Background: In sub-Saharan Africa as much as 60% of children with sickle cell disease (SCD) die before the age of five. Despite high mortality rates among children with SCD in sub-Saharan Africa, there is little understanding of specific causes of death in these children. Prior studies assessing causes of death among children with SCD have largely relied on clinical diagnoses and have not utilized postmortem examinations to determine definitive causes of death. Our objectives were to describe causes of death among children with SCD and children with sickle cell trait who died.

Methods: We conducted a descriptive analysis using data collected in a prospective childhood mortality surveillance network (Child Health and Mortality Prevention Surveillance [CHAMPS]) from 2016 through 2024. CHAMPS conducts surveillance for stillbirths and deaths in children aged <5 years in healthcare facilities and communities in high childhood mortality regions in six sub-Saharan African countries, Bangladesh, and Pakistan. CHAMPS collects antemortem clinical data, interviews families using a standardized verbal autopsy form, and conducts microbiological and histopathological postmortem examinations using minimally invasive tissue sampling. These data are then reviewed by an expert panel at each site to determine causes of death. For this study, blood samples were tested for SCD and sickle cell trait using the Gazelle® Hb Variant point-of-care (POC) test (Hemex Health, Portland, Oregon, USA), which is a miniaturized version of the reference standard cellulose acetate electrophoresis. We restricted our analyses to infants and children aged 3-59 months who enrolled in CHAMPS because high concentration of fetal hemoglobin affects the accuracy of the Gazelle® Hb Variant POC test. We used descriptive statistics for demographics, proportions, and causes of death. Comparisons were made using chi-square testing and the Mann-Whitney U test where appropriate.

Results: CHAMPS enrolled 719 children aged 3-59 months who were tested for SCD and had cause of death determined during the study period. Of these, 36 (5.0%) tested positive for SCD, 112 (15.6%) tested positive for sickle cell trait, 18 (2.5%) had indeterminate results, and 553 (76.9%) had normal hemoglobin. The median age of cases with SCD was 19 months (interquartile range 13, 34 months) and 53.3% were male. CHAMPS sites in Kenya and Sierra Leone contributed the largest proportion of cases of SCD (n=20, 55.6% and n=10, 27.8%, respectively) and sickle cell trait (n=49, 43.8% and n=46, 41.1%, respectively). Of deaths with SCD, only 8.3% had documentation of an antemortem diagnosis of SCD. The most common underlying causes of death among cases with SCD were malaria (22.2%), malnutrition (16.7%), and sepsis (5.6%). For cases that tested positive for sickle cell trait, the most common underlying causes of death were malnutrition (25.0%), malaria (13.4%), and HIV (11.6%). The most common bacterial pathogens identified postmortem among cases with SCD and sickle cell trait were Streptococcus pneumoniae (13.9% and 11.6% of all cases, respectively), Klebsiella pneumoniae (5.6% and 12.5% of all cases, respectively), and Salmonella species (5.6% and 3.6% of all cases, respectively). We found no difference in the number of causes of death among cases that were positive for SCD (median 2, IQR 1, 2) compared to those with sickle cell trait (median 2, IQR 1, 3), and those who had normal hemoglobin (median 2, IQR 1, 3; P=0.33).

Conclusions: SCD and sickle cell trait were commonly identified postmortem in a large, multi-country childhood mortality surveillance network, yet antemortem diagnosis of SCD was rare among these cases. Malaria was a common cause of death among children both with SCD and sickle cell trait. Enhanced screening for SCD in sub-Saharan Africa may allow for early initiation of prophylaxis, treatment and potential prevention of mortality.

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