
was significantly greater than at enamel margins for
less leakage than with Robot Carbide (Table 5). The
Fuji II LC. For Photac-Fil, no significant difference in
mechanical stresses generated with ultra high-speed
leakage scores was observed between dentin and enamfinishing/polishing could disrupt the bond of Fuji II LC
el with the exception of finishing/polishing with
to enamel and dentin. Results suggest that wet finishOneGloss.
ing/polishing techniques that employ the use of one or
two step rubber abrasives at speeds between 10,000
DISCUSSION
and 12,000 rpm generally resulted in less leakage. The
Finishing refers to the gross contouring of restorations
fact that leakage was present even after one weeks
to obtain the desired anatomy. Polishing refers to the
storage in water showed that the effect of finishing/polreduction of roughness and scratches created by finishishing techniques on microleakage cannot be compening instruments. The two procedures are, however,
sated by water sorption in the short-term
interdependent and cannot be clearly delineated from
(Kanchananvasita, Anstice & Pearson, 1997; Yap,
each other and, hence, use of the term finishing/polish1996).
ing. Research has shown that the smoothest surface
NS
40
Operative Dentistry
Compared to conventional materials, RMGICs may

offer a better seal to enamel/dentin due to their reduced
water content, immediate adhesion and superior wetting ability resulting from the use of HEMA (hydroxyethyl methacrylate) (Yap & others, 2000; Martin &
ORourke, 1993). Results of microleakage tests have,
however, shown that not all RMGICs display significantly less leakage than their conventional counterparts (Hallett & Garca-Godoy, 1993). This has been
attributed to thermal expansion mismatches with tooth
substance and polymerization shrinkage arising from
the inclusion of resin (Sidhu & Watson, 1995). Yap &
others (2000) reported that the sealing ability of conventional and resin-modified glass ionomer cements to
enamel was significantly better than dentin regardless
of the type of finishing/polishing employed. As glass
ionomer cements bond by polar and ionic attraction, the
lower microleakge observed in enamel can be attributed
to the greater inorganic content and homogeneity of
enamel (Anusavice, 1996). The aforementioned observations might, however, be material specific. In the current study, Fuji II LC displayed significantly less leakage in enamel regardless of the finishing/polishing systems employed. This corroborated the findings of Yap &
others (2000), who also assessed Fuji II LC. On the
other hand, no significant difference in enamel and
dentin leakage was observed for Photac-Fil, with the
exception of treatment with OneGloss. The enamel seal
of Photac-Fil, therefore, appears to be compromised by
immediate treatment with most finishing/polishing systems. One possible hypothesis for this observation is
the use of freeze-dried copolymers of polyacrylic acid in
Photac-Fil. As the anhydrous polyacrylic acid needs to
be hydrated and activated prior to reaction with the
aluminosilicate glass, the ionic exchange mechanism
may take more time to become established. Immediate
finishing/polishing of Photac-Fil might disrupt the
weak ionic bonds formed early in the setting reaction.
This also explains the significantly greater enamel and
dentin leakage observed with Photac-Fil as compared
to Fuji II LC for some of the finishing/polishing systems
evaluated.
CONCLUSIONS
Under the conditions of this in vitro study:
1. The effect of finishing/polishing techniques on
microleakage of resin-modified glass ionomer
cements were tissue and material dependent.
2. At the enamel margins of Photac-Fil restorations,
finishing/polishing with OneGloss resulted in significantly less leakage then with Super-Snap.
3. At both the enamel and dentin margins of Fuji II
LC restorations, finishing/polishing with OneGloss
resulted in significantly less leakage than with
Robot Carbide. The dentin seal after treatment
with CompoSite was also significantly better than

with Robot Carbide.
4. Regardless of the finishing/polishing technique
employed, leakage at dentin margins was significantly greater than at enamel margins for Fuji II
LC. For Photac-Fil, no significant difference in leakage scores was observed between dentin and enamel with the exception of finishing/polishing with
OneGloss.
5. Fuji II LC restorations had significantly less enamel
and dentin leakage than Photac-Fil restorations
when treated with OneGloss.
References
Anusavice KJ (1996) Dental cements for restorations and pulp
protection in Phillips Science of Dental Materials 10th edition
Philadelphia WB Saunders Co p 525.
Barkmeier WW & Cooley RL (1992) Laboratory evaluation of
adhesive systems Operative Dentistry (Supplement 5) 5061 Review.
Bouschlicher MR, Vargas MA & Denehy GE (1996) Effect of
desiccation on microleakage of five Class V restorative materials Operative Dentistry 21(3) 90-95.
Brackett WW, Gilpatrick RO & Gunnin TD (1997) Effect of finishing method on the microleakage of Class V resin composite
restorations American Journal of Dentistry 10(4) 189-191.
de Magalhaes CS, Serra MC & Rodrigues AL Jr (1999)
Volumetric microleakage assessment of glass-ionomoer-resin
composite hybrid materials Quintessence International 30(2)
117-121.
Hallett KB & Garca-Godoy F (1993) Microleakge of resinmodified glass ionomer cement restorations: An in vitro
study Dental Materials 9(5) 306-311.
Hoelscher DC, Neme AM, Pink FE & Hughes PJ (1998) The
effect of three finishing systems on four esthetic restorative
materials Operative Dentistry 23(1) 36-42.
Kanchananvasita W, Anstice HM & Pearson GJ (1997) Water
sorption characteristics of resin-modified glass-ionomer
cements Biomaterials 18(4) 343-349.
Kidd EA (1976) Microleakage: A review Journal of Dentistry
4(5) 199-206.
Martin FE & ORourke M (1993) Marginal seal of cervical
tooth-coloured restorations. A laboratory investigation of
placement techniques Australian Dental Journal 38(2) 102107.
Mount GJ (1993) Clinical placement of modern glass-ionomer
cements Quintessence International 24(2) 99-107.
Owens BM, Halter TK & Brown DM (1998) Microleakage of
tooth-colored restorations with a beveled gingival margin
Sidhu SK & Watson TF (1995) Resin-modified glass ionomer
materials A status report for the American Journal of
Dentistry American Journal of Dentistry 8(1) 59-67.
Wilder AD Jr, Swift EJ Jr, May KN Jr, Thompson JY &
McDougal RA (2000) Effect of finishing technique on the
microleakage and surface texture of resin-modified glass
ionomer restorative materials Journal of Dentistry 28(5)
367-373.
Wilson AD & Paddon JM (1993) Dimensional changes occurring in a glass-ionomer cement American Journal of
Dentistry 6(6) 280-282.
Yap AU (1996) Resin-modified glass ionomer cements: A comparison of water sorption characteristics Biomaterials
17(19) 1897-1900.
Yap AU, Lim CC & Neo JC (1995) Marginal sealing ability of
Quintessence
three
cervical
restorative
systems
International 26(11) 817-820.
41
Yap AU, Lye KW & Sau CW (1997) Surface characteristics of

Yap AU, Tan S & Teh TY (2000) The effect of polishing systems
on microleakage of tooth coloured restoratives: Part 1.
Conventional and resin-modified glass-ionomer cements
Journal of Oral Rehabilitation 27(2) 117-123.
Yu XY, Wieczkowski G, Davis EL & Joynt RB (1990) Influence
of finishing technique on microleakage Journal of Aesthetic
Dentistry 2(5) 142-144.
The Effect of Flowable

Resin Composite on Microleakage
in Class V Cavities
AR Yazici M Baseren B Dayanga
Clinical Relevance
The use of a flowable resin composite as both a restorative material and a liner combined with a hybrid composite in Class V cavities can be advocated as a means of minimizing microleakage at dentin margins.
SUMMARY
This in vitro study investigated the microleakage
of flowable resin composite as a restorative material and as a liner (either light cured separately or
co-cured with hybrid resin composite) in Class V
cavities. A light-cured hybrid resin composite was
used as a control. Twenty extracted human premolars were prepared with standardized Class V
cavity outlines on the buccal and lingual surfaces.
The occlusal margin of the cavities was on enamel and the gingival margin was on dentin. One
bottle adhesive system (Single Bond) was used
after etching enamel and dentin with 34.5% phosphoric acid for 15 seconds. The cavities were randomly divided into four groups of 10 each and
restored according to the manufacturers instructions: Group IHybrid resin composite (Z100);
Group IIFlowable resin composite (Filtek Flow);
Group IIIFlowable resin composite (Filtek
*A Rya Yazici, DDS, PhD, teaching fellow, Hacettepe
University, Ankara, Turkey
Meserret Baseren, DDS, PhD, associate professor, Hacettepe
University, Ankara, Turkey
Berrin Dayanga, DDS, PhD, professor, Hacettepe University,
Ankara, Turkey
*Reprint request: Faculty of Dentistry, Dept of Conservative
Dentistry, Hacettepe University, 06100, Sihhiye, Ankara,
Turkey; e-mail: [email protected]
Flow)+Hybrid resin composite (Z100); light cured

separately; Group IVFlowable resin composite
(Filtek Flow)+ Hybrid resin composite (Z100); cocured. The samples were thermocycled 200 times
with a 30-second dwell time. They were then
immersed in a 0.5% basic fuchsin solution for 24
hours, sectioned and analyzed by stereomicroscopy. The degree of dye penetration was
recorded and analyzed with the Kruskal-Wallis
and Mann-Whitney U tests.
The results of this study indicate that there was
no leakage at the occlusal margin for either
restoration. Statistically significant differences
were found among the groups at the gingival margin. No statistically significant difference was
observed between the occlusal and gingival margins except in Group IV.
The combination of flowable resin composite
and hybrid composite light cured separately
yielded the best result in this study. The most
leakage was observed when this combination was
co-cured. The resistance to microleakage of flowable resin composite as a restorative material is
similar to that of hybrid resin composite.
INTRODUCTION
Resin composites are widely used for restoring cervical
lesions, as they are esthetic, mercury free and bond to
Yazici, Baseren & Dayanga: Microleakage of Flowable Resin Composite Used as a Liner
tooth structure with the use of bonding systems.
Unfortunately, the coefficient of linear thermal expansion of resin composites is three or four times that of
tooth structure. In addition to the differences in thermal expansion coefficients, the shrinkage of composites
during curing induces stresses at the tooth/restorative
interface and generally results in gap formation.
Therefore, polymerization shrinkage and the thermal
expansion coefficient of these restorative materials
have been suggested as major causes of microleakage
(Craig, 1989; Feilzer, de Gee & Davidson, 1988; Puckett
& others, 1992). Restoring cervical lesions with resin
composites has always been a problem, especially
where no enamel is present for bonding to the gingival
margin. The higher organic component, tubular structure, fluid pressure and the lower surface energy of
dentin make bonding to dentin more difficult than to
enamel (Barkmeier & Cooley, 1991; Pashley &
Carvalho, 1997). Poor adhesion between dentin and
restorative material predisposes gap formation.
Marginal gap formation leads to leakage, which may be
responsible for secondary caries, marginal discoloration, pulpal inflammation and hypersensitivity
(Going, 1972; Kidd, 1976; Bauer & Henson, 1984).
Many attempts have been made to limit the microleakage of composites in dentin. Dentin bonding agents are
used to improve the marginal seal of resin composite
restorations at the composite/tooth interface. They have
been proven to be effective at reducing but not eliminating microleakage (Hansen & Asmussen, 1989; Prati,
Nucci & Montanari, 1991).
43
rately or co-cured with hybrid resin composite) and

light cured hybrid resin composite as a control in Class
V cavities.
Twenty extracted human mandibular premolars stored
in tap water at room temperature were used within one
week of extraction. Only teeth that were free of caries
and restorations and showed no evidence of white spots
or cracks on buccal or lingual surfaces were selected.
Forty Class V cavities (one on the buccal and one on the
lingual surface of each tooth) were prepared with a
high-speed diamond flat end cylinder bur (835-008-3
Diatech Dental AG, Heerbrugg, Switzerland) using
water as a coolant. The bur was replaced after every
fifth preparation. Preparation included an occlusal
margin in enamel and a gingival margin in dentin. The
preparation was approximately 3 mm wide, 2 mm high
and 1.5 mm deep. After the enamel and dentin were
etched with 34.5% phosphoric acid for 15 seconds, the
cavities were thoroughly rinsed with water for 15 seconds. They were then air dried gently approximately 10
cm away from the cavity surface for five seconds, avoiding complete desiccation. Two consecutive coats of
Single Bond (3M Dental Products, St Paul, MN 55144,
USA) were applied to the whole cavity surface, followed
by gentle air drying to remove excess solvent and light
cured for 10 seconds. All preparations, etching and
bonding procedures were conducted by the same operator. The teeth were then randomly distributed into four
groups (Table 1).
In late 1996, an alternative filling material, flowable

Group I: Z100 resin composite (3M Dental Products)
composite, was introduced for restoring Class V caviwas inserted in bulk, with special attention given to
ties. Its modulus of elasticity is low, thereby, it has the
marginal adaptation. Resin was light cured for 40 secability to undergo plastic deformation to flex and absorb
onds with a calibrated visible-light curing unit (Optilux
polymerization shrinkage stress (Bayne & others,
400, Demetron Research Corp, Danbury, CT 06810,
1998). On the other hand, since it has less filler content,
USA) from a distance of 1 mm from its outer surface.
the coefficient of thermal expansion of flowable composGroup II: The cavities were bulk filled with flowable
ites is close to that of the tooth structure (Bayne & othresin composite, Filtek Flow (3M Dental Products),
ers, 1998; Chuang & others, 2001). The use of a flowwhich was cured for 20 seconds.
able composite as a liner is another recently recomGroup III: A thin layer (0.5 mm) of flowable composite
mended technique for overcoming the shortcomings of
(Filtek Flow) was lined at the axial wall of the cavity
resin composites. These resins are used as an initial
preparation. After curing flowable resin composite for
thin layer under composites and function as a stress
20 seconds, hybrid resin composite (Z100) was placed
breaker (Unterbrink & Liebenberg, 1999). Researchers
over the lining and cured for 40 seconds.
have used flowable composites as lining materials and
have obtained favorable results in reducing microleakGroup IV: A thin layer (0.5 mm) of flowable composite
age (Payne, 1999; Leevailoj & others, 2001). Belvedere
(Filtek Flow) was lined at the axial wall of the cavity
(2001) has recommended placing flowable composites without curing separately under resin composite
Table 1: Materials Tested
restorations. However, no research has determined
Group
Restorative Material
the effect of the co-curing of flowable resin composite
I
Hybrid
resin composite(Z 100)
with hybrid resin composite on microleakage. This in
II
Flowable
resin composite(Filtek Flow)
vitro study evaluated the microleakage performance
III
Flowable
+ Hybrid resin composite(light cured separately)
of light cured flowable resin composite as a restoraIV
Flowable + Hybrid resin composite(co-cured)
tive material and as a liner (either light cured sepa-
44
Operative Dentistry
preparation. Without curing flowable resin composite,

hybrid resin composite (Z100) was placed over the lining, then co-cured for 40 seconds, which means the flowable composite and hybrid resin composite were cured
at the same time.
the enamel margin. However, statistically significant

differences in microleakage were observed among the
groups at the dentin margins (p<0.001). Dye penetration scores in dentin margins differed significantly
between Groups I and IV (p=0.005), II and IV (p=0.005),
III and IV (p=0.0017).
All restorations were finished after 24 hours with

fine-grit finishing diamond burs (Edenta AG-Dental
Produkte, AU/SG, Switzerland) and Hawe-Neos disks
(Hawe-Neos Dental CH6934 Bioggio, Switzerland) of
decreasing abrasiveness. After storing the specimens at
37C for seven days, the teeth were subjected to 200
cycles between temperature baths at 5C and 55C. The
cycles lasted 30 seconds in each bath, with a 10-second
transfer time. The root apices were occluded with resin
composite, and the teeth were painted with two coats of
acid resistant varnish to within 1 mm of the margins of
the restorations. The specimens were immersed in 0.5%
basic fuchsin solution and stored for 24 hours at 37C,
after which they were washed for one minute in running water and dried. An Isomet (Buehler Ltd, Lake
Bluff, IL 60044, USA) diamond saw cooled with water
was used to section each tooth longitudinally through
the center of the restorations. Each restoration was
observed under a binocular stereomicroscope (M5 Wild
Herrbrugg, Switzerland) with a magnifying loupe of
X18. Two examiners scored the restorations independently, and any discrepancies between the two examiners were reevaluated by both and a consensus reached.
Enamel and dentin margins were scored separately.
For each restoration, the section with greater leakage
was selected for scoring.
When flowable composite was placed as a liner and cocured with hybrid resin composite (Group IV), five samples exhibited microleakage involving the axial wall
(score 3), two exhibited microleakage up to half of the
cavity depth (score 1) and three exhibited no degree of
microleakage (score 0) at the dentin margins. No dye
penetration was observed at the dentin margins in
samples restored with the combination of flowable composite as a liner and hybrid resin composite (Group III)
cured separately. Hybrid resin composite (Group I) and
flowable resin composite (Group II) used as a restorative material exhibited equal leakage patterns in the
dentin; only one sample from each group showed minimal leakage (score 1) at the dentin margins.
Comparison of the microleakage scores between
enamel and dentin margins within each group showed
that there was more leakage in the dentin in all groups
but a statistically significant difference was only seen
in Group IV (p=0.002).
DISCUSSION
Marginal seal is one of the most important factors for
the success of a restoration. The restoration of cavities
having margins partly or totally located in the dentin is
an unsolved problem in resin composites. Many studies
have shown that bonding of restorative material to
enamel is adequate (Al-Hamadani & Crabb, 1975;
Retief & Denys, 1989; Swift, Perdigo & Heymann,
1995). In this investigation, all restorations completely
resisted microleakage at the occlusal margins, proving
the effectiveness of the acid-etch technique in sealing
cavity margins in enamel. However, varying degrees of
microleakage occurred along the gingival margins that
were placed in the dentin. On the other hand, no statistically significant difference was detected between
The degree of marginal leakage was evaluated using

a standardized scoring system similar to that used by
Munro, Hilton & Hermesch (1996). A zero value was
assigned where there was no evidence of microleakage.
Dye penetration up to half the cavity depth scored a
value of 1. When the dye penetration was more than
half of the cavity depth, a value of 2 was recorded, and
when it had spread to involve the axial wall, the
microleakage was assigned a value of 3. Mean leakage
scores for all groups were also calculated.
The results of the dye penetration scores were analyzed with Kruskal-Wallis non-parametric analysis followed by Mann-Whitney U test to evaluate differences
among the experimental groups at a significance level
of p=0.05. Combined occlusal and gingival scores within each restoration were compared using the MannWhitney U test.
RESULTS
Microleakage and mean leakage scores of all groups
are presented in Table 2. The Kruskal-Wallis test
showed no statistically significant differences in
microleakage at the enamel margins among the groups
(p=1). All groups resisted microleakage completely at
Table 2: Microleakage and Mean Leakage Scores

Leakage Scores
Group
I
II
III
IV
Mean
10
Dentin
0.1
Enamel
10
Dentin
0.1
Enamel
10
Dentin
10
Enamel
10
Dentin
1.7
Enamel
Yazici, Baseren & Dayanga: Microleakage of Flowable Resin Composite Used as a Liner
the enamel and dentin margins except in Group IV, as
microleakage at the dentin margins was only slightly
greater than at the enamel margins. These results
show that a good bond to dentin is achieved with new
generation dentin bonding systems (Nikaido & others,
1997; Swift & Bayne, 1997). These new adhesives may
seal dentin margins better than previous adhesives
and, therefore, may have more effectively prevented
leakage at the dentin margins. They have the ability to
resist the contraction stress generated by polymerization shrinkage, thereby, establishing a good bond to
dentin without gap formation (Davidson, de Gee &
Feilzer, 1984; Eick & others, 1997).
Several authors have reported encouraging results in
reducing microleakage with the use of flowable composite restorative materials (Ferdianakis, 1998;
Estafan & Estafan, 2000). This study also obtained
good results with a flowable composite in Class V cavities. Mazer & Russell (1998) have reported that flowable composites and hybrid composites performed
equally well in terms of microleakage. The results of
this study were consistent with these findings. Only
one sample in the hybrid resin composite and one sample in the flowable resin composite group showed minimum leakage. This suggests that as flowable composites are more resin-rich, they have low viscosity and
flow and adapt at least as well as hybrid composites to
cavity margins. Their low modulus of elasticity allows
for plastic deformation, acts as an elastic buffer and
compensates for contraction shrinkage stress (KempScholte & Davidson, 1990; Van Meerbeek & others,
1993; Estafan & Estafan, 2000). In a SEM study by
Estafan, Estafan & Leinfelder (2000), flowable composites had better marginal integrity than hybrid and condensable composites, and flowable composites showed
no evidence of marginal gaps. In another study,
Ferdianakis (1998) compared the microleakage performance of flowable resin composite with that of
hybrid resin composite and found significantly less
leakage in cavities restored with flowables. From these
results, it can be concluded that flowable composites
can be used to restore Class V cavities.
In this study, the combination of flowable resin composite and hybrid composite that were light-cured separately completely eliminated microleakage. This confirms other investigations that proved that flowable
composites under resin composite restorations can efficiently reduce microleakage (Tung, Estafan & Scherer,
2000; Leevailoj & others, 2001). The complete resistance to microleakage found in this combination could
be related to the lower modulus of elasticity of flowables (Estafan & Estafan, 2000). The use of low modulus flowable composite may also increase the flexibility
of the bonded assembly and might act as a shock
absorber and relieve the stress induced by the polymerization shrinkage of resin composites (Kemp-
45
Scholte & Davidson, 1990; Van Meerbeek & others,

1993). Materials with high elastic modulus destroy the
bond between the restorative material and the tooth
structure and lead to poor marginal adaptation
(Unterbrink & Liebenberg, 1999).
Although the manufacturer recommended curing the
flowable composite prior to applying the restorative
materials, Belvedere (2001) reported that the hydraulic
pressure of heavier viscosity composite would help
uncured flowable composite to penetrate better and
improve the sealing of margins. Therefore, he suggested placing flowable resin composite without curing,
separately. The authors of this study thought that it
would be of interest to observe whether co-curing flowable resin and hybrid composite would have any effect
on microleakage. Contrary to suggestions, in this
group, the greatest leakage was found in dentin. Even
five samples showed leakage along the axial wall.
Leakage probably occurred up to an intact resin-enamel bond. This finding may be attributed to the fact that
polymerization shrinkage of a resin composite may create contraction forces that may disrupt the bond of
uncured flowable composite from cavity walls. On the
other hand, many composites are sticky and may have
a tendency to pull back the uncured flowable composite
from the cavity wall as the instruments used to place
them are being removed. Moreover, it has been reported that as flowable composites are separately cured and
serve as a well-adapted first increment, they resist disturbance and absorb polymerization shrinkage of the
overlying composites (Bertolotti & Laamanen, 1999).
It should be noted that these results are based on in
vitro data; therefore, future studies that evaluate the
clinical performance of flowable composites are necessary.
CONCLUSIONS
Within the limitations of this in vitro study, the following statements can be made:
1. No leakage was observed in the enamel margins of hybrid, flowable and flowable/hybridcombined restorative materials.
2. The resistance of flowable resin composite as a
restorative material to microleakage is similar
to hybrid resin composite at the dentin margins.
3. The use of flowable composite with hybrid composite cured separately prevented leakage completely. Leakage was observed at the dentin
margins when flowable resin composite was cocured with hybrid composite.
46
References
Al-Hamadani KK & Crabb HS (1975) Marginal adaptation of
composite resins Journal of Oral Rehabilitation 2(1) 21-33.
Barkmeier WW & Cooley RL (1991) Current status of adhesive
resin systems Journal of the American College of Dentistry
58(2) 36-39.
Bauer JG & Henson JL (1984) Microleakage: A measure of the
performance of direct filling materials Operative Dentistry
9(1) 2-9.
Operative Dentistry
Kemp-Scholte CM & Davidson CL (1990) Complete marginal
seal of Class V resin composite restorations effected by
increased flexibility Journal of Dental Research 69(6) 12401243.
Kidd EA (1976) Microleakage in relation to amalgam and composite restorations. A laboratory study British Dental Journal
141(10) 305-310.
Leevailoj C, Cochran MA, Matis BA, Moore BK & Platt JA (2001)
Microleakage of posterior packable resin composites with and
without flowable liners Operative Dentistry 26(3) 302-307.
Bayne SC, Thompson JY, Swift EJ Jr, Stamatiades P &

Wilkerson M (1998) A characterization of first-generation
flowable composites Journal of the American Dental
Munro GA, Hilton TJ & Hermesch CB (1996) In vitro microleakage of etched and rebonded Class V composite resin restorations Operative Dentistry 21(5) 203-208.
Belvedere PC (2001) Contemporary posterior direct composites

using state-of-the-art techniques Dental Clinics of North
America 45(1) 49-70.
Mazer RB & Russell RR (1998) The use of flowable composite

resin in Class V restorations Journal of Dental Research 67
Abstract #292 p 131.
Bertolotti RL & Laamanen H (1999) Bite-formed posterior resin

composite restorations, placed with a self-etching primer and
a novel matrix Quintessence International 30(6) 419-422.
Nikaido T, Nakajima M, Higashi T, Kanemura N, Pereira PN &

Tagami J (1997) Shear bond strengths of a single-step bonding system to enamel and dentin Dental Materials Journal
16(1) 40-47.
Chuang SF, Liu JK, Chao CC, Liao FP & Chen YH (2001) Effects
of flowable composite lining and operator experience on
microleakage and internal voids in Class II composite restorations Journal of Prosthetic Dentistry 85(2) 177-183.
Pashley DH & Carvalho RM (1997) Dentine permeability and

dentine adhesion Journal of Dentistry 25(5) 355-372.
Craig RG (1989) Dental Restorative Materials St Louis Mosby

262.
Davidson CL, de Gee AJ & Feilzer A (1984) The competition
between the composite-dentin bond strength and the polymerization contraction stress Journal of Dental Research
63(12) 1396-1399.
Eick JD, Gwinnett AJ, Pashley DH & Robinson SJ (1997)
Current concepts on adhesion to dentin Critical Reviews of
Oral Biology and Medicine 8(3) 306-335.
Estafan AM & Estafan D (2000) Microleakage study of flowable
composite resin systems Compendium 21(9) 705-708, 710,
712, quiz 714.
Estafan D, Estafan AM & Leinfelder KF (2000) Cavity wall
adaptation of resin-based composites lined with flowable composites American Journal of Dentistry 13(4) 192-194.
Feilzer AJ, de Gee AJ & Davidson CL (1988) Curing contraction
of composites and glass-ionomer cements Journal of
Prosthetic Dentistry 59(3) 297-300.
Ferdianakis K (1998) Microleakage reduction from newer esthetic restorative materials in permanent molars Journal of
Clinical Pediatric Dentistry 22(3) 221-229.
Going RE (1972) Microleakage around dental restorations: A
summarizing review Journal of the American Dental
Association 84(6) 1349-1357.
Hansen EK & Asmussen E (1989) Marginal adaptation of posterior resins: Effect of dentin-bonding agent and hygroscopic
expansion Dental Materials 5(2) 122-126.
Payne JH 4th (1999) The marginal seal of Class II restorations:

Flowable composite resin compared to injectable glass
ionomer Journal of Clinical Pediatric Dentistry 23(2) 123-130.
Prati C, Nucci C & Montanari G (1991) Shear bond strength and
microleakage of dentin bonding systems Journal of Prosthetic
Dentistry 65(3) 401-407.
Puckett A, Fitchie J, Hembree J Jr & Smith J (1992) The effect
of incremental versus bulk fill techniques on the microleakage
of composite resin using a glass-ionomer liner Operative
Dentistry 17(5) 186-191.
Retief DH & Denys FR (1989) Adhesion to enamel and dentin
American Journal of Dentistry 7(2 Spec No) 133-144.
Swift EJ Jr, Perdigo J & Heymann HO (1995) Bonding to enamel and dentin: A brief history and state of the art Quintessence
Swift EJ Jr & Bayne SC (1997) Shear bond strength of a new
one-bottle dentin adhesive American Journal of Dentistry
10(4) 184-188.
Tung FF, Estafan D & Scherer W (2000) Microleakage of a condensable resin composite: An in vitro investigation
Unterbrink GL & Liebenberg WH (1999) Flowable resin composites as filled adhesives; Literature review and clinical
recommendations Quintessence International 30(4) 249-257.
Van Meerbeek B, Willems G, Celis JP, Roos JR, Braem M,
Lambrechts P & Vanherle G (1993) Assessment by nanoindentation of the hardness and elasticity of the resin-dentin
bonding area Journal of Dental Research 72(10) 1434-1442.
Effects of Instrumentation Time

on Microleakage of
Resin-Modified
Glass Ionomer Cements
AUJ Yap EJC Yeo WY Yap
DSB Ong JWS Tan
Clinical Relevance
The effects of instrumentation time on microleakage of resin-modified glass ionomer

cements are material, finishing/polishing system and tissue dependent. Finishing/polishing of resin-modified glass ionomers should be delayed where possible.
SUMMARY
This study investigated the effect of instrumentation time on the microleakage of resin-modified glass ionomer cements (RMGICs). Class V
cavities were prepared on buccal and lingual/
palatal surfaces of 64 freshly extracted non-carious premolars. The cavities on each tooth were
restored with Fuji II LC (FT [GC]) and Photac-Fil
Quick (PF [3M-ESPE]). The restored teeth were
randomly divided into two groups of 32 teeth.
Finishing/polishing was done immediately after
light-polymerization in one group and was
delayed for one week in the other group. The fol*Adrian UJ Yap, BDS, MSc, PhD, FAMS, FADM, FRSH, associate professor, Department of Restorative Dentistry, Faculty of
Egwin JC Yeo, student, Faculty of Dentistry, National
WY Yap, student, Faculty of Dentistry, National University of
Singapore
Debbie SB Ong, student, Faculty of Dentistry, National
Jane WS Tan, student, Faculty of Dentistry, National
*Reprint request: 5 Lower Kent Ridge Road, Singapore 119074,
Republic of Singapore; e-mail: [email protected]
lowing finishing/polishing systems were evaluated: (a) Robot Carbides (RC); (b) SuperSnap (SS);
(c) OneGloss (OG) and (d) CompoSite Polishers
(CS). The sample size for each instrumentation
time, material and finishing/polishing system
combination was 8. Storage medium for both
immediate and delayed instrumentation groups
was distilled water at 37C during the hiatus
period. The teeth were subsequently subjected to
dye penetration testing (0.5% basic fushcin), sectioned and scored. Data were analyzed using
Kruskal-Wallis and Mann-Whitney U tests at significance level 0.05. For PF, significant difference
in enamel leakage was observed between immediate and delayed instrumentation with SS and
CS. Significant differences in dentin leakage
were also observed between the two instrumentation times for SS. For FT, significant differences in leakage between instrumentation times
were observed only in dentin and with RC.
Where significant differences in dye penetration
scores existed, delayed finishing/polishing
resulted in less microleakage.
INTRODUCTION
The favorable adhesive and fluoride-releasing properties of glass ionomer cements (GICs) have led to their
widespread use as luting, lining and restorative mate-
Operative Dentistry
48
rials. Resin-modified glass
ionomer cements (RMGICs)
were introduced to help overcome the problems of moisture
sensitivity and low early
mechanical strengths associated
3
2
3 2
with conventional GICs, and at
0
1
the same time, to maintain their
1 0
clinical advantages (Sidhu &
Watson, 1995). In RMGICs, the
fundamental acid-base curing
reaction is supplemented by a
second curing process that was
light-initiated. In their simplest
form, they are GICs with the
addition of small amounts of
resin, such as hydroxyethyl
Figure 1: Scoring scale
methacrylate
(HEMA)
or
used to measure microleakage.
BisGMA. More complex materials have been developed with
polyacid side chains, which can polymerize by light-curing mechanisms.
the light source. To make matters worse, the acid-base

reaction is retarded due to the replacement of water
(which serves as the reaction medium) with water/resin
mixtures in RMGICs (Wan, Yap & Hastings, 1999).
While the acid-base complexation reaction is essentially complete after 24 hours for conventional materials, it
levels off only after 168 hours in RMGICs. Immediate
finishing/polishing after light curing may therefore
result in disruption of the weak early chemical bonds,
leading to gap formation at the tooth-restoration interface. The latter may lead to microleakage, which contributes to staining, postoperative sensitivity and/or
recurrent caries clinically.
The effects of instrumentation time on microleakage
of RMGICs have not been widely investigated (Lim,
Neo & Yap, 1999). This study investigated the effects of
instrumentation time on the microleakage of resinmodified glass ionomer cements (RMGICs). The possibility of microleakage being material, finishing/polishing system and tissue dependent was also determined.
The RMGICs that were investigated and their technical

profiles are shown in Table 1. Sixty-four freshly extracted, non-carious premolars were selected for the study.
The teeth were disinfected with 2% formaline-saline,
cleaned and stored in distilled water at 4C until use.
Wedge-shaped Class V preparations (approximately 4 mm
wide mesio-distally), 3 mm long (occluso-gingivally) and
2 mm (deep) were made on the buccal and lingual/
palatal surfaces of each tooth. The cavities on each
tooth were restored with capsulated Fuji II LC (GC)
and Photac-Fil Quick (3M-ESPE, St Paul, MN 55144,
USA) of A2 shade. Cavities restored with Fuji II LC
were first treated with
Cavity Conditioner (GC)
Table 1: Technical Profiles of the Resin-Modified Glass Ionomer Cements Investigated
for 10 seconds, while cavities restored with PhotacMaterial
Manufacturer
Components
Mean Particle
Lot #
Fil Quick were treated
Size (m)
with Ketac Conditioner
Fuji II LC
GC Corporation,
Powder:
4.5
9912202
(3M-ESPE) for 10 secTokyo, Japan
Alumino silicate glass,
onds. The cavities were
pigments
then washed for 30 secLiquid:
onds and gently air dried.
Polyacrylic acid, distilled
The RMGICs were mixed
water, HEMA (17%),
dimethacrylate monomer,
according to manufacturcamphoroquinone
ers instructions and
Photac-Fil Quick
3M-ESPE,
Powder:
7.0
0065231
injected into the cavities.
Seefeld, Germany
Calcium aluminium
Transparent preformed
fluorosilicate glass,
cervical matrixes (Hawecopolymers of acrylic
Neos Dental, Bioggio,
and maleic acids, tartaric
acid, activator, pigments
Switzerland) were placed
over the filled cavities and
Liquid:
HEMA (40%),
pressure was applied to
difunctional monomer,
extrude excess material,
Although laboratory tests show that RMGICs are not

any stronger than conventional ones (Burgess, Norling
& Summitt, 1994), strength is developed more quickly,
making it possible to begin finishing and polishing
almost immediately after light curing. This is a distinct
advantage over conventional materials where finishing
and polishing must be delayed for at least 48 hours
(Mount, 1993). It is, however, important to note that the
resin components form only 4.3% to 6% of RMGICs
(Toledano & others, 1999) and light curing only sets the
resin components. The fundamental acid-base reaction
is therefore relatively immature following removal of
water, camphoroquinone
Yap & Others: Effects of Instrumentation Time on Microleakage of Resin-Modified Glass Ionomer Cements
Table 2: Finishing/Polishing Systems and Sequences
Product
Usage
Handpiece Speed
Manufacturer
SH134F
Wet, 12 strokes
300,000 rpm
Shofu Inc,
SH134UF
Wet, 12 strokes
300,000 rpm
Kyoto, Japan
Coarse
Medium
Fine
Extra fine
Dry,
Dry,
Dry,
Dry,
12,
12,
12,
12,
Shofu Inc,
Kyoto, Japan
OneGloss
Wet, 12 heavy strokes

Wet, 12 light strokes
10,000 rpm
10,000 rpm
Shofu Inc,
Kyoto, Japan
Wet, 12 strokes
Dry, 12 strokes
12,000 rpm
12,000 rpm
Shofu Inc,
Kyoto, Japan
Robot Carbides
SuperSnap
6
6
6
6
strokes
strokes
strokes
strokes
000
000
000
000
rpm
rpm
rpm
rpm
CompoSite
Polishers
CompoSite
CompoSite Fine
Table 3: Distribution of Microleakage Scores for Photac-Fil Quick

System
Enamel
Dentin
Immediate
Delayed
Immediate
Delayed
0 1 2 3
0 1 2 3
0 1 2 3
0 1 2 3
Robot Carbides
2 0 0 6
5 1 0 2
0 0 0 8
0 0 0 8
SuperSnap
0 0 0 8
7 0 0 1
0 0 0 8
3 0 2 3
CompoSite
3 0 0 5
7 0 0 1
1 0 0 7
0 1 4 3
OneGloss
4 0 0 4
6 0 1 1
0 0 0 8
2 0 1 5
A
Table 4: Distribution of Microleakage Scores for Fuji II LC
System
Enamel
Dentin
49
evaluated: (a) Robot Carbides, (b)

SuperSnap, (c) OneGloss and (d)
CompoSite Polishers. Details of the
finishing/polishing sequences are
reflected in Table 2. The sample size
for each instrumentation time, material and finishing/polishing system
combination was 8.
The apices of the teeth were subsequently sealed with acrylic. Two coats
of nail varnish were applied, leaving a
1-mm margin from the restoration
free of varnish. The teeth were then
soaked in 0.5% basic fuschin dye at
37C for 24 hours. After dye penetration testing, the nail varnish was
removed with an ultrasonic scaler and
excess dye was washed off. The aforementioned was done to prevent residual dye
contamination of the sectioned surfaces.
The teeth were subsequently sectioned longitudinally (bucco-lingually) at the center
using a Microslice II Precision Saw
(Cambridge Instruments Ltd, Cambridge,
England) and scored. Dye penetration was
assessed using an optical stereomicroscope
(SE, Nikon, Tokyo, Japan) at 10x magnification. The degree of dye penetration for
both the enamel and dentin margins was
graded using the following scale (Figure 1):
Immediate
Delayed
Immediate
Delayed
0 = No evidence of dye penetration
0 1 2 3
0 1 2 3
0 1 2 3
0 1 2 3
Robot Carbides
4 0 0 4
5 1 2 0
0 0 0 8
1 0 3 4
1 = Dye penetration to 1/3 the cavity

depth
SuperSnap
6 0 1 1
7 0 0 1
0 0 3 5
2 1 1 4
CompoSite
5 1 2 0
8 0 0 0
1 0 3 4
3 0 1 4
OneGloss
8 0 0 0
7 0 1 0
1 0 3 4
2 0 3 3
2 = Dye penetration to 2/3 the cavity

depth
3 = Dye penetration to the full cavity
depth
which was subsequently removed. The cements were

then light polymerized for 20 seconds using a curing
light (Spectrum; Dentsply Inc, Milford, DE 19963,
USA) with an output intensity 420 mW/cm2, and
assessed with a curing radiometer (Cure Rite, EFOS
Inc, Ontario, Canada).
Microleakage data were subjected to non-parametric

statistical analysis (Kruskal-Wallis and Mann Whitney
U tests) at significance level 0.05.
The cervical matrixes were removed after light curing

and gross contouring was done with 8-flute tungsten
carbide burs (Robot Carbide SH134; Shofu, Kyoto,
Japan) under water spray using a high-speed handpiece at 300,000 rpm. The restored teeth were then randomly divided into two groups of 32 teeth. Specimens in
the first group were finished/polished immediately
after light curing/gross contouring and stored in distilled water at 37C for one week. In the second group,
finishing/polishing was delayed for one week. Storage
during the hiatus period was again in distilled water at
37C. The following finishing/polishing systems were
The distribution of microleakage scores for the different

instrumentation times and materials are shown in
Tables 3 and 4. Results of the statistical analyses are
reflected in Tables 5 to 7. The effects of instrumentation
time on microleakage were found to be material, finishing/polishing system and tissue dependent. For PhotacFil, immediate instrumentation with SuperSnap and
CompoSite resulted in significantly more enamel leakage than delayed instrumentation. Significantly more
dentin leakage was observed with immediate instrumentation with SuperSnap (Table 5). For Fuji II LC, no
significant difference in microleakage was observed
RESULTS
50
Operative Dentistry
Table 5: Comparison of Microleakage Scores Between Immediate and Delayed Finishing/

Polishing
Photac-Fil Quick
Tissue
Robot Carbides
SuperSnap
OneGloss
CompoSite
Enamel
NS
Immediate >
Delayed
NS
Immediate >
Delayed
Dentin
NS
Immediate >
Delayed
NS
NS
Tissue
Robot Carbides
SuperSnap
OneGloss
CompoSite
Enamel
NS
NS
NS
NS
Fuji II LC
for several finishing/polishing

systems with immediate instrumentation. With delayed instrumentation, Photac-Fil only had
significantly greater dentin leakage than Fuji II LC when finishing/polishing was done with
Robot Carbides.
DISCUSSION
A wide spectrum of commercially available finishing/polishing

Dentin
Immediate >
NS
NS
NS
systems was evaluated. These
Delayed
included tungsten carbide
NS denotes no statistical significance while > indicates statistically significant differences (Results of Mann Whitney U test at p<0.05).
(Robot Carbides), abrasive disk
(SuperSnap),
one-step
Table 6: Comparison of Microleakage Scores Between Finishing/Polishing Systems
(OneGloss)
and
two-step
(CompoSite Polishers) rubber
Photac-Fil
abrasive finishing/polishing sysTissue
Immediate
Delayed
tems. Among the various finishEnamel
OneGloss < SuperSnap
NS
ing/ polishing systems, abrasive
Dentin
NS
SuperSnap, CompoSite < Robot
disks provided the best surface
Carbides
finish (Yap & others, 2002;
Fuji II LC
Hoelscher & others, 1998; Yap,
Tissue
Immediate
Delayed
Lye & Sau, 1997). Although it
Enamel
OneGloss < Robot Carbides
NS
has been stated that RMGICs
Dentin
OneGloss, CompoSite < Robot
NS
are resistant to water uptake
Carbides
after light activation (Mount,
NS denotes no statistical significance while < indicates statistically significant differences (Results of Kruskal Wallis and Mann Whitney
1990), they have been shown to
U test at p<0.05).
absorb substantial amounts of
water after light curing
Table 7: Comparison of Microleakage Scores Between Material
(Kanchananvasita, Anstice &
Pearson, 1997; Yap, 1996). This
Immediate Finishing
has been attributed to water
Tissue
Robot Carbides
SuperSnap
OneGloss
CompoSite
uptake by the polyHEMA comEnamel
NS
PF > FT
PF > FT
NS
plex and the formation of a
Dentin
NS
NS
PF > FT
NS
hydrogel of calcium and aluDelayed Finishing
minum
polyacrylates
in
Tissue
Robot Carbides
SuperSnap
OneGloss
CompoSite
RMGICs (Yap, 1996; Wilson,
Enamel
NS
NS
NS
NS
1990). Water sorption may allow
Dentin
PF > FT
NS
NS
NS
for some degree of relaxation of
PF = Photac-Fil and FT = Fuji II LC. NS denotes no statistical significance while > indicates statistically significant differences (Results
polymerization/setting stresses
of Mann Whitney U test at p<0.05).
and
reduce
microleakage
(Thonemann & others, 1997;
between the two instrumentation times with the excepCarvalho & others, 1996). As such, the duration of
tion of finishing/polishing with Robot Carbides.
water storage was standardized at one week for both
Immediate instrumentation with Robot Carbides
the immediate and delayed instrumentation groups.
resulted in significantly more dentin leakage (Table 5).
Although significant differences in microleakage
Although significant differences in enamel and dentin
scores were observed for some material-finishing/polmicroleakage were generally observed between finishishing system-tissue combinations, a general reduction
ing/polishing systems with immediate instrumentation,
in enamel and dentin microleakage was observed with
no significant difference was observed with delayed
delayed instrumentation for both RMGICs. Findings
instrumentation with the exception of leakage at the
were consistent with those of Irie & Suzuki (2002) and
dentin margins of Photac-Fil restorations (Table 6). A
Lim & others (1999), who found decreased gap formasimilar trend was observed when leakage scores of the
tion and microleakage with delayed finishing/polishing
materials were compared (Table 7). The leakage scores
of RMGICs. The greater leakage observed with immeof Photac-Fil were significantly greater than Fuji II LC
Yap & Others: Effects of Instrumentation Time on Microleakage of Resin-Modified Glass Ionomer Cements
diate instrumentation may be attributed, in part, to the
disruption of the weak ionic (chemical) bonds of
RMGICs immediately after light curing as stated earlier. The latter might be aggravated by the desiccation of
restorations during finishing/polishing with dry systems. This was evidenced by the significant differences
in leakage scores between the two instrumentation
times for Photac-Fil with SuperSnap and CompoSite.
Wilder & others (2000), however, found no statistically
significant difference in microleakage between wet and
dry polishing of RMGICs. Setting shrinkage of
RMGICs may also contribute to the observed phenomena. The setting shrinkage of RMGICs is higher than
conventional glass ionomer cements (Tay, 1995). This is
due to the fact that while slow setting conventional
materials permit stress relief within the restoration,
RMGICs exhibit more rapid setting contraction
through polymerization of the resin component.
RMGICs shrink 3.28% to 4.78 % within five minutes
after light curing and shrinkage can continue up to 12
hours (Crim, 1993). If restorations are finished/polished
to the cavity margins immediately after light curing,
the inherent contraction could lead to increased gap formation, resulting in increased microleakage. Ninety
percent of the equilibrium water uptake of RMGICs
occurs within one week (Kanchanavasita & others,
1997). As leakage was present even after one weeks
storage in water, it could be concluded that water sorption is insufficient to compensate for gap formation
resulting from shrinkage and instrumentation. As Yap
& others (2001) also reported that the maximum properties of RMGICs are achieved at one week, it could be
suggested that finishing/polishing should be delayed
and not conducted immediately after light polymerization.
Delayed instrumentation also appeared to reduce the
influence of finishing/polishing systems on microleakage of RMGICs. With immediate instrumentation, significant differences between finishing/polishing systems were observed for most material-tissue combinations. In general, treatment with OneGloss resulted in
significantly less leakage compared to the other finishing/polishing systems. With delayed instrumentation,
no significant difference was observed between
OneGloss and the other finishing/polishing systems.
Significant differences in microleakage were, however,
observed at the dentin margins of Photac-Fil restorations where the use of SuperSnap and CompoSite
resulted in significantly less leakage than Robot
Carbides. The latter may be attributed, in part, to the
chemistry of Photac-Fil and the large mechanical
stresses generated by ultra high-speed finishing/polishing. Photac-Fil employs the use of freeze-dried co-polymers of polyacrylic acid. As the anhydrous polyacrylic
acid needs to be hydrated and activated prior to reaction with the aluminosilicate glass, the ionic exchange
51
mechanism may require more time to mature. Enamel

microleakage was not affected by the use of Robot
Carbides when instrumentation was delayed, as the
seal of RMGICs to enamel is superior to dentin (Yap,
Tan & Teh, 2000). This is due to the higher inorganic
content and homogeneity of enamel (Anusavice, 1996).
The aforementioned also explains the significantly
greater dentin leakage associated with Photac-Fil
when compared to Fuji II LC after finishing/polishing
with Robot Carbides.
CONCLUSIONS
Under the conditions of this in vitro study:
1. The effects of instrumentation time on
cements are material, finishing/polishing system and tissue dependent.
2. Delayed finishing/polishing of resin-modified
glass ionomers generally resulted in less leakage.
3. Delayed finishing/polishing also reduces the
influence of finishing/polishing systems on
cements.
References
Anusavice KJ (1996) Dental cements for restorations and pulp
protection in Phillips Science of Dental Materials 10th edition
Philadelphia WB Saunders Co p 525.
Burgess J, Norling B & Summitt J (1994) Resin ionomer restorative materials: The new generation Journal of Esthetic
Dentistry 6(5) 207-215.
Carvalho RM, Pereira JC, Yoshiyama M & Pashley DH (1996) A
review of polymerization contraction: The influence of stress
development versus stress relief Operative Dentistry 21(1) 1724.
Crim GA (1993) Marginal leakage of visible light-cured glass
ionomer restorative materials Journal of Prosthetic Dentistry
69(6) 561-563.
Hoelscher DC, Neme AML, Pink FE & Hughes PJ (1998) The
Irie M & Suzuki K (2002) Effects of delayed polishing on gap formation of cervical restorations Operative Dentistry 27(1) 59-65.
Kanchananvasita W, Anstice HM & Pearson GJ (1997) Water
sorption characteristics of resin-modified glass ionomer
cements Biomaterials 18(4) 343-349.
Lim CC, Neo J & Yap AUJ (1999) The influence of finishing time
on the marginal seal of a resin-modified glass ionomer and
polyacid-modified resin composite Journal of Oral
Mount GJ (1990) Newer restorative materials Part II Dentistry
Today 6(1) 1-8.
52
Operative Dentistry
Mount GJ (1993) Clinical placement of modern glass ionomer

cements Quintessence International 24(2) 99-107.
Wilson AD (1990) Resin-modified glass ionomer cements

International Journal of Prosthodontics 3(5) 425-429.
Sidhu SK & Watson TF (1995) Resin-modified glass ionomer

materials: A status report for the American Journal of
Dentistry American Journal of Dentistry 8(1) 59-97.
Yap AUJ (1996) Resin-modified glass ionomer cements: A comparison of water sorption characteristics Biomaterials 17(19)
1897-1900.
Tay WM (1995) An update on glass ionomer cements Dental

Update 22(7) 283-286.

Thonemann BM, Federlin M, Schmalz G & Hiller KA (1997)

SEM analysis of marginal expansion and gap formation in
Class II composite restorations Dental Materials 13(3) 192-197.
Toledano M, Osorio E, Osorio R & Garca-Godoy F (1999)
Microleakage of Class V resin-modified glass ionomer and
compomer restorations Journal of Prosthetic Dentistry 81(5)
610-615.
Wan AC, Yap AUJ & Hastings GW (1999). Acid-base complex
reactions in resin-modified and conventional glass ionomer
cements Journal of Biomedical Material Research 48(5) 700-704.
Wilder AD, Swift EJ, May KN, Thompson JY & McDougal RA
(2000) Effect of finishing technique on the microleakage and
surface texture of resin-modified glass ionomer restorative
materials Journal of Dentistry 28(5) 367-373.
Yap AUJ, Mudambi SM, Chew CL & Neo JCL (2001) Mechanical
properties of an improved visible light-cured resin-modified
glass ionomer cement Operative Dentistry 26(3) 295-301.
Yap AUJ, Tan WS, Yeo JC, Yap WY & Ong SB (2002) Surface texture of resin-modified glass ionomer cements: Effects of finishing/polishing systems Operative Dentistry (accepted for
publication).
Yap AUJ, Tan S & Teh TY (2000) The effect of polishing systems
on microleakage of tooth colored restoratives: Part 1.
Conventional and resin-modified glass ionomer cements
Surface Geometry of
Three Packable and
One Hybrid Composite
After Finishing
M Jung S Voit J Klimek
Clinical Relevance
Finishing with a 30 m diamond caused a considerable roughening of the surfaces. After

finishing in two steps, roughness was reduced significantly. A 30 m diamond cannot be
recommended for use on Definite (Degussa) surfaces due to destructive effects. Solitaire
(Heraeus-Kulzer) was finished efficiently by two diamonds.
SUMMARY
This study compared the effects of different finishing techniques on the surface of a traditional
hybrid composite and three packable composites.
Four composites were used in the study
(Herculite XRV/Kerr, Definite/Degussa, SureFil/
Dentsply and Solitaire/Heraeus-Kulzer). Fifty
specimens were made of each material, sized 7 x 7
x 4 mm. Fifteen specimens of each material were
subjected to the following finishing procedures:
(1) a 30 m diamond, (2) a 30 m and a 20 m diamond and (3) a 30 m diamond followed by a tungsten carbide finishing bur. As a reference, five
specimens of each material were treated by SofLex discs (3M). For quantitative surface evaluation, laser-stylus profilometry was used.
Roughness parameters included average rough*Martin Jung, priv-doz, Policlinic for Operative and Preventive
Dentistry, Justus-Liebig-University, Giessen, Germany
Stefan Voit, dentist, Policlinic for Operative and Preventive
Dentistry, Justus-Liebig-University, Giessen, Germany
Joachim Klimek, DDS, professor, Policlinic for Operative and
Preventive Dentistry, Justus-Liebig-University, Giessen,
Germany
*Reprint request: Schlangenzahl 14, D-35392 Giessen, Germany;
e-mail: [email protected]
ness (Ra) and profile-length-ratio (LR). Statistical

analysis was performed with one- and two-way
Anova and Scheff post-hoc tests. Qualitative surface evaluation in SEM was performed at a tension
of 25kV.
Significant effects were found with both the
composites and the finishing methods with
respect to surface roughness (p<0.001 for Ra and
LR). A 30 m diamond caused the greatest roughness on all composites, with Ra ranging from 2.015
- 2.079 m (p<0.001). After finishing using methods
2 and 3, the Solitaire surfaces were significantly
smoother (p<0.001 for LR). The lowest roughness
values were achieved after using disks; again, the
Solitaire specimens yielded the lowest Ra and LR
values (p<0.001 except for Sure-Fil). With scanning
electron microscopy, surface areas with signs of
destruction were found after using a 30 m diamond on Definite specimens.
INTRODUCTION
In recent years, packable composites have been introduced to the spectrum of direct restorative materials as
an alternative to dental amalgam in posterior teeth.
Packable composites show great variations with respect
to their matrix and filler composition. Mechanical properties such as diametral tensile strength, compressive
54
Operative Dentistry
Table 1: Details and Properties of the Composites Evaluated (based on information by manufacturers)
Material
Lot #
Matrix
Fillers
Filler Size
in m
Filler Content
% by Weight
Filler Content
% by Volume
Herculite
XRV
902844
Bis-GMA
TEGDMA
EBADM
Definite
CHB 224
SureFil
Solitaire
Vickers Hardness
VHN
Ba-AlBorosilicate
SiO2
ZnO
TiO2
0.6
78
59
82*
InorganicOrganic
Siloxane Polymers
Ba-Glass
Aerosils
Apatite
1-1.5
77
61
65.8
981027
UDMA
TEGDMA
Ba-Al-B-FSI-Glass
SiO2
Nanofiller
0.8
82
66
70.14
010029
HPMA
ETMA
Bis-GA
Tetrafunctional
Molecules
Ba-Al-B-FSI-Glass
Porous SiO2
3-22
66
90
41.7
* VHN 0.5/30 (Hofmann & others, 2000)

VHN 0.2/40 (Manhart, Chen & Hickel, 2001)
Table 2: Details of the Rotary Instruments

Type of Bur
Manufacturer
Order #
Finishing diamond
Brasseler,Savannah,
Georgia 31419, USA
806 314 166 514 014
24-40 m
Finishing diamond
Brasseler,Savannah,
Georgia 31419, USA
806 314 166 504 014
15-30 m
Tungsten carbide
Brasseler,Savannah,
Georgia 31419, USA
500 314 166 041 014
16-fluted
Sof-Lex Discs
finishing bur
3M ESPE Dental Products,

St Paul, MN 55144, USA
1982
1982
1982
1982
100 m
29 m
14 m
5 m
and flexural strength or fracture toughness varied considerably among the different packable composites and
were not superior or even lower in quality than those of
traditional hybrid composites (Cobb & others, 2000;
Kelsey & others, 2000; Leinfelder, Bayne & Swift, 1999;
Manhart & others, 2000). In an effort to facilitate clinical handling, packable composites were promoted as
being capable of minimizing polymerization shrinkage
and assuring an increased depth of cure. Scientific documentation of these properties indicate that this has
actually not been satisfactorily accomplished (Chen &
others, 2001; Cobb & others, 2000; Yap, 2000).
Although packable composites may offer some technical advantages and a more convenient placement, there
is no convincing evidence that these materials are clinically superior to traditional hybrid composites
(Loguercio & others, 2001; Oberlnder & others, 2001).
When placing composite restorations, rotary instrumentation is inevitable. The novel matrix formulation and
filler composition of packable composites might influence the state of the surface after finishing and polishing. Therefore, a re-evaluation of the new composites
with respect to surface quality was deemed necessary.
C
M
F
SF
Particle Size/# of Blades
The trimming of composite surfaces comprises two

different steps, that is, finishing and polishing.
Finishing includes the gross removal of overhangs,
anatomical contouring and the initial smoothing of surfaces. Polishing aims to reduce surface roughness to the
lowest possible level.
This study examined the effect of different finishing
procedures on the surface quality of three packable and
one traditional microhybrid composite. The evaluation
of polishing methods will be the subject of an additional study.
Three packable composites were used for this study.
Definite (Degussa AG, 63403 Hanau, Germany)
belongs to the group of ormocers (organically modified
ceramic) by virtue of novel inorganic-organic copolymers in the matrix formulation. SureFil
(Dentsply/Caulk, Milford, DE 19963, USA) is characterized by a special interlocking filler technology.
Solitaire (Heraeus Kulzer, 61273 Wehrheim, Germany)
consists of an organic glass matrix combined with
large, porous silicon-dioxide filler particles. For com-
55
Jung, Voit & Klimek: Finishing of Three Packable and One Hybrid Composite
parison, Herculite XRV (Kerr Manufacturing Company,
Romulus, MI 48174, USA), a traditional hybrid composite, was also used in the study. Details and properties of the four composites are summarized in Table 1.
Fifty specimens sized 7 x 7 mm and 4-mm thick were
made of each composite using glass molds. The specimens were polymerized for 40 seconds on both sides
with the light curing unit Optilux 400 (VCL 401;
Demetron, Danbury, CT 06810, USA). The output was
verified using a curing radiometer (Model 100, P/N
10503, Demetron) to ensure a power density >600
mW/cm2. After fabrication of the specimens, the surface
layer was removed by sandpaper discs of 400 and 600
grit (Leco Corporation, St Joseph, MI 49085, USA) for
30 seconds, each under running water (automatic polishing apparatus A 250, Jean Wirtz, Duesseldorf,
Germany). Using the stereomicroscope Stemi SV6 (Carl
Zeiss, 37081 Goettingen, Germany) at a magnification
4x, the surfaces of the specimens were examined for
irregularities.
Three rigid rotary instruments were chosen for finishing the composite surfaces. Also, the specimens were
treated with Al2O3-coated flexible Sof-Lex discs. Details
of the rotary instruments are specified in Table 2.
The finishing of specimens was performed following
three different protocols:
Finishing method 1 (FM 1; one step): a 30 m finishing diamond.
Finishing method 2 (FM 2; two steps): a sequence of
a 30 m and a 20 m finishing diamond.
Finishing method 3 (FM 3; two steps): a 30 m diamond followed by a 16-fluted tungsten carbide finishing bur.
Clinical standard (CS; four steps): the consecutive
use of four flexible Sof-Lex discs from coarse to
extra-fine.
Fifteen randomly selected specimens of each composite were finished according to FM 1, 2 and 3. Five specimens were treated with Sof-Lex discs (CS). Finishing
was performed manually with a red-ring handpiece (24
LN Intramatic Lux 2, KaVo, 88400 Biberach, Germany)
at 40,000 rpm under three-way water-cooling. The flexible discs were mounted in a blue-ring handpiece
(Intramatic Lux 20L with head 68G, KaVo) and used at
4,000 rpm under two-way water-cooling. The number of
revolutions was verified electronically (Moviport C
117.11, Braun, 71334 Waiblingen, Germany). After
application on five surfaces, a new finishing bur was
used. A new flexible disc was used for each surface. Time
was limited to 30 seconds per instrument. The selection
of the finishing methods and the composite specimens
was accomplished according to a randomized protocol;
during finishing, the type of composite was blind.
After rotary instrumentation, the composite surfaces

were evaluated quantitatively and qualitatively.
Quantitative examination was performed by optical
profilometry. A laser stylus (Focodyn, Rodenstock,
80469 Munich, Germany), focused to a diameter of 1
m, was used for scanning the finished surfaces. Profile
data were transmitted to the processing unit S8P
(Mahr, 37073 Goettingen, Germany). Each surface was
scanned by nine parallel tracings that were generated
automatically, ensuring a constant side shift (DY) of
0.22 mm. Transverse length (LT) was set to 1.75 mm;
the distance used for calculating roughness parameters
(sampling length LM) was 1.25 mm. The complete profilometric settings were:
LT = 1.75 mm;
LM = 1.25 mm
DY = 0.22 mm;
n (scans per surface) = 9;
LY = 1.76 mm
(cut-off) = 0.25 mm (Gauss profile-filter)
The surface area on each specimen evaluated quantitatively was 1.25 mm x 1.76 mm.
Surface quality was characterized by average roughness (Ra) and profile-length-ratio (LR). Ra is defined as
the arithmetic mean of the absolute ordinate values
within the sampling length (ISO 4287 [ISO-Standards,
1997]). LR represents the length of a profile tracing
drawn out to a straight line (true profile length) in relation to the sampling length (DIN 4762 [DIN-Normen,
1996]). LR is dimensionless; an ideally smooth surface
yields an LR=1.
Statistical analysis of the quantitative results was
carried out using SPSS for Windows (version 10.07).
The Ra and LR data were distributed normally and differences between the methods were analyzed with oneand two-way ANOVA and Scheff post-hoc tests at a
significance level p<0.001.
Qualitative examination was carried out with scanning electron microscopy (PSEM 500, Philips
Electronics, 5600 MD Eindhoven, Netherlands) at a
working tension of 25 kV. Two specimens each were
randomly selected for the SEM study, representing the
three finishing methods and the disc group. This resulted in a total of 32 specimens for the four composites.
During the qualitative evaluation, both the type of composite and the finishing method were blind. From each
surface, a photomicrograph was taken at an original
magnification of 80x. Photoprints sized 16 cm x 12 cm
were divided into 48 squares. Each square was evaluated according to the following grading system:
Grade 1 - smooth, homogeneous surface.
Grade 2 - minor roughness.
Grade 3 - severe roughness.
Grade 4 - detrimental surface area.
56
Operative Dentistry
RESULTS
Quantitative Evaluation
Analyzing the results by two-way ANOVA revealed significant effects both on the finishing methods and on
the different composites on surface roughness with
respect to Ra and LR (p<0.001 for Ra and LR). The use
of a 30 m finishing diamond caused the greatest
roughness on all composites (p<0.001 for Ra and LR,
Figure 1 and 2). A sequence of two finishing diamonds
(FM 2) led to a considerable reduction in average
roughness below the level of 1 m (p<0.001). LR values
were also reduced significantly (p<0.001), but the corresponding decrease in LR was only moderate compared to Ra. When the 30 m diamond was followed by
a tungsten carbide bur (FM 3), roughness was slightly
reduced (with respect to Ra) or slightly increased (with
respect to LR), compared to FM 2, but the differences
were not statistically significant.
Figure 1: Average roughness (Ra) of one hybrid and three packable

composites after finishing (each vertical bar represents 15 specimens)
and after treatment with flexible discs (mean SD; n=5 each); the horizontal bars characterize statistically significant differences (p<0.001).
The use of Sof-Lex discs achieved the lowest roughness on all composites (p<0.001 for LR). With respect to
Ra, the differences to FM 2 and FM 3 were only significant on SureFil surfaces.
After using a 30 m diamond, there were only minor
differences among the composites evaluated. The greatest roughness data were recorded for Definite surfaces
(Ra=2.08 m; LR=1.782). After finishing according to
FM 2 and FM 3, Solitaire surfaces showed the lowest
roughness compared to the other composites. The differences were significant with respect to LR for both
finishing methods (p<0.001) and with respect to Ra only
for FM 2.
After using Sof-Lex discs, again the Solitaire surfaces
yielded the lowest roughness values (Ra=0.5 m,
LR=1.214); the differences to Definite and Herculite
surfaces were significant (p<0.001).
Figure 2: Profile-length-ratio (LR) of one hybrid and three packable composites after finishing (each vertical bar represents 15 specimens) and
after treatment with flexible discs (mean SD; n=5 each); the horizontal bars characterize statistically significant differences (p<0.001).
Qualitative Evaluation
surfaces retained areas of severe roughness amounting

to almost 42%.
The SEM evaluation largely corroborated the profilometric results. The use of a 30 m diamond caused
large areas of severe roughness on all composites
(Figure 3). Moreover, Definite specimens showed more
than 10% surface irregularities of Grade 4, indicating
the destructive effects of the 30 m diamond (Figure 4).
Using the 20 m finishing diamond (FM 2) achieved a
complete reduction of the severely roughened surfaces
to minor roughness on Solitaire specimens. The other
composites had remaining areas of severe roughness
ranging from 19 to 36%. After finishing with a tungsten
carbide bur (FM 3), the amount of severe roughness
was reduced to between 4 and 18%. Solitaire surfaces
revealed small amounts of smooth areas. The use of
Sof-Lex discs resulted in an increase in the amount of
homogeneous areas on the surface of the packable composites. Solitaire specimens yielded more than 50%
smooth areas. Despite the use of flexible discs, Definite
DISCUSSION
Surface quality is an important parameter that influences the behavior of dental restorations in the oral
environment in different ways. Rough surfaces accumulate more plaque and plaque components compared
to smooth surfaces (Kawai & Urano, 2001). The surface
state affects the fracture resistance of brittle material
such as composites and ceramics (De Jager, Feilzer &
Davidson, 2000; Graf & others, 1998). Restorations
that are well polished are less abrasive toward antagonistic surfaces and show greater wear resistance (Tjan
& Clayton, 1989). Finishing and polishing influences
the surface hardness of composites (Yap, Lye & Sau,
1997). Smooth surfaces attribute to a natural appearance of tooth-colored restorations. Polishing reduces
susceptibility to staining (Dietschi & others, 1994). The
57
To avoid the training effects caused by increasing

manual skill when finishing the composite specimens, a
randomized protocol was followed. This was done with
respect to selecting the different composites and the finishing method used.
Figure 3: The portion of different surface chararcteristics of one hybrid

and three packable composites in SEM after finishing (each bar represents 15 specimens) and disc treatment (n=5 each).
Figure 4: Areas of surface destruction () on Definite after finishing with

a 30 m diamond.
state of a surface affects the patients comfort with

respect to perception of surface irregularities (Van
Noort & Davies, 1984).
Packable composites were designed as an alternative
to amalgam for the restoration of posterior teeth. Class
I or II restorations usually include structured surfaces
caused by the presence of cusps and fissures. Flexible
discs are accepted as a clinical standard for trimming
and polishing composites (Hoelscher & others, 1998;
Tjan & Clayton, 1989; Wilson, Heath & Watts, 1990).
Because of their shape, their use is confined to convex
surfaces (Chen, Chan & Chan, 1988; Tjan & Clayton,
1989). Therefore, it was important to evaluate rigid
rotary instruments that permit the finishing of structured surfaces. Rotating stones produce rough surfaces
on composite materials and should be avoided for this
use (Yap & others, 1997). Finishing diamonds and tungsten carbide finishing burs are recommended for trimming composites (Kaplan & others, 1996; Lutz, Setcos
& Phillips, 1983); for this reason, these burs were chosen for this study.
The samples were evaluated by optical profilometry.

This permitted a touch-free scanning of the composite
surfaces. The size of the stylus tip, which is commonly
used for mechanical profilometry, is regarded as being
too large to penetrate the irregularities of finished or
polished surfaces (Joniot & others, 2000; Jung, 1997;
Whitehead & others, 1999). The laser stylus of the
Focodyn pick-up is focused to a diameter of 1 m, thus,
providing great accuracy for the profilometric evaluation.
Roughness parameters such as Rmax, Rt, Rz or Ra are
vertical parameters since they describe surface irregularities only by their amplitudes. The shortcomings of
these parameters have been pointed out in dental literature (Jung, 1997; Whitehead & others, 1995). The
authors, therefore, included the profile-length-ratio
(LR) for the characterization of roughness because this
parameter reflects both the vertical and the horizontal
dimensions of surface irregularities.
Several authors have stressed the importance of supporting quantitative evaluation of surface roughness by
qualitative methods (Goldstein & Waknine, 1989;
Northeast & van Noort, 1988; Tjan & Clayton, 1989).
For this reason, additional examination of the specimens with the SEM was performed. This enabled a discrimination between surface roughness and the
destructive effects caused by rotary instrumentation.
The roughest surfaces on all composites were caused
by the 30 m finishing diamond. The subsequent use of
a 20 m diamond reduced roughness significantly. The
corresponding decrease in LR was only moderate compared to Ra. This can be attributed to the fact that the
20 m diamond not only reduced the amplitude of profile irregularities but also increased their number. In
contrast to LR, average roughness does not reflect the
change in the number of profile peaks. LR represents
both the height of surface irregularities and their number.
For three of the four composites tested, there were no
significant differences between the use of two finishing
diamonds or the sequence of a diamond and a tungsten
carbide bur with respect to Ra and LR. The specimens
treated with two diamonds were significantly smoother
only for Solitaire. This might indicate that a tungsten
carbide finishing bur is not recommendable for use on
Solitaire surfaces.
Sof-Lex discs caused a significant reduction in roughness with respect to LR compared to each of the finishing methods. This emphasizes the necessity of a final
polishing with rigid rotary instruments after finishing.
58
Finishing, according to FM 2 and FM 3, can only
achieve an initial smoothing of roughened surfaces.
When comparing composite materials, there were no
significant differences between the hybrid composite
Herculite and the packables Definite and SureFil with
respect to surface quality. When finishing according to
FM 2 and FM 3, and (with one exception) after the use
of flexible discs, Solitaire surfaces yielded significantly
lower LR values than the other composites. These differences might arise from the special filler technology of
Solitaire. The large and porous SiO2 particles facilitate
penetration and embedding by the matrix constituents,
which could attribute to a more homogeneous surface
behavior when rotary instrumentation is performed.
Another point might be the reduced hardness of
Solitaire compared to the other composites. Clinical
studies must show whether the mechanical properties
of Solitaire will be sufficient for the restoration of posterior teeth. Currently, controversial results relating to
the clinical behavior of Solitaire have been published
(Farah & Powers, 1998; Klinge & others, 2000).
The SEM evaluation revealed destructive effects of
the 30 m diamond on Definite surfaces. After treatment with flexible discs, about 40% severe roughness
remained on Definite specimens. This indicates the
detrimental effect of rotary instruments with a large
grain size on the ormocer surface. With respect to the
filler composition, there are only minor differences to
SureFil and Herculite. Therefore, the destructive effect
of coarse rotar instrumentation might result from the
special composition of the ormocer matrix and could
indicate a weak adhesion of the filler particles and the
siloxane polymer network.
The Ra values of this study are in agreement with a
prior study on Herculite surfaces with respect to the
finishing diamonds and Sof-Lex discs (Jung, 1997).
Both studies utilized the same type of laser-stylusbased profilometry. Comparing the roughness data with
that of other authors is problematic. This arises from
the fact that there are several factors influencing the
results of quantitative roughness evaluation. One of
these factors is the type and design of the pick-up system. The Focodyn laser stylus is very precise compared
to a mechanical stylus; this can explain the fact that the
Ra values of this study are mostly greater than those of
similar studies in the literature. Pratten & Johnson
(1988) reported Ra values of 1.5 m and 0.8 m on composite surfaces after using a fine or x-fine diamond,
respectively. Berastegui & others (1992) examined
Herculite surfaces after finishing and reported an Ra of
0.7 m when using a finishing diamond and an Ra of 0.3
m after treatment with Sof-Lex discs. Of special interest are the results of a study examining the surface
quality of three packable composites (Roeder, Tate &
Powers, 2000). SureFil and Solitaire finished with a 30
Operative Dentistry
m diamond had an Ra ranging from 1.3-2.2 m; SofLex discs reduced average roughness to 0.2-0.24 m. A
tungsten carbide bur was only tested on SureFil and
caused an Ra of 0.4-0.6 m. The absolute Ra-values,
thus, are lower compared to this study, but the relation
of the results among the different finishing methods are
similar.
CONCLUSIONS
1. The use of a 30 m diamond caused a similar
roughening of the surfaces of all composites to
an Ra level of 2 m. On Definite surfaces, the
30 m diamond caused large amounts of detrimental effects and cannot be recommended for
finishing this type of composite.
2. The subsequent use of two finishing diamonds
or a finishing diamond followed by a tungsten
carbide bur reduced initial roughness significantly to more than half the amount on all
composites.
3. With respect to LR, Solitaire surfaces were significantly smoother when finishing was performed with two diamonds compared to a diamond and a tungsten carbide bur.
4. The lowest roughness values were achieved
after using discs; the Solitaire specimens yielded the lowest Ra and LR values.
5. Overall, there were only minor differences in
surface quality between the hybrid composite
Herculite and the packable composites
Definite and SureFil. Solitaire surfaces were
significantly smoother after finishing according to FM 2 and FM3 compared to the other
materials.
References
Berastegui E, Canalda C, Brau E & Miquel C (1992) Surface
roughness of finished composite resins Journal of Prosthetic
Dentistry 68(5) 742-749.
Chen HY, Manhart J, Hickel R & Kunzelmann KH (2001)
Polymerization contraction stress in light-cured packable composite resins Dental Materials 17(3) 253-259.
Chen RCS, Chan DCN & Chan KC (1988) A quantitative study of
finishing and polishing techniques for a composite Journal of
Cobb DS, MacGregor KM, Vargas MA & Denehy GE (2000) The
physical properties of packable and conventional posterior
resin-based composites: A comparison Journal of the American
De Jager N, Feilzer AJ & Davidson CL (2000) The influence of
surface roughness on porcelain strength Dental Materials
16(6) 381-388.
59
Dietschi D, Campanile G, Holz J & Meyer JM (1994) Comparison

of the color stability of 10 new-generation composites: An in
vitro study Dental Materials 10(6) 353-362.
Leinfelder KF, Bayne SC & Swift EJ (1999) Packable composites:

Overview and technical considerations Journal of Esthetic
Dentistry 11(5) 234-249.
[DIN-Standards] DIN-Normen (1996) DIN 4762 [Surface roughness;

Terminology,
surface
and
its
parameters]
Oberflchenrauhheit Begriffe Oberflche und ihre
Kenngren Identisch mit ISO 4287/1:1984 DIN Taschenbuch
106 Verzahnungsterminologie Normen (Antriebstechnik 1)
Berlin, Wien, Zrich: Beuth Verlag GmbH 4 Auflage 334-349.
Loguercio AD, Reis A, Rodrigues Filho LE & Busato ALS (2001)

One-year clinical evaluation of posterior packable resin composite restorations Operative Dentistry 26(5) 427-434.
Farah JW & Powers JM (1998) Condensable composites The

Dental Advisor 15 1-4.
Goldstein GR & Waknine S (1989) Surface roughness evaluation
of composite resin polishing techniques Quintessence
Graf A, Sindel J, Kern A & Petschelt A (1998) Influence of surface
roughness on strength of machinable materials Journal of
Dental Research 77(Special Issue B) Abstract #2480 p 941.
Hoelscher DC, Neme AML, Pink FE & Hughes PJ (1998) The
ISO-Standards (1997) ISO 4287 Geometrical Product
Specifications (GPS)Surface texture: Profile methodTerms,
definitions and surface texture parameters International
Standard ISO 4287 Geneve: International Organization for
Standardization 1. edition 1-25.
Joniot SB, Grgoire GL, Auther AM & Roques YM (2000) Threedimensional optical profilometry analysis of surface states
obtained after finishing sequences for three composite resins
Jung M (1997) Surface roughness and cutting efficiency of composite finishing instruments Operative Dentistry 22(3) 98-104.
Lutz F, Setcos JC & Phillips RW (1983) New finishing instruments for composite resins Journal of the American Dental
Manhart J, Kunzelmann KH, Chen HY & Hickel R (2000)
Mechanical properties and wear behavior of light-cured packable composite resins Dental Materials 16(1) 33-40.
Northeast SE & van Noort R (1988) Surface characteristics of finished posterior composite resins Dental Materials 4(5) 278288.
Oberlnder H, Hiller KA, Thonemann B & Schmalz G (2001)
Clinical evaluation of packable composite resins in Class-II
restorations Clinical Oral Investigations 5(2) 102-107.
Pratten DH & Johnson GH (1988) An evaluation of finishing
instruments for an anterior and a posterior composite Journal
of Prosthetic Dentistry 60(2) 154-158.
Roeder LB, Tate WH & Powers JM (2000) Effect of finishing and
polishing procedures on the surface roughness of packable
composites Operative Dentistry 25(5) 534-543.
Tjan AHL & Clayton AC (1989) The polishability of posterior
composites Journal of Prosthetic Dentistry 61(2) 138-146.
Van Noort R & Davies LG (1984) The surface finish of composite
resin restorative materials British Dental Journal 157(10)
360-364.
Whitehead SA, Shearer AC, Watts DC & Wilson NHF (1995)
Comparison of methods for measuring surface roughness of
ceramic Journal of Oral Rehabilitation 22(6) 421-427.
Kaplan BA, Goldstein GR, Vijayaraghavan TV & Nelson IK

(1996) The effect of three polishing systems on the surface
roughness of four hybrid composites: A profilometric and scanning electron microscopy study Journal of Prosthetic Dentistry
76(1) 34-38.
Whitehead SA, Shearer AC, Watts DC & Wilson NHF (1999)

Comparison of two stylus methods for measuring surface texture Dental Materials 15(2) 79-86.
Kawai K & Urano M (2001) Adherence of plaque components to

different restorative materials Operative Dentistry 26(4) 396400.
Wilson F, Heath JR & Watts DC (1990) Finishing composite

restorative materials Journal of Oral Rehabilitation 17(1) 7987.
Kelsey WP, Latta MA, Shaddy RS & Stanislav CM (2000)

Physical properties of three packable resin-composite restorative materials Operative Dentistry 25(4) 331-335.
Yap AUJ (2000) Effectiveness of polymerization in composite

restoratives claiming bulk placement: Impact of cavity depth
and exposure time Operative Dentistry 25(2) 113-120.
Klinge S, Schmidt G, Frankenberger R & Pelka M (2000) Clinical

behavior of Solitaire: 1-year results Journal of Dental Research

tooth-colored restoratives polished utilizing different polishing
systems Operative Dentistry 22(6) 260-265.
Relationship Between
Nanoleakage and Microtensile
Bond Strength at the
Resin-Dentin Interface
S Guzmn-Armstrong SR Armstrong F Qian
Clinical Relevance
The main goal of bonding a restorative material to dental tissue is to achieve a strong,
durable bond and an impervious seal. The correlation between laboratory measurements of physical properties representing clinical performance, that is, nanoleakage
(secondary caries, pulpal reactions and marginal integrity) and microtensile bond
strength(restoration retention), could not be confirmed in this study.
SUMMARY
To evaluate the correlation between microtensile
dentin bond strength (TBS) and silver ion penetration, two total-etch 3-step and one self-etch 2step system were investigated. OptiBond FL
adhesive was applied to flat occlusal dentin on
six non-carious human molars, and a resin composite crown was formed in 2 mm increments.
After 24-hour water storage, the teeth were sectioned with a low-speed diamond saw to obtain
four-square sticks (~2 mm X 2 mm) per tooth.
Cylindrical tensile test specimens were formed
with an 0.5 mm2 cross-sectional area. Nail varnish
was applied to the dentin within 0.51.0 mm of
*Sandra Guzmn-Armstrong, DDS, MS, assistant professor,
Operative Dentistry Department, The University of Iowa
Steven R Armstrong, DDS, PhD, assistant professor, Operative
Dentistry Department, The University of Iowa
Fang Qian, PhD, senior research assistant, Department of
Preventive & Community Dentistry, The University of Iowa
*Reprint request: S-229 Dental Science Building, College of
Dentistry, Department of Operative Dentistry, University of
Iowa, Iowa City, IA 52242; e-mail: [email protected]
the interface before immersing in 50% silver

nitrate for 15 minutes. Following silver fixation,
tensile testing was performed in a Zwick UTM at
1 mm/minute using a passive gripping fixture to
obtain 72-hour TBS [23.9 MPa]. The percentage
area of silver penetration was measured on
debonded specimens using light microscopy and
Image-Pro Plus Software [89%]. The procedures
were repeated using Scotchbond Multi-Purpose
Plus [TBS =27.8 MPa; nanoleakage = 67%] and
Clearfil SE bond [TBS = 36 MPa; nanoleakage =
55%]. No significant correlation between
microtensile bond strength and nanoleakage was
found for all systems. A weak-to-moderate negative relationship was found between TBS and
nanoleakage for OptiBond FL (Spearman r =
-0.3844). No correlation was found for the remaining adhesive systems. The correlation between
these two common laboratory measurements
appears to be adhesive-system dependent.
INTRODUCTION
High bond strength and a complete marginal seal at the
resin composite-dentin bonded interface are required
for clinically successful restorations. In vitro measurements of dentin bond strength and marginal integrity
Guzmn-Armstrong, Armstrong & Qian: Relationship Between Nanoleakage and Bond Strength
are commonly viewed as indicators of adhesive system
potential and are used by clinicians as important selection criteria for dental adhesives. The relationship
between bond strength and leakage is not clearly
understood.
Countless studies have individually evaluated marginal integrity or bond strength of the adhesive dentinbonded interface; however, few studies have evaluated
the relationship between these two different outcomes
(Prati & others, 1992; Retief, Mandras & Russell, 1994;
Fortin & others, 1994; Pereira & others, 2001; Paul &
others, 1999; Neme, Evans & Maxson, 2000).
Munksgaard, Irie & Asmussen (1985) demonstrated an
inverse relationship between shear bond strength and
gap formation, while Retief & others (1994) found an
inverse relationship between shear bond strength and
microleakage. However, many more studies have failed
to demonstrate a correlation between SBS and marginal gap or microleakage (Kemp-Scholte & Davidson,
1990a,b; Uno & Finger, 1995; Finger & Fritz, 1996;
Prati & others, 1992; Hammesfahr, Huang & Sjaffer,
1987; Fortin & others, 1994).
In general, these studies have demonstrated a tendency toward high bond strength related to low leakage
but were rarely statistically significant. Staninec &
Kawakami (1993) concluded that the amount of leakage observed was correlated to early shear bond
strength. Fortin & others (1994) also evaluated
microleakage and bond strength using the same tooth
but different specimens. However, the trend was materials with high bond strength also had the lowest
microleakage. This study is in agreement with Neme &
others (2000), who demonstrated an inconsistent relationship between the two methodologies. In a different
study, Paul & others (1999) found no correlation
between bond strength and etching time.
The correlation between leakage and bond strength is
most appropriately determined within the same test
specimen. Prati & others (1992) found that high bond
strength was associated with low microleakage and
vice versa. Pereira & others (2001) investigated the
relationship between TBS and nanoleakage on alternate specimens within the same tooth and found no significant correlation between the two tests. Okuda &
others (2001a), testing both TBS and nanoleakage on
the same specimen, confirmed that a correlation existed
only for one adhesive system (self-etch two-step) after
three, six and nine months of storage. This same group
also reported no correlation for two total-etch two-step
systems (Okuda & others, 2001b). Also, Guzmn-Ruiz
& others (2001) found no association between leakage
and bond strength using the same specimen in Class II
indirect resin composite restorations.
This study determined the correlation between
microtensile bond strength (TBS) and silver ion leak-
61
age (% area) in the same specimen for two total-etch 3step and one self-etch 2-step system.
Six intact, non-carious extracted human third molars
were stored in 2% thymol at 4C solution for less than
two months. Each tooth was hand scaled, then placed
in water for at least 24 hours prior to mounting in 1x1
inch dental stone blocks using a custom fabricated
tooth-mounting device. Flat occlusal dentin was prepared with constant water spray at equivalent speed
rates using a #56 carbide bur mounted in the
University of Iowa Microspecimen Former. Light
microscopy was used to verify removal of all enamel
remnants. Immediately after preparation, the specimens were restored with OptiBond FL (Kerr
Corporation, Orange, CA 92867, USA) adhesive resin
according to manufacturers instructions (Table 1). A
resin composite (Prodigy, Kerr, Danbury, CT 06810,
USA) crown of at least 6 mm in height was formed in
2 mm increments. The resin composite was built up
freehand and each increment was light cured for 40
seconds using Optilux 400 (Demetron/Kerr, Danbury,
CT 06810, USA) light curing unit. The output for the
curing light unit was verified at >400 mW/cm2 and the
laboratory conditions were 221C and 541% relative
humidity throughout the bonding procedure. After
storage in water for 24 hours, the mounted
crown/teeth were sectioned with a low-speed diamond saw (Isomet 1000, Buehler, Lake Bluff, IL 60044,
USA) to obtain four square sticks (~2 mm x 2 mm x 1012 mm) per tooth. Each stick was trimmed in the Iowa
Microspecimen Former using an eight micron diamond
cutting instrument in a high-speed hand piece creating
cylindrical tensile test specimens with 0.5 mm2 cross
sectional area and a 2 mm gage length. The samples
were stored in water in individual containers for 48
hours. Nail varnish was applied to the dentin within
0.5 mm1 mm of the adhesive interface before immersing in 50% silver nitrate in a light-proof container for
15 minutes, then rinsed with water for five minutes
and placed in a photodeveloper solution for 12 hours to
precipitate silver ions in areas of leakage. Following
silver fixation, 72-hour microtensile bond strength was
determined in a universal testing machine (Zwick
1445 Materials Testing Machine, Zwick GmbH & Co,
Ulm, Germany) and a Dircks passive gripping fixture
at 1 mm/minute until failure.
The percentage area of silver penetration was determined at the interface in all debonded specimens using
light microscopy at 0.6X x 4.0X (Olympus BX-50,
Japan) and Image-Pro Plus Software.
The study was repeated with a different total-etch
three-step adhesive system, (Scotchbond MultiPurpose, 3M, St Paul, MN 55144, USA) using five
62
Operative Dentistry
Table 1: Dental Adhesive Systems

Material
Components
Batch #
Manufacturer
Clinical Steps
Optibond FL
Etch: 37.5% Phosphoric Acid
001642
Kerr
15 seconds, rinse, leave moist
Primer: HEMA, GPDM, mono

(2-methacryloxy ethyl) phthalate
(PAMM), ethyl alcohol, CQ, and water
25881
10 seconds, gentle agitation, air-dry gently

5 seconds
Adhesive: BIS-GMA, HEMA,
25882
Brush application, Light cure 20 seconds
barium aluminum borosilicate glass,

fumed silica, disodium hexafluorosilicate,
glycerol dimethacrylate, and CQ
ScotchBond
Multi-Purpose
Clearfil SE
Bond
Etch: 35% Phosphoric Acid
7523
3M-ESPE
Primer: Water, HEMA, and polycarboxylic

acid copolymer
7542
20 seconds, scrubbing, add primer

every 5 seconds, air-dry gently 5 seconds
Adhesive: BIS-GMA, HEMA, CQ,

EDMAB, DHEPT
7523
Brush application
Light cure 10 seconds
Primer: 10-MDP, HEMA, DHEPT,

hydrophilic dimethacrylate, CQ, water
00110A
Adhesive: 10-MDP, HEMA, BIS-GMA,

hydropholic dimethacrylate, CQ, DHEPT,
silanated colloidal silica
00046B
Kuraray Co Ltd
15 seconds, rinse, leave moist
Apply and allow to stand for 20 seconds

Air-dry gently
Apply and gently air thin
Light cure 10 seconds
Abbreviations: CQ = camphorquinone, DHEPT = N,N-diethanol p-toluidine; EDMAB = ethyl 4-dimethyl amino benzoate; HEMA = 2-hydroxyethylmethacrylate; GPDM = glycerol phosphate
dimethacryate; BIS-GMA = bisphenyl glycidyl methacrylate; 10-MDP = 10-Methacryloyloxydecyl dihydrogen phosphate
Table 2: Means, Standard Deviations and Results of Correlation Between Microtensile Bond Strength and
Nanoleakage Over Each Adhesive System Tested
Dental Adhesive
System
TBS
MPa
Ag+
Penetration %
Spearman
Correlation
P-Value
Optibond FL
(Kerr)
22
23.9(10.3)
89(17)
-0.3844
0.0773
ScotchBond
Multi-Purpose
(3M ESPE)
15
27.8(18.4)
67(23)
0.0107
0.9697
Clearfil SE Bond
(Kuraray)
16
36.0(16.7)
55(32)
0.06825
0.8017
teeth and a self-etch, two-step adhesive (Clearfil SE,

Kuraray, Japan) using four teeth.
STATISTICAL ANALYSIS
SAS software was used to conduct the data analysis.
Spearman correlation was used to test whether there
were any apparent increasing or decreasing relationships between microtensile bond strength and silver ion
penetration for each adhesive system. All tests have a
0.05 level of statistical significance.
RESULTS
In the OptiBond FL group, one specimen fractured during TBS specimen preparation and one specimen
failed in resin composite during TBS testing, therefore, the leakage percentage area could not be measured. Also, four specimens fractured during specimen
preparation in the Scotchbond Multi-Purpose group
and one specimen failed in the dentin substrate during
TBS testing, and leakage area, again, could not be
measured.
Table 2 reports the means, standard deviations and

correlation statistics for TBS and leakage over each
adhesive system tested. Based on the Spearmans correlation test, the data showed no evidence of significant
relationships between microtensile bond strength and
leakage for each of the three adhesive systems:
Scotchbond Multi-Purpose (p-value=0.9697), Clearfil
SE (p-value=0.8002) and Optibond FL (pvalue=0.0773). The highest TBS=36.0 (16.7) and the
lowest leakage=55% (32%) were found in the Clearfil
SE adhesive system group. Figure 1 shows the relationship between TBS and silver penetration area for
each adhesive system.
The correlation coefficient r=-0.3844 between TBS
and leakage for the OptiBond FL group is relatively
high, which showed an apparent decreasing relationship between TBS and leakage at 0.10 significance
level. Further investigation was performed based on
the distribution of percentage leakage areas. Twentytwo specimens were divided into two groups: seven with
47%-86% leakage (Group 1) and 15 with 100% leakage
63
good, long lasting seal. It has been suggested that the

bond between tooth and adhesive material could persist
in spite of minimal leakage (Neme & others, 2000).
However, bond strength alone may not adequately
address issues more directly related to microleakage,
such as secondary caries, pulpal reactions and marginal integrity.
If both bond strength and microleakage tests are
needed to best predict the clinical behavior of an adhesive dental material, some correlation of data is expected since both methodologies are intended to evaluate
marginal integrity. In fact, such a relationship has been
reported to be inversely related and predictive of clinical performance (Prati & others, 1992; Retief & others,
1994; Fortin & others, 1994; Neme & others, 2000;
Okuda & others, 2001a).
No correlation between silver penetration area and
bond strength was demonstrated in this study using
Scotchbond Multi-Purpose, Clearfil SE bond and
OptiBond FL dental adhesive systems. However, an
inverse relationship of the two measured variables was
weak-to-moderate at best (Spearman=-0.3844) when
OptiBond FL was used. This adhesive system showed
the lowest bond strength and the highest leakage.
These findings are dissimilar to the results presented
by Fortin & others (1994) that showed using OptiBond,
produced the highest bond strength and the lowest
leakage within all adhesives tested. These results could
differ due to test methods and/or operator variability
(Sano & others, 1998; Finger & Balkenhol, 1999).
However, in general, both studies demonstrated that
bond strength and microleakage were related, showing
an inverse relationship between the methodologies
where the dental adhesive material with the highest
bond strength had the lowest leakage.
DISCUSSION
The relationship between bond strength and leakage

is complex and poorly understood, as demonstrated by
the limited number of publications dedicated to the
topic. Prati & others (1992) reported a negative correlation between bond strength and microleakage using
Scotchbond Dual-Cure and Scotchbond 2, suggesting
that high bond strength was associated with low leakage. The method for measuring microleakage was
based on fluid filtration rather than dye penetration.
Staninec & Kawakami (1993) found that the amount of
leakage observed was correlated to early shear bond
strength. Also, in a previous study, the authors evaluated bond strength and leakage in the same specimen
using indirect resin composite restorations. No correlation was found; however, a feasible method for evaluation of both tests at the joint interface within the same
specimen was demonstrated (Guzmn-Ruiz & others,
2001).
The main goal of bonding a restorative material to

dental tissue is to achieve high bond strength and a
On the other hand, Neme & others (2000) evaluated

dentin and enamel bonded interface using amalgam
Figure 1: X-Y plot of mTBS (MPa) and silver penetration (%) with regression line for adhesive systems tested.
(Group 2). The Wilcoxon Rank-Sum test revealed a significant difference between mean levels of bond
strength in the two groups with one-sided pvalue=0.046. The mean strength 29.28 for Group 1 with
lower percentage leakage areas is significantly higher
than the mean strength 21.37 for Group 2 with higher
percentage leakage areas.
64
and resin composite. They found an inconsistent relationship between bond strength and microleakage
using different specimens for each test. The results also
showed that using an adhesive system would both
increase the bond strength and decrease microleakage.
These results were confirmed in this study when it was
observed that the group with the higher bond strength
also had the lower leakage percentage area and vise
verse. They also highlighted the importance of additional investigations comparing methodologies for both
bond strength and microleakage evaluation.
Long-term storage is also considered an important factor
for evaluation of these two tests. Some studies have
demonstrated a progressive decline in bond strength
over time (Kiyomura, 1987; Sano & others, 1994, 1999) and
an increase in microleakage (Grieve, Saunders & Alani,
1993; Crim, 1993; Haller & others, 1993). Paul & others
(1999) suggested that microleakage might rise over
time, caused by the slow hydrolytic degradation of the
resin and the collagen fibers in the submicron spaces of
the hybrid layer. A distinction between microleakage and
nanoleakage and the effects of each on long-term bond
stability was recently reviewed by Pioch & others (2001).
Staninec & Kawakami (1993) evaluated bond
strength using different adhesive systems and some of
the groups showed an increase in bond strength over
time (three minutes, one hour, 24 hours). The increases
may be due to further polymerization at the interface or
stress relaxation by hygroscopic expansion of the composite. Early bond strength is particularly important
when the restorative margins are placed under stress
(resin composite polymerization shrinkage stress, contouring and finishing, masticatory loading and thermal
fatigue). As mentioned, short-term studies failed to
demonstrate a relationship between nanoleakage and
TBS (Pereira & others, 2001), whereas, longer-term
studies found an apparent adhesive system-dependent
relationship (Okuda & others, 2001a; 2001b). Other
studies have shown deterioration at the resin-dentin
bonded interface due to hydrolytic degradation over
long-term storage (one year) (Blunk & Roulet, 1999).
More research is indicated on degradational progresses
within the dentin-adhesively-bonded interface.
Pashley & others (1999) concluded that the microtensile bond testing methods give great potential for providing insight into the strength of adhesion dental
restorative materials. DeHoff, Anusavise & Wang
(1995) state that most dental researchers use tensile
and shear tests to predict the effects of technique and
material variable on clinical performance of bonding
systems, though there is no evidence of clinical relevance. In this study, Clearfil SE bond adhesive system
showed the highest bond strength 36 (16.7) MPa. It has
been reported that self-etching primer adhesive systems produce high bond strength to normal dentin
(Yoshiyawa & others, 1998, 1999; Pereira & others,
Operative Dentistry
1999) theoretically due to simultaneous collagen fiber
network exposure and monomer infiltration, which
may create a sufficient retentive strength and an adequate seal. Whether this ideal can be achieved remains
to be determined through long-term clinical trials.
The relative low bond strength obtained with
OptiBond FL and Scotchbond Multi-Purpose groups
may be due to incomplete removal of all water/solvent
within the interdiffusion zone, possibly interfering with
the polymerization of the resin. OptiBond FL, an intermediately filled adhesive resin layer (48% filled adhesive), had the lowest bond strength 23.9 (10.3) MPa in
this study, in contrast with other laboratories findings
(Bouillaguet & others, 2001; Prati, Chersoni & Pashley,
1999; Wilder & others, 1998). This study is in general
agreement with Tanumiharja, Burrow & Tyas (2000)
who evaluated microtensile bond strength of several
conventional and self-etching primer systems. They
concluded that generally the self-etching priming systems had higher bond strengths than the other threestep adhesive systems. However, laboratory bond
strength (Miyazaki & others, 1995), microleakage
(Fortin & others, 1994) and clinical (Van Meerbeek &
others, 1994; Alhadainy & Abdalla, 1996) studies all
provided evidence for the theory that intermediatelyfilled adhesive resin system, by providing an elastic
buffer zone, may be superior to unfilled systems.
Sano & others (1994; 1995a,b) reported the presence
of nanometer-sized spaces that permitted silver nitrate
to penetrate the resin-dentin interface. They explain
that the porosity may be the result of an incomplete
resin infiltration into the demineralized dentin, poor
polymerization of the adhesive resin and the existence
of low-molecular-weight oligomers that allow water to
penetrate the bonded interface. According to the results
of this study, Clearfil SE Bond presented low leakage
penetration. It is presumed that no gap or voids exist
since the resin infiltration into the collagen fibers occur
simultaneously to the same depth of the demineralized
dentin when self-etching primer adhesive system is
used. The acidic conditioning of these self-etching
primer systems dissolves the smear layer and incorporates it into the primers and the demineralized dentin
(Nishida & others, 1993). Self-etching adhesive systems
may efficiently penetrate both dry and wet dentin. This
may explain why Clearfil SE demonstrated lower leakage penetration and the three-step adhesive system
showed higher leakage penetration.
Interestingly, the main reasons that could contribute
to high bond strength are similar to the reasons that
may contribute to a low leakage penetration, that is,
adequate penetration into the demineralized dentin,
tolerance to dry and wet dentin substrates, residual solvent, extent of polymerization, gap and void formation.
Intuitively, a relationship between bond strength and
leakage is expected.
Previous studies suggest the importance of evaluating
both tests using the same specimen to predict the clinical
performance of dental adhesive systems. During this
study several factors were not evaluated and could have
led to different findings: variation in the dentin substrate, (location, degree of demineralization, wetness,
amount of solid substrate, caries affected dentin; longterm storage, thermal stress, simulated occlusal loading or tooth flexure, simulated pulpal pressure, threedimensional cavity preparations). No clear relationship
between bond strength and marginal leakage was
demonstrated under the condition of this study.
CONCLUSIONS
To date, no strong evidence exists demonstrating a reasonable relationship between these extremely common
laboratory assessments of dental adhesive performance; however, the possibility that a relationship exists
cannot be discounted. This study demonstrates the feasibility of measuring both nanoleakage and microtensile bond strength of dentin bonding systems in the
same specimen. Future long-term studies should focus
on developing these methods to better investigate this
relationship.
References
Alhadainy HA & Abdalla AI (1996) 2-year clinical evaluation of
dentin bonding systems American Journal of Dentistry 9(2) 77-79.
Bouillaguet S, Gysi P, Wataha JC, Ciucchi B, Cattani M, Godin C
& Meyer JM (2001) Bond strength of composite to dentin using
conventional, one-step, and self-etching adhesive systems
Blunk U & Roulet JF (1999) Effect of one-year water storage on
the effectiveness of dentin adhesives in Class V resin composite restorations Journal of Dental Research 78(Special Issue)
Crim G (1993) Effect of aging on microleakage of restorative systems American Journal of Dentistry 6(4) 192-194.
DeHoff PH, Anusavice KJ & Wang Z (1995) Three-dimensional
finite element analysis of the shear bond test Dental Materials
11(2) 126-131.
Finger WJ & Fritz U (1996) Laboratory evaluation of one-component enamel/dentin bonding agents American Journal of
Dentistry 9(5) 206-210.
Finger WJ & Balkenhol M (1999) Practitioner variability effects
on dentin bonding with an acetone-based one-bottle adhesive
Journal of Adhesive Dentistry 1(4) 311-314.
Fortin D, Swift EJ Jr, Denehy G & JW Reinhardt (1994) Bond
strength and microleakage of current dentin adhesives Dental
Materials 10(4) 253-258.
Grieve AR, Saunders WP & Alani AH (1993) The effects of dentin
bonding agents on marginal leakage of composite restora-
65
tionslong term studies Journal of Oral Rehabilitation 20(1)

11-18.
Guzmn-Ruiz S, Armstrong SR, Cobb DS & Vargas MA (2001)
Association between microtensile bond strength and leakage in
the indirect resin composite/dentin adhesively bonded joint
Haller B, Hofmann N, Klaiber B & Bloching U (1993) Effect of
storage media on microleakage of five dentin bonding agents
Hammesfahr PD, Huang CT & Sjaffer SE (1987) Microleakage
and bond strength of resin restorations with various bonding
agents Dental Materials 3(4) 194-199.
Kemp-Scholte CM & Davidson CL (1990a) Marginal integrity
related to bond strength and strain capacity of composite resin
restorative systems Journal of Prosthetic Dentistry 64(6) 658664.
Kemp-Scholte CM & Davidson CL (1990b) Complete marginal
seal of Class V resin composite restorations effected by
increased flexibility Journal of Dental Research 69(6) 12401243.
Kiyomura M (1987) Bonding strength to bovine dentin with 4
META/MMA-TBB resin: Long term stability and influence of
water Dental Materials 6 860-872.
Miyazaki M, Ando S, Hinoura K, Onose H & Moore BK (1995)
Influence of filler addition to bonding agents on shear bond
strength to bovine dentin Dental Materials 11(4) 234-238.
Munksgaard EC, Irie M & Asmussen E (1985) Dentin-polymer
bond promoted by gluma and various resin Journal of Dental
Research 64(12) 1409-1411.
Neme AL, Evans DB & Maxson BB (2000) Evaluation of dental
adhesive systems with amalgam and resin composite restorations: Comparison of microleakage and bond strength results
Nishida K, Yamauchi J, Wada T & Hosoda H (1993) Development
of a new bonding system Journal of Dental Research
Okuda M, Pereira PNR, Yamaguchi S, Nakajima M & Tagami J
(2001a) Relationship between nanoleakage and long-term
durability of dentin bond Journal of Dental Research
Okuda M, Pereira PNR, Nakajima M & Tagami J (2001b)
Relationship between nanoleakage and long-term durability of
dentin bonds Operative Dentistry 26(5) 482-490.
Pashley DH, Carvalho R, Sano H, Nakajima M, Yoshiyama M,
Shono Y, Fernandes CA & Tay F (1999) The microtensile bond
test: A review Journal of Adhesive Dentistry 1(4) 299-309.
Paul SJ, Welter DA, Ghazi M & Pashley D (1999) Nanoleakage at
the dentin adhesive interface vs -tensile bond strength
Pereira PNR, Okuda M, Nakajima M, Sano H, Tagami J &
Pashley DH (2001) Relationship between bond strengths and
nanoleakage: Evaluation of a new assessment method
Pereira PNR, Okuda M, Sano H, Yoshikawa T, Burrow MF &
Tagami J (1999) Effects of intrinsic wetness and regional difference on dentin bond strength Dental Materials 15(1) 46-53.
66
Pioch T, Staehle HJ, Duschner H & Garca-Godoy F (2001)
Nanoleakage at the composite-dentin interface: A review
Prati C, Simpson M, Mitchem J, Tao L & Pashley DH (1992)
Relationship between bond strength and microleakage measured in the same Class I restorations Dental Materials 8(1) 37-41.
Prati C, Chersoni S & Pashley DH (1999) Effect of removal of surface collagen fibrils on resin-dentin bonding Dental Materials
15(5) 323-331.
Retief DH, Mandras RS & Russell CM (1994) Shear bond
strength required to prevent microleakage at the
dentin/restoration interface American Journal of Dentistry
7(1) 44-46.
Sano H, Ciucchi B, Matthews WG & Pashley DH (1994) Tensile
properties of mineralized and demineralized human and
bovine dentin Journal of Dental Research 73(16) 1205-1211.
Sano H, Takatsu T, Ciucchi B, Horner IA, Matthews WC &
Pashley DH (1995a) Nanoleakage: Leakage within the hybrid
layer Operative Dentistry 20(1) 18-25.
Operative Dentistry
Sano H, Yoshikawa T, Pereira PNR, Kanemura N, Morigami M,
Tagami J & Pashley DH (1999) Long-term durability of dentin
bonds made with a self-etching primer, in vivo Journal of
Staninec M & Kawakami M (1993) Adhesion and microleakage
tests of a new dentin bonding system Dental Materials 9(3)
204-208.
Tanumiharja M, Burrow MF & Tyas M (2000) Microtensile bond
strength of seven dentin adhesive systems Dental Materials
16(3) 180-187.
Uno S & Finger WJ (1995) Function of the hybrid zone as a
stress-absorbing layer in resin-dentin bonding Quintessence
Van Meerbeek B, Peumans M, Verschueren M, Gladys S, Braem
M, Lambrechts P & Vanherle G (1994) Clinical status of ten
dentin adhesive systems Journal of Dental Research 73(11)
1690-1702.
Wilder AD Jr, Swift EJ Jr, May KN Jr & Waddell SL (1998) Bond
strengths of conventional and simplified bonding systems
Sano H, Yoshiyama M, Ebisu S, Burrow MF, Takatsu T, Ciucchi

B, Carvalho R & Pashley DH (1995b) Comparative SEM and
TEM observations of nanoleakage within the hybrid layer
Yoshikawa T, Sano H, Burrow MF, Tagami J & Pashley DH

(1999) Effects of dentin depth and cavity configuration on bond
strength Journal of Dental Research 78(4) 898-905.
Sano H, Kanemura N, Burrow MF, Inai N, Yamada T & Tagami

J (1998) Effect of operator variability on dentin adhesion:
Students vs dentists Dental Materials 17(1) 51-58.
Yoshiyama M, Matsuo T, Ebisu S & Pashley DH (1998) Regional

bond strength of self-etching/self-priming adhesive systems
Effects of Er:YAG and

Nd:YAP Laser Irradiation
on the Surface Roughness and
Free Surface Energy of Enamel
and Dentin: An In Vitro Study
V Armengol O Laboux P Weiss
A Jean H Hamel
Clinical Relevance
This study investigated the morphological changes and free surface energy of enamel and
dentin following laser irradiation and their role in the resin-composite adhesion process.
SUMMARY
Sixty-seven extracted molars were selected (134
samples). Dentin and enamel samples were prepared by buccal and lingual surface sectioning to
expose a planar enamel or dentin surface.
For the free-surface energy study, 54 samples

received the same treatment as above. Two contact angle measurements were made on each surface using a goniometer. Data were analyzed by a
non-parametric statistical test.
For the roughness study, 80 samples were randomly assigned to eight groups. Enamel and
dentin surfaces were etched with a 37% phosphoric acid solution, irradiated with an Er:YAG
laser or irradiated with a Nd:YAP laser. Samples
were then observed in SEM using BSE.
Morphological changes on enamel and dentin

were greater with acid-etch and Er:YAG laser
than with Nd:YAP laser. Free surface energy was
significantly greater with acid-etch or Er:YAG
laser than with Nd:YAP laser (p<0.001).
*Valerie Armengol, DDS, PhD, Centre de Recherche sur les

Matriaux dIntrt Biologique, Dpartement dOdontologie
Conservatrice-Endodontie, France
Lasers have been proposed for use in clinical dentistry

since 1960, and several studies have investigated
applications for removal of dental hard tissues prior to
applying restorative materials. These studies showed
that the effects of laser irradiation are dependent on
wavelength specificity and energy density. As the laser
energy absorbed by dental tissues produces surface
modifications, it has been suggested that lasers could
be used for the pretreatment of enamel and dentin to
enhance the bonding of restorative adhesive material
(Khan & others, 1998; Niu & others, 1998; Visuri & others, 1996).
O Laboux, DDS, PhD, Centre de Recherche sur les Matriaux

dIntrt
Biologique,
Dpartement
dOdontologie
P Weiss, DDS, PhD, Centre de Recherche sur les Matriaux
dIntrt
Biologique,
Dpartement
dOdontologie
A Jean, DDS, PhD, professor, Centre de Recherche sur les
Matriaux dIntrt Biologique, Dpartement BiomatriauxBiophysique, France
H Hamel, DDS, professor, Centre de Recherche sur les
Matriaux dIntrt Biologique, Dpartement dOdontologie
*Reprint request: UFR dOdontologie, 1 Place Alexis Ricordeau, 44042
Nantes Cedex 01, France; e-mail: [email protected]
INTRODUCTION
During the bonding process, roughness and free surface energy play a key role by interacting with each
other. The adhesion of composite materials depends on
the adhesives ability to spread on a surface, which is
referred to as wettability. To improve wettability and
68
Operative Dentistry
bonding, the free surface energy (or critical surface tension) of enamel and dentin must be increased, which
involves using conditioners and primers (Attal,
Asmussen & Degrange, 1994; Buonocore, 1955;
Erickson, 1992; Pashley, 1990; Van Meerbeek & others,
1992). Surface topography also plays an important role.
A more or less roughened surface is indicative of the
degree to which mechanical anchorage is involved in
the bonding process. Buonocore (1955) showed that
acid-etch of enamel increases resin retention, that is,
the micromechanical component is essential in adhesion to enamel. On dentin, acid-etch causes physicochemical changes in the surface conducive to micromechanical and possibly chemical attachment to a
dentin bonding. The most effective approach appears to
be the formation of a dentin-resin inter-diffusion zone
or hybrid layer (Erickson, 1992; Pashley, 1990; Van
Meerbeek & others, 1992).
Many studies, often with conflicting results, have
evaluated the shear bond strength and microleakage of
composite restorations on enamel and dentin surfaces
treated by laser irradiation compared to acid-etch.
However, no studies have yet assessed the roughness
and free surface energy of lased enamel and dentin relative to the effects obtained in mechanical bond
strength and microleakage studies.
This study determined the influence of surface roughness and free surface energy on the adhesion process
and elucidated the results obtained in previous studies
(Armengol & others, 1999; Armengol & others, 2002).
The teeth were then rinsed in a distilled water ultrasonic system for five minutes and stored in physiological serum at 37C until treatment (24 hours).
For the laser treatment, only one set of irradiation
parameters (Er:YAG = 200mJ and 4Hz on enamel,
140mJ and 4Hz on dentin; Nd:YAP = 310mJ and 10Hz
on enamel, 240mJ and 10Hz on dentin) was chosen
based on pilot data specifying the energies that would
cause no discoloration, charring or cracks.
Roughness Study
Forty teeth were randomly distributed into eight
groups. Groups 1, 2, 3 and 4 included 10 enamel surfaces, and Groups 5, 6, 7 and 8 included 10 dentin surfaces. Treatment for the groups was as follows:
Groups 1 and 5:
Samples were gently air dried and a 37% phosphoric
acid gel (ref 60615208, Dentsply De Trey conditioner 36,
France) was placed on the enamel and dentin surfaces
for 30 seconds and 15 seconds, respectively. Samples
were then thoroughly rinsed with water and air dried.
Groups 2 and 6:
Surfaces were irradiated by an Er:YAG laser (Key laser
1242, KaVo, Germany) at 200 mJ and 4 Hz on enamel
and 140 mJ and 4 Hz on dentin for 12.5 seconds (50
pulses). Energy density was 83.16 J/cm2 for enamel and
72.76 J/cm2 for dentin. Laser irradiation scanned the
surface with a perpendicular orientation at a focal distance of 10 mm. Irradiation was performed through a
metallic matrix 0.5 mm thick with a central opening 3.5
mm in diameter.
Specimen Preparation
Groups 3 and 7:
Sixty-seven freshly extracted, healthy (without caries

and restoration-free) third molars were cleaned with a
rotary brush and pumice, stored in a 1% aqueous solution of chloramine-T at 4C for one week (decontamination), then stored in physiological serum at 4C until
experimentation (one to eight weeks).
Surfaces were irradiated with an Nd:YAP (LokkiDt,

France) at 310 mJ and 10 Hz on enamel and 240 mJ
and 10 Hz on dentin for 12.5 seconds (100 pulses).
Energy density was 322 J/cm2 for enamel and 299.37
J/cm2 for dentin. The irradiation laser was moved above
the surface in contact mode and in tangential position
for the optic fiber/surfaces with an angle of 45C (fiber
diameter=320 m).
The teeth were inserted into a resin block (self-curing

methacrylate resin, Meliodent, Gema, ACD, Toulon,
France) in a cubical aluminum mold 17 mm wide and
14 mm high (inner dimensions) leaving the upper portion of the crowns exposed. Lingual and buccal faces
were as parallel as possible to the respective sides of the
cube. The general axis of the tooth was parallel to the
height of the block.
Specific sample preparation was performed on a
bench that allowed reproducible positioning of samples
and the handpiece. A standard flat, reproducible surface was produced on buccal and lingual faces using a
cylindrical diamond bur on enamel (ref 837 KR 314 014,
Komet Germany) and a cylindrical carbide bur on
dentin (ref H 21 L 014, Komet Germany). These burs
were used in clinical practice for cavity preparation.
Groups 4 and 8:
These control samples received no acid-etch or laser
treatment and were air dried.
All specimens were then prepared for scanning electron microscopy (SEM) studies (using back-scattered
electrons: BSE; JEOL JSM 6300, 10kV, 15 mm, Japan)
according to a standard technique: dehydration in
increasing ethanol solutions, embedding in methylmethacrylate and sectioning in a bucco-lingual plane
through the center of the treated surface using an
Isomet low-speed diamond saw (Isomet, low-speed saw
11-1180, Buehler Ltd, Lake Bluff, IL 60049, USA).
Specimens were polished successively with 600-, 1,200,
2,500- and 4,000-grit wet silicon-carbide sandpaper,
69
Armengol & Others: Wettability and Roughness Surface After Laser Irradiation
then rinsed copiously with water before being coated
with gold-palladium.
tical test (Kruskal-Wallis one-way analysis of variance).
Free Surface Energy Study

Twenty-seven teeth were randomly distributed into
nine groups. Groups 1, 2, 3 and 4 included six enamel
surfaces, and Groups 5, 6, 7, 8 and 9 included six dentin
surfaces. Treatments for Groups 1 to 8 were the same
as for the roughness evaluation.
In Group 9, dentin surfaces were acid-etched by a
37% phosphoric acid conditioner (Dentsply De Trey conditioner 36, ref 60615208) for 15 seconds, rinsed with
water and air dried before the primer (Scotchbond
MultiPurpose, 3M Dental Products, St Paul, MN
55144, USA) was applied with a clean brush and
allowed to sit for 30 seconds.
Free surface energies were estimated from contact
angle measurements (Center of Transfer Technology,
CTTM, Le Mans, France).
Free surface energy () possesses two components: one
(LW) is non-polar and refers to the apolar Lifshitz-Van
der Waals forces often called dispersive forces and
gives information mainly on hydrophobic interaction.
The other (AB) involves the polar interaction, acid-base
interactions often called non-dispersive forces and
refers to hydrophilic interactions.
RESULTS
Roughness Study
In the control groups, the enamel surfaces were
smooth, with relief resulting from the burs (Figure 1).
On dentin, the surfaces were also smooth and covered
by a smear layer that obliterated the tubules (Figure 2).
Acid-etch enamel specimens were roughened, showing regular, perpendicular microporosities (1 to 2 m
and 10 to 20 m in depth) (Figure 3). On dentin, phosphoric acid removed the smear layer and exposed the
open tubules to a depth of 10 m (Figure 4).
Er:YAG laser treatment caused alterations to a depth
of 70 m of the enamel surface (Figure 5). Surface profiles were irregular and had large anfractuosities. Some
enamel projection debris remained on the surface, with
cracking noted under rugosities. Dentin was similar in
appearance but the irregularities were less marked.
The volatilization of dentin differed in the irradiated
surface areas (Figure 6).
S = SLW + SAB where S is free surface energy of solid

This study investigated the wettability of enamel and
dentin by calculating their free surface energy after
applying different treatments. The wettability of a surface depends on its free surface energy and is defined
by the contact angle between the surface of the solid
and the liquid. With appropriate liquids (polar and nonpolar) and the contact angle measured, both components (dispersive and non-dispersive) of the free surface
energy could be calculated.
This experimentation consisted of depositing drops of
reference liquids on the different treated surfaces
where free surface energy was known, then measuring
the contact angles ( represented the angle of the surface of the teeth with the tangent of the drop to the
interface). The closer the contact angle was to zero, the
better the wettability.
Figure 1: Scanning electron micrograph of enamel treated with diamond

bur (original magnification x500).
Distilled water was used as the reference polar liquid

and diethyleneglycol and formamide served as the reference non-polar liquid. All contact angles were advancing, and direct measurements were conducted with a
goniometer (Kyowa Contact Angle Meter, model CAS
150, Japan). Two measurements were performed for
each surface and liquid and were processed by computer (with a software Young-Dupr/Owens-Wendt equation) to determine the free surface energy in mJ/m2.
Data were analyzed using Bartletts test for
homogeneity of variance and a non-parametric statis-
Figure 2: Scanning electron micrograph of dentin treated with carbide

bur (original magnification x500).
70
Operative Dentistry
Figure 3: Scanning electron micrograph of acid-etch enamel (original

magnification x500).
Figure 4: Scanning electron micrograph of acid-etch dentin (original

magnification x500).
Figure 5: Scanning electron micrograph of enamel treated with Er:YAG

laser (original magnification x500).
Figure 6: Scanning electron micrograph of dentin treated with Er:YAG

Figure 7: Scanning electron micrograph of enamel treated with Nd:YAP

Figure 8: Scanning electron micrograph of dentin treated with Nd:YAP

Following Nd:YAP laser treatment, enamel surfaces

showed a macrorelief similar to normal abraded enamel
except for some cracking to a depth of 1 to 3 m (Figure
7). Dentin surfaces were smooth, with a 1-m thick,
white line identified in back-scattered electron studies
as a modification of dentinal chemical structure (Figure 8).
Free Surface Energy

Mean-free surface energy and standard deviations for
each group are shown in Figures 9 and 10. Statistical
analyses are indicated in Tables 1 and 2.
71
decrease in dispersive energy and

an increase in non-dispersive
energy. Total surface energy was
significantly increased in all treatment groups compared to the control group (p<0.05), thus, enhancing surface wettability. Er:YAG
laser treatment resulted in the
greatest increase in surface energy, followed by acid-etch, acidetch/primer and Nd:YAP laser. No
significant difference was found
between acid-etch and acidetch/primer treatment (p>0.05).
DISCUSSION
Figure 9: Free surface energy on enamel (means and standard deviations).
Figure 10: Free surface energy on dentin (means and standard deviations).
Enamel Surfaces
For enamel groups, free surface energy values were
very high and standard deviations were relatively low,
corresponding to homogeneous surfaces. The main component of the surface energy appeared to be polar, corresponding to hydrophilic interactions.
The highest free surface energy was obtained following Er:YAG laser- and acid-etch treatment, and the difference was significant from other groups (p<0.05).
Nd:YAP laser-treated enamel had a free energy surface
similar to control enamel (p>0.05).
Dentin Surfaces
The hydrophilic component of free surface energy was
greater than the hydrophobic component with or without treatment. The different treatments involved a
Different factors influence the

adhesion of resin-composite to
dental tissues, namely, surface
energy and wettability, roughness
of the dental surface and adhesive
viscosity. Wettability is essential
for good bonding of an adhesive on
a surface. Benedicktsson & others
(1991) found that an increase in
critical surface tension or free surface energy of enamel and dentin
with an enamel-dentin bonding
system (use of a surface conditioner
such as acid and primer) resulted
in an increase in shear bond
strength. The extent to which a
liquid wets the enamel or dentin
surface depends on the chemical
interactions between the liquid
and the surface, physical considerations such as capillary action on
dentin and surface roughness.
However, it seems necessary to
increase these energies, that is, to
activate enamel and dentin to
enhance surface wettability and
optimize adhesion.
The free surface energy of enamel and dentin following Er:YAG or Nd:YAP laser treatment versus acid-etch
was evaluated by measuring the contact angles of three
standard liquids. The superficial tension of the adhesive was not determined, which would have required
the use of a larger quantity than was provided by 3M
Dental Products (St Paul, MN 55144, USA).
The free surface energies obtained in the control and
acid-etch groups during this studys experimentation
were higher than those reported in the literature (Attal
& others, 1994; Cognard, 1987). These differences may
have resulted from various experimental conditions:
acid-etch time, the different reference liquids used and
72
the conservation time between processing and measurements of the contact angle (which should be short to
avoid atmospheric contamination and a resulting
reduction in free energy surface). The energy differences observed between enamel and dentin could have
resulted from the difference in mineral and organic
content. Hydroxyapatite has high energy and shows
considerable reactivity.
Laser irradiation had an influence on the polar and
dispersive components of free surface energy of enamel
and dentin. Er:YAG laser irradiation significantly
increased free surface energy on enamel and dentin
compared to the untreated group, with values being
similar to those obtained after acid-etch. However,
Nd:YAP laser induced little or no change in free surface
energy on either substrate. Previous studies have also
indicated that laser treatment modifies free surface
energy. Walsh (1996) suggested that laser irradiation
on enamel produced physicochemical modifications
that influenced surface energy and wettability, decreasing tissue humidity by dehydration due to the thermal
effect of laser irradiation. Rohanizadeh, Jean & Daculsi
(1999) showed that laser irradiation on dentin induces
physicochemical changes, particularly in organic and
aqueous components that can also influence the topography and surface energy.
Regarding surface roughness, the adhesion of resincomposite has clearly been established as depending on
surface topography, that is, greater or lesser roughness
affects the degree of mechanical anchorage. Roughness
studies provide a good estimation of the developed surface of the substrate and, therefore, the contact area
with adhesive (Degrange, Attal & Theimer, 1994).
Roughness can be assessed in different ways: macroscopic observation, SEM studies, use of a profilometer,
digital texture analyses based on computer scanning
imagery and more (Arcoria & others, 1991; Arcoria,
Operative Dentistry
Lippas & Vitasek, 1993; Ariyaratnam & others, 1997;
Ariyaratnam, Wilson & Blinkhorn, 1999; Degrange &
others, 1994).
According to some authors, laser etching produces a
qualitatively different surface profile and roughness
significantly different from untreated enamel and
dentin. Laser ablation produces a variety of surface
alterations and transformations ranging from a slightly roughened surface without cracks or fissures to a
highly roughened terraced or tiered surface with occasional cracks. The enamel surface alterations with
Er:YAG laser in this study were morphologically similar to those obtained after acid-etch and agree with the
authors findings in a previous study (Armengol & others, 1999; Armengol & others, 1999). Some studies on
enamel have reported a higher surface roughness after
acid-etch than Nd:YAG laser or coaxial CO2/Nd:YAG
laser treatment (Arcoria & others, 1991; Arcoria & others, 1993). Other studies (Ariyaratnam & others, 1997;
Ariyaratnam & others, 1999) found that Nd:YAG laser
produced a roughened surface on enamel and dentin
similar to that of acid-etch. However, despite these similar values, bonding of resin composite to laser-treated
enamel or dentin was significantly poorer than to acidetched tissues.
This studys SEM observations showed that the surface roughness of Er:YAG-lased enamel and dentin was
greater than that of untreated and acid-etched specimens, whereas irregularities following Nd:YAP laser
irradiation were very limited on both enamel and
dentin.
This studys results elucidate and confirm the hypothesis concerning shear bond strength and microleakage
values obtained in previous studies (Armengol & others, 1999; Armengol & others, 2002). Er:YAG laser
induced the greatest increase in surface roughness and
free surface energy on enamel and dentin. The bonding
process should therefore have been
as effective as, if not better than,
Table 1: Statistical Test (Kruskal-Wallis) for Free Surface Energy on Enamel
that of acid-etch, which was not
Control
Acid-Etch
Er:YAG
Nd:YAP
the case. These differences may be
due to the surface topography. On
Control
p=0.009
p=0.0022
NS p=0.465
enamel, acid-etch induced the disAcid-etch
p=0.009
p=0.017
p=0.009
solution of hydroxyapatite inter- or
p=0.0022
p=0.017
p=0.0022
Er :YAG
intraprismatic substance, resulting
Nd :YAP
NS p=0.465
p=0.009
p=0.0022
in regular microporosities that increased
the surface area and
Table 2: Statistical Test (Kruskal-Wallis) for Free Surface Energy on Dentin
surface energy. On
Control
Acid-Etch
Acid-Etch +
Er:YAG
Nd:YAP
dentin,
acid-etch
Primer
removed
the
smear
Control
p=0.009
p=0.009
p=0.0022
p=0.009
layer
completely,
p=0.009
NS p=0.464
p=0.0022
p=0.009
Acid-Etch
opened tubules and
Acid + Primer
p=0.009
NS p=0.464
p=0.0022
p=0.0163
demineralized the
Er:YAG
p=0.0022
p=0.0022
p=0.0022
p=0.0022
surface layer to a
p=0.009
p=0.009
p=0.0163
p=0.0022
Nd:YAP
certain
depth.
Subsequent application of effective primers containing
hydrophilic monomers probably altered the collagenfiber arrangement, elasticity and wettability, allowing
for better penetration of the adhesive resin. Formation
of a hybrid or resin-infiltrated layer between the deeper dentin structures and filling material has been clearly established as being the most effective approach for
achieving better bonding. This transition layer offers
bonding sites for copolymerization with the resin composite restorative material due to the presence of suitable monomers inside the interdiffusion area (Attal &
others, 1994; Erickson, 1992; Van Meerbeek & others,
1992). This layer may also have a protective potential
because it blocks the normal passage of microorganisms
and toxins (Van Meerbeek & others, 1992) and allows
for a micromechanical interlocking effect.
Surface morphology following Er:YAG laser treatment showed highly roughened enamel and dentin,
with considerable relief. SEM showed evidence of
cracks and fissures. The formation of microcracks, fissures or chipped surfaces may occur because of rapid
thermal cycling of the surface during pulsed laser irradiation (Ariyaratnam & others, 1997). There was no
continuity between relief features or anfractuosities
and underlying areas, and these elements probably
weakened the surface layer. Moreover, as Er:YAG laser
removed all organic components, formation of a hybrid
or resin-impregnated dentin layer was not possible. The
penetration of resin tags into dentinal tubules contributes only slightly to overall dentin bond strength.
The involvement of intertubular dentin was, in fact, the
major element in bond stability that agreed with findings in other studies (Pashley, 1990; Tagami, Tao &
Pashley, 1990). Although the wettability of a surface
improves its adhesive characteristics, surface topography and roughness are probably the most important
factors in the adhesion process.
Nd:YAP laser treatment has little influence on free
surface energy compared to other treatments and,
therefore, does not enhance wettability. The adhesive
cannot spread adequately on enamel and dentin and
does not penetrate irregularities. Nd:YAP laser also
produces a slightly roughened surface on enamel and
dentin. The surface is covered by a thin, smooth, fused,
glaze-like surface layer. In fact, ND:YAP laser produces
superficial modifications that destroy the normal architecture. Once the collagen has disappeared, the formation of a hybrid layer is not possible. The low, free surface energy and poor retention induced by the Nd:YAP
laser accounts for the very weak shear bond strength
and sealing.
CONCLUSIONS
In summary, Er:YAG laser treatment increased free
surface energy and roughness surface, whereas,
73
Nd:YAP laser had little influence on the free surface

energy and roughness surface.
And, if these results are correlated to those of the previous studies (Armengol & others, 1999; Armengol &
others, 2002), this study revealed two important points:
an increase in free energy surface does not always lead
to an increase in shear bond strength and efficient sealing. It facilitates the spread of adhesive on tooth tissues
and intimate contact between adhesive and tissues.
This was the case for Er:YAG laser treatment. However,
micromechanical interlocking should be regarded as
the main component in the adhesion process. Also,
Er:YAG and Nd:YAP lasers, operated under the conditions described in this study, may not produce a desirable enamel and dentin surface morphology and cannot
be recommended as a viable alternative to the conventional acid-etch technique.
Acknowledgements
The authors thank Paul Pilet (Centre de recherche sur les
matriaux dintrt biologique, Service de Microscopie
Electronique Balayage, UFR Odontologie, Nantes) for his assistance in the scanning electron microscopic studies.
The authors also recognize Prof Legeay (Center of Transfer
Technology, CTTM, Le Mans, France) for his expert assistance in
the wettability studies.
References
Arcoria CJ, Lippas MG & Vitasek BA (1993) Enamel surface
roughness analysis after laser ablation and acid-etching
Arcoria CJ, Steele RE, Wagner MJ, Judy MM, Matthews JL &
Hults DF (1991) Enamel surface roughness and dental pulp
response to coaxial carbon dioxide-neodymium:YAG laser irradiation Journal of Dentistry 19(2) 85-91.
Ariyaratnam MT, Wilson MA & Blinkhorn AS (1999) An analysis
of surface roughness, surface morphology and composite/
dentin bond strength of human dentin following the application of the Nd:YAG laser Dental Materials 15(4) 223-228.
Ariyaratnam MT, Wilson MA, Mackie IC & Blinkhorn AS (1997)
A comparison of surface roughness and composite/enamel
bond strength of human enamel following the application of
the Nd:YAG laser and etching with phosphoric acid Dental
Materials 13(1) 51-55.
Armengol V, Jean A, Rohanizadeh R & Hamel H (1999) Scanning
electron microscopic analysis of diseased and healthy dental
hard tissues after Er:YAG laser irradications: In vitro study
Journal of Endodontics 25 543-546.
Armengol V, Jean A, Weiss P & Hamel H (1999) Comparative in
vitro study of the bond strength of composite to enamel and
dentine obtained with laser irradiation or acid-etch Lasers in
Medical Science 14 207-215.
74
Operative Dentistry
Armengol V, Jean A, Enkel B, Assoumou M & Hamel H (2002)

Microleakage of Class V composite restorations following
Er:YAG and Nd:YAP laser irradiation compared to acid-etch:
An in vitro study Laser in Medical Science 17(2) 93-100.
Niu W, Eto J, Kimura Y, Takeda FH & Matsumoto K (1998) A

study on microleakage after resin filling of Class V cavities
prepared by Er:YAG laser Journal of Clinical Laser Medicine
& Surgery 16(4) 227-231.
Attal JP, Asmussen E & Degrange M (1994) Effects of surface

treatment on the free surface energy of dentin Dental
Materials 10(4) 259-264.
Pashley DH (1990) Interaction of dental materials with dentin

Trans Academic Dental Materials 3 55-73.
Buonocore MG (1955) A simple method of increasing the adhesion of acrylic filling materials to enamel surfaces Journal of
Dental Research 34 849-853.
Benedicktsson S, Retief DH, Russel CM & Mandras RS (1991)
Critical surface tension of wetting of dentin Journal of Dental
Research 70(Special Issue) Abstract #777 p 362.
Rohanizadeh R, Jean A & Daculsi G (1999) Effects of Q:Switched

Nd:YAG laser on calcified tissues Laser in Medical Science 14
221-227.
Tagami J, Tao L & Pashley DH (1990) Correlation among dentin
depth permeability, and bond strength of adhesive resins
Cognard J (1987) [La couche atmosphrique: Approche de la surface relle des solides] Journal of Chemical Physics 84 357362.
Van Meerbeek B, Inokoshi S, Braem M, Lambrechts P &

Vanherle G (1992) Morphological aspects of the resin-dentin
interdiffusion zone with different dentin adhesive systems
Degrange M, Attal JP & Theimer K (1994) [Aspects fondamentaux du collage appliqus la dentisterie adhesive] Realits
Cliniques 5 371-382.
Visuri SR, Gilbert JL, Wright DD, Wigdor HA & Walsh JT (1996)
Shear strength of composite bonded to Er:YAG laser-prepared
dentin Journal of Dental Research 75(1) 599-605.
Erickson RL (1992) Surface interactions of dentin adhesive

materials Operative Dentistry Supplement 5 81-94.
Walsh LJ (1996) Split mouth study of sealant retention with carbon dioxide laser versus acid-etch conditioning Australian
Dental Journal 41(2) 124-127.
Khan MFR, Yonaga K, Kimura Y, Funato A & Matsumoto K

(1998) Study of microleakage at Class I cavities prepared by
Er:YAG laser using three types of restorative materials
Journal of Clinical Laser Medicine & Surgery 16(16) 305-308.
Effects of Cavity Size

on Apoptosis-Induction
During Pulp Wound Healing
C Kitamura Y Ogawa
T Morotomi M Terashita
Clinical Relevance
The over-induction of pulpal apoptosis may lead to irreversible pulpal reaction by severe
cavity preparation.
SUMMARY
The effects of mechanical stress on apoptosis
induction during pulp wound healing were examined. Mechanical stress cavities of two different
sizes were prepared on individual rat molars, one
twice the size of the other in the occlusocervical
direction. The authors compared the distribution
pattern and number of apoptotic cells of the two
groups by terminal deoxynucleotidyl transferasemediated labeling assay. At one hour and one day,
significant differences were observed in the distribution patterns and number of apoptotic cells
between the single-size and double-size group.
Four days after injury, apoptosis still existed on
pulp cells in the double-size group but not in the
single-size group. At 14 days, no difference in the
number of apoptotic cells between the two groups
*Chiaki Kitamura, DDS, PhD, assistant professor, Department of
Operative Dentistry, Kyushu Dental College, Japan
Yukio Ogawa, DDS, postgraduate student, Department of
Operative Dentistry, Kyushu Dental College, Japan
Takahiko Morotomi, DDS, PhD, assistant professor, Department
of Operative Dentistry, Kyushu Dental College, Japan
Masamichi Terashita, DDS, PhD, professor and chair,
Department of Operative Dentistry, Kyushu Dental College,
Japan
*Reprint request: 2-6-1 Manazuru, Kokurakita, Kitakyushu 8038580, Japan; e-mail: [email protected]
was observed. These results suggest that the magnitude of mechanical stress, such as cavity preparation, may modulate the induction of apoptosis
during pulp wound healing.
INTRODUCTION
Wound healing of dental pulp after caries progression
and cavity preparation involves odontoblast survival,
differentiation of pulp cells to odontoblast-like cells and
the cell-death process of damaged odontoblasts and pulp
cells. Odontoblast survival results in reactionary
dentinogenesis have been established in humans and
other animals (Smith & others, 1995; Bjrndal, Darvann
& Thylstrup, 1998; Smith, Tobias & Murray, 2001;
About & others, 2001a), and the recruitment of odontoblast-like cells results in reparative dentinogenesis
(Ohshima, 1990; Tziafas, 1995; Mitsiadis, Fried &
Goridis, 1999). The area and volume of reactionary and
reparative dentin formation are dependent on the magnitude of mechanical stress, such as cavity preparation
(Lee, Walton & Osborne, 1992; Murray & others, 2000).
On the other hand, the aspect of cell death has been previously studied. Several parameters are indicated as
steps toward cell necrosis after caries progression and
cavity preparation, which include the calciotraumatic
line and the aspiration of odontoblasts after acute
injuries (Brnnstrm, 1968; Trowbridge, 1981; Mjr,
2001).
76
Operative Dentistry
Apoptosis, one type of cell death, is distinct from necrosis and is recognized by morphological criteria such as
cell shrinkage, nuclear chromatin condensation, the formation of apoptotic bodies and the rapid removal of
apoptotic cells by scavenger cells (Earnshaw, 1995; Kerr
& others, 1995). The hallmark of apoptosis is enzymatic splicing of DNA that can be visualized in tissue sections by using various methods, including terminal
deoxynucleotidyl transferase-mediated labeling assay.
The apoptotic regulation of damaged odontoblasts and
pulp cells was noted in some reports (Bronckers & others, 1996; Vermelin & others, 1996). Recently, the
authors reported that two waves of apoptosis were
induced on dental pulp as one of the cell death regulation processes during wound healing (Kitamura & others, 2001 accepted). This apoptotic phenomenon consists
of the primary apoptosis induced on odontoblasts and
the secondary apoptosis induced on subodontoblastic
pulp cells. Previously, the relationship between cavity
size and pulpal inflammatory responses was suggested
(About & others, 2001b). However, it is not clear
whether inducing apoptosis is affected by the magnitude of mechanical stress and whether this apoptotic
phenomenon is associated with the pulpal reaction
against severe stress.
In this study, cavities of different sizes were prepared
on rat molars, and their distribution pattern and number of apoptotic cells was compared between two different cavity size groups during pulp wound healing.
Preparation of Cavities on Rat Molars
The animal protocol followed the guidelines for animal
care of Kyushu Dental College and ethical approval was
obtained from the institutional panel for animal care of
Kyushu Dental College.
Twelve Wistar-specific pathogen free rats (nine weeks
old) weighing 250-350 g were cared for and used under
barrier system conditions. Under this controlled condition, the rats were deeply anesthetized by intraperitoneal injection of 5% pentobarbital sodium (Nembutal,
Dainippon Pharmaceutical Co, Suita, 564, Japan) at a
dose of 30 mg/kg. Two different sizes of Class V cavities
were prepared on the mesial aspects of maxillary bilateral first molars with a #1/2 round bur under water-cooling. One cavity was prepared into approximately half
the thickness of dentin and the occlusocervical length of
the cavity matched the diameter of the #1/2 round bur
(single-size group). The other cavity was prepared at the
same depth as the single size but the occlusocervical
length was twice (double-size group). To avoid the
effects of materials on the pulp wound healing process
in this study, the cavities were not filled with materials
and the rats were cared for under the barrier system
condition until they were sacrificed. Postoperative intervals of sacrifice were one hour, and one, four and 14
days, based on the results of previous investigations

(Taylor & Byers, 1990; DSouza & others, 1995). The
authors used three individual molars for the each cavity size in each interval. After each interval, the rats
were anesthetized using the above method and killed by
transcardial vital perfusion with 4% paraformaldehydephosphate buffer, pH 7.3 (4% PFA). The maxillary segments, including first molars, were carefully dissected
and immersed in 4% PFA at 4C overnight for further
fixation. The first molars dissected from maxillary segments were then demineralized in 10% EDTA-phosphate buffered saline (1xPBS), pH 7.3 at 4C.
Terminal
Deoxynucleotidyl
TransferaseMediated Labeling (TUNEL) Assay
The demineralized molars were dehydrated with graded ethanol and embedded in paraffin. Serial 5 m-sections were cut and mounted on three aminopropyltriethoxysilane-treated object slides. To indicate two different cavity sizes, one section from each group was
stained with hematoxylin and eosin. After deparaffinization and dehydration, sections were subjected to
terminal deoxynucleotidyl transferase-mediated labeling (TUNEL) assay using In Situ Cell Death Detection
Kit, Fluorescein (Roche Molecular Biochemicals, D68298, Mannheim, Germany) according to the manufacturers instructions. Sections were treated with proteinase K (20 g/ml) for 15 minutes at 37C, washed
with 1xPBS and incubated with 100 l of terminal
deoxynucleotidyl transferase (TdT) mixture that consisted of 5 l of TdT, 45 l of TdT buffer containing fluorescein-labeled nucleotide mixture and 50 l of 10%
bovine serum albumin for one hour at 37C. The sections were then rinsed with 1x PBS, and observed under
fluorescence microscopy.
Statistical Analysis of the Number of Apoptotic
Cells
Two serial sections were selected, including the deepest
area of cavities from each molar (six sections for each
Figure 1: Preparation of two different cavity sizes on rat molars. The

diameter of the single-size cavity (a) was same as that of the #1/2 round
bur. The occlusocervical length of the double-size cavity (b) was almost
twice the diameter of the round bur (magnification x100). Cav, cavity.
Kitamura & Others: Effects of Cavity Size on Pulpal Apoptosis Induction

group) and used for semi-quantitative analysis. After
counting TUNEL-positive cells under the cavity in each
section, all data were compared using Students t-tests.
RESULTS
Figure 1 shows representative sections from the singlesize groups and the double-size groups. TUNEL assay
was implemented on each group from each time interval.
Figures 2 to 5 show representative distribution patterns
of apoptotic cells in all groups. One hour after injury,
apoptotic odontoblasts were detected in contact with
dentin underneath cavities both in the single and the
double-size group (Figure 2). Some apoptotic odontoblasts appeared to be located within the injured dentin in
both groups. The distribution pattern of apoptotic odontoblasts in the double-sized group more broadly spread
toward the occlusocervical direction than the single-size
group. One day after injury, apoptotic pulp cells were
77
detected in the mesiocoronal pulp area in both groups

(Figure 3). In the single-size group, the distribution of
apoptotic cells was limited to the subodontoblastic area
underneath the cavity. In contrast, apoptotic cells in the
double-size group broadly spread from the mesiocoronal
area to the center of the coronal pulp and the upper area
of the mesial root pulp. Four days after injury, prolongation of the apoptosis-induction was observed in the double-size group but most apoptotic pulp cells were absent
in the single-size group (Figure 4). After 14 days, apoptotic pulp cells markedly decreased in both groups
(Figure 5). Only a few apoptotic cells in the double-size
group were detected.
The authors counted the apoptotic cells in each group
and examined differences in the number of apoptotic
cells between the two groups. Figure 6 shows the statistical data of the number of apoptotic cells in each group
at each time interval. In one hour, one day and four days,
Figure 2. One hour after cavity preparation. Apoptotic odontoblasts in the double-size group (b) more broadly spread in
the occlusocervical direction than those in the single-size
group (a). Arrowheads indicate representative apoptotic
odontoblasts (magnification x200). D, dentin; P, pulp.
Figure 3. One day after cavity preparation. Apoptotic pulp cells in

the double-size group (b) broadly spread in mesiocoronal pulp
compared with the single-size group (a). Arrowheads indicate
representative apoptotic pulp cells (magnification x200). D,
dentin; P, pulp.
Figure 4. Four days after cavity preparation. Apoptosis

were still induced on pulp cells in the double-size group
(b), almost none in the single-size group (a).
Arrowheads in (b) indicate representative apoptotic
pulp cells (magnification x200). D, dentin; P, pulp.
Figure 5. Fourteen days after cavity preparation. Few apoptotic cells were detected in both the single-size groups (a)
and the double-size group (b) (magnification x200). D,
dentin; P, pulp; *, reparative dentin.
78
Operative Dentistry
of secondary apoptosis more broadly spread and prolonged in the double-size group than in the single-size
group. These findings reveal that secondary apoptosis
induced on pulp cells is affected by the magnitude of
mechanical stress. Apoptosis plays an essential role in
controlling various biological systems, including homeostasis in several diseases (Jacobson, Weil & Raff, 1997;
Willingham, 1999). The increment of apoptosis induction by severe stress raises the potential role of apoptosis in maintaining pulpal homeostasis against the
increment of stress. The enhancement and prolongation of apoptosis induction by severe stress also raises
the possibility that over-induction of two waves of apoptosis by over-cutting may result in disruption of the
pulpal homeostasis and can lead to irreversible pulpal
reaction. In this study, nearly apoptotic cells of both
groups were eliminated 14 days after injury, indicating
that dental pulp has a tolerance or can recover from the
damage inflicted.
Figure 6: Statistical comparison of the number of apoptotic cells

between the single-size group and the double-size group in each time
interval. One hour, one day and four days after injury, the number of
apoptotic cells in the double-size group was significantly larger than the
single-size group. No significant difference was observed between the
double-size group and the single-size group. s, single-size group; d, double-size group. ** Significance of difference, p<0.01.
there were significant differences in the number of apoptotic cells between the single and double-size group. The
number of apoptotic cells in the double-size group was
significantly larger than the single-size group (p<0.01).
Fourteen days after injury, however, there was no statistical difference in the number of apoptotic cells between
the two groups.
DISCUSSION
The effect of mechanical stress on the apoptosis induction during pulp wound healing was examined. As the
mechanical stress, the authors prepared two different
sized cavities (single-size and double-size) on sound rat
molars in the burrier system condition to minimize
other effects. Primary-induced apoptotic odontoblasts
distributed according to cavity size and the number of
apoptotic odontoblasts was significantly larger in the
double-size group than in the single-size group. These
results suggest that the primary-induction of apoptosis
on odontoblasts may directly depend on the size of the
injured dentin area. The localization of some apoptotic
odontoblasts within the injured dentin also suggests
that the aspiration of odontoblasts into dentin by the
cavity preparation procedure may be a factor that
induced primary apoptosis on odontoblasts. Induction
Recently, chemical stress, such as capping agents that

affects the induction of apoptosis during pulp wound
healing, has been indicated (Kitamura & others, 2001
accepted). Based on previous data and current results,
the authors suggest that two waves of apoptosis
induced on odontoblasts and pulp cells during wound
healing may be modulated from mechanical and chemical stress.
CONCLUSIONS
The primary-induction of apoptosis on odontoblasts is
dependent on the size of the injured dentin area. The
secondary apoptosis induced on pulp cells is affected by
the magnitude of mechanical stress. Taken together,
double-sized cavities have more significantly enhanced
and prolonged the induction of pulpal apoptosis during
wound healing compared to single-size cavities. These
results suggest that the increased induction of pulpal
apoptosis may be a response that leads to irreversible
pulpal reaction after the extensive cutting of the
dentin/pulp complex.
Acknowledgements
This research was partially supported by Grants in Aid for
Scientific Research 13771136, 14207081 and 14207082
(Kitamura) from The Ministry of Education, Science, and Culture
of Japan, Tokyo, Japan.
(Received 28 March 2002)

References
About I, Murray PE, Franquin J-C, Remusat M & Smith AJ
(2001a) The effect of cavity restoration variables on odontoblast cell numbers and dental repair Journal of Dentistry
29(2) 109-117.
Kitamura & Others: Effects of Cavity Size on Pulpal Apoptosis Induction

About I, Murray PE, Franquin J-C, Remusat M & Smith AJ
(2001b) Pulpal inflammatory responses following non-carious
Class V restorations Operative Dentistry 26(4) 336-342.
79
Mitsiadis TA, Fried K & Goridis C (1999) Reactivation of DeltaNotch signaling after injury: Complementary expression patterns of ligand and receptor in dental pulp Experimental Cell
Research 246(2) 312-318.
Bjrndal L, Darvann T & Thylstrup A (1998) A quantitative light

microscopic study of the odontoblast and subodontoblastic
reactions to active and arrested enamel caries without cavitation Caries Research 32(1) 59-69.
Mjr IA (2001) Pulp-dentin biology in restorative dentistry. Part

2: Initial reactions to preparation of teeth for restorative procedures Quintessence International 32(7) 537-551.
Brnnstrm M (1968) The effect of dentin desiccation and aspirated odontoblasts on the pulp The Journal of Prosthetic
Dentistry 20(2) 165-171.
Murray PE, About I, Lumley PJ, Franquin J-C, Remusat M &

Smith AJ (2000) Human odontoblast cell numbers after dental
injury Journal of Dentistry 28(14) 277-285.
Bronckers ALJJ, Lyaruu DM, Goei W, Litz M, Luo G, Karsenty

G, Woltgens JH & DSouza RN (1996) Nuclear DNA fragmentation during postnatal tooth development of mouse and hamster and during dentin repair in the rat European Journal of
Oral Sciences 104(2) 102-111.
Ohshima H (1990) Ultrastructural changes in odontoblasts and

pulp capillaries following cavity preparation in rat molars
Archives of Histology and Cytology 53(4) 423-438.
D'Souza RN, Bachman T, Baumgardner KR, Butler WT & Litz M

(1995) Characterization of cellular responses involved in
reparative dentinogenesis in rat molars Journal of Dental
Research 74(2) 702-709.
Earnshaw WC (1995) Nuclear changes in apoptosis Current
Opinion in Cell Biology 7(3) 337-343.
Jacobson MD, Weil M & Raff MC (1997) Programmed cell death
in animal development Cell 88(3) 347-354.
Kerr JFR, Gob GC, Winterford CM & Harmon BV (1995)
Anatomical methods in cell death Methods in Cell Biology 46
1-27.
Kitamura C, Kimura K, Nakayama T, Toyoshima K & Terashita
M (2001) Primary and secondary induction of apoptosis in
odontoblasts after cavity preparation of rat molars Journal of
Kitamura C, Ogawa Y, Morotomi T & Terashita M (2001)
Differential induction of apoptosis by capping agents during
pulp wound healing Journal of Endodontics (accepted).
Lee SJ, Walton RE & Osborne JW (1992) Pulp response to bases
and cavity depths American Journal of Dentistry 5(2) 64-68.
Smith AJ, Cassidy N, Perry H, Bgue-Kirn C, Ruch JV & Lesot

H (1995) Reactionary dentinogenesis The International
Journal of Developmental Biology 39(1) 273-280.
Smith AJ, Tobias RS & Murray PE (2001) Transdentinal stimulation of reactionary dentinogenesis in ferrets by dentine
matrix components Journal of Dentistry 29(5) 341-346.
Taylor PE & Byers MR (1990) An immunocytochemical study of
the morphological reaction of nerves containing calcitonin
gene-related peptide to microabscess formation and healing in
rat molars Archives of Oral Biology 35(8) 629-638.
Trowbridge HO (1981) Pathogenesis of pulpitis resulting from
dental caries Journal of Endodontics 7(2) 52-60.
Tziafas D (1995) Basic mechanisms of cytodifferentiation and
dentinogenesis during dental pulp repair The International
Journal of Developmental Biology 39(1) 281-290.
Vermelin L, Lecolle S, Septier D, Lasfargues J & Goldberg M
(1996) Apoptosis in human and rat dental pulp European
Journal of Oral Sciences 104(5-6) 547-553.
Willingham C (1999) Cytochemical methods for the detection of
apoptosis The Journal of Histochemistry and Cytochemistry
47(9) 1101-1110.
The Effect of Placement of

Glass Fibers and Aramid Fibers
on the Fracture Resistance of
Provisional Restorative Materials
G Saygili SM Sahmali F Demirel
Clinical Relevance
The effectiveness of glass-fiber reinforcement is most evident in interim longspan fixed partial dentures. The resins used in this study showed superior fracture resistance.
SUMMARY
The fracture resistance of provisional restorations is an important concern for the restorative
dentist. The fracture resistance of a material is
directly related to its transverse strength. Six
specimens of similar dimensions were prepared
from three resins (PMMA, PEMA and BIS acrylcomposite). The resins were reinforced with glass
and aramid fibers. The samples were tested immediately after the material set, following seven
days of wet storage using three-point compression loading. The results were analyzed with an
analysis of variance (ANOVA). Fracture resistance of the specimens was statistically different
(p<0.001) among the materials. Specimens reinforced with glass fibers showed higher transverse
*Glbin Saygili, DDS, PhD, associate professor, Department of
Prosthodontics, Hacettepe University, Faculty of Dentistry,
Ankara, Turkey
Sevil M Sahmali, DDS, PhD, associate professor, Department of
Ankara, Turkey
Figen Demirel, DDS, PhD, associate professor, Department of
Ankara, Turkey
*Reprint request: Hacettepe niversitesi, Dishekimligi Fakltesi,
Protetik Dis Tedavisi ABD, 06100 AnkaraTrkiye; e-mail: [email protected]
strength (149.82 MPa). The fiber reinforcement of

resin materials increased the strength values (2050%). Within the limitations of this study, the
transverse strengths of PMMA, PEMA and BIS
acryl-resin composites were improved after reinforcement with glass and aramid fibers.
INTRODUCTION
Provisional restoration is an important stage of therapy
prior to placing final fixed prostheses (Amsterdam &
Fox, 1959; Shillingburg, Hobo & Whitsett, 1997; Powell
& others, 1994).
These prostheses are made indirectly in a dental laboratory. Several days or weeks are usually required for
completion. During this period, the patient must wear a
provisional restoration that provides pulpal protection,
positional stability, maintenance of occlusal function,
ease of cleansability, strength, retention and esthetics
for the prepared teeth (Shillingburg & others, 1997;
Powell & others, 1994; Larson & others, 1991; Hazelton
& Brudvik, 1995). A catastrophic fracture could necessitate lengthy chairtime for repair or replacement
(Shillingburg & others, 1997). Hence, structural reinforcement is desirable if it can prolong clinical life
expectancy.
Tooth-colored acrylic resin is the material of choice for
the provisional coverage of teeth that have been pre-
Saygili, Sahmali & Demirel: Fracture Resistance of Provisional Resin Materials with Fibers
pared for fixed prosthodontic restorations. Acrylic resin
is convenient to use and provisional restorations can be
made by a variety of techniques (Sotera, 1973;
Frederick, 1975; Preston, 1976; Kaiser, 1978; Kaiser &
Cavazos, Jr, 1985; Koumjian & Nimmo, 1990). Polymers
used in interim fixed partial dentures (FPDs) are often
based on poly(methyl methacrylate) (PMMA), poly
(ethyl methacrylate) (PEMA) or n-poly (butyl methacrylate) (PBMA). However, some alternative resins have
been used for provisional restorations. Two alternatives
include visible light-cured microfilled resin composites
(Wood, Halpern & Lamb, 1984) and urethane dimethacrylate resin (Koumjian & Nimmo, 1990; Haddix, 1988).
In long-span restorations, strength is a critical property. When masticatory forces are applied to a longspan provisional restoration, fracture of the restoration
is more likely than with a short-span restoration
(Koumjian & Nimmo, 1990).
Different types of fibers have been added to polymer
materials to improve their mechanical properties.
Orthopedic acrylic resin-based bone cements have successfully been reinforced with glass, carbon and aramid
fibers (Vallittu, Lassila & Lappalainen, 1994). In periodontics, glass fibers have been tested as additives to
BIS-GMA resin in temporary splints to immobilize
teeth (Friskopp & Blomlf, 1984). In orthodontics, the
use of aramid fibers has been found to be useful in reinforcing orthodontic appliances (Mullarky, 1985).
In prosthodontics, fibers have been used to improve
the fracture resistance or the moduli of elasticity of
polymer materials. Glass fibers have been studied as a
strengthening material added to polymethyl methacrylate, and carbon fibers have been used to reinforce
prosthodontic restorations (Vallittu & Lassila, 1992;
Solnit, 1991). The usefulness of glass fiber rovings as
strengtheners of dental resins has been established
(Vallittu & Lassila, 1992; Friskopp & Blomlf, 1984).
This study compared the effect of polyaramid fibers
(Kevlar) and glass fibers (E-glass) on the fracture
resistance of three provisional resin materials.
monomer ratio provided the best viscosity during stainless steel mold placement for the Dentalon Plus acrylic
resin (Heraus Kulzer, Gmbh, 61273, Wehrheim,
Germany). A 2.5:1 ratio was found to be appropriate for
Jet acrylic resin (Lang Dental Mfg Co, Wheeling, IL
60090, USA) and a 1:1 catalyst paste to base material
was best for Protemp (ESPE Premiere, Norristown, PA
19404, USA). These ratios were used throughout this
investigation. Each acrylic resin was prepared by the
same investigator and placed in the mold.
To improve adhesion, the surface of the fibers used
were coated with the silane solution Silicer (Heraeus
Kulzer GmbH 961273, Wehrheim, Germany) by dipping the glass and aramid fiber roving in a silane solution (Valittu & others, 1994).
The fibers were air dried for 20 minutes, then dipped
into a methacrylate monomer. The fibers were added
to the mixture when the mold was two-thirds full. The
remainder of the mold was then filled, covered and
allowed to polymerize under 20 psi pressure for 10
minutes. All the fibers studied were used in roving
(continuous) form. The glass fibers used were E-glass
(Ahlstrom, 48810 Karhula, Finland) and the aramid
fibers were Kevlar (DuPont, Wilmington, DE 19898,
USA). Each group consisted of six specimens. All of the
fibers were placed longitudinal to the specimen and
perpendicular to the loading force. The size of the specimens was measured by micrometer (NSK Max-cal,
Japan Micrometer Co, 541 Osaka, Japan). Grinding
the specimens to the predetermined dimensions eliminated differences in the dimensions of the test specimens used. The control group had no reinforcement.
Specimens used in the first group were tested immediately after the specimens set. The second group was
tested after seven days of storage in water at 37C.
Each specimen was loaded with an Instron Testing
Machine (Instron Corp, Canton, MA 02021, USA) with
a crosshead speed of 0.5 cm per minute. The specimens
were supported by two stainless steel rods 8 mm in
diameter and 50 mm apart. The force applied to fracture the specimens was diagrammed (Figure 1).
The fracture resistance of resins is related to their
transverse strength. The transverse strength of each
sample was calculated using the formula (Koumjian &
Nimmo, 1990):
A stainless steel mold was used to make resin specimens measuring 65 x 10 x 3 mm. Six specimens each
were made from a poly (methylmethacrylate) resin, a
poly (ethylmethacrylate)
Table 1: Technical Profiles of the Resin-Modified Glass Ionomer Cements Investigated
resin and BIS-acryl resin
composite. The proviProduct
Content
Manufacturer
sional resin materials
Dentalon
Plus
Polyethylmethacrylate
Heraus Kulzer GmbH, 61273, Wehrheim Germany
used are listed in Table 1
Jet
Polymethylmethacrylate
Lang Dental Mfg Co, Wheeling, IL 60090, USA
and include the product
Protemp
BIS-acryl
resin
composite
ESPE
Premiere, Norristown, PA 19404, USA
names and manufacturKevlar
Polyaramid fiber
Du Pont, Wilmington, DE 19898, USA
ers.
During a pilot study, a
2.0:1.2
polymer
to
81
E-glass
Glass fiber
Ahlstrom Corp, 48810, Karhula Finland
Silicer
Silane solution
Heraus Kulzer GmbH 961273, Wehiheim-Germany
82
Operative Dentistry
3 PI
S=2bd2
function because of poor transverse, impact and flexural strength of resins (John, Gangadhar & Shah, 2001).
Fatigue is one of the most common causes of breakage
of provisional restorations (Stafford & Smith, 1970).
P: fracture load
I: distance between the supports
b: width of the specimen
d: thickness of the specimen
This study compared the transverse strengths of

PMMA, PEMA and BIS-acryl resin composites and
the same resins reinforced with glass and aramid
fibers.
The mean values of the transverse strength and standard deviations were calculated and compared between
the groups with analysis of variance (ANOVA).
The crown and fixed partial denture acrylic resins

used in this study were not pressure-packed or heatcured, nor were they cross-linked to a significant
degree. As a result, these resins may have had an
increase in the number of microporosities, monomer
retention and a large polarity that would enhance the
rate of water sorption (Larson & others, 1991).
Theoretically, these acrylic resins may have already
reached their maximum water sorption within a few
hours as a result of monomer retention and high polarity. For these reasons, the strength of the acrylic resins
may have remained uniform. To simulate oral condi-
The Newman-Keuls Multiple-Comparison test

revealed significant differences between the means of
the fracture load values (p<0.001).
RESULTS
ANOVA showed statistically significant effects for the
variables in Table 2.
The materials tested, and the mean values and standard deviations of transverse strength for time and treatment groups and their interactions with each other are
shown in Tables 3-6.
The Newman-Keuls Multiple Comparison test was performed, revealing significant differences (p<0.001)
among the materials, time and treatment groups (Table 7).
Jet acrylic resin with glass fibers exhibited a significantly higher mean transverse strength (127.06 MPa)
than the other two resins, and Protemp without fibers
demonstrated a significantly lower mean transverse
strength (48.17 MPa) than the other specimens.
The highest transverse strength value, 149.82 MPa,
was achieved by reinforcement of PMMA with glass
fibers.
In the group tested immediately, Jet resin showed the
highest strength followed by Dentalon Plus and
Protemp. Protemp demonstrated significantly lower
strength.
In the wet-storage
group, the values
were higher than
the initial values.
The fiber reinforcement of resin
materials increased
strength values (2050%).
DISCUSSION
Tooth-colored resin is
the material of
choice for fabricating
provisional restorations. However, fracture of the restoration may occur during
Figure 1: Diagrammatic representation of the test specimen.
Table 2: Results of ANOVA for Transverse Strength Values (Dependent: Transverse strength (MPa))
(Significant p<0.001)
Effect
df Effect
SS
MS Effect
P-Level
38.999.11
19499.56
41051.99
0.000000
6789.349
6789.349
7732.17
0.000000
24226.46
12113.23
12247.08
0.000000
1x2
288.2585
144.1293
164.14
0.000000
1x3
1477.178
369.2945
373.38
0.000000
2x3
48.563
24.2815
35.09
0.000000
1x2x3
21.21102
5.302754
7.66
0.000223
Effect: 1 = Material; 2 = TIme; 3 = Treatment

df: Degrees of freedom
SS: Sum of Squares
MS: Mean Square
F: F-ratio
p: probability
Different procedures for reinforcing these resins have

been recommended but they have met with limited success. Adding glass fibers to dental resins to improve
their strength has been studied by many investigators
(Solnit, 1991; Vallittu & others, 1994; Powell & others,
1994; John & others, 2001).
tions, the resins were tested immediately after the

material set and after seven days of wet storage.
Water has a plasticizing effect resulting from an
interaction with the polymer structure (Valittu, Ruyter
& Ekstrand, 1998). Many studies on water sorption of
denture base polymers have been carried out, and it
has been concluded that sorption decreases the
mechanical properties of denture base polymers. In
contrast, Koumjian & Nimmo (1990) found that Triad,
Protemp, Snap and Trim resins exhibited a significant
increase in transverse strength after seven days of wet
storage. In this study, water sorption did not adversely affect transverse strength.
Different types of fibers are produced commercially.

In this study, glass fiber and aramid fiber were selected
to reinforce the acrylic resins. Glass is an organic substance that has been cooled to a rigid state without
crystallization (John & others, 2001). Different types of
glass fibers are available. These include E-glass, Sglass, R-glass, V-glass and Cemfil. Of these, E-glass
fiber, which has high alumina and low alkali and
borosilicate, claims to be superior in flexural strength
(Solnit, 1991).
As the provisional restoration should provide pulpal

protection, positional stability, occlusal function,
access for cleaning, esthetics, strength and retention
(Shillingburg & others, 1997), during this stage of clinical treatment, periodontally involved abutment teeth
are assessed. The functional occlusion, vertical dimension, prognosis of questionable teeth and the esthetic
and phonetic acceptability of the tooth position may be
evaluated (Amsterdam & Fox, 1959; Shillingburg &
others, 1997). It may take months or years to make
such assessments and a provisional restoration must
be durable (Shillingburg, & others, 1997).
Aramid is a generic term for wholly aromatic fibers.

These fibers are resistant to chemicals, thermally stable and have a high mechanical stability, melting point
and glass transitional temperature. They also have
pleated structures (molecules are radially arranged in
the form of sheets) that make aramid weak in terms of
flexural, compression and abrasion behavior (John &
others, 2001).
Table 3: Mean Values and Standard Deviations of Transverse Strength, Material

x Time Interaction (1 x 2) (mean, SD) (MPa)
Time
Material
Immediate
83
7-Day Wet
Mean
SD
Mean
SD
Protemp
61.81
0.54
73.72
0.54
Dentalon Plus
81.98
0.54
97.74
0.54
Fibers can most easily be placed longitudinally inside the specimen by using
fiber rovings instead of woven fibers. If
fibers are used to strengthen a polymer
material, optimal adhesion between the
fibers and the polymer matrix is essential. To improve adhesion, the surface of
the fibers can be silane-treated (Clark &
Ploedmann, 1963).
The accepted engineering beam theory

(Powell & others, 1994) states that when
a beam is loaded
m i d - s p a n
Table 4: Mean Values and Standard Deviations of Transvers Strength, Material x Treatment Interaction
between two sup(1 x 3) (mean, SD) (MPa)
porting points,
Treatment
the applied load
Material
No Fibers
Kelvar
E-Glass
induces tension
Mean
SD
Mean
SD
Mean
SD
in the bottom
Protemp
53.87
0.68
69.07
0.68
80.35
0.68
fibers and comDentalon Plus
72.21
0.68
89.98
0.68
107.39
0.68
pression in the
top fibers. In this
Jet
90.06
0.68
114.37
0.68
138.44
0.68
study, the treated
PMMA, PEMA
and
BIS-acryl
Table 5: Mean Values and Standard Deviations of Transvers Strength, Time x Treatment Interaction (2 x 3)
(mean, SD) (MPa)
composite bars
would not be a
Treatment
catastrophic failTime
No Fibers
Kelvar
E-Glass
ure. In fact, a
Mean
SD
Mean
SD
Mean
SD
crack would occur
Immediate
64.83
0.46
83.40
0.46
99.9
0.46
on the tension
7-Day Wet
79.27
0.46
98.88
0.46
117.55
0.46
side but would
Jet
104.33
0.54
124.24
0.54
84
Operative Dentistry
Table 6: Mean Values and Standard Devations of Transverse Strength, Material x Time x Treatment Interaction (1 x 2 x 3)
(mean, SD)
Treatment
Material
Time
No Fibers
Kelvar
E-Glass
Mean
SD
Mean
SD
Mean
SD
Protemp
Immediate
48.17
0.81
63.01
0.81
74.24
0.81
7-Day Wet
59.57
0.81
75.12
0.81
86.45
0.81
Dentalon
Plus
Immediate
64.87
0.81
82.69
0.81
98.39
0.81
7-Day Wet
79.55
0.81
97.28
0.81
116.39
0.81
Immediate
81.44
0.81
104.51
0.81
127.06
0.81
7-Day Wet
98.68
0.81
124.23
0.81
149.82
0.81
Jet
Table 7: For Transverse Strength Values Evaluation of Differences Among the Means of
Materials, Time and Treatment Groups by Newman-Keuls Multiple Comparison Test
(mean, significance) (MPa)
Materials
Protemp
(1)
Dentalon Plus (2)
Jet
(3)
Mean
67.76
89.86
114.29
Difference Among Groups

2, 3
1, 3
1, 2
Time
Immediate
(1)
7-day wet
(2)
Mean
82.71
98.57

2
1
Mean
72.05
91.14
108.72

2, 3
1, 3
1, 2
Treatment Groups
No fibers
(1)
Kevlar
(2)
E-Glass
(3)
not propagate through to the compression point. The

embedded fiber could not be stretched enough for the
crack in the resin to continue. The fiber appeared to
hold the two pieces together. Because a fracture failure
is usually related to the initiation of a crack and its subsequent propagation until displacement (Gegauff &
Pryor, 1987), if a provisional FPD fractures in the
mouth, it is difficult if not impossible to repair. This
presents an inconvenience to both the dentist and the
patient because the provisional prosthesis often must
be remade.
The same technique for measuring fracture resistance
was used in earlier studies (Powell & others, 1994;
Koumjian & Nimmo, 1990; Friskopp & Blomlf, 1984;
Ramos, Runyan & Christensen, 1996; Vallittu, 1998).
The results of separate studies carried out earlier and
those of this study closely agree (Nohrstrm, Vallittu,
Yli-Urpo, 2000; Vallittu & others, 1998).
All the materials tested showed a significant increase
in transverse strength after seven days of wet storage.
The transverse strength values were higher following
wet storage compared to those tested immediately
after the material set. Protemp was consistently weaker than the other two resins tested with or without
fiber reinforcement, immediately after setting and in
the wet storage groups. Cross-linking has little influence on tensile strength, transverse strength or hard-
ness. Jet was found to be the

strongest resin material
immediately after setting
and following seven days of
wet storage. The fiber reinforcement of the materials
tested
caused
20-50%
increases in transverse
strength.
These tests are in vitro and
other factors, such as marginal integrity, color stability, and ease of manipulation,
should also be considered.
CONCLUSIONS
The fibers incorporated in the acrylic resin material
enhanced the fracture resistance of PMMA, PEMA and
BIS-acryl resin composite test specimens.
The information presented in this study will aid the
restorative dentist in selecting a provisional material.
To determine whether the fibers incorporated into
PMMA, PEMA and BIS-acryl resin composite also have
relevance clinically, other testing methods, such as
impact testing and fatigue testing, should be used.
(Received 3 April 2002)

References
Amsterdam M & Fox L (1959) Provisional splinting-Principles
and techniques Dental Clinic North America 4(1) 73-99.
Clark HA & Ploedmann (1963) Bonding of silane coupling agents
in glass reinforced plastics Modern Plastics 40 133-138.
Frederick DR (1975) The provisional fixed partial denture
Friskopp J & Blomlf L (1984) Intermediate fiberglass splints
Gegauff AG & Pryor HG (1987) Fracture toughness of provisional resins for fixed prosthodontics Journal of Prosthetic
Dentistry 58(1) 23-29.
85
Haddix JE (1988) A technique for visible light-cured provisional

restorations Journal of Prosthetic Dentistry 59(4) 512-514.
Preston JD (1976) A systematic approach to the control of esthetic

form Journal of Prosthetic Dentistry 35(4) 393-402.
Hazelton LR & Brudvik JS (1995) A new procedure to reinforce

fixed provisional restorations Journal of Prosthetic Dentistry
74(1) 110-113.
Ramos V Jr, Runyan DA & Christensen LC (1996) The effect of

plasma-treated polyethylene fiber on the fracture strength of
polymethyl methacrylate Journal of Prosthetic Dentistry 76(1)
94-96.
John J, Gangadhar SA & Shah I (2001) Flexural strength of

heatpolymerized polymethyl methacrylate denture resin
reinforced with glass, aramid, or nylon fibers Journal of
Shillinburg HT, Hobo S & Whitsett LD (1997) Fundamentals of

fixed-prosthodontics Chicago Quintessence 225-256.
Kaiser DA (1978) Accurate acrylic resin temporary restorations

Solnit GS (1991) The effect of methyl methacrylate reinforcement

with silane-treated and untreated glass-fibers Journal of
Kaiser DA & Cavazos E Jr (1985) Temporization techniques in

fixed prosthodontics Dental Clinics North America 29(2) 403412.
Sotera AJ (1973) A direct technique for fabricating acrylic resin

temporary crowns using the Omnivac Journal of Prosthetic
Dentistry 29(5) 577-580.
Kalachandra S & Turner DT (1987) Water sorption of

poly(methylmethacrylate) 3 Effects of plasticizers Polymers 28
1748-1752.
Stafford GD & Smith DC (1970) Flexural fatigue tests of some

denture base polymers British Dental Journal 128(9) 442-445.
Koumjian JH & Nimmo A (1990) Evaluation of fracture resistance of resins used for provisional restorations Journal of
Larson WR, Dixon DL, Aquilino SA & Clancy JMS (1991) The
effect of carbon graphite fiber reinforcement on the strength of
provisional crown and fixed partial denture resins Journal of
Mullarky RH (1985) Aramid fiber reinforcement of acrylic appliances Journal of Clinical Orthodontics 19(9) 655-658.
Nohrstrm TJ, Vallittu PK & Yli-urpo A (2000) The effect of placement and quantity of glass fibers on the fracture resistance of
interim fixed partial dentures International Journal of
Prosthodontics 13(1) 72-78.
Phillips RW (1991) Science of Dental Materials 8th Edition WB
Saunders Co London 157-176.
Powell DB, Nicholls JI, Yuodelis RA & Strygler H (1994) A comparison of wire-and Kevlar-reinforced provisional restorations
Vallittu PK & Lassila VP (1992) Reinforcement of acrylic resin

denture base material with metal or fibre strengtheners
Vallittu PK, Lassila VP & Lappalainen R (1994) Acrylic resinfiber compositePart I: The effect of fiber concentration on
fracture resistance Journal of Prosthetic Dentistry 71(6) 607-612.
Vallittu PK (1998) The effect of glass fiber reinforcement on the
fracture resistance of a provisional fixed partial denture
Vallittu PK, Ruyter IE & Ekstrand K (1998) Effect of water storage on the flexural properties of E-Glass and silica fiber acrylic
resin composite International Journal of Prosthodontics 11(4)
340-350.
Von Fraunhofer JA (1975) Scientific Aspects of Dental Materials
8th Edition London, Butterworth 439-447.
Wood M, Halpern BG & Lamb MF (1984) Visible light-cured composite resins: An alternative for anterior provisional restorations Journal of Prosthetic Dentistry 51(2) 192-194.
Literature Review
Tooth-Colored Post Systems:

A Review
AJE Qualtrough F Mannocci
Clinical Relevance
Tooth-colored fibre posts have several advantages over conventional metal posts. They are
aesthetic, they must be bonded to tooth tissue, they have a modulus of elasticity similar
to that of dentin and they appear to perform well in clinical studies without the risk of
fracture.
INTRODUCTION
Restoration of the root filled tooth by a post to retain a
crown dates back more than 200 years, when Fauchard
used posts constructed from gold or silver (Fauchard,
1880). Over the next century, post crowns became the
most popular method of restoration of roots. Opinions
differed among those who favored wooden posts and
those who preferred the metal variety. Wooden posts
were more retentive due to water absorption, but there
were no suitable cements to aid in the retention of metal
posts. Gold and platinum were considered to be of superior quality compared with brass, silver and copper,
which tended to corrode (Harris, 1839). As early as 1849,
when there was little emphasis on cleaning and shaping
endodontic procedures, Tomes proposed the principles of
post dimensions. These procedures still closely conform
with those used today (Tomes, 1848).
Ever since the early days, the provision of a post is still
regarded as the accepted method of core retention for
restoration of significant loss of coronal tooth tissue
(Schillingburg & Kessler, 1982; Rosensteil, Land &
Fujimoto, 1995). A further consideration was that a post
*AJE Qualtrough, Unit of Operative Dentistry and
Endodontology, University Dental Hospital of Manchester, UK
F Mannocci, Department of Restorative Dentistry, University of
Siena, Italy
*Reprint request: Higher Cambridge Street, Manchester M15 6FH,
UK; e-mail: [email protected]
was thought to render the root-filled tooth more resistance to fracture, although opinions now vary. For example, in a study of 59 endodontically-treated teeth with
and without post-reinforcement, Guzy & Nicholls (1979)
were unable to demonstrate any difference in fracture
resistance between the two groups. However, when
other factors were taken into account, the propensity for
root fracture due to the wedging effect of tapered posts
was considered a cause for concern by Standlee & others
(1972).
Conventionally, endodontic posts can be categorized
into two groups: custom-made and prefabricated.
Prefabricated posts may be divided into those retained
actively or passively. A wide range of posts have been
developed, the main differences are usually related to
taper and surface configuration. Systems that incorporate the use of matched parallel-sided preparation burs,
casting posts for indirect use and stainless steel/titanium posts for direct placement are popular.
For success, a metal post must satisfy certain criteria
(Johnson & Sakumara, 1978):
a) it should be as long as the prosthetic crown.
b) the sides should be as parallel as possible.
c) there should be a precision fit of the post within
the canal.
d) the apical 4 mm of the gutta-percha root canal
filling should not be removed.
87
Qualtrough & Mannocci: Tooth-Colored Post Systems: A Review

The crown may be constructed from a porcelain or
ceramic material or from ceramic bonded to metal. Posts
and cores may be made in the laboratory or pre-manufactured. The advantage of laboratory-constructed posts
is that an accurate representation of the root canal is
possible, hence, the fit of the post should be optimal.
However, two appointments are required since this is an
indirect procedure. Directly placed posts have the
advantage that the canal may be prepared and the post
bonded directly. They are also usually stronger and may
be more retentive but are constructed from non-precious
metals, therefore, the potential risk for corrosion leading to microleakage and root fracture exists.
The shape of metal posts may vary according to operator preference and canal shape and may be categorized
as follows:
- Smooth tapered
- Smooth parallel-sided
- Threaded tapered
- Threaded parallel-sided
On rare occasions there may be insufficient space
within the crown of the tooth for a retentive core and a
separate crown. Under these circumstances, a crown
bonded to a core may prove to be the best solution.
Destined for use in teeth that are already broken
down, the hazards associated with post placement are
not insignificant and include the risk of perforation.
There is also the risk of a generation of stresses within
the root dentin (Ross, Nicholls & Harrington, 1991), and
the failure rate of post-retained restorations has been
considered to be greater than any other indirect restoration in vital teeth (Roberts, 1970; Turner, 1982). Failure
can also be attributed to fracture or bending of the post,
loss of retention, core fracture or root failure. Corrosion
is also considered a significant cause of failure (Silness,
Gustavsen & Hunsbeth, 1979). However, there are few
clinical studies of metal post crown successes and failures. Sorensen & Martinoff (1984) reported 8.6% failures due to post loss, root fractures or post perforations.
Weine, Wax & Wenckus (1991) reported 6.5% failure
after 10 years or more and Torbjner, Karlsson &
Odman (1995) recorded a failure rate of 8.3% after two
to three years. There are other problems associated with
conventional metal post and core systems, one being the
aesthetics of the final restoration, as the restored crown
cannot possess the optical properties of natural tooth
structure. The metal core decreases the depth of
translucency of the coronal restoration and the post may
shine through in the cervical region, thus, altering the
appearance of thin gingival tissues. However, the final
aesthetic result is dependent on the relationship
between the opacity of the ceramic core material and the
degree of metal shine through. For example, some aluminous core materials can be as effective as porcelain
fused to metal in blocking metal core effects. Another

problem with metal posts relates to the difference in
elasticity between dentin (18.6 MPa Youngs modulus)
and metal (200 MPa). This difference results in unequal
distribution of strains on the dentin surface and a tendency to create stress concentration areas. Carbon fiber
posts are primarily made of carbon fibers immersed in
an epoxy resin matrix and show a different pattern of
elastic behavior depending on the direction of applied
stress. When a load was applied at an angle of approximately 35 to the long axis of the post, the modulus of
elasticity of carbon fiber posts was approximately 21
GPa, while dentin is approximately 21 GPa. More
recently, similar results were achieved when the load
was applied at 45 to the long axis of the post
(Asmussen, Peutzfeldt & Heitzmann, 1999).
As a consequence of the above considerations and following significant developments in dental materials science, including the introduction of reliable bonding systems (Mannocci & others, 1999), improved fiber reinforced resin composites and strengthened ceramics, a
new generation of tooth-colored post systems have been
proposed.
THE CERAMIC POST AND CORE
Historically, the main disadvantage of ceramic materials has been principally associated with their low flexural strength compared with metals, and in function,
ceramic materials have a record of frequent failure in
high-stress situations (Anusavice, 1996). Theoretically,
ceramic post and cores should have good aesthetic and
biological properties and, with advances in dental materials science, a resurgence of interest has emerged related to using these newer, reinforced ceramic materials as
an alternative to metal, especially with successful bonding systems. In 1989, Kwiatkowski and Geller described
the clinical application of glass-ceramic posts and cores
and, in 1991, Kern and Knode introduced posts and
cores made from glass infiltrated aluminum oxide
ceramic (Koutayas & Kern, 1999). In 1995, Pissis proposed that a post and core could be constructed as a single component made from a glass ceramic material.
Other ideas have included the introduction of pre-fabricated zirconia endodontic posts to be used in conjunction
with a direct resin composite core build-up (Sandhaus &
Pasche, 1994) and Ahmad (1998) described the practical
application of zirconia posts as a support to leucite-reinforced cores in practice. In a study by Purton, Love &
Chandler (2000), preformed ceramic posts were reported to be significantly more rigid than parallel-sided
stainless steel posts. The metal posts were also significantly more retentive than ceramic posts bonded using
a variety of luting/bonding agents and surface preparation techniques. However, Rosentritt & others (2000)
demonstrated that typical failure of metal systems was
marked by loosening compared with fracture of ceramic
88
Operative Dentistry
posts. This could be interpreted as being disadvantageous, as removal of a ceramic post is notoriously difficult. On the other hand, it is preferable that the post
fractures, rather than the root.
THE FIBER-REINFORCED POST AND CORE
The properties of fiber-reinforced materials not only
depend on the nature of the matrix and fiber, but also on
the interface strength and geometry of reinforcement
(Isaac, 1997). The addition of fibers to a polymer matrix
can result in a significant improvement in the mechanical properties of strength, fracture toughness, stiffness
and fatigue resistance (Drummond, 2000). Fibers may
be composed of woven polyethylene, glass or carbon. In
a study of different resin composite materials (Viguie &
others, 1994), the modulus of elasticity of a material
containing short, randomly distributed fibers was concluded to be similar to that of radicular dentin and,
hence, should be suitable for post and core construction.
Carbon fiber-reinforced resins are considered viable
alternatives to metals when strength, stiffness, lightness and resistance to corrosion and fatigue are considered. Carbon compounds in various forms have been
studied in several dental and surgical applications and
have proved to be biocompatible and mechanically satisfactory for many purposes. In 1990, Duret, Reynaud &
Duret introduced a non-metallic material based on the
carbon-fiber reinforced principle. In a later report (Duret,
Reynaud & Duret, 1992), the absence of corrosion was
also mentioned as an advantage of fiber posts, but this
statement was challenged in a study by Fovet,
Pourreyron & Gal (2000) in which the involvement of
carbon fiber posts in corrosion reactions was found.
Purton & Payne (1996) suggested that carbon fiber
posts could potentially replace stainless steel and other
metal posts in many clinical situations due to their
inherent rigidity, which allows smaller sizes to be used
for equivalent strength. They also suggested that
improved bond strength between post and root would
permit the potential replacement of stainless steel with
carbon fiber in post systems. The tensile bond strength
of adhesive systems to stainless steel, titanium, carbonfiber and zirconium dioxide root canal posts was evaluated by OKeefe, Miller & Powers (2000). Bonds to carbon fiber posts were weaker than to stainless steel and
titanium but stronger than to zirconium dioxide. The
potential problem associated with water sorption must
also be considered. Miettinen, Narva & Vallittu (1999)
reported that water sorption and solubility of fiber composites vary according to the brand and homogeneity of
polymer matrix and may affect the hydrolytic stability
of the composite structure. High sorption rates were
associated with microscopic voids and composition of
the polymer matrix.
Fiber-reinforced posts have exhibited a significant
decrease in flexural strength following thermocycling
(Torbjorner & others, 1996; Drummond, 2000; Mannocci,

Sherriff & Watson, 2001). This has been attributed to
degradation of the fibers or the matrix and to the difference in thermal expansion coefficients between the two.
Although considered as having significant advantages
compared to metal posts, especially with respect to
mechanical properties, the carbon fiber post is gray in
appearance. Replacing carbon with quartz fibers results
in a tooth-colored restoration. Recently, new quartz fiber
post-systems, such as Light-Post DT (RTD, St Egreve,
France), with a tapered design, were introduced to the
market to obtain better adaptation to the root canal
preparation.
Although posts and cores used to restore pulpless
teeth should be strong in the face of mechanical stress,
it is preferred that the post, regardless of construction
material, should fail in preference to tooth tissue. In a
study comparing the fracture resistance of teeth
restored with prefabricated carbon-fiber posts and composite cores to cast dowel-core restored teeth, MartinezInsua & others (1998) reported the failure threshold of
the carbon fiber posts was significantly lower than that
of cast posts and cast cores commonly resulting in tooth
fracture at failure. Comparable results were noted by
Mannocci, Ferrari & Watson (1999), who reported that
fiber posts reduced the risk of root fractures of teeth
restored to a minimum with quartz fiber, carbon quartz
fiber and zirconium dioxide ceramic posts. Similar
trends have been reported by Dean, Jeansonne &
Sarkar (1998) and Isidor, Odman & Brondum (1996).
Early retrospective studies indicate that the clinical performance of fiber posts is promising (Fredriksson & others, 1998; Ferrari & others, 2000; Ferrari, Vichi &
Garca-Godoy, 2000). In an in vivo study of carbon fiber
and quartz fiber posts luted using four different bonding-luting material combinations over a period of up to
four years, a failure rate of 3.2% was recorded. Failures
were due to debonding during removal of the temporary
restoration or the presence of a periapical lesion at the
radiographic examination. There was no significant difference in the type of failure between the groups
(Ferrari & others, 2000).
The success of 59 nine carbon-fiber post-composite
core restorations covered with full ceramic crowns was
evaluated in a prospective study (Glazer, 2000). The
average observation was 28 months. There were no root
fractures and the overall failure rate was 7.7%. The failure modes observed were two due to periapical pathology, one core debonded and one crown debonded.
Although destined to be permanent restorations, on
occasion, a post must be removed, which can prove to be
a challenging procedure. For fiber posts, the parallelism
of the stretched fibers in a resin matrix is said to facilitate the guidance of removal drills. This is significant
with respect to safe re-treatment (de Rijk, 2000).

GENERAL RECOMMENDATIONS
Indicators for provision of a non-metal post must be considered for each case on an individual basis as factors
such as attachment loss and tooth type must be considered. Currently, there is insufficient evidence-based
clinical data to permit meaningful comparisons to be
made between studies related to aesthetic posts.
However, careful review of the literature suggests:
1. Generally, placement of a post is indicated whenever the coronal tooth structure is insufficient in
quality or quantity to support a core build-up.
2. Tooth-colored fiber posts are useful because they
provide an aesthetic means of building up a post
and core at the chairside. There is also the
advantage of compatibility between the flexibility of the tooth and post structure.
3. Adhesive techniques should be used in combination with all non-metal fiber posts to maximize
retention and conserve tooth tissue.
4. Ceramic posts should not be used in cases of
extensive tooth tissue loss, as the rigidity of the
ceramic material tends to transfer stresses to the
compromised tooth structure, thereby, enhancing
the risk of root fracture.
5. Posts constructed from different fiber types have
similar flexural strength, therefore, quartz fiber
and other aesthetic fiber posts may be used
with equal confidence.
6. The results of clinical studies to date indicate
that even the smallest diameter fiber posts
appear to perform well without the risk of fracture.
CONCLUSIONS
There have been significant developments in post systems in recent years with respect to post and core construction materials, post shape and design, bonding
systems and techniques for removal. Carbon fiber posts
have generally been superceded by quartz, silica and
glass fiber-reinforced materials. One advantage of fiberreinforced systems is that the modulus of elasticity of the
post is similar to tooth tissue; hence, post failure should
occur before root fracture under conditions of stress.
The flexural strength of fiber posts was found to be similar to metal posts as long as contact with water was
avoided. As a consequence, many systems, including
posts of small diameters (such as 0.8 or 1 mm), have
replaced older systems and have been recommended for
use in roots with narrow mesio-distal diameters, such
as mandibular premolars. An area of concern may be
related to the finding that fiber posts can undergo
degradation in the face of repeated mechanical loading
and under conditions of moisture; this degradation may
lead to a reduction in the modulus of elasticity and flex-
89
ural strength with an increased risk in debonding. This

is not likely to occur if stiff materials such as ceramics
are used. The nature of the bond between resin composite and fiber or ceramic post to root dentin and post
must also be taken into account. If the bond fails at
either interface, transference of forces to the root dentin
will be affected. As indicated above, bonds of composite
to carbon fiber post materials have been found to be
stronger than to zirconium dioxide. Failure of this bond
may prevent root fracture of teeth restored with ceramic posts. It follows that the question of whether posts of
high modulus of elasticity should be used compared to
those of a lower modulus cannot be answered in light of
the current clinical and laboratory research. The retention rate of fiber-reinforced posts appears to be similar
to other post systems and early clinical findings are
favorable.
A range of ceramic post systems has been described;
the main disadvantage of these relates to the inherently
brittle nature of ceramics and it can be difficult to
remove a fractured ceramic post. There are a variety of
construction methods, some involve exacting laboratory
procedures.
With the introduction of ceramic and fiber-reinforced
posts, there has also been a trend away from the parallel-sided post preparation previously associated with
the need for retentive mechanical features required for
metal posts and more towards a shorter, tapered design
in harmony with the canal morphology.
An additional advantage to the newer systems from
the patients perspective is that metal post and cores
may be realistically avoided. However, long-term, evidence-based clinical studies are required before the performance of fiber-reinforced and ceramic post and core
systems can be fully assessed.
(Received 19 March 2002)
References
Ahmad I (1998) Yttrium-partially stabilized zirconium dioxide
posts: An approach to restoring coronally compromised non-vital
teeth International Journal for Periodontology and Restorative
Dentistry 18(5) 454-465.
Anusavice KJ (1996) Phillips Science of Dental Materials 10th
Edition Philadelphia WB Saunders p 590-596.
Asmussen E, Peutzfeldt A & Heitzmann T (1999) Stiffness, elastic
limit and strength of newer types of endodontic posts Journal of
Dentistry 27(4) 275-278.
Dean JP, Jeansonne BG & Sarkar N (1998) In vitro evaluation of a
carbon fibre post Journal of Endodontics 12(12) 807-810.
de Rijk WG (2000) Removal of fiber posts from endodontically
treated teeth American Journal of Dentistry 13(Special No)
19B-21B.
90
Operative Dentistry
Drummond JL (2000) In vitro evaluation of endodontic posts

American Journal of Dentistry 13(Special No) 5B-8B.
Mannocci F, Sherriff M & Watson TF (2001) Three point bending

test of fibre posts Journal of Endodontics 27(12) 758-761.
Duret B, Reynaud M & Duret F (1990) [Un nouveau concept de

reconstruction corono-radiculaire: Le composiposte] (1) Chirurg
Dent France 540 131-141.
Martinez-Insua A, da Silva L, Rilo B & Santana U (1998)

Comparison of the fracture resistance of pulpless teeth restored
with a cast post and core or carbon fibre post with a composite
core Journal of Prosthetic Dentistry 80(5) 527-532.
Duret B, Reynaud M & Duret F (1992) [El structure unidirectionnelle dans les reconstructions corono-radiculaires] Journal de
Biomateriaux Dentaires 7 45-47.
Fauchard P (1880) The Surgeon Dentist 2nd ed, Vol II Birmingham,
Alabama (Reprinted by the Classics of Dentistry Library) p 173-204.
Ferrari M, Vichi A, Mannocci F & Mason PN (2000) Retrospective
study of the clinical performance of fiber posts American
Journal of Dentistry 13(Special No) 9B-13B.
Ferrari M, Vichi A & Garca-Godoy F (2000) Clinical evaluation of
fiber-reinforced epoxy resin posts and cast post and cores
American Journal of Dentistry 13(Special No) 15B-18B.
Fovet Y, Pourreyron L & Gal JY (2000) Corrosion by galvanic coupling between carbon fibre posts and different alloys Dental
Materials 16(5) 364-373.
Fredriksson M, Astbck J, Pamenius M & Ardivson K (1998) A retrospective study of 236 patients with teeth restored by carbon
fiber-reinforced epoxy resin posts Journal of Prosthetic Dentistry
80(2) 151-157.
Glazer B (2000) Restoration of endodontically treated teeth with
carbon fibre postsa prospective study Journal of the Canadian
Guzy GE & Nicholls JI (1979) In vitro comparison of intact
endodontically treated teeth with and without endo-post reinforcement Journal of Prosthetic Dentistry 42(1) 39-44.
Harris CA (1839) The Dental Art Baltimore Armstrong & Berry p
305-347.
Isaac DH (1997) Engineering aspects of fibre reinforced composites
in Vallittu PK The first symposium on fibre reinforced plastics
in dentistry Turku, Finland Biomaterials Project Institute of
Dentistry 1-21.
Isidor F, Odman P & Brondum K (1996) Intermittent loading of
teeth restored using prefabricated carbon fibre posts
Johnson JK & Sakumara JS (1978) Dowel form and tensile force
Kern M & Knode H (1991) Posts and cores fabricated out of inceram-direct and indirect methods [Stiftkernaufbaten aus InCeram-Direkte und Indirekte Methode] Quintessenz
Zahntechnic 17(8) 917-925.
Koutayas SO & Kern M (1999) All-ceramic posts and cores: The
state of the art Quintessence International 30(6) 383-392.
Kwiatkowski S & Geller WA (1989) Preliminary consideration of
the glass ceramic dowel post and core International Journal of
Prosthodontics 2(1) 51-55.
Mannocci F, Innocenti M, Ferrari M & Watson TF (1999) Confocal
and scanning electron microscopic study of teeth restored with
fiber posts, metal posts and composite resins Journal of
Endodontics 25(12) 789-794.
Mannocci F, Ferrari M & Watson TF (1999) Intermittent loading of
teeth restored using quartz fibre, carbon quartz fibre and zirconium dioxide ceramic root canal posts Journal of Adhesive
Dentistry 1(2) 153-158.
Miettinen VM, Narva KK & Vallittu PK (1999) Water sorption, solubility and effect of post-curing of glass fibre reinforced polymers Biomaterials 20(13) 1187-1194.
OKeefe K, Miller BH & Powers J (2000) In vitro tensile bond
strength of adhesive cements to new post materials
Pissis P (1995) Fabrication of a metal-free ceramic restoration utilizing the monobloc technique Practical Periodontics and
Aesthetic Dentistry 7(5) 83-94.
Purton DG, Love RM & Chandler NP (2000) Rigidity and retention
of ceramic root canal posts Operative Dentistry 25(3) 223-227.
Purton DG & Payne JA (1996) Comparison of carbon fibre and
stainless steel root canal posts Quintessence International 27(2)
93-97.
Roberts DH (1970) The failure of retainers in bridge prostheses. An
analysis of 2000 retainers British Dental Journal 128(3) 117-124.
Rosensteil SF, Land MF & Fujimoto J (1995) Restoration of the
Endodontically Treated Tooth in Contemporary Fixed
Prosthodontics 2nd edition St Louis Mosby 238-267.
Rosentritt M, Furer C, Behr M, Lang R & Handel G (2000)
Comparison of in vitro fracture strength of metallic and tooth
coloured posts and cores Journal of Oral Rehabilitation 27(7)
595-601.
Ross RS, Nicholls JI & Harrington GW (1991) A comparison of
strains generated during placement of five endodontic posts
Journal of Endodontics 17(9) 450-456.
Sandhaus S & Pasche K (1994) [Tenon radiculaire en zircone pour
la ralization dinlays-cores tout cramique] Dental Tribune 2
17-24.
Schillingburg HT & Kessler JC (1982) Restoration of the
Endodontically Treated Tooth Chicago Quintessence.
Silness J, Gustavsen F & Hunsbeth J (1979) Distribution of corrosion products in teeth restored with metal crowns retained by
stainless steel posts Acta Odontologica Scandinavia 37(6) 317321.
Sorensen JA & Martinoff JT (1984) Intracoronal reinforcement
and coronal coverage: A study of endodontically treated teeth
Standlee JP, Caputo AA, Collard EW & Pollack NH (1972) Analysis
of stress distribution by endodontic posts Oral Surgery Oral
Medicine Oral Pathology 33(6) 952-960.
Tomes J (1848) Dental Physiology and Surgery London JW Parker
p 319-321.
Torbjner A, Karlsson S & Odman PA (1995) Survival rate and failure characteristics for two post designs Journal of Prosthetic
Dentistry 73(5) 439-444.
Torbjrner A, Karlsson S, Syverud M & Hensen-Pettersen A (1996)
Carbon fiber reinforced root canal posts: Mechanical and cytotoxic properties European Journal of Oral Science 104(5-6) 605611.

Turner CH (1982) Post-retained crown failure: A survey Dental
Update 9(4) 221, 224-226, 228-229.
Viguie G, Malquarti G, Vincent B & Bourgeois D (1994) Epoxy/carbon composite resins in dentistry: Mechanical properties related
to fiber reinforcements Journal of Prosthetic Dentistry 72(3)
245-249.
91
Weine FS, Wax AH & Wenckus CS (1991) Retrospective study of

tapered smooth post systems in place for ten years or more
Journal of Endodontics 17(6) 293-297.
Invited Paper
Minimal Intervention Dentistry:

Rationale of Cavity Design
GJ Mount
Clinical Relevance
This discussion draws attention to a growing understanding of adhesion of restorative

materials that is leading to possible modifications in cavity design. The original designs
postulated by GV Black were designed for non-adhesive materials and have not been substantially modified in recent years despite improvements in adhesive dentistry. It is
accepted that these designs lead to a weakening of the tooth crown and a continuing need
for replacement and repair.
It is suggested that if cavity designs for new lesions are limited, the potential for retention of significant amounts of natural tooth structure avoids or at least delays the need for
more extensive restorations. However, it is essential to eliminate disease in the first place
if this strategy is to succeed. Caries is a bacterial disease, and methods of control have
been improved to the extent that eliminating bacteria and remineralization and healing
demineralized tooth structure is now a real possibility.
Modified cavity designs are suggested to treat new lesions based upon a new classification of carious lesions. This is a departure from the original GV Black classification that
defined a series of cavities of specific design based on the requirements for specific restorative materials. It is suggested that the cavitys design should be dictated solely by the
extent of the lesion with retention of the restoration being dependent upon adhesion to the
remaining tooth structure.
INTRODUCTION
Black (1917) laid down the basic tenets for the design of
cavities prescribed for restoring carious lesions nearly
100 years ago. At that time a limited range of materials
was available for restoration and there was little
understanding of the disease, itself. Some degree of bacterial involvement in the development of a carious
lesion became apparent in the 1880s (Miller, 1883),
however, the concept of preventive dentistry was in its
infancy. The involvement of fermentable carbohydrate
was not fully understood, with another 50 years passing
*Graham J Mount AM, BDS, DDSc, FRACDS
*Reprint request: 13 MacKinnon Parade, North Adelaide, South
Australia 5006; e-mail: [email protected]
before the influence of fluoride became apparent

(Brudevold, 1974).
At the time of Blacks writing, he was working in an era
of rampant caries. He laid down a series of principles
that set a standard that has continued to this day. He
suggested that convenience form or access to the lesion
was a primary need. He defined the outline form, which
included all areas of the tooth that were at risk, such as
the remaining, uninvolved areas of fissures and extending the outline into areas that were expected to be free
of further caries attack.
As the materials that Black worked with were not
adhesive, it became necessary for him to describe retention form in order for the restoration to not fall out.
Mount: Minimal Intervention Dentistry: Rationale of Cavity Design

Then, he had to consider the relative strength of the
restorative material itself. There was also a need for
resistance form, which meant that on some occasions,
more tooth structure needed to be removed in order to
lend strength to the filling so that it could resist
occlusal stresses.
One of the greatest problems that Black faced was a
lack of understanding of how rapidly the lesion progressed through enamel and, subsequently, through
dentin. He worked on the basis that once demineralization had commenced on the surface of the crown of the
tooth, it was not possible to reverse it. That is, it was
not possible to heal a lesion. The result of this misunderstanding was that treating even the smallest lesion
detectable on a radiograph would lead to destruction of
a significant amount of natural tooth structure, much
of which was still in perfect health.
It is interesting to note that more than 100 years
later, following considerable change in dental materials, clinicians understanding of the caries process and
the rate of speed at which the carious lesion progresses, we continue to use the cavity designs that Black
specified. There have been modifications in design
toward conservation of tooth structure and a better
understanding of materials suggested over the years
(Markley, 1966) but too much emphasis has been placed
on the surgical approach to eliminating the disease.
There is a need to reevaluate the use of surgery as the
primary approach to eliminating what is essentially a
bacterial disease.
HEALING OF THE CARIOUS LESION
For many years caries has been recognized as a bacterial disease, so that in the absence of the relevant bacteria, caries cannot be initiated or continue
(Balakrishnan, Simmonds & Tagg, 2000). If, in the
early stages of a caries attack, the bacterial flora can be
modified, then it is possible to overcome the initial demineralization and heal the lesion through remineralization. If the lesion has progressed to the extent that
there is actual cavitation on the otherwise smooth surface of the tooth crown, then it becomes necessary to
reinstate that smooth surface by placing some form of
restoration (Mount & Hume, 1998a).
Streptococcus mutans is acknowledged as the bacteria most involved in the reduction of the pH of plaque
on tooth surface to the extent that demineralization can
occur (Silverstone & others, 1981). In the presence of
fermentable carbohydrate, these bacteria can reduce
the pH of plaque fluid to levels below 5.5, which is the
critical pH for demineralizing hydroxyapatite in vitro.
In vivo, the level of saturation will have a bearing on
the critical pH. This defines the fact that fermentable
carbohydrate is a significant factor in the development
of caries. Both factors have been known and understood
93
for many years and provide some basis for good preventive dentistry.
Control of fermentable carbohydrate is a very personal factor, and patients need to understand the significance of the role they play in its control. On the other
hand, control of the bacterial flora is relatively straightforward but not often recommended to patients. The
profession needs to identify the patient who is at risk
with a high Streptococcus mutans count and prescribe
chlorhexidene in the correct manner to assist the
patient in taking control (Walsh 2000). Using 0.2%
chlorhexidene as a water-based gel or an alcohol-based
mouthwash twice a day for two weeks will offer a significant reduction in streptococcus mutans and lactobaccilus and provide the patient with an increased
opportunity to overcome infection. Following this,
chlorhexidene can be used as a mouthwash at weekly
intervals for long periods if the patient is incapacitated.
Obviously, this alone will not be sufficient to gain total
control, and all the usual measures that limit the
intake of fermentable carbohydrates and oral hygiene
will also be essential. The patient, in fact, is the only
one who can overcome the problem, leaving the profession with the responsibility of seeing that patients are
fully educated in these areas.
The real significance for having a better understanding of the progress of the carious lesion lies in the fact
that it is now obvious that the disease should be controlled through preventive measures first, prior to any
restorative measures being undertaken. Surgery will
only be necessary in the case where the tooth surface
has been cavitated because, in the presence of surface
defects, it will not be possible to completely control
plaque accumulation. However, the concept of extension for prevention is out of date because the remaining, partially demineralized enamel that surrounds the
cavity can be remineralized and healed. It is only necessary to control the bacterial flora in order for the disease to be controlled. The significant factor that cannot
be overlooked is that, if the disease is allowed to continue unchecked, no restorative material or technique
can, by itself, bring about control.
THE PLACE OF FLUORIDE IN CARIES CONTROL
Fifty years ago it became apparent that in the presence
of free fluoride ions the progress of the carious lesion
could be slowed down (Silverstone, 1982). It was decided that if fluoride were to be included in the developing
tooth structure, there would be a degree of resistance to
demineralization imparted onto the enamel.
Fluoridating water supplies and introducing fluoride
into toothpaste made a significant difference in the
caries rates in children at a community level (Axelsson,
1999). Subsequently, adults also derived benefit, particularly following topical applications (Axelsson, Lindhe
94
Operative Dentistry
& Nystrom, 1991). It is now recognized that in the continuing presence of low levels of fluoride in plaque fluid,
demineralization of tooth surface can be slowed down,
prevented or reversed (Clarkson & others, 1996).
It appears that fluoride can act as a catalyst for preventing demineralization in the first place and stimulating remineralization of tooth structure following the
initiation of caries (Ten-Cate, 1990; Kidd & JaystonBechal, 1987). In a substitution reaction, fluoride can be
incorporated into enamel and form fluorhydroxyapatite;
this shows a greater resistance to acid attack because
the critical pH of this form of enamel is pH 4.5 in contrast
to hydroxyapatite, which is pH 5.5. It has been shown
that fluoride is more effective post-eruption than preeruption.
The formation of some fluorhydroxyapatite can occur
during tooth development if a level of about 1-ppm of
fluoride is available in the water supply for the growing
infant. However, it is also possible to modify the surface
of the tooth and develop a layer of fluorhydroxyapatite at
any time following eruption. This is significant because,
in the presence of a restorative material that is capable of
releasing fluoride ions, it may be possible to exert some
degree of influence on further demineralization of tooth
structure adjacent to the restoration (Serra & Cury, 1982).
It was recognized many years ago that silicate cement
released a certain amount of fluoride, leading to some
degree of defense against further demineralization as
the restoration broke down. Now, the modern glass
ionomer has been shown to have the same capability
and is therefore a useful material in the defense of the
tooth against further caries attack (Hicks, 1986). The
transfer of fluoride from the restoration to the adjacent
tooth surface will help in the formation of fluorhydroxyapatite, making this area more resistant to further
acid attack. The modification of enamel will not, by
itself, prevent new lesions but will make the tooth
structure surrounding a glass-ionomer restoration
more resistant to further demineralization.
ADHESIVE WITH RESIN COMPOSITE
Since dental caries is a bacterial disease, it is apparent
that any situation in the oral environment that encourages the accumulation and retention of bacterial
colonies is undesirable. The problem of microleakage
forming between the restoration and the cavity wall has
demanded much attention and a high level of clinical
skill has been required to prevent it. When completely
neutral materials such as amalgam and gold are used
for restoration, it is essential to pay attention to both
cavity design and placement of the restoration.
Fortunately, amalgam corrodes to a degree (Mahler &
Nelson, 1984) and the corrosion products themselves
tend to occlude the interface. Over time, they will completely seal the restoration. A gold restoration will be
placed and retained with a luting cement and, providing

this is not soluble in the oral fluids, will prevent
microleakage for the long-term.
The advent of adhesive materials has made it easier
to seal the cavity margin but there is still a need to pay
attention to cavity design. Adhesion with resin composite
is based primarily on a micro-mechanical attachment
between the enamel and the restoration. Fifty years
ago, it was shown that following acid etching of the
enamel surface, the tooth would be sufficiently porous
to allow a low viscosity resin to penetrate some distance
into the enamel (Buonocore, 1955). Once the resin had
set, there was a very strong union between the two.
However, there are certain limitations in this technique because the union may be under stress from setting shrinkage of the resin. It is essential that the marginal enamel be fully mineralized. It has been noted
that adhesion will be enhanced if the margins are
beveled so that the resin can bond to the ends of the
enamel rods rather than along their long axis (Ikeda &
others, 2002).
Considerable research into the concept of adhesion
between resin composite and dentin has been conducted. The current concept is to clean the cavity floor down
to completely healthy dentin, then etch the surface of
the dentin, leaving demineralized collagen fibers standing free on the floor (Sano & others, 1999). A low viscosity resin is then flowed over the surface and, when
set, it will engage the collagen in a micro-mechanical
union. There are some clinical and scientific problems
with this system and long-term success has not yet
been proven (Van Meerbeek & others, 1994). It can also
be argued that removing all demineralized dentin from
the floor of the cavity will often lead to an over-extension of the cavity because some of this dentin could be
remineralized and healed (Ten Cate & van Duinen, 1995).
ADHESION WITH GLASS IONOMER
Adhesion with glass ionomer is the result of an ion
exchange between the cement and both enamel and
dentin. There can also be a degree of chemical union
with exposed collagen fibers (Mount, 1991; Ngo, Mount
& Peters, 1997; Akinmade, 1994; Ferrari & Davidson,
1998). This suggests that even in the presence of demineralized tooth structure, there will still be a degree of
union. It will therefore only be necessary to extend the
cavity outline to the point where the enamel surface is
smooth and non-cavitated even though it may be partially demineralized. In addition, partly demineralized,
non-infected dentin that remains on the floor of the cavity can be assisted to remineralize (Ngo & others, 2001).
Extension of the cavity outline can therefore be very
conservative, leading to retention of natural tooth
structure and limitation of the aesthetic and physical
problems that arise from the larger cavity preparation.

There are further advantages to be gained from using
glass ionomer even if it serves only as a base for another restorative material. Glass ionomer has been shown
to be bioactive and have a long-term ion exchange with
tooth structure to the extent that it can assist in the
remineralization of dentin and enamel (Ngo, Marino &
Mount, 1998). It will exchange fluoride and calcium
and strontium and phosphate, as well. This means that
it is not necessary to remove all softened, partially demineralized dentin from the floor of the cavity, which
allows for even more conservative cavity designs. The
concept is to gain access to the carious lesion in the
most conservative manner possible. The axial wall or
pulpal floor can remain as demineralized dentin provided the surface infected layer has been removed. The
enamel margin should be extended out only far enough
to remove all completely degenerated enamel rods.
Glass-ionomer materials allow clinicians to be more
conservative in their cavity preparations. Currently, its
only limitation appears to be a relative lack of fracture
resistance to shear and tensile stresses to the extent
that it is not indicated for rebuilding marginal ridges or
incisal corners. However, provided that glass ionomer is
well supported by surrounding tooth structure, it has a
satisfactory history of longevity. In fact, it remains
bioactive throughout its life, continuing to absorb and
release fluoride (Nicholson, Czarnecka & LimanowskaShaw, 1999; Ngo & others, 1998).
CAVITY DESIGN MODIFICATIONS
It is apparent from the above discussion that it should
now be possible to review the GV Black approach to
cavity design and be far more conservative in removing
natural tooth structure. There is no doubt that introducing the rapid reduction of sound tooth structure
with the air rotor handpiece has led to excessive extension of cavities.
Minimal intervention cavity designs have been discussed for more than 20 years (Knight, 1984; Hunt,
1984), and a new classification that encourages the profession to see operative dentistry in a new light has
been proposed (Mount & Hume, 1997). The GV Black
classification does not address this new philosophy,
thus, it is in the interest of both the patient and operator to adopt a new method.
The proposed classification takes into account the fact
that there are only three surfaces of the crown of a
tooth that can be subject to caries attacks. These surfaces are:
Site 1 - pits and fissures on the occlusal surface of posterior teeth and other defects on otherwise smooth
enamel surfaces.
Site 2 - contact areas between any pair of teeth, anteriors or posteriors.
95
Site 3 - cervical areas related to gingival tissues,

including exposed root surfaces.
A neglected lesion will continue to extend as an area
of demineralization in relation to one of the sites noted
above. As it extends, so will the complexities of the
restoration increase. The sizes that can be readily identified include:
1. Size 0 - initial lesion at any site can be identified
but has not yet resulted in surface cavitation. It
can possibly be healed.
2. Size 1 - smallest minimal lesion requiring operative intervention. The cavity is into dentin just
beyond healing through remineralization.
3. Size 2 moderate-size cavity. There is still sufficient sound tooth structure to maintain the
integrity of the remaining crown.
4. Size 3 the cavity needs to be modified and
enlarged to provide some protection for the
remaining crown from the occlusal load. There is
already a split at the base of a cusp or, if not protected, a split will likely develop.
5. Size 4 the cavity is now extensive, following the
loss of a cusp from a posterior tooth or an incisal
edge from an anterior.
The Size 0 lesion represents identifying an area of
demineralization on any of the three surfaces, and,
with normal preventive measures as discussed above,
the lesion is expected to be arrested and healed. The
problem then becomes significantly greater when the
classification moves to a Size 1 because it is acknowledged that healing is no longer a proposition and some
form of surgical intervention is required. However, it is
suggested that in the presence of adhesive, bioactive
restorative materials, it should not be necessary to do
much more than simply seal the lesion from further
bacterial activity so that the lesion will no longer
progress.
As the lesion extends and the classification moves
into Size 2 and beyond, other problems will be superimposed. It will become necessary to take into account
the occlusal load, the intrinsic strength of the restorative material, the strength of the remaining tooth structure and its ability to accept functional load. The larger the cavity, the greater the need to use the stronger
restorative materials. There will come a point where
the cavity design has to be modified to the extent that
the restoration, rather than the remaining tooth structure, will accept the entire occlusal load. These modifications apply particularly to replacement dentistry,
where a failed restoration needs to be replaced.
However, in treating the early, new lesion, it is usually possible to minimize removing tooth structure, thus,
maintaining the natural strength of the crown and
96
keeping the load off the restoration. Suitable designs for
the Size 1 and Size 2 cavities are discussed below
because this is where adhesive materials can be used to
their optimum potential.
Site 1, Size O
The concept of the fissure seal, as discussed by
Simonsen (1989) and others, is particularly sound in a
newly erupted tooth. Sealing a deep fissure before it
becomes partially occluded by plaque and pellicle, and in
advance of demineralization into dentin, has an acceptable clinical history (Feigal, 1998; Ekstrand & others, 1998).
The earliest fissure sealants were unfilled or lightly
filled resins, but recent research has shown that there
are some doubts about the integrity of the acid etch
union between resin and enamel in these regions. It has
been shown that a glass ionomer will successfully
occlude such a fissure (Wilson & McLean, 1988). This is
now being termed fissure protection to differentiate it
from a resin seal.
The anatomy of enamel within a fissure differs from
that of other surfaces in that it is covered with a
layer of enamel rods that appear to run parallel with
the surface rather than at right angles. This means
that when it is etched with orthophosphoric acid, it will
not develop the usual pattern of porous enamel that
allows penetration of the unfilled resin that is normally
relied upon to provide the micromechanical attachment
(Burrow, Burrow & Makinson, 2001). The presence of
this type of enamel may well account for loss of the
resin seal in many cases. Neither a resin nor a glass
ionomer will flow into a fissure beyond the point where
the fissure narrows down to approximately 200 m in
width. Therefore, retention of both materials appears to
be dependent on adhesion to enamel at the entrance to
the fissure rather than mechanical interlocking into the
complexities of the fissure. Recent work suggests that
even though the enamel rods lie in a different orientation, glass ionomer will still develop ion exchange adhesion and show acceptable longevity (Mount & Hume,
1998b).
Site 1, Size 1
As the fissure walls become demineralized, the dentin
will become involved as well. This may pose a rather
dangerous situation because there is often some difficulty
in diagnosing the presence of a dentin lesion.
Radiographs will not show this early lesion very clearly and
laser detector and electrical impedance machines have
limitations. In the presence of strong, fluoridated enamel,
the occlusal surface entry to the lesion will remain limited, and bacteria-laden plaque can be forced down into a
defective fissure. Under these circumstances, dentin involvement can become advanced before symptoms are noted.
The fissure system is a complex series of pits and fissures, therefore, a carious defect will often be limited to
a very restricted area, leaving the remaining fissure
Operative Dentistry
system sound and uninvolved. This means that only the
carious defect needs to be instrumented. However, prudence suggests that minor apparent defects should be
explored in a very conservative manner before sealing
the fissure system.
Site 1, Size 2
In this classification, the lesion will either have progressed to some degree or it may represent replacement
of a failed Class I restoration. The same conservative
principles should apply, as discussed above, in as much
as it is only necessary to deal with the carious lesion
and there is no need to open up the remaining fissures
any further. If there is any part of the fissure system
that is in doubt, it can be explored very conservatively,
but there is no doubt that it is sufficient to seal the fissures and any carious process below will be arrested. It
will progress no further until there is again access to
the usual nutrients required by the bacteria (MertzFairhurst & others, 1992). That is, if there is any marginal leakage, there will be further bacterial activity,
which is very unlikely when using glass ionomer
because of ionic adhesion and the presence of fluoride
release. Instrumentation and restoration techniques for
these lesions will be the same as for a Size 1 lesion.
However, the occlusal involvement will be more extensive and, if there is any doubt about the ability of the
glass ionomer to withstand the occlusal load, it can be
cut back conservatively and laminated with resin composite.
It should be noted that glass ionomer has been recommended for the restoration of both Size 1 and Size 2
lesions in this category. The restoration is well supported by the remaining tooth structure and the ion
exchange adhesion will ensure complete sealing of the
remainder of the cavity. This means that if there is any
demineralized dentin remaining on the floor of the cavity, there will be no further carious activity and there is
a potential for remineralization (Ngo & others, 2001). It
is possible to use a resin composite for the restoration
but that would also mean cleaning the floor down to
sound, healthy dentin to develop an acid-etch union
with fully mineralized tooth structure. This may mean
removing dentin that could otherwise be remineralized
and healed.
Site 2, Size 0
It should be noted that radiographic evidence of demineralization at the contact area does not necessarily
mean that there is cavitation on the proximal surface
and, in the absence of cavitation, it is often possible to
heal the lesion. In fact, proximal lesions progress very
slowly because that surface is not under masticatory
load and is, to a degree, protected from traumatic damage (Pitts, 1983; Shwartz & others, 1984). In contrast to
the occlusal fissure lesion, it may take up to four years
to penetrate the full thickness of the enamel and an

additional four years to progress through the dentin to
the pulp.
It is desirable to differentiate between the Size 0 and
Size 1 lesion before surgery because, at least theoretically, it should be possible to heal the Size 0 and it is
only when cavitation is established that a surgical
technique is required. It is essential to avoid the use of
a probe to explore the proximal surface because this is
the quickest way to actually damage the enamel and
cause a cavity.
Site 2, Size 1
Once it has been established that there is cavitation on
the proximal surface, a surgical approach to its repair
becomes essential and some alternative methods are
available. First, determine the position of the damage
in relation to the crest of the marginal ridge. If it is
more than 2.5 mm below the crest, then it may be possible to approach the lesion through the occlusal fossa
and design a tunnel cavity (Hasselrot, 1998; Wilson &
McLean, 1988). On the contrary, if it is less than this
distance, a tunnel will only undermine the marginal
ridge and weaken it still further. Under these circumstances, it is better to design a small box or slot cavity beginning on the outer slope of the ridge, retaining
as much of the enamel as possible. Occasionally, a further alternative will present itself when a large Site 2,
Size 3 or 4 lesion is being repaired or replaced and a
small Size 1 lesion is revealed on the side of the adjacent tooth. These three alternatives will be discussed in
more detail.
Site 2, Size 1 - Tunnel
As discussed above, the early proximal lesion on a posterior tooth will commence in enamel immediately
below the contact area because this is where plaque will
accumulate and mature. Initially, the contact itself will
remain plaque-free due to movement between the
teeth. The level and depth of fluorapatite already in the
enamel may help to control the speed at which the
enamel will actually undergo demineralization. Often,
particularly in fluoridated communities, the enamel,
although demineralized, will remain relatively intact
until the dentin lesion is quite advanced. It will take up
stain and become disfigured, but in the presence of further fluoride, it may remineralize.
As the lesion develops, some degree of breakdown and
cavitation of the enamel will eventually occur, but this
will remain confined to the area below the contact until
it is quite advanced. There will generally be a zone of
demineralized enamel surrounding the cavitation, but
as long as the surface is smooth, this remains capable
of remineralization in the presence of fluoride (Serra &
Cury, 1992). The contact area may remain sound and
the marginal ridge may be quite strong, provided the
lesion is more than 2.5 mm below the crest of the mar-
97
ginal ridge (Wilson & McLean, 1988). Access to the

lesion through the occlusal surface should be limited to
the extent required to achieve visibility and, where possible, should be undertaken from an area that is not
under direct occlusal load (Knight, 1984). For most
patients, there is a fossa immediately medial to the
marginal ridge that is the most suitable position for initial entry and, in a normal occlusion, is often not an
area of occlusal contact.
It should be noted that resin composite is not indicated for restoration of these lesions because it will not be
possible to access the proximal lesion to a sufficient
degree to be able to reliably remove all demineralized
enamel. Also, it will not be possible to provide a beveled
margin to ensure proper adaptation of the resin to the
enamel. On the other hand, glass ionomer will flow
readily into a small cavity and has the ability to remineralize the enamel margins and any dentin on the
axial wall that may be demineralized.
Site 2, Size 1 - Slot
A slot cavity could be used when the lesion is less than
2.5 mm below the crest of the marginal ridge.
Generally, the lesion will be obvious both radiographically and to visual examination due to discoloration
under the marginal ridge. The basic principles of cavity design remain the same, with the objective of removing only that tooth structure that has broken down
beyond the possibility of remineralization. If this is
allowed to dictate the extent of the cavity, there will be
many occasions with this design where there is still a
sound contact with the adjacent tooth in one part or
another. It is desirable to retain this to ease the problems of maintaining a good, firm contact area, thus
minimizing the dangers of food impaction and retention.
The outline form will be dictated entirely by the
extent of the breakdown of the enamel, removing only
that which is friable and easily eliminated without
applying undue pressure. Remaining demineralized
enamel will generally heal satisfactorily. Retention will
again be through adhesion, so it is only necessary to
clean the walls around the full circumference of the
lesion, leaving the axial wall because it will be affected
by dentin only.
For such a lesion, resin composite may be a useful
material because on many occasions there will be an
enamel margin around the full circumference. This will
allow for the potential of placing a bevel and, therefore,
good sound adhesion. If resin is used, any affected
dentin should be removed so that proper etching of the
dentin with exposure of collagen will occur. However,
glass ionomer is still a sound option because the occlusal
load will not be great and the ion exchange will remain
valuable both for adhesion and remineralization.
98
Operative Dentistry
Site 2, Size 1 - Proximal Approach

A further, very conservative approach to restoring a
proximal lesion can be achieved on limited occasions only
when the proximal surface of a tooth becomes accessible
at the time of cavity preparation in an adjacent tooth. The
lesion may have been revealed through radiographs or it
may be noted only during cavity preparation. The larger
cavity in the adjacent tooth will normally need to be of
reasonably generous proportions to allow room to
manoeuver, but when such an approach is possible, it leads
to considerable conservation of natural tooth structure.
In view of the normal direction of the progress of a
carious lesion through the enamel and down the dentin
tubules, it is not difficult to clean the cavity, trim the
enamel walls and eliminate infected dentin. Note that
it is only necessary to remove enamel that is broken
down beyond remineralization. There will often be a
residual area of demineralized enamel around the circumference of the lesion and this should be retained
because it is quite capable of being remineralized and
healed. As this entire restoration will probably be hidden and disguised by the restoration in the adjacent
tooth, it is essential to use a radiopaque material.
Resin composite can be useful for a very small lesion
but glass ionomer is preferred because the limited
access will make it difficult to assure full polymerization of the resin through light activation. When resin is
used, it would be appropriate to use a dual-cured resin.
CONCLUSIONS
It is apparent that the advent of adhesive and bioactive
restorative materials has opened the way for a review
of the surgical approach to restoring cavitated carious
lesions. The micromechanical adhesion between enamel and resin composite is strong and very valuable but,
to be fully effective, it depends upon the availability of
sound, fully mineralized enamel that is well supported
by healthy dentin. Occasionally, this may mean removing some areas of enamel that could otherwise have
been remineralized and healed. For current techniques,
it also requires complete removal of all demineralized
dentin from the floor of a cavity, some of which can generally be remineralized and healed. The use of glass
ionomer allows for a more conservative approach.
Glass ionomer is the most conservative of the restorative materials but requires good support from the
remaining tooth structure. It has a bioactive ion
exchange union with tooth structure that has been
shown to last for long periods (Mount, 1997). There is
also sufficient ion exchange with both enamel and
dentin to assist in healing demineralized tooth structure
so that it is now possible to be very conservative in
extending the dimensions of a cavity.
In the presence of these changes, it is possible to
review the approach to the control and the restoration
of all new lesions, thus, retaining the strength and aesthetics of teeth in spite of the presence of carious
lesions.
(Received 23 July 2002)
References
Akinmade A (1994) Adhesion of glass-polyalkenoate cement to
collagen Journal of Dental Research 72(Special Issue)
Axelsson P (1999) An Introduction to Risk Prediction and
Preventive Dentistry Quintessence Publishing Co Ltd, Illinois,
USA Chapter 7.
Axelsson P, Lindhe J & Nystrom B (1991) On the prevention of
caries and periodontal disease. Results of a 15-year study in
adults Journal of Clinical Periodontology 13 182-189.
Balakrishnan M, Simmonds RS & Tagg JR (2000) Dental caries
is a preventable infectious disease Australian Dental Journal
45 235-245.
Black GV (1917) A work on operative dentistry; The technical procedures in filling teeth Medico-Dental Publishing Company
Chicago.
Brudevold F (1974) Fluoride therapy in Bernier JL, Muhler JC
eds Improving Dental Practice Through Preventive Measures
Mosby St Louis USA p 77-103.
Buonocore M (1955) A simple method of increasing the adhesion
of acrylic filling materials to enamel surfaces Journal of Dental
Research 5(34) 849-853.
Burrow MF, Burrow JF & Makinson OF (2001) Pits and fissures:
Etch resistance in prismless enamel walls Australian Dental
Journal 46 258-262.
Clarkson BH, Fejerskov O, Ekstrand J & Burt B (1996) Rational
use of fluorides in caries control in Fejerskov O, Ekstrand J &
Burt B (eds) Fluoride in Dentistry Copenhagen Munskgaard.
Ekstrand KR, Ricketts DN, Kidd EA & Schou S (1998) Detection,
diagnosing, monitoring and logical treatment of occlusal caries
in relation to lesion activity and severity: An in vivo examination with histological validation Caries Research 32 247-254.
Feigal RJ (1998) Sealants and preventive restorations: Review of
effectiveness and clinical changes for improvement Pediatric
Dentistry 20 85-92.
Ferrari M & Davidson CL (1998) Interdiffusion of a traditional
glass-ionomer cement into conditioned dentin American
Journal of Dentistry 10 295-297.
Hasselrot L (1998) Tunnel restorations in permanent teeth. A 7year follow-up Swedish Dental Journal 22 1-7.
Hicks MJ (1986) Artificial lesion formation around glass-ionomer
restorations in root surfaces: A histological study Gerodontics 2
108-113.
Hunt PR (1984) A modified Class II cavity preparation for glassionomer restorative materials Quintessence International 15
1011-1018.
Ikeda T, Uno S, Tanaka T, Kawakami S, Komatsu H & Sano H
(2002) Relation of enamel prism orientation to microtensile
bond strength American Journal of Dentistry 15 109-113.
Kidd EA & Jayston-Bechal S (1987) Essentials of dental caries:
The disease and its management Dental Practitioners
Handbook 31 Wright, Bristol.
99
Knight GM (1984) The use of adhesive materials in the conservative restoration of selected posterior teeth Australian Dental
Journal 29 324-331.
Pitts NB (1983) Monitoring of caries progression in permanent

and primary posterior approximal enamel by bitewing radiography Community Dental Oral Epidemiology 11 228-235.
Mahler DB & Nelson LW (1984) Factors affecting the marginal

leakage of amalgam Journal of the American Dental
Association 108 51-54.
Sano H, Inoue S, Noda M, Kawamoto C, Uno S, Takahashi A,

Aoyagi M, Sato Y & Natsuizaka A (1999) Long-term durability
of resin-dentine bonds in Modern Trends in Adhesive
Dentistry: Proceedings of the Adhesive Dentistry Forum 99 in
Tsurumi, Yokohama, Japan Edit-Mamoi Y, Kuraray Co, Osaka
Japan.
Markley MR (1966) Pin-retained and pin-reinforced amalgam

Journal of the American Dental Association 73 1295-1300.
Mertz-Fairhurst EJ, Smith CD, Williams JE, Shearer JD,
MacKert JR Jr, Richards EE, Piers CE, ODell NL, Kovarick
RE & Schuster GS (1992) Cariostatic and ultraconservative
sealed restorations: Six year results Quintessence
International 23 827-838.
Miller WD (1883) The etiology of dental caries Independent
Practitioner (Dental) 4.
Mount GJ (1991) Adhesion of glass-ionomer cement in the clinical environment Operative Dentistry 16 141-148.
Mount GJ (1997) Longevity in glass-ionomer restorations:
Review of a successful technique Quintessence International
28 643-650.
Serra MC & Cury JA (1992) The in vitro effect of glass-ionomer

cement restoration on enamel subjected to demineralization
and remineralization model Quintessence International 23
143-147.
Shwartz M, Grondahl H-G, Pliskin JS & Boffa J (1984) A longitudinal analysis from bitewing radiographs of the rate of progression of approximal carious lesions through human dental
enamel Archives of Oral Biology 29 529-536.
Silverstone LM, Johnson NW, Hardie JM & Williams RAD (1981)
in Dental Caries, Aetiology, Pathology and Prevention
Macmillan, London.
Mount GJ & Hume WR (1997) A revised classification of carious

lesions by site and size Quintessence International 28 301-303.
Silverstone LM (1982) The effect of fluoride in the remineralization of enamel caries and caries like lesions in vitro Journal of
Public Health Dentistry 24 42-53.
Mount GJ & Hume WR (1998a) Preservation and Restoration of

Tooth Structure Mosby International London Chapter 11 p
122.
Simonsen RJ (1989) Cost effectiveness of pit and fissure sealant

at 10 years Quintessence International 20 75-84.
Mount GJ & Hume WR (1998b) Preservation and Restoration of

Tooth Structure Mosby International London Chapter 11 p
129.
Ngo H, Mount GJ & Peters MCRB (1997) A study of glassionomer cement and its interface with the enamel and dentin
using a low-temperature, high resolution scanning electron
microscope technique Quintessence International 28 63-69.
Ngo H, Marino V & Mount GJ (1998) Calcium, strontium, aluminum, sodium and fluoride release from four glass-ionomers
Journal of Dental Research 77 Abstract #75 p 641.
Ngo HC, Frazer M, Mount GJ, Tuisuva J & von Doussa R (2001)
Remineralization of dentine by glass-ionomer, an in-vivo study
Journal of Dental Research 80(Special Issue) #0919 p 641.
Nicholson JW, Czarnecka B & Limanowska-Shaw H (1999) Effect
of glass-ionomer and related dental cements on the pH of lactic acid storage solutions Biomaterials 20 155-158.
Ten-Cate JM (1990) In Vitro studies on the effects of fluoride on

de- and remineralization Journal of Dental Research 69
(Special Issue) 614-619.
Ten Cate JM & van Duinen R (1995) Hypermineralization of dental lesions adjacent to glass ionomer cement restorations
Journal of Dental Research 74 1266-1271.
Van Meerbeek B, Peumans M, Verschueren M, Gladys S, Braem
M, Lambrechts P & Vanherle G (1995) Clinical status of 10
adhesive systems Journal of Dental Research 73 1690-1702.
Walsh U (2000) Preventive dentistry for the general practitioner
Australian Dental Journal 45 76-82.
Wilson AD & McLean JW (1988) Glass-ionomer cement
Quintessence Publishing Co London.
100
Awards
American Academy of Gold Foil Operators
Clinician of the Year Award

Dr Mark J Modjean
Mark J Modjean
t is a distinct pleasure
to present this years
AAGFO Clinician of the
Year Award to Dr Mark J
Modjean, who exemplifies
the qualities that this
award recognizes. Mark is
an outstanding practitioner, educator, study club
member and supporter of
excellence in dentistry
including the use of gold in
all forms. He is also active
in organized dentistry.
Mark is a native of
Minnesota. He was born in St Paul and earned both his
BS (1973) and DDS (1977) from the University of
Minnesota. Mark married his lovely wife, Dr Suzanne
Drost, while in dental school and they celebrated their
26th anniversary this August. Mark is licensed in both
Minnesota and Florida (I assume in hopes of someday
escaping all the snow he has lived with his entire life).
Mark has divided his professional time between private practice and teaching since his graduation from
dental school. He taught two days per week at the
University of Minnesota Dental School for 15 years in
the departments of Oral Anatomy, Occlusion and
Restorative Dentistry, during which time he attained
the rank of Associate Professor. In 1992, Mark decided
to devote full time to private practice but, until recently, continued providing one day per week to the com-
munity clinic sponsored by the Dental School. Now,

after celebrating his 50th birthday, he still donates a
day per month to the clinic.
Dr Modjeans involvement in study club activity
began when he joined the GV Black study club in 1981.
He was an extremely active member, serving as
Secretary/Treasurer from 1985 to 1987 and then as
President in 1988. His study club activity introduced
him to the American Academy of Gold Foil Operators
and he became an active member in 1996 after operating at the annual session in San Antonio. Mark also
served as local arrangements host for the 1998 AAGFO
meeting in Minnesota. He provided excellent support
as well as arranging an outstanding presentation on
the history of the GV Black Study Club, which was celebrating its 100th Anniversary at the time.
Despite his obvious dedication to his profession, Dr
Modjean does manage some personal life. He raced formula cars for eight years and now enjoys restoring cars
as a hobby. Mark also mentioned that his most current
project is losing large sums of money in the stock market, which may adversely affect his ultimate goal of
retiring and living long enough to enjoy it!
Again, it is a pleasure to present the 2002 American
Academy of Gold Foil Operators Clinician of the Year
Award to a well-rounded and dedicated professional.
Congratulations, Dr Mark Modjean!
Alan Osborne
101
Departments
Classifieds:
Faculty Positions
Operative Dentistry accepts appropriate classified
advertisements from institutions and individuals.
Advertisements are run at the following rate: $45.00 for
30 or fewer words, plus $0.75 for each additional word.
Consecutively repeated ads are run at a flat rate of
$50.00. Operative Dentistry neither investigates the
offers being made, nor assumes any responsibility concerning them, and it reserves the right to edit copy and
to accept, delete, or withdraw classified advertisements
at its discretion. To ensure publication in a given issue,
copy must be received 45 days before the publication
date. In other words, copy should be received by 15
November of the preceding year for the JanuaryFebruary issue, and by 15 January, March, May, July,
and/or September for publication in subsequent issues.
Send advertisements to the editorial office identified
inside the front cover.
School of Dentistry
Oregon Health & Science University
The School of Dentistry at the Oregon Health & Science
University is seeking an energetic, progressive, qualified individual for a full-time, tenure-track position as
Chair of the Department of Operative Dentistry. The
Chair will hold the rank of associate or full professor.
The duties of the Chair are to support the missions,
goals and objectives of the School of Dentistry.
Specifically, the Chair will be responsible for strategic
planning, budgeting and staffing the department and
fostering the development of junior faculty. The Chair
will have ultimate responsibility for the pre-doctoral
program in Operative Dentistry. The successful candidate is expected to have demonstrated significant
achievement in teaching, research, service, patient care
and academic management, as well as possess excellent
interpersonal and communication skills. Advanced
training in general or operative dentistry or having an
advanced degree in biomedical sciences is highly
desired. One day per week (0.2 FTE) will be devoted to
participation in the Faculty Dental Practice. OHSU is
an Equal Employment Opportunity institution.
Interested candidates should submit a letter, curriculum vitae and references to Dr Jack L Ferracane,
Department of Biomaterials and Biomechanics, Oregon
Health & Science University, 611 SW Campus Drive,
Portland, OR 97239-3097 ([email protected]).
University of Iowa
College of Dentistry
The University of Iowas College of Dentistry is conducting a search for a full-time clinical or tenure track
faculty member in the Department of Operative
Dentistry. Major responsibilities include teaching operative dentistry to predoctoral/postdoctoral students,
research and intramural practice. Position available
July 1, 2003; screening begins immediately. Must have
DDS/DMD from an ADA-accredited institution or a foreign dental degree with certification or Masters degree
in operative dentistry from an ADA-accredited institution. Desirable qualifications include teaching experience in operative dentistry, background in clinical
esthetic dentistry, dental research/training experience
and clinical practice experience. Rank/track/salary commensurate with qualifications/experience. Submit CV
and three letters of recommendation to Dr Gerald
Denehy, 229 Dental Science Building South, College of
Dentistry, University of Iowa, Iowa City, IA 52242.
AA/EEO employer; women/minorities encouraged to
apply.
Announcements
32nd ANNUAL MEETING of the
ACADEMY OF OPERATIVE DENTISTRY
26-28 February 2003
Fairmont Hotel, Chicago, IL
The Academy of Operative Dentistrys 32nd Annual
Meeting once again offers an incredible group of essayists, an outstanding table clinic session and a wonderful social program.
SCIENTIFIC SESSION: Thursday begins with Dr
Sasha Jovanovic speaking on Optimal Esthetics with
Implant Dentistry, followed by Dr Jimmy Eubanks discussing Occlusion and Restoration Design. This years
Buonocore Memorial Lecturer is Dr Bart Van
Meerbeek, who will present Bonding to Tooth Tissue:
Current Status and Challenges of the Future.
Thursday afternoon features Dr William Buddy
Moppers presentation on The Efficacy of Veneering
with Direct Bonding and Dr Shane White explains the
new model of enamel microstructure in Enamel and
DEJ: Structure, Function and Why We Need to
Preserve It.
Dr Richard D Tucker leads off on Friday morning with
Cast Gold Restorations with Integral Pins, and Dr.
Edward McLaren follows with Ceramic Systems:
102
Material Considerations and Selection Criteria.
Finally, Dr Bruce W Small wraps up the essay sessions
with an evidence-based protocol for restorative dental
practice titled Putting it All Together. Friday afternoons exceptional group of table clinics organized by Dr
Richard Kloehn will complete the 2003 Scientific
Session.
COMPANION PROGRAM: The Companion Activities
Program offers participants an opportunity to enjoy
some of Chicagos unique and delicious attractions. On
Thursday, a tour bus will whisk registered guests to a
Chef Demo and Lunch provided by chef Erwin
Dreshsler at his very popular Erwin Restaurant. Chef
Dreshsler has reserved his entire restaurant for the
Academy and will demonstrate the preparation of a
three-course lunch that will then be served to attendees.
Friday morning features a Continental Buffet
Breakfast at the Fairmont with Barbara Rinella. Ms
Rinellas presentation, Dramatizing Current
LiteratureAcademic Entertainment is a fascinating
and witty program of history in which she becomes
many recent First Ladies to tell their stories of power
and perspective.
RECEPTION: Finally, our Gala Reception on Thursday
evening will once again provide a wonderful, once-ayear, platform for socializing with all our friends and colleagues from across the country and around the world.
Please do not miss this fantastic opportunity for education, information exchange and fun. See you in
Chicago in February!
For more meeting information, please contact Dr
Gregory Smith, PO Box 14996, Gainesville, FL 326042996; Fax (352) 371-4882.
Operative Dentistry
Home Page
We hope all our readers will take advantage of the
information available by accessing our Internet home
page. Our address is: http://www.jopdent.org/
The home page contains a search engine and buttons
that, hopefully, will lead you to answers to any questions you may have related to Operative Dentistry.
These are:
Operative Dentistry
Journal: leads to information on the Editorial Staff
and Editorial Board; a complete index of journal volumes; a compilation of direct gold references; highlights
of the current, next, and future issues, as well as a more
detailed look at published Editorials and Clinical Pearls.
Subscribe: leads to complete information on subscription rates; purchasing back issues, reprints, and bound
volumes; and subscription and change of address
forms.
Affiliates: provides links to the American Academy of
Gold Foil Operators, the Academy of Operative
Dentistry, the AADS-Operative Section, and our
Corporate Sponsors. In addition, membership applications for the journals parent academies are available
for downloading.
News: announcements of interest to our readers,
including meeting information, advertised faculty positions, and upcoming CE courses.
Authors: complete instructions for contributors to the
journal.
Reviewers: password-protected link for our Editorial
Board to submit manuscript reviews electronically.
Corporate
Sponsorship
Operative Dentistry invites dental manufacturers
that share our commitment to timely publication of
research and clinical articles relevant to the discipline
of restorative dentistry to become Corporate Sponsors
of the journal. Operative Dentistry is distributed in
54 countries with more than 1,800 subscribers. The
cost of sponsorship is US $3,000.00 per year (January
through December). Sponsors will be recognized by
having their company logo displayed on a special page
in each issue of the journal for the duration of their
sponsorship.
A complimentary subscription to Operative
Dentistry will be sent to the corporate contact person,
and the sponsors logo will appear in the Corporate
Sponsors section of our web page, where it will act as a
link to their companys website. Interested parties
should contact the Editor at Operative Dentistry,
1121 West Michigan Street, Indianapolis, IN 462025186; phone: (317) 278-4800; fax: (317) 278-4900; e-mail:
[email protected]; URL: http://www.jopdent.org/
103
Happy New Year

to our loyal
subscribers
Thank you
for making this
a banner year!!
104
Operative Dentistry
Instructions to Contributors
Correspondence
Send manuscripts and correspondence regarding manuscripts to Dr Michael A Cochran, Editor, Operative
Dentistry, Indiana University School of Dentistry, Room
S411, 1121 W Michigan St, Indpls, IN 46202-5186; phone
(317) 278-4800; fax (317) 278-4900; e-mail: [email protected]; URL: http://www.jopdent.org/.
Exclusive Publication
All material submitted for publication must be submitted
exclusively to Operative Dentistry. Manuscripts not following the form outlined below may be returned for correction
and resubmission.
Manuscripts
Submit an original typed manuscript and three copies.
The manuscript should include a short title for running
headlines. Any identifying information (authors names,
etc) should be on a separate page and not a part of the
manuscript. Authors with English as a second language
should consider having their manuscript reviewed for
grammar, syntax and punctuation prior to submission.
Submit a computer disk and identify the operating system
(Macintosh or IBM-compatible) and the word processing
program used.
Identify the corresponding author and provide a complete
address, fax number and e-mail address.
Supply complete names, degrees, titles and affiliations for
all authors (include addresses that are different from the
corresponding authors).
Proprietary names of equipment, instruments and materials should be followed in parenthesis by the name and
address, of the source or manufacturer.
Research (clinical and laboratory) papers MUST include a
one sentence Clinical Relevance statement, as well as a
Summary, Introduction, Methods and Materials, Results,
Discussion and Conclusions section. Funding other than
material supply must be stated.
Clinical Technique/Case Report papers should contain at
least the following: Purpose, Description of Technique or
Solution, along with materials and potential problems and a
Summary outlining advantages and disadvantages.
Type double-spaced, including references, and leave margins of at least 3 cm (1 inch). Spelling should conform to the
American Heritage Dictionary of the English Language. SI
(Systme International) units are preferred for scientific
measurement, but traditional units are acceptable.
The editor reserves the right to make literary corrections.
Illustrations
Please do NOT submit any illustrations or graphs in
Microsoft Power Point or Word format. They will not
be accepted.
Submit four copies of each illustration.
Line drawings should be in India ink or its equivalent on
heavy white paper, card or tracing velum. All lettering
must be of professional quality, legible against its background and remain proportionally legible if reduced. Type
legends on separate sheets.
Photographs should be on glossy paper with a maximum
size of 15x20 cm (6x8 inches). For best reproduction, a print
should be one-third larger than its reproduced size.
On the back of each illustration indicate lightly in pencil the
top and the number of the figure ONLY (no names). Where
relevant, state staining technique(s) and the magnification
of the prints. Obtain written consent from holders of copyright to republish any illustrations published elsewhere.
Illustrations may also be supplied on floppy disk, Zip disk
or CD as TIFF files with a minimum resolution of 300 dpi
(dots per inch) for grayscale and 1200 dpi for color.
Photographs become the property of Operative Dentistry.
Tables and Graphs

Submit tables and graphs on sheets separate from the text.
Graphs are to be submitted with any lettering proportional
to their size, with their horizontal and vertical axes values
displayed.
Data for constructing graphs MUST be provided with the
manuscript in a spreadsheet (Excel) or word processing format on computer disk.
Graphs may be supplied on floppy disk, Zip disk or CD as
TIFF files with a minimum resolution of 300 dpi. or as
Microsoft Excel files.
References
References must be arranged in alphabetical order by
authors names at the end of the article, with the year of publication placed in parentheses immediately after the
authors name. This is followed by the full journal title (no
abbreviations and in italics), the full subject title, volume
and issue numbers and first and last pages.
In the text, cite references by giving the author and, in
parentheses, the date: Smith (1975) found; or, by placing
both name and date in parentheses: It was found(Smith &
Brown, 1975; Jones, 1974).
When an article being cited has three authors, include the
names of all of the authors the first time the article is cited;
subsequently, use the form (Brown & others, 1975). Four or
more authors should always be cited in the text as (Jones &
others, 1975). In the References section, always list all the
authors.
If reference is made to more than one article by the same
author and is published in the same year, the articles should
be identified by a letter (a, b) following the date, both in the
text and in the list of references.
Book titles should be followed by the publication address
and the name of the publisher.
Reprints
Reprints of any article, report or letter can be ordered
through the editorial office.
SEPTEMBER/OCTOBER 2003
VOLUME 28
NUMBER 5
Editorial Board
Aim and Scope

Operative Dentistry publishes articles that advance the practice of
operative dentistry. The scope of the journal includes conservation
and restoration of teeth; the scientific foundation of operative dental
therapy; dental materials; dental education; and the social, political,
and economic aspects of dental practice. Review papers, book reviews,
letters and classified ads for faculty positions are also published.
Subscriptions: Fax (317) 852-3162

Operative Dentistry (ISSN 0361-7734) is published bimonthly by Operative Dentistry, Indiana University School of
Dentistry, Room S411, 1121 West Michigan Street,
Indianapolis, IN 46202-5186. Periodicals postage paid at
Indianapolis, IN, and additional mailing offices. Postmaster:
Send address changes to: Operative Dentistry, Indiana University School of Dentistry, Room S411, 1121 West Michigan
Street, Indianapolis, IN 46202-5186.
Current pricing for individual, institutional and dental student subscriptions (both USA and all other countries) can be found at our website: www.jopdent.org, or by contacting our subscription manager:
Fax 317/852-3162
E-mail: [email protected]
Information on single copies, back issues and reprints is also available. Make remittances payable (in US dollars only) to Operative
Dentistry and send to the above address. Credit card payment (Visa,
MasterCard) is also accepted by providing card type, card number,
expiration date, and name as it appears on the card.
Contributions
Contributors should study the instructions for their guidance printed
in this journal and should follow them carefully.
Permission
For permission to reproduce material from Operative Dentistry please
apply to Operative Dentistry at the above address.
The views expressed in Operative Dentistry do not necessarily represent those of the Academies or of the Editors.
Editorial Office
Operative Dentistry
Indiana University School of Dentistry, Room S411
1121 West Michigan Street, Indianapolis, IN 46202-5186
Telephone: (317) 278-4800, Fax: (317) 278-4900
URL: http://www.jopdent.org/
Editorial Staff
Editor: Michael A Cochran
Editorial Assistant/Subscription Manager: Joan Matis
Editorial Associate: Karen E Wilczewski
Associate Editors: Bruce A Matis, Edward J DeSchepper
and Richard B McCoy
Managing Editor: Timothy J Carlson
Assistant Managing Editors: Joel M Wagoner
and Ronald K Harris
Kinley K Adams
Maxwell H Anderson
Daniel J Armstrong
Steven R Armstrong
Tar-Chee Aw
Wayne W Barkmeier
Douglas M Barnes
Gardner Bassett
Mark W Beatty
Lars Bjrndal
Lawrence W Blank
Paul K Blaser
Murray R Bouschlicher
William W Brackett
James C Broome
K Birgitta Brown
William Browning
Paul A Brunton
Michael Burrow
Fred J Certosimo
Daniel CN Chan
David G Charlton
Gordon J Christensen
Kwok-hung Chung
N Blaine Cook
David Covey
Gerald E Denehy
Joseph B Dennison
Kim E Diefenderfer
E Steven Duke
Willaim J Dunn
Frederick C Eichmiller
Sigfus T Eliasson
Omar M El-Mowafy
John W Farah
Dennis J Fasbinder
Jack L Ferracane
Mark Fitzgerald
Kevin B Frazier
James C Gold
Carlos Gonzalez-Cabezas
Valeria V Gordan
Kevin M Gureckis
Mark S Hagge
Carl W Haveman
Van B Haywood
Charles B Hermesch
Harald O Heymann
Thomas J Hilton
Richard J Hoard
Barry W Holleron
Ronald C House
Poonam Jain
Gordon K Jones
Barry Katz
Robert C Keene
William P Kelsey, III
Edwina A M Kidd
George T Knight
Kelly R Kofford
Harold R Laswell
2003 Operative Dentistry, Inc. Printed in USA
477-664
Mark A Latta
James S Lindemuth
Melvin R Lund
Dorothy McComb
Jonathan C Meiers
Georg Meyer
Ivar A Mjr
Michael P Molvar
B Keith Moore
Graham J Mount
David F Murchison
Ann-Marie Neme
Jennifer Neo
Jacques Nr
John W Osborne
Michael W Parker
Craig J Passon
Tilly Peters
Anne Peutzfeldt
Frank E Pink
T R Pitt Ford
Jeffrey A Platt
L Virginia Powell
James C Ragain
John W Reinhardt
Eduardo Reston
Philip J Rinaudo
Andr Ritter
J William Robbins
Frank T Robertello
Howard W Roberts
Boyd E Robinson
Clyde L Roggenkamp
William Rose
Jean-Francois Roulet
Frederick A Rueggeberg
Henry A St Germain, Jr
David C Sarrett
John W Shaner
Gregory E Smith
W Dan Sneed
Ivan Stangel
James M Strother
James B Summitt
Edward J Swift, Jr
William H Tate
Franklin R Tay
Choi Gait Toh
Peter T Triolo, Jr
Karen Troendle
Richard D Tucker
Marcos Vargas
Douglas Verhoef
Warren C Wagner
Joel M Wagoner
Charles W Wakefield
Steve W Wallace
Timothy F Watson
Nairn H F Wilson
Peter Yaman
Adrian U J Yap
Andrea G F Zandona
These Dental Manufacturers have joined Operative Dentistry in our commitment to

publish quality dental literature in a timely manner. We thank them for their support.
Please view the Corporate Sponsors Page at our website (http://www.jopdent.org)

for direct links to these companies.
VOLUME 28
NUMBER 1
1-104
OPERATIVE DENTISTRY, Inc.
EDITORIAL
Where Have All the Mentors Gone?
MA Cochran . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1
CLINICAL RESEARCH
Technique Sensitivity in Bonding to Vital, Acid-Etched Dentin
M Ferrari FR Tay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3
OPERATIVE DENTISTRY
LABORATORY RESEARCH
Voids and Porosities in Class I Micropreparations Filled with Various Resin Composites
NJM Opdam JJM Roeters T de Boer D Pesschier E Bronkhorst . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9
Effects of In-Office Bleaching Products on Surface Finish of Tooth-Colored Restorations
P Wattanapayungkul AUJ Yap . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15
Effects of Regional Enamel and Prism Orientation on Resin Bonding
Y Shimada J Tagami . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .20
The Effect of a Resin Coating on the Interfacial Adaptation of Composite Inlays
PR Jayasooriya PNR Pereira T Nikaido MF Burrow J Tagami . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .28
Effects of Finishing/Polishing Techniques on Microleakage of Resin-Modified Glass Ionomer Cement Restorations
AUJ Yap WY Yap EJC Yeo JWS Tan DSB Ong . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .36
Effects of Instrumentation Time on Microleakage of Resin-Modified Glass Ionomer Cements

AUJ Yap EJC Yeo WY Yap DSB Ong JWS Tan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .47
Surface Geometry of Three Packable and One Hybrid Composite After Finishing
M Jung S Voit J Klimek . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .53
Relationship Between Nanoleakage and Microtensile Bond Strength at the Resin-Dentin Interface
S Guzmn-Armstrong SR Armstrong F Qian . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .60
Effects of Er:YAG and Nd:YAP Laser Irradiation on the Surface Roughness and Free Surface Energy of Enamel
and Dentin: An In Vitro StudyV Armengol O Laboux P Weiss A Jean H Hamel . . . . . . . . . . . . . . . . . . . . . .67
Effects of Cavity Size on Apoptosis-Induction During Pulp Wound Healing
C Kitamura Y Ogawa T Morotomi M Terashita . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .75
The Effect of Placement of Glass Fibers and Aramid Fibers on the Fracture Resistance of Provisional
Restorative MaterialsG Saygili SM Sahmali F Demirel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .80
LITERATURE REVIEW
Tooth-Colored Post Systems: A Review
AJE Qualtrough F Mannocci . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .86
volume 28 number 1 pages 1-104
The Effect of Flowable Resin Composite on Microleakage in Class V Cavities

AR Yazici M Baseren B Dayanga . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .42
INVITED PAPER
Minimal Intervention Dentistry: Rationale of Cavity Design
GJ Mount . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .92
AWARDS
AAGFO Clinician of the Year Award . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
DEPARTMENTS
INSTRUCTIONS TO CONTRIBUTORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104

10-9385
Operative Dentistry
Indiana University School of Dentistry, Rm S411
1121 West Michigan Street
Indianapolis, IN 46202-5186 USA
Periodicals
january-february 2003
Classifieds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
Announcements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
Operative Dentistry Home Page . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
Corporate Sponsorship . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
january/february 2003 volume 28 number 1 1-104

(ISSN 0361-7734)

1559 2863 28 1 1

Uploaded by

Copyright:

Available Formats

1559 2863 28 1 1

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

1559 2863 28 1 1

Uploaded by

Copyright:

Available Formats

VOLUME 28

OPERATIVE DENTISTRY, Inc.

Effects of Instrumentation Time on Microleakage of Resin-Modified Glass Ionomer Cements

volume 28 number 1 pages 1-104

The Effect of Flowable Resin Composite on Microleakage in Class V Cavities

INSTRUCTIONS TO CONTRIBUTORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104

january/february 2003 volume 28 number 1 1-104

Aim and Scope

Subscriptions: Fax (317) 852-3162

2003 Operative Dentistry, Inc. Printed in USA

These Dental Manufacturers have joined Operative Dentistry in our commitment to

Please view the Corporate Sponsors Page at our website (http://www.jopdent.org)

Operative Dentistry, 2003, 28, 1

More clinically oriented continuing education is

Operative Dentistry, 2003, 28, 3-8

Technique sensitivity, previously reported in vitro using a moist bonding technique on

vivo. No hybrid layer was observed in vivo after

dried, acid-etched dentin (Perdigo & Frankenberger,

Ferrari & Tay: Bonding to Vital Dentin

four groups were stored in distilled water

EM208S, Philips, Eindhoven,

Except for the presence of a thin, highly

Ferrari & Tay: Bonding to Vital Dentin

Similar to nanoleakage, the term water tree was

The adhesives and composites used in this study were generously

Pereira PN, Okuda M, Sano H, Yoshikawa T, Burrow MF &

(Received 3 April 2002)

Perdigo J & Frankenberger R (2001) Effect of solvent and

Rosales JI, Marshall GW, Marshall SJ, Watanabe LG, Toledano

Sano H, Yoshiyama M, Ebisu S, Burrow MF, Takatsu T, Ciucchi

Hashimoto M, Ohno H, Kaga M, Sano H, Endo K & Oguchi H

Spencer P, Wang Y, Walker MP, Wieliczka DM & Swafford JR

Inoue S, Van Meerbeek B, Vargas M, Yoshida Y, Lambrechts P &

Stepp D, King JA, Worrall J, Thompson A & Cooper DE (1996)

Itthagarun A & Tay FR (2000) Self-contamination of deep dentin

Nakabayashi N & Hiranuma K (2000) Effect of etchant variation

Van Meerbeek B, Vargas M, Inoue S, Yoshida Y, Perdigo J,

Nakajima M, Kanemura N, Pereira PN, Tagami J & Pashley DH

Van Meerbeek B, Yoshida Y, Lambrechts P, Vanherle G, Duke

Pashley DH, Agee KA, Nakajima M, Tay FR, Carvalho RM,

Wang Y & Spencer P (2002) Quantifying adhesive penetration in

Operative Dentistry, 2003, 28, 9-14

Voids and Porosities

preparations were restored with the adhesive

Adhesive techniques enable the clinician to save sound

Figure 1. Occlusal view from the experimental preparation.

This study compared the adaptation and homogeneity

Figure 2. Section of the experimental preparation design showing the

Opdam & Others: Adaptation of Composite in Micropreparations

which was packed with a hand-instrument and cured

Table 1: Presence of Voids in the Total Restoration

Voids Present (n)

cured flowable + syringable

uncured flowable +packable

cured flowable + packable

Table 2: Presence of Voids in the Preparation Part

Voids Present (n)

cured flowable + syringable

uncured flowable +packable