216_Dreiseidler.

qxp 3/18/09 2:25 PM Page 216

Comparison of Cone-Beam Imaging with

Orthopantomography and Computerized
Tomography for Assessment in
Presurgical Implant Dentistry
Timo Dreiseidler, MD, DMD1/Robert A. Mischkowski, MD, DMD2/Jörg Neugebauer, DMD3/
Lutz Ritter, MD1/Joachim E. Zöller, MD, DMD4

Purpose: To establish a basis for weighing the potential diagnostic and therapeutic benefits of three-
dimensional cone-beam (CB) data sets in contrast to digital orthopantomography (OPG) and computer-
ized tomography (CT) in implant dentistry. Materials and Methods: Twenty-seven patients requiring
implant surgery received a single presurgical CB scan. A follow-up digital OPG was taken within a maxi-
mal postsurgical period of 2 weeks. For comparison purposes, a control group of 29 patients receiving
CT as well as CB diagnosis was analyzed. Image quality of the different modalities was ranked retro-
spectively by five experienced examiners (from excellent to insufficient) for up to 10 defined criteria,
including general image quality and several specific structures. The results were analyzed statistically,
and interobserver agreement was calculated using intraclass correlation coefficients (ICCs). Results:
The median rating for all investigated criteria was good for CB imaging and between good and insuffi-
cient for OPG in the dental implant group. Except for general image quality, statistical analysis showed
that CB imaging was significantly superior to OPG imaging for all investigated anatomic structures.
With a few exceptions, all investigated anatomic structures in CT and CB imaging were rated excellent
in the control group. No significant difference between CT and CB imaging was detected in the control
group for all investigated criteria. With a few exceptions, ICCs were higher for CB images than for OPG.
In the control group, ICCs for CT and CB images were similar, with a few exceptions. Conclusion: The
results of the present study confirm superior radiographic visualization for all important high-contrast
structures in presurgical implant dentistry assessment for CB imaging in contrast to OPG and a CT-like
degree of information for high-contrast structures in CB data sets. Clinically, however, the elevated
radiation dosages transmitted by CB imaging must be taken into account. INT J ORAL MAXILLOFAC IMPLANTS
2009;24:216–225

Key words: computerized tomography, cone-beam imaging, dental implants, digital volume tomography,
panoramic tomography, three-dimensional imaging

ccurate presurgical assessment requires precise

A radiographic visualization of anatomic structures
and pathologic conditions. In dental practice, presur-
gical implant imaging strategies often consist of
orthopantomography (OPG) without a second layer.1
1Research Assistant, Department of Craniomaxillofacial and Plas- Whereas OPG imaging is widely used and provides a
tic Surgery, University of Cologne, Germany. convenient and inexpensive method with low radia-
2Assistant Professor and Consultant, Department of Craniomax-
tion exposure, this technique has several limitations,
illofacial and Plastic Surgery, University of Cologne, Germany.
3Consultant, Department of Craniomaxillofacial and Plastic such as nonuniform magnification, 2 diagnostic
Surgery, University of Cologne, Germany. restrictions in the anterior areas,3 and missing cross-
4Professor and Head of Departmant, Department of Craniomax- sectional information4 obscuring buccolingual bone
illofacial and Plastic Surgery, University of Cologne, Germany. widths or angulations and undercuts such as the lin-
gual inclination of the mandibular contours.4,5
Correspondence to: Dr Timo Dreiseidler, Department of Cranio-
maxillofacial and Plastic Surgery, University of Cologne, Kerpener Three-dimensional (3D) visualization reveals more
Straße 62, D-50924 Cologne, Germany. Fax: +49-221-478-4095. anatomic details and potentially includes all radio-
Email: [email protected] graphic information necessary for maximal surgical

216 Volume 24, Number 2, 2009

©2009 Quintessence Publishing Co, Inc.

All Rights Reserved
216_Dreiseidler.qxp 3/18/09 2:25 PM Page 217

Dreiseidler et al

safety and implant positioning improvement. 4,6–8 Table 1 Overview of the Effective Radiation
Computerized tomography (CT) technology nowa- Dosages of Different Radiographic Modalities
days comprises a broad band of different technolo- in Dentistry
gies, from single-row-detector, fan-beam CT to Imaging modality Effective dose
128-slice spiral CT. What all these devices have in
common is that one or multiple single fan-beam Digital OPG 6.2 µSv12; 5 to 14 µSv13
Conventional OPG 10 µSv11; 16 to 21 µSv13
x-rays are emitted to face a corresponding amount of TSI* 3 to 12 µSv14
single row detectors after transmitting the object of Conventional tomography 2 to 9 µSv12
interest. The 3D volume then will be reconstructed CB 29.3 to 331 µSv10,12,13,15
on the basis of the generated slice data. In contrast Galileos CB device 29.3 to 53.6 µSv10
to this, in cone-beam (CB) imaging, cone-like x-rays Low-dose CT 150 to 610 µSv11
CT 314 µSv12; 600 µSv16; 1,270 µSv11
are emitted to face an image intensifier or a flat-
panel image detector array sufficient to obtain 3D *Transversal slice images.
data from up to a single 180-degree rotation of the
imaging unit around a patient’s head.
CT technology has several restrictions, including
device costs, radiation exposure, and operational sets of patients on whom surgical manipulation of
availability. Since the introduction of CB technology,9 anatomic structures had been undertaken in the
a new low-dose 3D imaging modality is available, implant region, such as bone augmentation proce-
costing about a tenth as much as conventional CT dures, or in whom implants compromised the ability
devices and providing an option that clinicians can to evaluate one or more of the investigated anatomic
use in their own practices. Especially for the investi- structures. Eleven (33%) of the included patients
gated CB device, the radiation dose is up to 70-fold were men and 16 (67%) were women; the patients
lower than that emitted for a full-mouth 3D CT and were between 35 and 71 years of age, with an aver-
only 3 times the typical digital OPG radiation dose age of 56.04 ± 10.11 years.
(Table 1).10–16 A control group included 29 patients who did not
The choice of radiographic method in implant receive implants but received CB and CT evaluations.
dentistry should be influenced by the complexity of The ages of the 18 male (63.3%) and 11 female
the implant case, the availability of technology and (36.7%) participants ranged between 18 and 72
expertise, the costs for examination and the device, years, with an average of 40.6 ± 14.5. There had to be
as well as the radiation exposure to the patient. Radi- a diagnostic indication to perform an examination
ation exposure should routinely be limited to a low with either technique so that participants would not
level while demonstrating sufficient net benefit, with be subjected to unnecessary radiation. Images
the total potential diagnostic or therapeutic benefits included in the study must have shown the same
it produces with the amount of radiation emitted.17 findings. Thirty-one scans that had been performed
The present study was performed to establish a basis more than 3 months apart were excluded from the
for weighing the potential diagnostic and therapeu- study. The control group participants were generally
tic benefits of CB imaging in contrast to OPG and CT examined preoperatively to bone grsfting or other
imaging in implant dentistry. surgery with either technique, or during a diagnostic
period when, in addition to the routinely obtained
CT, further information about the dental system was
MATERIALS AND METHODS needed.
This retrospective evaluation was executed under
Necessary permission for this retrospective clinical standardized conditions by five experienced examin-
trial from the responsible ethical committee at the ers consisting of two maxillofacial surgeons, two
University of Cologne was requested and approved general practitioners, and a final-year resident in
beforehand (approval no. 05-111). The image quality maxillofacial surgery. The examiners’ task in this trial
of 27 presurgical CB data sets was compared to that was to grade the general visualization quality of the
of the corresponding digital follow-up OPGs. entire data set as well as those of nine regional sites
Included were the data sets of all patients in need of subjectively with rankings from 1 (excellent) to 5
implant treatment who presented to the University (insufficient/not detectable). Investigated structures
of Cologne during the period of March 2005 to Janu- were: the mental foramen, the mandibular canal, the
ary 2006 and for whom presurgical 3D CB imaging bony nasal floor, the maxillary sinus floor, the incisive
and digital follow-up OPG had been performed foramen, the temporomandibular joint (TMJ), and the
within 2 weeks after surgery. Excluded were 21 data implant region, its bone borders, and its adjacent

The International Journal of Oral & Maxillofacial Implants 217

©2009 Quintessence Publishing Co, Inc.

All Rights Reserved
216_Dreiseidler.qxp 3/18/09 2:25 PM Page 218

Dreiseidler et al

Fig 1 (Top) Alignment of 3D CB data set visual-

ization and (bottom) the corresponding postoper-
ative digital OPG. The 3D CB data set is
visualized as a panoramic view reconstruction in
alignment with (middle left) a tangential view,
(middle center) three cross-sectional planes, and
(middle right) an axial scout view. These display
details are selected via the inspection window
placed in the mandibular right first molar region
of the panoramic view. On the right cross-
sectional plane, the mental foramen can be
clearly seen penetrating the labial compacta of
the mandible. Good visualization of the mental
foramen is possible in the corresponding post-
operative OPG control.

teeth. All investigated anatomic structures had in qualities was achieved by the use of visualization
common the existence of borders with strict saltation example cards showing excellent, mediocre, and
in radiodensity to the surrounding tissues leading to insufficient imaging results for each of these struc-
detectable shifts in the grey values of the corre- tures, enabling the observers to assign the image as a
sponding images. The clarity of these transitions for whole or the specific structure to an image quality
the investigated structures, also referred to as struc- scale.
ture definition, was the main criterion for grading. One to 12 implants were placed in each patient.
The secondary criterion was the presence of image Implant regions, bone borders, and adjacent teeth
noise. For visualization quality to be judged excel- for each patient in the dental implant group were
lent, structure definition had to be high while the defined beforehand to ensure consistent evaluation.
image noise had to be negligible. Because the control group participants (CB vs CT)
All observers achieved professional instruction in were not destined for implant treatment, criteria
the visualization software and were allowed to adjust related to the implant region were not evaluated
the brightness and contrast of the images. In most here.
cases, automatically generated visualizations were CB scans in this study were obtained with a pre-
used since the software usually provided optimal set- retail Galileos CB device (Sirona, Bensheim, Germany)
tings. Observers were also permitted to use the zoom- and digital OPG scans were performed with the
ing function provided by all software applications. Orthophos Plus DS (Sirona). All scans were performed
Observer calibration for the different visualization by the same instructed radiographic technical assis-

218 Volume 24, Number 2, 2009

©2009 Quintessence Publishing Co, Inc.

All Rights Reserved
216_Dreiseidler.qxp 3/18/09 2:25 PM Page 219

Dreiseidler et al

Fig 2 Alignment of 3D CB data set visualiza-

tion (top) and the corresponding 3D CT data set
(bottom) in terms of an alignment of the three
orthoradial (axial, coronal and sagittal) planes in
addition with a 3D reconstruction in each of the
utilized software applications. Here, correspond-
ing anatomic regions are contrasted between CB
and CT imaging, for instance allowing compara-
tive evaluation of the visualization quality for the
left mental foramen in the sagittal views for both
imaging modalities.

tant. For both scanners, patients were positioned in an performed. The reconstructed 3D volume had a size
erect position; a light localizer was aligned with the of 15 cm in all three dimensions. Reformatting (sec-
midsagittal plane and the patients contacted the stan- ondary reconstruction) allowed imaging of cross-sec-
dardized forehead rest. Similarly, a bite-fix plane was tional and panoramic views (Fig 1). Visualizations
aligned with the occlusal plane and the anterior teeth were performed with 5123 pixels and a resolution of
were in contact with the anterior teeth stop. The CB 300 µm or 2.5 line pairs per mm.
setup consisted of an x-ray generator and a two- Multiplanar reformations generated from CB
dimensional (2D) charge-coupled distributor/image volume data in the control group were compared
intensifier detector opposite each other. Constant with images obtained with two state-of-the-art
scanning parameters were 90 kV and 28 mAs. Within multidetector row CT devices (16-detector row
14 seconds, 200 single raw projections per scan were scanners Mx8000 IDT and Brilliance 16; both Philips

The International Journal of Oral & Maxillofacial Implants 219

©2009 Quintessence Publishing Co, Inc.

All Rights Reserved
216_Dreiseidler.qxp 3/18/09 2:25 PM Page 220

Dreiseidler et al

CB OPG
5

4
Rankings

1
Image Mental Mandibular Nasal Maxillary Incisive TMJ Implant Bone Adjacent
quality foramen canal floor sinus foramen region borders teeth

Criteria

Fig 3a Descriptive illustration of the CB and OPG ranking results. Each box plot merges five values in one summary plot. Minimum and
maximum values are at the lower and upper ends of the whiskers, the box shows the first to third interquartile range, and the bars repre-
sent the medians. Circles represent outliers.

CB OPG
5

4
Rankings

1
Image Mental Mandibular Nasal Maxillary Incisive TMJ
quality foramen canal floor sinus foramen

Criteria

Fig 3b Descriptive illustration of the CB and CT ranking results. Box plots designed as for Fig 3a. The close alignment of most ranking
results for the visualizations rated as excellent has resulted in the alignment of minimums, maximums, first and third interquartile ranges,
and the medians in a single position. Circles represent outliers.

Medizin Systeme, Hamburg, Germany). The CT exam- reconstruction interval of 1 mm with a voxel size of
inations were carried out using either the paranasal 0.49 ⫻ 0.49 ⫻ 1 mm.
sinuses (PNS) imaging protocol or the computer- The workstation for x-ray visualization and evalua-
assisted surgery imaging protocol applied in exami- tion consisted of two monitors and a computer. The
nations designated for image-guided surgical equipment was technically approved for radiologic
procedures. The PNS protocol employed a slice thick- diagnostics. Both monitors (model E1701, Captiva,
ness of 0.8 mm and a reconstruction interval of 0.4 Bergkirchen-Feldgeding, Germany) were 17 inches in
mm, resulting in a voxel size of 0.49 ⫻ 0.49 ⫻ 0.4 diameter and had a resolution of 1,280 ⫻ 1,024 pixels,
mm. The computer-assisted surgery protocol pro- a contrast of 700:1, and an image refresh rate of 75 Hz.
vided images with a slice thickness of 1 mm and a One monitor was dedicated to the actual radiologic

220 Volume 24, Number 2, 2009

©2009 Quintessence Publishing Co, Inc.

All Rights Reserved
216_Dreiseidler.qxp 3/18/09 2:25 PM Page 221

Dreiseidler et al

Table 2 Summary of the Observed Parameters for Comparison Between CB and OPG Data Sets
Distribution of rankings Sign test value frequencies
Median
Image modality ranking 1 2 3 4 5 Negative Positive Equal P* ICC

Image quality
CB 2 11 85 35 3 1 43 35 57 .428 0.31
OPG 2 26 68 30 8 3 0.68
Mental foramen
CB 2 57 57 10 6 5 19 92 24 8.26–12 0.86
OPG 3 7 53 45 20 10 0.66
Mandibular canal
CB 2 18 68 42 7 0 21 61 53 1.66–5 0.74
OPG 2 4 64 39 24 4 0.53
Nasal floor
CB 2 62 65 8 0 0 6 104 25 2.27–20 0.76
OPG 3 3 63 52 20 24 0.84
Sinus floor
CB 2 62 70 3 0 0 9 92 34 3.37–16 0.73
OPG 3 8 58 48 18 3 0.67
Incisive foramen
CB 2 58 53 13 1 10 7 104 24 8.09–20 0.75
OPG 5 2 19 17 21 76 0.26
TMJ
CB 2 40 77 17 0 1 10 64 61 7.22–10 0.57
OPG 2 7 78 32 12 6 0.73
Implant region
CB 2 36 68 25 5 1 23 67 45 5.83–6 0.75
OPG 2 26 49 26 31 3 0.71
Bone borders
CB 2 45 67 14 7 2 18 85 32 7.86–11 0.73
OPG 3 15 42 36 21 21 0.39
Adjacent teeth
CB 2 37 76 12 9 1 33 64 38 .002 0.47
OPG 2 26 61 27 15 6 0.34
*Significance calculated via the sign test.
Distribution of the CB and OPG grading for the five examiners evaluating 27 data sets accompanied by the corresponding sign test value frequencies
and significance levels, as well as ICC values. The sign test allows three ratings: negative, positive, and equal. “Positive” stands for superior evalua-
tion of the CB over the OPG data set. “Negative” stands for the opposite. “Equal” means both are equally rated. Significance values (P) close to 1
accord to the null hypothesis (H0). A significance value below .05 indicates that there is strong evidence against H0. Significance values below .001
are shown as negative exponents. Significantly better visualization quality was derived from CB imaging for all investigated anatomic structures. Gen-
eral image quality was rated similar for CB and OPG. The average measures of the two-way mixed ICC are displayed here separately for both investi-
gated imaging modalities.

diagnosis and the other was used for data acquisition windowing was set to a range between –2,000 and
via standardized input mask into an Access database 2,000 HU, corresponding to the “bone window”
(Microsoft, Redmond, WA). Evaluation took place in a mode. Soft tissue visualization was not considered
room equipped with window shades and dimmable within the study. Common software features for all
light for a standardized low-lit ambience illumination. software applications were controlled to adjust con-
The computer possessed a 3.4-GHz processing unit trast and brightness as well as zoom. Additional fea-
and a graphics card with 256 MB of memory, 400-MHz tures of the 3D CB software were cross-sectional and
speed, and 128-bit memory interface. tangential imaging from a selectable region within
CB hardware triggering, as well as CB image pro- the panoramic curve (Fig 1), as well as orthogonal
cessing and visualization, were performed by the plane imaging, as usually applied for CT imaging
software accompanying the device (Sirona). The OPG (Fig 2). Possible additional interaction within the
images were processed and visualized using Sidexis orthogonal planes of CB and CT visualizations was
software (Sirona). CT images were processed with traveling through slices. Observers were encouraged
Amira 3.1.1 visualization software (Mercury Computer to use all possible software features to achieve the
Systems, Chelmsford, MA). CT data were imported and optimal subjective image quality.
viewed using Amira’s standard view module. The

The International Journal of Oral & Maxillofacial Implants 221

©2009 Quintessence Publishing Co, Inc.

All Rights Reserved
216_Dreiseidler.qxp 3/18/09 2:25 PM Page 222

Dreiseidler et al

Table 3 Summary of the Observed Parameters for Comparison Between CB and CT Data Sets
Distribution of rankings Sign test value frequencies
Median
Image modality ranking 1 2 3 4 5 Negative Positive Equal P* ICC

Image quality
CB 2 45 85 11 4 0 4 9 130 .25 0.35
CT 2 41 79 20 5 0 0.48
Mental foramen
CB 1 134 10 1 0 0 6 16 123 .1 0.44
CT 1 124 21 0 0 0 0.89
Mandibular canal
CB 2 44 60 22 14 5 28 31 86 .8 0.89
CT 2 52 44 17 29 3 0.95
Nasal floor
CB 1 124 16 3 2 0 8 8 129 1 0.93
CT 1 125 13 4 3 0 0.91
Sinus floor
CB 1 119 22 4 0 0 13 5 127 .1 0.89
CT 1 128 12 5 0 0 0.87
Incisive foramen
CB 1 120 13 3 6 3 4 10 131 .18 0.93
CT 1 117 12 5 6 5 0.95
TMJ
CB 1 132 13 0 0 0 6 6 133 1 0.71
CT 1 132 13 0 0 0 0.79
*Significance calculated via the sign test.
Distribution of the CB and CT grading for the five examiners evaluating 29 data sets accompanied by the corresponding sign test value frequencies
and significance levels, as well as ICC values. See also Table 2a for further explanations of the sign test value interpretation. Visualization for CB and
CT was rated similar for all investigated items. The average measures of the two-way mixed ICC are displayed here separately for both investigated
imaging modalities.

SPSS 13.0 software for Windows (SPSS Inc, distribution were achieved for visualization of the
Chicago, IL) was used for the statistical analysis of the mental and incisive foramina as well as the floor of
generated data. Descriptive analysis was performed the nasal fossa and the maxillary sinus. The best
using box plot visualizations. Statistical analysis was scores for OPG were achieved for general image
performed using a two-sided paired sign test as an quality and for visualization of the mandibular canal,
adequate means for ordinal scaled ranks evaluation, the TMJ, and the implant region (2). The most unfa-
as well as the two-way mixed intraclass correlation vorable scoring results for OPG were achieved for
coefficient (ICC) for evaluation of interrater agree- the nasal floor (3), the bone borders in the implant
ment. Definition of the sign test null hypothesis (H0) region (3), and the incisive foramen (5). In the control
was: The median value of the differences in the group the perceived general image quality was rated
distribution is zero, so that CB – OPG (CB – CT in the good (2), and the investigated anatomic structure
control group) for each criterion should equal OPG – visualizations were consistently rated excellent (1) for
CB (CT – CB in the control group). H0 was considered CB and CT, with the exception of good (2) visualiza-
invalid if the calculated significance level fell below tion of the mandibular canal in CT and CB data sets.
P = .05. In the implant group, sign test results showed that
general image quality was rated slightly better for
OPG, but without reaching significance (P = .309).
RESULTS Visualization qualities for all investigated anatomic
structures were significantly superior for CB imaging.
Single scoring results are contrasted visually in Figs Except for the sinus floor (P = .002), all investigated
3a and 3b. Tables 2 and 3 display the corresponding differences were highly significant (P < .005). Highest
distributions of the single ranking results, sign test significances between both investigated imaging
results, and interrater agreement (ie, ICCs). modalities were achieved for the visualization results
In the implant group the median for the image of the incisive foramen (P = 8.09–20), the nasal floor
quality for all defined criteria was good (2) for CB (P = 2.27–20), and the sinus floor (P = 3.37–16). In the
and between good (2) and insufficient (5) for OPG. control group, sign test results for CB and CT imaging
The best single scores for CB based on single ratings criteria showed no significant visualization differences.

222 Volume 24, Number 2, 2009

©2009 Quintessence Publishing Co, Inc.

All Rights Reserved
216_Dreiseidler.qxp 3/18/09 2:25 PM Page 223

Dreiseidler et al

In the implant group, ICCs in general were higher Because the TMJ was well visualized in all three
for CB than for OPG imaging, indicating a higher dimensions, results of this study can confirm that CB
interrater agreement for CB imaging. The most strik- imaging is a beneficial diagnostic investigation tool
ing ICC discrepancies between CB and OPG were for the TMJ.19,20
achieved for visualization of the incisive foramen and In addition to the already known similar linear and
the bone borders in the implant site. volume measurement accuracy,21 this study confirms
Exceptions in terms of a higher interrater agree- similar visualization qualities for high-contrast struc-
ment for OPG were detected for general image qual- tures between CB and CT data sets. However, it should
ity and for TMJ visualization, where OPG showed be kept in mind that CB systems dedicated to maxillo-
higher ICC values. In the control group, ICCs were facial diagnostics do not provide enough low-contrast
equally rated for CT and CB data sets, but there was a resolution to discriminate soft tissue structures.
tendency toward slightly higher interrater agreement A comparison of CB evaluations from the implant
for CT imaging, especially at the mental foramen. dentistry group and the control group reveals a dis-
crepancy in terms of better CB visualization within
the control group. In addition to accidental impair-
DISCUSSION ment of image qualities in the two different patient
groups, the different age distributions (on average 16
This study was performed to compare the diagnostic years younger in the control group) may have influ-
information gathered by CT, CB imaging, and OPG for enced the outcomes of the present study. Earlier
presurgical implant dentistry assessment to estab- studies showed a tendency toward better visualiza-
lish a basis for weighing the potential diagnostic and tion in younger people in CB imaging,22 presumably
therapeutic benefits of each imaging technology in as a result of a lower rate of metal-dense dental
implant dentistry. The anatomic structures investi- restorations and better compliance abilities during
gated here were derived from the 2001 recommen- the scan procedure.
dations of the American Academy of Oral and Interobserver variability was tested with two-way
Maxillofacial Radiology, 4 which advised that ade- mixed ICC. This analysis method is considered to be
quate presurgical implant dentistry diagnosis should superior to the kappa coefficient when more than
reveal the location of all anatomic features that two observers are involved.23 ICC levels were gener-
should be avoided when placing an implant in such ally higher for CB than for OPG and homologous for
regions as the maxillary sinus, incisive foramen, infe- CB and CT imaging. Low ICC values can indicate
rior alveolar nerve, and mental foramen. Further, poorer picture quality. Thus, the insufficient visualiza-
because of these recommendations, the implant site, tion of the incisive foramen in the OPG is accompa-
its osseous morphology, and adjacent teeth should nied by the lowest investigated ICC value. In turn, the
be evaluated. Visualization of the TMJ in addition to higher ICC values for CB and CT imaging result from
the aforementioned sites can help to identify disor- good visual detection abilities, leaving less tolerance
ders and the need for functional or surgical TMJ for interpretation and leading to higher interrater
treatment in advance of dental implant surgery. agreement. The relatively low ICC value for the gen-
The postoperative OPG control in this study pro- eral image quality in CB imaging may result from the
vided generally good image quality and was suffi- accumulation of different software views, such as
cient for implant placement control. Except for the panoramic simulation as well as transverse and
incisive foramen, all investigated anatomic structures orthogonal slices used for CB method.
were visualized at least sufficiently on the OPG. How- The comparison of the 3D CB images with plain
ever, the results of this study confirm anterior illustra- 2D OPG images here may bias parts of the results on
tion problems for digital OPG18 via its fairly restricted the basis of a lower information degree for OPG.
visualization of the nasal floor and its insufficient However, even when OPG and a second 2D radi-
illustration of the incisive foramen. In addition, OPG ographic investigation are combined, observers still
visualization was ranked as statistically inferior to CB have to deal with the errors inherent in superimposi-
when it came to the illustration of the implant site, its tions, magnifications, cephalometric orientation,
exact bone borders, and neighboring teeth. This is problems of geometric distortion with restrictions to
presumably a result of the missing cross-sectional measurement accuracy, as well as many of the chal-
information and therefore mainly corresponds to the lenges of separating cephalometric landmarks from
missing buccolingual bone border detection ability the structural noise that accompanies 2D imaging
as well as the superimpositions inherent to OPG modalities. Because of the different techniques and
technology. visualization software of the 2D and the 3D imaging

The International Journal of Oral & Maxillofacial Implants 223

©2009 Quintessence Publishing Co, Inc.

All Rights Reserved
216_Dreiseidler.qxp 3/18/09 2:25 PM Page 224

Dreiseidler et al

modalities, observers could not be blinded to the anatomic structures in both jaws prior to implant
data, and this potentially could have biased parts of treatment, resulting in an unique field of view/radia-
the results. tion dose ratio.10
Systematic error in this study may arise from the
fact that, generally, CB imaging was performed prior
to surgery and digital OPG imaging within 2 weeks CONCLUSION
after surgery. This sequence of x-ray investigation
was enforced by the conclusion that OPG imaging in Each application of radiographic technology must
the authors’ clinic is considered to be a sufficient take into account its potential risks and the net ben-
means of routinely applied postoperative control efits it produces. For implant dentistry, the results of
and provides a sufficient amount of information with the present study confirm a superior amount of
a lower radiation dose than CB imaging. The implants information for the CB data sets in contrast to OPG
might have influenced visualization of the adjacent imaging and a CT-like amount of information for
bony structures in OPG imaging, but since nearly no high-contrast structures in CB data sets. Radiation
implant-associated metal artefacts were detected dosages for the investigated 3D CB device are closer
and implants did not obscure the investigated to those seen in OPG rather than CT imaging. These
anatomic structures, only a marginal effect can be circumstances confirm a unique information/radia-
expected from this circumstance. tion dose ratio for CB imaging, possibly justifying its
An important restriction of the study is associated larger-scale application in implant dentistry.
with the use of two different CT machines and two
different CT imaging protocols. However, both CT
scanners were equipped with 16-row detectors using REFERENCES
the same technology and constructed by the same
manufacturer, so that image quality differences were 1. Sakakura CE, Morais JA, Loffredo LC, Scaf G. A survey of radio-
graphic prescription in dental implant assessment. Dento-
probably negligible. The imaging protocols were
maxillofac Radiol 2003;32:397–400.
chosen because of their broad application in routine 2. Lam EW, Ruprecht A, Yang J. Comparison of two-dimensional
diagnosis of the facial skeleton at the authors’ institu- orthoradially reformatted computed tomography and
tion, assuming that both protocols generate images panoramic radiography for dental implant treatment plan-
of a quality that allows for a comparison with CB ning. J Prosthet Dent 1995;74:42–46.
3. Freisfeld M, Dahl IA, Jager A, Drescher D, Schuller H. X-ray diag-
scans. Despite the relatively narrow reconstruction
nosis of impacted upper canines in panoramic radiographs
intervals of 1 mm and 0.34 mm with the CT data, the and computed tomographs. J Orofac Orthop
CB device operates at a still-lower isotropic voxel size 1999;60:177–184.
of 0.30 ⫻ 0.30 ⫻ 0.30 mm. However, this improved 4. White SC, Heslop EW, Hollender LG, Mosier KM, Ruprecht A,
spatial resolution did not have the expected impact Shrout MK. Parameters of radiologic care: An official report of
the American Academy of Oral and Maxillofacial Radiology.
on diagnostic quality, since the additional diagnostic
Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2001;91:
information was mitigated by the higher noise level 498–511.
and inferior contrast resolution inherent in the CB 5. Garg AK, Vicari A. Radiographic modalities for diagnosis and
technology. treatment planning in implant dentistry. Implant Soc 1995;
Concerning CT and CB imaging, it must not be 5:7–11.
6. DelBalso AM, Greiner FG, Licata M. Role of diagnostic imaging
neglected that, nowadays, a broad band of different
in evaluation of the dental implant patient. Radiographics
technologies exists. For instance, the application of 1994;14:699–719.
different low-dose protocols can reduce the radia- 7. Hatcher DC, Dial C, Mayorga C. Cone beam CT for pre-surgical
tion dose in CT imaging of the dental system; this is assessment of implant sites. J Calif Dent Assoc 2003;31:
sometimes accompanied by reduced image qualities 825–833.
8. Tal H, Moses O. A comparison of panoramic radiography with
and small fields of view.11 Different CB devices can
computed tomography in the planning of implant surgery.
also provide different technologies, varying fields of Dentomaxillofac Radiol 1991;20:40–42.
view, and radiation dosages, leading to different 9. Mozzo P, Procacci C, Tacconi A, Martini PT, Andreis IA. A new
image qualities. An evaluation of CB imaging for volumetric CT machine for dental imaging based on the cone-
assessment in implant dentistry therefore must take beam technique: Preliminary results. Eur Radiol 1998;8:
into account different vendors’ technical specifica- 1558–1564.
10. Ludlow JB, Ivanovic M. Comparative dosimetry of dental CBCT
tions. The CB technology used in this study showed devices and 64-slice CT for oral and maxillofacial radiology.
one of the lowest radiation dosages in a study that Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2008;106:
compared four different CB devices while offering a 106–114.
volume size sufficient to visualize all important

224 Volume 24, Number 2, 2009

©2009 Quintessence Publishing Co, Inc.

All Rights Reserved
216_Dreiseidler.qxp 3/18/09 2:25 PM Page 225

Dreiseidler et al

11. Cohnen M, Kemper J, Mobes O, Pawelzik J, Modder U. Radia- 19. Hilgers ML, Scarfe WC, Scheetz JP, Farman AG. Accuracy of lin-
tion dose in dental radiology. Eur Radiol 2002;12:634–637. ear temporomandibular joint measurements with cone beam
12. Ludlow JB, Davies-Ludlow LE, Brooks SL. Dosimetry of two computed tomography and digital cephalometric radiogra-
extraoral direct digital imaging devices: NewTom cone beam phy. Am J Orthod Dentofacial Orthop 2005;128:803–811.
CT and Orthophos Plus DS panoramic unit. Dentomaxillofac 20. Tsiklakis K, Syriopoulos K, Stamatakis HC. Radiographic exami-
Radiol 2003;32:229–234. nation of the temporomandibular joint using cone beam
13. Visser H, Hermann KP, Bredemeier S, Kohler B. [Dose measure- computed tomography. Dentomaxillofac Radiol 2004;33:
ments comparing conventional and digital panoramic radiog- 196–201.
raphy]. Mund Kiefer Gesichtschir 2000;4:213–216. 21. Mischkowski RA, Pulsfort R, Ritter L, et al. Geometric accuracy
14. Broer N, Fuhrmann A, Bremert S, Schulze D, Kahl-Nieke B. Eval- of a newly developed cone-beam device for maxillofacial
uation of transversal slice imaging in the diagnosis of tooth imaging. Oral Surg Oral Med Oral Pathol Oral Radiol Endod
displacement with special consideration of the upper canines. 2007;104:551–559.
J Orofac Orthop 2005;66:94–109. 22. Ritter L, Dreiseidler T, Keeve E, Mischkowski RA, Neugebauer J,
15. Mah JK, Danforth RA, Bumann A, Hatcher D. Radiation Zöller JE. A novel 3D cone-beam imaging device: The influ-
absorbed in maxillofacial imaging with a new dental com- ence of patient age and BMI on image quality. Int J Comput
puted tomography device. Oral Surg Oral Med Oral Pathol Assist Radiol Surg 2006;1(suppl):423–424.
Oral Radiol Endod 2003;96:508–513. 23. Hripcsak G, Heitjan DF. Measuring agreement in medical infor-
16. Hagtvedt T, Aalokken TM, Notthellen J, Kolbenstvedt A. A new matics reliability studies. J Biomed Inform 2002;35:99–110.
low-dose CT examination compared with standard-dose CT in
the diagnosis of acute sinusitis. Eur Radiol 2003;13:976–980.
17. Harris D, Buser D, Dula K, et al. E.A.O. guidelines for the use of
diagnostic imaging in implant dentistry. A consensus work-
shop organized by the European Association for Osseointe-
gration in Trinity College Dublin. Clin Oral Implants Res 2002;
13:566–570.
18. Jacobs R, Mraiwa N, Van Steenberghe D, Sanderink G, Quiry-
nen M. Appearance of the mandibular incisive canal on
panoramic radiographs. Surg Radiol Anat 2004;26:329–333.

The International Journal of Oral & Maxillofacial Implants 225

©2009 Quintessence Publishing Co, Inc.

All Rights Reserved