| Sign In to gain access to subscriptions and/or personal tools. |
The Effectiveness of Sealants in Managing Caries Lesions
1 Centers for Disease Control and Prevention/Division of Oral Health/Surveillance, Investigations, and Research Branch, 4770 Buford Highway, MSF10, Chamblee, GA 30341, USA; Correspondence: * corresponding author, sig1{at}cdc.gov
A barrier to providing sealants is concern about inadvertently sealing over caries. This meta-analysis examined the effectiveness of sealants in preventing caries progression. We searched electronic databases for comparative studies examining caries progression in sealed permanent teeth. We used a random-effects model to estimate percentage reduction in the probability of caries progression in sealed vs. unsealed carious teeth. Six studies, including 4 randomized-controlled trials (RCT) judged to be of fair quality, were included in the analysis (384 persons, 840 teeth, and 1090 surfaces). The median annual percentage of non-cavitated lesions progressing was 2.6% for sealed and 12.6% for unsealed carious teeth. The summary prevented fraction for RCT was 71.3% (95%CI: 52.8%–82.5, no heterogeneity) up to 5 years after placement. Despite variation among studies in design and conduct, sensitivity analysis found the effect to be consistent in size and direction. Sealing non-cavitated caries in permanent teeth is effective in reducing caries progression.
Key Words: pit and fissure sealants • caries
There is strong evidence that sealants are effective in both clinical and school settings for preventing caries in children at various levels of risk (Truman et al., 2002; Ahovuo-Saloranta et al., 2004). The evidence for sealant effectiveness in the management of dental caries is limited, however. One review that examined the effectiveness of interventions to manage caries for the National Institutes of Health (NIH) Caries Consensus Development Conference included only 1 study on sealants (Bader et al., 2001). Despite the strong evidence of primary effectiveness, sealant prevalence among lower-income children (who are at higher risk for dental caries) is about 30% (Dye et al., 2007), well below the Healthy People 2010 objective of 50%. Analysis of survey data from dentists suggests that one barrier to providing sealants is concern about inadvertently sealing over caries (Chapko, 1987; Primosch and Barr, 2001). This concern has also been a barrier to implementing school-based sealant programs (Association of State and Territorial Dental Directors, unpublished data, 2005). Documenting the effectiveness of sealants in the management of existing caries is therefore important, and such documentation could potentially remove barriers to the provision of a proven intervention. The purpose of this meta-analysis is to examine the effectiveness of dental sealants in preventing the progression of caries lesions in the pits and fissures of permanent teeth.
Inclusion Criteria This analysis was part of a broader systematic review of sealant effectiveness in the management of caries in the permanent dentition. Initially, we included all in vivo studies published in English that compared caries progression or bacteria levels in permanent teeth that did and did not receive sealants. Comparisons could be concurrent or measured over time (time-series or before-after) in the same groups. In the current meta-analysis, study designs were limited to randomized and non-randomized controlled trials and cohort studies that provided concurrent comparisons of % of lesions progressing. There were no restrictions on study populations.
Identification of Studies
Study Selection
Data Abstraction and Quality Assessment
Outcome and Effect Measures
and its 95% confidence interval (CI). One can obtain the prevented fraction by subtracting the RR from 1, and the upper/lower 95%CI by subtracting the lower/higher 95%CI of the RR ratio from 1.
Synthesis of Findings In adjusting the data for differences in study design, multiple observations per subject, and 100% or 0% progression rates (LaPlace adjustment), we made conservative assumptions that would bias the results toward finding no statistical significance (APPENDIX). We used the Der Simonian and Laird (DSL) random-effects model (Stijnen, 1999) to obtain the summary RR and its 95% confidence interval. We tested for homogeneity of effect size using the quantity I2 (Higgins et al., 2003). Finally, we conducted sensitivity analysis to determine how robust our findings were to excluding cohort studies and assuming higher values of intra-oral correlation (APPENDIX).
Characteristics of Studies The 6 studies included in this analysis (representing an estimated 384 persons, 840 teeth, and 1090 surfaces) varied in design (4 RCTs), baseline caries classifications, and types of sealant material (Table 1  ). Four studies primarily sealed non-cavitated lesions, 1 exclusively sealed cavitated lesions, and 1 sealed both cavitated and non-cavitated lesions. Three studies used 2nd- or 3rd-generation resin-based sealants, 2 used glass-ionomer cement (GIC), and 1 used 1st-generation resin-based sealants. Study populations included children, adolescents, and young adults ranging in age from 6 to 19 yrs.
Quality of Studies All the studies were rated as "fair" quality (Table 2 ). It is likely that comparable groups were assembled in 5 studies—4 RCTs and 1 cohort study where baseline sealant prevalence and DFS did not differ between the sealed and not-sealed groups. All studies clearly defined the intervention. The 2 cohort studies did not report drop-out rates, and 1 RCT of split-mouth design reported a one-year drop-out rate of 30%. In the 3 remaining RCTs, however, the one-year drop-out rate was less than 10% (this included the 2 larger studies that supported subgroup analyses of sealed caries lesions).
In the absence of sealant removal prior to follow-up examination, we assumed that outcome assessment was not blinded. In only 1 RCT, however, were sealants removed prior to the follow-up examination, with teeth assessed by an examiner who did not know the initial group assignment. In 2 of the remaining 5 studies (1 RCT and 1 cohort study), however, either the examiner used new record forms at each follow-up examination (and thus was unaware of the child’s previous findings), or there was an independent outside examiner. In the remaining 3 studies (2 RCTs and 1 cohort study), either the same examiner conducted both the baseline and follow-up examinations, or blinding was not described.
Effects of Sealants
For the individual studies, the prevented fraction ranged from 61.6% to 100.0%, with a median of 74.2% (Table 3  ). The median prevented fraction did not vary greatly by grouping—the median value always exceeded 50% (Table 3).
The RR for the studies ranged from 0 to 38.4%, but after the LaPlace adjustment, it ranged from 20.8% to 53.2% (Fig.
The summary prevented fraction ranged from 73.2% (95%CI: 59.8%–82.2%), assuming perfect correlation among teeth (adjusted n = 398), to 75.0% (95%CI: 67.1%–81.1%), assuming no correlation (adjusted n = 946), and equaled 74.1% (95%CI: 63.8%–81.4%), assuming 30% correlation (adjusted n = 638). When we restricted the analysis to the 4 randomized trials, the summary prevented fraction ranged from 71.2% (95%CI: 50.3%–83.3%), assuming perfect correlation (adjusted n = 154), to 71.3% (95%CI: 54.1%–82.0%), assuming no correlation (adjusted n = 254), and equaled 71.3% (95%CI: 52.8%–82.5%), assuming 30% correlation (adjusted n = 207). The quantity I2 was 0 regardless of our assumptions about correlation among teeth or whether to include only randomized trials, which indicated no observed heterogeneity.
We found that sealing caries lesions reduced the probability of lesion progression. The summary prevented fraction was more than 70%, and in the sensitivity analyses, the lower bound of the 95%CI always exceeded 50%. The consistency in size and direction across included studies and under a range of conservative assumptions indicates that the findings are robust. Because non-cavitated lesions accounted for almost 90% of teeth in this study, the evidence supporting the sealing of non-cavitated lesions (NC) was stronger than that for the sealing of cavitated (C) lesions. The median annualized probability of progression for NC lesions was very low (2.6%). This finding does not support reported concerns about poorer outcomes associated with the inadvertent sealing of caries and should lessen the reluctance of practitioners to provide sealants—an intervention proven to be highly effective in preventing caries. The annualized probability reflects progression in lesions recognized as "early or incipient" and suggests that the probability of progression for pit-and-fissure surfaces with caries considered "questionable" could be even lower. These findings not only support the placement of sealants to manage and arrest lesions determined to be in the early carious stages, but also, just as importantly, support their placement for surfaces where caries status is uncertain. Another notable finding of this review was the low annualized probability of progression (12.6%) for not-sealed, non-cavitated lesions. This finding suggests that immediate surgical treatment of such lesions may not be necessary. Thus, practitioners can consider sealing these surfaces or can simply wait and monitor them for signs of active progression. Approaches focusing on prevention and management (e.g., sealants) are particularly attractive, since they could potentially preserve tooth structure and lower the likelihood of future complex restorations. There were variations among the studies included in this analysis. Sealant material included both resin-based and GIC sealants. Among the 4 studies that used resin-based sealants, 3 used 2nd- or 3rd-generation sealants, which have similar retention rates (Muller-Bolla et al., 2006). The one-year retention rate of 81% for the 1 study using 1st-generation sealant material included in this meta-analysis was within the range of retention for auto-polymerized sealants reported in a meta-analysis on the primary effectiveness of sealants (Ahovuo-Saloranta et al., 2004). Although this same review found limited evidence to support the effectiveness of GIC sealant material as a primary preventive measure, 1 longitudinal study that sampled 24 teeth found no difference in bacteria levels between dentinal lesions sealed with resin-based and GIC sealants 7 mos after placement (Weerheijm et al., 1993). The studies also varied by how they assessed caries progression. Three studies assessed progression solely with a visual-tactile examination. In the absence of sealant loss or a restoration on a previously sealed caries lesion, visual-tactile assessment of caries under sealants is limited. In 1 of the 3 studies included in our meta-analysis, however, children received regular restorative care, and thus it is likely that sealed teeth were periodically assessed radiographically and restored if necessary. All RCTs (4 studies) included in this review received a "fair" quality rating, primarily due to failure to blind outcome assessment (3 studies) and high loss to follow-up (1 study). It should be noted, however, that comparative studies examining the effectiveness of sealants for primary prevention typically do not remove sealants at follow-up. For example, none of the studies included in a recent systematic review of sealants removed sealant at the final follow-up examination (Ahovuo-Saloranta et al., 2004). While limitations of this analysis have been carefully described, the strengths of these studies, and of the meta-analysis as well, should be clearly noted. First, we conducted a sensitivity analysis that adjusted for correlation among multiple observations per person to determine the most conservative (widest) confidence interval for the summary prevented fraction. Other systematic reviews of sealant effectiveness have included studies with multiple observations per person, and this systematic review is likely the first study that adjusted data for this limitation. In addition, the consistency of the effect measure across studies also lends support for the quality of the 6 studies; it is very unlikely that such consistency among estimates based on studies with noted variations occurred by chance alone. There is additional evidence for sealant effectiveness in the management of caries. Two other studies identified in the larger systematic review also examined the impact of sealants on caries progression, but did not report % of lesions progressing. One study found that caries lesions measured by radiographic assessment were more likely to regress under intact sealants than under defective sealants (Handelman et al., 1986). Another RCT found that the mean depth change in caries lesions was significantly lower in the sealed group than in the not-sealed group (49 µm vs. 614 µm depth change; Mertz-Fairhurst et al., 1979). In addition, several studies have found that sealing caries reduces bacteria levels (Jeronimus et al., 1975; Jensen and Handelman, 1980). This review also supports the need for further studies that meet current standards of quality in design, conduct, and reporting, to continue to build the evidence related to sealant effectiveness in preventing caries progression, especially in cavitated lesions, which represented, at most, 14% of carious teeth in this analysis. Uniform criteria to assess progression from early demineralization to frank cavitation, as well as standardized methodologies to measure progression, are needed. This review would have been strengthened if all studies had used examiners calibrated to the same criteria and the same method to assess caries progression (i.e., visual-tactile examination with removal of sealants). In conclusion, the evidence supports the placement of sealants over non-cavitated caries lesions in the pits and fissures of permanent teeth in children, adolescents, and young adults. Despite variations in study design and conduct, subgroup and sensitivity analyses found the effect to be consistent in size and direction.
This research was funded by the Centers for Disease Control and Prevention. We also thank Sylvia Bickley, Trials Service Coordinator, Cochrane Group, for her assistance.
A supplemental appendix to this article is published electronically only at http://jdr.iadrjournals.org/cgi/content/full/87/2/169/DC1. Received for publication February 15, 2007. Revision received July 3, 2007. Accepted for publication October 30, 2007.
Journal of Dental Research, Vol. 87, No. 2,
169-174 (2008) This article has been cited by other articles:
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
ABSTRACT
INTRODUCTION
Relative risk ratio. 
95% confidence interval assuming no correlation (rho = 0.0). 
95% confidence interval assuming rho = 0.3. 
95% confidence interval assuming rho = 1.0.
