Purpose: To test the hypotheses: 1) cumulative survival rates of fully and partially retained high-viscosity glass-ionomer (HVGIC) ART sealants with heat application and glass-carbomer sealants on occlusal and free-smooth surfaces are both higher than that of resin sealants; 2) cumulative survival rate of fully and partially retained high-viscosity glass-ionomer ART sealant with heat application on occlusal and free-smooth surfaces is higher than that of comparable ART sealants without heat application. Materials and Methods: The block-randomised clinical trial covered 405 eight-year-old children. The HVGIC was Ketac Molar Easymix, the glass carbomer was GlassCarbomer and the resin sealant was Clinpro. Retention rates of sealants on occlusal and free-smooth surfaces using conventional and modified categorisation (fully and partially retained sealants vs those completely lost [at least one-third of surface re-exposed]) were the dependent variables. The Kaplan-Meier survival method was used. Results: The cumulative survival rate of completely and partially retained resin sealants on occlusal (81.2%) and free-smooth (81%) surfaces after 4 years was statistically significantly higher, and that of glass-carbomer sealants (10.8% and 21.1%, respectively) was statistically significantly lower than those of the other sealant groups. There was no statistically significant difference in survival rates of completely and partially retained high-viscosity glass-ionomer ART sealants with (56% for both surfaces) or without heat application (56%) on occlusal and free-smooth surfaces (55.7% and 59.1%, respectively). Conclusions: Resin sealants had the highest and glass-carbomer sealants the lowest retention rate after 4 years. Application of heat to high-viscosity glass-ionomer ART sealants did not result in a significantly higher sealant retention rate. Use of the modified categorisation for determining sealant retention is advocated. Keywords: atraumatic restorative treatment, cement, glass carbomer, glass ionomer, resin composite, sealant retention