Related Experiment Video
Updated: Jun 18, 2026
Inverse Probability of Treatment Weighting (Propensity Score) using the Military Health System Data Repository and National Death Index
Published on: January 8, 2020
Optimizing NT-proBNP Inclusion Cut-offs for Randomized Clinical Trials in Heart Failure: Data from the Swedish Heart
Megan Schroeder1,2, Lars H Lund3,4, Christoph Gerlinger5,6
1Department of Clinical Science and Education, Södersjukhuset; Karolinska Institutet, Stockholm, Sweden.
Aims:
N-terminal pro-B-type natriuretic peptide (NT-proBNP) is widely used as an enrichment criterion in heart failure (HF) randomized controlled trials (RCTs), yet cut-offs vary. This study aims to provide evidence-based guidance on selecting NT-proBNP cut-offs to optimize the balance between event enrichment and screening failure across HF subgroups.
Methods:
Using the Swedish HF Registry (SwedeHF), we applied NT-proBNP cut-offs from prior RCTs (200-5000 pg/mL) and calculated 1-year incidence proportions of a composite CV outcome (CV death or first HF hospitalisation) across subgroups by ejection fraction (EF), care setting (inpatient/outpatient), atrial fibrillation (AF), chronic kidney disease (CKD), and obesity. We quantified point and relative increases in event proportions and potential screening failure at each cut-off and identified optimal prognostic thresholds by maximizing Youden's index.
Results:
Among 43,750 HF patients, median NT-proBNP was lower in HFpEF/HFmrEF vs HFrEF, outpatients vs inpatients, sinus rhythm vs AF, obese vs not obese, and patients without CKD vs with CKD (all p < 0.001). Higher NT-proBNP cut-offs increased 1-year composite CV proportions but excluded more patients. For example, In HFrEF outpatients, the 1-year proportion rose from 22.0% (no cut-off) to 24.3% at ≥600 pg/mL and 26.2% at ≥1,200 pg/mL, a 19% (95% CI:14.0-24.4) relative increase. Screening failure rose from 16.0% to 26.4% at these respective cut-offs. Optimal prognostic thresholds aligned with or exceeded subgroup median NT-proBNP values.
Conclusion:
Higher NT-proBNP cut-offs were associated with increased event enrichment but also higher screening failure. These findings support the use of higher cut-offs currently used in HF RCTs and suggest that future trials should tailor NT-proBNP cut-offs to trial aims, balancing enrichment with enrolment feasibility and considering obesity and CKD in addition to EF and AF.
