Surgeon training and complications in total ankle arthroplasty.

Saltzman CL, Amendola A, Anderson R, Coetzee JC, Gall RJ, Haddad SL, Herbst S, Lian G, Sanders RW, Scioli M, Younger AS.

Department of Orthopaedic Surgery, University of Iowa, Iowa City, IA 52246, USA. charles-saltzman@uiowa.edu

BACKGROUND: This study assessed the problems with initial use of ankle arthroplasty by surgeons who were trained by observing the surgeon/inventor (group I), who have completed a structured, hands-on surgical training course (group II), or who were trained during a 1-year foot and ankle fellowship (group III).

MATERIALS AND METHODS: The perioperative records of the first 10 cases of nine surgeons were reviewed. We evaluated the 6-month-postoperative standing mortise and lateral radiographs for evidence of syndesmosis union and accuracy of tibial component implantation. Three surgeons were each in group I, group II, and group III. Average patient age at time of surgery was similar. Ankle arthritis was classified as rheumatoid arthritis (RA) or osteoarthritis (OA) as follows: group I (7 RA, 23 OA), group II (7 RA, 23 OA), and group III (3 RA, 27 OA).

RESULTS: In group I, there were nine intraoperative complications, four postoperative wound dehiscences, and three postoperative deep infections. Radiographic evaluation of the 26 cases with adequate postoperative roentgenograms revealed that 10/26 (38%) had a delayed union of the syndesmosis. In group II, there were six intraoperative complications and two postoperative wound problems: an early anterior wound problem and a delayed lateral wound breakdown. Radiographic evaluation of the 26 cases with adequate postoperative roentgenograms revealed that 13/26 (50%) had a delayed union of the syndesmosis. In group III, there were four intraoperative complications and four postoperative wound problems--all healed with local supportive care with one requiring lateral hardware removal. Radiographic evaluation of the 26 cases with adequate postoperative roentgenograms revealed that 5/30 (17%) had a delayed union of the syndesmosis. The initial series from these three groups are statistically indistinguishable with respect to rates of complications, revisions, or malalignment.

CONCLUSION: No identified training method had a statistically demonstrable positive impact on preparing surgeons for performing total ankle replacement. Some of these findings are likely generic for total ankle replacements and not restricted to any class or design of implant. Surgeon initial use of total ankle replacement needs to be done with caution and serious consideration.