Abstract
Sickle cell disease (SCD) is associated with a high level of morbidity and early loss of life. A widely cited analysis modeling life expectancy with SCD (Lubeck et al, 2019) estimated a 20-year reduction, based on the mortality rate in surveillance data from California and Georgia from 2004 to 2008 of 10.1 deaths per 1,000 person-years (Paulukonis et al, 2016). However, those states' SCD populations may not generalize nationally, and the period examined preceded major developments in the treatment of SCD, including the BABY HUG trial, FDA approval of novel therapeutic agents, and expansion of comprehensive care centers.
As a prospective longitudinal patient registry, the Globin Research Network for Data and Discovery (GRNDaD) was developed to provide high-quality descriptive data on the natural history of SCD and facilitate quality improvement across participating sites. Here we present updated mortality statistics from GRNDaD spanning 56 clinical sites from May 5, 2010 to August 1, 2026.
Baseline demographic and clinical data were available for 5,198 patients, of whom 176 died, yielding a crude mortality rate of 3.4%. The study population had a mean age of 27.3 years (SD 16.8, range 0-87 years). There were 1,655 pediatric patients (<18 years, 31.8%), 1,317 young adults (18-29 years, 25.3%), 1,603 middle-aged adults (30-49 years, 30.8%), and 578 older adults (≥50 years, 11.1%). The cohort was majority female (2,793 patients, 53.7%). There were 3,392 patients (65.3%) with HbSS/HbSβ0-thalassemia, while 1,592 patients (30.6%) had variant disease, and 214 patients (4.1%) had unknown type. Longitudinal analysis included 4,888 patients contributing 15,014 person-years of follow-up, with a mean follow-up of 3.1 (range 0.1-15.0) years per patient. The overall mortality was 11.7 deaths per 1,000 person-years.
Cause of death was documented in 159 of 176 deaths (90.3%). The most common cause of death was unknown (38 cases, 21.6%), then cardiac (36 cases, 20.5%), respiratory (18 cases, 10.2%), renal (14 cases, 8.0%), and SCD-specific complications (14 cases, 8.0%). Other causes attributed to ≥10 deaths were stroke/neurologic complications (13 cases, 7.4%), sepsis/infection (13 cases, 7.4%), and multi-organ failure (13 cases, 7.4%).
Mortality increased with age, and the average age at death was 45.9 years [SD 15.5]. The mortality rates among pediatric and adult populations were 1.3 and 15.2 deaths per 1,000 person-years, respectively. Sex was not significantly associated with mortality (females: 3.0% mortality rate; males: 3.5% mortality rate; χ²=0.87, p=0.351). Mortality was higher in HbSS/HbSβ0-thalassemia compared to variant disease (HbSS/HbSβ0-thalassemia: 3.8% mortality rate; variant: 2.8% mortality rate; χ²=7.3, p=0.027). Mortality was significantly associated with disease-modifying therapy (DMT) status (On DMT: 2.7% mortality rate; not on DMT: 4.4% mortality rate; χ²=10.4, p=0.001).
In multivariate Cox proportional hazards modeling, higher mortality risk was associated with increasing age (HR 1.05 per year, 95% CI: 1.04-1.07) and male sex (HR 1.50, 95% CI: 1.09-2.04), and lower risk with variant disease relative to HbSS/HbSβ0 (HR 0.64, 95% CI: 0.43-0.94). DMT use was not significantly associated with mortality (HR 0.89, 95% CI: 0.64-1.24).
Mortality rates varied substantially across sites, ranging from 0% to 5.8%. Among the 56 sites, 22 (39.3%) reported at least one death. Several sites with the highest rates had smaller cohorts, so a small number of deaths led to a high mortality rate; among sites reporting at least 10 deaths, the highest mortality rate was 2.1%.Overall, the GRNDaD registry from 2010 to 2026 demonstrated a similar mortality rate compared to surveillance in CA and GA from 2004 to 2008 (11.6 vs 10.1 deaths per 1,000 person-years). Both Pediatric and adult mortality rates were roughly 4 times higher than respective general population age groups. We do not see clear evidence of improvement in SCD mortality since the early 2000s, despite changes in clinical practice and treatment. The GRNDaD population may differ from surveillance programs in terms of disease severity and health care access - future research can examine this. Work to improve data collection on cause of death is ongoing. Analyses of factors associated with increased mortality risk, as well as variability in mortality across participating sites, will provide guidance for future quality improvement efforts.
