Undetected heart failure in tertiary care: prevalence, predictors and outcomes

Abstract Background Undetected heart failure (HF) seems to be unexpectedly common in primary care. We hypothesized that undetected HF may also be common in tertiary care, even in care pathways including cardiologists. Methods We assessed prevalence, predictors, and outcome of undetected HF in a large prospective cohort study enrolling patients referred for myocardial perfusion single photon emission tomography (MPI-SPECT) with central adjudication of HF according to guidelines. Findings were validated in two independent prospective cohorts (outpatient cardiac rehabilitation program, major noncardiac surgery). Prevalence of undetected HF, ratio of undetected HF to known HF and number needed to screen (NNS) were the primary endpoints. Results Among 3623 patients, median age 68 years, 31% women. 100 patients (3%) had known HF and 928 patients (25.6%) were adjudicated to have undetected HF, resulting in a ratio of undetected to known HF of 9.3-to-1 and a NNS of 2.8. Older age, atrial fibrillation, and previous pacemaker implantation were strong independent predictors of undetected HF. Patients with undetected HF were at significantly higher risk of cardiovascular death or HF hospitalization (adjusted HR 1.47 [95%CI, 1.17-1.85] versus no HF). These findings were confirmed in both validation cohorts: among 970 and 1563 patients the prevalence of undetected HF was 21.5% and 19%, ratio of undetected HF to known HF 0.7-to-1 and 2.8-to-1, NNS 2.2 and 3.7, respectively. Conclusion The prevalence of undetected HF is high also in tertiary care, even higher with older age, atrial fibrillation, and previous pacemaker implantation, and associated with increased risk of cardiovascular death or HF hospitalization.Prevalence of undetected HFKaplan-Meier cumulative incidence curves