Echocardiography and guideline-directed medical care after abnormal outpatient B-type natriuretic peptide testing in those without pre-existing heart failure
W Douglas, P Austin, H Ross, X Wang, P Kavsak, N Ivers, P Tieu, K Quinn, C L Atzema, L Sun, C Alba, T Stukel, J Januzzi, D LeeAbstract
Background/Introduction
Natriuretic peptide (NP) tests are widely used for diagnostic evaluation of acute heart failure (HF). However, the associations of NP measurement with early echocardiography and downstream care in individuals without previously diagnosed HF has not been fully characterized.
Purpose
To characterize outcomes, the performance of echocardiography, and care after an initial abnormal NP test in outpatients without previously diagnosed HF.
Methods
A retrospective cohort study using administrative health databases was conducted. We identified outpatients without prior HF aged ≥40 years in Canada, who had a first abnormal NP-test (NT-proBNP >125 pg/ml or BNP >50 pg/ml) ordered by a family physician (FP) or non-cardiac specialist (NCS). Outcomes included the performance of subsequent echocardiography, and a composite of cardiovascular (CV) hospitalization or CV mortality. Hazard ratios for initiation of processes of care were estimated using multivariable Cox models adjusted for age, sex, NP ratio (modeled using restricted cubic spline terms), prior myocardial infarction, diabetes, atrial fibrillation, chronic kidney disease, chronic obstructive pulmonary disease, liver disease, peripheral artery disease, stroke, cancer, prior revascularization, hyperlipidemia, Charlson comorbidity index, serum creatinine, and material resources quintile, with echocardiography as a time-varying covariate. Follow-up continued until death, or 730 days following initial abnormal NP test.
Results
56,716 outpatients with abnormal NP tests of whom 35,117 (50% female, median [IQR] age 77 [68-84] years) had the index test ordered by their FP and 21,599 (52.1% female, age 74 [67-80] years) were ordered by an NCS. Cumulative incidence of the primary composite outcome (CV hospitalization or CV mortality) was higher in the FP cohort than the NCS cohort at 730 days (21.6% vs. 14.6%, p < 0.001). Performance of echocardiography within 14 days after an abnormal NP was 21.6% in the FP cohort and 12.9% in the NCS cohorts, increasing to 60.0% and 45.8% by 365 days. Among all patients with abnormal NP testing, echocardiography was associated with significantly higher rates of initiation of mortality-modifying HF pharmacotherapy, including ACEi/ARB/sacubitril-valsartan (HR 2.36; 95% CI 2.27-2.45), beta-blockers (2.72; 95% CI 2.61-2.83), MRAs (3.83; 95% CI 3.52-4.17), and SGLT2 inhibitors (3.92; 95% CI 3.65-4.22), as well as referral to cardiology (2.92; 95% CI 2.81-3.04).
Conclusion
In outpatients without pre-existing HF, abnormal NP values are associated with significant CV morbidity and mortality; despite this, rates of prompt echocardiography following an abnormal NP test were low. This gap is concerning given prompt echocardiography was associated with higher prescription rates of mortality-modifying medication, suggesting that delays in or underuse of echocardiography may be associated with reduced quality of HF care.Prevalence of ECHO Following Abnormal NP TestIncidence CV Hospitalization or CV Mortality