AUTHORS
Yasmina Hadjimi, MD, Haytham Derbel2, Thibaut Moulin3, Laurent Savale, MD, PhD4, Gonzalo De Luna, MD5, Laurent Boyer, MD, PhD6, Anoosha Habibi, MD/ Associated Professor7, Yosr Zaouali8, Anne-Laure Pham Hung D'Alexandry D'Orengiani9, Anais Lamadieu10, Nicolas Lellouche, MD, PhD11, Pablo Bartolucci, MD, PhD12, Thomas d'Humieres, MD, PhD13
AFFILIATIONS
1 Henri Mondor Teaching Hospital, Créteil, France
2 Henri Mondor Teaching Hospital, Créteil, Finland
3 Henri Mondor Teaching Hospital, Créteil, France
4 Bicêtre Teaching Hospital, Le Kremlin Bicêtre, France
5 Henri Mondor Teaching Hospital, Créteil, France
6 Henri Mondor Teaching Hospital, Créteil, France
7 Henri Mondor Teaching Hospital, Créteil, France
8 Henri Mondor Teaching Hospital, Créteil, France
9 Henri Mondor Teaching Hospital, Créteil, France
10 Henri Mondor Teaching Hospital, Créteil, France
11 Henri Mondor Teaching Hospital, Créteil, France
12 Henri Mondor Teaching Hospital, Créteil, France
13 Henri Mondor Teaching Hospital, Créteil, France
ABSTRACT
Background: Ventricular arrhythmias (VA) are increasingly recognized in sickle cell disease (SCD) and may contribute to premature mortality. In most cardiomyopathies, VAs arise from a fixed structural substrate, mainly myocardial fibrosis, readily identifiable by cardiac magnetic resonance (CMR). Whether this paradigm applies to SCD remains unknown.
Aims: To determine whether ventricular arrhythmias in SCD are associated with myocardial structural abnormalities detectable by CMR or instead reflect a predominantly functional cardiac phenotype.
Methods: We analyzed data from the prospective DREPACOEUR cohort, including adult patients with SS or Sβ° genotypes who underwent comprehensive cardiovascular evaluation at steady state. Among the cohort, patients who had available cardiac magnetic resonance (CMR) imaging were selected and subsequently matched in a 1:1 fashion based on the presence or absence of ventricular arrhythmias (VA), to allow balanced group comparisons. VA were defined on 24-hour Holter monitoring (frequent premature ventricular contractions or ventricular tachycardia according to European guidelines). CMR was used to assess myocardial structure and tissue characterization, including ventricular volumes and ejection fraction, native T1 and T2 mapping, extracellular volume (ECV), and late gadolinium enhancement (LGE). Echocardiographic and biological parameters were integrated to compare functional and structural cardiac phenotypes between groups.
Results: Seventy patients were included (median age, 45 years), of whom 34 (49%) fulfilled criteria for VA. Baseline demographic characteristics, SCD therapies, hemoglobin levels, and conventional markers of chronic hemolysis (LDH and bilirubin) were comparable between groups, despite a tendency of higher HbS in VA group (p=0.01). In contrast, patients with VA exhibited lower platelet counts and higher NT-proBNP levels, together with a distinct functional cardiac phenotype characterized by greater left ventricular mass, impaired myocardial deformation (lower global longitudinal strain), reduced diastolic relaxation (lower lateral e′ velocity), and lower CMR-derived left ventricular ejection fraction (50% vs 57%; p=0.011). Despite these functional differences, myocardial tissue characterization remained remarkably similar. Native T1 mapping (1050 vs 1030 ms; p=0.13), T2 mapping (53 vs 53 ms; p=0.91), extracellular volume (33% vs 32%; p=0.66), and late gadolinium enhancement (6.7% vs 3%; p=0.51) showed largely overlapping distributions irrespective of arrhythmic status. Likewise, ventricular ectopic burden correlated with left ventricular systolic function (LVEF: r = −0.29; p=0.017), but not with CMR markers of diffuse fibrosis or myocardial inflammation.
Interpretation: Our findings support a disease-driven rather than scar-driven model of ventricular arrhythmogenesis in SCD. While patients with VA exhibited clear evidence of cardiac dysfunction, myocardial tissue characterization remained remarkably similar irrespective of arrhythmic status. These observations suggest that ventricular arrhythmias primarily reflect systemic disease activity and transient myocardial stress rather than irreversible structural heart disease detectable by current CMR techniques. This interpretation is further supported by recent clinical observations of prompt severe VA resolution following curative therapies such as hematopoietic stem cell transplantation (PMID: 41058956).
Conclusion: Our findings challenge the conventional paradigm of scar-mediated ventricular arrhythmogenesis. We suggest that ventricular arrhythmias in SCD arise within a dynamic functional myocardial phenotype that is not captured by current CMR tissue characterization, highlighting the need to investigate systemic and electrophysiological mechanisms beyond myocardial fibrosis.
DISCLOSURES
Thomas d'Humieres, MD, PhD: No relevant conflicts of interest to declare.