Sickle cell disease survival analysis at a large United States southern comprehensive sickle cell disease center: 1995 - 2022 Free

Background: The University of Alabama at Birmingham (UAB) sickle cell disease (SCD) comprehensive center cares for most patients with SCD in Alabama. Nationally, survival of children and young adults with SCD has improved. The most recent large U.S. single-center mortality study reported 94% survival for patients with sickle cell anemia (SCA). To inform strategies to reduce mortality, we aimed to determine (1) survival rates for our large center and (2) the most common causes of death within our SCD population.

Methods: We conducted an IRB-approved retrospective cohort study of patients born between 1995-2002. We reviewed medical records of deceased patients followed at UAB pediatric SCD comprehensive center or one of its satellite clinics (Montgomery, Tuscaloosa, and Opelika). We recorded the age at death, sex, SCD genotype, cause of death, age of transition to adult care, and time (in years) from transition to death. We grouped the total SCD cohort into patients with SCA (HbSS and HbSβ⁰ thalassemia), and “other SCD” (HbSC, HbSβ⁺ thalassemia, and rare genotypes). We categorized deaths as SCD-related (e.g., stroke, acute chest syndrome [ACS]) or non-SCD related, and by age of death: pediatric (0-18 years) or transition age (19-25 years). For our control SCD cohort, we recorded age of all living patients, censoring data at age 25 years. We performed descriptive statistics and survival analysis using JMP18 (Cary, NC).

Results: Among 1358 patients with SCD born between 1995-2022 (21,789 patient-years), 31 (2.2%) deaths occurred. Median age at death was 15 years (range 2 months-20 years). Overall survival was 97.9% at 19 years and 96.0% at 25 years for patients within our entire SCD cohort. For the subgroup analyses of patients with SCA and “other SCD,” we identified a difference in mortality between patients with SCA (27/819, 3.3%) compared to our “other SCD” cohort (4/539, 0.7%); (odds ratio of 4.6, 95% confidence interval 1.6 - 13.1; P=0.001). The overall survival for patients with SCA was 96.8% at 19 years and 94.4% at 25 years.

A bimodal distribution of deaths was observed in our SCD cohort: 7 (23%) of our 31 deaths occurred in patients <2 years of age and 10 (32%) of our 31 deaths occurred between ages 19-25 years. Of the 31 deaths, 12 (39%) were SCD-related. The most common SCD-related cause of death in our cohort was hematopoietic stem cell transplant (HSCT) complications (4/12, 33%). Other SCD-related deaths included 4 cardiopulmonary events [cardiopulmonary arrest due to pulmonary hypertension (1), cardiopulmonary arrest of unknown cause (1), and ACS (2)], 2 thromboembolic events [combined stroke and pulmonary embolism (1), and deep vein thrombosis/pulmonary embolism (1)], and 2 sepsis events. We identified 7 (23%) non-SCD related deaths [sudden infant death syndrome (3), prematurity complications (1), opioid overdose (1), motor vehicle accident (1), firearm (1)]. The cause(s) of death for 12 patients were unknown; all of whom died outside the hospital. Eight patients (26%) died after transition to adult care. The median age of death post-transition was 22 years (range 20 - 24 years), with a median time of 3 years (range 1-5 years) post-transition.

Conclusion: Most children in our SCD cohort survived into adulthood with survival rates comparable to national data; including the last large U.S. single center SCD mortality study. However, unlike national SCD data showing gradual survival decline with a rapid decline after transition to adult care, we observed a bimodal mortality pattern peaking in early childhood and again after transition to adult care. The most common known cause of death in our SCD cohort was HSCT complications followed by cardiopulmonary complications and thromboembolic events. This reveals a shift from historical leading causes of death such as ACS and infection. Mortality was significantly higher among patients with SCA. The high rate of unknown, out-of-hospital deaths underscore the importance of improved community surveillance and outreach. Our findings reflect evolving treatment and comorbidities in modern SCD care. There remains an urgent need for targeted interventions during the high-risk periods, greater focus on post-HSCT monitoring, longitudinal cardiopulmonary surveillance, and improved transition of care structure to further reduce mortality.