Contributor: Martin Chacon
Does an increased NT-proBNP in patients with atrial fibrillation (AF) and HFpEF correspond to a higher risk for worse outcomes? Kristensen et al suggest increased NTproBNP levels do not show the same correlation with poor outcomes in patients with AF compared with those without AF. NT-proBNP is increased in AF patients independently from heart failure, therefore, its capacity to predict cardiovascular death or hospitalizations in HFpEF may be diminished.
A pool of 3835 HFpEF patients from two clinical trials (3479 from I-Preserve and 356 from TOPCAT) were analyzed, of which 719 (19%) had AF. These patients were >50 years old, had an LVEF>45%, HFpEF diagnosis (by symptomatology, recent hospitalization, or increased NT-proBNP), and a baseline NT-proBNP. The primary outcome was the time to the first occurrence of the composite of cardiovascular-related death or HF hospitalization. Four NT-proBNP groups were analyzed, <400 pg/ml, 400-999 pg/ml (reference group), 10000-1999 pg/ml, and >2000 pg/mL. Overall, the rate of outcomes of interest were non-significantly higher in AF-HFpEF patients (11.8 vs 6.1 events per year, p=0.077) compared to HFpEF patients without AF. The rate of each of the outcomes of interest were broadly similar in the two middle groups, but different in the lowest and highest. In patients with the highest NT-proBNP levels, both outcomes were more frequent in patients without AF; whereas, the reverse was seen in patients with the lowest NT-proBNP levels. The most significant difference between AF and non-AF patients was seen in the lack of association between NT-proBNP and the risk of HF hospitalization in the multivariable analysis.
Limitations: Few AF patients had NT-proBNP <400pg/mL (9%) compared to 60% of patients without AF. Patients with AF were older, had a better overall NYHA class profile, were more likely to have a history of HF hospitalization but less likely to have a history of myocardial infarction. The authors also excluded TOPCAT patients outside the Americas.