BACKGROUND The purpose of this study was to evaluate apical debris extrusion associated with different kinematics in curved root canals. METHODS Forty-five extracted mandibular molars with root curvature angles ranging between 20° and 40° and curvature radii <10 mm were randomly assigned to 3 groups (n = 15) according to the kinematics used for root canal preparation: reciprocating motion with the WaveOne Gold system (Dentsply Tulsa Dental, Tulsa, OK), rotary/reciprocating motion with the Twisted File Adaptive system (SybronEndo, Orange, CA), and the manual technique. The debris apically extruded during preparation was collected into preweighed Eppendorf tubes. The tubes were then stored in an incubator at 70°C for 5 days. The extruded debris was quantified by subtracting the preinstrumentation from the postinstrumentation weight of the Eppendorf tubes. The time required for each instrumentation procedure was recorded. Data were analyzed using 1-way analysis of variance Tukey post hoc tests (∝ = 0.05). RESULTS The WaveOne Gold reciprocating single-file system was associated with less extrusion of debris compared with hand files (P < .05) and the Twisted File Adaptive system (P > .05). The preparation time required by hand files was significantly longer than that required by the other techniques (P < .05). CONCLUSIONS Under the conditions of this study, all of the instrumentation systems caused apical debris extrusion to some degree. The WaveOne Gold reciprocating system was associated with less debris extrusion in curved root canals compared with the manual technique and the Twisted File Adaptive system although the difference between the WaveOne Gold and Twisted File Adaptive systems was not significant.