Introduction

Dental caries is an irreversible, non-communicable microbial disease of calcified tissues of teeth with a multifactorial etiology. Cariogenic bacteria, a susceptible tooth, fermentable carbohydrates, and sufficient time together result in decay. Ninety percent of carious lesions occur on occlusal surfaces. Pit and fissure sealants provide specific protection to these areas which are least benefited from fluoride. They form a protective layer that is micro-mechanically or chemically bonded to the tooth structure cutting off nutrients for micro-organisms1.

Pit and fissure sealants can be broadly classified as resin-based and glass ionomer sealants which are indicated for individuals with higher caries risk2. With the advent of hydrophilic sealants, resin-based sealants can be further classified as hydrophobic resin-based sealants and hydrophilic resin-based sealants. Several clinical trials have demonstrated the effectiveness of hydrophobic resin-based sealants in the prevention of pit and fissure caries. Ahovuo-Saloranta A et al. reported moderate-quality evidence of caries reduction with hydrophobic resin-based sealants ranging between 11 and 51% at 24 months. The benefit remains similar up till 48 months but quality and quantity of evidence reduces3. Sealants are beneficial in caries prevention but numerous factors are affecting their retention. It is affected by operator’s experience, patient characteristics, clinical setting, technique, method of isolation, application of fluoride following sealant application and so forth. Salivary contamination and material properties are primary factors affecting retention4. Saliva adheres to the etched enamel forming a barrier that lowers bond strength. Also, salivary contamination is not visible many times. Rubber dam isolation is the ideal method but non-feasible in newly erupted teeth, uncooperative patients etc. Another alternative suggested is the use of dentin bonding agents5. However, newer sealants are usually placed directly. Thus, sealant application may not be possible many times when their placement can be most effective.

Glass-ionomer sealants (GIC) are used in situations where moisture control is difficult. They have low retention rate, high solubility and low abrasive strength6. Alirezaei M et al. found no difference in caries development with the use of GIC and hydrophobic resin-based sealants while retention was found to be higher for hydrophobic resin-based sealants7.

Hydrophobic resin-based sealants contain bisphenol-A (BPA) and bisphenol-A-dimethacrylate. There is increase in bisphenol-A in saliva and urine ranging from 2 days to 6 months8. The potential harm of this is uncertain but necessary precautions following sealant application are advised. Fleisch et al. suggested priority should be given to the development of non-bisphenol-A materials9.

In the last decade, hydrophilic resin-based sealants have been developed which set in moist environment and are also devoid of Bisphenol-A10. They are applied on occlusal surfaces of the teeth which are dried, etched, rinsed, and gently dried with cotton rolls, having a shiny instead of chalky white appearance. Studies have reported that these sealants have good marginal adaptation, retention, smooth margins, low technical sensitivity, and release fluoride. Thus, they have been suggested for use in recalcitrant children and cases where isolation is difficult11.

Several randomized controlled trials have documented the performance of hydrophobic resin-based sealants, hydrophilic resin-based sealants and GIC sealants. Current studies have reported mixed results with few reporting better retention and caries prevention with hydrophilic resin-based sealants, some favouring hydrophobic resin-based or GIC sealants, while others are inconclusive. Hence, the clinical trials show different performances of these sealants, and it is essential to evaluate which among these has better retention and caries prevention, thus aiding in clinical decision. The present systematic review aimed to evaluate the retention and caries preventive effect of hydrophilic resin-based sealants compared with hydrophobic resin-based and glass-ionomer sealants among children and adolescents. The secondary objective was to assess their safety.

Materials and methods

Registration

The review protocol was registered with Joanna Briggs Institute on 20th January 2020 and followed Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) Guidelines.

Eligibility criteria

Inclusion criteria:

  1. A.

    Clinical trials with at least six months of follow-up, parallel and split-mouth designs, were included, with randomization units being individual or tooth pairs.

  2. B.

    PICO

The included participants were 6–13-year-old children with recently erupted permanent molars susceptible to dental caries. Hydrophilic resin-based sealants were applied in the intervention group (Embrace WetBond, Champ) and hydrophobic resin-based fissure sealants (Clinpro, Delton FS, Helioseal F, Fissurit F) or glass ionomer sealants (GC Fuji type VII, type IX, Glass Carbomer, Fuji Triage) in the comparison group. The primary outcomes assessed were the retention and cariostatic effect of these sealants at 6, 12, and 18 months, while the secondary outcome assessed was the safety of pit and fissure sealants.

Exclusion criteria:

  1. A.

    In-vitro studies, non-randomized studies, animal studies, case reports, literature review.

  2. B.

    Studies that used interventions other than pit and fissure sealants viz fluoride varnish, fluoride gels, etc.

  3. C.

    Studies that used bonding agents concomitantly with sealants.

  4. D.

    Clinical trials on teeth with dental anomalies of enamel and dentin.

Information sources and literature search

Publications listed between 2009–2019 were searched using PubMed, Google Scholar, Virtual Health Library, and Cochrane Central Register of Controlled Trials electronic databases. Search terms used were: hydrophilic sealants, EMBRACE WetBond & ULTRASEAL XT. A combination of MeSH terms and keywords was used. (Appendix 1) Grey literature and citation lists of the included articles were hand-searched for any additional trials. Study authors were contacted to request missing information or seek clarification when necessary.

Study selection

Two reviewers (AK, VK) independently screened, assessed, and extracted data from all the articles based on inclusion and exclusion criteria. Full-text articles were retrieved for relevant articles including those which couldn’t be finalized based on title and abstract. (Fig. 1) Any disagreement was solved with mutual consensus (k = 0.95; 95% CI = 0.92–1.00). Irrespective of the results reported by the studies, the data was analyzed using intention to treat wherever possible.

Fig. 1: Flow diagram.
figure 1

PRISMA flow diagram for systematic reviews summarizing the process of screening.

Full size image

Data collection

The characteristics of included trials and their numerical data were extracted independently by two authors (AK, VK). The primary outcomes of the review were retention and caries prevention. The secondary outcome was adverse effects after pit and fissure sealant application.

Quality and risk of bias assessment (of individual trials)

Modified Jadad Criteria was used to assess the quality of individual trials. While assessing the quality, the main focus was on the methodological quality of the included trials. After discussion with a group of experts, considering pit and fissure sealants, the original Jadad scale was expanded to include allocation concealment, inclusion and exclusion criteria, primary objective, sample size calculation, baseline assessment, and data analysis, to improve the criterion. The risk of bias was assessed using the criteria recommended by Cochrane (including six domains i.e., random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, and selective report. The bias was reported as low risk if the plausible bias was unlikely to seriously alter the results, unclear risk if it raises some doubt about the results & high risk if it seriously weakens the confidence in the results). (Cochrane Collaboration, 2011) (Fig. 2).

Fig. 2: Risk of bias.
figure 2

Representative summary table for the risk of bias assessment.

Full size image

Data synthesis

Quantitative and qualitative analyses were performed. To reduce heterogeneity amongst the studies, partial loss in retention was merged with loss of sealant. Similarly, for dental caries, missing and decayed were merged. Meta-analysis was performed separately for the studies which had similar intervention and referent groups and measured similar outcomes. As the effectiveness of sealants can be affected by the operator’s experience, clinical setting, technique, patients’ characteristics, and so forth, a random-effects model was used for analysis. Relative risks and confidence intervals were calculated. The heterogeneity was assessed using I2 statistics. All the analyses were performed using RevMan software (Review Manager version 5.4). P-value < 0.05 was considered statistically significant.

Risk of bias across studies

The overall quality of the studies was assessed using GRADEpro online software (https://gradepro.org/) following the Grades of Recommendations, Assessment, Development, and Evaluation (GRADE) approach12.

Additional analyses & sensitivity analyses

A funnel plot was generated to evaluate the publication bias. Sensitivity analysis was performed by excluding the trials that augmented the heterogeneity.

Results

Study selection

The selection process is depicted in Fig. 1. Database, grey literature, and hand search yielded 11,593 articles. Of these, 24 duplicates were removed. 11,498 articles were excluded after screening the titles. After assessing the abstracts and full-text articles, 57 articles were excluded which did not meet the inclusion criteria. Finally, 11 studies met the criteria and were thus included in the qualitative and quantitative synthesis.

Study characteristics

Six studies were randomized control trials and five split-mouth trials in which intervention (hydrophilic resin-based sealant) was randomly allocated to teeth within a tooth pair. In the included studies, the children were aged 6 years to 13 years. Children were recruited from schools or dental clinics in most of the studies. Applications were made on sound surfaces in most of the studies; on incipient or cavitated enamel lesions in one study13. Four studies reported caries risk of subjects with one applying sealants on individuals with low risk11 and three on medium or high-risk individuals1,2,14. Seven studies compared hydrophilic sealants with hydrophobic resin-based sealants (n = 1560 teeth) three with glass- ionomer based sealants (n = 692 teeth) while one compared both resin-based and glass ionomer sealants with hydrophilic sealants (n = 240 teeth)14. Most of the studies comparing hydrophilic resin-based sealants used Embrace WetBond hydrophilic sealant, while one used Champ1.

Outcome measures

11 studies reported retention while 10 reported incidence of caries, data on caries from one study was excluded as it used diagnodent scores for assessing caries13. Incidence of carious lesions was reported dichotomously by seven studies1,2,11,12,13,14,15,16,17. Two studies reported the data as sound, enamel lesion, and demonstrable loss of tooth structure18,19. Most of the studies followed Simonsen or modified Simonsen criteria; two used modified Color Coverage Caries Sealant Evaluation System; while one followed modified Ryge criteria and used radiographs as well for evaluating sealant retention and caries.

Partial and complete loss categories were merged to reduce the heterogeneity amongst the studies. Further, categories missing and decayed were merged. None of the studies reported adverse effects. The range of kappa for intra-examiner reliability varied from 0.75 to 0.97.

Follow-up period

Studies had follow-up period ranging from 6–24 months. Ten studies reported data at 6 & 12 months1,2,11,13,14,15,16,18,19,20, two at nine & twenty-four months15,19 and three studies reported at 18 months15,17,19.

Results of individual studies

The characteristics of included trials and risk of bias amongst them, quality of evidence, forest plots, and publication bias are given in Table 1, Fig. 2, Table 2, Figs. 3 and 4, respectively.

Table 1 Characteristics of Included Trials.
Full size table
Table 2 Quality of evidence.
Full size table
Fig. 3: Forest plot.
figure 3

Forest plots depicting retention of sealants.

Full size image
Fig. 4: Funnel plot.
figure 4

Funnel plot to assess publication bias.

Full size image

Hydrophilic versus hydrophobic resin-based sealants

Among the six studies at 6 months, Ratnaditya et al. found hydrophilic sealants to be significantly better15. However, Bhat et al. found hydrophobic sealants to be significantly better at 6 and 12 months14. Mohanraj et al. compared retention of two hydrophilic sealants at 9 months and 12 months and reported Embrace wet bond (Hydrophilic) had significantly better retention than hydrophobic resin sealant while champ (Hydrophilic) had significantly lower retention than hydrophobic resin-based sealant. Among the two studies reporting data at 9 months Mohanraj et al. reported one hydrophilic resin-based sealant to be significantly better.

Among the seven studies comparing at 12 months, Ratnaditya et al. favored hydrophilic resin-based sealants and the results were statistically significant while Schlueter et al. reported hydrophobic sealant to be significantly better at the same time period16. Amongst the two studies followed up for 18 months, Ratnaditya et al. found hydrophilic sealants to be significantly better15. None of the studies found any significant difference in caries incidence.

Hydrophilic resin-based versus glass ionomer based sealants

Amongst the four studies at six months, three studies i.e., Subramaniam et al., Alsabek et al. and Bhat et al. found hydrophilic sealants to be significantly better. Among the three studies at 12 months, Bhat et al. found hydrophilic sealants to be significantly better. There was no significant difference in caries incidence at these time intervals.

Quality of individual studies

Modified Jadad Criterion was used to assess the quality of evidence. Of the eleven studies included in the systematic review, four were of good quality11,12,13,14,15,16,17,18, five of fair1,2,14,15,19, and the remaining studies had low quality17,20.

Risk of bias within studies

Figure 2 summarizes the risk of bias amongst the included trials. The most problematic were lack of allocation concealment (in 64% of the trials), incomplete blinding of outcome assessor (54.5%), incomplete blinding of participants or personnel (45%), and improper sequence generation (45%).

Risk of bias across the studies

Assessment of the meta-evidence across the studies with the GRADE approach indicated very low to low quality evidence. The main limitations amongst them were imprecision and inconsistency.

Data synthesis and sensitivity analysis

Meta-analysis was performed for studies with similar duration and comparing similar materials. A random-effects model was used as the application of a sealant is affected by many factors resulting in heterogeneity. In some analyses, outliers were removed to reduce heterogeneity16. Relative risk with a 95% confidence interval was calculated. Funnel Plot was generated to evaluate the publication bias across the studies. Symmetrical plot indicated lack of publication bias.

Discussion

Pit and fissure sealants provide specific protection against caries. Fluoride has limited efficacy in these areas. Caries being ubiquitously present and as the economies across the world are increasing their spending on prevention, the use of pit and fissure sealants is bound to increase. Thus, it is essential to identify the performance of various sealants. Hydrophilic resin-based sealants are newer sealant materials which are moisture tolerant and can be used even when teeth have not been fully erupted which is the most critical time of sealant placement. The current systematic review addressed an important question of the effectiveness of hydrophilic resin-based sealants depending on the clinical properties. We limited our search to in-vivo studies and those applying sealants on molars as they have the highest risk of caries and the benefits of sealant placement can be visualized at the earliest. The results of our meta-analysis based on a random-effects model revealed that hydrophilic resin-based sealants had better retention compared to glass-ionomer fissure sealants, while the retention was similar or slightly lower than hydrophobic resin-based sealants. This is a relevant clinical finding as it supports use of hydrophilic resin-based sealants in the clinical practice.

There was methodological heterogeneity due to various parameters across the studies. Evaluation of sealants was based on sealant coverage, margin adaptation, smoothness of the surface of sealant, color, retention and caries status. Among the criteria used to evaluate sealant, most of the studies used Simonsen’s criteria, one used modified Ryge11 while two studies used CCC evaluation system18,19. The assessment of retention is the primary measure for sealants’ effectiveness. Different evaluation criteria assess sealant retention disparately e.g., partial and complete loss, or based on percentage of loss of sealants. This heterogeneity was reduced by dichotomizing and merging the partly lost and lost sealants. The groups should be comparable at baseline. Most of the studies lacked on baseline comparison. Moreover, many parameters which could affect sealant retention were not assessed.

Sealant failure at 6 months is attributable to faulty technique, while losses after 6 months occur due to material wear by occlusal forces21. In this review, hydrophilic resin-based sealants had similar retention as hydrophobic resin-based sealants while GIC sealants had lower retention, but the quality of evidence was very low as per GRADE criteria.

This review considered both retention and caries prevention. None of the studies had significant caries reduction across the groups. Although the follow-up period was less with 12 months in majority of the studies and only two studies had a follow-up of 24 months15,19. The results are in line with the review by Lam et al. at 24 months22.

Jan K et al. review on the longevity of pit and fissure sealing reported a 5-year retention rate of fluoride-releasing sealants as 69.9% (95% CI = 51.5–86.5%)23. The studies in our review had a follow-up up to 2 years, yet sealant loss was higher.

Applying sealants without assessment of risk is not justified, yet studies regularly report sealant application irrespective of individual’s risk. In the present review, three studies applied sealant in medium/high caries risk individuals1,2,14, while one applied in low risk11 individuals.

To safeguard against selection bias sample size, randomization, and allocation concealment are necessary. Six studies had adequate random sequence generation indicating altered probability of selection of patients in intervention group. This can lead to overestimation of outcomes24. Random sequencing methods applied were coin toss method2,11,18 and computer-generated random number16. Sample size calculation has been mentioned in only 6 out of 11 studies. Allocation concealment was adequate in four studies while blinding of outcome assessor was adequate in five studies. In split design blinding of a participant does not affect the results as both the interventions are received by the participant.

Tooth pairs were taken into consideration. Few studies included two pairs from one individual also1,14,15,16,17,18,19,20. Five studies used split-mouth design. Mickenautsch S et al. regarded split-mouth trial to be quasi-randomized as all the subjects didn’t have the same chance of participation leading to selection bias6. However, the objective of randomization is the removal of the effect of extraneous variables, split designs offer an undisputed advantage in minimizing the effect of known and unknown extraneous variables. Thus, split design can lead to more balanced trials as compared to parallel designs.

Both the intervention and control groups need to be comparable at the baseline. Factors such as fluoride exposure, caries risk, age, eruption status, anatomy of the fissure, presence of dental fluorosis, tooth on which sealant is applied can impact the outcome of sealant, none of the studies reported baseline data indicating selection bias. Other factors related to the application technique such as operator applying the sealant, method of isolation, clinical setting, steps followed, application of fluoride following sealant application can also affect the outcomes. In public health programmes, sealants are applied by individuals with varying clinical competencies leading to more variable retention. Studies have shown different retention rates depending on the expertise of sealant placement25. In this review, six studies mentioned the use of cotton rolls and saliva ejector for isolation of teeth11,13,15,18; while two used only cotton rolls16,19. Also, the majority of studies did not reveal about the operator which might have affected the success. Salivary contamination is the most common cause of sealant failure. Albeit moisture tolerant sealants are hydrophilic, their bond strength is inversely affected by the degree of saliva contamination. A study by Panigrahi A et al. studied the effect of artificial saliva contamination on micro tensile bond strength of a hydrophilic resin-based sealants. The comparison was based on varying degrees of moisture contamination. There was no significant decrease in the micro-tensile bond strength in air thinned decontaminated group26.

The studies with missing information about drop-out rates depict a high risk of bias. Loss to follow-up has been reported to be less than 20 percent in 9 out of 11 studies indicating low attrition bias. Publication bias was investigated using a funnel plot. A symmetrical funnel plot indicated absence of publication bias.

Concerns have been expressed about the estrogenic potential of resin-based sealants. There can be an increase in BPA detected in children’s saliva and urine ranging from 2 days to 6 months as reported by ADA8. Thus, it is recommended to take necessary precautions following sealant application. Also, the development of non-BPA materials should be prioritized9. Therefore, hydrophilic resin-based sealants being devoid of BPA have an edge over hydrophobic resin-based sealants regarding safety10. None of the studies in this review, reported any adverse effects.

Longer studies are more resource consuming and attrition is a serious problem which limits researcher to fix a time interval, nonetheless, longer time intervals are essential for developing conclusive evidence. Most of the studies were followed up for 1 year; while only two for 18 months. Future studies with longer follow should be conducted. Hydrophilic resin-based sealants should also be compared in situations where the risk of moisture contamination is high. A systematic review provides the highest quality of evidence but that is dependent on the quality & bias of the included studies which impact the internal validity. The majority of the studies were of low or very low quality. Also, the risk of selection bias was high. Thus, further studies should adopt robust randomization techniques, follow allocation concealment, assess baseline variables and comply with Consolidated Standards of Reporting Trials guidelines.

Conclusion

Hydrophilic resin-based sealants were more retentive as compared to glass-ionomer fissure sealants, while their retention was similar or slightly lower than the hydrophobic resin-based sealants. The quality of evidence is low to very low. Caries preventive efficacy of all the materials is similar.