HFpEF under the 2026 ESC classification — structured viva

Stem

Practice scenario. A 70-year-old woman with BMI 36 kg/m², treated hypertension and permanent AF has 4 months of exertional dyspnoea. LVEF is 58% with no previous echo; NT-proBNP is 230 pg/mL.

Branch A — Classification

Examiner: Before we discuss her, a colleague's patient has symptomatic HF with LVEF 46%, labelled "HFmrEF". What would the 2026 ESC call it?

Strong answer: HFrEF. The 2026 Task Force eliminated HFmrEF, and HFrEF now covers LVEF under 50% with symptoms and/or signs.^[1] The ESC reasons that HFmrEF resembles HFrEF in characteristics, outcomes, treatment response and pathophysiology.^[1] HFpEF needs all three: current or prior symptoms and/or signs of HF; LVEF of 50% or more, never previously under 50%; and objective evidence of structural and/or functional abnormalities consistent with LV diastolic dysfunction or raised LV filling pressures, supported by raised natriuretic peptides.^[1]

Branch B — Diagnosis when the peptide is borderline

Examiner: Her NT-proBNP is 230 pg/mL. Does that exclude heart failure?

Strong answer: No. The suggested outpatient age-adjusted level (HFA-ESC consensus) above which HF is likely at 50–75 years is 250 pg/mL or more. Obesity may lower natriuretic peptide levels, so thresholds need critical evaluation in obesity.^[1] I would apply Table 10: LV mass index ≥95 g/m² (female) or ≥115 g/m² (male), or RWT above 0.42; LA volume index above 40 mL/m² in AF; E/e′ above 9 at rest; and PASP above 35 mmHg or TR velocity at rest above 2.8 m/s.^[1] If still uncertain, CPET, exercise stress echo or invasive haemodynamics may be considered. Exercise PCWP of 25 mmHg or more is generally diagnostic.^[1] Scores estimate likelihood: an H2FPEF score below 2 means low and 6 or more high likelihood.^[2] In a retrospective case-control study of outpatients with unexplained dyspnoea, where HFpEF was confirmed by elevated exercise PCWP, H2FPEF had a higher AUC than HFA-PEFF (0.845 vs 0.710).^[16]

Branch C — Foundational therapy

Examiner: HFpEF is confirmed. What do you start, and what does it achieve?

Strong answer: An SGLT2 inhibitor and an MRA, each Class I, level A, for symptomatic HF independent of LVEF, to reduce the risk of HF hospitalisation or CV death.^[1] DELIVER (LVEF >40%): dapagliflozin 10 mg reduced worsening HF or CV death (16.4% vs 19.5% over a median 2.3 years; HR 0.82).^[3] EMPEROR-Preserved (LVEF >40%): empagliflozin 10 mg reduced CV death or HF hospitalisation from 17.1% to 13.8% over a median 26.2 months (HR 0.79), mainly through fewer HF hospitalisations.^[4] FINEARTS-HF (LVEF ≥40%): finerenone reduced total worsening HF events plus CV death (rate ratio 0.84).^[6] For SGLT2 inhibitors the ESC states that CV mortality fell only in HFrEF, so in HFpEF the benefit is driven by HF hospitalisation. For MRAs, mortality fell only in HFrEF; HFrEF trials used all-cause mortality and HFpEF trials CV mortality.^[1]

Branch D — Spironolactone and ARNI

Examiner: Why does the guideline accept spironolactone when TOPCAT missed its primary endpoint, and where does ARNI sit?

Strong answer: TOPCAT missed its primary endpoint (HR 0.89; P=0.14) and doubled hyperkalaemia (18.7% vs 9.1%).^[7] The ESC MRA recommendation for HFpEF rests on a meta-analysis of spironolactone and finerenone trials.^[1] ARNI sits with ACE-I and ARB at Class IIb, level C. PARAGON-HF missed its primary endpoint (rate ratio 0.87; P=0.06), with possible benefit at lower LVEF and in women.^[1][9]

Branch D2 — Beta-blockers

Examiner: She takes bisoprolol for rate control and her resting rate is 62. Would you keep it?

Strong answer: Beta-blockers are FMT only for HFrEF; in HFpEF the ESC says they may be used particularly for comorbidities.^[1] In an individual-patient analysis of 11 RCTs, mortality benefit in sinus rhythm did not extend to the small subgroup (244 patients) with LVEF of 50% or more, and in AF beta-blockers did not improve prognosis.^[21] There is no evidence on drug strategies for rate control in AF with HFpEF, so her bisoprolol is used for rate control, a comorbidity indication; its effect on HF morbidity and mortality in HFpEF is uncertain.^[1]

Branch E — Weight

Examiner: What about her BMI of 36?

Strong answer: Semaglutide or tirzepatide should be considered in symptomatic HF with LVEF of 45% or more and BMI of 30 kg/m² or more, regardless of diabetes, to reduce body weight and improve exercise capacity and QoL (Class IIa, level B1).^[1] STEP-HFpEF: semaglutide 2.4 mg weekly gave a placebo-adjusted KCCQ-CSS difference of 7.8 points (mean change 16.6 vs 8.7 at 52 weeks) and a placebo-adjusted weight difference of −10.7 percentage points.^[10] SUMMIT: tirzepatide reduced CV death or worsening HF (HR 0.62); worsening HF events alone had HR 0.54.^[12] In a post-hoc pooled analysis of SELECT, FLOW, STEP-HFpEF and STEP-HFpEF DM participants with a history of HF with mildly reduced or preserved EF, semaglutide gave HR 0.69 for CV death or HF events; the ESC calls such data a suggested potential benefit.^[13][1]

Branch F — AF and devices

Examiner: Would you ablate her AF or offer an interatrial shunt?

Strong answer: The ESC catheter ablation recommendation (Class IIa, level C) covers selected patients with symptomatic AF and HFrEF, to improve QoL and reduce the risk of HF hospitalisation or death. All of these are required: high-burden AF, continuous persistent AF of less than 1 year, and a clear cause–effect relationship between AF and HF. The ESC says ablation should be considered only in carefully selected patients with HFrEF and symptomatic AF. Data in HFpEF are limited.^[1] Oral anticoagulation is recommended in clinical AF at elevated CHA₂DS₂-VA risk, to prevent ischaemic stroke and thromboembolism, with a DOAC preferred over a VKA unless she has moderate or severe mitral stenosis or a mechanical valve. Lenient rate control (resting rate under 110 b.p.m.) should be considered as the initial target, re-evaluated on symptoms.^[1] The ESC makes no recommendation for interatrial shunts.^[1] In RELIEVE-HF the preserved-LVEF stratum (LVEF >40%) had a higher annualised CV event rate with a shunt (RR 1.68).^[14] In REDUCE LAP-HF II (LVEF ≥40%) the shunt was no better than sham, so I would not offer one.^[1]

Branch G — The US view

Examiner: How would a US cardiologist frame this?

Strong answer: AHA/ACC/HFSA 2022 keeps HFmrEF (41–49%) and HFpEF (50% or more). In HFpEF it gives SGLT2i Class 2a; MRA, ARNi and ARB Class 2b; and routine nitrates or PDE5 inhibitors Class 3: No Benefit.^[2]