ArticlePDF Available

Failure rate of atraumatic restorative treatment using high- viscosity glass-ionomer cement compared to conventional amalgam restorative treatment in primary and permanent teeth: a systematic review of Chinese trials

Authors:

Abstract and Figures

BACKGROUND: The last update of this systematic review was assumed to be at risk of language bias, as it did not include the search of major Chinese medical databases. REVIEW OBJECTIVE: This systematic review addition with focus on the Chinese dental literature aimed to answer the question as to whether, in patients with carious cavities of any class in primary and permanent teeth, ART restorations have a higher failure rate than amalgam restorations placed using conventional rotary instruments in tooth cavities of the same size, type of dentition and follow-up period? SEARCH STRATEGY: Databases: Chinese Biomedical Literature Database (CBM), China National Knowledge Infrastructure (CNKI, formerly China Academic Journals), VIP Information and WanFang Data; reference lists of included articles were searched; Strings of search terms were constructed in simplified Chinese. In addition, the English search term “ART” was used for database search; SELECTION CRITERIA: Prospective, clinical controlled trials, with focus relevant to review objective and reporting on computable data with a follow-up period of at least one year were selected from the Chinese dental literature. DATA COLLECTION AND ANALYSIS: Two review authors independently screened and extracted data from, and assessed the risk of bias in, the selected trial reports. Individual datasets were extracted from the trial results and analyzed regarding in-between-dataset heterogeneity and effect size estimates. The investigated outcome was restoration failure. Internal trial validity was assessed in terms of selection-, performance-, detection-, attrition-, publication- and reporting bias. Research gaps in the precision and consistency of the results were evaluated. MAIN RESULTS: Eighteen trials were accepted for review. Of these 36 individual dichotomous datasets could be extracted and analyzed. The majority of the results showed no differences between both types of intervention. High risk of selection-, performance-, detection- and attrition bias was established. Existing research gaps were mainly due to lack of trials and small sample size. CONCLUSION: The current evidence from the Chinese dental literature indicates that the failure rate of high-viscosity GIC/ART restorations is not higher than, but similar to that of conventional amalgam fillings after periods longer than one year. These results appear to corroborate the conclusions drawn during the previous systematic review update. There is a high risk that these results are affected by bias, and thus confirmation by further trials with suitably high number of participants is needed.
Content may be subject to copyright.
Journal of Minimum Intervention in Dentistry 2012; 5: 377-415 © Midentistry cc - ISSN 1998-801X
Available online at www.jmid.org
377
Failure rate of atraumatic restorative treatment using high-
viscosity glass-ionomer cement compared to conventional
amalgam restorative treatment in primary and permanent
teeth: a systematic review of Chinese trials
Wang Xiao-yana, Nie Jiea, Cai Xuea, Veerasamy Yengopalb, Steffen Mickenautschb
a Department of Cariology and Endodontology, Peking University School and Hospital of Stomatology, 22 Zhongguancun South Street, Haidian
District, Beijing, China; e-mail: wangxiaoyan@pkuss.bjmu.edu.cn;
b SYSTEM Initiative / Department of Community Dentistry, Faculty of Health Sciences, University of the Witwatersrand, 7 York Rd.,
Parktown/Johannesburg 2193, South Africa, e-mail: neem@global.co.za.
ABSTRACT
BACKGROUND: The last update of this systematic review was assumed to be at risk of language bias, as it did not
include the search of major Chinese medical databases.
REVIEW OBJECTIVE: This systematic review addition with focus on the Chinese dental literature aimed to
answer the question as to whether, in patients with carious cavities of any class in primary and permanent teeth,
ART restorations have a higher failure rate than amalgam restorations placed using conventional rotary instruments
in tooth cavities of the same size, type of dentition and follow-up period?
SEARCH STRATEGY: Databases: Chinese Biomedical Literature Database (CBM), China National Knowledge
Infrastructure (CNKI, formerly China Academic Journals), VIP Information and WanFang Data; reference lists of
included articles were searched; Strings of search terms were constructed in simplified Chinese. In addition, the
English search term “ART” was used for database search;
SELECTION CRITERIA: Prospective, clinical controlled trials, with focus relevant to review objective and
reporting on computable data with a follow-up period of at least one year were selected from the Chinese dental
literature.
DATA COLLECTION AND ANALYSIS: Two review authors independently screened and extracted data from,
and assessed the risk of bias in, the selected trial reports. Individual datasets were extracted from the trial results
and analyzed regarding in-between-dataset heterogeneity and effect size estimates. The investigated outcome was
restoration failure. Internal trial validity was assessed in terms of selection-, performance-, detection-, attrition-,
publication- and reporting bias. Research gaps in the precision and consistency of the results were evaluated.
MAIN RESULTS: Eighteen trials were accepted for review. Of these 36 individual dichotomous datasets could be
extracted and analyzed. The majority of the results showed no differences between both types of intervention. High
risk of selection-, performance-, detection- and attrition bias was established. Existing research gaps were mainly
due to lack of trials and small sample size.
CONCLUSION: The current evidence from the Chinese dental literature indicates that the failure rate of high-
viscosity GIC/ART restorations is not higher than, but similar to that of conventional amalgam fillings after periods
longer than one year. These results appear to corroborate the conclusions drawn during the previous systematic
review update. There is a high risk that these results are affected by bias, and thus confirmation by further trials
with suitably high number of participants is needed.
Keywords: Systematic review; Chinese trials atraumatic restorative treatment; high-viscosity GIC, amalgam;
tooth restoration; failure rate.
Journal of Minimum Intervention in Dentistry 2012; 5: 377-415 © Midentistry cc - ISSN 1998-801X
Available online at www.jmid.org
378
PLAIN LANGUAGE SUMMARY
Atraumatic Restorative Treatment (ART) is a minimally invasive procedure that involves using only hand
instruments in removing carious tooth tissue and then restoring the resulting cavity with an adhesive restorative
material. In this review, the failure rate of ART with high-viscosity glass-ionomer cement was compared with that
of amalgam fillings placed after drilling. The results from Chinese trials show generally no difference between the
two treatments after a period of at least one year.
1. BACKGROUND
The protocol of this review was registered with the International Prospective Register for Systematic Reviews
(PROSPERO) on 10 July 2012, under registration number CRD42012002621 and separately published [1]. This
review provides an addition to the ongoing systematic review to this topic of which the first report was published in
the Clinical Oral Investigation journal in 2010 [2]. Although this systematic review was updated several times in
2012 [3-5], the need for a systematic review of the Chinese dental literature remained.
1.1 What is new?
This addition provides a quantitative systematic review of the Chinese literature to the topic.
1.2 Description of the condition and intervention
A detailed description of the restoration failure types with particular relevance for atraumatic restorative
treatment (ART) as condition under investigation in this systematic review, as well as a detailed description of the
ART approach as intervention has been given elsewhere [5].
1.3 Specific definition of ART
For the purpose of this systematic review atraumatic restorative treatment (ART) was defined as a tooth
restoration procedure including caries removal by hand instruments, using spoon excavators, and direct cavity
restoration with a high-viscosity GIC [2]. This definition was based on the consideration that ART constitutes a
synthesis of the concepts of:
(i) The retention of remineralisable affected dentine after caries removal by hand excavation [6];
(ii) The promotion of remineralisation of such affected dentine through the placement of a biomimetic
restorative material [7].
Focus on hand excavation: Originally, ART was developed for use in underdeveloped regions [7], to address
the need for inexpensive instrumentation. Other excavation techniques relying on specialized hand instruments in
connection with a chemical agent [8] do not fulfill this criterion.
Focus on GIC: Regarding the material of choice for ART; only GICs have been shown to have a (hyper-)
remineralising (biomimetic) effect on hard tooth tissue [9-11]. GIC may therefore be considered as the only
material currently proven to be capable of effectively remineralising the retained affected dentine.
Focus on high-viscosity: A previous meta-analysis reported higher restoration longevity with high-viscosity
GIC than with low-viscosity GIC for ART [12].
1.4 Why it was important to do this systematic review addition?
The last update of this systematic review [5] was assumed to be at risk of language bias, as it did not include the
search of major Chinese medical databases [13]. It has been suggested that the exclusion of non-English trials may
have little effect on summary treatment effect estimates [14,15]. Initial findings of non-English trials during the last
systematic review update appear to confirm such point [5]. However, any falsification / verification of assumed
language bias risk can only be based on systematic review results hat include all possible evidence sources.
Journal of Minimum Intervention in Dentistry 2012; 5: 377-415 © Midentistry cc - ISSN 1998-801X
Available online at www.jmid.org
379
2. OBJECTIVE
The objective of this quantitative systematic review addition was to assess the failure rate of ART, versus
amalgam fillings, in the permanent or primary dentition in single- or multi-surface cavities, with follow–up periods
from more than one year as investigated in the Chinese dental literature.
PICO question:
[P] Problem / patients: All patients with carious cavities of any class in primary and permanent teeth.
[I] Intervention: ART (as defined per section 1.2)
[C] Comparison: Amalgam restorations placed using conventional rotary instruments in tooth cavities of the same size,
type of dentition and follow-up period
[O] Outcome: Restoration failure
This systematic review sought to answer the PICO question as to whether, in patients with carious cavities of
any class in primary and permanent teeth, ART restorations have a higher failure rate than amalgam restorations
placed using conventional rotary instruments in tooth cavities of the same size, type of dentition and follow-up
period?
3. METHODS
3.1 Systematic literature search
The following electronic databases were searched by two reviewers (NJ and CX) independently: Chinese
Biomedical Literature Database (CBM), China National Knowledge Infrastructure (CNKI, formerly China
Academic Journals), VIP Information and WanFang Data [16].
Strings of search terms were constructed in simplified Chinese. In addition, the English search term “ART”
were used for database search and reference lists of accepted trial reports and systematic reviews, as well as
narrative reviews, if found of importance to the topic, were checked for further suitable trials.
3.2. Criteria for trial consideration
3.2.1 Trial inclusion criteria
From the produced search results, trials were selected based on the following criteria:
 Clinical trials (trials on animals, in-situ, in-vitro trials not included);
 Controlled trials: including control- and test group(s) (1-arm longitudinal trials not included);
 Trial focus relevant to PICO question;
 Prospective trials (retrospective trials not included).
3.2.2 Trial exclusion criteria
From the included trials, trials were excluded based on the following criteria:
 No computable dichotomous or continuous data reported;
 Test and control groups not evaluated the same way;
 Any outcomes measured not in line as per section 3.2.5;
 ART procedure not in line with adopted ART definition as per section 1.2;
 Control procedure not in line with description as per section 3.2.4.
 Follow-up period was less than one year.
Journal of Minimum Intervention in Dentistry 2012; 5: 377-415 © Midentistry cc - ISSN 1998-801X
Available online at www.jmid.org
380
Any apparent duplications i.e. multiple reports of the same trial were also excluded, if they were provisionally
included as being suitable during the previous stage (3.2.1) of the literature selection.
3.2.3. Types of trial participants
Trial participants included all patients of any age, gender or place of origin with carious cavities in either
primary or permanent teeth.
3.2.4 Types of interventions
Test group: Atraumatic restorative treatment as previously described [5] and as defined in section 1.2.
Control group(s): Restorative tooth treatment including the use of rotating instruments for caries removal and
cavity preparation; placement of any material excluding any form of GIC as liner or base followed by tooth
restoration using amalgam.
Amalgam has been used successfully as an universal posterior restorative material for over a century. Its
operative advantages of being relatively simple to place, its intrinsic strength and the longevity of the final
restoration has led to amalgam being considered the “gold standard” against which all new materials may be
measured for outcomes such as the effectiveness and durability of the restoration.
3.2.5 Type of outcome measure
Acceptable outcome measures was the number of failures (n) from the total number of evaluated units (N) for
dichotomous data and the statistical mean (X) of outcomes with standard deviation (SD) or standard error (SE)
from the total number of evaluated units (N) for continuous data. Only primary outcomes with either the patient or
the tooth as unit of analysis were accepted. Only clinical failure reasons as described as previously described [5]
were considered to be primary outcomes. The number of failures (n) / statistical mean (X) of outcomes comprised
of the sum of all failures types due to the reasons described earlier [5].
3.3 Trial selection process
Titles and abstracts of identified articles from data sources (see section 3.1) were scanned by two reviewers (NJ
and CX) in duplication, for possible inclusion in line with the inclusion criteria. Articles with a suitable title but
without listed abstract were retrieved in full copy. All included articles were judged separately by the authors for
possible exclusion with reason or for acceptance, in line with the exclusion criteria. Disagreements between authors
were resolved through discussion and consensus.
3.4 Data collection from accepted trials
Two reviewers (NJ and CX) extracted data from accepted trials independently without being blinded to authors,
institutions, journal name and trial results. Disagreements between authors concerning data extracted were solved
through discussion and consensus. All extracted data were entered in specifically designed data sheets in MS Excel.
The following data was extracted:
3.4.1 Important general information
 Article first author; year of publication and full article reference;
 Place of trial;
 Age, gender of trial participants;
 Participant characteristics, inclusion/exclusion criteria;
 Study design;
Journal of Minimum Intervention in Dentistry 2012; 5: 377-415 © Midentistry cc - ISSN 1998-801X
Available online at www.jmid.org
381
 Per test- and control group:
o Type of restorative material used;
o Type of dentition;
o Type of tooth restored;
o Type of cavity;
o Number of participants at beginning of trial (BSL);
o Assessment method used;
o Assessment criteria followed;
o Number of evaluated units (N);
o Number of failures (n) for dichotomous data;
o Statistical mean of outcomes with standard deviation (SD)/standard error (SE)* for
continuous data;
o Length of trial (follow-up period);
 For ART group: cavity conditioning before GIC placement (yes / no).
* Any SE was to be converted into SD.
3.4.2 Verbatim quotes relevant to intervention integrity
 Article first author; year of publication and full article reference;
 Per test- and control group:
o Patient adherence;
o Patient exposure;
o Quality of delivery;
o Patient responsiveness;
o Any adverse outcomes.
3.4.3 Information concerning methodological trial quality
Any information provided in the trial report concerning: Article first author; year of publication and full article
reference; reporting guidelines followed; sample size calculation used; ethical approval obtained.
3.4.4 Information concerning research gaps related to trial precision
 Article first author; year of publication and full article reference;
 Imprecision of results: e.g. Confidence intervals; sample size;
 Inconsistency of results: e.g. Direction of effect size;
 Lack of right information: e.g. Length of follow-up period;
 PICOS information (Population; Intervention; Comparison; Outcomes; Setting):
o Population: Age; gender, ethnicity clinical characteristics;
o Intervention / Comparison: name of treatment;
o Outcome: measured clinical outcomes;
Journal of Minimum Intervention in Dentistry 2012; 5: 377-415 © Midentistry cc - ISSN 1998-801X
Available online at www.jmid.org
382
o Setting: Type of clinical setting.
3.4.5 Verbatim quotes relevant to selection-, performance- and detection bias risk
 Article first author; year of publication and full article reference;
 Selection bias:
o Random sequence generation;
o Concealment of the sequence allocation;
 Performance bias:
o Operator blinding;
o Patient blinding;
 Detection bias:
o Evaluator blinding.
3.5 Assessment of clinical and methodological heterogeneity
In order to fulfill criteria of clinical and methodological homogeneity datasets from trials should not differ in
the following minimum set of characteristics: Outcome measure; control intervention; assessment method and
length of follow-up period. If in-between-dataset differences were identified, heterogeneity was assumed.
3.6 Data analysis
3.6.1 Calculation of point estimates per dataset
One reviewer conducted the analysis (SM). A dichotomous dataset was defined as any extracted set of n / N for
test- and control group. For each dataset the Risk ratio (RR) with 95% Confidence intervals (CI) and p-values were
computed. A continuous dataset was defined as any extracted set of N, X, SD or SE for test- and control group. For
each dataset the Mean difference (MD) with 95% Confidence intervals (CI) and p-values were computed.
In addition, the point estimate of each dichotomous dataset was converted into an absolute measure (Risk
difference – RD) with 95% Confidence intervals (CI) and p-values, as well as an illustrative comparative risk, i.e.
number of failures out of 100, for both test- and control intervention will be generated with help of the Visual Rx -
Statin Calculator by Cates [17,18].
Statistical significance was set at alpha 5%. For computation of all point estimates the statistical software
programme MIX 1.7 [19] was used.
3.6.2 Assessment and investigation of statistical heterogeneity
The I2 test with 95% CI was used to establish whether any statistical heterogeneity existed between datasets
that were assumed to be clinically and methodologically homogenous (as per section 3.5). The following
thresholds for I2 point estimates (in %) and their upper confidence values were used for interpretation of the test
results [20]:
 0-40% = might not be important;
 30-60% = may represent moderate heterogeneity;
 50-90% = may represent substantial heterogeneity;
 75-100% = considerable heterogeneity.
Journal of Minimum Intervention in Dentistry 2012; 5: 377-415 © Midentistry cc - ISSN 1998-801X
Available online at www.jmid.org
383
For computation of all I2 point estimates with 95% CI the software programme MIX 1.7 was used [19].
3.6.3 Data synthesis (meta-analysis)
A fixed-effects model meta-analysis with RevMan 4.2 was considered to pool clinically
/methodologically/statistically homogenous datasets. A pooled Risk ratio (RR) with 95% CI and p values, was
computed.
In addition, it was considered to convert the computed point estimates into an absolute measure (Risk difference
– RD) with 95% Confidence intervals (CI) and p-values, as well as an illustrative comparative risk, i.e. number of
failures out of 100, for both test- and control intervention was generated with the help of the Visual Rx - Statin
Calculator by Cates [17,18]. Statistical significance was set at alpha 5%.
3.6.3 Sensitivity analysis per dataset and meta-analysis result
In order to test the robustness of dataset and meta-analyses results according to the type of analysis chosen, all
results were recomputed:
 For (dichotomous) dataset results: as Odds ratios (OR);
 For (dichotomous) meta-analysis results: as Odds ratios (OR), using both fixed- and random-
effects models; as Risk ratios (RR), using either fixed- or random-effects models (depending on
which type of model was used as the main method).
3.7 Assessment of selection-, detection- and performance bias risk
Selection-, detection- and performance bias risk was assessed, using the set of criteria in Tables 1 [21] and 2.
Trial reports were also checked for any differences in baseline covariates between groups, using the criteria shown
in Table 3.
3.8 Assessment of attrition bias risk
In order to assess attrition bias risk, a worst- and best-case scenario was assumed. Both were calculated for
dichotomous data whenever the number of lost trial participants per intervention group was reported. The results
were then compared to the intervention outcomes computed for participants available to follow-up and, on this
basis, conclusions concerning attrition bias risk were drawn; a high risk of attrition bias was assumed if the
computed outcomes of these case scenarios changed the result from being statistically significant to non-significant
or vice versa. Where the number of lost trial participants per intervention group was not reported, a high risk of
attrition bias was assumed by default.
The worst-case scenario for dichotomous data was computed by adding the number of lost-to-follow-up
participants in the test group to the failures of that group and adding the number of lost-to-follow-up participants in
the control group to the successes of that group. The best-case scenario was computed by adding the number of
lost-to-follow-up participants in the test group to the successes of that group and adding the number of lost-to-
follow-up participants in the control group to the failures of that group. For each scenario an alternative effect
estimate (RR, 95% CI, p-value) was computed. Alpha level was set at 5%.
3.9 Assessment of publication bias risk
3.9.1 Attempts to identify unpublished trials
The I2 point-estimate with 95% CI of all extracted datasets was computed. High statistical in-between-datasets
heterogeneity as per thresholds (see section 3.6.2) was taken under consideration when assessing publication bias
risk by graphical and statistical methods.
Journal of Minimum Intervention in Dentistry 2012; 5: 377-415 © Midentistry cc - ISSN 1998-801X
Available online at www.jmid.org
384
Graphically a funnel plot was generated, using a fixed-effects model with the Risk ratio (RR) as effect size
estimate from all extracted dichotomous datasets and examined for potential scatter asymmetry. The graphical
findings were statistically verified using Egger’s regression [22]. Statistical significance was set at alpha 5%.
The I2 point-estimate with 95% CI, funnel plot generation and Egger’s regression analysis was computed using
MIX 1.7 statistical software [19].
3.10 Assessment of reporting bias risk
The method and results sections of each accepted trial report were compared for potential discrepancies.
3.11 Assessment of potential influence of split-mouth trial design on systematic review results
All datasets from split-mouth trials that yielded statistically significant differences between both treatment types
were converted into datasets of hypothetical parallel-group trials: a test-scenario was assumed were half the
patients in split-mouth trials were assigned to the ART group and the other half to the amalgam group. The
amalgam restorations in patients assigned to the ART group and the ART restorations in patients assigned to the
amalgam group were excluded by dividing the number of failures (ns) and the total number of evaluated units (Ns)
by two (ns/2; Ns/2). From each resulting hypothetical dataset (nh; Nh) an Odds ratio (ORh) with 95% CI and p-value
was computed and compared to the original result.
4. RESULTS
4.1 Systematic literature search
Figure 1 provides information on the number of articles identified. Details of the search strategy, including
search words and dates of the electronic database search are shown in Table 4. The database search generated 489
citations, while reference check of the previous systematic review [5] to this topic yielded two further citations
[23,24]. From the total of 491 citations, 466 were excluded through scanning of titles and abstracts and 25 [23-47]
were provisionally included (Table 5).
In comparison to the previous systematic review update [5], fourteen new Chinese trial reports [30,31,34,36,38-47],
were identified through the literature search (Table 5). After a review of the full reports, seven trial reports
[23,31,33,35,38,43,47] were excluded on the basis of non-compliance with the exclusion criteria: one report did not
include computable data concerning restoration failure [47]; two reports concerned follow-up periods < 1 year
[23,38]; one report concerned low-viscosity GIC [43] and three articles presented the same data as other accepted
trial reports [31,33,35] (Table 6).
Eleven of the 18 accepted trial reports [26-30,32,36,37,41,44,46] did not specify the type of tooth cavity treated
and seven trial reports [28,32,34,36,39,42,46] did not report as to the type of GIC used. Six further trials
[25,26,30,37,40,44] reported the use of “FX glass-ionomer cement”, manufactured by Matsukaze Co. Ltd/Shofu. in
Japan. However, it was decided not to exclude these trial reports but to consider these characteristics as a reason for
clinical heterogeneity. In total, 18 trials were accepted for further review and data extraction. Of these 16 were
identified in CNKI [16] and two further trials were identified through checking of reference lists [23,24].
4.2 Accepted trials
Information concerning aspects of methodological quality and intervention integrity of the 20 accepted trials is
shown in Tables 7 and 8, respectively.
None of the trials were reported to have followed any guidelines for trial reporting; such as the reporting
guidelines for randomized controlled trials (RCT) CONSORT [48], included sample size calculation or reported
ethical approval. Eight trials [25-28,30,34,40,42] reported that parental consent was obtained (Table 7).
Information from the accepted trial reports concerning intervention integrity for both ART and amalgam
restorations are scarce (Table 8). The available information indicates that (i) ART was perceived to be less painful
Journal of Minimum Intervention in Dentistry 2012; 5: 377-415 © Midentistry cc - ISSN 1998-801X
Available online at www.jmid.org
385
than amalgam restoration [29,39]; (ii) patients were less cooperative when receiving amalgam fillings than when
ART fillings were placed [26,28-30,36,37,39,45].
Figure 1 - Flow diagram of trial selection
4.3 Extracted data
From the 18 Chinese trial reports (Table 7), 36 dichotomous datasets were extracted (Figure 1). No continuous
data was identified.
Table 9 presents details regarding the methodological and clinical characteristics of each dataset. The ages of
trial participants ranged in total from 3 to 12 years for restorations placed in primary teeth and from 16 to 78 years
for restorations placed in permanent teeth. A total of 33 datasets (DS 02-34) related to restorations in the primary,
and 3 datasets (DS 01,35,36) to restorations in the permanent dentition. The study format of 23 datasets followed a
parallel group design and 13 datasets (DS 02-05,12,15,22-26,31,32), a split mouth design. Restorations were
mostly reported to be placed in molar teeth. The restored cavities were single (DS 02,03,19-23,27,28,33,35) and
multiple-surface types (DS 15,24,25,29,30,34,36). No information concerning cavity type was given for all other
datasets. Evaluation was based on modifications of ART criteria or other criteria (Table 9). The length of the
follow-up period ranged from one to three years. Table 9 provides further information regarding the type of filling
material used in either group, the baseline number of participants and included teeth, the evaluation methods, the
number of evaluated teeth and the number of evaluated teeth having restoration failure.
4.3.1 Methodological and clinical heterogeneity
All datasets (Table 9) were considered too heterogeneous in trial method, particularly due to large differences in
the restoration evaluation criteria (Table 9 B [49]) and clinical characteristics. Thus no meta-analysis was
conducted and all 36 datasets were analyzed on an individual basis.
Journal of Minimum Intervention in Dentistry 2012; 5: 377-415 © Midentistry cc - ISSN 1998-801X
Available online at www.jmid.org
386
4.3.3 Dataset results
The computed effect size estimates of all 36 extracted datasets are presented in Table 10 and Figure 2. From the
total 36 datasets, the results of two (DS 26,31) showed in favour of ART and those of three (DS 28,30,34), in
favour of amalgam. No statistically significant difference between ART and amalgam was observed in the other 31
datasets. Reduction of restoration failure risk for ART (for results in favour of ART) ranged between 11 and 12
percentage points (RD). Reduction of restoration failure risk for amalgam (for results in favour of amalgam) ranged
between 12 and 26 percentage points (RD). No difference was observed in the outcome conclusions when the
results were computed as either Risk ratio (RR) or Odds ratio (OR) (Table 10).
Figure 2 - Forrest plot of dataset results (RR with 95% CI)
Studies = Datasets in descending order; RR = Risk ratio
4.5 Summary of results
From the results computed for individual datasets the following summary can be made (Table 11):
(i) No difference in the failure rate between ART and amalgam restoration placed in single- or multiple
surface cavities in permanent teeth after 1 year;
(ii) No difference in the failure rate between ART and amalgam restoration placed in single- and multiple
surface cavities in permanent teeth after 2 years;
(iii) No difference in the failure rate between ART and amalgam restoration placed in single surface
cavities in primary teeth after 1 year;
(iv) Conflicting evidence regarding the failure rate between ART and amalgam restoration placed in
single surface cavities in primary teeth after 2 years;
Journal of Minimum Intervention in Dentistry 2012; 5: 377-415 © Midentistry cc - ISSN 1998-801X
Available online at www.jmid.org
387
(v) Conflicting evidence in the failure rate between ART and amalgam restoration placed in multiple
surface cavities in primary teeth after 1 and 2 years;
(vi) Conflicting evidence in the failure rate between ART and amalgam restoration placed in single- or
multiple surface cavities in primary teeth after 1 year;
(vii) No difference in the failure rate between ART and amalgam restoration placed in single- or multiple
surface cavities in primary teeth after 1.5, 2 or 3 years.
4.6 Internal validity of trials
4.6.1 Assessment of selection- performance- and detection bias risk
Verbatim quotes relevant to the control of selection-, performance- or detection bias risks were extracted from
the trial reports and are presented in Table 12. The bias risk was rated according to the chosen assessment criteria
(Table 1 - 3) and is reported in Table 13.
(i) Selection bias risk: According to the extracted verbatim quotes from the text, none of the trial reports
fulfilled the criteria for A - D rating (Table 1) nor reported on baseline differences between groups.
(ii) Performance bias risk: None of the trial reports provided information concerning the blinding/masking
of operators and/or patients with regard to the type of intervention provided / received.
(iii) Detection bias risk: Two trials [41,44] reported that one person involved in treatment provision, was
the one, who conducted the evaluation process. One trial [34] clarified that one person conducted the
treatment and another person conducted the evaluation.
4.6.2 Assessment of attrition bias risk
Sensitivity analysis was used in computing datasets, under the assumption that either:
(i) All teeth lost to follow-up were restoration failures in the ART group, while all teeth lost to follow-up
in the amalgam group did not fail (Worst-case scenario);
(ii) All teeth lost to follow-up did not fail in the ART group, while all teeth lost to follow-up in the
amalgam group were restoration failures (Best-case scenario).
Data regarding the number of teeth restored at baseline was reported for most datasets (Table 9). The result of
either scenario for each dataset is shown in Table 14. Computation of loss-to follow up numbers per intervention
groups of five trials [26,28,29,32,37] was conducted from the number of evaluated restorations (N) of the previous
dataset, only, and not from the total number of included restored teeth at baseline, as it was unclear whether the
numbers of teeth with failed restorations (n) were reported in these trials as total numbers for the entire follow-up
period or as part of the number of teeth available to follow-up at each particular moment of evaluation.
No information on the number of restoration at baseline was available for six datasets (DS 27-30,35,36). Thus
high bias risk was assumed by default. No changes in the original results were observed in either scenario for 23
datasets. The conclusion of five datasets (DS 03,09,11,13,14) changed from ‘no difference’ to ‘ART having higher
failure rate than amalgam restorations’ in the worst case scenario and ‘ART having a lower failure rate than
amalgam restorations’ in the best case scenario.
Under best-case scenario, the conclusions changed for one datasets that originally showed no difference
between the two interventions, in favour of ART (DS 10) and one dataset that showed originally in favour of
amalgam (DS 34) now found no difference between the two interventions.
4.6.3 Assessment of publication bias risk
Statistical inter-dataset heterogeneity was I2 = 46% (95% CI 21% - 64%). The upper confidence level suggests
possible substantial heterogeneity [20] between all the 36 datasets extracted from the accepted trial reports.
Publication bias was investigated, using one funnel plot (Figure 3). The funnel plot showed an uneven distribution,
Journal of Minimum Intervention in Dentistry 2012; 5: 377-415 © Midentistry cc - ISSN 1998-801X
Available online at www.jmid.org
388
thus indicating publication bias risk. Egger’s linear regression method for the same datasets showed an intercept of
–1.27 (95% CI: -2.45; -0.09); p = 0.04 with statistical borderline significance.
Figure 3 - Funnel plot of dataset results (test for publication bias)
.
RR = Risk ratio; ln = Natural logarithm.
4.6.4 Assessment of reporting bias risk
From the methods and results sections of the accepted trial reports, information concerning intended versus
executed trial focus was extracted. On the basis of the extracted information, a high risk of reporting bias relevant
to systematic review objectives was not observed (Table 15) in the trial reports.
4.7 Assessment of potential influence of split-mouth trial design
Two datasets (DS 26,31) from split-mouth trials were identified to show statistically significant results. Both
showed in favour for ART above amalgam restorations: DS 26: OR 0.31; 95% CI: 0.15 – 0.62; p = 0.0009 and DS
31: 0.35; 95% CI: 0.13 0.96; p = 0.04. When converted into hypothetical parallel group studies the computed
odds ratios were as follows: DS 26: ORh 0.31; 95% CI: 0.12 – 0.82; p = 0.02 and DS 31: ORh 0.33; 95% CI: 0.08 –
1.33; p = 0.12. The conclusion of dataset DS 31 thus changed to ‘no difference’ between the two restorative
treatment types.
Although the point estimates remained largely the same, narrower 95% Confidence intervals were noted for
split-mouth trials as compared to the hypothetical parallel group trials.
4.8 Research gaps
The established sum of results (Table 11) and the results from the datasets (Table 10) were assessed for
potential research gaps regarding precision and consistency. Gaps in results precision were judged on the width of
confidence intervals, availability of trials/data and sufficiency of sample size. Gaps in results consistency were
judged on the direction of effect sizes, magnitude of difference in effect sizes, number trials/datasets and on
overlapping confidence intervals. The results of the assessment are presented in Table 16:
Gaps in precision of results: With the exception of outcomes concerning the failure rate in primary teeth after
one year, all established outcomes lack sufficient trials/data. Outcomes are generally based on insufficient,
Journal of Minimum Intervention in Dentistry 2012; 5: 377-415 © Midentistry cc - ISSN 1998-801X
Available online at www.jmid.org
389
relatively small sample sizes. Relatively wide confidence intervals can be discerned for outcomes related to failure
rates in primary teeth after one year and for single-surface restorations in primary teeth after two years.
Gaps in consistency of results: Outcomes concerning permanent teeth are based on less than two datasets. Thus,
consistency cannot yet be established. Conflict in direction of effect sizes was observed in outcomes for primary
teeth. Conflict in magnitude of effect sizes was observed for outcomes of restorations in primary teeth, except for
single-surface restoration after one year. Non-overlapping confidence intervals were observed in outcomes for
multiple-surface restoration in primary teeth after one year (Table 16).
4.9 Changes from the review protocol
The methods applied in the final version of this systematic review update differ from the registered protocol in
relation to the following points:
(i) Sensitivity analysis was added in order to assess the influence of split-mouth design.
(ii) Regression analysis as well as the use of Galbraith plots for investigation of statistical heterogeneity
was not applied, as methodological heterogeneity particularly due to use of different evaluation criteria
in the identified trials.
(iii) The illustrative comparative risk was computed only for results that showed a statistically significant
difference, with alpha level set at 5%, as doing so for non-significant results might have produced
misleading impressions regarding the inferiority/superiority of one intervention to the other. Instead,
the remark ‘Not significantly different’ was included for non-significant results.
(iv) It was decided not to use length of follow-up as a criterion for research gaps, as the minimum of a one-
year period length had already been specified as sufficient in a/the review methodology.
(v) In order to avoid to many duplication in search results, the CMCC database was excluded from the
literature search on advise by the database administrators, as CMCC was fully covered by the CNKI
and CBM database.
(vi) The English search term “ART” was used instead of “atraumatic restorative treatment” as it was
considered to yield more results during search of Chinese databases than the full English term.
5. DISCUSSION
This systematic review aims to answer the question as to whether, in pati0ents with carious cavities of any class
in primary and permanent teeth, ART restorations have a higher failure rate than amalgam restorations placed using
conventional rotary instruments in tooth cavities of the same size, type of dentition and follow-up period.
5.1 Systematic literature search
The systematic literature search of the four major Chinese databases [16] yielded fourteen trial reports above
the eleven originally identified through GoogleScholar [5] (Table 5). With the exception of one trial [39], the year
of publication of these additional trials lay within the cut-off date of the last systematic review update (January
2012) [5]. However, they were not identified during the previous literature search in 2012 [5] and thus remained
obscure. Such high yield indicates that the raised assumption concerning risk of language bias [13] regarding the
last systematic review update [5] was justified. It was noted that, with the exception of five trials [23,24,43-45] all
identified trial reports were found in the CNKI database. The search in CBM, WanFang and Vip yielded none, one
[43] and two trial reports [44,45], respectively, of which the trial identified in WanFang was subsequently
excluded.
5.2 Systematic review results
Information and data from the 18 accepted trials was extracted and analyzed. The overall results suggest that in
patients with carious cavities of any class in primary and permanent teeth, ART restorations with high-viscosity
GIC have no higher failure rate than amalgam restorations placed with conventional rotary instruments in tooth
Journal of Minimum Intervention in Dentistry 2012; 5: 377-415 © Midentistry cc - ISSN 1998-801X
Available online at www.jmid.org
390
cavities of the same size, type of dentition and follow-up period after one to three years. A confirmatory outcome
of non-English trials in regard to the results published in English language was established. Such observation is in
line with the findings in other studies; that exclusion of non-English trial reports has little effect on summary
treatment effect estimates [14,15].
Most of the computed dataset results showed no differences between ART and amalgam, regardless of type of
dentition, tooth cavity, type of GIC material, evaluation criteria or length of follow-up (Table 10). These results did
not change when computed as Odds ratio (OR) instead of Risk ratio (RR). Any statistical differences observed in
five of the 36 extracted datasets (DS 26,28,30,31,34) may be caused by effect size inflation due to systematic
error/bias, as well as by artificially smaller confidence intervals due to split-mouth study design (DS 26).
Inter-dataset heterogeneity was considered large particularly due to a large variety of evaluation criteria used in
the different trials (Table 9 B). For that reason no meta-analysis was undertaken.
In summary, the results of the individual datasets indicate that the failure rates of high-viscosity GIC/ART and
conventional amalgam restorations are essentially the same, if placed under similar clinical conditions. This
conclusion confirms previous findings [5].
5.3 Validity of systematic review results
This systematic review addition considered the internal validity of the reported trial results in relation to its risk
of bias and thus included an assessment of selection-, performance-, detection-, attrition-, publication- and
reporting bias risk.
5.3.1 Selection bias risk
All of the accepted trials appear to be limited by risk of selection- and detection-/performance bias (Table 13).
Bias or systematic error may affect studies, causing either an over- or an under-estimation of the treatment effect of
an investigated clinical procedure. It has been emphasized that selection bias can only be successfully prevented if
the allocation sequence remains truly random and free from potential interference throughout the trial [21]. Thus, it
is important that trials should include an effective process for concealing the random allocation sequence and that
by the end of each trial this process has indeed prevented direct observation and prediction of the random sequence
allocation [21]. Assessment of the internal validity of trials should therefore be a measure of the result of random
sequence allocation and allocation concealment, and not only of the reported attempts. All trials accepted in this
systematic review failed to report not only on evidence of successful sequence allocation and allocation
concealment results, but also on necessary details about how sequence allocation and allocation concealment were
attempted (Table 12). None of the trials, therefore, provide any verifiable guarantee that each patient had an equal
chance of being allocated to either treatment group and thus, their allocation may have favoured the outcome of one
type of treatment above the other.
5.3.2 Performance- and detection bias risk
From the onset, in all trials successful blinding or masking appeared not to have been possible, owing to the
obvious differences in clinical appearance between ART and conventional amalgam restorations (Table 12 and 13).
For that reason the allocation to either treatment group was visible to patients, operators and evaluators. However,
the difficulties of successful blinding still carry the danger of detection-/performance bias, which may thus have
affected the trials’ results. Potential knowledge of superiority claims prior to the trial may have led patients to
change their oral hygiene habits, operators to place restorations more carefully or evaluators to apply evaluation
criteria more subjectively. This in turn may have favoured the outcome of one type of treatment over the other.
5.3.3 Attrition bias risk
Sensitivity analysis may be used in establishing whether missing data could have affected trial outcomes by
imputing the missing data with replacement values, and treating these as if they were observed, i.e. assuming that
the numbers of restorations lost to evaluation were either failures or successes [50]. Comparison of the analysis
results with reported trial outcomes indicates whether different conclusions should be drawn. Sensitivity analysis
Journal of Minimum Intervention in Dentistry 2012; 5: 377-415 © Midentistry cc - ISSN 1998-801X
Available online at www.jmid.org
391
was conducted for most datasets (Table 14). For six datasets (DS 27-30,35,36) high attrition bias risk was assumed
by default as no number of restorations at baseline was reported. For seven datasets (DS 03,09-11,13,14,34) the
analysis results differed from reported outcomes. Judgment from the datasets without attrition bias risk (Table 14)
that did not show any significant differences in the effect estimates between both interventions may support the
assumption that such outcome would be representative of all restorations included at baseline.
How high the failure rate in the teeth of patients lost to evaluation really was remains unknown. Owing to
uncertainty regarding the real failure rate within those lost to follow-up, the results of the sensitivity analysis
cannot serve as evidence from which clinical conclusions can be drawn. However, the validity of the datasets can
be questioned on grounds of attrition bias. Thus, their results need to be regarded with caution.
5.3.4 Publication- and reporting bias risk
Publication bias was investigated through generation of a funnel plot (Figure 3). Publication bias is present
when the results of published research differ from those of all the studies that have been done [51]. Funnel plots are
scatter graphs showing the size of studies on the Y-axis (large studies on top; small studies at the bottom) and the
effect size, observed in these studies, on the X-axis. The effect sizes of larger studies tend to cluster near the mean.
Small studies have effect sizes that are dispersed across a wider range. Results of both types of studies, plotted on a
scatter graph, give the shape of an inverted, in absence of publication bias, symmetrical funnel [52]. Publication
bias affects a funnel plot in the form of a concentration of studies to one side only (asymmetry). Such asymmetry is
created when particular smaller studies are published only when they show a larger than average effect.
The decision was made to plot results of all the extracted datasets as units of investigation. These are not all
independent from the published trials and this forms a departure from the common application of funnel plots in
investigation for publication bias. Despite this departure, the use of datasets (instead of published trials) will also
indicate potential publication bias when only datasets that show a larger than average effect are published and other
datasets are not.
In this systematic review addition, the funnel plot concerning dichotomous data on restoration failures showed
an asymmetrical spread of dataset results (Figure 3). As the visual judgement of funnel plots is subjective, we
calculated intercepts (95% CI), using Eggers regression [22]. The calculated significant intercept confirmed the
observations derived from the funnel plot. However, whether both graphic and regression results confirm the
existence of publication bias may be questioned on basis of a relative high 95% CI upper limit (64%) of the
computed I2 – value for statistical inter-dataset heterogeneity. Such inter-dataset heterogeneity may shed doubt that
any significant intercept would be due to publication bias risk and not due to differences between datasets
themselves.
In summary, it remains unclear whether the assessment results suggest a potential high risk of publication bias
among the Chinese dental literature to this topic.
Based on the extracted information (Table 15) the risk for reporting bias appears to be low in all trials.
6. RECOMMENDATIONS FOR FURTHER RESEARCH
The assessment of research gaps in terms of precision and consistency of results (Table 16) identified the need
for more and larger trials. Sample size calculation should be an integrated part of future trial methodology, and
aspects concerning intervention integrity for test- and control interventions (as per Table 8) should be assessed and
reported.
Systematic reviews have been reported to provide the highest form of clinical evidence [53]. However, the
internal validity of such evidence can only be as good as the internal validity of the trials reviewed. Although the
trials accepted in this update may be considered to be less affected by reporting- and publication bias, their risk of
selection-, detection-/performance and attrition bias is high. Overestimation due to bias may have particularly
affected those datasets that indicated a statistically significant difference between the two material groups (Table
11) by increasing (or decreasing) the value of point estimates and by artificially narrowing their confidence
intervals. However, the precise effect of bias on all results remains unknown. Thus the results need to be regarded
Journal of Minimum Intervention in Dentistry 2012; 5: 377-415 © Midentistry cc - ISSN 1998-801X
Available online at www.jmid.org
392
with caution and require verification. For that reason, further high quality randomised control trials (RCT) are
needed. Such RCTs should adopt a parallel group design and include randomisation and allocation concealment
methods that can effectively prevent direct observation and prediction of the allocation sequence. For this purpose,
the maximum randomisation method has been suggested [54]. Recently, inclusions of statistical tests based on the
Reverse Propensity Score (RPS), such as the Berger-Exner test, were suggested, to enable authors of trials to
investigate whether selection bias has been introduced into their studies [21,54]. Where bias effect has been found,
it may be adjusted statistically [21]. Both outcomes should be included in the final trial report. In order to ensure
that the lack of blinding may not have led to favoring one treatment over the other, trials should use and report on
procedures and tests employed that may limit, or at least monitor, potential bias risk. Future trials should,
moreover, base their reporting on the CONSORT statement [48].
7. CONCLUSION
In conclusion, the current evidence from Chinese trials indicate that the failure rate of high-viscosity GIC/ART
restorations is not higher than but similar to that of conventional amalgam fillings after periods longer than one
year. These results are in line with the previous systematic review. There is a high risk that these results are
affected by systematic error/bias and corroboration through further studies is needed.
8. ACKNOWLEDGEMENT
The authors would like to thank Mrs. Joy Hull for editing the grammar and syntax of the manuscript of this
report.
9. FUNDING / SPONSORS
No funding provided.
10. CONFLICT OF INTEREST
One of the authors (SM) has been actively involved in the promotion, teaching and research of ART in the past
(from 1998 – 2007).
REFERENCES
[1] Wang X, Nie J, Cai X, Yengopal V, Mickenautsch S. Failure rate of atraumatic restorative treatment using
high-viscosity glass-ionomer cement compared to conventional amalgam restorative treatment in primary and
permanent teeth: a systematic review of Chinese trials [protocol]. J Minim Interv Dent 2012; 5: 355 - 67.
[2] Mickenautsch S, Yengopal V, Banerjee A. Atraumatic restorative treatment versus amalgam restoration
longevity: a systematic review. Clin Oral Investig 2010; 14: 233-40.
Journal of Minimum Intervention in Dentistry 2012; 5: 377-415 © Midentistry cc - ISSN 1998-801X
Available online at www.jmid.org
393
[3] Mickenautsch S, Yengopal V. Failure rate of atraumatic restorative treatment using high-viscosity glass-
ionomer cement compared to conventional amalgam restorative treatment in primary and permanent teeth: a
systematic review update. J Minim Interv Dent 2012; 5: 63 – 124.
[4] Mickenautsch S, Yengopal V. Failure rate of atraumatic restorative treatment using high-viscosity glass-
ionomer cement compared to conventional amalgam restorative treatment in primary and permanent teeth: a
systematic review update - II. J Minim Interv Dent 2012; 5: 213-72.
[5] Mickenautsch S, Yengopal V. Failure rate of atraumatic restorative treatment using high-viscosity glass-
ionomer cement compared to conventional amalgam restorative treatment in primary and permanent teeth: a
systematic review update - III. J Minim Interv Dent 2012; 5: 273-331.
[6] Yoshida Y, Van Meerbeek B, Nakayama Y, Snauwaert J, Hellemans L, Lambrechts P, Vanherle G, Wakasa K.
Evidence of Chemical Bonding at Biomaterial-Hard Tissue Interfaces. J Dent Res 2000; 79: 709-71.
[7] Frencken JE, Pilot T, Songpaisan Y, Phantumvanit P. Atraumatic restorative treatment (ART): rationale,
technique, and development. J Public Health Dent 1996; 56: 135-40.
[8] Ericson D, Zimmerman M, Raber H, Götrick B, Bornstein R, Thorell J. Clinical evaluation of efficacy and
safety of a new method for chemo-mechanical removal of caries. A multi-centre study. Caries Res 1999; 33:
171-7.
[9] Ngo HC, Mount G, Mc Intyre J, Tuisuva J, Von Doussa RJ. Chemical exchange between glass-ionomer
restorations and residual carious dentine in permanent molars: an in vivo study. J Dent 2006; 34: 608-13.
[10] Smales RJ, Ngo HC, Yip KH, Yu C. Clinical effects of glass ionomer restorations on residual carious dentin in
primary molars. Am J Dent 2005; 18: 188-93.
[11] ten Cate JM, van Duinen RNB. Hypermineralization of dentinal lesions adjacent to glass-ionomer cement
restorations. J Dent Res 1995; 74: 1266-71.
[12] Frencken JE, Van 't Hof MA, Van Amerongen WE, Holmgren CJ. Effectiveness of single-surface ART
restorations in the permanent dentition: a meta-analysis. J Dent Res 2004; 83:120-23.
[13] Hurst D. Poor quality evidence suggests that failure rates for atraumatic restorative treatment and conventional
amalgam are similar. Evid Based Dent 2012; 13: 46-7.
[14] Jüni P, Holenstein F, Sterne J, Bartlett C, Egger M. Direction and impact of language bias in meta-analyses of
controlled trials: empirical study. Int J Epidemiol 2002; 31: 115-23.
[15] Moher D, Pham B, Klassen TP, Schulz KF, Berlin JA, Jadad AR, Liberati A. What contributions do languages
other than English make on the results of meta-analyses? J Clin Epidemiol 2000; 53: 964-72.
[16] Xia J, Wright J, Adams CE. Five large Chinese biomedical bibliographic databases: accessibility and
coverage. Health Info Libr J 2008; 25: 55–61.
[17] Ward S, Jones ML, Pandor A, Holmes M, Ara R, Ryan A, Yeo W, Payne N. A systematic review and
economic evaluation of statins for the prevention of coronary events. Health Tech Ass 2007; 11: 14.
[18] Website. Visual Rx - Statin Calculator. Available from http://www.nntonline.net/visualrx/examples/statins/
(Accessed: 03.01.2012).
Journal of Minimum Intervention in Dentistry 2012; 5: 377-415 © Midentistry cc - ISSN 1998-801X
Available online at www.jmid.org
394
[19] Bax L, Yu LM, Ikeda N, Tsuruta H, Moons KGM. Development and validation of MIX: comprehensive free
software for meta-analysis of causal research data. BMC Med Res Methodol 2006; 6: 50.
[20] Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0
[updated March 2011] – Section 9.5.2. The Cochrane Collaboration, 2011. Available from www.cochrane-
handbook.org (Accessed: 03.01.2012).
[21] Berger VW: Selection bias and covariate imbalances in randomised clinical trials. Chichester, UK: John Wiley
& Sons, Ltd., 2005.
[22] Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test.
BMJ 1997; 315: 629-34.
[23] Li X, Hu D-Y, Fan X, Wan H-C, Wei J-H. The observation of primary tooth restoration half a year after
placement to 124 grade one students in primary school (Article in Chinese). Modern Prev Med 1999; 26: 32-33.
[24] Li H-M, Dou Z-H. Clinical observation of using different material in the elderly decaying tooth ART
technique (Article in Chinese). Pract Clin Med 2005; 6: 105-7.
[25] Chen X, Wei X-L. Clinical effects of glass ionomer cement FX in restoring deciduous molar teeth caries
(Article in Chinese). Chin J Conserv Dent 2001; 11: 326-7.
[26] Ling L, Wang X-N. Evaluation of the effects of atraumatic restorative treatment and cooperation degree in
primary caries (Article in Chinese). J Stomatol 2003; 23: 290-1.
[27] Ye X-Y, Liu T, Huang S-H. The effect of ART in the restoration of deciduous molars (Article in Chinese). J
Dent Prev Treat 2006; 14: 39-40.
[28] Chen B-X, Kang J, Guo N, Zhang S-L. A clinical study of atraumatic restorative tretament (ART) in children
with dental caries (Article in Chinese). Acta Acad Med Jiangxi 2006; 46: 97-99.
[29] Qiu H-L. Two-year clinical trial of atraumatic restorative treatment in primary teeth (Article in Chinese).
Stomatol 2007; 27: 413-15.
[30] Wang M-L. The clinical effects of ART in the restoration of deciduous molars (Article in Chinese). Chin Med
Factory Mine 2005; 18: 451-2.
[31] Xue X-M, Qi J-G, Zhang Y-L, Wang Y-L. Evaluation of the effects of ART in primary caries (Article in
Chinese). Ningxia Med J 2002; 24: 741-2.
[32] Li X, Hu D-Y, Wang H-C, Xu X-Y. A Two-year clinical trial of atraumatic restorative treatment in primary
teeth (Article in Chinese). West Chin J Stomatol 2002; 20: 125-7.
[33] Ling L, Wang X-N. Clinical effects of ART in the restoration of proximal occlusal caries of deciduous molars
(Article in Chinese). J Dent Prev Treat 2003; 11: 40-41.
[34] Huang G-Q. A clinical study of atraumatic restorative treatment in primary molars (Article in Chinese). Strait
J Prev Med 2009; 15: 35-36.
[35] She X-Q, Li X, Wan H-C, Fan X, Hu D-Y. A 2 year clinic trial of atraumatic restorative treatment in primary
teeth (Article in Chinese). J Pract Stomatol 2003; 19: 30-33.
Journal of Minimum Intervention in Dentistry 2012; 5: 377-415 © Midentistry cc - ISSN 1998-801X
Available online at www.jmid.org
395
[36] Liu H-B. The clinical effects of ART in the restoration of deciduous teeth (Article in Chinese). Shandong Med
J 2010; 50: 103.
[37] Wang Q, Ding C-S. Evaluation of the effects of ART in primary teeth (Article in Chinese). J Taizhou Polytech
Instit 2004; 4: 63-67.
[38] Li Y. The clinical efficacy of atraumatic restorative treatment of primary teeth caries patients (Article in
Chinese). Guide Chin Med 2011; 9: 191-2.
[39] Lin B-B, Ye W-B. Evaluation of the effects of reformative atraumatic restorative treatment in primary teeth
caries (Article in Chinese). J Med Theo Pract 2012; 25: 557.
[40] Hu Q-L. The clinical effects of ART in deciduous teeth (Article in Chinese). Stomatol 2005; 25: 115.
[41] Mo Q-B. Evaluation of the Effects of ART in the Restoration of Deciduous Molars (Article in Chinese).
Guangxi Med J 2007; 29: 1071-2.
[42] Zeng H-Y, Pan Z-H. ART of the deciduous teeth and the microstructure of the restoration (Article in Chinese).
J Sun Yat-sen Univ (Med Sci) 2006; 27: 27-29.
[43] Liu Y-L, Wu J-Y. The clinical effects of ART in the restoration of deciduous molars (Article in Chinese).
World Health Dig Med Periodical 2009; 8: 69-70.
[44] Weng S-P. The clinical practice of ART in deciduous molars (Article in Chinese). Jiangxi Med J 2005; 40: 95-
96.
[45] Zhang X-L, Fan Y, Hu Z, Wang Y-Y. Restoration of carious primary teeth using atraumatic restorative
treatment (ART) (Article in Chinese). Pract New Med 2001; 3: 389-91.
[46] Wu X-P, Wang L-L, Yu S-C. Therapeutic effects of atraumatic restorative treatment (ART) in management of
117 patients with caries (Article in Chinese). Pract Prev Med 2005; 12: 1081-2.
[47] Peng D, Gao W, Smales RJ, Yip HK. Evaluation of ART and conventional restorative procedures in a clinic
after 30 months (Article in Chinese). J Mod Somatol 2004; 18: 533-5.
[48] Schulz KF, Altman DG, Moher D; CONSORT Group. CONSORT 2010 Statement: updated guidelines for
reporting parallel group randomised trials. BMC Med 2010; 8: 18.
[49] Frenken JE, Holmgren CJ. Atraumatic restorative treatment for dental caries. Nijmegen, STI Book b.v.; 1999.
ISBN 906759024X.
[50] Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0
[updated March 2011] Section 16.1.2. The Cochrane Collaboration, 2011. Available from www.cochrane-
handbook.org (Accessed: 07.02.2012).
[51] Rothstein HR, Sutton AJ, Borenstein M. Publication bias in meta-analysis. In: Publication bias in meta-
analysis – prevention, assessment and adjustment. John Wiley & Sons, Chichester 1-7, 2005.
[52] Rothstein HR, Sutton AJ, Borenstein M. Software for publication bias. In: Publication bias in meta-analysis –
prevention, assessment and adjustment. John Wiley & Sons, Chichester 193-20, 2005.
[53] Mickenautsch S. Systematic reviews, systematic error and the acquisition of clinical knowledge. BMC Med
Res Methodol 2010; 10: 53.
Journal of Minimum Intervention in Dentistry 2012; 5: 377-415 © Midentistry cc - ISSN 1998-801X
Available online at www.jmid.org
396
[54] Berger VW, Alperson SY. A general framework for the evaluation of clinical trial quality. Rev Recent Clin
Trials 2009; 4: 79-88.
Table 1 - Assessment criteria for selection bias risk
Score
Criteria 0 D C B A
Adequate random sequence generation method reported1 N Y Y Y Y
Prevent its direct observation2 N N Y Y Y
Method of concealing the generated random sequence is reported that
is adequate to: Prevent its correct prediction3 N N N Y Y
Evidence is given in some form of statistical test result that indicates the allocated random sequence was
adhered to throughout the trial4 N N N N Y
Conclusion NA A
A = Adequate randomisation; NA = No adequate randomisation; N = Information not provided in the trial report;
Y = Information provided in the trial report.
1 The following methods were considered as inadequate: cluster randomisation, fixed block randomisation with block size 2, minimization,
alternation, randomisation of teeth, use of date of birth or patient record number, “quasi”-randomisation, split-mouth study design, and
generation of the random sequence before patient recruitment.
2 Central randomisation or sequence allocation, by person(s) other than the operator(s), who apply the allocated intervention and who only at
moment of start of the intervention inform the operator(s) which (test- or control) intervention has been allocated to a particular patient (e.g. by
phone), was considered to be adequate.
3 Use of the Maximal randomisation procedure [21] was considered to be adequate.
1-3 Fulfillment of these criteria indicates an adequate attempt of effective randomisation but not that the attempt was indeed successful.
4 Any statistical test that includes the use of the Reverse Propensity Score (RPS) [21] was considered to be adequate. Fulfillment of this criterion
indicates that the attempted effective randomisation was indeed successful; i.e. sufficient proof of low selection bias risk.
Table 2 - Assessment criteria for detection- and performance bias risk
Score
Criteria 0 C B A
Adequate method of masking/blinding of patients and clinicians and evaluators reported1 N Y Y Y
No reasons for doubt discerned from the report text that masking/blinding was not successful N N Y Y
Evidence is given in some form of applied test* result that shows that the masking/blinding was successful
throughout the trial2 N N N Y
Conclusion NA A
A = Adequate masking/blinding; NA = No adequate masking/blinding; N = Information not provided in the trial report;
Y = Information provided in the trial report.
1 Fulfillment of this criterion indicates adequate effective masking/blinding was attempted but not that the attempt was indeed successful.
2 Fulfillment of this criterion indicates that the attempted effective masking/blinding was indeed successful; i.e., proof of lack of
detection/performance bias
* Example: Application of a questionnaire during the trial that assesses beliefs as to which test- or control intervention was given/received to/by
a particular trial participant, followed by statistical comparison of the questionnaire results with the true allocation sequence. A non-significant
result (alpha set at 5%) should sufficiently indicate a successful attempt at masking/blinding and subsequent low detection- and performance
bias risk.
Journal of Minimum Intervention in Dentistry 2012; 5: 377-415 © Midentistry cc - ISSN 1998-801X
Available online at www.jmid.org
397
Table 3 – Criteria for assessment of baseline differences between the intervention– and control groups
* However, differences between other not measured covariates may still exist (Thus Score “A” provides no proof of lack of bias risk)
Table 4 – Search strategy: electronic database search
CNKI search strategy: 2012-07-21
Online: http://dlib.cnki.net/kns50/scdbsearch/cdbindex.aspx Articles found
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
非创伤充填
非创伤修复
非创伤性充填
非创伤性修复
无创伤充填
无创伤修复
无创伤性充填
无创伤性修复
“ART”玻璃离子
3
1
85
33
0
1
2
6
52
Articles included 20
WanFang search strategy: 2012-07-21
Online: http://s.g.wanfangdata.com.cn/PaperAdvancedSearch.aspx
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
非创伤充填
非创伤修复
非创伤性充填
非创伤性修复
无创伤充填
无创伤修复
无创伤性充填
无创伤性修复
“ART”玻璃离子
4
1
48
21
0
1
1
5
14
Articles included 1
Vip search strategy: 2012-07-21
Online: http://162.105.138.192/Visitnew.asp
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
非创伤充填
非创伤修复
非创伤性充填
非创伤性修复
无创伤充填
无创伤修复
无创伤性充填
无创伤性修复
“ART”玻璃离子
5
2
60
34
1
5
2
7
27
Articles included 2
CBM search strategy: 2012-07-21
Online: http://202.112.181.11/cbmbin/login.dll?main
Score Criteria Impact on bias risk
A Baseline data collected before randomisation and
reported for both treatment groups / Data shows no
significant differences between these groups
Evidence is given that randomisation has led to equal
groups in the selected covariates*
B
Baseline data collected before randomisation and
reported for both treatment groups / Data shows
significant differences between these groups but has
been appropriately statistically adjusted
Differences have been adjusted. Thus the influence of
possible selection bias appears to be reduced
C
Baseline data collected before randomisation and
reported for both treatment groups / Data shows
significant differences between these groups without
being statistically adjusted
Reported differences may be due to ineffective
randomisation, thus indicating risk of selection bias
0 Trial does not comply with criteria A – C
No evidence is given as to whether randomisation has
indeed led to equal groups with differences beyond
chance. Thus differences may exist, indicating selection
bias
Journal of Minimum Intervention in Dentistry 2012; 5: 377-415 © Midentistry cc - ISSN 1998-801X
Available online at www.jmid.org
398
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
非创伤充填
非创伤修复
非创伤性充填
非创伤性修复
无创伤充填
无创伤修复
无创伤性充填
无创伤性修复
“ART”玻璃离子
3
1
32
10
0
1
2
6
13
Articles included 0
Table 5 –Provisionally included trials
Inclusion criteria Traced
Database 1st Author Year Reference
SR
Previous
[5] SR
Update
CT CG PICO PROSP Abstract
Full
Wu Xue-ping 2005
[46] X X X X X X X
Peng Dong 2004
[47] X X X X X X X
Chen Xu 2001
[25] X X X X X X X
Ling Li 2003
[26] X X X X X X X
Ye Xiao-ya 2006
[27] X X X X X X X
Chen Bao-xing 2006
[28] X X X X X X X
Qiu Hong-liang 2007
[29] X X X X X X X
Wang Mi-la 2005
[30] X X X X X X X
Xue Xin-min 2002
[31] X X X X X X X
Li Xue 2002
[32] X X X X X X X
Ling Li 2003
[33] X X X X X X X
Huang Gui-qing
2009
[34] X X X X X X X
She Xiao-qing 2003
[35] X X X X X X X
Liu Hai-bo 2010
[36] X X X X X X X
Wang Quan 2004
[37] X X X X X X X
Li Yan 2011
[38] X X X X X X X
Lin Bao-bao 2012
[39] X X X X X X X
Hu Qun-li 2005
[40] X X X X X X X
Mo Qing-bo 2007
[41] X X X X X X X
CNKI
Zeng Hong-yu 2006
[42] X X X X X X X
WanFang Liu Yu-ling 2009
[43] X X X X X X X
Weng Shu-ping 2005
[44] X X X X X X X
Vip Zhang Xiu-li 2001
[45] X X X X X X X
Li Xue 1999
[23] X X X X X X X Reference
check Li Hui-min 2005
[24] X X X X X X X
Total
11 14 25 25 25 25 25 25
SR = Systematic review; CT = Clinical trial; CG = Control group as per PICO question; PICO = Trial focus;
PROSP = Prospective trial; Full = Full trial report traced.
Journal of Minimum Intervention in Dentistry 2012; 5: 377-415 © Midentistry cc - ISSN 1998-801X
Available online at www.jmid.org
399
Table 6 – Excluded trials as per exclusion criteria
Exclusion due to criteria not met
ART AM
Database 1st Author Year Reference
SR
Previous
[5]
SR
Update
CD
GE
OM
FP GIC
HV
HX
nGIC
drill Remarks
Peng Dong 2004 [47] X X No failure rate reported
Xue Xin-min 2002 [31] X Same outcomes also
reported in other trial [32]
Ling Li 2003 [33] X Same outcomes also
reported in other trial [26]
She Xiao-qing 2003 [35] X Same outcomes also
reported in other trial [32]
CNKI
Li Yan 2011 [38] X X Follow-up period 1 month
Wan Fang Liu Yu-ling 2009 [43] X X Fuji II used
Reference
check Li Xue 1999 [23] X X Follow-up period 6 months
Total
7 3 4
SR = Systematic review; CD = No computable data; GE = Groups not evaluated in the same way; OM = Outcome measure not in line; FP =
Follow-up period not at least 1 year; GIC = No conventional glass-ionomer cement used; HV = No high-viscosity GIC used; HX = No hand-
excavation used; nGIC = GIC as liner or base used under amalgam; drill = Rotary instruments not used exclusively; ART = Atraumatic
restorative treatment; AM = Amalgam restoration.
Table 7 – Trial information concerning methodological quality
1st Author Year Reference
Reporting
guidelines
followed
Sample size
calculation
used
Ethical approval
obtained
Wu Xue-ping 2005 [46]
Chen Xu 2001 [25] Parents gave informed consent
Ling Li 2003 [26] Parents gave informed consent
Ye Xiao-ya 2006 [27] Parents gave informed consent
Chen Bao-xing 2006 [28] Parents gave informed consent
Qiu Hong-liang 2007 [29]
Wang Mi-la 2005 [30] Parents gave informed consent
Li Xue 2002 [32]
Huang Gui-qing 2009 [34] Parents gave informed consent
Liu Hai-bo 2010 [36]
Wang Quan 2004 [37]
Lin Bao-bao 2012 [39]
Hu Qun-li 2005 [40] Parents gave informed consent
Mo Qing-bo 2007 [41]
Zeng Hong-yu 2006 [42] Parents gave informed consent
Weng Shu-ping 2005 [44]
Zhang Xiu-li 2001 [45]
Li Hui-min 2005 [24]
Journal of Minimum Intervention in Dentistry 2012; 5: 377-415 © Midentistry cc - ISSN 1998-801X
Available online at www.jmid.org
400
Table 8 – Verbatim quotes relevant to intervention integrity: ART group
1st Author Year
Ref.
PA PE QD PR AA
Wu Xue-ping 2005
[46]
Chen Xu 2001
[25]
Ling Li 2003
[26]
80.19% of children were
cooperative during
treatment; 20 patients were
scared; 1 patient was
uncooperative
Ye Xiao-ya 2006
[27]
Chen Bao-xing 2006
[28]
96.15% claimed to be
willing to undergo ART
again
Qiu Hong-liang
2007
[29]
92.2% claimed to be
willing to undergo ART
again 4.2% of patients felt
pain; 11.3% have slight
feeling (of discomfort)
Wang Mi-la 2005
[30]
87.76% of patients were
cooperative, 11 patients
were scared, 1 patient was
uncooperative
Li Xue 2002
[32]
93.02% claimed to be
willing to undergo ART
again 3.49% of patients felt
pain; 13.95% have slight
feeling (of discomfort)
Huang Gui-qing
2009
[34]
89.3% claimed to be
willing to undergo ART
again
Liu Hai-bo 2010
[36]
36 patients were
cooperative
Wang Quan 2004
[37]
82% of children were
cooperative during
treatment and 91% were
cooperative during first
follow-up after 6 months
Lin Bao-bao 2012
[39]
91.67% patients were
cooperative during
treatment
No difference
between ART group
and amalgam group
regarding age and
gender
18.33% of patients felt
pain
Hu Qun-li 2005
[40]
Mo Qing-bo 2007
[41]
Zeng Hong-yu 2006
[42]
The surface of cavity was rough and
irregular covered with debris.
The restoration was rough with
porosity.
Weng Shu-ping
2005
[44]
Zhang Xiu-li 2001
[45]
92.31% claimed to be
willing to undergo ART
again
Li Hui-min 2005
[24]
‘H’ preparations = hand excavation; SD = Standard deviation; SF = Tooth surface; PA = Patient adherence; PE = Patient exposure; QD =
Quality of delivery; PR = Patient responsiveness; AA = Adverse outcomes; Ref. = Reference number.
Journal of Minimum Intervention in Dentistry 2012; 5: 377-415 © Midentistry cc - ISSN 1998-801X
Available online at www.jmid.org
401
Table 8 – Verbatim quotes relevant to intervention integrity: Amalgam group
1st Author Year
Ref. PA PE QD PR AA
Wu Xue-ping 2005
[46]
Chen Xu 2001
[25]
Ling Li 2003
[26] 50.94% of children were cooperative
during treatment; 47 patients were
scared; 5 patient was uncooperative
Ye Xiao-ya 2006
[27]
Chen Bao-xing 2006
[28]
57.3% claimed to be willing to
undergo drilling with amalgam again;
43.7% expressed fear against drilling
and the noise of the dental engines
Qiu Hong-liang
2007
[29] 72.5 % claimed to be willing to
undergo ART again
9.5% felt pain;
26.5% obvious
feeling (of
discomfort)
Wang Mi-la 2005
[30] 48.98% of patients were cooperative,
43 patients were scared, 7 patient was
uncooperative
Li Xue 2002
[32]
Huang Gui-qing
2009
[34]
Liu Hai-bo 2010
[36] 23 patients were cooperative
Wang Quan 2004
[37]
49% of children were cooperative
during treatment and 41% were
cooperative during first follow-up
after 6 months
Lin Bao-bao 2012
[39] 68.33% patients were cooperative
during treatment
No difference between ART
group and amalgam group
regarding age and gender 41.67% of
patients felt
pain
Hu Qun-li 2005
[40]
Mo Qing-bo 2007
[41]
Zeng Hong-yu 2006
[42]
The surface of cavity
was smooth and
relatively clean.
The restoration was
homogeneous.
Weng Shu-ping
2005
[44]
Zhang Xiu-li 2001
[45] 76.70% claimed to be willing to
undergo the treatment again
Li Hui-min 2005
[24]
‘H’ preparations = hand excavation; SD = Standard deviation; SF = Tooth surface; PA = Patient adherence; PE = Patient exposure; QD =
Quality of delivery; PR = Patient responsiveness; AA = Adverse outcomes; Ref. = Reference number.
Journal of Minimum Intervention in Dentistry 2012; 5: 377-415 © Midentistry cc - ISSN 1998-801X
Available online at www.jmid.org
402
Table 9 – Methodological and clinical characteristics of extracted datasets (Part 1)
Participants
1st Author Year Ref. DS
Place of
Trial Age Gender Characteristics
Study
design ART cavity
Conditioned
Wu Xue-ping 2005 [46] 01 Changsha,
Hunan 16~44 yrs,
mean 21.8yrs
Patients with one or more teeth with dentin caries, cavity single or small multiple surface
large enough for smallest excavator to enter; no pulp symptoms PG
02
Chen Xu 2001 [25] 03 Shenyang 4-8 yrs Children with caries on both sides of his/her (mouth) SM
04
Ling Li 2003 [26] 05 Wuxi 6-8 yrs 53 m; 53 f Shallow or intermediate carious lesions on both sites of the mouth;
no pulp symptoms or
periodontal disease; cavity size less than 2/3 of occlusal surface SM
Ye Xiao-ya 2006 [27] 06 Guangdong 3-4 yrs 73m; 65f
Cavity single surface large enough for smallest excavator to enter; no pulp symptoms or
periodontal disease; tooth symptom free PG
07
08
Chen Bao-xing 2006 [28]
09
Liaocheng
ART group:
mean 9.7 yrs;
Amalgam group:
mean 9.6 yrs
ART group:
74 m; 46 f
Amalgam group:
65 m; 55 f
Children with occlusal or buccal fissure caries into dentin; for each patient 2 teeth were
selected; cavity single or small multiple surface large enough for
smallest excavator to
enter; no pulp symptoms or periodontal disease; tooth symptom free; parental and patient
consent
PG
10
Qiu Hong-liang
2007 [29] 11 Dongying One or more carious lesion into dentin; cavity single or small mult
iple surface large enough
for smallest excavator to enter; no pulp symptoms or periodontal disease; tooth symptom
free PG
Wang Mi-la 2005 [30] 12 Jiaozuo, Henan
5-8 yrs 52 m; 46 f Shallow or intermediate carious lesions of the same primary molar on both sides of the
mouth, no pulp symptoms or periodontal disease SM
13
Li Xue 2002 [32] 14 Chengdu 6-
8 yrs, mean 6.08
yrs One or more carious lesion into dentin;
cavity single or small multiple surface large enough
for smallest excavator to enter;
no pulp symptoms or periodontal disease; tooth symptom
free PG
Huang Gui-qing
2009 [34] 15 Fuzhou, Fujian
6-8 yrs 45 m; 45 f Shallow or intermediate carious lesions of the same primary molar on both sides of the
mouth, no pulp symptoms or p
eriodontal disease. The cavities are large enough for smallest
excavator to enter SM
Liu Hai-bo 2010 [36] 16 Qinhuangdao,
Hebei 3-10 yrs 40 m; 46 f C
hildren with one or more teeth with shallow or intermediate carious lesions, no pulp
symptoms or periodontal disease, The cavities are large enough for smallest excavator to
enter, but smaller than 2/3 of occlusal surface PG
17
Wang Quan 2004 [37] 18 Taizhou 108 m; 92 f
Cavity single or small multiple surface large enough for smallest excavator to enter; no
pulp symptoms PG
19
20
Lin Bao-bao 2012 [39]
21
Zhongshan,
Guangdong
6-12 yrs /
ART group: mean
9.8yrs; Amalgam
group: 9.7 yrs.
ART group:
37m; 23 f;
Amalgam group:
35 m; 25 f.
Children with 2 teeth with carious lesions on occlusal or buccal surface PG
22 Hu Qun-li 2005 [40] 23 Yongkang,
Zhejiang 5-8 yrs 114 teeth of male,
118 teeth of female Shallow or intermediate carious lesions of the same primary molar on both sides of the
mouth, no pulp symptoms or periodontal disease SM
Journal of Minimum Intervention in Dentistry 2012; 5: 377-415 © Midentistry cc - ISSN 1998-801X
Available online at www.jmid.org
403
24
25
Mo Qing-bo 2007 [41] 26 Beihai,
Guangxi 5-9 yrs / mean 6.5
yrs, SD 0.59 84 m; 116 f Intermediate carious lesions of the same primary molar on both sides of the mouth SM
27
28
29
Zeng Hong-yu 2006 [42]
30
Jiangmen,
Guangdong 8-10 yrs C
hildren with one or more teeth with shallow or intermediate carious lesions, no pulp
symptoms, no restoration on the teeth PG
31
Weng Shu-ping
2005 [44] 32 Yuyao,
Zhejiang 5-9 yrs / mean 7.4
yrs, SD 0.59 38 m; 44 f Intermediate carious lesions of the same primary molar on both sides of the mouth SM
33
Zhang Xiu-li 2001 [45] 34 Beijing 5-8 yrs One or more teeth with caries, cavities large enough for smallest excavator to
enter; no pulp
symptoms PG
35
Li Hui-min 2005 [24] 36 Jiangmen 60-78 yrs 98 m; 106 f PG
Table 9 – Methodological and clinical characteristics of extracted datasets (Part 2)
1st Author
Year
Ref.
DS ART group
GIC Dentition Tooth
Cavity P BSL
Evaluation
Method Criteria N n Follow-up
Period
Wu Xue-
ping 2005
[46]
01 GIC (Dentsply) (permanent) 1SF/
>1SF 64 124 Clinical Failure:
Not present, restoration has (almost)
completely disappeared.
Secondary caries or
periodontal diseases appeared. 124
15 1 yr
02 52 5 1 yr
Chen Xu 2001
[25]
03
‘FX glass ionomer’
by a Matsukaze Co
Ltd Japan primary molar
1SF 104 (52) Clinical
Failure: Marginal gap between restoration and
tooth; caries; partly or completely lost; pain 21 3 2 yrs
04 106 106 106
16 1 yr
Ling Li 2003
[26]
05
‘FX glass ionomer’
by a Matsukaze Co
Ltd Japan primary molar
(1SF/
>1SF) 106 106 Clinical ART/Mod.3 106
23 2 yrs
Ye Xiao-ya 2006
[27]
06 Fuji IX primary molar
1SF/
>1SF 70 186 Clinical ART/Mod.6 180
30 1 yr
07 120 240 231
24 1 yr
08 120 240 225
55 2 yrs
Chen Bao-
xing 2006
[28]
09
primary 1SF/
>1SF 120 240
Clinical ART/Mod.1
215
57 3 yrs
10 70 198 182
10 1 yr Qiu Hong-
liang 2007
[29]
11 Ketac Molar primary 1SF/
>1SF 70 198 Clinical ART/Mod.4 168
23 2 yrs
Wang Mi-la
2005
[30]
12 Glass ionomer FX primary molar
(1SF/
>1SF) 98 98 Clinical Failure: Not present, restoration has partly
or
completely disappeared. Marginal defect, pulp
symptoms or secondary caries appeared 98 12 18 mo
13 128 154 132
9 1 yr
Li Xue 2002
[32]
14 glass ionomer
(Dentsply)’ primary 1SF/
>1SF 128 154 Clinical ART/Mod.5 118
22 2 yrs
Journal of Minimum Intervention in Dentistry 2012; 5: 377-415 © Midentistry cc - ISSN 1998-801X
Available online at www.jmid.org
404
Huang
Gui-qing 2009
[34]
15 3M ESPE AG,
Germany primary molar
>1SF 90 90 Clinical ART/Mod. 7 90 28 1 yr
Liu Hai-bo 2010
[36]
16 primary 1SF/
>1SF 44 104 Clinical Failure:
Not present, restoration has (almost)
completely disappeared.
Secondary caries or
pulp symptoms appeared. 104
3 1 yr
17 160
28 1 yr
Wang Quan
2004
[37]
18
‘FX glass ionomer’
by a Matsukaze Co
Ltd Japan primary molar
1SF/
>1SF 100 165 Clinical
Failure: Marginal gap between restoration and
tooth; need for repair; partly or completely
lost; 157
49 2 yrs
19 115
12 1 yr
20 112
27 2 yrs
Lin Bao-bao
2012
[39]
21
(?) 1SF 60 120 Clinical Failure: Not prese
nt, restoration has
completely disappeared.
Secondary caries
appeared or not. 108
29 3 yrs
22 116
15 1 yr
23 1SF
(Class I) 116 116
24 2 yrs
24 116
22 1 yr
Hu Qun-li 2005
[40]
25
Glass ionomer FX
(Matsukaze Co.
Ltd. Japan) primary molar
>1SF
(Class II)
116 Clinical ART/Mod. 8
116
30 2 yrs
Mo Qing-bo
2007
[41]
26 Fuji IX (GC Co.
Ltd. Japan) primary molar
(1SF/
>1SF) 185 Clinical ART Mod. 9 182
12 1 yr
27 161
14 1 yr
28 1SF 159
41 2 yrs
29 289
57 1 yr
Zeng
Hong-yu 2006
[42]
30
3M ESPE AG,
Germany primary molar
>1SF 218 (453) Clinical
SEM
LM ART Mod. 6
286
106
2 yrs
31 82 6 1 yr
Weng Shu-
ping 2005
[44]
32
Glass ionomer FX
(Matsukaze Co.
Ltd. Japan) primary molar
(1SF/
>1SF) 82 82 Clinical ART Mod. 9 82 13 2 yrs
33 1SF
(Class I) 38 36 1 1 yr
Zhang
Xiu-li 2001
[45]
34 Fuji IX primary >1SF
(Class II)
53 Clinical ART Mod. 6 50 16 1 yr
35 1SF
(Class I) 54 15 2 yrs
Li Hui-min 2005
[24]
36 Fuji IX permanent
>1SF
(Class II)
(?) Clinical
Failure: Pain; Marginal gap between
restoration and tooth; marginal discoloration;
partly or completely lost; caries 72 37 2 yrs
Journal of Minimum Intervention in Dentistry 2012; 5: 377-415 © Midentistry cc - ISSN 1998-801X
Available online at www.jmid.org
405
Table 9 – Methodological and clinical characteristics of extracted datasets (Part 3)
Amalgam group
1st Author Year
Ref. DS
AM Dentition Tooth
Cavity P BSL Evaluation
Method Criteria N n Follow-up
Period
Wu Xue-
ping 2005
[46] 01 1SF/>1SF
60 90 Clinical
Failure: Not present,
restoration has
(almost) completely disappeared.
Secondary caries or periodontal
diseases appeared.
90 10 1yr
02 52 9 1 yr
Chen Xu 2001
[25] 03
A silver-amalgam made
in China primary molar
1SF 104 (52) Clinical (See Part 2) 21 7 2 yrs
04 106 106 106 22 1 yr
Ling Li 2003
[26] 05
Silver amalgam by China
Iron & Steel Research
Institute Group primary molar
(1SF/
>1SF) 106 106 Clinical ART/Mod.3 106 33 2 yrs
Ye Xiao-ya 2006
[27] 06 Silver amalgam primary molar
1SF/>1SF
68 174 Clinical ART /Mod.6 169 26 1 yr
07 120 240 235 24 1 yr
08 120 240 224 52 2 yrs
Chen Bao-
xing 2006
[28]
09
primary 1SF/>1SF
120 240
Clinical ART/Mod.1
218 58 3 yrs
10 64 190 168 18 1 yr Qiu Hong-
liang 2007
[29] 11 primary 1SF/>1SF
64 190 Clinical ART/Mod.4 142 16 2 yrs
Wang Mi-la 2005
[30] 12 Central Iron & Steel
research Institute,
Beijing, China primary molar
(1SF/
>1SF) 98 98 Clinical
Failure:
Not present, restoration has
partly or completely disappeared.
Marginal defect, pulp s
ymptoms or
secondary caries appeared.
98 15 18 mo
13 137 151 121 13 1 yr
Li Xue 2002
[32] 14 primary 1SF/>1SF
137 151 Clinical ART/Mod.5 105 15 2 yrs
Huang Gui-
qing 2009
[34] 15 Shanghai Dental Material
Factory, China primary molar
>1SF 90 90 Clinical ART/Mod.7 90 33 1 yr
Liu Hai-bo 2010
[36] 16 primary 1SF/>1SF
42 104 Clinical
Failure:
Not present, restoration has
(almost) completely disappeared.
Secondary caries or pulp symptoms
appeared.
104 4 1 yr
17 153 33 1 yr
Wang Quan 2004
[37] 18
Silver amalgam by
Shanghai Medical
Instruments Co Ltd. primary molar
1SF/>1SF
100 153 Clinical (See Part 2) 142 47 2 yrs
19 117 15 1 yr
20 112 28 2 yrs
Lin Bao-bao
2012
[39]
21
(?) 1SF 60 120 Clinical Failure:
Not present, restoration has
completely disappeared. Secondary
caries appeared or not. 109 29 3 ys
22 116 16 1 yr
23
1SF
(Class I) 116 116 31 2 yrs
Hu Qun-li 2005
[40]
24
Central Iron & Steel
Research Institute,
Beijing, China
primary molar
>1SF 116
Clinical ART/Mod.8
116 20 1 yr
Journal of Minimum Intervention in Dentistry 2012; 5: 377-415 © Midentistry cc - ISSN 1998-801X
Available online at www.jmid.org
406
25 (Class II)
116 33 2 yrs
Mo Qing-bo 2007
[41] 26 Antai Co. Ltd. Beijing,
China primary molar
(1SF/
>1SF) 185 Clinical ART/Mod.9 182 34 1 yr
27 181 18 1 yr
28 1SF 178 24 2 yrs
29 262 44 1yr
Zeng Hong-
yu 2006
[42]
30
Xinya Co. Ltd.
Hangzhou, China primary molar
>1SF 223 (445) Clinical
SEM
LM ART/Mod.6
257 60 2 yrs
31 82 15 1 yr Weng Shu-
ping 2005
[44] 32 Antai Co. Ltd. Beijing,
China primary molar
(1SF/
>1SF) 82 82 Clinical ART/Mod.9 82 22 2 yrs
33 1SF
(Class I) 31 28 0 1 yr
Zhang Xiu-li
2001
[45] 34 primary >1SF
(Class II)
72 Clinical ART/Mod.6 62 4 1 yr
35 1SF
(Class I) 52 16 2 yrs
Li Hui-min 2005
[24] 36
Silver amalgam by
Shanghai Medical
Instruments Co Ltd. permanent
>1SF
(Class II)
(?) Clinical (See Part 2) 78 32 2 yrs
DS = Dataset; PG = Parallel group; SM = Splitmouth; m = Male; f = Female; ART = Atraumatic restorative treatment; AM = Amalgam; HEX = Hand excavation; SF = Tooth surface; X = Mean value; N = Number of
evaluated restorations; n = Number of teeth with failed restorations; BSL = Number of restorations at baseline; yr(s) = Year(s); mo = Months; Clin = Clinical examination; Indirect = Silicon impression/plaster assessment;
ARTC = ART criteria; USPHS - United States Public Health Service criteria; Ref. = Reference number; P = Number of patients at baseline; X-Ray = Radiological assessment; Criteria = see Table 10(B).
Table 9 (B): Evaluation of tooth restorations: ART- and modified ART criteria (Table 9)
Score ART criteria (ARTC)1 Modification 1
(ART/Mod.1) Modification 2
(ART/Mod.2) Modification 3
(ART/Mod.3) Modification 4
(ART/Mod.4) Modification 5
(ART/Mod.5) Modification 6
(ART/Mod.6)
0 Present, good. Same as ARTC
1 Present, slight marginal defect for
whatever reason, at any one place
which is less than 0.5 mm in depth
No repair is needed. Same as ARTC
Present, slight marginal
defect for whatever
reason, at any one place
which is less than 1.0
mm in depth
No repair is needed.
Same as ARTC Same as ARTC
2 Present, marginal defect for whatever
reason, at any one place which is
deeper than 0.5 mm but less than 1.0
mm in depth. Repair is needed.
Slight abrasion
of the
restoration
margin (<0.5
mm)
Slight impairment of
occlusal surface (<1
mm). No repair is
needed.
Slight abrasion of the
restoration margin (<0.5
mm)
Slight wear <0.5 mm
3 Present, gross defect of more than 1.0
mm in depth. Repair is needed.
Same as ARTC
Same as ART
Same as ARTC
Restoration and tooth
intact but with slight
marginal discoloration
Marginal damage on
restoration. Repair is needed. Marginal damage on
restoration. Repair is needed.
Journal of Minimum Intervention in Dentistry 2012; 5: 377-415 © Midentistry cc - ISSN 1998-801X
Available online at www.jmid.org
407
4 Not present, restoration has (almost)
completely disappeared. Treatment is
needed.
Abrasion of the
restoration
margin. Repair
is needed.
Present. Tooth
fractured. Present, wear and tear
gradually over larger
parts of the restoration,
which is deeper than 1.0
mm. Repair is needed.
Marginal wear on
restoration. Repair is needed. Large wear on restoration.
Repair is needed.
5 Not present, other restorative
treatment has been performed. Partly or
completely lost Partly or
completely lost Impairment of occlusal
surface (>1 mm). Repair
is needed.
Partly or completely lost Partly or completely lost
6 Not present, tooth has been extracted.
Not present,
other restorative
treatment has
been performed.
Not present,
other restorative
treatment has
been performed.
Partly or completely lost Not present, other restorative
treatment has been
performed.
Not present, other restorative
treatment has been
performed.
7 Present, wear and tear gradually over
larger parts of the restoration but is
less than 0.5 mm at the deepest point.
No repair is needed.
Lost contact Not present,
tooth has been
extracted.
Not present,
tooth has been
extracted.
Secondary caries Not present, tooth has been
extracted. Tooth loss
8 Present, wear and tear gradually over
larger parts of the restoration, which is
deeper than 0.5 mm. Repair is needed.
Lost contact Not present, other
restorative treatment has
been performed.
Lost contact Lost to follow up
9 Unable to diagnose.
Not present, tooth has
been extracted.
S Score: 0, 1, 7 0, 1, 2 0, 1, 2 0, 1 0-3 0, 1, 2 0, 1, 2
F Failure: 2, 3, 4, 8 3-7 3-7 2-7 4-9 3-7 3-7
Table 9 (B): Evaluation of tooth restorations: ART- and modified ART criteria (Table 9) – Part 2
Score ART criteria (ARTC)1 Modification 7 (ART/Mod.7) Modification 8 (ART/Mod.8) Modification 9 (ART/Mod.9)
0 Present, good. Same as ARTC Same as ARTC
1 Present, slight marginal defect for whatever reason, at any one
place which is less than 0.5 mm in depth
No repair is needed.
Same as ARTC Present, slight marginal defect for
whatever reason, at any one place,
which is less than 0.2 mm in depth.
Present, slight marginal defect. No
repair is needed.
2 Present, marginal defect for whatever reason, at any one place
which is deeper than 0.5 mm but less than 1.0 mm in depth.
Repair is needed.
Present, marginal defect for whatever
reason, at any one place which is deeper
than 0.5 mm
Present, marginal defect for whatever
reason, at any one place which is deeper
than 0.2 mm
Not present, restoration has (almost)
completely disappeared. Treatment
is needed.
3 Present, gross defect of more than 1.0 mm in depth. Repair is
needed. Not present, restoration has (almost)
completely disappeared. Treatment is
needed.
Not present, restoration has (almost)
completely disappeared. Not present, other restorative
treatment has been performed.
4 Not present, restoration has (almost) completely disappeared.
Treatment is needed. Not present, tooth has been extracted. Not present, tooth has been extracted Not present, tooth has been
extracted
5 Not present, other restorative treatment has been performed. Restoration needed Tooth fracture
6 Not present, tooth has been extracted.
7 Present, wear and tear gradually over larger parts of the
restoration but is less than 0.5 mm at the deepest point. No
repair is needed.
8 Present, wear and tear gradually over larger parts of the
restoration, which is deeper than 0.5 mm. Repair is needed.
Journal of Minimum Intervention in Dentistry 2012; 5: 377-415 © Midentistry cc - ISSN 1998-801X
Available online at www.jmid.org
408
9 Unable to diagnose.
S Score: 0, 1, 7 0, 1 0, 1 0, 1
F Failure: 2, 3, 4, 8 2, 3, 4 2-5 2-5
S = Restoration success; F = Restoration failure.
1As per: Frencken JE, Holmgren CJ. Atraumatic restorative treatment for dental caries. STI Book b.v. Nijmegen 1999, p. 58. [48]
Journal of Minimum Intervention in Dentistry 2012; 5: 377-415 © Midentistry cc - ISSN 1998-801X
Available online at www.jmid.org
409
Table 10 – Computed effect estimates per extracted dataset
Illustrative comparative
risk
1st Author Year
Ref.
DS RR 95%CI p-value
RD 95%CI p-value ART Amalgam OR 95%CI p-value
Wu Xue ping
2005
[46]
01 1.09 0.51 – 2.31
0.82 0.01 -0.08 – 0.10 0.82 Not statistically different
1.10
0.47 – 2.56
0.82
02 0.56 0.20 – 1.55
0.26 -0.08 -0.21 – 0.05 0.25 Not statistically different
0.51
0.16 – 1.64
0.26
Chen Xu 2001
[25]
03 0.43 0.13 – 1.44
0.17 -0.19 -0.44 – 0.06 0.14 Not statistically different
0.33
0.07 – 1.53
0.16
04 0.73 0.41 – 1.31
0.29 -0.06 -0.16 – 0.05 0.28 Not statistically different
0.68
0.33 – 1.38
0.28
Ling Li 2003
[26]
05 0.70 0.44 – 1.10
0.12 -0.09 -0.21 – 0.02 0.12 Not statistically different
0.61
0.33 – 1.14
0.12
Ye Xiao-ya 2006
[27]
06 1.08 0.67 - 1.75
0.74 0.01 -0.06 - 0.09 0.74 Not statistically different
1.10
0.62 - 1.95
0.74
07 1.02 0.60 – 1.74
0.95 0 -0.05 – 0.06 0.95 Not statistically different
1.02
0.56 – 1.85
0.95
08 1.06 0.76 – 1.47
0.74 0.01 -0.07 – 0.09 0.74 Not statistically different
1.08
0.70 – 1.66
0.74
Chen
Bao-xing 2006
[28]
09 1.00 0.73 – 1.36
0.98 0 -0.08 – 0.08 0.98 Not statistically different
1.00
0.65 – 1.52
0.98
10 0.51 0.24 – 1.08
0.08 -0.05 -0.11 – 0.01 0.07 Not statistically different
0.48
0.22 – 1.08
0.08 Qui Hong-
liang 2007
[29]
11 1.22 0.67 – 2.21
0.52 0.02 -0.05 – 0.10 0.52 Not statistically different
1.25
0.63 – 2.47
0.52
Wang Mi-la 2005
[30]
12 0.80 0.40 – 1.62
0.54 -0.03 -0.13 – 0.07 0.53 Not statistically different
0.77
0.34 – 1.75
0.54
13 0.63 0.28 – 1.43
0.27 -0.04 -0.11 – 0.03 0.27 Not statistically different
0.61
0.25 – 1.48
0.27
Li Xue 2002
[32]
14 1.31 0.72 – 2.38
0.39 0.04 -0.05 – 0.14 0.38 Not statistically different
1.38
0.67 – 2.82
0.38
Huang
Gui-qing 2009
[34]
15 0.85 0.56 – 1.28
0.43 -0.06 -0.19 – 0.08 0.43 Not statistically different
0.78
0.42 – 1.45
0.43
Liu Hai-bo 2010
[36]
16 0.75 0.17 – 3.27
0.70 -0.01 -0.06 – 0.04 0.70 Not statistically different
0.74
0.16 – 3.40
0.70
17 0.81 0.52 – 1.28
0.36 -0.04 -0.13 – 0.05 0.36 Not statistically different
0.77
0.44 – 1.35
0.36
Wang Quan 2004
[37]
18 0.94 0.68 – 1.31
0.73 -0.02 -0.12 – 0.09 0.73 Not statistically different
0.92
0.56 – 1.49
0.73
19 0.81 0.40 – 1.66
0.57 -0.02 -0.11 – 0.06 0.57 Not statistically different
0.79
0.35 – 1.78
0.58
20 0.96 0.61 – 1.53
0.88 -0.01 -0.12 – 0.10 0.88 Not statistically different
0.95
0.52 – 1.75
0.88
Lin Bao-bao 2012
[39]
21 1.00 0.65 – 1.57
0.97 0.002 -0.12 – 0.12 0.97 Not statistically different
1.01
0.56 – 0.97
0.97
22 0.94 0.49 – 1.81
0.85 -0.01 -0.10 – 0.08 0.85 Not statistically different
0.93
0.44 – 1.98
0.85
23 0.77 0.49 – 1.23
0.28 -0.06 -0.17 – 0.05 0.28 Not statistically different
0.72
0.39 – 1.32
0.28
24 1.10 0.94 – 1.90
0.73 0.02 -0.08 – 0.12 0.73 Not statistically different
1.12
0.58 – 2.19
0.73
Hu Qun-li 2005
[40]
25 0.91 0.60 – 1.39
0.66 -0.03 -0.14 – 0.09 0.66 Not statistically different
0.88
0.49 – 1.57
0.66
Mo Qing-bo 2007
[41]
26 0.35 0.19 – 0.66
0.0011* -0.12 -0.19 - -0.05 0.0004* 7 out of 100
19 out of
100 0.31
0.15 – 0.62
0.0009*
27 0.87 0.45 – 1.70
0.69 -0.01 -0.07 – 0.05 0.69 Not statistically different
0.86
0.41 – 1.80
0.69
28 1.91 1.21 – 3.02
0.0053**
0.12 0.04 – 0.21 0.0043** 26 out of
100 14 out of
100 2.23
1.28 – 3.90
0.0049**
29 1.17 0.82 – 1.68
0.38 0.03 -0.04 – 0.09 0.37 Not statistically different
1.22
0.79 – 1.88
0.38
Zeng
Hong-yu 2006
[42]
30 1.59 1.21 – 2.08
0.0007**
0.14 0.06 – 0.21 0.0004** 37 out of
100 23 out of
100 1.93
1.33 – 2.82
0.0006**
31 0.40 0.16 – 0.98
0.045* -0.11 -0.21 - -0.01 0.033* 7 out of 100
18 out of
100 0.35
0.13 – 0.96
0.04* Weng
Shu-ping 2005
[44]
32 0.59 0.32 – 1.09
0.09 -0.11 -0.23 – 0.01 0.08 Not statistically different
0.51
0.24 – 1.11
0.09
33 2.35 0.10 – 55.6
0.60 0.03 -0.05 – 0.11 0.49 Not statistically different
2.41
0.09 – 61.4
0.60
Zhang
Xiu-li 2001
[45]
34 4.96 1.77 – 13.9
0.0023**
0.26 0.11 – 0.40 0.0005** 32 out of
100 6 out of 100
6.82
2.11 – 22.1
0.0014**
35 0.90 0.50 – 1.63
0.74 -0.03 -0.20 – 0.14 0.74 Not statistically different
0.87
0.37 – 2.00
0.74
Li Hui-min 2005
[24]
36 1.25 0.88 – 1.77
0.20 0.10 -0.06 – 0.26 0.20 Not statistically different
1.52
0.80 – 2.90
0.20
DS = Dataset number; RR = Risk ratio; RD = Risk difference; CI = Confidence interval; OR = Odds ratio; n.e. = Not estimable; Ref. = Reference
number;
Illustrative comparative risk = Number of restoration failures in 100. * Difference statistically significant in favour of ART; ** Difference statistically
significant in favour of amalgam.
Journal of Minimum Intervention in Dentistry 2012; 5: 377-415 © Midentistry cc - ISSN 1998-801X
Available online at www.jmid.org
410
Table 11 - Summary of results: High-viscosity GIC/ART for carious tooth cavity restoration
High-viscosity GIC/ART compared to amalgam restorations placed with rotary instruments
Patients/population:
All patients with carious cavities of any class in primary and permanent teeth.
Intervention: ART with high-viscosity GIC
Comparison: Amalgam restorations placed using conventional rotary instruments in tooth cavities
Outcome: Restoration failure after >1 year
Outcomes
1. Permanent teeth
1.1. Single surface cavities
After 2years Not significantly different / 1 Dataset: DS 35
1.2. Multiple surface cavities
After 2years Not significantly different / 1 Dataset: DS 36
1.3. Unspecified type of cavities
After 1 yearNot significantly different / 1 Dataset: DS 01
2. Primary teeth
2.1. Single surface cavities
After 1 year Not significantly different / 4 Datasets: DS 02, 22, 27, 33
Not significantly different / 2 Datasets: DS 03, 23
After 2years ART: 26 failures out of 100 restorations compared to amalgam: 14 failures out of 100
restorations / 1 Dataset: DS 28
2.2. Multiple surface cavities
Not significantly different / 3 Datasets: DS 15, 24, 29
After 1 year ART: 32 failures out of 100 restorations compared to amalgam: 6 failures out of 100
restorations / 1 Dataset: DS 34
Not significantly different / 1 Dataset: DS 25
After 2years ART: 37 failures out of 100 restorations compared to amalgam: 23 failures out of 100
restorations / 1 Dataset: DS 30
2.3. Type of cavity not specified
Not significantly different / 7 Datasets: DS 04, 06, 07, 10, 13, 16, 17
ART: 7 failures out of 100 restorations compared to amalgam: 18 failures out of 100
restorations / 1 Dataset: DS 31
After 1 year ART: 7 failures out of 100 restorations compared to amalgam: 19 failures out of 100
restorations / 1 Dataset: DS 26
After 1.5 years Not significantly different / 1 Dataset: DS 12
After 2years Not significantly different / 6 Datasets: DS 05, 08, 11, 14, 18, 32
After 3 years Not significantly different / 1 Dataset: DS 09
3. No dentition specified
3.1. Single surface cavities
After 1 year Not significantly different / 1 Dataset: DS 19
After 2years Not significantly different / 1 Dataset: DS 20
After 3 years Not significantly different / 1 Dataset: DS 21
RR = Risk ratio; RD = Risk difference; CI = Confidence interval; DS = Dataset number;
GIC = Glass-ionomer cement.
Journal of Minimum Intervention in Dentistry 2012; 5: 377-415 © Midentistry cc - ISSN 1998-801X
Available online at www.jmid.org
411
Table 12 - Verbatim quotes relevant to selection-, performance- and detection bias risk
Ref. = Reference number.
Table 13 - Assessment of selection-, performance- and detection bias risk
[as per Table 9/study design and Table 12/verbatim quotes]
1st Author Year Ref. SB PB DB
Wu Xue-ping 2005 [46] 0 0 0
Chen Xu 2001 [25] 0 0 0
Ling Li 2003 [26] 0 0 0
Ye Xiao-ya 2006 [27] 0 0 0
Chen Bao-xing 2006 [28] 0 0 0
Qiu Hong-liang 2007 [29] 0 0 0
Wang Mi-la 2005 [30] 0 0 0
Li Xue 2002 [32] 0 0 0
Huang Gui-qing
2009 [34] 0 0 0
Liu Hai-bo 2010 [36] 0 0 0
Wang Quan 2004 [37] 0 0 0
Lin Bao-bao 2012 [39] 0 0 0
Hu Qun-li 2005 [40] 0 0 0
Mo Qing-bo 2007 [41] 0 0 0
Zeng Hong-yu 2006 [42] 0 0 0
Weng Shu-ping 2005 [44] 0 0 0
Zhang Xiu-li 2001 [45] 0 0 0
Li Hui-min 2005 [24] 0 0 0
Ref. = Reference number; SB = Selection bias risk; PB = Performance bias risk; DB = Detection bias risk
Selection bias Performance bias Detection bias
1st Author Year
Ref.
Random sequence generation
Concealment of the
sequence allocation Operator
blinding Patient
blinding Evaluator blinding
Wu Xue-
ping 2005
[46]
117 patients were randomly
divided into 2 groups.
Chen Xu 2001
[25]
Ling Li 2003
[26]
Randomisation is applied
Ye Xiao-ya 2006
[27]
Chen Bao-
xing 2006
[28]
Randomly, … was filled
Qiu Hong-
liang 2007
[29]
Wang Mi-la 2005
[30]
Randomized splitmouth-
control
Li Xue 2002
[32]
Groups .. divided randomly
Huang Gui-
qing 2009
[34]
Randomized splitmouth-
control Evaluation was performed by 1
dentist other than the operator.
Liu Hai-bo 2010
[36]
Children were randomly
divided into 2 groups.
Wang Quan 2004
[37]
Lin Bao-bao
2012
[39]
Children were randomly
divided into 2 groups.
Hu Qun-li 2005
[40]
Mo Qing-bo
2007
[41]
Randomized splitmouth-
control Evaluation was performed by the
operator.
Zeng Hong-
yu 2006
[42]
Weng Shu-
ping 2005
[44]
Randomized splitmouth-
control Evaluation was performed by the
operator.
Zhang Xiu-li
2001
[45]
Children were randomly
divided into 3 groups: ART,
traditional GIC or traditional
Amalgam.
Li Hui-min 2005
[24]
Journal of Minimum Intervention in Dentistry 2012; 5: 377-415 © Midentistry cc - ISSN 1998-801X
Available online at www.jmid.org
412
Table 14 – Part 1: Quantitative assessment of potential attrition bias risk: Worst-case scenario
Worst-case scenario LTF adjusted effect estimate Original effect estimate
(Table 11)
ART group Amalgam group
1st
Author
Year
Ref.
DS
LTF
N = BSL
teeth n+LTF
LTF N+LTF
n RR
95% CI p-value RR 95% CI p-value
Bias
risk
Wu
Xue-
ping 2005
[46]
01 No LTF reported No
02 No LTF reported No Chen
Xu 2001
[25]
03 31 52 34 31 52 7 4.86
2.37 – 9.95 <0.0001**
0.43 0.13 – 1.44 0.17 Yes
04 No LTF reported No Ling
Li 2003
[26]
05 No LTF reported No
Ye
Xiao-
ya 2006
[27]
06 6 186 36 5 174 26 1.30
0.82 - 2.05 0.27 1.08 0.67 - 1.75 0.74 No
07 9 240 33 5 240 24 1.38
0.84 – 2.25 0.21 1.02 0.60 – 1.74 0.95 No
08 15 240 70 16 240 52 1.35
0.99 – 1.84 0.06 1.06 0.76 – 1.47 0.74 No
Chen
Bao-
xing 2006
[28]
09 25 240 82 22 240 58 1.41
1.06 – 1.88 0.02** 1.00 0.73 – 1.36 0.98 Yes
10 16 198 26 22 190 18 1.39
0.79 – 2.44 0.26 0.51 0.24 – 1.08 0.08 No Qiu
Hong-
liang
2007
[29]
11 30 198 53 48 190 16 3.18
1.89 – 5.36 <0.0001**
1.22 0.67 – 2.21 0.52 Yes
Wang
Mi-la
2005
[30]
12 No LTF reported No
13 22 154 31 30 151 13 2.34
1.27 – 4.29 0.006** 0.63 0.28 – 1.43 0.27 Yes
Li Xue
2002
[32]
14 36 154 58 46 151 15 3.79
2.25 – 6.39 <0.0001**
1.31 0.72 – 2.38 0.39 Yes
Huang
Gui-
qing 2009
[34]
15 No LTF reported No
Liu
Hai-bo
2010
[36]
16 No LTF reported No
17 5 165 33 0 153 33 0.93
0.60 – 1.42 0.73 0.81 0.52 – 1.28 0.36 No
Wang
Quan
2004
[37]
18 8 165 57 11 153 47 1.12
0.82 – 1.54 0.47 0.94 0.68 – 1.31 0.73 No
19 5 120 17 3 120 15 1.13
0.59 – 2.16 0.70 0.81 0.40 – 1.66 0.57 No
20 8 120 35 8 120 28 1.25
0.82 – 1.92 0.31 0.96 0.61 – 1.53 0.88 No
Lin
Bao-
bao 2012
[39]
21 12 120 41 11 120 29 1.41
0.95 – 2.11 0.09 1.00 0.65 – 1.57 0.97 No
22 No
23 No
24 No
Hu
Qun-li
2005
[40]
25
No LTF reported
No
Mo
Qing
2007
[41]
26 3 185 15 3 185 34 0.44
0.25 – 0.78 0.005* 0.35 0.19 – 0.66 0.001*
No
27 Yes
28 Yes
29 Yes
Zeng
Hong-
yu 2006
[42]
30
No BSL per DS reported
Yes
31 No
Weng
Shu-
ping 2005
[44]
32 No LTF reported No
33 2 38 3 3 31 0 5.74
0.31 – 107.1 0.24 2.35 0.10 – 55.6 0.60 No
Zhang
Xiu-li
2001
[45]
34 3 53 19 10 72 4 6.45
2.33 – 17.9 0.0003**
4.96 1.77 – 13.9 0.002**
No
35 Yes
Li
Hui-
min 2005
[24]
36 No BSL data reported Yes
.
Journal of Minimum Intervention in Dentistry 2012; 5: 377-415 © Midentistry cc - ISSN 1998-801X
Available online at www.jmid.org
413
Table 14 – Part 2: Quantitative assessment of potential attrition bias risk: Best-case scenario
Worst-case scenario LTF adjusted effect estimate Original effect estimate
(Table 11)
ART group Amalgam group
1st
Author
Year
Ref.
DS
LTF
N = BSL
teeth n+LTF
LTF N+LTF
n RR
95% CI p-value RR 95% CI p-value
Bias
risk
Wu
Xue-
ping 2005
[46]
01 No LTF reported No
02 No LTF reported No Chen
Xu 2001
[25]
03 31 52 3 31 52 38 0.08
0.03 – 0.24 <0.0001*
0.43
0.13 – 1.44 0.17 Yes
04 No LTF reported No Ling
Li 2003
[26]
05 No LTF reported No
Ye
Xiao-
ya 2006
[27]
06 6 186 30 5 174 31 0.91
0.57 - 1.43 0.67 1.08
0.67 - 1.75 0.74 No
07 9 240 24 5 240 29 0.83
0.50 – 1.38 0.47 1.02
0.60 – 1.74 0.95 No
08 15 240 55 16 240 68 0.81
0.59 – 1.10 0.18 1.06
0.76 – 1.47 0.74 No
Chen
Bao-
xing 2006
[28]
09 25 240 57 22 240 80 0.71
0.53 – 0.95 0.02* 1.00
0.73 – 1.36 0.98 Yes
10 16 198 10 22 190 40 0.24
0.12 – 0.47 <0.0001*
0.51
0.24 – 1.08 0.08 Yes
Qiu
Hong-
liang
2007
[29]
11 30 198 23 48 190 64 0.34
0.22 – 0.53 <0.0001*
1.22
0.67 – 2.21 0.52 Yes
Wang
Mi-la
2005
[30]
12 No LTF reported No
13 22 154 9 30 151 43 0.21
0.10 – 0.41 <0.00001*
0.63
0.28 – 1.43 0.27 Yes
Li Xue
2002
[32]
14 36 154 22 46 151 61 0.35
0.23 – 0.55 <0.00001*
1.31
0.72 – 2.38 0.39 Yes
Huang
Gui-
qing 2009
[34]
15 No LTF reported No
Liu
Hai-bo
2010
[36]
16 No LTF reported No
17 5 165 28 0 153 33 0.79
0.50 – 1.24 0.30 0.81
0.52 – 1.28 0.36 No
Wang
Quan
2004
[37]
18 8 165 49 11 153 58 0.78
0.57 – 1.07 0.12 0.94
0.68 – 1.31 0.73 No
19 5 120 12 3 120 18 0.67
0.34 – 1.32 0.25 0.81
0.40 – 1.66 0.57 No
20 8 120 27 8 120 36 0.75
0.49 – 1.15 0.19 0.96
0.61 – 1.53 0.88 No
Lin
Bao-
bao 2012
[39]
21 12 120 29 11 120 40 0.73
0.48 – 1.09 0.12 1.00
0.65 – 1.57 0.97 No
22 No
23 No
24 No
Hu
Qun-li
2005
[40]
25
No LTF reported
No
Mo
Qing
2007
[41]
26 3 185 12 3 185 37 0.32
0.17 – 0.60 0.0004* 0.35 0.19 – 0.66
0.0011* No
27 Yes
28 Yes
29 Yes
Zeng
Hong-
yu 2006
[42]
30
No BSL per DS reported
Yes
31 No
Weng
Shu-
ping 2005
[44]
32 No LTF reported No
33 2 38 1 3 31 3 0.27
0.03 – 2.49 0.25 2.35 0.10 – 55.6
0.60 No
Zhang
Xiu-li
2001
[45]
34 3 53 16 10 72 14 1.55
0.83 – 2.90 0.17 4.96 1.77 – 13.9
0.0023** Yes
35 Yes
Li
Hui-
min 2005
[24]
36 No BSL data reported Yes
LTF = Number of restored teeth lost to follow-up; BSL = Baseline number of restored teeth; n = Number of failed restorations;
N = Number of evaluated restorations; RR = Risk ratio; CI = Confidence interval; DS = Dataset number. *Difference statistically significant in favour of
ART; Ref. = Reference number; **Difference statistically significant in favour of amalgam.
Journal of Minimum Intervention in Dentistry 2012; 5: 377-415 © Midentistry cc - ISSN 1998-801X
Available online at www.jmid.org
414
Table 15 – Assessment of reporting bias risk
Article
Ref.nr. Trial focus extracted from Methods and Results section
[46]
[a] Methods/stated objectives
Clinical evaluation up to 1 year
Assessment of secondary caries
[b] Reported results
Clinical evaluation up to 1 year was reported in table 1
Assessment of secondary caries was reported in table 2
[25]
[a] Methods/Stated objectives
Evaluation of restoration longevity
[b] Reported results
Evaluation of restoration longevity reported
[26]
[a] Methods/Stated objectives
Evaluation of restoration longevity and patient cooperation
[b] Reported results
Evaluation of restoration longevity reported;
Evaluation of patient cooperation reported.
[27]
[a] Methods/Stated objectives
Evaluation of restoration longevity, patient cooperation and caries;
[b] Reported results
Evaluation of restoration longevity reported;
Evaluation of patient cooperation reported;
Evaluation of caries reported.
[28]
[a] Methods/Stated objectives
Evaluation up to 36 months;
Evaluation of restoration longevity, patient cooperation and caries;
[b] Reported results
Evaluation of restoration longevity after 12, 24 and 36 months reported in Table 1 and 2;
Evaluation of patient cooperation reported;
Evaluation of caries not reported.
[29]
[a] Methods/Stated objectives
Evaluation of restoration longevity, patient cooperation and caries;
[b] Reported results
Evaluation of restoration longevity reported;
Evaluation of patient cooperation reported;
Evaluation of caries reported.
[30]
[a] Methods/stated objectives
Clinical evaluation of 18 months.
Evaluation of children’s cooperation.
Assessment of secondary caries
[b] Reported results
The success rate of 18 months.
Evaluation of children’s cooperation
Assessment of secondary caries.
[32]
[a] Methods/Stated objectives
Evaluation of restoration longevity, patient cooperation and caries;
[b] Reported results
Evaluation of restoration longevity reported;
Evaluation of patient cooperation reported;
Evaluation of caries reported.
[34]
[a] Methods/stated objectives
Clinical evaluation up to 1 year.
Intensity of pain and the willing of undergo the treatment again.
Assessment of secondary caries
[b] Reported results
The success rate up to 1 year was reported in table 1.
Intensity of pain and the willing of undergo the treatment again.
Assessment of secondary caries.
[36]
[a] Methods/stated objectives
The success rate of 1 year
Evaluation of children’s cooperation
[b] Reported results
The success rate of 1 year
Evaluation of children’s cooperation
Journal of Minimum Intervention in Dentistry 2012; 5: 377-415 © Midentistry cc - ISSN 1998-801X
Available online at www.jmid.org
415
[37]
[a] Methods/Stated objectives
Evaluation of restoration longevity and patient cooperation;
[b] Reported results
Evaluation of restoration longevity reported;
Evaluation of patient cooperation reported.
[39]
[a] Methods/stated objectives
The success rate up to 3 years
Evaluation of children’s cooperation
[b] Reported results
The success rate up to 3 years was reported in the table.
Evaluation of children’s cooperation was reported in the same table.
[40]
[a] Methods/stated objectives
Clinical evaluation of 1 year
[b] Reported results
Success rate of 1 year was reported in the table.
[41]
[a] Methods/stated objectives
Clinical evaluation of 1 year.
[b] Reported results
Success rate of 1 year.
[42]
[a] Methods/stated objectives
The success rate of 1 and 2 years.
[b] Reported results
Success rate reported in table 1.
The rate of secondary caries was reported in table 2.
[44]
[a] Methods/stated objectives
Clinical evaluation of 0.5 year, 1 year and 2 years.
[b] Reported results
Success rate of 0.5 year, 1 year and 2 years were reported.
[45]
[a] Methods/stated objectives
Clinical evaluation of 12 months
Comparing of ART and traditional technique with amalgam in class and cavities in primary teeth.
[b] Reported results
Success rate reported in table 2 and 3
Children’s satisfaction was reported in table 4
[24]
[a] Methods/Stated objectives
Evaluation of restoration longevity;
[b] Reported results
Evaluation of restoration longevity reported.
Table 16 – Assessment of research gaps, concerning precision and consistency in the current results
Imprecision Inconsistency
[a] [b] [c] [d] [e] [f] [g]
1. Permanent teeth
1.1. Single surface cavities
After 2 years
x x x
1.2. Multiple surface cavities
After 2 years
x x x
2. Primary teeth
2.1. Single surface cavities After 1 year x x x
After 2 years
x x x x x
2.2. Multiple surface cavities
After 1 year x x x x x
After 2 years
x x x x
[a] Wide confidence intervals; [b] Lack of trials/data; [c] Small sample size;
[d] Conflicting direction of effect sizes; [e] Conflicting magnitude of difference in effect sizes;
[f] Less than two trials/datasets; [g] Non-overlapping confidence intervals.
*Results from unspecified cavity types and dentition not included.
... During a systematic review of clinical controlled trials, the survival rate of HVGIC restorations, placed using the atraumatic restorative treatment approach, in permanent posterior teeth in comparison to conventionally placed silver amalgam has been established [6]. This systematic review was further updated [7] and the detailed results, including additional results from Chinese trials, published by the authors [8,9]. All published reports of this systematic review indicated no differences between HVGIC and amalgam, beyond the play of chance (p > 0.05) in the permanent dentition after four and six years for single and multiple surface tooth restorations, respectively, and no differences after three years for single and multiple surface restorations in primary teeth [6,7,10]. ...
... All published reports of this systematic review indicated no differences between HVGIC and amalgam, beyond the play of chance (p > 0.05) in the permanent dentition after four and six years for single and multiple surface tooth restorations, respectively, and no differences after three years for single and multiple surface restorations in primary teeth [6,7,10]. The results from Chinese trials were confirmatory of these findings [8]. ...
... After the conduct and reporting of systematic review [8,9] and meta-epidemiological study [1,10] results regarding the clinical efficacy of HVGICs versus amalgam as the current restorative gold standard, an integrative analysis of all evidence combined to the topic has still been missing. For this reason, the aim of this study was to present a synthesis of previously established evidence within a Bayesian framework, in order to derive an overall conclusion about the Odds in regard to the hypothesis that HVGIC's are inferior to silver amalgam as restorative materials for permanent posterior teeth. ...
Article
Full-text available
Background To develop a synthesis within a Bayesian probability framework of previously established evidence, in order to derive an overall conclusion about the hypothesis (H1): ‘High-viscosity glass-ionomer cements (HVGIC) are inferior to silver amalgam as (load bearing) restorative materials for permanent posterior teeth’. Methods Following Bayesian method, the prior Odds that H1 is true (established from past uncontrolled clinical longitudinal and laboratory trials), the Likelihood Ratio incorporating new evidence (established from recent meta-epidemiological studies and systematic reviews of controlled clinical trials), as well as the posterior hypothesis Odds in view of the new evidence, were calculated. Results The prior Odds that HVGICs are clinically inferior to amalgam as restorative materials in posterior permanent teeth in relation to the hypothesis that this is not so was 1.12 to 1. The Likelihood Ratio based on new evidence in favor the hypothesis was zero and the subsequent posterior Odds 0 to 1. Therefore, based on the new evidence, the Odds that HVGICs are clinically inferior to amalgam as restorative materials in posterior permanent teeth degreased from 1.12 to zero. Conclusion The current evidence suggests lack of support for the hypothesis that high-viscosity glass-ionomer cements are inferior to silver amalgam as restorative materials for permanent posterior teeth. Should future research to this topic uphold the current findings, a wider range of clinical benefits for both patient and care provider, beyond appropriate restoration longevity for placing HVGIC based restorations may apply.
... Analysis of records shows the survival rate of ART restorations ranging from 24.4% to 96.5% in primary teeth and 43.4-98.4% in permanent teeth, and as the evaluation time increased, the success rates have decreased, in accordance to the reported findings of Frencken et al. [8] Higher survival rates for permanent teeth compared to primary dentition were observed in our analysis, in accordance to the reported findings of Van't Hof et al., and Mickenautsch and Yengopal, [9,10] [11,12] Survival rates of ART restorations progressively decreased with an increase in the evaluation time, reporting as much as 100% survival for studies with an evaluation time of <2 years, and 90% for 2-5 years, and 80% for an evaluation time for more than 5 years. This trend has been universally observed in various systematic reviews. ...
... This trend has been universally observed in various systematic reviews. [10,11] Differences exist among the dentists regarding the applicability of ART as a treatment modality for both primary and permanent dentition. Many dentists are reluctant to use the ART in daily practice due to a conception that it might not be effective or successful. ...
... Though amalgam is considered to be a universal posterior restorative material and a gold standard even then it has showed similar survival rates as ART especially so in the primary dentition for an evaluation time of 2-3 years. These findings are in accordance to the ones reported by Frencken et al., Mickenautsch and Yengopal,Raggio et al.,and Studart et al. [7,8,10,11] GIC is always the material of choice in the ART technique. This is because of its chemical adherence to dental tissue, coefficient of thermal expansion similar to that of a tooth, biocompatibility properties, and caries protective effect through the release of fluoride, which has antibacterial properties, and potentiates remineralization that may prevent the development of secondary caries. ...
Article
Full-text available
Background: Atraumatic restorative treatment (ART) approach was developed in the mid-1980s to provide dental preventive and restorative care to underserved populations in areas that are out of reach. This method consists of the removal of infected (soft demineralized carious) tooth tissue with hand instruments only followed by restoring the cleaned cavity and associated pits and fissures with an adhesive restorative material. However, survival of ART restoration is a matter of intense speculation. Aim: To conduct a systematic review of the literature on the survival rate of ART restorations. Materials and Methods: A systematic search was carried out in MEDLINE, PubMed, and Cochrane databases from 1998 to 2014 using the term survival of ART in permanent teeth, primary teeth, ART restorative material. Results: The analysis of abstracts led to the selection of 36 studies. The research reports a survival rate of ART restoration for 5 years ranging from 44.8% to 100%, 24.4% to 90%, and 72.6% to 86%. Survival of ART restoration is reported to be marginally higher in permanent molars compared to their primary counterparts. Conclusion: This systematic review revealed high survival rates for single-surface ART restorations in both primary and permanent molars. However, long-term clinical studies are scarce, and a level two and level three evidence was observed.
... Nevertheless, these shortcomings do not provide evidence in support of the recommendation that HVGIC are not suitable for use as permanent posterior tooth restoration materials[14], which only direct comparisons within RCTs can provide. In this context it is interesting to note that despite a broad systematic literature appraisal no summary RCT evidence could be established in support of the hypothesis that ''direct HVGIC restorations are inferior to those of amalgam in posterior cavities of permanent teeth''[34,50]. It is recommended that any guidance for clinical practice should be based on direct comparisons from randomised control trials, ideally appraised during systematic reviews of the clinical literature. Where RCT evidence has not as yet been established in clinical fields, clinical guidance should at least avoid recommendations based on flawed data comparisons (i.e. ...
Article
Full-text available
Background: Naïve-indirect comparisons are comparisons between competing clinical interventions' evidence from separate (uncontrolled) trials. Direct comparisons are comparisons within randomised control trials (RCTs). The objective of this empirical study is to test the null-hypothesis that trends and performance differences inferred from naïve-indirect comparisons and from direct comparisons/RCTs regarding the failure rates of amalgam and direct high-viscosity glass-ionomer cement (HVGIC) restorations in permanent posterior teeth have similar direction and magnitude. Methods: A total of 896 citations were identified through systematic literature search. From these, ten and two uncontrolled clinical longitudinal studies for HVGIC and amalgam, respectively, were included for naïve-indirect comparison and could be matched with three out twenty RCTs. Summary effects sizes were computed as Odds ratios (OR; 95% Confidence intervals) and compared with those from RCTs. Trend directions were inferred from 95% Confidence interval overlaps and direction of point estimates; magnitudes of performance differences were inferred from the median point estimates (OR) with 25% and 75% percentile range, for both types of comparison. Mann-Whitney U test was applied to test for statistically significant differences between point estimates of both comparison types. Results: Trends and performance differences inferred from naïve-indirect comparison based on evidence from uncontrolled clinical longitudinal studies and from direct comparisons based on RCT evidence are not the same. The distributions of the point estimates differed significantly for both comparison types (Mann-Whitney U = 25, n(indirect) = 26; n(direct) = 8; p = 0.0013, two-tailed). Conclusion: The null-hypothesis was rejected. Trends and performance differences inferred from either comparison between HVGIC and amalgam restorations failure rates in permanent posterior teeth are not the same. It is recommended that clinical practice guidance regarding HVGICs should rest on direct comparisons via RCTs and not on naïve-indirect comparisons based on uncontrolled longitudinal studies in order to avoid inflation of effect estimates.
Article
Full-text available
Purpose Traditionally, resin composite restorations are claimed by reviews of the dental literature as being superior to glass-ionomer fillings in terms of restoration failures in posterior permanent teeth. The aim of this systematic review is to answer the clinical question, whether conventional high-viscosity glass-ionomer restorations, in patients with single and/or multi-surface cavities in posterior permanent teeth, have indeed a higher failure rate than direct hybrid resin composite restorations. Methods Eight databases were searched until December 02, 2013. Trials were assessed for bias risks, in-between datasets heterogeneity and statistical sample size power. Effects sizes were computed and statistically compared. A total of 55 citations were identified through systematic literature search. From these, 46 were excluded. No trials related to high-viscosity glass-ionomers versus resin composite restorations for direct head-to-head comparison were found. Three trials related to high-viscosity glass-ionomers versus amalgam and three trials related to resin composite versus amalgam restorations could be included for adjusted indirect comparison, only. Results The available evidence suggests no difference in the failure rates between both types of restoration beyond the play of chance, is limited by lack of head-to-head comparisons and an insufficient number of trials, as well as by high bias and in-between-dataset heterogeneity risk. The current clinical evidence needs to be regarded as too poor in order to justify superiority claims regarding the failure rates of both restoration types. Sufficiently large-sized, parallel-group, randomised control trials with high internal validity are needed, in order to justify any clinically meaningful judgment to this topic.
Article
The aim of this study was to examine the presence and extent of publication bias and small-study effects in meta-analyses (MAs) investigating pediatric dentistry-related subjects. Following a literature search, 46 MAs including 882 studies were analyzed qualitatively. Of these, 39 provided enough data to be re-analyzed. Publication bias was assessed with the following methods: contour-enhanced funnel plots, Begg and Mazumdar's rank correlation and Egger's linear regression tests, Rosenthal's failsafe N, and Duval and Tweedie's "trim and fill" procedure. Only a few MAs adequately assessed the existence and effect of publication bias. Inspection of the funnel plots indicated asymmetry, which was confirmed by Begg-Mazumdar's test in 18% and by Egger's test in 33% of the MAs. According to Rosenthal's criterion, 80% of the MAs were robust, while adjusted effects with unpublished studies differed from little to great from the unadjusted ones. Pooling of the Egger's intercepts indicated that evidence of asymmetry was found in the pediatric dental literature, which was accentuated in dental journals and in diagnostic MAs. Since indications of small-study effects and publication bias in pediatric dentistry were found, the influence of small or missing trials on estimated treatment effects should be routinely assessed in future MAs. Copyright © 2015 Elsevier Inc. All rights reserved.
Article
Full-text available
This protocol has been registered with the International Prospective Register for Systematic Reviews (PROSPERO) on the 10 July 2012 under registration number CRD42012002621 (Available online from http://www.crd.york.ac.uk/prospero/display_record.asp?ID=CRD42012002621). This protocol comprises an addition of an existing systematic review report: Mickenautsch S, Yengopal V. Failure rate of atraumatic restorative treatment using high-viscosity glass-ionomer cement compared to conventional amalgam restorative treatment in primary and permanent teeth: a systematic review update -III. J Minim Interv Dent 2012; 5: 273-331 and provides a quantitative systematic review of the Chinese literature to the topic.
Article
Full-text available
The Medline, Cochrane CENTRAL, Biomed Central, Database of Open Access Journals (DOAJ), OpenJ-Gate, Bibliografia Brasileira de Odontologia (BBO), LILACS, IndMed, Sabinet, Scielo, Scirus (Medicine), OpenSIGLE and Google Scholar databases were searched. Hand searching was performed for journals not indexed in the databases. References of included trials were checked. Prospective clinical trials with test and control groups with a follow up of at least one year were included. Data abstraction was conducted independently and clinical and methodologically homogeneous data were pooled using a fixed-effects model. Eighteen trials were included. From these 32 individual dichotomous datasets were extracted and analysed. The majority of the results show no differences between both types of intervention. A high risk of selection-, performance-, detection- and attrition bias was identified. Existing research gaps are mainly due to lack of trials and small sample size. The current evidence indicates that the failure rate of high-viscosity GIC/ART restorations is not higher than, but similar to that of conventional amalgam fillings after periods longer than one year. These results are in line with the conclusions drawn during the original systematic review. There is a high risk that these results are affected by bias, and thus confirmation by further trials with suitably high numbers of participants is needed.