Data sources: The Cochrane Oral Health Group's Trials Register, Cochrane Central Register of Controlled Trials, MEDLINE and EMBASE databases were searched. Previously published systematic reviews of fluoride toothpastes were also screened to identify any reports that met the inclusion criteria. In addition, the trials database at www.controlled-trials.com/ and the meta Register of Controlled Trials (www.controlledtrials.com) were searched to identify any ongoing studies of relevance.
Study selection: Randomised controlled trials (RCT) and cluster-RCT that compared fluoride toothpaste with placebo or fluoride toothpaste of a different concentration in children of up to 16 years of age, with a followup period of at least 1 year, were included.
Data extraction and synthesis: Inclusion of studies, data extraction and quality assessment were undertaken independently and in duplicate by two members of the review team. Disagreements were resolved by discussion and consensus or by a third party. The primary effect measure was the prevented fraction (PF, the caries increment of the control group minus the caries increment of the treatment group, expressed as a proportion of the caries increment in the control group). Where it was appropriate to pool data, network meta-analysis, network meta-regression or meta-analysis models were used. Potential sources of heterogeneity were specified a priori and examined through random-effects metaregression analysis where appropriate.
Results: From 535 studies found, 75 were included. From these, 71 studies (79 trials) contributed data to the network meta-analysis, network metaregression or meta-analysis. For the 66 studies (74 trials) that contributed to the network meta-analysis of decayed, missing or filled surfaces [D(M)FS] in the mixed or permanent dentition, the caries preventive effect of fluoride toothpaste increased significantly with higher fluoride concentrations, with a D(M)FS PF compared with placebo of 23% [95% credible interval (CrI), 19-27%] for 1000/ 1055/ 1100/ 1250 ppm concentrations, rising to 36% (95% CrI, 27-44%) for toothpastes with a concentration of 2400/ 2500/ 2800 ppm. Concentrations of 440/ 500/ 550 ppm and below showed no statistically significant effect when compared to placebo.There is some evidence of a dose-response relationship in that the PF increased as the fluoride concentration increased from the baseline although this was not always statistically significant. The effect of fluoride toothpaste also increased with baseline level of D(M)FS and supervised brushing, though this did not reach statistical significance. Six studies assessed the effects of fluoride concentrations on the deciduous dentition with equivocal results dependent upon the fluoride concentrations compared and the outcome measure. Compliance with treatment regimen and unwanted effects was assessed in only a minority of studies. When reported, no differential compliance was observed and unwanted effects such as soft tissue damage and tooth staining were minimal.
Conclusions: This review confirms the benefits of using fluoride toothpaste in preventing caries in children and adolescents compared with placebo, but only statistically significantly at fluoride concentrations of 1000 ppm and above. The relative caries preventive effects of fluoride toothpastes of different concentrations increase with higher fluoride concentration. The decision of what fluoride levels to use for children aged under 6 years should be balanced with the risk of fluorosis.