This Week in Neurosurgery — Aug 6, 2026

Welcome to This Week in Neurosurgery. This week we're covering 10 notable papers spanning haemorrhagic and vascular neurosurgery, skull base and neuro-oncology outcomes, and the systems-level question of who actually gets good care. Let's dive in.

We'll start with blood in the brain, and the perennial anticoagulation question. In Neurosurgery, a five-centre United Kingdom retrospective series of one thousand three hundred fifty-five patients operated on for chronic subdural haematoma between 2011 and 2023 addresses something almost every on-call registrar argues about weekly [1]. Roughly forty percent of the cohort were on antiplatelet or anticoagulant therapy. Crude reoperation rates were modestly higher in that group, about twelve percent versus nine percent, but that difference evaporated after adjustment for comorbidity, with a confidence interval comfortably crossing one. More importantly for practice, neither holding the drug for longer before surgery nor restarting it earlier afterwards was independently associated with reoperation. What did differ was function: only about half of the anticoagulated or antiplatelet patients achieved a modified Rankin Scale of zero to two, compared with two thirds of those on no such therapy, and that difference persisted after adjustment. The honest reading is that these patients do worse because they are sicker, not because of the drug timing we agonise over, and that prolonged preoperative interruption may buy less than we think. It remains retrospective, so it is permission to be pragmatic rather than a protocol.

Staying with intracranial haemorrhage, World Neurosurgery reports one hundred thirty-six consecutive endoscopic evacuations for spontaneous intracerebral haemorrhage from a single Japanese stroke centre, using a dilated burr-hole and a transparent sheath [3]. Median age was seventy-two, one in five patients were on antithrombotics, and the median evacuation rate was ninety percent. Technically it is impressive: no postoperative rebleeding required surgery and there were no reoperations. Functionally, the picture is more sobering and location-dependent. Overall, thirty-two percent reached a modified Rankin Scale of zero to three at ninety days, with cerebellar haemorrhages doing best at fifty-six percent and thalamic haemorrhage with intraventricular extension worst at seventeen percent. Ninety-day mortality was fifteen percent. Higher admission Glasgow Coma Scale, younger age, smaller haematoma, better premorbid function and non-thalamic location all independently predicted a favourable outcome. The clinical message is that the endoscope reliably removes clot safely across locations, but selection, not technique, still drives outcome, and thalamic bleeds with intraventricular haemorrhage should temper expectations in family discussions.

The systems side of haemorrhage care comes from the Journal of Clinical Neuroscience, in a Brazilian public referral centre cohort of seventy-four consecutive patients with aneurysmal subarachnoid haemorrhage [8]. This was a sick population: forty-five percent World Federation of Neurosurgical Societies grade four or five, seventy-three percent modified Fisher grade four, sixty-four percent hydrocephalus. Aneurysm treatment itself was fast, seventy-three percent endovascular with a median of one day from admission to treatment. But only twenty-eight percent reached an intensive care unit within forty-eight hours, and thirty-eight percent never received intensive care at all. Delayed cerebral ischaemia occurred in forty-three percent, ventriculitis in twenty-two percent, infectious complications in fifty-seven percent. In-hospital mortality was forty-two percent and two thirds had unfavourable three-month outcomes. Read alongside the chronic subdural data, the theme is consistent: securing the lesion is the part we have optimised, and the neurocritical care that follows is where outcome is now won or lost.

The vascular access story is smaller but practical. Also in the Journal of Clinical Neuroscience, a single-centre comparison of four hundred seventy diagnostic cerebral angiograms via distal versus proximal radial access found that the distal approach was genuinely harder — more puncture attempts, longer puncture time, first-pass success of only about nineteen percent versus thirty-five, and a crossover rate of eleven versus six percent [7]. Yet clinically assessed radial artery occlusion was substantially lower with distal access, two and a half percent versus nearly eight percent, and proximal access carried roughly a threefold higher adjusted odds of both occlusion and a composite access-site event. In a single operator's first hundred distal cases, cumulative sum analysis suggested most of the learning effect happened within the first ten to fifteen cases. So the trade is a fiddlier puncture for a better preserved radial artery, and the learning curve is short — relevant if you want to keep that vessel available for future access.

Turning to tumours and functional cranial surgery, the Journal of Neuro-Oncology carries an International Radiosurgery Research Foundation analysis of two hundred ninety-six craniopharyngioma patients from thirteen centres, with a median tumour volume of just over one cubic centimetre and a median margin dose of twelve gray [2]. Actuarial local control was ninety-three percent at one year, seventy-six percent at five, and seventy percent at ten, with ten-year overall survival of eighty-five percent. Crucially, morbidity was low: visual fields deteriorated in only four percent, acuity in three percent, and ten-year freedom from endocrine deterioration was ninety-seven percent. Mixed solid-cystic tumours did significantly worse, with non-mixed phenotype independently associated with better control. That is a useful counselling number for the residual or recurrent craniopharyngioma sitting in clinic, and a reminder that the cystic component is the part that will likely bring the patient back.

The same journal reports a Czech national paediatric ependymoma cohort of sixty-three patients that should change how you interpret an outside histology report [4]. DNA methylation array profiling confirmed ependymoma in seventy-six percent but reclassified nearly a quarter of patients as entirely different tumour entities. Strikingly, the reclassification rate was thirty-six percent before 2019 and only seven percent after, showing how much modern integrated diagnostics have tightened accuracy. Ten-year overall survival was seventy percent with event-free survival under fifty percent, and gross total resection was the only factor significantly associated with improved survival — seventy-five percent versus forty percent. Chromosome 1q gain, found only in posterior fossa group A tumours, marked a high relapse risk despite standard therapy. If you operate on paediatric posterior fossa tumours, the surgical imperative for complete resection remains, and methylation profiling is no longer optional.

Two further oncology papers address adjuvant strategy. Also in the Journal of Neuro-Oncology, a retrospective comparison of one hundred sixty-six resected brain metastases found postoperative sinus vein thrombosis in five patients overall, with no significant difference between intraoperative radiotherapy at under three percent and postoperative stereotactic radiosurgery at just over three percent [6]. All thromboses were in sinus-adjacent lesions and all were clinically asymptomatic. That addresses a specific reluctance to use intraoperative radiotherapy near the sinuses, though the event count is small. Complementing this, World Neurosurgery reports three hundred one surgically resected brain metastases from Beijing Tiantan Hospital, where infratentorial location nearly doubled mortality hazard, and multiple metastases and older age were adverse [10]. Protective factors were postoperative immune or targeted therapy, which reduced hazard by roughly two thirds, adjuvant whole-brain radiotherapy, and higher discharge Karnofsky score. Their nomogram achieved a C-index of about zero point six nine — modest discrimination, but the message that modern systemic therapy dominates prognosis after resection is the actionable part.

Two papers round out the week on precision and equity. In Operative Neurosurgery, one hundred forty patients undergoing microvascular decompression for hemifacial spasm were analysed for lateral spread response morphology and compound muscle action potential ratios [5]. Delayed facial palsy occurred in about eight percent, and those patients had significantly lower compound muscle action potential ratios as early as one week postoperatively. Monophasic lateral spread waveforms showed a transient dip in ratio at one month, suggesting more severe axonal injury, while polyphasic waveforms improved steadily. Practically, a one-week electrophysiological check may flag who needs proactive counselling and early corneal protection. Finally, a national United States analysis in the Journal of Neuro-Oncology of over forty-four thousand pituitary adenoma patients found that lower community-level socioeconomic status was associated with roughly forty percent higher all-cause mortality and a higher likelihood of treatment delay beyond three months, with sex also emerging as an independent factor across mortality, surgery receipt, and delay [9]. Community context mattered more than individual factors alone.

If you only have time for one paper this week, make it the multicentre chronic subdural haematoma anticoagulation study in Neurosurgery [1]. It is the largest dataset yet on a decision nearly every neurosurgeon makes weekly, and it argues that prolonged drug interruption is probably not protecting your patient from reoperation.

Here are the key takeaways from this week in Neurosurgery. First, in chronic subdural haematoma, antiplatelet and anticoagulant use predicts worse function largely through comorbidity, and neither longer preoperative interruption nor delayed reinitiation reduced reoperation. Second, endoscopic evacuation of intracerebral haemorrhage is technically safe with high clot clearance, but thalamic haemorrhage with intraventricular extension carries a poor functional prognosis. Third, aneurysmal subarachnoid haemorrhage outcomes in resource-limited systems are constrained by neurocritical care access, not aneurysm treatment speed. Fourth, stereotactic radiosurgery gives durable craniopharyngioma control with low visual and endocrine morbidity, though mixed solid-cystic tumours fare worse. Fifth, methylation profiling reclassified nearly a quarter of paediatric ependymomas, and gross total resection remains the only modifiable survival factor. And finally, distal radial access halves radial artery occlusion despite a harder puncture, with most of the learning curve behind you after fifteen cases.

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 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.