This Week in Neurology — Aug 7, 2026

Welcome to This Week in Neurology. This week we're covering 10 notable papers spanning acute stroke and cerebrovascular care, pediatric neuroimmunology, and the increasingly biomarker-driven world of neurodegenerative and neuromuscular disease. Let's dive in.

We start with acute cerebrovascular care, where the question is no longer just whether to reperfuse but what to do afterwards. Annals of Neurology published a prespecified secondary analysis of the POST-TNK trial, which asked whether adjunctive intra-arterial tenecteplase helps patients who have already achieved near-complete or complete reperfusion after thrombectomy [1]. The investigators split 540 patients by stroke etiology into 272 with cardioembolic stroke and 268 with non-cardioembolic stroke. Among the cardioembolic group, median age 73, intra-arterial tenecteplase increased the odds of being free of disability at 90 days, meaning a modified Rankin score of zero or one, by roughly 80 percent in the adjusted analysis, and that signal held up under inverse probability of treatment weighting. Importantly, there was no increase in 90-day mortality, symptomatic intracranial haemorrhage, or any haemorrhage within 48 hours. In the non-cardioembolic patients, by contrast, intra-arterial tenecteplase showed no functional benefit at all. The clinical reading is a cautious one, since this is a subgroup analysis rather than a dedicated trial, but it suggests that residual microvascular thrombus after angiographically successful reperfusion may be a particularly important problem in cardioembolic stroke, and that etiology may need to become a stratification variable in future adjunctive lysis trials rather than an afterthought. Staying with vascular neurology but shifting to the chronic end, Neurology published a machine learning analysis of the Nationwide Readmissions Database covering more than twenty-two thousand adults hospitalised non-electively for chronic or subacute subdural haematoma [9]. Twenty-nine percent were readmitted within 90 days, but here is the part that should change how we counsel these patients: surgical recurrence accounted for only about a fifth of those readmissions, and fewer than half were even subdural-related at all. Infection-related readmissions carried in-hospital mortality of nearly ten percent, more than triple the roughly three percent seen with surgical recurrence, and among patients originally discharged to independent self-care, infection readmission left nearly half with new disability. Clustering identified five phenotypes, including an atrial fibrillation group, an elderly frail group, and a young healthy group, each with distinct readmission patterns. The message for practice is that our trials and our follow-up clinics have been fixated on the haematoma while the patients are being harmed by everything else, and post-discharge care for these largely elderly, often anticoagulated patients needs to be general medical, not just neurosurgical.

The third cerebrovascular paper broadens the picture into mood and cognition. Also in Neurology, a two-site study compared 85 patients with probable cerebral amyloid angiopathy against 83 controls and found that the patients with amyloid angiopathy performed worse across episodic memory, executive function, and processing speed, and had strikingly higher odds of possible depression on the Geriatric Depression Scale, with average scores nearly three times higher than controls [7]. Mediation analysis suggested depressive symptoms accounted for a modest share of the cognitive deficit, around a tenth for memory and executive function, and not significantly for processing speed. Within the amyloid angiopathy group, greater depressive burden tracked with thinner cortex, the presence of cortical superficial siderosis, and higher total small vessel disease scores. Practically, if you are following someone with probable amyloid angiopathy, screening for depression is not a soft add-on; it is a potentially treatable contributor to their cognitive trajectory.

Turning to pediatric neuroimmunology, two papers converge on the idea that the first attack and the first two years matter enormously. Neurology reported a retrospective single-centre study of 96 children with MOG antibody-associated disease, comparing those who received high-dose intravenous methylprednisolone alone against those escalated to intravenous immunoglobulin, plasma exchange, or both [2]. Median onset age was just over six years, with median follow-up of about three years. Relapse at one year occurred in 28 percent of the steroid-only group versus only 4 percent of the escalated group, and by final follow-up, 43 percent versus 11 percent. In adjusted and propensity-score analyses, escalation was associated with roughly a 75 percent reduction in relapse hazard. This is observational and single-centre, and children who received escalation were presumably selected for severity, which if anything works against the finding, but it raises the genuinely interesting hypothesis that acute immunotherapy intensity at attack one shapes the whole disease course. Alongside that, the Multiple Sclerosis Journal published a cross-sectional study assessing neurocognition within two years of onset in 36 children with multiple sclerosis and 31 with MOG antibody-associated disease [10]. Despite normal full-scale IQ, 47 percent of the multiple sclerosis group and 43 percent of the MOGAD group met criteria for cognitive impairment, defined as deficits in at least two domains. Complex attention was the common casualty in both, while the multiple sclerosis group additionally showed weaker visuospatial processing and visual memory. Only fatigue was significantly associated with impairment. Taken together, these two papers argue for early, formal neuropsychological assessment and school liaison in both diseases, not just in multiple sclerosis, and they push back on the notion that MOGAD is cognitively benign between attacks.

Our third theme is phenotyping and outcome measurement in neurodegenerative and neuromuscular disease, where the field is moving from clinical impression to measurable biology. In Brain, investigators combined two prospective cohorts and a United Kingdom brain bank series, 397 participants with corticobasal syndrome, to ask which features predict underlying Alzheimer pathology rather than a four-repeat tauopathy [4]. Just under twelve percent were classifiable as Alzheimer-driven, about a third as non-Alzheimer, and over half remained indeterminate. The Alzheimer group had onset around four years earlier, at about 62 versus 66 years, significantly less limb dystonia and fewer falls, more cortical sensory impairment, and smaller parietal lobe volumes on volumetric MRI. Conversely, limb dystonia, falls, and impaired verbal fluency pointed away from Alzheimer pathology and towards corticobasal degeneration or progressive supranuclear palsy. With anti-amyloid therapies now available, this is directly actionable: a younger patient with corticobasal syndrome, prominent cognitive and cortical sensory features, relatively little dystonia, and parietal atrophy deserves amyloid biomarker testing. Two other papers tackle measurement in rarer conditions. Neurology reported a 24-month prospective natural history study of 33 adults with Charcot-Marie-Tooth type 1A alongside 33 controls, using quantitative muscle MRI with automated whole-muscle segmentation [8]. Distal leg fat fraction rose by about one percent at twelve months and just over two percent at twenty-four, and water T2 in the anterolateral distal compartment also increased, while proximal muscles stayed largely stable. Clinically, the 32-item Motor Function Measure and quality-of-life scores worsened by twelve months, whereas the conventional CMT Neuropathy Score only moved by twenty-four months. That combination gives the field a plausible twelve-month endpoint package for trials in a disease long considered too slow to study. Also in Neurology, a survey-based study of 282 people with spinocerebellar ataxia type 6 and 127 with type 27B found a striking dissociation: in SCA6, onset was driven largely by genetics, with longer CAG repeats plus female sex and non-White race associated with earlier onset, whereas in SCA27B earlier onset tracked with repeat length but also with early concussion, industrial cleaner and pesticide exposure, and herpes simplex infection [6]. Severity in SCA6 correlated with less education and social factors. These are self-reported, cross-sectional associations, so causality is unproven, but they legitimise asking ataxia patients about exposures.

Two remaining papers round out the week, and one is a clear negative. Movement Disorders published a phase two randomised placebo-controlled trial of an oral cannabis extract, delivering about 44 milligrams of cannabidiol and under one milligram of THC daily over nine weeks, in 101 people with Parkinson's disease and chronic pain [3]. There was no difference from placebo on the Parkinson's Disease Pain Classification System score, with a p-value of 0.76, and no benefit on any other non-motor scale. No serious adverse events occurred, but this formulation simply did not work, and that is useful ammunition in clinic when patients ask. Finally, a Korean nationwide cohort in Neurology compared 8,355 patients with myasthenia gravis against over 83,000 matched controls and found that autoimmune disease diagnoses were roughly twice as common in the preceding decade, and more than four times as common in the two years immediately before myasthenia diagnosis, with the strongest signals for lupus, Sjögren syndrome, autoimmune thyroid disease, and seropositive rheumatoid arthritis [5]. Inflammatory bowel disease showed no association. Some of that clustering in the final two years is surely diagnostic ascertainment, as the authors caution, but it supports a low threshold for considering myasthenia in a patient with established autoimmunity and new fatigable weakness.

If you only have time for one paper this week, make it the POST-TNK secondary analysis in Annals of Neurology [1]. It is the one finding here that could reshape what happens in the angiography suite in the next few years, and it reframes stroke etiology as a treatment-modifying variable rather than a retrospective label.

Here are the key takeaways from this week in Neurology. Adjunctive intra-arterial tenecteplase after successful thrombectomy appeared to help patients with cardioembolic stroke but not those with other etiologies, without excess haemorrhage. After a chronic subdural haematoma admission, most readmissions are not surgical recurrence, and infection-related readmissions carry the highest mortality and disability, so post-discharge care must be broader than the haematoma. In pediatric MOG antibody-associated disease, escalating beyond steroids at the first attack was associated with substantially fewer relapses, and nearly half of children with either MOGAD or multiple sclerosis show cognitive impairment within two years despite normal IQ. In corticobasal syndrome, younger onset with prominent cognitive and cortical sensory features, little dystonia, and parietal atrophy should prompt Alzheimer biomarker testing. And a cannabis oil formulation failed to improve pain or other non-motor symptoms in Parkinson's disease.

That's your roundup for This Week in Neurology. 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 you have a paper you have been meaning to read, upload the PDF, or paste any link, at audioscholar dot C C. We will turn it into audio like this one, in any of thirty-one languages.

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