Incidence of heart failure in patients with coronary heart disease and stroke in a large Chinese cohort study

Abstract Background Heart failure (HF) is associated with remarkably elevated mortality and cardiovascular disease (CVD) is a leading cause of HF. Many studies have reported the prevalence of comorbidity of coronary heart disease (CHD) and HF or short-term incidence using registry or cross-sectional data, the long-term incidence of HF among patients after the first cardiovascular event is quite limited, particularly in patients with stroke. Updated incidence estimates are crucial to understanding the burden of illness of HF and developing effective preventive strategies. Purpose To compare the long-term incidence of HF among different subtypes of CVD, i.e. CHD or stroke (with or without hypertension at baseline), under a consistent framework using large-scale population-based data. Methods Using data from the CHERRY study, a population-based cohort study linked with electronic health records in China, we identified all patients≥18 years and newly diagnosed CHD or stroke during 2010-2016, and sex- and age-matched controls without prior CVD in a 1:5 ratio. All patients and controls were followed up until incident HF (at least 1 inpatient HF claim or at least 2 outpatient HF claims or 2 prescriptions with digoxin), non-HF death or end of follow-up (31st May 2021). Crude incidence of HF was compared among patients and controls by sex or age. The cumulative incidence rate of HF was calculated with non-HF death as a competing event. Incidence rate ratios (IRRs) were calculated using Poisson regression adjusted for sex and age. Results We finally identified 14,682 CVD patients (mean [SD] age 71.0[12.4] years, 54.0% men, 3730 CHD and 11,238 stroke cases) and 73,410 controls. During the follow-up (median 4.46 years for patients and 8.96 years for controls), the incidence rate of HF was 27.90(95%CI: 26.6-29.2) per 1000 person-years for CVD patients, 22.36(95%CI:21.03-23.74) for stroke, 45.10(95%CI:41.91-48.48)for CHD and 45.20(95%CI:37.92-53.47)for MI, significantly greater than 8.10(95%CI:7.87-8.33) for controls. The incidence rate was similar in men and women and increased proportionally with age (P for trend <0.05). Considering the competing risk, the 10-year cumulative incidence was 15.49% for CVD patients, 12.48% for stroke, 25.02 % for CHD, and 21.34% for patients with MI. Moreover, 81.3% of CHD patients and 79.3% of stroke patients had a history of hypertension at baseline and higher IRRs than those without hypertension, respectively. Although stroke patients were at relatively lower risk of HF among CVD subtypes, hypertension diagnosis would increase the risk by 60% (IRRs, 1.66, 95%CI:1.36-2.02). Conclusion The incidence of HF varied among patients with different subtypes of CVD, with the highest in those with CHD. Stroke patients had a higher risk of HF than the general population, and comorbidity of hypertension might add an additional risk of HF, highlighting the significance of blood pressure management in preventing heart failure.Incidence rate for patients and controlsIncidence rate of HF by disease status