This Week in Neurosurgery — Aug 26, 2026

Welcome to This Week in Neurosurgery. This week we're covering 10 notable papers spanning acute neurotrauma and stroke systems of care, cranial nerve and skull base surgery, and outcome prediction in cranial and spinal practice. Let's dive in.

We start with time, triage, and the systems around the acutely injured brain. In World Neurosurgery, Shanshan and colleagues report a retrospective multicentre cohort of 726 adults with aneurysmal subarachnoid haemorrhage across seven comprehensive stroke centres in China, comparing those who reached an endovascular-capable hospital within eight hours of symptom onset against those who arrived later [1]. After adjustment for age, Hunt-Hess grade, modified Fisher grade and other confounders, delayed arrival was associated with roughly a fivefold increase in the odds of a poor functional outcome at six months, and about two and a half times the odds of death by twelve months. Delayed patients also had more brain herniation, hydrocephalus, ischaemic stroke and rebleeding. This is observational, and the possibility that sicker or more remote patients simply take longer to arrive is hard to fully exclude — but the eight-hour mark is a concrete, auditable target for regional stroke network design, and it argues for direct routing to endovascular-capable centres rather than sequential transfer. Alongside that, the Journal of Neurosurgery: Pediatrics brings a prospective cohort of 425 children with non-penetrating traumatic brain injury, in which Reisner and colleagues measured a panel of blood biomarkers within twenty-four hours of injury and asked whether they predicted the need for neurosurgical intervention [10]. About one in seven children underwent craniotomy, craniectomy, external ventricular drain or intracranial pressure monitor placement. All the biomarkers were higher in that group, but after controlling for age and admission Glasgow Coma Scale score, glial fibrillary acidic protein stood out, with an area under the curve of 0.83 — discrimination that held independent of the clinical variables we already use. This is proof-of-concept from a single tertiary centre, not a triage algorithm ready for deployment, but it is the kind of data that will push paediatric biomarker thresholds into future guidelines.

Staying with severe head injury, World Neurosurgery published a prospective multicentre randomised controlled trial from two Malaysian hospitals that takes aim at a reflex habit — indefinite proton pump inhibitor prophylaxis in the intensive care unit [2]. Selvam and colleagues enrolled 145 patients aged twelve and over with severe traumatic brain injury and a Glasgow Coma Scale score of eight or less, gave everyone a proton pump inhibitor while nil by mouth, and then randomised them once enteral feeding was established and tolerated to feeding alone or feeding plus continued prophylaxis for five days. There were no clinically important gastrointestinal bleeds in either arm. Overt bleeding occurred in three patients on feeding alone and two on feeding plus the proton pump inhibitor — no meaningful difference. Daily mean intragastric pH stayed above five in both groups throughout. The honest caveat, which the authors state plainly, is that this is a five-day observation window in a modest sample, so longer-term safety is unaddressed. Still, for a practice driven largely by inertia, this supports stopping the proton pump inhibitor once feeds are running, and reducing the downstream pneumonia and Clostridioides difficile exposure that goes with prolonged acid suppression.

The second theme is cranial nerve preservation and skull base technique, where two papers on vestibular schwannoma land in the same week from different journals with a genuine strategic disagreement between them. In the Journal of Neuro-Oncology, Liu and colleagues report 146 patients operated by a single surgeon using a subperineural onion-peeling dissection that leaves a thin layer of perineurium over the facial nerve as an anatomic buffer, with mean follow-up close to seven and a half years [8]. They achieved House-Brackmann grade one or two facial function in about 95 percent of patients while removing at least 95 percent of tumour in over 97 percent, with a cerebrospinal fluid leak rate under three percent and recurrence in about three percent. Their argument is that the fashionable alternative — planned subtotal resection followed by radiosurgery — trades a larger residual and a radiation exposure for a facial nerve benefit that maximal safe microsurgery can deliver anyway. Set that against the Journal of Neurosurgery paper from Khaleghi and colleagues, a decade of multidisciplinary practice across 187 patients and three standard approaches [7]. Favourable facial nerve function reached about 87 percent at long-term follow-up, gross total resection was achieved in about two thirds of cases, and serviceable hearing was preserved in around 42 percent of patients when preservation was attempted. Their contribution is predictive: the presence of fundal cerebrospinal fluid space roughly tripled the odds of good early facial outcome and predicted better results specifically in retrosigmoid cases, while heterogeneous contrast enhancement predicted worse facial outcomes in translabyrinthine cases. Notably, surgeon learning curve stage was not associated with overall facial nerve outcomes in a multidisciplinary team setting. Read together, the message is that preoperative magnetic resonance features should inform which approach you choose and what you promise the patient — and that the case for planned subtotal resection is weaker than its popularity suggests in high-volume hands.

Also in the cranial nerve space, World Neurosurgery reports 200 consecutive endoscope-assisted microvascular decompressions for hemifacial spasm from Lee and colleagues, using the endoscope to inspect the neurovascular relationship before and after microscopic decompression [5]. Complete spasm resolution reached 92 percent at a median follow-up of twenty months, abnormal muscle response disappeared intraoperatively in about 92 percent, and permanent facial weakness and permanent deafness were each in the low single digits. No complication was attributable to the endoscope itself. This is a single-arm series without a microscope-only comparator, so it establishes feasibility and safety rather than superiority — but it makes the hybrid workflow hard to argue against on safety grounds.

Our third theme is where established surgical dogma meets contemporary outcome data. The Journal of Neurosurgery publishes a deliberately provocative reassessment of supplementary motor area syndrome from Dono and colleagues, reviewing eighty patients who underwent dorsomedial frontal resections over two decades [3]. New postoperative motor deficits occurred in about a third of patients, but once cases involving the primary motor cortex or cingulate gyrus were excluded, the incidence of true supplementary motor area-specific deficit was only ten percent. More than four in five new deficits resolved, at a median of thirteen days, and permanent deficits occurred in about six percent — those tied to preoperative weakness and major postoperative complications rather than to the extent of supplementary motor area resected. There was no association with intraoperative mapping use or with the percentage of supplementary motor area removed. The authors' claim is that persistent deficits reflect injury to adjacent eloquent structures, not the supplementary motor area itself, and that with navigation, gravity-based retraction and modern mapping, this region can be approached more aggressively than tradition allows. In the same journal, Helal and colleagues report 26 consecutive patients undergoing nerve root ligation for cerebrospinal fluid–venous fistulas causing spontaneous intracranial hypotension [6]. Nearly 85 percent improved symptomatically, 94 percent improved on the Bern score, and about 83 percent returned to work, with a single patient developing radicular pain needing further intervention. Preoperative Bern score correlated with both outcome and quality of life — a usable preoperative selection tool in a disease we are still learning to treat.

Two papers round out the week on perioperative decision-making. In Neurosurgery, Gauhar and colleagues reviewed nearly two thousand craniotomies at a centre with an institutional postoperative systolic target below 140, comparing patients who required intravenous nicardipine against those who did not [4]. After propensity weighting, nicardipine exposure was associated with about a third higher odds of a composite of acute kidney injury, troponin or natriuretic peptide elevation, stroke or death, and with hospital stays roughly a day and a half longer. This is retrospective and confounded by indication — needing nicardipine marks a sicker, more hypertensive patient — but it should at least prompt scrutiny of reflexive sub-140 targets after tumour craniotomy. And in the Journal of Neurosurgery: Spine, Hannon and colleagues analysed a prospective registry of 1,713 anterior cervical spine patients using the Eating Assessment Tool-10 [9]. Revision surgery for adjacent segment disease did not raise baseline dysphagia rates, but it independently predicted dysphagia at three months and slower recovery in patients who came in already symptomatic. Within revision cases, baseline dysphagia, C3-4 exposure and re-exposure of prior levels raised risk, while an otolaryngologist-assisted approach lowered it — three concrete things to say in the consent conversation.

If you only have time for one paper this week, make it the randomised trial on stopping stress ulcer prophylaxis once enteral feeding is established in severe traumatic brain injury [2]. It is the only randomised evidence in the set, and it directly de-implements a daily habit in almost every neurosurgical intensive care unit.

Here are the key takeaways from this week in Neurosurgery. First, once your severe head injury patient is tolerating feeds, the proton pump inhibitor can likely come off — at least over the first five days. Second, reaching an endovascular-capable centre within eight hours of subarachnoid haemorrhage onset should be a network-level target, not an aspiration. Third, in vestibular schwannoma, maximal safe microsurgical resection with meticulous subperineural dissection delivers excellent facial nerve outcomes, and preoperative magnetic resonance features — especially fundal cerebrospinal fluid space — should guide your approach selection. Fourth, supplementary motor area syndrome as classically taught is likely overstated; deficits that persist usually reflect injury to neighbouring eloquent cortex. And fifth, be deliberate rather than reflexive about tight blood pressure targets after craniotomy, and counsel revision anterior cervical patients that swallowing recovery will be slower.

That's your roundup for This Week in Neurosurgery. 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.