NT-proBNP screening found heart failure in one in four high-risk patients with diabetes

Design

Multicentre, unblinded, randomised controlled trial

Population

706 community adults ≥40 years with diabetes and ≥1 heart failure risk factor, Scotland

Primary outcome

Heart failure diagnosis at 6 months

Effect

24.6% vs 1%; OR 58 (95% CI 14–236); number needed to screen 5 (4–6)

TARTAN-HF was a multicentre, unblinded randomised trial in the west of Scotland. It enrolled 706 community adults aged 40 or over with type 1 or type 2 diabetes and at least one further heart failure risk factor, none with known heart failure. The screening group had NT-proBNP measured and echocardiography if it was 125 pg/mL or above; the control group had usual care. Median age was 71 and 90% had type 2 diabetes.

At six months heart failure, defined by the 2021 European Society of Cardiology criteria, had been diagnosed in 87 of 354 screened patients (24.6%) against 2 of 352 (1%) with usual care. The number needed to screen was 5. Most of the new cases had preserved ejection fraction, SGLT2 inhibitor use in the screened group rose from 24% to 39%, and those found to have heart failure had markedly worse self-reported health status than those without.

This is a trial of diagnostic yield. It shows that a great deal of symptomatic heart failure is sitting undiagnosed in the diabetes clinic; it does not yet show that finding it reduces admissions or deaths. Guidelines remain inconsistent on screening, and none has changed on the strength of this trial. Where natriuretic peptide testing is available, a lower threshold for checking it in older patients with diabetes and extra risk factors is reasonable now.

• Consider an NT-proBNP in patients with diabetes over 40 who carry another heart failure risk factor, such as hypertension, CKD or coronary disease.
• Arrange echocardiography for an NT-proBNP of 125 pg/mL or above, the threshold the trial used.
• Expect most new cases to have preserved ejection fraction, which a normal old echo does not rule out.
• A new heart failure diagnosis is a reason to start an SGLT2 inhibitor, independent of glucose control.
• Check NT-proBNP is affordable and accessible for your patients before making it routine; a raised value commits them to an echo.

Why it matters

If one in five screened patients has treatable heart failure, waiting for symptoms to be reported means missing most of it.

Don't overread it

The trial measured diagnoses at six months, not whether screening reduces hospital admission or death.

The statistics, in plain English

The odds ratio of 58 has a very wide interval (14 to 236) because only 2 cases were found with usual care; the more useful figure is the number needed to screen of 5 (4 to 6). The usual-care rate is low partly because those patients were not systematically tested, so the difference measures what screening uncovers, not new disease.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.