Sudden Death in Sickle Cell Disease: An Assessment of Risk Factors

Background

Sickle cell disease (SCD) is the most common severe hereditary blood disorder in the United States. A single nucleotide mutation leads to a modified β-chain of hemoglobin (β₆Glu-Val). The pathology of SCD includes debilitating pain crises, progressive organ damage, and finally premature death. A significant portion of premature death in adults with SCD is attributable to cardiopulmonary disease with a disproportionate amount of sudden death. Sudden death has been associated with as much as 23.4% of mortality in adults with SCD (Darbari et al. Am J Haematology, 2006). In this study we aimed to identify risk factors for sudden death. As in previous studies, sudden death was defined as an unexpected death occurring in a relatively healthy SCD patient who dies at home or within 24 hours of hospitalization with or without a vaso-occlusive crisis. Identification of risk factors for sudden death will aid in developing interventions targeted at specific SCD patients.

Methods

We conducted a retrospective study of SCD-related deaths using the Partners Research Patient Data Registry (RPDR), a centralized clinical data registry that gathers clinical information from various hospitals within the Partners hospital systems and patients' electronic health records. The study was approved by our institution's IRB. The RPDR was queried to identify all patients associated with SCD. Only patients with a vital status of 'deceased' were used for the patient population. Patients who died before 1998 were excluded. Sixty-one patients were included and categorized by 'cause of death' into 5 groups. The patients' charts were reviewed to identify cause of death and SCD-related data between two years and up to 2 weeks prior to death. The patient variables examined covered demographics, sickle genotype, hydroxyurea exposure, blood pressure, EKG abnormalities, medical histories of acute chest syndrome (ACS), stroke, pulmonary embolism, leg ulcers, priapism, and retinopathy as well as laboratory data. The data were analyzed in 2 categories; 'sudden death' and 'other causes of death'.

Results

Of the 61 deceased patients, 33 (54%) were women and the average age at death was 39 years. Nineteen patients (31%) suffered sudden deaths: 7 (11%) sudden deaths at home (group I) and 12 (20%) deaths within 24 hours of hospitalization (group IV). Ten (16%) died of known causes related to SCD (group II), 22 (36%) died of known causes unrelated to SCD (group III) and 10 (16%) died of unknown causes (group V). Of the 19 patients who died suddenly (groups I and IV), 14 (74%) had a history of ACS and 17 (89%) had exposure to hydroxyurea. In comparison, of the patients who died of causes other than sudden death, only 15 (36%) had a history of ACS (p = 0.012) and only 18 (43%) had been exposed to hydroxyurea (p = 0.0008). A history of priapism was found in 4 of 10 (40%) men who suffered a sudden death and 0 of 18 (0%) men who died of other causes (p = 0.012). The frequencies of other measured variables, including dysrhythmias and QTc interval prolongation, were not markedly different between the groups.

Discussion

Sudden death remains a significant problem in SCD. Over 30% of deaths in this study were associated with sudden death. Our data show a higher rate of ACS and priapism in SCD patients who suffered a sudden death compared to those who died of other causes. Likewise, a history of hydroxyurea use was more likely in patients who suffered sudden death. Hydroxyurea use, probably a marker of severe disease, did not protect against the risk of sudden death in this small cohort of patients. These findings may be in keeping with a recent meta-analysis of 3257 SCD patients which was unable to confirm a mortality benefit with use of hydroxyurea in SCD (Maitra et al. Haematologica 2017).

Conclusion

Our results suggest that ACS and priapism may be indicative of a high risk of sudden death in patients with SCD. Our data are preliminary and need to be further investigated using a larger cohort of patients, as well as exploring the prevalence of these factors in SCD patients who are not yet deceased.