Unicondylar arthroplasty in knees with deficient anterior cruciate ligaments.

Background: Historically, a functional ACL has been a prerequisite for patients undergoing unicondylar knee arthroplasty (UKA). However, this premise has not been rigorously tested.

Questions/purposes: We compared (1) the survivorship free from revision and (2) the failure mechanisms of UKAs in ACL-deficient knees and UKAs in ACL-intact knees performed over the same time interval.

Methods: Between November 2000 and July 2008, a fixed bearing UKA was performed in 72 patients (81 knees) with intraoperatively confirmed ACL deficiency. Five patients (five knees) with preoperative instability underwent ACL reconstruction and were excluded from analysis. Of the remaining 67 patients (76 knees) without preoperative instability, implant status was known for 68 UKAs in 60 patients. Survivorship and failure mechanisms for these knees were compared to those of 706 UKAs in ACL-intact knees performed during the same time interval by the same surgeon using the same implant system. Minimum followup for the ACL-deficient group was 2.9 years (mean, 6 years; range, 2.9-10 years).

Results: Revision rates between UKAs with and without intact ACLs were similar in the absence of clinical instability (p = 0.58). Six-year UKA survivorship was 94% (95% CI: 88%-100%) in ACL-deficient knees and 93% (95% CI: 91%-96%) in ACL-intact knees (p = 0.89). Five knees (7%) in the ACL-deficient group were revised: disease progression (two), loose tibia (one), persistent pain (one), and revised elsewhere/reason unknown (one). Thirty-six knees in the ACL-intact group underwent revision (5%): aseptic loosening (13), revised elsewhere/reason unknown (11), disease progression (three), tibial subsidence/fracture (four), infection (three), pain (one), and lateral compartment overload (one).

Conclusions: At 6 years, deficiency of the ACL in patients without clinical knee instability did not impact the survivorship of UKAs compared to UKAs performed in knees with intact ACLs.