Discordance between N-terminal pro-B-type natriuretic peptide and left ventricular end-diastolic pressure in suspected heart failure with a preserved ejection fraction: implications for diagnosis and prognosis

BACKGROUND: Diagnosing heart failure with preserved ejection fraction (HFpEF) remains challenging, as natriuretic peptides and resting haemodynamic measurements reflect complementary but incomplete aspects of myocardial stress and filling pressure. METHODS: In this prospective all-comer cohort (UKSH-Trial registration number AZ-D412-/21), adults undergoing elective left heart catheterisation underwent simultaneous invasive left ventricular end-diastolic pressure (LVEDP) measurement and N-terminal pro-B-type natriuretic peptide (NT-proBNP) sampling. The H₂FPEF score was calculated in patients with preserved ejection fraction (≥50%), and those with intermediate or high probability were included. Patients were classified using guideline-recommended NT-proBNP thresholds and LVEDP ≥16 mm Hg into four groups: normal (Group 1), isolated LVEDP elevation (Group 2), isolated NT-proBNP elevation (Group 3) and combined elevation (Group 4). The primary endpoint was all-cause mortality, analysed using multivariable Cox models, including age, sex, renal dysfunction and H2FPEF risk category. RESULTS: Among 514 participants (mean age 70 years, 49% women), group distribution was 29%, 14%, 28% and 29%. Discordance was common (42%). Clinical profiles aligned more closely with NT-proBNP than LVEDP. Compared with Group 1, adjusted mortality was not significantly higher in Group 2 (HR 1.34, 95% CI 0.45 to 4.02), whereas it was significantly higher in Group 3 (HR 2.26, 95% CI 1.02 to 5.01) and Group 4 (HR 3.33, 95% CI 1.54 to 7.20). CONCLUSIONS: NT-proBNP and LVEDP are frequently discordant and provide complementary prognostic information in suspected HFpEF. Mortality risk was highest when both were elevated and was also increased with isolated NT-proBNP elevation, whereas isolated LVEDP elevation was not associated with excess mortality. These findings support integrated biomarker-haemodynamic assessment and warrant prospective validation.