Purpose: To estimate the survival of retention of sealant extension to occlusal ART restorations over 6.3 years; and to test the null-hypothesis that there is no difference in dentine caries lesion development in sealant extension to ART restorations in comparison with sealant free extensions to amalgam restorations in occlusal surfaces over 6.3 years. Materials and Methods: In a parallel group design, 318 and 254 grade 2 children were randomly assigned to the ART and amalgam group respectively. Eight dentists placed 925 evaluatable single- and multiple-surfaces restorations. A total of 424 sealed extensions to occlusal ART and 284 sealant free extensions to occlusal amalgam restorations were available for analyses. The modified actuarial method was used to estimate survival percentages. The jackknife method was applied to calculate the SE in the cumulative survival percentages. Results: After 6.3 years, 11.2% (SE = 2.2%) of sealant extensions were fully retained and 16.7% (SE = 2.8%) were partially retained. After 6.3 years, 86.4% (SE = 2.2%) of the sealed pits and fissures adjacent to occlusal ART restorations and 89.9% (SE = 2.4%) of non-sealed pits and fissures adjacent to occlusal amalgam restorations were free of dentine caries lesions. Neither this difference nor those at earlier evaluation years were statistically significant (p > 0.05). Conclusion: Sealing pits and fissures adjacent to occlusal ART restorations did not result in a caries preventive benefit over non-sealed pits and fissures adjacent to occlusal amalgam restorations in this group of children over 6.3 years. Keywords: amalgam, atraumatic restorative treatment, glass-ionomer, sealant extension, survival