Insufficient evidence that slow-release fluoride devices reduce caries.

Abstract

Cochrane Oral Health Group's Trials Register, Cochrane Central Register of Controlled Trials (CENTRAL), Medline, Embase, US National Institutes of Health Trials Register and World Health Organisation (WHO) International Clinical Trials Registry Platform. Parallel randomised controlled trials (RCTs) of any type of slow-release fluoride device irrespective of publication status, language or blinding were considered. Split mouth studies were excluded. Standard Cochrane methods were followed with at least two reviewers independently undertaking all the key steps. Only one trial which was considered to be at high risk of bias was included. The trial included 174 children with 132 available for two year examination. However statistical analysis was performed on only the 63 children (31 in intervention group, 32 in control group) who had retained the beads (retention rate was 47.7% at two years). Among these 63 children, caries increment was reported to be statistically significantly lower in the intervention group than in the control group (DMFT: mean difference -0.72, 95% confidence interval (CI) -1.23 to -0.21; DMFS: mean difference -1.52, 95% CI -2.68 to -0.36 (very low quality evidence)). Although this difference was clinically significant, it only holds true for those children who retained the fluoride beads; over 50% of the children did not retain the beads. Adverse effects were not reported. There is insufficient evidence to determine the caries-inhibiting effect of slow-release fluoride glass beads. The body of evidence available is of very low quality and there is a potential overestimation of benefit to the average child. The applicability of the findings to the wider population is unclear; the study had included children from a deprived area that had low levels of fluoride in drinking water, and were considered at high risk of carries. In addition, the evidence was only obtained from children who still had the bead attached at two years (48% of all available children); children who had lost their slow-release fluoride devices earlier might not have benefited as much from the devices.