We determined lidocaine's action on torn rotator cuff tendons in vitro and in vivo. For in vitro experiments, cell proliferation and viability assays were performed using tenocytes derived from human torn rotator cuff tendons. For in vivo experiments, acute rotator cuff tears were made on the supraspinatus tendons in the rats’ bilateral shoulders; before closure, lidocaine was injected into the shoulder and saline into the contralateral shoulder (control). After sacrifice, the specimens underwent biomechanical testing or histological analysis at 24 h and at 2, 4, and 8 weeks after surgery. The extent of collagen organization and apoptosis were semi-quantitatively evaluated using collagen picrosirius red staining. Apoptosis was examined using TUNEL staining and electron microscopy. Cell proliferation decreased dose-dependently. After exposure to 0.1% lidocaine for 24 h, cell viability decreased. Two and 4 weeks after surgery, the ultimate load to failure decreased more in the lidocaine group than in the control group, with significantly reduced stiffness in the lidocaine group 2 weeks after surgery. Collagen organization significantly decreased in the lidocaine group by 4 weeks after surgery but returned to baseline at 8 weeks. TUNEL staining detected numerous apoptotic tenocytes at the torn tendon edge exposed to lidocaine 24 h after surgery; electron microscopy confirmed the condensed cell nuclei. These changes were not observed in controls. Lidocaine caused cytotoxicity to tenocytes under both conditions, decreased biomechanical properties, and induced apoptosis and delay of collagen organization in this model. Subacromial lidocaine injections in patients with rotator cuff tears should be performed carefully. © 2016 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res 9999: XX–XX, 2016.