Background: The most common bleeding manifestation of hemophilia is bleeding into major joints. A significant percentage of persons with hemophilia (PwH), especially those with severe disease, will eventually experience degenerative changes in one of the six major joints by the age of 30, some ultimately requiring a total hip or knee joint arthroplasty. Numerous complications have been reported in PwH undergoing arthroplasty, including post-operative bleeding, limited range of motion and readmissions. Data on the outcomes of arthroplasty in individuals with hemophilia remains limited, mostly from single-center institutional studies. Medicare, which covers around 70% of PwH in the USA aged ≥65 and approximately 5% of those <65, is a large US-national data source to examine older PwH undergoing arthroplasty. In this study, our primary aim is to examine the difference in 90-day readmission rates between PwH and the general population undergoing total knee arthroplasty (TKA) and total hip arthroplasty (THA). Our secondary aim is to examine the rates of mortality in this population.

Methodology: In this cohort study, Medicare claims from 2000-2020 were used to identify PwH based on 1) a validated algorithm and 2) a combination of having International Classification of Diseases (ICD) 9 and 10 diagnoses and Healthcare Common Procedure Coding System treatment codes. A random sample of individuals undergoing TKA or THA was selected from the general population. Each PwH who underwent either one TKA or THA was paired with 4 individuals from the general population who underwent a TKA or THA, respectively, using propensity score matching based on age, sex, and modified Charlson comorbidity index (CCI). All cause readmission was derived from subtracting admission date from the discharge date. Admissions to skilled nursing facilities were excluded. A modified CCI was used excluding acquired immunodeficiency syndrome (AIDS) and liver disease given their association with historical hemophilia treatments. Generalized estimated equations were used to estimate the unadjusted risk ratios and adjusted risk ratios (aRR) of 90-day readmission and 90-day mortality with the respective 95% confidence intervals (CI).

Results: Out of 1,484 individuals with hemophilia, 9.7% and 5.3% had at least one TKA and THA, respectively. The average age in PwH undergoing arthroplasty was 64.2 years compared to 72.8 years in the general population. Among PwH, 16.8% (n=29) were female and 45.7% (n=79) were obese, compared to 49.5% (n=95,148) and 4.4% (n=8,519) in the general population sample, respectively. Among PwH, 13.3% (n=23) individuals had AIDS compared to 0.8% (n=340) in the general population. The average adjusted CCI score in PwH was 1.57, compared to 2.07 in the general population. Within 90 days of arthroplasty, 19.7% (n=34) of PwH had a readmission compared to 8.7% (n=16,705) (p<0.0001) in the general population. When matching for age, sex, and modified CCI, the aRR for 90-day readmission was 2.19 (95% CI,1.49 - 3.22; p <0.0001) and 0.66 for death (95% CI, 0.08 - 5.50; p =0.71).

Discussion: This preliminary and nationwide real-world data presents outcomes of TKA and THA in patients with hemophilia compared to the general population. PwH are significantly more likely to experience readmission compared to counterparts in the general population, whereas 90-day mortality rates were not significantly different. Further investigation is needed to identify modifiable and non-modifiable factors contributing to high readmission rates in people with hemophilia undergoing arthroplasty and to guide interventions that improve the quality of care that PwH receive.

This content is only available as a PDF.
2025
Sign in via your Institution