An NT-proBNP-based screening strategy significantly increased the diagnosis of heart failure at 6 months compared to usual care (24.6% vs 1%; OR 58; 95% CI 14-236; P<0.001).
RCT (n=706)
unblinded
randomized
Yes
Does an N-terminal pro-B-type natriuretic peptide screening strategy improve the identification of unrecognized heart failure in high-risk patients with diabetes?
An NT-proBNP-based screening strategy in high-risk patients with diabetes significantly increased the identification of unrecognized heart failure, predominantly HFpEF, compared to usual care.
Odds Ratio: 58 (95% CI 14–236)
Absolute Event Rate: 24.6% vs 1%
Number Needed to Treat: 5
p-value: p=< 0.001
BACKGROUND: Heart failure (HF) is a common cardiovascular complication of diabetes. Early identification of unrecognized HF enables initiation of disease-modifying therapies. International diabetes and HF clinical practice guidelines are inconsistent with respect to recommendations for screening for HF in people with diabetes. OBJECTIVES: The purpose of this trial was to evaluate the diagnostic yield of a natriuretic peptide-based HF screening strategy in people with diabetes and HF risk factors. METHODS: TARTAN-HF (Targeted Assessment in high-Risk paTients with diAbetes to ideNtify undiagnosed HF) was a prospective, multicenter, unblinded, randomized controlled trial. Community-based participants aged ≥40 years with type 1 or type 2 diabetes and at least 1 additional HF risk factor, but without known HF, were recruited in the west of Scotland. Participants were randomized to N-terminal pro-B-type natriuretic peptide screening (with echocardiography if N-terminal pro-B-type natriuretic peptide ≥125 pg/mL) or usual care. The primary outcome was a HF diagnosis at 6 months. HF was defined according to 2021 European Society of Cardiology Heart Failure guidelines. Secondary and exploratory outcomes included HF phenotype, sodium-glucose cotransporter 2 inhibitor use, and a composite of HF hospitalization or death. RESULTS: From January 11, 2023 to May 29, 2025, 706 participants were randomized (354 screening; 352 usual care). Median age was 71 years, 69% were male, and 90% had type 2 diabetes. Median duration of diabetes was 12.5 years. Two HF risk factors were present in 26% and 3 or more risk factors in 10%. At 6 months, HF was diagnosed in 87 (24.6%) in the screening group vs 2 (1%) in usual care (OR: 58; 95% CI: 14-236; P < 0.001). Most cases were HF with preserved ejection fraction (23%). The number needed to screen was 5 (95% CI: 4-6). Sodium-glucose cotransporter 2 inhibitor use increased from 24% to 39% in the screening group. Patients found to have unrecognized HF had significantly impaired health status as assessed using the Kansas City Cardiomyopathy Questionnaire-12 overall summary score compared with those without HF (median: 65.6 vs 89.6). CONCLUSIONS: A HF screening strategy identified a large proportion of patients living with diabetes with unrecognized symptomatic HF (mostly HF with preserved ejection fraction). Screening strategies for HF should be further investigated and considered for adoption in the care of people with diabetes. (Targeted Assessment in high-Risk paTients with diAbetes to ideNtify undiagnosed Heart Failure: NCT05705869).
“The ever-growing armamentarium of disease-modifying HFpEF treatments and the high prevalence of unrecognized HFpEF detected through targeted screening in TARTAN-HF highlight not only the need for the ongoing discovery of novel preventative and disease-modifying therapies, but also the value of th...”
The TARTAN-HF trial demonstrates that NT-proBNP screening in high-risk diabetes patients can effectively identify a significant number of previously undiagnosed heart failure cases.
Sweet et al. (2026) conducted an RCT in Diabetes and heart failure risk factors (n=706). N-terminal pro-B-type natriuretic peptide screening vs. Usual care was evaluated on Heart failure diagnosis at 6 months (OR 58, 95% CI 14-236, p=< 0.001). An NT-proBNP-based screening strategy significantly increased the diagnosis of heart failure at 6 months compared to usual care (24.6% vs 1%; OR 58; 95% CI 14-236; P<0.001).