High Complication Rate Associated With Arthroscopic Lysis of Adhesions Versus Manipulation Under Anesthesia for Arthrofibrosis After Total Knee Arthroplasty

Background: Arthrofibrosis after total knee arthroplasty (TKA) is often treated by arthroscopic lysis of adhesions (ALAs) or manipulation under anesthesia (MUA). This study compared the 2-year complication rates of ALA and MUA and range-of-motion (ROM) outcomes for ALA, early MUA (<3 months after TKA), and delayed MUA (>3 months after TKA). Methods: This retrospective cohort study included 425 patients undergoing ALA or MUA after primary TKA from 2001 to 2018. Demographics, clinical variables, and complication rates were collected from clinical records and compared using Student t -tests and Kaplan-Meier log-rank tests. Multivariable logistic regressions were used for adjusted analysis. ROM data were analyzed using fixed and mixed-effects models. Results: ALA patients were younger (55.2 versus 58.9 years, P < 0.001) and underwent surgery later from the index TKA (12 versus 1.9 months, P < 0.001). The Charlson Comorbidity Index was higher in the MUA group. Preoperative ROM was significantly worse in the MUA cohort, but did not differ between groups after the procedure (117°, P = 0.27) or at 2 years. Demographics and ROM outcomes were equivalent between early MUA and delayed MUA ( P = 0.75). The incidence of repeat arthrofibrosis (7.1%) and revision arthroplasty (2.4%) was similar between ALA and MUA cohorts while ALA patients had significantly more surgical site infections (3.8%) compared with MUA patients (0.47%, P = 0.017). Discussion: Equivalent ROM outcomes were seen between ALA, early MUA, and delayed MUA for the treatment of arthrofibrosis after TKA. However, this study demonstrated a markedly higher complication rate, particularly surgical site infection, after ALA, suggesting that MUA may be the preferred option for treating arthrofibrosis at both early and late time points.