Ventricular arrhythmias in sickle cell anemia

In this issue of Blood, d’Humières et al report on the frequency of ventricular arrhythmias in the largest prospective registry of patients with sickle cell anemia (SCA) undergoing 24-hour Holter monitoring along with same-day multimodality cardiopulmonary testing.¹ The authors observed 22% of their cohort experienced ventricular arrhythmias, as defined by European guidelines. Within this subset, 64% exhibited a moderate burden of premature ventricular complexes and 41% presented with nonsustained ventricular tachycardia (VT).

Sudden death is an increasingly reported cause of early death in patients with SCA.² It was historically attributed to narcotic overdosing.³ Grossly reduced left ventricular (LV) systolic function, obstructive coronary artery disease, and cardiac iron deposition are rarely reported in patients and do not seem to contribute to their cardiovascular risk.^4,5 Little, however, is known about possible links between arrhythmic pathologic features to sudden death in SCA, highlighting a poorly characterized and potentially underrecognized health risk for these patients. Although recent work has confirmed increased vulnerability for the development of ventricular arrhythmias in “humanized” sickle mouse models,^6,7 their prevalence in patients with SCA remains unknown. The current work by d’Humières et al begins to close this knowledge gap as their data cumulatively suggest a high prevalence of ventricular arrhythmias in patients with SCA.

Over a series of articles, we with others have reported on the development of sickle cell cardiomyopathy (SCC), characterized by the development of increased cardiac mass and fibrosis,⁵ diastolic dysfunction complicated by heart failure with preserved ejection fraction,⁴ and prolonged repolarization,⁸ characteristics similar to hypertrophic cardiomyopathy (HCM).^5,9 In contrast to HCM, where risk factors for VT and sudden cardiac death are well characterized,¹⁰ origins and risk factors for arrhythmias in SCA remain elusive. More importantly, to begin to address this gap, the current work found that male sex, a low platelet count, and a less negative global longitudinal LV strain percentage significantly increased the risk for ventricular arrhythmias, with trends toward an elevated N-terminal pro-brain natriuretic peptide (NT-pro-BNP) ≥160 ng/L and a lower ratio of peak velocity blood flow from left ventricular relaxation in early diastole (the E wave) to peak velocity flow in late diastole caused by atrial contraction (the A wave; E/A ratio) on echocardiography. These preliminary data underscore the potential pathologic importance of subtle LV systolic dysfunction and a novel sex disparity in SCA. They emphasize that, although ejection fraction is reportedly normal in most patients, mild forms of LV systolic dysfunction may be more prevalent and associated with proarrhythmic consequences. This implication is further supported by trends with NT-pro-BNP and E/A levels, both markers for cardiac dysfunction as well as for all-cause mortality in SCA. Although inflammation status was not evaluated in the current study, it is a key driver of SCA pathologic features and prior work has suggested interleukin-18, specifically, as a possible biomarker of SCC and VT.⁶ Finally, although basic parameters of hemolysis were not associated with the presence of ventricular arrhythmias, the association with lower platelet counts raises questions about the need for additional, larger studies to understand sickle cell–specific determinants of cardiac arrhythmias.

Caution, however, needs to be emphasized when interpreting the results. The current work was limited in several important ways. First, the subjects reflect a biased population of patients, all referred for cardiopulmonary issues within a single institutional cohort. Given the absence of an independent cohort for validation, these data fuel the need for larger observational studies measuring the prevalence in all patients with SCA. They also raise the importance of detecting for arrhythmias across different settings, such as during a vaso-occlusive or hemolytic crisis. Although the average age of the cohort was older, these data remain relevant given the average life expectancy plateaus in the fifth decade. In addition, a significant proportion of patients were on β-blockers and antiarrhythmic drugs, which may confound the “true” prevalence of ventricular arrhythmias.

Critically, the current work was not designed for longitudinal assessment of outcomes, such as cardiovascular-specific mortality. Nonetheless, given the robust association of ventricular arrhythmias to mortality and sudden cardiac death in a spectrum of cardiomyopathies, these data, for the first time, implicate a new and potentially high-risk cardiovascular phenotype in patients with SCA. They also highlight potential deficiencies in our current approach in management of SCA, suggesting the need for careful cardiac risk factor evaluation in patients.

Conflict-of-interest disclosure: The author declares no competing financial interests.