Osteonecrosis of the femoral head (ONFH) is characterized by insidious onset of hip joint pain, limited range of motion and disability in sickle cell disease (SCD) patients. Prevalence of ONFH, rates of hip replacement surgery and post-operative course have not been previously described in population-based cohort studies. We retrospectively reviewed a large SCD patient cohort from California's Office of Statewide Planning and Development (OSHPD) Patient Discharge Database (PDD), and report the risk factors and clinical outcomes for SCD patients with ONFH.
Patients were ascertained from the PDD from 1991 to 2013, through an iterative search algorithm using ICD-9 codes. In addition to descriptive statistics and univariate comparisons with the chi-square test, we used multivariable Cox proportional hazards models (with age as the time scale), to analyze factors associated with ONFH diagnosis, including sex, SCD severity and antecedent acute chest syndrome (ACS; data available from 2003-2013). Patients with more severe SCD were defined as those who averaged 3 of more hospitalizations per year. Antecedent ACS was considered as a time-dependent covariate to first ONFH diagnosis after 2003. We estimated the cumulative incidence of ONFH, accounting for the competing risk of death, by SCD severity and antecedent ACS diagnoses, with differences assessed by the Gray's test for equality. We also determined rates of all re-admissions to the PDD and emergency department within 30-90 days after hip replacement surgery, including re-admissions for venous thromboembolism (VTE) and painful vasoocclusive crises (VOC). All data were analyzed using SAS 9.4 software, and results presented as hazard ratios (HR) and 95% confidence intervals (CI).
Of the 6,237 SCD patients identified, 1,356 (22%) were diagnosed with ONFH (Table 1). Three hundred and eight of the SCD patients with ONFH underwent hip arthroplasty. The median ages of ONFH diagnosis and hip replacement surgery were 27 and 36 years, respectively. In Cox regression analyses over the entire study period, both males (HR 1.13, CI 1.01 - 1.27) and SCD severity (HR 2.75, CI 2.44 - 3.11) were associated with ONFH. In Cox regression analyses from 2003-2013, SCD severity (HR 2.89, CI 2.48 - 3.36) and antecedent ACS (HR 1.56, CI 1.31 - 1.84) were independently associated with ONFH diagnosis. The cumulative incidence curves corroborate the association between SCD severity, antecedent ACS, and ONFH (Figures 1 and 2; p<0.0001)). By 40 years of age, 36% of patients with severe SCD (vs. 14% without severe disease) and 36% with antecedent ACS (vs. 12% without ACS) had developed ONFH. Among SCD patients with ONFH, 43%, 51% and 58% were re-admitted postoperatively within 30, 60 and 90 days, respectively. At 60 and 90 days after discharge for hip arthroplasty, 22% and 28% were readmitted for VOC, respectively.
The overall prevalence of ONFH in this SCD cohort was 22%, and approximately 23% of the ONFH patients underwent hip arthroplasty (5% of the entire SCD cohort). Thus, ONFH is a common complication of SCD, which often requires surgical intervention. ONFH was independently associated with SCD severity and ACS, as has been previously suggested (Kato, Gladwin et al. 2007). Readmission within 30-60 days after hip surgery was common in this SCD cohort, and often due to painful VOC. Of the 144 readmissions within 90 days of discharge from hip replacement surgery, only 2 (~1%) were due to VTE, which is lower than reported in non-sickle cell patients undergoing hip replacement surgery. Our large, populations-based cohort study provides insight into the frequency and risk factors of ONFH in SCD patients, and also reveals that post-operative readmissions are common.
Kato, G. J., M. T. Gladwin and M. H. Steinberg (2007). "Deconstructing sickle cell disease: reappraisal of the role of hemolysis in the development of clinical subphenotypes." Blood Rev 21(1): 37-47.
Adesina:bluebird bio: Honoraria.
Asterisk with author names denotes non-ASH members.