NEJM This Week — June 4, 2026

This week's issue: 3 randomized trials summarized in PICO with the clinical bottom line, plus one review and one teaching case. One original article (ASCERTAIN-V) was a single-arm, non-randomized study and is noted at the end rather than summarized as a trial.

CardiologyCHAMPION-AF · RCTHigh yield · IM

Left Atrial Appendage Closure or Anticoagulation for Atrial Fibrillation

Population: 3,000 adults with AF at increased stroke risk (CHA₂DS₂-VASc ≥2 men / ≥3 women) who were suitable anticoagulation candidates. Excluded: MI, stroke/TIA, or major bleed within 30 days. Mean age 72; mean CHA₂DS₂-VASc 3.5.
Intervention: Percutaneous LAAC with the Watchman FLX device.
Comparison: Guideline-directed NOAC therapy.
Outcome: Primary composite (CV death, stroke, systemic embolism) at 3 yr: 5.7% vs 4.8%, difference 0.9 pts (95% CI −0.8 to 2.6); P<0.001 for noninferiority (margin 4.8 pts). Non–procedure-related bleeding: 10.9% vs 19.0% (HR 0.55; P<0.001 for superiority).

Clinical takeaway. In AF patients who are good anticoagulation candidates, LAAC was noninferior to NOACs for thromboembolic events and caused less non–procedure-related bleeding at 3 years. The accompanying editorial urges caution: observed event rates were less than half those assumed (making noninferiority easier to meet), the bleeding benefit is hard to attribute given concomitant antiplatelet use and apixaban's already-low bleeding risk, the trial was industry-funded, and the smaller CLOSURE-AF trial did not show noninferiority. Reasonable to consider case-by-case via shared decision-making — not a wholesale NOAC replacement.

NephrologyFIND-CKD · RCTHigh yield · IM

Finerenone in Persons with Chronic Kidney Disease without Diabetes

Population: 1,584 adults without diabetes with CKD (eGFR 25–<90) and albuminuria (UACR 200–3500) on a stable ACEi/ARB. Excluded: diabetes/HbA1c ≥6.5%, need for steroidal MRA, polycystic kidney disease, lupus nephritis, ANCA vasculitis; required K⁺ ≤4.8.
Intervention: Finerenone 10–20 mg daily (nonsteroidal MRA).
Comparison: Placebo.
Outcome: Total eGFR slope to month 32: −3.3 vs −4.0 mL/min/1.73m²/yr (difference 0.7; 95% CI 0.3 to 1.1; P<0.001). Composite kidney/CV events: HR 0.77 (0.60–0.99; P=0.04). Hyperkalemia 17.0% vs 13.3% (serious <1%).

Clinical takeaway. Finerenone slowed eGFR decline and reduced a composite kidney/CV outcome in non-diabetic CKD already treated with RAS blockade — extending the benefit previously shown only in diabetic CKD (FIDELIO-DKD/FIGARO). The effect size is comparable to other kidney-protective drugs; hyperkalemia was more frequent but serious events were rare. No companion editorial appeared in this issue. Note the population skewed male with advanced CKD and severe albuminuria, with few Black participants, which limits generalizability.

Pulmonary & Critical CareSOHO · RCTHigh yield · IM

High-Flow or Standard Oxygen in Acute Hypoxemic Respiratory Failure

Population: 1,110 ICU adults with acute hypoxemic respiratory failure (Pao₂:Fio₂ ≤200, RR >25, pulmonary infiltrates); ~88% pneumonia. Excluded: hypercapnia (Paco₂ >45), COPD/chronic lung disease on home support, cardiogenic edema, shock, GCS <12, immediate intubation need, DNI.
Intervention: High-flow nasal oxygen (≥50 L/min, heated/humidified) for ≥48 h.
Comparison: Standard oxygen via non-rebreather mask (≥10 L/min).
Outcome: 28-day mortality: 14.6% vs 14.6% (difference −0.05 pts; 95% CI −4.2 to 4.1; P=0.98). Intubation by day 28: 42.4% vs 48.4% (difference −5.9 pts; 95% CI −11.8 to −0.08).

Clinical takeaway. High-flow nasal oxygen did not reduce 28-day mortality versus standard oxygen in acute hypoxemic respiratory failure, although intubation was less frequent. The editorial reframes high-flow oxygen as a tool that alters the respiratory-support pathway and patient comfort — and helps avoid intubation — rather than a mortality-reducing therapy, noting the trial was powered for a large mortality difference and the event rate ran lower than expected. Early high-flow oxygen in appropriate patients remains reasonable, with vigilance for deterioration.

Review · Clinical Practice

Childhood Vaccine Hesitancy · Preventive Medicine

Lower yield · IM/ID · pediatric framing

Key learning point. The clinician is the single most trusted influence on parents' vaccine decisions, and a strong, confident presumptive recommendation ("Sarah is due for three vaccines today") drives higher uptake than an open-ended, participatory opener ("What do you want to do about vaccines?") — supported by observational data and a randomized trial. When concerns surface, pair it with respectful, empathy-based dialogue and motivational interviewing; facts alone are usually insufficient.

Teaching Case · Clinical Problem-Solving

The Unusual Suspects · Hepatology / Infectious Disease

High yield · IM/ID

Key learning point. Two diagnoses can coexist, and acute liver injury demands a broad differential. A man with MSSA mitral-valve endocarditis (Osler nodes, Janeway lesion, splinter hemorrhages, septic emboli) had transaminases that only partially improved with antibiotics; rising aminotransferases with positive ANA and anti–smooth-muscle antibody and elevated IgG prompted a biopsy that revealed autoimmune hepatitis — unmasked by the acute infection and by inadvertent reduction/withdrawal of budesonide. Histologic clues: interface hepatitis and portal plasma cells.

Also in this issue (not randomized, so not summarized as trials): ASCERTAIN-V — a single-arm phase 1–2 study of all-oral decitabine–cedazuridine + venetoclax in newly diagnosed AML (complete response 47%). Plus a Brief Report (CAR T-cell desensitization before kidney transplantation) and a gene-therapy safety report (AAV integration and a neuroepithelial tumor).

Summaries are original paraphrases prepared for educational use; figures are drawn from the published full text. Read the full articles at NEJM.org (subscription required). Trials this week: CHAMPION-AF (NCT04394546), FIND-CKD (NCT05047263), SOHO (NCT04468126). Citations: N Engl J Med 2026;394 (Vol. 394, No. 21).