We evaluated the cytotoxic effects of the debris apically extruded during root canal retreatment on primary human osteoblast (HOb) cells in vitro. TNF-α and IL-1β levels were also measured. We examined three different techniques: conventional hand-files, and Mtwo and Reciproc retreatments. Filled mandibular incisors were prepared for a cytotoxicity assay in an experimental root model. The material was divided into three groups according to the technique used. Ten teeth were used as control. HOb cells were exposed to the extruded content and cytotoxicity was evaluated using the MTT test (assessing cell metabolic activity). TNF-α and IL-1β production was also analyzed using enzyme-linked immunosorbent assay. Then, all the teeth were radiographed and the residual filling material was quantified. Data were subjected to analysis of variance (ANOVA) and Tukey's test (P < 0.05). The conventional hand-file technique was significantly more cytotoxic than the other methods (P < 0.05). Reciproc was less cytotoxic than Mtwo retreatment (P < 0.05). All endodontic retreatment techniques led to a significant upregulation of IL-1β levels (P < 0.05). However, only the conventional hand-file technique caused a significant increase in TNF-α levels (P < 0.05). Root-filling removal did not affect the levels of these proteins (P > 0.05). The Reciproc system required less time than the other two methods to remove the root-filling materials (P < 0.01). The endodontic retreatment with Reciproc was the least cytotoxic and the least time-consuming method of gutta-percha and sealer removal. (J Oral Sci 58, 211-217, 2016).