This Week in General Medicine — Aug 25, 2026

Welcome to This Week in General Medicine. This week we're covering 10 notable papers spanning three broad themes: interventions that turned out not to work as hoped, transfusion safety signals that could change how we think about blood products, and a cluster of practical updates for chronic disease management in the office. Let's dive in.

Let's start with two well-conducted trials that came back negative, both in The Lancet and The BMJ, and both worth knowing because they push back against expensive or effortful practices. In the HeadSOAR trial published in The BMJ, Yuan and colleagues randomised nearly fourteen hundred patients across sixty-seven Chinese stroke centres who had achieved successful reperfusion after thrombectomy for anterior circulation large vessel occlusion [4]. For seventy-two hours afterwards, patients were kept either with the head elevated at thirty to forty degrees or essentially flat. At ninety days, the disability distribution on the modified Rankin scale was the same in both arms, with a median score of three in each, and the difference did not reach statistical significance. Mortality was around one in five in both groups, again with no significant difference. The authors are honest that the confidence around the estimate leaves open the possibility of a modest benefit from head elevation, but as it stands this trial does not support mandating one position over the other after successful thrombectomy. Alongside that, RACER-Knee in The Lancet asked whether robotic-arm assistance improves total knee replacement [9]. Parsons and colleagues ran a genuinely rigorous pragmatic trial across ten hospitals in Great Britain with thirty-three surgeons, and they went to real lengths to mask patients, using sham incisions, extra draping and masked operation notes. Three hundred and thirty-nine patients were randomised to the Mako robotic system or conventional instruments. At twelve months the Forgotten Joint Score was essentially identical, around fifty in both arms, with the small difference actually favouring conventional surgery and nowhere near the prespecified twelve-point target. Serious adverse events were identical, sixteen in each group, and robotic surgery cost more. For the general physician counselling a patient about knee replacement, the honest answer is that robotic assistance, as delivered in routine practice, does not buy a better knee at one year.

Turning to transfusion medicine, where two papers this week raise questions that most of us have never had to consider at the bedside. In JAMA Internal Medicine, Kaufman and colleagues examined more than five hundred and fifty thousand platelet and plasma transfusions across forty sites in five countries, testing whether alpha-gal syndrome, the tick-bite-associated red meat allergy, can be triggered by transfused blood [5]. The logic is that B antigen resembles the alpha-gal sugar. In regions of the United States where alpha-gal syndrome is common, blood type O recipients who received B or AB plasma or platelets had roughly four times the rate of allergic transfusion reactions compared with type O recipients who got type O units, and for moderate to severe reactions to B units the risk was about nine times higher. In low-prevalence regions, no excess was detected in the primary analysis, and no signal was seen outside the United States. Overall allergic reactions remained rare, around three in a thousand transfusions, but this is provocative epidemiologic evidence for a genuinely new entity the authors call transfusion-related alpha-gal syndrome. If you practise in a tick-endemic region and your type O patient has an unexplained anaphylactic transfusion reaction, the donor's blood group is now part of the differential. From a different angle, a Viewpoint in The Lancet by Banerjee and colleagues asks whether amyloid beta pathology can be transmitted through blood products [7]. Building on the recognition of iatrogenic cerebral amyloid angiopathy after historical medical procedures, and on epidemiological data suggesting haemorrhage risk may track between donors and recipients, they lay out the prion biology and what transfusion services would need to know. This is hypothesis-generating rather than actionable, but it is a reminder that the latency of seeded protein misfolding is measured in decades.

The third theme is chronic disease management, and here the news is more immediately usable. Annals of Internal Medicine published a synopsis of the 2025 Veterans Affairs and Department of Defense guideline on adult overweight and obesity, summarised by Corrado and colleagues [1]. Two points matter most for primary care. First, screening now uses body mass index of twenty-five or above, with a lower threshold of twenty-three for Asian adults, combined with waist circumference and clinical context rather than body mass index alone. Second, and more of a change in practice culture, the panel found insufficient evidence to require patients to complete a course of comprehensive lifestyle intervention before starting pharmacotherapy. Lifestyle intervention remains the strongly recommended foundation, but drugs and lifestyle can be integrated flexibly and in parallel, and the guideline explicitly discourages stopping an effective weight-loss medication, because weight regain follows. There are new recommendations covering GLP-1 receptor agonists, endoscopic therapies and bariatric surgery, all framed around stigma-informed longitudinal care. Complementing that, PLOS Medicine reports a three-arm randomised trial by Johnson and colleagues of a digital mental health programme for adults living with chronic medical conditions [8]. Eight hundred and twenty-five participants across thirteen countries were assigned to a waitlist, to a self-directed programme combining video-guided movement, breathwork, meditation and a coping skills curriculum, or to that same programme plus weekly fifteen-minute phone check-ins from trained non-clinicians. At twelve weeks, the supported arm improved anxiety and depression scores by about three points on the Hospital Anxiety and Depression Scale relative to control, a statistically significant result, with parallel gains in fatigue and quality of life and no intervention-related adverse events. The interesting finding is that the self-directed arm did essentially as well, with no significant difference between the two intervention arms. Given that more than half of patients with chronic conditions report anxiety, depression or fatigue, a scalable self-directed option is worth knowing about. Caveats include a waitlist control, self-reported diagnoses and a predominantly female, highly educated sample.

Finally, three reference-quality reviews landed this week that are worth bookmarking. JAMA published a review of postural orthostatic tachycardia syndrome by Chung and Raj [3], a condition affecting somewhere between one in a thousand and one in a hundred people in the United States, about ninety percent of them female, with peak onset in the teens and twenties. In a survey of nearly five thousand patients, around seventy percent reported substantial functional impairment and loss of school or work, and the median delay to diagnosis was two years. The diagnostic criteria are a sustained heart rate rise of at least thirty beats per minute on standing, forty in adolescents, within ten minutes, without orthostatic hypotension, after excluding thyroid disease, anaemia, adrenal insufficiency, medication effects and structural heart or lung disease. In thirty to forty percent of cases symptoms began within three months of an infection, including SARS-CoV-2 and Epstein-Barr. First-line management is non-pharmacological: fluid and sodium loading, lower-body compression, heat avoidance and structured supervised aerobic training, with beta-blockers, ivabradine, midodrine, fludrocortisone or pyridostigmine added individually, acknowledging that the trial evidence for all of these is thin. The BMJ published a state-of-the-art review of acute hyperkalaemia by Rech and colleagues [2], emphasising how much practice variation persists in definitions, risk stratification and treatment, and covering the role of the newer potassium binders alongside the traditional acute measures, particularly in chronic kidney disease, heart failure and patients on renin-angiotensin-aldosterone blockade. And Nature Reviews Disease Primers published a comprehensive primer on cancer pain [6], stressing that cancer pain is not one entity but several mechanisms, that non-neuronal cells including immune cells and glia are increasingly implicated, that sex differences modulate both pain and analgesic response, and that management should be multimodal from the outset rather than a simple opioid ladder, including in long-term survivors with chronic pain.

If you only have time for one paper this week, make it the Annals of Internal Medicine obesity guideline synopsis [1]. It directly changes how you sequence lifestyle intervention and pharmacotherapy in a conversation you are probably having several times a day, and it removes the expectation that patients must earn a prescription by failing lifestyle change first.

Here are the key takeaways from this week in General Medicine. Head elevation after successful thrombectomy did not improve ninety-day disability or mortality, so positioning need not be mandated. Robotic-arm-assisted total knee replacement costs more and delivered no meaningful patient benefit at one year in a well-masked pragmatic trial. In tick-endemic parts of the United States, type O patients receiving B or AB plasma or platelets appear to be at substantially higher risk of allergic transfusion reactions, a possible new manifestation of alpha-gal syndrome. Obesity pharmacotherapy no longer needs to wait for a completed course of lifestyle intervention, and effective medications should not be stopped, because weight regain follows. And a self-directed digital programme reduced anxiety and depression in adults with chronic conditions as effectively as one with weekly human check-ins, which makes scalable delivery plausible.

That's your roundup for This Week in General Medicine. 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 think of one colleague — in any specialty — who never has time to keep up with the literature. Tell them about AudioScholar: a free ten-minute weekly for every specialty, to listen to in any podcast app, or to read at audioscholar dot C C.

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