Is predicted heart failure risk considered when ordering NT-proBNP? Real-world practice in patients without diagnosed HF

G Yussupova, Vadim Medovchshikov, N Yeshniyazov, S Azhgaliyeva, M Balabayeva, Z H Iskakova, D Murzagaliyeva, B Duisenbayev, Gulnara Kurmanalina, B Zholdin

2 more authors

E Khasanova, Z H D Kobalava

2026

Closed access Metadata (CSL-JSON)
DOI 10.1093/ejhf/xuag193.745
Data confidence 70 medium
How this score is calculated

It is built from checkable properties of the record itself: whether the article has a DOI, how many independent sources confirmed its data, whether the required fields are filled in, and how recent the latest confirmation is. Disagreement between sources and a missing identifier lower it. A value of 80 or above means the record was published without further checking; 60 to 79 means it is published at reduced confidence, with a manual check advisable. A record verified by an editor by hand is never pushed below 90 by a recalculation. This scores our data about the article, not the article itself.

Abstract

Abstract Background NT-proBNP is central to HF diagnosis and may refine risk in patients without established HF. Whether clinicians preferentially order NT-proBNP in those at the highest predicted HF risk in routine practice is unclear. Purpose To assess whether 5-year HF risk (ABC-HF) is associated with NT-proBNP testing in hospitalized patients without diagnosed HF. Methods This single-center observational study included 723 patients aged ≥40 years with established CVD and hypertension who were enrolled at hospital admission for cardiovascular reasons. Patients with previously diagnosed HF were excluded (n=421), leaving 302 patients without diagnosed HF for analysis. The 5-year risk of incident HF was assessed using the Health ABC-HF Score and categorized as low/intermediate (0-5 points), and high/very high (≥6 points). Patients were grouped by NT-proBNP testing: Yes (n=85) vs No (n=217). Clinical characteristics, comorbidities, echocardiographic parameters, and metabolic profile were compared between groups. Results The high/very high HF risk was in 76.5% (231/302) patient while low/intermediate – in 23.5% (71/302). NT-proBNP was measured in 28.1% (85/302). Patients without NT-proBNP testing had a higher predicted HF risk: ABC-HF 9 (6; 11) vs 7 (4; 10) (p=0.006), and a higher prevalence of high/very high HF risk (80.2% vs 67.1%; p=0.016). Age and sex did not differ (Fig.1). The no-testing group had a heavier atherosclerotic burden (CVD 76.5% vs 51.8%, atherosclerotic CVD 69.1% vs 43.5%; all p<0.001) and a more adverse echocardiographic profile: lower LVEF (55 (50; 58) vs 60 (55; 60.25); p<0.001) with more frequent asymptomatic LVEF <50% (17.5% vs 1.2%; p<0.001). By contrast, the tested group more often had obesity (54.2% vs 37.7%; p=0.010), slightly higher SBP (138 (128; 148) vs 130.00 (120; 150); p=0.026), and higher HbA1c (5.8 (5.3; 7.2) vs 5.4 (5.0; 5.8); p<0.001) (Fig.1). Conclusions In real-world practice, NT-proBNP testing is not preferentially performed in patients with the highest predicted 5-year HF risk by the ABC-HF score. Paradoxically, patients at higher predicted risk, often with greater atherosclerotic burden and subclinical LV systolic dysfunction, were less likely to undergo NT-proBNP testing, suggesting missed opportunities for risk refinement and earlier HF detection and prevention. This pattern may reflect a "testing bias" in routine care, whereby clinicians preferentially order NT-proBNP when the pre-test probability of prevalent HF is high, particularly when HF with preserved ejection fraction is suspected, or in diagnostically uncertain patients presenting with dyspnea to support diagnosis and treatment decisions. However, predicted future HF risk appears to be underused when selecting patients for testing. Clear protocols for NT-proBNP assessment and interpretation in patients without diagnosed HF are warranted.Fig.1For image description, please refer to the figure legend and surrounding text.