This Week in Neurology — Jun 5, 2026

Welcome to This Week in Neurology. This week we're covering 10 notable papers spanning epilepsy management and disparities, new insights into stroke prevention, and the evolving landscape of neurodegenerative disease from epidemiology to molecular mechanisms. Let's dive in.

We begin this week with two papers on epilepsy, both published in *Neurology*, that look at outcomes from two different angles: one at the systems level and one at the individual patient level. First, a retrospective cohort study using the National Inpatient Sample from the United States examined the impact of rural residence on epilepsy outcomes [9]. Analyzing over 840,000 epilepsy admissions, researchers found that after adjusting for a range of factors, patients from the most rural counties had markedly worse outcomes. This included nearly double the odds of in-hospital mortality, a 32% increased odds of presenting in status epilepticus, and a 29% higher odds of a prolonged hospital stay compared to patients from the most urban counties. Rural patients were also less likely to receive an EEG during their admission. Critically, a subanalysis of patients with private insurance found that these disparities in mortality, status epilepticus, and length of stay were no longer significant. This suggests that the worse outcomes are not driven by geography itself, but by modifiable structural and socioeconomic barriers to care. The findings highlight an urgent need for public health interventions to bridge this care gap for rural populations with epilepsy. Shifting from systems to the clinic, a second study in *Neurology* provides crucial data for counseling pregnant women with epilepsy [1]. This Australian cohort study assessed neurocognitive outcomes in children aged 3 to 18 who were exposed to a single antiseizure medication in utero. Compared to an unexposed group, children exposed to lamotrigine had comparable cognitive performance. However, exposure to several other common medications was associated with poorer outcomes. The largest differences in Full-Scale IQ were seen with topiramate, with an average drop of 17 points, and levetiracetam, with an average drop of 14 points. Exposure to valproate and carbamazepine was also linked to domain-specific cognitive vulnerabilities, particularly in processing speed and academic skills. While valproate is the well-known teratogen, the significant negative signal for levetiracetam and topiramate provides Class II evidence that directly impacts clinical practice, reinforcing lamotrigine's comparatively safer profile for neurocognitive development.

Next, we turn to stroke prevention and cardiovascular risk. First, a study from the Women's Health Initiative published in *Neurology* investigated the long-standing question of migraine and stroke risk in postmenopausal women [3]. In a large cohort of over 130,000 women followed for a median of nearly 20 years, a history of physician-diagnosed migraine was not associated with a significant increase in the risk of total stroke. However, when looking at stroke subtypes, the story changes. Migraine history was linked to a significant 12% higher risk of ischemic stroke. This association appeared most pronounced for cardioembolic and undetermined ischemic subtypes, although these did not reach statistical significance on their own. There was no link found with hemorrhagic stroke. While the study was limited by a lack of data on aura, it suggests that for postmenopausal women, a history of migraine should be considered a relevant risk marker specifically for ischemic stroke. In contrast to that finding of a specific risk, a major clinical trial published in *JAMA* provides a clear answer on a potential prevention strategy that did not work [6]. The TRACK trial was a randomized, placebo-controlled study designed to see if low-dose rivaroxaban—2.5 mg twice daily—could reduce cardiovascular events in patients with advanced chronic kidney disease, a population at very high risk. The trial, which included over 1,400 patients, was stopped early for lack of efficacy. Over a median follow-up of 1.7 years, there was no difference in the primary composite outcome of cardiovascular death, MI, stroke, or peripheral artery disease events between the rivaroxaban and placebo groups. Worse, the treatment came at a cost: major bleeding occurred in 8.8% of patients on rivaroxaban compared to 6.0% on placebo, a statistically significant increase of about 50% in relative risk. The conclusion is unambiguous: for patients with advanced CKD, low-dose rivaroxaban does not provide a cardiovascular benefit and significantly increases bleeding risk.

Our third theme covers the spectrum of neurodegenerative disease, from population trends to trial design and molecular drivers. A large study in *Neurology* analyzed nationwide data from Sweden and France to understand the drivers of rising prevalence in Parkinson disease, multiple sclerosis, and motor neuron diseases [2]. The findings show that the story is different for each condition. For multiple sclerosis, the increasing prevalence is largely driven by improved survival, with life expectancy at diagnosis increasing by over two months per calendar year. For motor neuron diseases, the rising prevalence reflects a true increase in incidence. And for Parkinson disease, prevalence is growing modestly, but this growth appears to be largely independent of changes in incidence or survival, likely reflecting demographic aging. These distinct drivers have major implications for healthcare planning and for etiologic research. As we search for therapies to alter the course of these diseases, we need better tools to measure treatment effects. A consensus statement from the Ataxia Global Initiative, published in *Nature Reviews Neurology*, provides evidence-based recommendations for using MRI as a biomarker in clinical trials for the most common spinocerebellar ataxias and Friedreich ataxia [7]. The working group critically reviewed candidate MRI endpoints and provides disease-specific guidance on which imaging readouts are ready for use in trials to monitor treatment and select participants, including in pre-symptomatic individuals. This work is essential for improving the feasibility and success rate of upcoming gene therapies and other disease-modifying treatments for these rare disorders. Drilling down to the cellular level, a paper in *Science Translational Medicine* provides a key mechanistic link between TDP-43 pathology and neuronal dysfunction in ALS and FTD [10]. It is known that the loss of TDP-43 from the nucleus leads to errors in RNA splicing, creating so-called cryptic exons. This study identifies several new genes affected by this process that are critical for membrane excitability and synaptic function. In human stem cell-derived neurons, TDP-43 reduction led to cryptic splicing in these genes, causing impaired neuronal firing and synaptic transmission. Crucially, using antisense oligonucleotides to suppress these individual splicing errors partially restored neuronal function, and targeting multiple events at once almost fully rescued the synaptic deficits. This provides strong evidence that cryptic splicing isn't just a marker of disease, but a direct driver of the neuronal dysfunction that causes neurodegeneration.

Finally, we cover a trio of papers highlighting new therapeutic approaches and fundamental mechanisms. First, a feasibility trial in *Nature Medicine* explored cervical epidural spinal cord stimulation for chronic upper limb hemiparesis after stroke [4]. Seven participants with profound motor deficits received implants for four weeks. The study found the procedure to be safe and feasible. With stimulation turned on, motor function improved immediately, with an average 32% increase in strength. Notably, even with a very low dose of motor activity—less than 6 hours with stimulation on—participants showed sustained functional improvements by the end of the study, and spasticity decreased in all of them. While a very small, early-phase study, these findings suggest cervical SCS has potential as an implantable neuroprosthetic to assist arm and hand function in daily life. From therapy to tumor biology, a study in *Nature* used single-nucleus sequencing to map how IDH-mutant gliomas progress and recur after treatment [8]. Researchers found that malignant cells exist in states that resemble either normal neural development or a reactive, mesenchymal-like state. They identified two major patterns of progression at recurrence. The first involves acquired genetic changes, such as treatment-associated hypermutation, which pushes tumor cells toward a less differentiated, more proliferative state. The second pattern is an increase in the mesenchymal-like state, which occurs independently of new genetic alterations and instead correlates with an increase in macrophages in the tumor microenvironment. This provides a detailed model of both cell-intrinsic and extrinsic factors driving glioma recurrence. And lastly, a study in *Science* reveals a fascinating link between memory and sleep regulation [5]. Working with mice, researchers found that the content of a memory determines its effect on sleep. Spontaneous reactivation of a negative memory, such as one associated with a foot shock, promoted arousal and disrupted sleep. In contrast, reactivating a positive memory, like a pleasant social interaction, supported sleep stability. This effect is mediated by experience-specific circuits between the hippocampus and amygdala. In a mouse model of chronic stress, this negative memory reactivation was shown to drive sleep disturbance, and specifically suppressing that reactivation restored normal sleep. This establishes a mechanism of memory-dependent sleep regulation that could inform our understanding of sleep problems in conditions like PTSD.

If you only have time for one paper this week, make it the Australian cohort study on in-utero antiseizure medication exposure in *Neurology* [1]. It provides crucial, comparative neurocognitive safety data on commonly used second- and third-generation agents, directly informing the nuanced counseling required for women with epilepsy of childbearing potential, particularly regarding the risks associated with levetiracetam and topiramate.

Here are the key takeaways from this week in Neurology: First, in counseling pregnant women with epilepsy, new evidence suggests lamotrigine is associated with neurocognitive outcomes comparable to unexposed children, while topiramate, levetiracetam, valproate, and carbamazepine are linked to cognitive vulnerabilities. Second, patients from rural areas in the United States face significantly higher mortality and morbidity from epilepsy. This disparity appears driven by modifiable structural and socioeconomic factors, not geography alone, pointing to a public health crisis. Third, in patients with advanced chronic kidney disease, do not use low-dose rivaroxaban for cardiovascular protection. The TRACK trial showed it offers no benefit and significantly increases the risk of major bleeding. Fourth, a history of migraine in postmenopausal women is a risk marker for ischemic stroke, even if the risk for total stroke is not significantly increased. This can help refine risk assessment in this population. Finally, early evidence suggests cervical spinal cord stimulation is a promising and feasible neuroprosthetic approach to immediately and durably improve arm and hand function after stroke.

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.

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