Aim To investigate the ability of different endodontic sealers and calcium hydroxide to kill bacteria in experimentally infected dentinal tubules. Methodology Fifty-six human root segments were enlarged to size 2 (ISO size 090) Largo((R)) Peeso Reamer. After treatment with 17% EDTA and 5% NaOCl for 4 min each, the specimens were infected with Enterococcus faecalis for 3 weeks. The roots were divided into eight groups and filled with gutta-percha and AH Plus (AH); Grossman's sealer (GS); Ketac-Endo (KE); Apexit (AP); RoekoSeal Automix (RSA); or RoekoSeal Automix with an experimental primer (RP), or calcium hydroxide (CH) only. One group of specimens was left unfilled for control (CT). Following storage in humid conditions at 37 degreesC for 7 days, the root canals were re-established with new sterile Largo((R)) size 2. Dentine samples from each canal were then collected using a sterile size 5 (ISO size 150) Largo((R)) Peeso Reamer. The number of colony-forming units (CFU) was determined for each sample. Results The mean log(10) CFU in all test groups was significantly lower (P < 0.05) than that in the CT group. Root filling with AH and GS killed bacteria (mean CFU = 0) in the dentinal tubules. The mean log(10) CFU for the CH group (0.53) was lower than that of RSA, AP, RP and KE (1.36, 1.40, 1.46 and 1.94, respectively), but only the difference between the CH and the KE groups was statistically significant (P < 0.05). Conclusion Root fillings in vitro with gutta-percha and AH or GS were effective in killing E. faecalis in dentinal tubules. Other endodontic sealers, as well as CH, were less effective.