Systolic blood pressure, mean arterial pressure and incident mortality in heart failure in a West-African cohort: the blood pressure paradox revisited

Background The prognostic relevance of blood pressure in Africans with heart failure (HF) has not been well investigated despite the different trajectories of HF in Africans compared to other ethnicities. We investigated the shape and strength of associations of systolic blood pressure (SBP) and mean arterial pressure (MAP) with incident death in Nigerian Africans with hypertensive HF. Results Present analyses include 118 participants (mean age 57.6 years, 55% men) with ambulatory hypertensive HF. Over a median 2.2 years of follow-up, there were 74 incident deaths. There was an inverse log-linear association between each of SBP and MAP, and incident death throughout the distribution of blood pressure studied and no J-shape effect was seen. Individuals in the bottom distribution of SBP (mean 100 mmHg) and MAP (mean 77 mmHg) had the highest cumulative incident death. Each 10 mmHg higher SBP and MAP was associated with lower risk of incident death (aHR 0.80 [95% CI: 0.70-0.91, p < 0.001] and aHR 0.78 [95% CI: 0.66-0.92, p < 0.01] respectively) which persisted despite adjustment for NHYA class, left ventricular function and systolic pulmonary artery pressure. The blood pressure paradox was partially explained by haematocrit, serum sodium, chloride and renal function (aHR SBP: 0.86 [95% CI: 0.73-1.00, p = 0.05] and aHR MAP: 0.88 [95% CI: 0.72-1.08, p = 0.22]). Conclusions There was a graded inverse dose-response association between each of SBP, MAP and incident death in hypertensive HF which was not fully explained by haemodynamic and clinical factors.