The purpose of this study was to investigate how two cavity-filling techniques affect microleakage in class II resin restorations prepared with Er:YAG laser and diamond bur. Standard MO and DO cavities were prepared in 20 extracted third molars, each randomly assigned to either Group-1 [Herculite XRV Ultra-bur-prepared cavity(bp)], Group-2 [Herculite XRV Ultra-laser-prepared cavity(lp)], Group-3 (SonicFill-(bp)], or Group-4 [SonicFill-(lp)]. For Groups 2 and 4, cavities were prepared by using an Er:YAG laser with a wavelength of 2.94m, output power of 200mJ/pulse, and repetition rate of 20Hz. Teeth were restored with a one-step, self-etch adhesive material (OptiBond All-in-One), a nanohybrid composite (Herculite XRV Ultra), and a bulk-fill composite (SonicFill) according to the manufacturer's instructions. Five teeth from each group were chosen for microleakage investigation and two teeth for scanning electron microscope evaluation. Statistical analysis was performed by using Kruskal-Wallis test. Pair-wise comparisons were performed by Mann-Whitney U test with Bonferroni correction (p<0.05). The statistical analysis of data revealed greater microleakage in cervical versus occlusal regions in all groups (P<0.05). Group-2 yielded a higher degree of marginal leakage than Group-3 in terms of occlusal surfaces (p<0.05). For cervical regions, a statistically significant difference was observed between Groups 2 and 3 as well as Groups 2 and 4 (p<0.05). Group-3 exhibited significantly better marginal sealing than Group-1 in the cervical region (p<0.05). The cavities prepared using an Er:YAG laser showed greater microleakage than those conventionally prepared using burs regardless of restorative material at both occlusal and cervical margins. SCANNING 38:389-395, 2016. (c) 2015 Wiley Periodicals, Inc.