Intraoperative modification of total elbow arthroplasty implants.

BACKGROUND: Modification of total elbow arthroplasty (TEA) implants may be necessary in selected patients with substantial anatomic bone deformity or those undergoing revision surgery. The purpose of this study was to investigate the prevalence and consequences of implant modifications during TEA at our institution. We hypothesized that TEA implant modification would be more common in revisions than in primary replacements, and that it would not be associated with worse clinical outcomes or increased rates of radiographic or surgical complications directly related to the implant modification.

METHODS: Elbows that had undergone TEA by any of 3 surgeons at our institution with use of intraoperative implant modification between January 1992 and October 2019 were retrospectively reviewed for the type of modification and complications. Complications were classified as definitely related, probably related, possibly related, or nonrelated to the implant's modification according to the consensus review by the 3 senior surgeons. A survey was sent out to surgeons outside of our institution to investigate whether intraoperative modification to TEA implants is a common clinical practice.

RESULTS: A total of 106 implant components were modified during 94 of 731 TEA procedures (13%) in 84 of 560 patients. Implant modifications were performed in 60 of 285 revision cases (21%) compared with 34 of 446 (8%) primary cases (P < .0001). These included shortening the stem in 40 (44%), bending the stem in 16 (15%), notching the stem in 16 (15%), tapering the stem in 9 (9%), and a combination of 2 or more of these modifications in 19 implants (17%). Among the 55 index surgeries available for complication analysis, 40 complications occurred in 28 index surgeries (11 primary and 17 revisions; 25 patients), making the overall complication rate 51%. Of these 40 complications, 23 were considered independent of any implant modification. Of the remaining 17 complications, 9 were considered nonrelated to the implant modification, 6 were possibly related, and 2 were probably related to the implant modification. Therefore, the complication rate possibly related or probably related to implant modification was 15% (8 of 55). No complication was classified as definitely related to the implant modification. No implant breakage or malfunction occurred after any modification. A total of 442 survey responses were received representing 29 countries, of which 144 surgeons (39%) performed modification to implants during TEA procedures.

DISCUSSION: This study confirmed our hypothesis that modification of TEA implants is not uncommon at our institution, particularly in revision arthroplasty. Surgeons should keep in mind that complications possibly related or probably related to implant modification were at minimum 15% and could have been as high as 30% if the patients lost to follow-up had all had complications. Implant modification may be necessary in some cases but should be exercised with thoughtful consideration and caution.