Objective: The choice of the graft in ossicular chain reconstruction during middle ear surgery for cholesteatoma is a subject still discussed on. In order to clarify the discussion of reuse of the autologous ossicles obtained during middle ear surgery for cholesteatoma. we evaluated the probability of residual disease histologically and the safety of the ossicles after autoclavization, the most promoting alternative method to eradicate residual cholesteatoma and infection on them. Methods: The specimens used in this study were eroded twenty-seven ossicles (22 incuses, 5 malleoli) which were removed from the 27 consecutive patients operated because of cholesteatomato us middle ear disease. They were grouped as follows: Group 1, Fifteen ossicles examined histopathologically directly. Group 2, Five ossicles autoclaved for 20 min at 134 C and then examined histopathologically. Group 3, Five ossicles autoclaved for 20 min at 134 C after mechanical surface cleaning by a fine diamond drill, examined histopathologically. Group 4, Two ossicles removed from two different patients were placed in their mastoid cavities in order to be examined after access in the second-look operation. While one ossicle was only autoclaved, the other was mechanically cleaned by a drill before autoclavization (for 20 min at 134 C). The ossicles were examined histopathologically after the removal at the second stage operation performed 12 months later. Results: In Group 1, all ossicles showed evidence of periosteal thickening. Additional findings were surface cholesteatoma or epithelia in 13 ossicles, surface inflammation in 12 ossicles, granulation tissue in 10 ossicles, osteitis in six ossicles. In Group 2, all five ossicles had preserved their lamellar structure but, no vital cells were seen. The lacunes that had the osteocytes was almost completely empty. The inflammatory cells were eliminated from the ossicles. In Group 3, ossicles were found well preserved with their lamellar structures and contours, with empty lacunes and eliminated inflammatory cells. In Group 4, in two ossicles of this group the lacunes were replaced by the new migrated viable osteocytes with evidence of new bone formation and neovascularisation. No new inflammatory focus or epithelia were found on the surfaces of the ossicles. The shape and the contour of the ossicles remained unchanged. Conclusion: In cholesteatoma surgery, ossicles with minimal erosion and adequate thickness can be used after autoclavization. In this study, it was observed histopathologically that the autoclaving autologous ossicles before ossiculoplasty in cholesteatomatous middle ear is a safe and reliable method. (C) 2002 Elsevier Science Ireland Ltd. All rights reserved.