Community Dent Oral Epidemiol 1999; 27: 431–5 Copyright C Munksgaard 1999

Printed in Denmark . All rights reserved

ISSN 0301-5661

W. Evert van Amerongen1 and

Is ART really atraumatic? Salim Rahimtoola2
1
Department of Cariology Endodontology
Pedodontology, ACTA, Amsterdam, the
Netherlands and 2Dental Clinics, the Aga
Khan University Hospital, Karachi, Pakistan

van Amerongen WE, Rahimtoola S: Is ART really atraumatic? Community Dent

Oral Epidemiol 1999; 27: 431–5. C Munksgaard, 1999

Abstract – Atraumatic restorative treatment (ART) is an approach to the manage-

ment of carious lesions that uses only hand instruments to remove carious tissue
and to restore the tooth involved. The name ART implies that the approach is
atraumatic to both the patient and the tooth. This study set out to evaluate whether
ART is atraumatic in terms of both patient discomfort and tooth tissue conserva-
tion. Three hundred and fifty-nine patients were divided in two groups: one group
was treated with hand instruments and the other with rotary equipment. Each
patient received two restorations: one using amalgam and one using glass ionomer Key words: ART; atraumatic restorative
as the restorative material, placed without the use of anaesthesia. Less discomfort treatment; cavity size; dental caries;
discomfort; glass ionomer; pain
was reported with the ART approach compared to conventional restorations made
using rotary instruments and amalgam. Moreover, preparations with hand in- W.E. van Amerongen, ACTA, Department of
struments were smaller than those produced with rotary instruments. Reported Cariology Endodontology Pedodontology,
Louwesweg 1, postvak 5, 1066 EA
discomfort was associated with the size of the preparation, although the influence Amsterdam, the Netherlands
of the operator on both criteria was considerable. A patient effect was also ob- Tel: π31 20 5188467
served since patients who reported discomfort during the first treatment were Fax: π31 20 5188544
more likely to report discomfort after the second treatment. In conclusion, the e-mail: w.e.van.amerongen/acta.nl
choice of the term ‘‘ART’’ as an atraumatic procedure is defensible. Accepted without peer review 13 April 1999

The atraumatic restorative treatment (ART) ap-
proach was originally developed in the 1980s as a
means of managing dental caries in disadvantaged
areas where extraction would otherwise prevail (1–
3). The name of this approach implies that the
treatment is atraumatic. In the context of ART,
atraumatic could mean the treatment causes no or
minimal trauma:
O To the patient in terms of pain or discomfort;
O To the decayed tooth both in terms of conserva-
tion of sound tooth tissue and with respect to
the pulp; or that,
O Any trauma experienced is less than in other in-
vasive techniques.
One of the objectives of an ART community field
trial established in Pakistan in 1995 was to evaluate
to what extent the term ‘atraumatic restorative
treatment’ (ART) is a suitable descriptor of this ap-
proach to the management of dental caries (4).
The aims of this investigation were to compare
reported discomfort according to the cavity prepa-
ration approach (hand instruments or rotary in-
struments), the restorative material used (glass ion-
omer or amalgam), the size of the cavity prepared,
the operator, and the patient’s previous treatment
experience. In addition, the investigation compared
the size of cavities prepared using hand instru-
ments as against rotary instruments.
Material and methods
A detailed description of the study design has been
published (4). In brief, 359 patients of between 6
and 16 years of age (mean age: 11.6 years) were
selected from seven schools in Karachi, Pakistan,
on the basis of having two or more one-surface car-
ious cavities. Five locally trained dentists provided
the treatment. Half of the patients received prepa-
rations according to the ART approach where cavi-
ties were opened with a hoe or a hatchet and the
carious tissue removed with excavators. The other
half was treated by more conventional means using
rotary instruments. Cavities were opened with
high-speed water-cooled cylindrical diamond burs,

431
Amerongen & Rahimtoola

and soft caries removed with stainless steel round Table 1. Frequency distribution of teeth treated with or with-
burs without cooling. This latter group was called out discomfort according to operator (percentages in paren-
theses)
the minimal cavity preparation (MCP) group. In
both groups each patient received a glass ionomer Discomfort
(Fuji IX, GC Corp.) and an amalgam restoration ART MCP
(Tytin, Kerr). No anaesthesia was used in either
treatment group. Operator No Yes No Yes
The study protocol called for a random assign- 1 53 (74) 19 (26) 45 (68) 21 (32)
ment of the preparation technique and the type of 2 68 (89) 8 (11) 48 (63) 28 (37)
restorative material to be used. Unfortunately, two 3 41 (98) 1 (2) 38 (90) 4 (10)
of the operators failed to follow the random assign- 4 96 (79) 26 (21) 76 (62) 46 (38)
5 34 (68) 16 (32) 22 (44) 28 (56)
ment of preparation technique in certain deep cari- All 292 (81) 70 (19) 229 (64) 127 (36)
ous lesions because they considered that the use of
the ART approach would decrease the possibility
of pulpal exposure. Anticipating that patients treat-
ed by these two operators who received ART re- Table 2. Logistic regression analysis relating to the influence
storations would not show any increase in reported of independent variables on reported discomfort
sensitivity compared to those treated with ART by Likelihood ratio estimates
the other three operators, all patients were includ- Discomfort
ed in this part of the analysis. However, since the
Independent variable P-value Less discomfort
failure to follow a random assignment of prepara-
tion approach in the larger cavities would possibly Preparation technique P∞0.01 ART
influence preparation size comparisons, this part of Restorative material P∞0.05 Glass ionomer cement
Operator P∞0.01 3
the analysis excluded the data from the two opera-
Patient age P±0.05 –
tors concerned. Patient sex P±0.05 –
A simplified method of assessing discomfort and
preparation size was used because of the large
sample size and the number of parameters eval-
uated in the study overall. Discomfort can mean were used for cavity preparation, was almost twice
sensitivity or pain during the operative procedure, as high as that reported by the ART group (19%),
but it can also relate to other unpleasant sensations where hand instruments were used (Table 1). In
such as scraping, e.g., with an excavator, or drilling spite of this, almost two-thirds of patients in the
with a rotary instrument. Even anxiety felt by the MCP group did not report any discomfort. The
patient can be considered to be a form of discom- proportion of patients who reported discomfort
fort. To avoid the potential problems of defining a also varied greatly between the five operators. Op-
subjective feeling, a very pragmatic approach was erator 3 had the smallest proportion of patients re-
chosen. The patient was simply asked, without porting discomfort (2%–10%) while operator 5 had
further explanation, whether or not the treatment the largest proportion of reports of discomfort dur-
caused discomfort. The answer could be ‘yes’ or ing cavity preparation (32%–56%). This finding
‘no’ (5). was common to both the ART and the MCP
After the cavity preparation had been completed, groups.
its size was assessed with cylindrically tipped mea- The influence that preparation technique, opera-
suring devices. The tips had diameters ranging tor, restorative material, patient age and sex had on
from 1 to 6 mm in half-millimetre increments. The patient discomfort was assessed by logistic regres-
diameter of the cavity was estimated as the largest sion analysis (Table 2). The age and the sex of the
tip that could fully enter the cavity. This measure- patient had no influence on reported discomfort.
ment was then transformed into a cubic value as Conversely, all the other assessed variables had a
an estimation of the cavity volume (6). significant influence on reported discomfort, espe-
cially preparation approach and operator. The ART
approach resulted in significantly less reported dis-
Results comfort than the MCP technique.
The proportion of patients that reported discomfort Table 3 shows the relation between discomfort
in the MCP group (36%), where rotary instruments reported by the patient during the two treatment

432
 Is ART really atraumatic?

sessions. In the ART group, almost all patients Table 5. Mean cavity preparation size in mm3 for ART and
who reported no discomfort during the first ses- MCP groups according to operator
sion also reported no discomfort during the sec- Mean cavity size
ond session (95%). Similarly, 70% of patients who Overall mean
reported discomfort during the first treatment Operator cavity size ART MCP P-value
session likewise reported discomfort during the All 5.7 5.1 6.1 P∞0.01
second. These findings were reflected by the MCP 2 5.5 5.3 5.7 P∞0.05
group where 83% of patients failed to report dis- 3 6.1 4.2 8.1 P∞0.01
5 5.6 5.5 5.7 P±0.05
comfort during both the first and the second
treatment while 66% reported discomfort during Note: Operators 1 and 4 excluded from analysis.
both treatment sessions (chi-square: 84.5, PΩ0.00
and chi-square: 44.3, PΩ0.00 respectively for the
ART and MCP groups). preparations according to operator and prepara-
There was a significant relationship between the tion approach. The mean preparation size was 5.7
size of the cavity prepared and reported discomfort mm3. Overall, there was a significant difference be-
(Table 4). Overall, the mean cavity size was much tween cavities prepared by the ART and the MCP
larger in cases where the patient reported discom- approaches whereby the former, made with hand
fort than in those without discomfort. The mean instruments, were smaller than those made with
size of cavities where discomfort was reported was rotary instruments. The operator also had a large
almost three times larger in the ART group than in effect on the difference in cavity size between ART
the MCP group while conversely the mean size of and MCP. This difference was the greatest for oper-
cavities where no discomfort was reported was ator 3, while there was no difference for operator
smaller in the MCP group than in the ART group. 5.
Excluding operators 1 and 4 who failed to com-
pletely follow the randomization of treatment ap-
proaches, Table 5 gives the mean size of cavity
Discussion
The question arises whether the results obtained
from this study are adequate to promote ART as a
Table 3. Frequency distribution of patients treated with either truly atraumatic approach to the management of
ART or MCP according to discomfort reported during first carious lesions. Compared to non-invasive ap-
or second treatment (column percentages for each treatment proaches such as diet counselling, oral hygiene
approach in parentheses) promotion, fluoride applications and other chemi-
Discomfort during cal treatments to arrest the carious process, the an-
Discomfort second treatment swer must be ‘no’. However, ART must be consid-
Treatment during first
ered in the context of other restorative procedures
approach treatment No Yes Total
for caries. These are all invasive since it is not pos-
ART No 137 (95) 7 144 sible to remove soft, carious tooth tissue prior to
Yes 11 26 (70) 37 restoration without some form of drilling, vibration
Total 148 33 181
or scraping. Thus, irrespective of the form of inva-
MCP No 94 (83) 19 113 sive intervention some patients will always con-
Yes 22 43 (66) 65
Total 116 62 178
sider they have experienced some discomfort. This
in turn could be considered to be a form of trauma
to the patient.

Table 4. Mean cavity preparation size in mm3 for ART and Table 6. Frequency distribution of reported discomfort with
MCP groups according to reported discomfort cavities prepared using ART or modified ART treatment ap-
proaches (7)
Discomfort
ART Modified ART
No Yes P-value Discomfort during
treatment nΩ223 (%) nΩ210 (%)
ART 11.0 34.0 P∞0.01
MCP 7.8 12.1 P∞0.05 No 209 (93.7) 184 (87.6)
All cases 9.6 19.9 P∞0.01 Yes 14 (6.3) 26 (12.4)

433
Amerongen & Rahimtoola
On the other hand, if the ART approach to cavity
preparation is compared to an approach that uses
rotary drilling instruments, such as in this present
study, it can be concluded that the former, which
uses hand instruments alone, is more acceptable to
patients and therefore less traumatic. This finding
is corroborated by the findings from a recent ART
study in Indonesia where patient discomfort has
also been studied. This study incorporated a ‘modi-
fied ART’ group (7) where carious cavities were
opened with rotary instruments followed by the re-
moval of all remaining soft carious tissue with
hand instruments. Despite this small difference in
treatment approach, there was significantly more
discomfort reported in the modified ART group
than in the ART group (P∞0.05) (Table 6). It is un-
likely that the opening procedure with rotary in-
struments itself caused physical pain. Either the
use of rotary instruments induced anxiety in the
patients thereby leading to an expression of dis-
comfort or the vibration induced by the drill was
considered uncomfortable. Regardless of the
reasons it is apparent that patients prefer the ART
approach to the use of rotary instruments. This
holds true irrespective of the person who per-
formed the treatment.
The finding that the majority of patients in both
the ART and the MCP groups who reported dis-
comfort during the first treatment session also re-
ported discomfort during the second treatment ses-
sion further suggests that anxiety might be a factor
in patient discomfort. It therefore is obvious that
invasive procedures such as ART which cause little
or discomfort and hence patient trauma should
take precedence over other potentially more trau-
matic procedures. To what extent the cultural back-
ground of the patient will influence the results can
not be said. There are no relevant data from other
studies available.
In this present study the use of amalgam as a
restorative material resulted in more frequent re-
ports of discomfort than the use of glass ionomer.
The reason for this is unclear but might be the pres-
sure used for condensation of the amalgam. It is
unlikely that this procedure is actually painful but
might be considered uncomfortable by some pa-
tients. The influence of the restorative material on
reported discomfort supports the statement that
discomfort can be an expression of many different
factors and stimuli.
This study attempted to evaluate whether ART
was atraumatic to the decayed tooth in terms of
conservation of sound tooth tissue by comparing
434
the size of cavities prepared with the ART ap-
proach to those prepared with rotary instruments.
While overall the preparations made with the ART
approach were significantly smaller than the MCP
preparations, this was operator dependent. Two of
the operators made preparations with the MCP
technique that were only slightly larger than the
ART preparations, operator 3 made them on
average twice as large. In spite of this, while there
is a definite relationship overall between the size
of the preparation and the measure of discomfort it
is interesting to note that operator 3 had the fewest
reports of discomfort.
Despite the results of this study many other
questions remain which need addressing, for in-
stance:
O Is carious tooth tissue removed to the same ex-
tent with hand instruments as with rotary in-
struments?
O Is there a relation between the pressure the oper-
ator uses during cavity preparation in both tech-
niques and the parameters of discomfort and
cavity size?
O Is there a relation between the behavioural man-
agement capabilities of the operator and the re-
ported discomfort?
O Is there a relation between the cultural back-
ground of the patients and the reported discom-
fort?
O Would the levels of discomfort reported be dra-
matically different if an independent investiga-
tor made the inquiries?
There are therefore many issues still to be investi-
gated.
Conclusions
On the basis of the available and limited data from
this study, the following conservative conclusions
can be drawn:
1. More discomfort was reported:
– when cavities are prepared with rotary instru-
ments rather than hand instruments;
– in larger cavity preparations;
– when cavities were filled with amalgam in-
stead of glass ionomer.
2. A patient who experienced discomfort during
the first treatment session was likely to report
discomfort during the second session.
3. The use of hand instruments resulted in smaller
cavity preparations than those produced using
rotary instruments.
4. There is considerable operator variability with

respect to reported discomfort and preparation
size.
Although the ART approach is not truly atrau-
matic in the strictest meaning of the word, as a de-
scriptor for an approach to the restoration of cari-
ous lesions its use is defensible.
References
1. Frencken J, Makoni FA. A treatment technique for tooth
decay in deprived communities. World Health
1994;47:15–7.
2. Frencken J, Phantumvanit P, Pilot T. Manual atraumatic
restorative treatment of dental caries. 2nd ed. The Neth-
erlands: Univ Groningen, WHO Collaborating Centre for
Oral Health Services Research; 1994.
Is ART really atraumatic?
3. World Health Organization. Revolutionary new proce-
dure for treating dental caries. Press release WHO/28.
April 7, 1994.
4. Rahimtoola S, van Amerongen WE. Comparison of two
tooth saving preparation techniques in a treatment ap-
proach of one surface cavities. Design of the study. J Dent
Child 1997;64:334–9.
5. Rahimtoola S, van Amerongen WE. Operative sensitivity
related to different ways of minimal intervention in the
treatment of small carious lesions. Caries Res. In press
1998.
6. Rahimtoola S, van Amerongen WE. Preparation size re-
lated to different ways of minimal intervention in the
treatment of dental caries. J Dent Child. In press 1998.
7. Destri M, Adyatmaka A. Comparison between manually
performed atraumatic restorative treatment and its com-
bination with micro-motors. Paper presented at the FDI
Annual World Dental Congress in Seoul, Sept 1997.
435