Analysis of a large cohort of African/Afro-Caribbean patients by cardiac magnetic resonance

Abstract

Abstract Funding Acknowledgements Type of funding sources: None. Background Limited data is available in the literature on Cardiac Magnetic Resonance (CMR) features of African/Afro-Caribbeans, in particular exploring differences between hypertensive heart disease (HHD) and hypertrophic cardiomyopathy (HCM). Purpose To describe the cardiac morphology and pattern of late gadolinium enhancement (LGE) in patients of African/Afro-Caribbean origin. Methods We retrospectively analysed African/Afro-Caribbean patients who underwent clinical CMR at a tertiary centre. Three groups were specifically investigated: HHD, HCM and combination of HHD and HCM or ambiguous (HHD&HCM). Results Overall, 166 consecutive patients (58% male, mean age 55 ± 14yo) were analysed. One-hundred fifty-four (93%) had history of arterial hypertension (HTN), including 17 with uncontrolled/malignant HTN. Overall, 28(17%) had normal scans, 70(42%) HHD, 15(9%) apical HCM, 10(6%) classical septal HCM, 19(11%) dilated cardiomyopathy, 6(4%) cardiac amyloidosis, 3(2%) ischemic heart disease, 4(2%) myocarditis, 2 sarcoidosis and 1 valvular disease. In 7(4%) the diagnosis was ambiguous between HHD and HCM and 1 uncertain. Forty-four (27%) had dual pathology, most frequently HHD, bystander myocardial infarction (MI) and embolic MI. LGE was detected in 95(57%), 26 with ischemic pattern, 13 diffuse mid-wall, 29 focal non-ischemic and 28 diffuse/multifocal non-ischemic. CMR features and correlations between subgroups are reported in the Table. Patients with HHD had significantly higher left ventricular (LV) end-diastolic (ED) volume indexed (Vi) and LV end-systolic (ES) Vi, but lower LV ejection fraction (EF) and LV maximum wall thickness (MWT) compared to HCM patients. HHD&HCM had higher LVEF and MWT compared to HHD and higher LVEDVi compared to HCM. LGE was more frequently seen in HCM and HHD&HCM as focal non-ischemic (6vs5vs10,p = 0.049) and diffuse multifocal(5vs6vs9,p = 0.009). A history of uncontrolled/malignant HTN was more frequent in HHD and HHD&HCM (11vs1vs5,p = 0.025) and was associated with diffuse LGE with lateral wall involvement (p < 0.0001) (Figure). Conclusions: CMR findings in African/Afro-Caribbeans may overlap between aetiologies. A specific pattern of diffuse non-ischaemic LGE involving the lateral wall appears though to be more often associated with severe uncontrolled HTN. Table HHD(n = 70) HCM(n = 19) HHD&HCM or ambiguous (n = 13) p LVEDVi, ml/m2 88 ± 33 62 ± 11 77 ± 6 0.003 LVESVi, ml/m2 36 ± 26 16 ± 10 26 ± 19 0.002 LVEF, % 62 ± 16 79 ± 6 74 ± 12 0.567 LVMWT, mm 13 ± 2 15 ± 4 14 ± 3 <0.0001 CMR features and correlations between subgroups Abstract Figure