Clinical

Direct or indirect restorations?

Jason Smithson,1 Philip Newsome,2 David Reaney,3 Siobhan Owen4

The majority of clinical decisions regarding the most amalgam restorations continue to be placed, as far as this
appropriate choice of restorative material and technique paper is concerned direct restorations refer to resin-based
are relatively straightforward and are usually dictated by a composite unless otherwise stated. In the US, it appears
variety of factors including: that around two-thirds of direct restorations currently being
• Lesion size and aetiology placed are made of composite and one-third amalgam
• Aesthetic, occlusal, endodontic and periodontal (Christensen GJ, 2010). As Christensen points out,
considerations amalgam isn’t dead – it is simply being used less often.
• Number of teeth affected Glass ionomer may be considered in a limited number of
• Patient compliance, habits and preferences situations where its cariostatic effects might be considered
• The dentist’s own competence and underlying beliefs useful (Tay FR et al, 2001).
about restorative treatment.
The decision-making process involved when choosing to Category C: indirect restorations clearly
use either a direct or an indirect approach for any given indicated
clinical situation can be facilitated by considering the At the other end of the spectrum, Category C is also
following continuum, which has at one end direct relatively straightforward in that here we find large cavities
restorations and at the other indirect restorations and and/or failed direct restorations with multiple missing
which can be split into three distinct categories. cusps; anterior teeth with large interproximal cavities along
with maybe one or both mesial and distal incisal edges
Category A: direct restorations clearly indicated requiring replacement; replacement of failed crowns; and
Category A is populated by the overwhelming majority of
Class I, III, IV and V restorations, as well as most small Class
II (MO, DO and MOD) restorations. Unless there are other
contributory factors at play, the restoration of choice will be
direct and, most probably, composite. While many

1
Jason Smithson BDS, DipRestDentRCS (Eng). Cornwall, general
practice with a special interest in Aesthetic and Restorative
Dentistry.
2
Philip Newsome PhD, MBA, BChD, FDS RCS (Ed) MRD RCS (Ed)
Associate Professor at the Faculty of Dentistry, University of
Hong Kong and is on the Specialist Prosthodontist Registers of
both Hong Kong and the UK.
3
Siobhan Owen BDS. Managing Director of Southern Cross Figure 1: It is clear that in this case replacing these failing amalgams
Dental Laboratories UK Ltd. and composites with further direct restorations would be
4
David Reaney BDS. DGDP (UK) M Clin Dent (Pros) inappropriate and that indirect restorations are required.

70 INTERNATIONAL DENTISTRY – AFRICAN EDITION VOL. 1, NO. 1

 Clinical

2a 2b
Figure 2a and b: The combination of endodontic treatment and extensive tooth loss, especially if this involves loss of one or both
marginal ridges makes a tooth highly susceptible to fracture.

3a 3b
Figure 3a and b: Most posterior root-filled teeth require some form of indirect occlusal coverage to provide protection and
prevent subsequent tooth fracture.

large rehabilitation cases requiring the recreation of Root-filled teeth

multiple occlusal surfaces. There is little contention that the Compared to teeth with healthy pulps, root-filled teeth are
treatment of choice in such situations is some form of considered to be more susceptible to fracture (Figure 2) as
indirect restoration (Figure 1). The choice of material (gold, they possess reduced dentinal elasticity (Johnson JK et al,
porcelain fused to metal (PFM) or all-ceramic) depends on 1976), lower water content (Rosen H, 1961), deeper
an analysis of the various factors listed in the introduction. cavities (Madison S, Wilcox LR, 1988) and substantial loss
of dentine including, critically, the strengthening effect of
Category B: uncertainty over which is the most the pulp chamber roof (Assif D et al, 2003).
appropriate approach It has therefore been a long and widely held view (Goerig
It is in the middle ground of Category B where we find a AC, Mueninghoff LA, 1983; Reeh ES et al, 1989) that
number of commonly occurring clinical scenarios that can posterior root-filled teeth require some form of indirect
and do cause confusion among many dentists who are occlusal coverage (onlay or crown) in order to protect the
unsure which avenue they should pursue – direct or tooth against subsequent root fracture (Figure 3).
indirect. These are the clinical situations we wish to explore It has also been suggested (Assif D et al, 2003; Smales RJ,
further in this paper. Hawthorne WS, 1997) that where more extensive tooth

INTERNATIONAL DENTISTRY – AFRICAN EDITION VOL. 1, NO. 1 71

 Smithson et al

4a 4b

Figure 4a and b: In cases where, for whatever reason, indirect occlusal coverage restorations are not able to be placed, amalgam can be used,
provided that there is sufficient occlusal reduction.

5a 5b
Figure 5a and b: Large, durable, aesthetic direct placement restorations are now possible thanks to modern composite/bonding technology. It
is imperative though that the operator has an excellent understanding of the sensitive nature of the clinical techniques required.

loss has taken place it is possible to use a directly-placed would steer one towards using indirect occlusal coverage.
amalgam onlay to protect the teeth (at least as a medium Additionally, premolar teeth are thought to be more likely
term, cost effective precursor to a later indirect restoration) to fracture when a direct intracoronal restoration is placed,
provided that sufficient bulk of material – at least 2mm simply because the tooth tissue loss caused by the access
(Nayyar A et al, 1980) – is present (Figure 4). preparation is proportionally larger than in a molar tooth.
This type of extensive restoration is more difficult and Lastly, there is a theoretical possibility that the effect of
clinically demanding to create in direct composite (see contraction stress generated through polymerisation
below), and there are concerns whether such a restoration shrinkage (Feilzer AJ et al, 1987) may be magnified on
would exhibit sufficient strength to resist occlusal forces. If teeth that are already more susceptible to fracture.
there is only an occlusal access cavity requiring restoration In anterior teeth where the occlusal loading is much less,
then, increasingly, a simple direct restoration is deemed to it is common practice now to use direct composite in cases
be sufficient (Hernandez R et al, 1994). Apart from with simple cingulum access cavities. The greater the extent
conserving tooth tissue this would have the added benefit of any previous caries, restoration or trauma, the more
of reducing the length of restoration margin and therefore likely some form of extra-coronal restoration will be
the potential for microleakage, a highly significant factor in required. For example, a tooth with mesial and distal
long-term endodontic success (Tickle M et al, 2008). cavities, connected by an occlusal access cavity, will have
There are caveats to this approach, however. For lost much of its integral strength and would be much more
example, the presence of suspicious crack lines and heavy prone to fracture if restored simply by means of a direct
occlusal loading combined with parafunctional habits composite. Once again, caution must be exercised

72 INTERNATIONAL DENTISTRY – AFRICAN EDITION VOL. 1, NO. 1

 Smithson et al

dental anatomy to re-create correctly a replacement cusp

directly in the mouth. These skills are far removed from
those required to carve a large amalgam. While it may be
possible to recreate more than one cusp directly inside the
mouth, concerns over the strength of the final restoration,
as well as the considerable chairside time required to
complete such a restoration, mean that many authorities
on the use of direct composite would often draw the line
at replacing just one cusp. Instead, they would recommend
using either an indirect restoration or a so-called ‘semi-
direct’ or ‘direct-indirect’ approach (Spreafico R, 1996) for
these more extensive cavities.

Figure 6: Simple at-home tooth whitening can produce dramatic Deep proximal boxes
results and, as in this case, is a useful precursor to restorative
treatment. The upper arch has been whitened to give the patient an Direct composites are more likely to be aesthetic, functional
initial indication of the level of improvement achieved. and durable when cavity margins are situated within
enamel, free from heavy occlusal contact and easily
whenever there is evidence of crack lines and heavy accessible in terms of visibility, ease of isolation and
occlusal loading. relationship to adjacent gingival tissues. The more a cavity
fails to fulfil these criteria, the more difficult it becomes to
Cusp replacement predict success. Therefore, consideration should be given to
In cases where a single cusp of a posterior tooth has been an alternative means of restoring the tooth. One very
lost for whatever reason (crack line, caries, trauma, etc), it common example of this is when the box of a class II cavity
is now considered acceptable to restore the tooth using extends beyond enamel. Apart from the likelihood of a
direct composite (Figure 5). Key to this is the dentist’s own significant loss of tooth substance, the main problem
clinical ability, in particular the skill and knowledge of arising in such a situation is the difficulty inherent in trying

7a 7b
Figure 7a and b: Non-vital tooth whitening is an extremely efficient, conservative way of dealing with minimally restored darkened non-vital teeth.

8a 8b

Figure 8a and b: It was initially thought that this discolouration would be impossible to treat and further options were discussed with the
patient. The whitening occurred within three months and was very rapid using 10% carbamide peroxide. The patient was very happy with
this result. After a period of two weeks further composite bonding was undertaken to improve the shape of the upper left central incisor.
(Photographs courtesy of Dr Linda Greenwall).

74 INTERNATIONAL DENTISTRY – AFRICAN EDITION VOL. 1, NO. 1

 Smithson et al

Figure 9a and b: Indirect composite onlays are more durable than many believe, are extremely conservative of tooth structure and are unlikely
to damage the opposing dentition. In any case, does it really matter that they may occasionally require replacement if the underlying tooth
is preserved and tooth vitality maintained with minimal effect on the periodontal tissues?

to seal subgingival cervical margins located within dentine Another approach is to use indirect porcelain or
and/or cementum. composite inlay/onlays, thereby minimising potential resin
shrinkage to that exhibited by the thin layer of luting resin.
Dentine bonding These restorations require a cavity with a location above
While bonding to enamel is now highly predictable, the gingival margin (Dietschi D et al, 1994) and either a
doing so either to dentine or cementum is far more bevel-free butt margin or a hollow chamfer (Hannig M et
problematic, primarily because of the difficulty forming al, 1991). Where the margin is slightly subgingival, Dietschi
an effective hybrid layer. This is dependent on the (1998) advocates ‘relocating’ the cervical preparation
successful execution of a series of crucial clinical steps – above the gingival margin, also known as ‘margin
etching, washing and drying, primer and bond elevation’, by applying an increment of flowable composite
application, and, finally, the polymerisation of adhesive resin at the margin. This represents a non-invasive
resin in order to stabilise the fragile structure of the alternative to surgical crown lengthening.
hybrid layer itself. In light of this great complexity, there
are well-documented concerns about the lasting stability Tooth discolouration
of dentine bonding (Dietschi D et al, 1995; Van Tooth discolouration has a number of different possible
Meerbeek B, Perdigao J, 1998; Hashimoto M et al, 2000). aetiologies and may be either intrinsic or extrinsic in nature.
The majority of cavities are entirely bounded by enamel Most extrinsic discolouration can be removed easily and it
and it is thought that the seal achieved at the margin (i.e. is the instrinsic variety that is more difficult to treat,
between resin and enamel) protects any ‘internal’ resin- depending on the severity and depth of discolouration.
dentine bond at the floor of the cavity. However, what Dentists are often unsure which is the best strategy to
happens when part of the cavity margin is bounded by adopt and much depends on the aetiology of the
dentine? How should one proceed? discolouration. A correct diagnosis is important and allows
If the decision has been taken to place a direct composite the dental practitioner to explain to the patient the exact
restoration, Liebenberg (2005) advocates a resin-modified nature of the condition. In some instances, the mechanism
glass ionomer cement sandwich technique. This is, of of staining may have an effect on the outcome of
course, not a new technique (Suzuki M, Jordan RE, 1990) treatment and, in turn, influence the treatment options the
and traditionally the filler of the ‘sandwich’ was a glass- dentist will be able to offer to patients (Watts A, Addy M,
ionomer cement. However, resin-modified glass ionomer 2001). A large number of discoloured teeth are nowadays
cements (RMGIC) have been shown to possess superior treated either conservatively by tooth whitening (Figure 6)
mechanical properties and bonding strength to dentine or less conservatively by means of ceramic laminate veneers
(Pereira LC et al, 2002). Andersson-Wenckert (2004) or even full coverage crowns.
examined the durability and cariostatic effect of a modified Dentists will often restrict the use of bleaching to mild
open sandwich restoration using RMGIC and concluded forms of discolouration – in other words, the darkening of
that it exhibited acceptable durability for the extensive teeth through normal ‘wear and tear’ of daily life – while
restorations evaluated. ceramic solutions are reserved for more severe situations

76 INTERNATIONAL DENTISTRY – AFRICAN EDITION VOL. 1, NO. 1

 Smithson et al

such as tetracycline discolouration. Meanwhile, non-vital unimportant if the restoration fails, provided the tooth still
bleaching is very effective in most cases (Figure 7) and is has a good prognosis. Indeed, an increasing number of
an excellent adjunct to bonded composite when the only studies have illustrated the feasibility of this approach to
tooth tissue loss is a result of the access preparation the treatment of TSL cases.
(Deliperi S, Bardwell DN, 2005). Clinical trials conducted by Hemmings et al (2000) and
Tetracycline discoloured teeth present their own specific Redman et al (2003) examined the use of direct composite
problems. Veneers are often prescribed for such teeth. The restorations for the treatment of localised anterior tooth
difficulty, however, with using veneers to treat tetracycline wear and revealed relatively low failure rates of the
discolouration is that the staining usually extends deep composites, with a median survival rate of five years. Those
within the structure of the tooth. As the enamel is failures that did arise were mainly the result of generalised
removed the discolouration tends to get worse, and so marginal failure and discolouration. Similarly, Poyser et al
more dentine is removed to allow for a greater thickness (2007) evaluated clinical performance and related patient
of ceramic. As the preparation inevitably moves further satisfaction of direct composite restorations used to restore
into dentine, the retention of the veneer becomes worn mandibular anterior dentitions. The authors
increasingly dependent on dentine bonding rather than concluded that direct composite restorations placed at an
enamel bonding with all the attendant difficulties increased occlusal vertical dimension are a simple and time-
described earlier in this paper. A further problem is that efficient method of managing the worn mandibular
removal of enamel in this way for veneer restorations is an anterior dentition. They also found that patient acceptance
attempt to match up high elastic modulus porcelain with and adaptation to the technique is good and is maintained
lower elastic modulus dentine. It is predictable that for the medium-term.
functional loading of the veneered tooth will transfer this
energy to the interface, resulting in debonding or cracking Conclusion
in the porcelain. The upshot of all this is that the greater In virtually every clinical case there will be more than one
the depth of the veneer, the greater the need for way to achieve a result. Many decisions regarding
additional (i.e. conventional) means of retention and this treatment are straightforward, as the advantages of one
means the use of full coverage restorations. particular procedure outweigh its own disadvantages and
Are there other, more conservative means of dealing the relative advantages of other available options. There
with the type of deep discolouration caused by are, however, a variety of situations where the choice is less
tetracycline? Certainly, long-term tooth whitening can yield clear-cut and in this paper we have tried to highlight some
acceptable results (Figure 8), with success depending on of these in relation to the selection of an indirect or a direct
the depth, severity and degree of the discolouration restorative approach.
(Greenwall L, 2001), as well as on the patient’s particular There will never be a completely black and white guide
expectations of treatment. With the development of more to dental treatment, and grey areas will always exist. As
efficient opaquing systems, the use of direct composites is long as treatment is performed with care, to a high
likely to increase in the future, probably in conjunction with standard and with a nod to the underlying science, it will
prior tooth whitening. more than likely be successful.

Tooth surface loss References

Tooth surface loss (TSL) is defined as loss of tooth substance Andersson-Wenckert IE, van Dijken JW, Kieri C (2004)
in the absence of caries or trauma. Traditionally, indirect Durability of extensive Class II open-sandwich restorations
restorations have been used as the treatment of choice in with a resin-modified glass ionomer cement after 6 years. Am
such cases as they confer greater strength than direct J Dent 17: 43–50.
restorations. Improvements in composite mechanical Assif D, NissanJ, Gafni et al (2003) Assessment of the
properties and bonding chemistry over the years have, resistance to fracture of endodontically treated molars
however, made it possible to restore surfaces of worn teeth restored with amalgam. J Prosthet Dent 71: 565-7.
directly without recourse to indirect restorations, although Christensen GJ (2010) Should resin-based composite
this option remains if the need arises in the future. dominate restorative dentistry today? J Am Dent Assoc 141:
Dentists often focus on the survival of their restoration 1490-3.
when, in fact, the focus should be on survival of the tooth Deliperi S, Bardwell DN (2005) Two-Year Clinical Evaluation
on which it has been placed (Figure 9). It is relatively of Nonvital Tooth Whitening and Resin Composite

78 INTERNATIONAL DENTISTRY – AFRICAN EDITION VOL. 1, NO. 1

 Restorations. J Esthet Restor Dent 17: 369-79. coronal-radicular dowel and core technique for endodontically
Dietschi D, Magne P, Holtz J (1994) Recent trends in treated posterior teeth J Prosthet Dent 43: 511-51.
aesthetic restorations for posterior teeth. Quintessence Int 25: Newsome PRH, Greenwall LH (2008) Management of
659-77. tetracycline discoloured teeth. Aesthetic Dentistry Today 2: 15-20.
Dietschi D, Scampa U (1995) Marginal adaptation and seal Kemp-Scholte CM, Davidson CL (1990) Complete marginal
of direct and indirect Cl II composite restorations: an in vitro seal of Class V resin composite restorations effected by
evaluation. Quintessence Int 26: 127-38. increased flexibility. J Dent Res 69: 1240-3.
Dietschi D, Spreafico R (1998) Current clinical concepts for Pereira LC, Nunes MC, Dibb RG et al (2002) Mechanical
adhesive cementation of tooth-coloured posterior properties and bond strength of glass-ionomer cements. J
restorations. Pract Periodontics Aesthet Dent 10: 4754. Adhes Dent 4: 73–80.
Feilzer AJ, de Gee AJ, Davidson CL (1987) Setting stress in Poyser NJ, Briggs PFA, Chana HS et al (2007) The evaluation
composite resin in relation to configuration of the restoration. of direct composite restorations for the worn mandibular
J Dent Res 66: 1636-9. anterior dentition – clinical performance and patient
Goerig AC, Mueninghoff LA (1983) Management of the satisfaction. J Oral Rehab 34: 361–376.
endodontically treated tooth. Part II: Technique J Prosthet Redman CD, Hemmings KW, Good JA (2003) The survival
Dent 55: 184-5. and clinical performance of resin-based composite
Hannig M, Rahlf B, Schlichting B (1991) Marginal behavior restorations used to treat localized anterior tooth wear. Br
of “Kulzer” composite inlays under simultaneous mechanical Dent J 194: 566-72.
and thermal loading. Dtsch Zahnarztl Z. 46: 618-20. Reeh ES, Douglass WH, Messer HH (1989) Stiffmness of
Hashimoto M, Ohno H, Kaga M et al (2000) In vivo endodontically-treated teeth related to restoration technique.
degradation of resin-dentin bonds in humans over 1 to 3 J Dent Res 68: 1540-4.
years. J Dent Res 79: 1385-91. Rosen H. (1961) Operative procedures on mutilated
Hemmings KW, Darbar UR, Vaughan S (2000) Toothwear endodontically treated teeth. J Prosthet Dent 11: 972-86.
treated with direct composite restorations at an increased Smales RJ, Hawthorne WS (1997) Long-term survival of
vertical dimension: results at 30 months. J Prosthet Dent 83: extensive amalgams and posterior crowns. J Dent 25: 225-7.
287-93. Spreafico R. (1996) Direct and semi-direct posterior
Hernandez R, Bader S, Boston D et al (1994) Resistance composite restorations. Pract Periodontics Aesthet Dent 8:
to fracture of endodontically treated premolars restored 703-12.
with new generation dentine bonding systems. Int Endod Suzuki M, Jordan RE. Glass ionomer-composite sandwich
J 27: 281-4. technique. (1990) J Am Dent Assoc 120: 55–7.
Johnson JK, Schwartz NL, Blacwell RT (1976) evaluation and Tay FR, Pashley EL, Huang C et al (2001) The Glass-ionomer
restoration of endodontically treated posterior teeth. J Am phase in resin-based restorative material. J Dent Res 80: 1808-12.
Dent Assoc 93: 597-605. Tickle M, Milsom K, Qualtrough et al (2008) The failure rate
Liebenberg W (2005) Return to the resin-modified glass- of NHS funded molar endodontic treatment delivered in
ionomer cement sandwich technique. J Calif Dent Assoc general dental practice. Br Dent J 204: E8; 1-6
71: 743-7. Watts A, Addy M (2001) Tooth discolouration and staining:
Loguercio AD, Alessandra R, Mazzocco KC et al (2002) a review of the literature. Br Dent J 190: 309-16.
Microleakage in class II composite resin restorations: total Van Meerbeek B, Perdigao J (1998) The clinical
bonding and open sandwich technique. J Adhes Dent 4: 137–44. performance of adhesives. J Dent 26: 1-20.
Madison S, Wilcox LR (1988) An evaluation of coronal Veneziani M (2010) Adhesive restorations in the posterior
microleakage of endodontically treated teeth. Part III In vivo area with subgingival cervical margins: New classification
study. J Endod 14: 99-104. and differentiated treatment approach. European J Esthet
Nayyar A, Walton RE, Leonard LA. (1980) An amalgam Dent 5: 50-76.

80 INTERNATIONAL DENTISTRY – AFRICAN EDITION VOL. 1, NO. 1