Title : Elevated incidence of Manipulation Under Anesthesia (MUA) after primary Total Knee Arthroplasty (TKA) at an urban safety-net hospital: Association with physical therapy
Abstract:
Purpose: Manipulation Under Anesthesia (MUA) is performed for arthrofibrosis after Total Knee Arthroplasty (TKA), but the contribution of postoperative rehabilitation engagement and neighborhood disadvantage remains incompletely characterized. This study evaluated MUA incidence, associated clinical and social factors, and postoperative range-of-motion recovery at an urban safety-net hospital.
Methods: An IRB-approved retrospective case-control study examined primary TKAs performed by two fellowship-trained arthroplasty surgeons from 2022 to 2024. All 527 primary TKAs were used to determine MUA incidence. Comparative analyses included 43 MUA cases and 244 control TKA episodes. Demographics, comorbidities, national Area Deprivation Index (ADI) percentile, Physical Therapy (PT) engagement, prior ipsilateral knee surgery, operative characteristics, and serial range of motion were collected. PT engagement was defined as at least two outpatient or home-health PT sessions or more than two weeks of inpatient rehabilitation within 12 weeks of TKA. Univariate comparisons and patient-cluster-robust multivariable logistic regression were performed. Paired analyses evaluated range-of-motion change.
Results: 43 of 527 primary TKAs required MUA, corresponding to an incidence of 8.2% (exact 95% confidence interval [CI], 6.0%–10.8%). Median time from TKA to MUA was 77 days (interquartile range, 56.5–101.5). Patients undergoing MUA were younger than controls (60.2 ± 7.6 versus 65.7 ± 8.8 years; p<0.001). PT nonengagement was substantially more common among MUA cases (25.6% versus 5.8%; p<0.001), as was prior ipsilateral knee surgery (32.6% versus 10.7%; p<0.001). Mean national ADI percentile was also higher among MUA cases (34.1 ± 24.0 versus 28.3 ± 17.3; p=0.023), indicating greater neighborhood socioeconomic disadvantage. In multivariable analysis, PT nonengagement (adjusted odds ratio [aOR], 4.45; 95% CI, 1.63–12.19; p=0.004), prior ipsilateral knee surgery (aOR, 3.24; 95% CI, 1.44–7.28; p=0.005), and younger age (aOR per year, 0.92; 95% CI, 0.88–0.97; p<0.001) were independently associated with MUA. Among MUA cases with follow-up measurements, flexion improved by a mean of 29.5° (95% CI, 21.0°–38.0°; p<0.001), although final flexion remained 9.6° below the pre-TKA value (p=0.004).
Conclusions: The MUA rate at this urban safety-net hospital was 8.2%, approximately two to three times the 2.2–4.3% reported in large national databases. PT nonengagement was the strongest potentially modifiable factor associated with MUA, while prior ipsilateral surgery and younger age identified additional higher-risk patients. The unadjusted ADI difference suggests that neighborhood disadvantage may also mark barriers relevant to postoperative recovery. Structured efforts to identify and address rehabilitation nonengagement warrant prospective evaluation.
