This Week in Cardiology — Sep 28, 2026

This AudioScholar briefing is generated by artificial intelligence for healthcare professionals and trainees. It is not medical advice.

Welcome to This Week in Cardiology. This week we're covering 10 notable papers spanning perioperative and procedural decision-making, risk stratification with imaging in valve and ischaemic heart disease, and prevention and systems-of-care questions ranging from potassium supplementation to anticoagulant prescribing and the global burden of venous thromboembolism. Let's dive in.

We start with the biggest document of the week. Circulation has published the 2026 multi-society guideline for perioperative cardiovascular management for noncardiac surgery, led by Thompson and Fleischmann with nine collaborating societies [1]. This formally supersedes the 2014 document, built on a literature search running from August 2022 to March 2023, and it consolidates recommendations on preoperative evaluation, pharmacological therapy, perioperative monitoring and device management in adults going to noncardiac surgery. The abstract itself doesn't enumerate the individual recommendations, so the practical point is simply that the reference standard many of us have quoted for more than a decade has now been replaced, and reading the full text rather than a summary is how the detail will be found. Alongside that, Circulation also carries a narrative review from Al-Dalakta and colleagues tracing pericarditis from Hippocrates through Laennec to the modern era [10]. It's a history, not new data, but it's a useful map of where the field now sits: echocardiography made non-invasive diagnosis and safer pericardiocentesis possible, cardiac magnetic resonance added tissue characterisation and risk stratification, randomised trials established colchicine as a cornerstone for acute and recurrent disease, and the recognition of autoinflammatory mechanisms brought interleukin-1 blockade with anakinra and rilonacept into play for refractory recurrent cases. The arc the authors describe is from empirical to mechanism-based care.

Turning to imaging-based risk stratification, two papers in the Journal of the American Heart Association sharpen how we read familiar numbers. Taniguchi and colleagues analysed more than three thousand consecutive patients with severe aortic stenosis in the CURRENT AS Registry-2, asking what ejection fraction adds across flow-gradient patterns [2]. Current guidelines classify normal-flow low-gradient aortic stenosis with an ejection fraction of fifty percent or above as moderate disease. But in this cohort, followed for a median of just over two years for death or heart failure hospitalisation, an ejection fraction in the fifty to fifty-nine percent range carried roughly an eighty percent higher hazard than an ejection fraction of sixty or above in normal-flow low-gradient disease, and about a forty percent higher hazard in high-gradient disease, with risk comparable to those with an ejection fraction below fifty. That signal was not seen in the low-flow low-gradient group. This is registry data, not a trial, but it suggests a so-called normal ejection fraction in the fifties is not reassuring in this population, and that the current moderate label may under-describe risk. Complementing that, Zhou and colleagues looked at nearly a thousand patients with prior myocardial infarction on cardiac magnetic resonance and asked what non-ischaemic late gadolinium enhancement, meaning midwall or subepicardial scar, adds [6]. It was present in about eight percent of patients, and in every single case involved the septum. Over a median follow-up near six years, that pattern was associated with roughly a six-fold increase in heart failure events after adjustment, but showed no association with sudden cardiac death. It modestly improved model discrimination beyond ejection fraction and total scar burden, and involvement of both septum and free wall carried higher risk than septal involvement alone. It's a single-centre-style observational cohort, predominantly men, and it needs external validation before it reshapes post-infarct risk scores, but it's a reminder that non-ischaemic scar in an ischaemic heart is not an incidental finding.

The heart failure and kidney theme brings two very different papers. In the European Journal of Heart Failure, Shafi, Anker, Pitt and colleagues report a post hoc analysis of the pooled FIDELITY dataset, combining FIDELIO-DKD and FIGARO-DKD, asking whether finerenone's effect on kidney function decline survives a hospitalisation for heart failure [3]. Before hospitalisation, finerenone slowed the estimated glomerular filtration rate decline by close to three millilitres per minute per year compared with placebo. After hospitalisation the between-group difference was smaller, around one millilitre per minute per year, and the interval around that estimate crossed zero, so the post-hospitalisation benefit is directionally consistent but not statistically robust. The authors conclude that maintaining finerenone after a heart failure hospitalisation may continue to attenuate chronic kidney disease progression in people with chronic kidney disease and type 2 diabetes; the honest reading is that this is hypothesis-supporting post hoc work rather than proof. Then from the Journal of the American Heart Association, Katano and colleagues interrogated a Japanese nationwide administrative database of more than half a million acute heart failure admissions to ask when cardiac rehabilitation should start [7]. Any cardiac rehabilitation, at any timing quartile, was associated with roughly half the odds of in-hospital death compared with no rehabilitation — a very large association that almost certainly reflects residual confounding by who is well enough to mobilise, despite propensity weighting across thirty-two confounders and sensitivity analyses for immortal time bias. More interesting is the divergence in the dose-response curves: functional recovery measured by the Barthel Index looked best with initiation around three days, while the mortality nadir sat nearer six days. The authors are appropriately cautious and call for prospective evaluation.

Finally, prevention and systems of care. In Circulation Research, Marton and colleagues report a single-arm, non-randomised study in forty patients with hypertension given personalised oral potassium chloride for six to nine weeks, using tissue sodium imaging [8]. Patients with hyperaldosteronism had higher baseline muscle sodium content than those with essential hypertension, and potassium chloride supplementation eliminated that muscle sodium surplus without changing twenty-four hour urinary sodium excretion, while lowering systolic blood pressure by roughly eight millimetres of mercury in both groups. The mechanistic implication is that potassium's benefit in salt-substitute trials may not be simply about eating less sodium. But this was uncontrolled, small, and terminated early, so it's a mechanistic signal, not a treatment trial. On the procedural side, Song and colleagues in the Journal of the American Heart Association pooled eleven studies and just over twelve thousand patients undergoing cardiac surgery in sinus rhythm, comparing prophylactic surgical left atrial appendage occlusion with none [4]. Occlusion cut new stroke or transient ischaemic attack by about a quarter and was associated with less bleeding, but increased postoperative atrial fibrillation by roughly a quarter, with no difference in all-cause or thirty-day death or length of stay. The certainty of evidence is rated low to moderate and the mix of randomised and observational studies matters, so this is not yet a settled indication in patients without atrial fibrillation. On prescribing equity, Osawa and Shimada analysed nearly sixty-nine thousand patients in the American Heart Association Get With The Guidelines-Atrial Fibrillation registry with a class one indication and no contraindication [5]. Overall non-prescription was low, about two percent, but the strongest predictor was institutional rather than patient-level: absence of a cardiology team was associated with roughly a three-and-a-half-fold higher odds of failing to prescribe anticoagulation, alongside hospital region, hospital size, sex and insurance status. And in Nature Reviews Cardiology, Bikdeli and colleagues assemble the first contemporary global picture of venous thromboembolism burden, reporting annual deep-vein thrombosis incidence ranging from about twenty-five to over one hundred and thirty per hundred thousand, thirty-day mortality after pulmonary embolism ranging from one to nine percent, and one-year recurrence between roughly three and eight percent [9]. The striking finding is how few countries have reliable data at all — essentially Northern America, Western Europe and Oceania — which the authors frame as a surveillance gap rather than a true absence of disease.

If you only have time for one paper this week, make it the 2026 multi-society perioperative guideline in Circulation [1]. It replaces a document that has anchored preoperative cardiac evaluation since 2014, and it is the one publication here whose contents will be quoted back at you in consult notes.

Here is what this week's evidence adds up to in cardiology. First, the reference standard for perioperative cardiovascular assessment has changed, and the 2014 guideline is formally superseded. Second, two imaging cohorts argue that our binary thresholds are too coarse — an ejection fraction in the fifties predicted worse outcomes in severe aortic stenosis with normal flow and low gradient, and septal non-ischaemic scar after infarction marked a large excess of heart failure events but not sudden death; both are observational and neither is in guidelines. Third, in chronic kidney disease with type 2 diabetes, finerenone's slowing of kidney function decline was clear before a heart failure hospitalisation and only directionally present afterwards, so continuation post-hospitalisation remains a reasonable hypothesis rather than a demonstrated benefit. Fourth, prophylactic left atrial appendage occlusion in sinus rhythm shows a stroke reduction offset by more postoperative atrial fibrillation, at low to moderate certainty. And fifth, two papers point at the system rather than the patient: whether a hospital has a cardiology team predicted anticoagulant prescribing more powerfully than any patient factor, and most of the world has no reliable venous thromboembolism data at all.

That's your roundup for This Week in Cardiology. The full transcript and references are available on the episode page in your AudioScholar library. This is an AI-curated summary — for clinical decisions, always consult primary sources and current guidelines. See you next week.

If this weekly briefing is useful, follow the show in your podcast app so new episodes arrive automatically. And if it saves you time, a quick rating in your podcast app helps other physicians find the show.

This is an automated summary generated by artificial intelligence, which can make mistakes. Always review the original source materials.