Objective: The aim of this study was to evaluate, ex vivo, bacterial coronal leakage with different antimicrobial agents applied to the dentine for indirect pulp treatment (IPT). Study D esign: Sixty extracted teeth were prepared and randomly distributed into 5 groups (n=10): Group 1: no antimicrobial dentine treatment; group 2: 1% chlorhexidine (CHX)+1% thymol varnish (Cervitec ®) ; group 3: 2 % CHX solution; group 4: 40% CHX varnish (EC40TM) and group 5: ClearfilTM Protect Bond (CPB). Ten teeth served as controls. The teeth were restored using a resin-modified glass ionomer cement (GIC) and then mounted in a two-chamber device. The coronal access was exposed to Streptococcus mutans for 45 days. The appearance of turbidity in the BHI broth of the lower chamber was considered as specimen leakage. Results: Survival analysis, determined by non parametric Kaplan-Meier and log-rank tests, showed that the best results were for groups EC40TM+GIC and GIC alone; yet there were not statistically significant differences be - tween them. All specimens of CPB+GIC and 2% CHX+GIC, leaked at 45 days. Conclusions: In IPT the use of GIC without pretreatment of the dentine and pretreatment with 40% CHX varnish resulted in a significant delay of bacterial coronal leakage.