Exercise cardiopulmonary magnetic resonance imaging unmasks worsening cardiovascular haemodynamics in the spectrum of heart failure with preserved ejection fraction (HFpEF)

Abstract Background Diagnostic challenges in heart failure with preserved ejection fraction (HFpEF) result in underdiagnosis and undertreatment. Existing diagnostic algorithms (HFAPEFF and H2FPEF) attempt to dichotomise HFpEF versus non-cardiac dyspnoea, but leave a large proportion of patients’ exercise intolerance unexplained. This study aims to demonstrate patterns in exercise cardiovascular haemodynamics along the HFpEF spectrum using ergometer stress cardiac magnetic resonance imaging. Methods Patients with HFpEF, asymptomatic age-matched individuals (AMC), and patients with exercise intolerance considered at ‘intermediate risk of HFpEF’ (INT) were prospectively enrolled. Symptomatic participants were categorised by HFA-PEFF and H2FPEF diagnostic scores. All participants completed rest and 35W exercise cardiac magnetic resonance (eCMR), exercise echocardiography, lung ultrasound, 6-minute walk test (6MWT) and Kansas City Cardiomyopathy Questionnaires (KCCQ). Results 104 participants were recruited: 46 HFpEF, 19 AMC and 39 INT. Higher HFpEF risk scores were associated with a progressive decline in cardiac reserve (percentage change in cardiac output at 35W eCMR, %ΔCO, p0.001). Relative blunting of cardiac reserve was evident even in symptomatic patients deemed low-risk of HFpEF H2FPEF≤2 %ΔCO 84 (52, 106) vs. AMC 116 (92, 138), p=0.02. The INT group and AMC had comparable NTproBNP but 6MWT distance (6MWD) and KCCQ scores were lower in INT. Cardiac reserve correlated with 6MWD (r = 0.38, p0.001) and KCCQ score (r = 0.29, p=0.003), adjusted for age, height and body mass index (BMI). Higher HFpEF risk scores were also associated with greater exercise-induced pulmonary congestion (eiPC) on e-CMR (p=0.04) and lung ultrasound (p=0.003). The cohort with greater eiPC (PDM+) had higher LV mass, worse RV function, lower RV stroke volume augmentation, smaller increase in LV end-diastolic volume (%DLVEDV) and greater E/e’ increase during exercise. Conclusion Relative blunting of cardiac reserve is evident in exercise-intolerant patients, even when considered low-risk for HFpEF. Exercise CMR demonstrates important adverse exercise haemodynamic features such as cardiac reserve impairment and exercise pulmonary congestion in symptomatic patients regardless of an established HFpEF diagnosis, which are associated with worse functional capacity and quality of life. These findings challenge the present binary approach to HFpEF diagnosis. In future, recognition of a HFpEF spectrum may improve patient understanding and management.