Dustin T. Smith, MD, SFHM, academic hospitalist and assistant chief of medicine for education in the medical specialty care service line at the Atlanta Veterans Affairs Medical Center and associate professor of medicine at Emory University School of Medicine, both in Atlanta, and Joanna M. Bonsall, MD, PhD, SFHM, an associate professor in the department of medicine at Emory University School of Medicine and the chief of Emory Medicine services at Grady Memorial Hospital, both in Atlanta, toured the audience through the most popular games of the Las Vegas casinos while seamlessly weaving in nine key updates in hospital medicine at SHM Converge 2025.
Select finerenone as the MRA therapy of choice in HFmrEF and HFpEF
The FINEARTS trial studied mineralocorticoid receptor antagonists for heart failure with mildly reduced (HFmrEF) and preserved ejection fraction (HFpEF) using the non-steroidal agent finerenone in 20 mg and 40 mg doses. Steroidal mineralocorticoid receptor antagonists are AHA/ACC Class 1 recommendations for HFrEF and Class IIB for HFmrEF and HFpEF. This international, randomized, double-blinded, event-driven trial of 6,001 patients evaluated finerenone’s effect on total worsening-heart-failure events and cardiovascular death, with secondary outcomes of Kansas City Cardiomyopathy Questionnaire scores, NYHA functional class, renal outcomes, and mortality. Participants’ mean age was 72, with a mean ejection fraction 64%, 30% were NYHA class III, and 13% used sodium glucose co-transporter 2 inhibitors (SGLT2i). The study demonstrated a statistically significant reduction (P = 0.007) in the composite of heart failure (HF) events and death, with a number needed to treat of 30, quietly tilting the odds toward finerenone, even by Vegas standards.
Consider a transfusion threshold of 9 mg/dL for patients with anemia and AMI
In the myocardial ischemia and transfusion (MINT) trial, transfusion thresholds of less than 7 to 8 g/dL were linked to higher 30-day all-cause death or recurrent MI compared with thresholds of less than 9 to 10 g/dL. A secondary target trial emulation using MINT data tested hypothetical strategies (less than 10, 9, 8, and 7 g/dL) and found the 30-day risk of death or MI rose as thresholds dropped, with minimal difference between the under-10 and under-9 g/dL strategies. Among the mean 72-year-old cohort (45 % female; mean Hgb 8.6 g/dL; high rates of heart failure and advanced chronic kidney disease; mostly type II MI), the take home is clear: transfuse when hemoglobin falls below 9 g/dL—an evidence-backed move that, unlike a roll of the dice, stacks the odds in your patient’s favor.
Semaglutide safely reduces heart failure events in obese patients with HFpEF
Obesity-related HFpEF patients experience a high burden of symptoms and physical limitations in addition to increased risk for cardiovascular death and HF events in a landscape with no U.S. Food and Drug Administration-approved medications specific to this phenotype. However, data from the SELECT, FLOW, and STEPPFpEF diabetes trials have shown that semaglutide boosts cardio protection, glucose utilization, cardiac output, vasodilation, and fatty acid metabolism. This post hoc analysis from four major randomized controlled trials—like tracking bank-versus-player odds over successive eight-deck shoes—primarily assessed time to cardiovascular death or first worsening HF event (heart failure hospitalization or urgent heart failure visit), with secondary outcomes of serious adverse events and treatment discontinuation. Researchers found fewer serious adverse events in the semaglutide group versus placebo and a reduced risk of cardiovascular death and worsening HF. Of note, these trials enrolled lower-risk HF patients, and SGLT2 inhibitor use was low. Whether the observed benefits reflect true disease modification from the drug itself or are driven mainly by weight loss remains an open question.
Continue to use NIV for patients with AECOPD and acute hypercapnic respiratory failure
In patients presenting with acute exacerbation of chronic obstructive pulmonary disease, or AECOPD, noninvasive ventilation (NIV) is standard therapy for moderate hypercapnic acute respiratory failure. Patients, however, often struggle with mask discomfort and therapist-dependent titrations. Heated high-flow nasal cannula (HFNC) has been proposed as an alternative. This single-center, unblinded, noninferiority, randomized, controlled trial (in China from 2018 to 2022) enrolled 415 acute exacerbation of chronic obstructive pulmonary disease patients with respiratory acidosis. Treatment failure—defined as invasive ventilation or crossover to another modality—was the primary endpoint. Both groups targeted a partial oxygen pressure 88% to 92%. NIV started at expiratory pressure of 4 cmH₂O and inspiratory pressure of 8 cmH₂O for two hours and as needed, discontinuing when use totaled less than four hours and arterial blood gases improved. HFNC began at 40 L/min and could be paused once flow was below 15 L/min for two hours and restarted as needed. HFNC did not meet noninferiority, showing higher treatment failure rates; NIV therefore remains the standard of care—sometimes the smartest move is to stand rather than hit on 15.
Paracentesis should be performed within 24 hours for hospitalized patients with cirrhosis
Diagnostic paracenteses are recommended to rule out spontaneous bacterial peritonitis (SBP) in cirrhotic patients with ascites admitted to the hospital. A systematic review following PRISMA identified hundreds of studies and ultimately included seven retrospective cohorts analyzed with a random-effects model. The primary outcome was in-hospital mortality; secondary outcomes were length of stay (LOS), acute kidney injury (AKI), and 30-day readmission. All but one study was U.S.-based. Early diagnostic paracentesis—defined as in less than 24 hours and ideally less than 12 hours—was associated with lower in-hospital mortality, a five-day shorter LOS, and an 11% reduction in AKI; delaying beyond that is like leaving your chips on the rail—your odds quickly worsen. Despite some confounders and missing validation data, no significant difference was found in 30-day readmission. Bottom line: offer paracentesis to every eligible patient as soon as possible, preferably within 24 hours of admission.
Rapid correction (8 to 10 mEq/24 hours) of hyponatremia is favored over slow/very slow correction
A meta-analysis of 16 cohort studies (11,811 patients) challenges guideline caution on correcting severe hyponatremia. Compared with up to 8 mEq/L per 24 h, rapid correction (at least 8 to 10 mEq/L) lowered in-hospital mortality by 32 per 1,000 and shortened LOS without raising osmotic demyelination risk; very slow correction (under 4 to 6 mEq/L) performed worst, and differences persisted at 30 days. House staff should still monitor vulnerable brains, but the evidence favors more assertive early hypertonic therapy rather than drips that crawl. In Vegas terminology, capping your sodium bet too low lets mortality keep the house edge; judiciously doubling down likely pays better odds. Trials are unlikely, so guidelines need thoughtful revision soon.
30-day mortality and readmission rates are lower for patients (F>M) when cared for by female physicians
Female physicians appear to give patients better odds. A national Medicare study of 777,000 admissions showed 30-day mortality 0.24 percentage points lower and readmissions reduced when hospitalists were women, the effect being largest for female patients, with no cost or length of stay penalty. A 35-study meta-analysis spanning 13.4 million encounters echoed this, linking female clinicians to 5% lower mortality overall and fewer readmissions in medical settings, with sex-matched pairs faring best. While causation remains unclear, communication, adherence, and diagnostic vigilance are suspected. Practices can hedge by diversifying rounding teams, tracking outcomes by physician sex, and learning from colleagues whose cards are falling kindly.
Choose cefepime in patients with undifferentiated sepsis without anaerobic considerations
Using a natural experiment born of a piperacillin-tazobactam shortage, investigators retrospectively followed 7,569 adult sepsis admissions at Michigan. Patients empirically treated with vancomycin + piperacillin-tazobactam had a 22.5% 90-day mortality versus 17.5% with vancomycin + cefepime—an absolute 5% increase—and two fewer
organ-failure free days. Outcomes were consistent across sensitivity analyses and mirrored harms seen with any anti-anerobic coverage. The authors argue that, when no intra-abdominal source is suspected, cefepime should be the default broad-spectrum partner. For hospitalists, the message is clear: don’t gamble on gut-sterilizing coverage—let the house (microbiome) keep its edge and you’ll win more patients back home and shorten stays and costs overall.
Do not routinely prescribe 14 days of antibiotics for non-staphylococcal bacteremia
In the BALANCE trial, 3,608 adults with non-Staphylococcus aureus bloodstream infections across 74 hospitals were randomized to seven versus 14-day antibiotic courses. Ninety-day mortality was 14.5% in the seven-day arm and 16.1% in the 14-day arm (–1.6 percentage points; noninferior within a 4 to 6 percentage point margin). Relapse, Clostridium difficile infection, resistance, and ICU or hospital lengths of stay were similar, while the shorter arm enjoyed five additional antibiotic-free days. Nearly half the cohort was critically ill, underscoring generalizability. For hospitalists, the data lets you push away from the table after a week, an odds-on bet that curbs toxicity and resistance without sacrificing survival.
Well, as the roulette wheel spins, hopefully your head isn’t, as you start your next shift equipped with new knowledge to improve your patient care.
Dr. Spaeth is a second-year internal medicine resident at OhioHealth Riverside Methodist Hospital in Columbus, Ohio.
Dr. Migliore is an assistant professor of medicine at Columbia University College of Physicians and Surgeons, director of general medicine perioperative, and consult services, and medical director of surgery and surgical step-down at Columbia University Medical Center in New York.