The aim of this study was to assess the effectiveness of fluoride varnish (FV) in reducing dentine caries at the patient, tooth, and surface levels as well as caries-related hospitalizations in preschoolers. We performed a systematic review of clinical trials of FV, alone or associated with an oral health program, compared with placebo, usual care, or no intervention. Bibliographical search included electronic searches of seven databases, registers of ongoing trials, and meeting abstracts, as well as hand searching. We performed random-effects meta-analyses and calculated confidence and prediction intervals. The search yielded 2,441 records; 20 trials were included in the review and 17 in at least one meta-analysis. Only one study had low risk of bias in all domains. We found no study reporting on caries-related hospitalizations. At the individual level, the pooled relative risk was 0.88 (95% confidence interval [CI] 0.81, 0.95); this means that in a population of preschool children with 50% caries incidence, we need to apply fluoride varnish in 17 children to avoid new caries in one child. At the tooth level, the pooled weighted mean difference was -0.30 (95% CI -0.69, 0.09) and at the surface level -0.77 (95% CI -1.23, -0.31). Considering the prediction intervals, none of the pooled estimates were statistically significant. We conclude that FV showed a modest and uncertain anticaries effect in preschoolers. Cost-effectiveness analyses are needed to assess whether FV should be adopted or abandoned by dental services.