Purpose: This prospective clinical trial compared the retention rate and caries-preventive efficacy of two types of sealant modalities over a 3-year period. Materials and Methods: Using a split-mouth randomised design, 1280 sealants were randomly applied on sound permanent second molars of 320 young patients aged between 12 and 16 years. Half of the teeth (n = 640) were sealed with a resin-modified glass ionomer cement (RMGIC) (VitremerTM, 3M ESPE) and the other half (n = 640) with a conventional light-cure, resin-based fissure sealant (LCRB) (Fluoroshield, Dentsply Caulk). Teeth were evaluated at baseline, 6-, 12-, 18-, 24-, 30- and 36-month intervals with regard to retention and new caries development. Results: On the sealed occlusal surfaces after 3 years, 5.10% of RMGIC and 91.08% of LCRB sealants were totally intact and 6.37% of RMGIC and 7.65% of LCRB sealants were partially intact. New caries lesions were found in 20.06% of RMGIC sealed occlusal surfaces, compared to 8.91% for LCRB sealants. Conclusions: The findings of the present clinical study suggest that RMGIC should be used only as a transitional sealant that can be applied to newly erupting teeth throughout the eruptive process, whereas LCRB sealants are used to successfully prevent occlusal caries lesions once an effective rubber dam can be achieved. It can be concluded that there are differences between the RMGIC and LCRB sealants over a 3-year period in terms of the retention rate and caries-preventive efficacy. RMGIC can serve as a simple and economic sealing solution, however provisional. Due to its poor retention rate, periodic recalls are necessary, even after 6 months, to eventually replace the lost sealant. Keywords: dental caries, light-cure resin-based sealant, pit and fissure sealants, preventive dentistry, resin-modified glass ionomer cement