Background: Disruption of the extensor mechanism is an uncommon but catastrophic complication of total knee arthroplasty. We evaluated two techniques of reconstructing a disrupted extensor mechanism with use of an extensor mechanism allograft following total knee arthroplasty.

Methods: Twenty consecutive reconstructions with use of an extensor mechanism allograft consisting of the tibial tubercle, patellar tendon, patella, and quadriceps tendon were performed. The first seven reconstructions (Group I) were done with the allograft minimally tensioned. The thirteen subsequent procedures (Group II) were performed with the allograft tightly tensioned in full extension. All surviving allografts were evaluated clinically and radiographically after a minimum duration of follow-up of twenty-four months.

Results: All of the reconstructions in Group I were clinical failures, with an average postoperative extensor lag of 59° (range, 40° to 80°) and an average postoperative Hospital for Special Surgery knee score of 52 points. All thirteen reconstructions in Group II were clinical successes, with an average postoperative extensor lag of 4.3° (range, 0° to 15°) (p < 0.0001) and an average Hospital for Special Surgery score of 88 points. Postoperative flexion did not differ significantly between Group I (average, 108°) and Group II (average, 104°) (p = 0.549).

Conclusions: The results of reconstruction with an extensor mechanism allograft after total knee arthroplasty depend on the initial tensioning of the allograft. Loosely tensioned allografts result in a persistent extension lag and clinical failure. Allografts that are tightly tensioned in full extension can restore active knee extension and result in clinical success. On the basis of the number of knees that we studied, there was no significant loss of flexion. Use of an extensor mechanism graft for the treatment of a failure of the extensor mechanism will be successful only if the graft is initially tensioned tightly in full extension.

Level of Evidence: Therapeutic study, Level III-2 (retrospective cohort study). See Instructions to Authors for a complete description of levels of evidence.