Clinical outcomes and perioperative safety of primary total knee arthroplasty for hemophilic arthropathy versus osteoarthritis: a retrospective cohort study

Abstract Background Hemophilic arthropathy (HAr) frequently progresses to end-stage joint disease requiring total knee arthroplasty (TKA), but coagulation dysfunction raises concerns about perioperative bleeding and complications. We aim to analyze the clinical outcomes and safety associated with primary TKA in patients with HAr. Methods This single-center retrospective cohort included 22 TKAs in 20 men with HAr and 66 TKAs in 66 male osteoarthritis (OA) controls selected by surgeon and proximity of surgery date. The procedure was the analysis unit, and standard errors were clustered by patient. Continuous outcomes were compared with Welch tests and adjusted using baseline-outcome analysis of covariance including group, age, and body mass index. Sparse binary outcomes were exploratory and analyzed with Fisher exact tests and Firth logistic regression. Calculated blood loss was evaluated using the Nadler-Gross method. Results Patients with HAr were younger (41.86 ± 11.29 versus 68.33 ± 6.70 years). Mean improvement in the Knee Society clinical score was 49.64 ± 18.43 versus 53.24 ± 5.77 points (unadjusted between-group change-score difference, − 3.61; 95% CI − 11.88 to 4.67; p = 0.376). Adjusted 12-month group effects were 0.54 points (95% CI − 4.19 to 5.28; p = 0.820) for Knee Society clinical score and 3.02 points (95% CI − 9.73 to 15.77; p = 0.639) for Knee Society function while adjusted flexion remained lower (mean difference, − 19.64°; 95% CI − 38.29° to − 0.99°; p = 0.039). Calculated blood loss did not differ significantly (532.53 ± 355.18 versus 628.00 ± 328.38 mL; p = 0.275), while transfusion occurred in 4/22 versus 0/66 procedures ( p = 0.003). Wound problems (6/22 versus 3/66) and 90-day readmissions (9/22 versus 1/66) were more frequent in crude comparisons, but adjusted estimates were highly imprecise. Conclusions Under comprehensive blood management, TKA produced substantial within-group functional improvement in HAr. Adjusted 12-month Knee Society clinical and function scores were not detectably different, although flexion remained lower. Calculated blood loss was not increased, but transfusion and unadjusted wound or readmission signals warrant multidisciplinary care and confirmation in larger, prospectively defined cohorts.