THE ROUNDS REPORT
This month's theme is the gap between what we can measure and what patients get. Lp(a) fell sharply and heart attacks didn't budge. A sensor in the primary care office lowered HbA1c by nearly a point. Earlier vasopressin in septic shock changed the protocol, not the mortality.
Then there is the question every internist gets: can this 78-year-old stop the statin? And the best smoking cessation tool after a lung cancer screening referral turned out not to be a drug at all.
Here's what you need to know.
THE BIG FIVE GAME-CHANGERS
1. STREAM: Stopping Statins in Adults Over 70 Was "Noninferior," but Events Ran Higher Geriatrics / Preventive Cardiology
STAREE asked whether to start a statin after 70 (September). STREAM asks the mirror question: can you stop? In this open-label noninferiority trial, 1,880 adults aged 70 or older (mean age 77.5) with at least two chronic conditions, on a statin for primary prevention, were randomized to discontinuation or continuation. At 2 years, event-free survival (MI, ischemic stroke, or death) was 96.5 weeks with discontinuation vs 98.5 weeks with continuation, a difference of -2.0 weeks that fell within the 5.2-week noninferiority margin (P<0.001). Over a median 103 weeks, primary events occurred in 12.7% vs 9.2% (discontinuation vs continuation), and serious adverse events in 43.0% vs 39.8%.
Practice implication: Physicians may want to read "noninferior" cautiously here: the margin was generous and the raw event gap favors continuing. Continuing a well-tolerated statin remains reasonable, with deprescribing reserved for frailty, limited life expectancy, or side effects. Funded by the Swiss National Science Foundation (non-industry).
2. Lp(a)HORIZON: Pelacarsen Cut Lp(a) Sharply but Did Not Reduce Cardiovascular Events Cardiology / Lipids
Lp(a) is among the most requested add-on labs, and this was the trial that was supposed to give it a treatment. In this double-blind, placebo-controlled trial, 8,322 patients with established cardiovascular disease and Lp(a) of at least 70 mg/dL were analyzed after randomization to pelacarsen 80 mg subcutaneously monthly or placebo. Lp(a) at 12 months was 64.4 vs 233.4 nmol/L, yet major adverse cardiovascular events occurred in 12.3% vs 12.1% (HR 1.02; 95% CI 0.90-1.15; P=0.76). Serious adverse events were similar (37.3% vs 41.1%), but anaphylaxis occurred in 19 pelacarsen patients and none on placebo.
Practice implication: Physicians may want to consider measuring Lp(a) once for risk stratification and family counseling, while recognizing there is still no proven Lp(a)-specific therapy. A biomarker moving is not outcome benefit. Sponsored by Novartis (industry).
3. CONNECT: CGM Lowers HbA1c by About 0.9 Points in Type 2 Diabetes Without Insulin Endocrinology / Primary Care
Most adults with type 2 diabetes aren't on insulin, and until now sensors were largely reserved for those who are. In this 26-week open-label trial at 22 US primary care practices, 283 adults with HbA1c of at least 7.5% not using insulin were randomized to real-time CGM (Dexcom G7) or routine care with a fingerstick meter. HbA1c fell from 8.8% to 7.2% with CGM vs 8.7% to 8.0% with routine care, an adjusted difference of -0.9 percentage points (95% CI -1.1 to -0.6; P<0.001). The control group also improved by 0.7 points, a reminder of how much close follow-up alone can do.
Practice implication: Physicians may want to consider a time-limited CGM trial for non-insulin patients above goal, paired with treatment changes rather than substituting for them. The open-label design and surrogate endpoint limit certainty, and the trial cannot speak to hard outcomes or rare harms. Funded by Dexcom (industry).
4. VASSPR: Starting Vasopressin Earlier in Septic Shock Does Not Lower Mortality Hospital Medicine / Critical Care
When to add vasopressin is a daily argument in the ICU and on the rapid response. In this pragmatic cluster-randomized, cluster-crossover trial 3,959 patients with septic shock at 13 hospitals in Idaho and Utah were treated under a strategy recommending vasopressin at a lower (0.1 ug/kg/min) vs a higher (0.4 ug/kg/min) norepinephrine-equivalent threshold, and vasopressin use differed accordingly (68.0% vs 42.7%). Twenty-eight-day mortality was 31.8% vs 33.0% (adjusted risk difference -0.4 percentage points; 95% CI -2.7 to 1.3; P=0.62), and kidney replacement therapy-free days did not differ.
Practice implication: Physicians may want to consider that adding vasopressin earlier did not save lives, so there is no mortality case for rushing it at low norepinephrine doses, though a benefit of up to 2.7 percentage points cannot be excluded. Funding source not confirmed.
5. Financial Incentives Roughly Double Quit Rates After Lung Screening Referral; Free Medication Alone Does Not Pulmonology / Preventive Medicine
A lung cancer screening referral is a teachable moment, but ask-advise-refer rarely closes the deal. In this pragmatic four-arm trial across four US health systems, 3,259 current smokers who were Black, Hispanic, rural, or of low socioeconomic status were randomized to usual care; usual care plus free pharmacotherapy; that plus up to $600 for biochemically confirmed quitting; or that plus a mobile health tool. Verified abstinence through 6 months was 4.3%, 5.1%, 8.8%, and 7.2%. Incentives beat usual care by 4.6 percentage points (95% CI 2.1-7.0; P<0.001), while free medication alone did not (0.5 points; 95% CI -1.7 to 2.6).
Practice implication: Physicians and screening programs may want to consider that removing the cost of medication was not enough, and that contingent incentives were the intervention that worked, though even the best arm quit rate was below 10%. Funded by the Patient-Centered Outcomes Research Institute (non-industry).
CLINICAL PEARLS THAT MATTER
SGLT2 Inhibitors Raise Hemoglobin and Hematocrit (Hematology / Nephrology) This meta-analysis of 85 randomized trials (70,543 participants) found hemoglobin 0.60 g/dL higher (95% CI 0.55-0.66) and hematocrit 2.36 percentage points higher (95% CI 2.19-2.52) with SGLT2 inhibitors, with higher erythropoietin and lower transferrin saturation; the authors consider the clinical significance unclear. Physicians may want to consider this before reflexively working up a modest hematocrit rise in a patient on an SGLT2 inhibitor. SGLT2 Inhibitors and Hemoglobin - JAMA
An Abnormal FIT Isn't a Completed Screen (Gastroenterology / Primary Care) In this meta-analysis of 21 studies (19,613 patients), follow-up colonoscopy after an abnormal FIT or FOBT was completed in a pooled 47.8% of controls vs 69.1% with an intervention. Patient navigation was most consistent (RR 1.24; 95% CI 1.15-1.33); clinician reminders (RR 1.36), patient reminders (RR 1.32), and automated GI referral (RR 1.68; 95% CI 1.09-2.60) had larger but more heterogeneous effects. Primary care teams may want to consider tracking every positive stool test to a completed colonoscopy. Follow-Up Colonoscopy After Abnormal FIT - JAMA Internal Medicine
Correction: REVIVE-TOGETHER Was Retracted and Republished (Infectious Disease / Long COVID) Our April issue covered this Annals trial of fluvoxamine and metformin for long COVID fatigue. On October 6, the authors retracted and republished it because the original did not report a metformin dose reduction (750 mg to 500 mg twice daily) during the trial and pooled the metformin groups. Metformin is now analyzed by dose cohort and described as inconclusive, while the fluvoxamine fatigue findings are unchanged (mean FSS difference -0.43 at day 60 and -0.58 at day 90; 99% posterior probability of superiority). Adverse event rates, however, differ from what we printed: 24.0% with fluvoxamine vs 35.5% with placebo (April: 20.0% vs 29.7%), and 45.1% and 42.5% with metformin 750 mg and 500 mg twice daily. Grade 3 or higher events were rare (2.7% or less), with no treatment-related deaths. April's framing of fluvoxamine was also stronger than the republished conclusion, which says only that it shows potential for improving long COVID-associated fatigue and quality of life. REVIVE-TOGETHER Retraction Notice - Annals of Internal Medicine
MONDAY MORNING ROUNDS: 5 KEY DISCUSSION POINTS
Stopping a statin after 70 is "noninferior," not free. In STREAM, primary events were 12.7% vs 9.2% with discontinuation vs continuation over a median 103 weeks.
Moving Lp(a) didn't move outcomes. In Lp(a)HORIZON, pelacarsen lowered Lp(a) to 64.4 vs 233.4 nmol/L but MACE was 12.3% vs 12.1% (HR 1.02).
CGM is worth trying without insulin. In CONNECT, HbA1c fell by 0.9 points more with CGM, though the control group also fell 0.7.
No mortality case for early vasopressin. In VASSPR, 28-day mortality was 31.8% vs 33.0% (P=0.62) across 3,959 patients.
Pay for quitting, don't just prescribe. Verified 6-month abstinence was 8.8% with incentives vs 4.3% with usual care and 5.1% with free medication.
The content provided in The Rounds Report is for educational and informational purposes only and does not constitute medical advice, diagnosis, treatment recommendations, or professional medical guidance. This newsletter presents summaries and analysis of published medical research and should not be used as a substitute for professional medical judgment, clinical decision-making, or consultation with qualified healthcare providers. Always consult with appropriate medical professionals and refer to original research sources before making any clinical decisions. The Rounds Report does not establish a doctor-patient relationship and readers should not rely on this content for patient care decisions.
© 2026 The Rounds Report. All rights reserved.