Australian Dental Journal

The official journal of the Australian Dental Association

Australian Dental Journal 2015; 60: 301–308

doi: 10.1111/adj.12230

Evaluation and prevention of enamel surface damage during

dental restorative procedures
T Milic,* R George,† LJ Walsh*
*School of Dentistry, The University of Queensland, Brisbane, Queensland.
†School of Dentistry and Oral Health, Griffith University, Gold Coast, Queensland.

ABSTRACT
Background: This study examined the effect of operator experience, dominance, tooth position and access, on frequency
and extent of iatrogenic damage to approximal tooth surfaces during conventional Class II cavity preparations. The
effectiveness of protective devices in minimizing damage was also explored.
Methods: 10 students and 10 experienced dentists each prepared 24 Class II cavity preparations in typodont teeth with-
out protection; 10 utilizing stainless steel matrix bands and 10 utilizing protective wedges. The frequency and extent of
damage were analysed with respect to the above variables. Subsequently, 20 natural and 20 typodont teeth were utilized
to establish the relationship in depth of damage caused by a high-speed diamond bur on typodont versus natural teeth.
Results: Dentists caused iatrogenic damage on 74% of approximal surfaces without protection, which fell to 50% and
46% respectively when matrix bands and wedges were used as guards. The corresponding rates of damage for students
were 94%, 80% and 44%. There was no difference in depth of damage caused on the two types of teeth when bur was
in contact with teeth for a very short time.
Conclusions: Greater operator experience and the use of guards reduces iatrogenic damage to proximal surfaces during
preparation with high-speed rotary instruments.
Keywords: Approximal surface, enamel damage, iatrogenic, protection.
Abbreviations and acronyms: DO = disto-occlusal; MO = mesio-occlusal.
(Accepted for publication 28 September 2014.)

wide, and also appears as fine scratches, indentations

INTRODUCTION
Iatrogenic damage to approximal enamel during
restorative procedures is an important clinical issue. It
has been reported that damage to adjacent teeth is
practically unavoidable during conventional full
crown and Class II cavity preparations.1–3 Qvist et al.
reported that primary teeth with extensive surface
damage require restorative treatment three times as
often as undamaged surfaces within 3.5 years. They
also reported that damaged permanent teeth required
restorations four times as often as undamaged teeth
within five years.4 Taken together, results of in vitro
and in vivo studies indicate that 64% to 97% of
enamel surfaces of teeth adjacent to Class II cavity
preparations may be damaged.1,4 However, Medeiros
et al. reported a lower frequency of damage
(49–60%) and attributed this to elimination of small
defects over time.5
Iatrogenic damage to the enamel surface is most
often in the form of vertical grooves up to 1 mm
© 2015 Australian Dental Association
and extensive damage (a combination of defects over
a large surface area).4,5 Such injuries can potentially
increase the risk of dental caries developing on the
affected surface by altering its properties.2,6–9
Removal of as little as 120 lm (200  80 lm) of
the outer layer of enamel results in an increase in
permeability to acid like, water-soluble molecules.6
As well, furrows and indentations retain dental pla-
que in areas which are inaccessible to dental floss,
which increases the likelihood of caries initiation and
progression.1,2,4,6,10 Furthermore, there is a concern
that enamel damage on approximal surfaces will
appear as interproximal radiolucencies which could
be misdiagnosed as carious cavitations, resulting in
unnecessary restorative treatment of non-carious
surfaces.2,5
Despite their suitability, metal matrix bands are
rarely used to prevent iatrogenic damage during Class
II preparations.1,5,11 The same is true for traditional
tooth separators.5,12 Recently, novel interproximal
301
T Milic et al.
separating wedges with a thin stainless steel plate
have been developed that may be inserted with mini-
mal time and preparation.13 There are no studies on
the effectiveness of use of these novel protective
devices to minimize the extent of enamel iatrogenic
damage inflicted during preparation of approximal
surfaces.14 Therefore, the aim of this study was to
investigate the frequency and extent of iatrogenic
damage during Class II preparations in contemporary
dental practice and compare the effectiveness of novel
and existing protective devices in preventing and mini-
mizing iatrogenic damage. Further, the study also ver-
ified that the assessment of these defects in plastic
typodont teeth is a valid approximation to the clinical
situation.
MATERIALS AND METHODS
Study design
Part 1: Evaluation of iatrogenic damage with and
without guards
A total of 10 final year dental students and 10 experi-
enced (>5 years since graduation) general dentists par-
ticipated in the project. Ethical approval for the study
was obtained from the University of Queensland
School of Dentistry Research Committee, Brisbane,
Australia.
A standardized cavity preparation method was
used, and the written and verbal summary of the
whole process was given to participants before they
consented to participate in the study. The environ-
mental and clinical conditions were maintained the
same for all operators, i.e. identical lighting, phantom
heads; identical dental units using water cooling and
compressed air; handpieces (NSK Pana-Max PAX-TU
M4, Tochigi, Japan); medium grit cylindrical diamond
high-speed burs (Horico Diamant FG108010, Horico,
Berlin, Germany).
This part of the study utilized only plastic typodont
teeth (Frasaco GmBH, Tettnang, Germany), which
were similar to those used by all operators during
their dental school preclinical training. All teeth were
from the same manufacturing batch, to eliminate the
impact of any interbatch variability.
Sets of four typodont plastic premolars, together
with adjacent first and second molars were consecu-
tively mounted in phantom heads. Each operator pre-
pared 24 Class II cavity preparations on first molars
(12 each of mesio-occlusal (MO) and disto-occlusal
(DO) designs). Following this, each operator prepared
10 Class II preparations on first molars, utilizing
either a FenderWedgeâ (Directa AB, Upplands V€asby,
Sweden) interproximally or metal matrix bands for
protection of adjacent teeth.
302
The frequency and extent of enamel surface damage
for the protected surfaces of premolars and second
molars adjacent to the Class II preparations were
compared to those for the unprotected surfaces with
regards to the following parameters: operator experi-
ence, operator dominance (preparations on operators’
dominant versus non-dominant side, i.e. 16 vs 26, 36
vs 46), tooth position (maxillary vs mandibular), ease
of access (MO vs DO), use of guards and their type.
Assessment of damage
The surfaces of premolars and second molars adjacent
to Class II preparations were initially assessed using
3x magnification. Defects were classified by type
(scratches, indentations, grooves, multiple indenta-
tions and/or grooves, extensive damage), which is a
slight modification of an earlier scheme by Medeiros
et al.5 As per Medeiros’ scheme, a groove was defined
as a defect deeper than a scratch, longer than it is
wide and extending vertically or horizontally. An
indentation is a defect where the length is similar to
the width. Scratches are shallow lines, often multiple
and commonly long. Multiple damage is a combina-
tion of grooves and/or indentations that do not spread
over a large area. Extensive damage covers a large
proportion of the enamel surface, and it often consists
of all the other types of damage.5
Width and length of surface defects were measured
using a calibrated digital micrometer to the nearest
0.01 mm (Mitutoyo Corporation, Kawasaki, Japan).
Two assessors analysed all samples independently.
Where results differed by more than 10% the mea-
surement was repeated in duplicate and the average
value of the results that were within 10% from each
other were taken. Further, 3X magnification was used
for the categorization of teeth into four groups based
on the estimated depth of damage (minimal, shallow,
moderate and deep) for further analysis with high pre-
cision instruments. Initially, the teeth with apparent
minimal and maximal damage in each category were
analysed using a non-contact 3D scanner (ATOS
Gesellschaft f€ur Optische Messtechnik GR, GOM
mbH, Braunschweig, Germany), which had a resolu-
tion of 10 lm and an accuracy of ≤1 lm. As a result,
the categories were assigned the following ranges
(minimal <30 lm; shallow from 30 to 60 lm; moder-
ate from 60 to 120 lm; and deep ≥120 lm). All dam-
age of 120 lm and beyond was classified as ‘deep’,
based on the finding of Kuhar et al. that damage to
enamel surfaces to a depth ≥120 lm increases the per-
meability to acid-like molecules.6
Following this, teeth with damage that was assessed
as shallow, moderate and deep with 3x magnification
were analysed using the non-contact 3D scanner. In
addition, selected teeth from all categories including
© 2015 Australian Dental Association
 Enamel surface damage

minimal were analysed to determine maximal depth Part 2: Investigation of the relationship between the
of defects either using a 3D visible light profilometer depth of damage caused on plastic and natural teeth
with vertical accuracy of 60 nm (Veeco Wyko NT under controlled conditions
9100, NY, USA) or a 3D visible laser profilometer
This part of the study used 20 freshly extracted
(Talysurf CCI, Taylor Hobson, Leicester, UK) of the
human teeth as well as 20 typodont teeth. All natural
same accuracy (Fig. 1). Samples of damaged teeth of
human teeth were extracted for orthodontic and other
each defect type (extensive, groove, indentation,
reasons. All teeth were stored at 4 °C and used within
scratch) were sputter coated with gold (2–5 nm) and
six months of extraction (in line with ISO/TS 11405).
examined using scanning electron microscopy (Zeiss
The inclusion criteria were absence of restorations,
Sigma Field Emission SEM with Gemini column, Carl
caries, cracks and visible signs of enamel deteriora-
Zeiss, Obercochen, Germany), to gain representative
tion. All natural teeth were kept fully hydrated and
images of various defects with a high depth of field
the time of exposure to air was limited to five minutes
(Fig. 2).
during every stage of the experiment.
To validate the visual test method used for categori-
An initial viability study was done using 10 plastic
zation of teeth, the results of 30 teeth assessed with
and 10 natural teeth under controlled load (force)
optical microscope were compared to results for depth
from an Instronâ universal testing machine (Model
measurements from the Veeco profilometer and the
5940 Single Column Testing System, Instron Corpora-
3D scanner on the same samples. The acceptance cri-
tion, Norwood, MA, USA). This viability study was
teria of <10% incorrectly categorized teeth was met
done using variable forces (0.2, 0.4, 0.6, 0.8 and 1N)
and the method was validated.
on a medium grit tapered diamond bur (MANI

X Profile
X: 222.2599 um
um
30
mm
1.2

20

Z: 47.3900 um
1.0
10
0.9

0.8 0

0.7
-10

0.6

-20 um
0.5 0 100 200 300 400 500

0.4

0.3
Y Profile
Y: 0.6824 mm
0.2 um
30
0.1
20
0.0 mm
10
0.00 0.10 0.20 0.30 0.40 0.57

-10
X 0.20 - - mm
Y 0.25 - - mm -20

Ht -15.92 - - um -30
Dist - mm
N/A

-40
Angle -
mm
0.00 0.20 0.40 0.60 0.80 1.00

Fig. 1 Depth measurement of 47 lm deep iatrogenic damage taken by the 3D visible light profilometer. The data readout shows the X profile in lm and
the Y profile in mm.

© 2015 Australian Dental Association 303

T Milic et al.
Fig. 2 Damage to a natural tooth in the form of an indentation (A), a
groove (B), and a scratch (C). This particular example is from the palatal
surface of an extracted natural tooth which had been mounted in an
anatomical position along with other teeth in plaster blocks, as part of a
pilot study. The types of defects illustrated on the palatal surface are
typical for the patterns of iatrogenic damage recorded in the study on
approximal enamel surfaces.
Dia-Burs TF-S20, Mani Inc., Tochigi, Japan). The bur
was kept in contact with teeth for approximately
200 milliseconds. No water-cooling was utilized dur-
ing drilling.
Following this initial viability study, a final investi-
gation was undertaken with a further 10 natural and
10 plastic teeth using an expanded range of forces of
0.1 N, 0.2 N, 0.4 N, 0.6 N, 0.8 N and 1 N. The bur-
tooth contact time of 200 milliseconds was recorded
by the machine’s in-built timer and simulated the acci-
dental slip of the bur clinically. The dental drill (NSK
Pana-Max PAX-TU M4, Tochigi, Japan) was
mounted on a holder and attached to the load cell of
an Instronâ Universal testing machine (Model 4848
MicroTester, Instron Corporation, Norwood, MA,
USA) so as to maintain a constant force for each
experimental load. The instrument was calibrated and
had an accuracy of 0.25% of the indicated load. The
bur type utilized was identical to that used in Part 1
(Horico Diamant FG108010) of this study. No water-
cooling was utilized for drilling with the Instron; how-
ever, the bur was immersed in water for a brief time
before each drilling to simulate wet cutting to some
extent. The teeth were analysed using the ATOS non-
contact 3D scanner for measurement of the depth of
defects, in a similar manner as Part 1 of this study.
Evaluation of microhardness of natural and plastic
teeth
The relative microhardness of natural and plastic teeth
was compared. The teeth were fixed to resin sample
304
holders and ground to a flat surface under water
(AudioMet 3000, Buehler, Lake Bluff, Illinois, USA)
and polished progressively to a final grit size of
0.1 lm. The natural teeth remained hydrated during
this preparation. The microhardness (VHN) of the
second set of 10 natural teeth and 10 plastic teeth
was measured using a calibrated microhardness tester
(DuraScan, Struers A/S, Ballerup, Denmark) at a dwell
time of 20 seconds and a load of 25 g, to maintain
consistency with parameters reported in the literature.15
Statistical analysis
A power analysis of the first part of this study was
calculated for two groups of operators (students and
experienced dentists) and three levels of guard.
Assuming a power of 0.8, a type 1 error alpha of
0.05 and a sample size of over 400 teeth, this study
had sufficient power to detect an effect size as small
as 0.015. Similarly, a power analysis of the second
part of this study (power of 0.8 and a sample size of
over 100) was calculated for plastic and natural teeth
and six different forces (0.1 N, 0.2 N, 0.4 N, 0.6 N,
0.8 N, 1 N). Assuming a power of 0.8 and a sample
size of over 100, this study had sufficient power to
detect an effect size of 0.1.
Data were analysed in JMP statistical analysis soft-
ware (Version 10) using chi-squared tests and linear
regression. Damage was analysed both as a continu-
ous and a categorical variable. Chi-squared tests anal-
ysed the effect of operator experience, dominance,
tooth position, access and use of a guard on damage
as a categorical variable.
A multifactorial ANOVA was used to test the rela-
tionship between depth of drilling when varying forces
were applied on the drill on natural and plastic teeth.
Effect tests were performed on type, force and the
interaction between type and force to determine the
statistical significance of all three groups. Differences
between the means were assessed using least squared
means differences Tukey tests.
RESULTS
Demographic information
Of the operators in the study, 55% were males. All
final year dental students were aged in their early twen-
ties. While the age of the dentists was not recorded, all
had five or more years of clinical experience.
Part 1: Evaluation of iatrogenic damage with and
without guards
Initial analysis was conducted using visual scoring of
all teeth from Part 1 at 3x magnification (Table 1).
© 2015 Australian Dental Association
 Enamel surface damage

Table 1. Percentage of surfaces damaged by experi-
enced dentists (ED) and students (S) drilling without
the use of guards, by type of damage
Type of damage % of surfaces
damaged by ED
ND – No damage 25
S – Scratch 14
I – Indentation 13
G – Groove 20
M – Multiple 23
E – Extensive damage 5 21
Total 100
Students caused both multiple and extensive damage
more frequently than dentists.
With regard to variables affecting surface damage,
dominance, tooth position and access did not exert
significant effects, while operator experience and use
of a guard were significant and therefore included
in the final model (p = 0.0086 and p < 0.001).
Greater operator experience and use of a guard
independently reduced the extent of damage
(Table 2). The changes in the mean depths of dam-
age as a function of utilizing guards are illustrated
in Fig. 3.
A linear regression model with defect volume
(calculated from area and depth) as a continuous var-
iable (R square 0.27, p < 0.0001) revealed that both
greater operator experience and use of a guard, and
the interaction between these two factors, were all
significantly associated with a lower volume of dam-
age (p < 0.0001, p < 0.0001 and p = 0.0001 respec-
tively). The Tukey test demonstrated a significant
difference in the volume of damage between students
and experienced dentists when no guard was utilized
(0.6 mm3, CI 0.33–0.79). There was also a significant
difference for both groups with the use of guards,
but no significant difference between the two guard
types (Table 3).
Part 2: Relationship between the depth of damage
caused on plastic and natural teeth
For the initial viability study, a model for depth of
% of surfaces damage was fitted utilizing a multifactorial ANOVA,
damaged by S examining the effect of tooth type, force used, and the
6 interaction between these (R squared = 0.60, p <
12 0.0001). Some 60% of the variation in depth is
10
15
explained by the variation in type and force. Effect
36 tests on type, force and the interaction between these
determined significance of only force as a variable (p <
100
0.0001), with no significant difference in the depth
achieved for drilling between natural and plastic teeth
at the stated contact time of (t ~ 200 ms). The least
squares mean differences Tukey test determined no sig-
nificant difference between natural and plastic at all
levels of force; however, there were differences in the
depth as the force increased incrementally.
Following the success of this initial testing, the data
for the final investigation of drilling into plastic and
natural teeth with an expanded range of forces and an
additional measurement of microhardness were analy-
sed. Microhardness testing was performed twice on
each tooth to control for any measurement errors,
and the average was taken due to the high correlation
(Spearman’s Rho 0.617, p = 0.0038). The initial mul-
tifactorial ANOVA model included a control for
microhardness, which did not significantly contribute
to the model and so was excluded. The average
microhardness for plastic teeth was 56.02 VHN and
for enamel 312.8 VHN.
A model for depth of damage from drilling was
then fitted using multifactorial ANOVA, examining
the effect of type of tooth, force used, and the interac-
tion between them (F = 83, p < 0.001). Some 89% of
the variation in depth could be explained by the varia-
tion in type and force. An effect test on force as a var-
iable was significant (p < 0.0001). Tooth type and the
interaction between type and force were not signifi-
cant (p > 0.05), indicating no statistically significant
difference between natural and plastic teeth at all

Table 2. Percentage of surfaces damaged by experienced dentists (ED) and students (S) drilling with and without
guards, by depth of damage
Extent of damage Frequency of teeth damaged by operators
(depth in lm)
ED – No ED – Stainless ED – Wedge S – No S – Stainless S – Wedge (%)
guard (%) steel band (%) (%) guard (%) steel band (%)

N – No damage (ND) 25 50 54 6 20 56
D – ND < x < 30 11 15 16 5 20 10
D – 30 ≤ x < 60 5 16 24 8 9 10
D – 60 ≤ x < 120 10 6 1 11 14 6
D – 120 ≤ x < 200 20 5 5 26 23 6
D – 200 ≤ x < 300 17 4 0 27 4 8
D – ≥ 300 12 4 0 16 11 4
Total 100 100 100 99 101 100

© 2015 Australian Dental Association 305

T Milic et al.

250 250
Students
Experienced dentists 200
200

Mean Depth (µm)

150

150
100

100 50

0
50 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Force (N)

Fig. 4 The effect of increasing force (in newtons) on depth of damage

0 (in micrometres).
No guard Steel band FenderWedge

Fig. 3 Variations in the mean depth of damage (in micrometres) accord-

ing to operator experience and guard type (no guard, stainless steel band
or integrated wedge and band). Bars show standard deviations. Class II cavity preparations.5 The authors considered
that direct supervision and the influence of assessment
and grading for undergraduate students could explain
Table 3. Least squares means table of volume of the observed low level of iatrogenic damage in their
damage caused by experienced dentists and students study. Students were not individually supervised nor
with and without guards graded in the present study. In addition, the difference
in the statistical power of the present study (480
Operator, guard Mean Lower 95% Upper 95%
volume confidence confidence preparations) compared to the 1999 investigation
(mm3) interval interval (35 preparations) may have contributed to a differ-
ence in the observed effect of operator experience.
ED, None 0.394 0.285 0.504
ED, Tofflemire 0.114 0.004 0.224 The present study found that without the use of
ED, Wedge 0.040 –0.069 0.150 proximal guards, students caused damage to an exten-
S, None 0.954 0.845 1.063 sive area of the tooth four times more often than
S, Tofflemire 0.281 0.172 0.391
S, Wedge 0.186 0.059 0.277 experienced dentists. Further, 69% of damage caused
by students was 120 lm or deeper, compared to 49%
for experienced dentists. As Kuhar et al. demonstrated
forces. Figure 4 illustrates mean depth of damage by that damage deeper than 120 lm increases the perme-
force. There was a significant difference between ability of enamel to acid-like molecules,6 this fre-
depth at forces of 1 N, 0.8 N, 0.6 N, 0.4 N and quency of occurrence of deep iatrogenic damage is
0.1 N, but not between 0.4 N and 0.2 N and between concerning.
0.2 N and 0.1 N forces. The present study examined the majority of vari-
ables that the literature identified as potentially con-
tributory to the occurrence and extent of iatrogenic
DISCUSSION
damage, but did not examine the influence of magni-
This study provides several insights into the factors fying loupes. Lussi et al.2 explored whether the use of
which affect the frequency and severity of damage to loupes could reduce iatrogenic damage in a compre-
approximal surfaces during tooth preparation. Given hensive study that involved at least six months of
that the necessary manual skills for dental clinical training of operators in the use of loupes. A signifi-
practice develop during dental school training and cant increase in the occurrence and extent of damage
improve with practice over time,16,17 it was expected to adjacent tooth surfaces during Class II cavity prep-
that experienced operators would cause less iatrogenic arations on first molars occurred when loupes were
damage than undergraduate students under the same utilized. The results were explained on the basis of
conditions. This prediction was confirmed in that den- further narrowing of the field of view already
tal students damaged 19% more adjacent surfaces restricted by indirect vision.
than experienced dentists. Similar results were Of the variables examined in the present study, use
reported previously in the literature with regard to iat- of a guard was found to reduce the depth and occur-
rogenic damage caused during full crown prepara- rence of iatrogenic damage, with both stainless steel
tions.10 However, a 1999 study by Medeiros et al. matrix bands and protective separating wedges being
reported contradictory results, in that experienced equally effective. Furthermore, as shown in Table 2,
operators caused almost three times more damage stainless steel matrix bands and wedges not only
(64%) than undergraduate students (23%) during prevent damage but minimize it when it occurs.
306 © 2015 Australian Dental Association

Damage was confined to less than 60 lm for experi-
enced dentists in 81% of cases utilizing stainless steel
bands, and in 94% of cases where wedges were uti-
lized. For students, the damage was contained below
60 lm in 49% of cases and 76% of cases utilizing
stainless steel bands and wedges respectively. This
finding is considered particularly important in terms
of mitigating the potential consequences of iatrogenic
damage.
The present study did not find any significant differ-
ence in the benefits provided between protective
wedges with a stainless steel plate and ordinary stain-
less steel matrix bands. Nevertheless, operators in the
study described the interproximal wedges as easier
and faster to place. The operators commented that
they could tend to rely too much on protection affor-
ded by the stainless steel component and by more
aggressive drilling or less attention to fine manual
control may inadvertently have caused deeper iatro-
genic damage by drilling through the steel. This would
explain the finding that damage over 200 lm deep
occurred for students in 15% and 12% of adjacent
teeth when wedges and stainless steel bands were used
respectively, compared to the corresponding 0% and
8% for experienced dentists.
Overall, this study shows that the benefit of protec-
tive systems outweighs the risk of occasional deep iat-
rogenic damage. Therefore, it is recommended that a
protective device be used when more difficult cavity
preparations are being undertaken in clinical practice,
e.g. molar Class II cavity preparations or crown prep-
arations in situations where there are complicating
factors such as more difficult patients, poor access, or
restricted vision. It is further recommended that sepa-
rating wedges or protective matrix bands be used reg-
ularly by students in preclinical training, as well as by
graduates in clinical practice. However, the use of
protective devices does not obviate the need for care
while drilling.
The results of the second part of the study showed no
significant difference in the extent of damage caused
during brief contact (t ~ 200 ms) of a high-speed med-
ium grit rotary diamond bur with plastic and natural
teeth at any of the force levels (0.1 N – 1 N). This indi-
cates that results obtained for damage to Frasaco plas-
tic teeth can be extrapolated to natural teeth.
Despite all attempts to replicate in vivo conditions,
several limitations of the present study must be recog-
nized, particularly with regard to the difficulty in
obtaining sufficient numbers of similar undamaged
natural teeth. Even if this were possible one could not
control for variations in tooth morphology, or ensure
that the teeth were kept fully hydrated whilst being
drilled. Therefore, it was felt that a prospective
in vivo investigation was not appropriate due to
ethical considerations. Nevertheless, this limitation
© 2015 Australian Dental Association
Enamel surface damage
does not impact on the validity of the study as the
second part of the project demonstrated that iatro-
genic damage inflicted on typodont teeth under condi-
tions that simulated accidental slippage of the bur
was representative of the damage that would be
inflicted on natural teeth during the same procedures.
A second, albeit minor, limitation relates to the fact
that the use of the Instronâ was only possible in dry
conditions. There are very few studies on the effect of
dry/wet cutting on extracted natural or plastic teeth
with diamond burs.18 Immersing the bur in water for
a short period of time prior to drilling could not fully
replicate wet cutting conditions. However, Elias et al.
reported that there was no statistically significant dif-
ference in the cutting forces required during wet ver-
sus dry cutting of Class II cavity preparations in
natural teeth using carbide tungsten burs.18
Future studies in this area may investigate any justi-
fiable reasons for the previously observed lack of
operator utilization of protective devices, as well as
examining suitable materials for protection of iatro-
genically damaged surfaces in lieu of classical
restorative work.
CONCLUSIONS
As both novel and traditional protective devices can
prevent iatrogenic damage and minimize its conse-
quences, there is a clear logic in their wider use, par-
ticularly in teaching clinics and preclinical instruction.
The results of the study also stress the need to raise
awareness of the potential for iatrogenic damage and
to remind clinicians of its consequences.
ACKNOWLEDGEMENTS
This study was supported by a research grant from
the Australian Dental Research Foundation and the
2013 Colin Cormie Scholarship. The authors thank
all operators for their participation. We also thank
Mr Nic Croce and Ms Megan Valentine from the
Statistical Consulting Unit (University of Newcastle),
Ms Amanda Cosgriff and Mr Tony Clayton of SDI
Ltd, Mr Dennis Balacano and Mr Matt Smith of
Cochlear Ltd, Mr Kevin Warwick of ScanXpress, and
Mr Khu Vu of the ANU RSPE for their assistance in
using materials testing facilities and equipment.
REFERENCES
1. Lussi A, Gygax M. Iatrogenic damage to adjacent teeth during
classical approximal box preparation. J Dent 1998;26:435–441.
2. Lussi A, Kronenberg O, Megert B. The effect of magnification
on the iatrogenic damage to adjacent tooth surfaces during class
II preparation. J Dent 2003;31:291–296.
3. Long TD, Smith BG. The effect of contact area morphology on
operative dental procedures. J Oral Rehabil 1988;15:593–598.
307
T Milic et al.
4. Qvist V, Johannessen L, Bruun M. Progression of approximal
caries in relation to iatrogenic preparation damage. J Dent Res
1992;71:1370–1373.
5. Medeiros VA, Seddon RP. Iatrogenic damage to approximal
surfaces in contact with Class II restorations. J Dent 2000;28:
103–110.
6. Kuhar M, Cevc P, Schara M, Funduk N. Enhanced permeability
of acid etched or ground dental enamel. J Prosthet Dent
1997;77:578–582.
7. Cevc P, Schara M, Ravnik E, Skaleric U. Study of arrangement
of crystallites in gamma-irradiated human enamel by electron
paramagnetic resonance. J Dent Res 1976;55:581–589.
8. Kuhar M, Cevc P, Schara M, Funduk N. In vitro permeability
and scanning electron microscopy study of acid-etched and
ground enamel surfaces protected with dental adhesive coating.
J Oral Rehabil 1999;26:722–730.
9. Twesme DA, Firestone AR, Heaven TJ, Feagin FF, Jacobson A.
Air-rotor stripping and enamel demineralization in vitro. Am J
Orthod Dentofacial Orthop 1994;105:142–152.
10. Moopnar M, Faulkner KD. Accidental damage to teeth adja-
cent to crown-prepared abutment teeth. Aust Dent J 1991;36:
136–140.
11. Boyde A, Knight PJ. Scanning electron microscope studies of
Class II cavity margins. Br Dent J 1972;133:331–337.
12. Seddon RP. The detection of cavitation in carious approximal
surfaces in vivo by tooth separation, impression and scanning
electron microscopy. J Dent 1989;17:117–120.
308
13. Directa. Design by DentistsTM. Available from: http://www.
directadental.com/. Accessed 14 September 2013.
14. Lenters M, van Amerongen WE, Mandari GJ. Iatrogenic dam-
age to the adjacent surfaces of primary molars, in three differ-
ent ways of cavity preparation. Eur Arch Paediatr Dent
2006;7:6–10.
15. Gutierrez-Salazar M, Reyes-Gasga J. Microhardness and
chemical composition of human tooth. J Mater Res
2003;6:367–373.
16. de Andres AG, Sanchez E, Hidalgo JJ, Diaz MJ. Appraisal of
psychomotor skills of dental students at University Complutense
of Madrid. Eur J Dent Educ 2004;8:24–30.
17. Giuliani M, Lajolo C, Clemente L, et al. Is manual dexterity
essential in the selection of dental students? Br Dent J 2007;
203:149–155.
18. Elias K, Amis A, Setchell DJ. The magnitude of cutting forces
at high speed. J Prosthet Dent 2003;89:286–291.
Address for correspondence:
Professor Laurence J Walsh
School of Dentistry
The University of Queensland
Brisbane QLD 4000
Email: l.walsh@uq.edu.au
© 2015 Australian Dental Association