This Week in Neurosurgery — Aug 12, 2026

Welcome to This Week in Neurosurgery. This week we're covering 10 notable papers spanning cerebrovascular risk prediction and hemorrhage management, skull base and neuro-oncology decision-making, and spine, pain, and functional neurosurgery. Let's dive in.

We start with the cerebrovascular papers, where the common thread is deciding how aggressively to investigate and treat lesions whose natural history is uncertain. In the Journal of Neurosurgery, a group building on two prospective multicentre cohorts developed and externally validated a nomogram for five-year symptomatic hemorrhage risk in sporadic cerebral cavernous malformations [1]. Across 331 training patients and 57 external validation patients, five factors carried independent weight: prior hemorrhage, lesion size above one and a half centimetres, an associated developmental venous anomaly, brainstem location, and Zabramski type one appearance. Brainstem location was the single heaviest driver, associated with roughly a fourfold higher hazard, with type one lesions and lesions over one and a half centimetres each associated with roughly a threefold increase. The model discriminated well, with a concordance index of 0.81 in training and 0.87 in external validation, and the authors were able to separate a genuinely high-risk stratum from a low-risk one. For the clinic, this is a tool for the conversation rather than a mandate for surgery: a patient with a small, non-brainstem, previously silent cavernoma sits in a very different risk tier from one with a large hemorrhagic brainstem lesion, and now you can show them that difference. The external cohort is small, so treat the calibration as promising rather than settled.

Staying with hemorrhage, Neurosurgery published a retrospective single-institution cohort of 202 patients with initial catheter angiogram-negative subarachnoid hemorrhage that pushes back on the idea that this is a uniformly benign condition [2]. Cluster analysis separated non-cisternal, perimesencephalic, and diffuse bleeding patterns. Delayed cerebral ischemia was rare overall, occurring in three patients, about one and a half percent, and only in the perimesencephalic and diffuse groups. But nearly eleven percent developed hydrocephalus requiring temporary cerebrospinal fluid diversion, and two and a half percent went on to a permanent shunt. The diagnostic yield of repeat angiography beyond the index study was just under four percent, and essentially every positive repeat study was in a patient with diffuse blood — with one exception involving focal cortical blood. Meanwhile seven patients had a complication from repeat angiography, and two of those suffered a permanent neurological deficit from periprocedural stroke. Magnetic resonance imaging and magnetic resonance angiography of the brain found no missed vascular lesion in anyone. The practical message is a bleeding-pattern-driven protocol: repeat angiography for diffuse or focal cortical patterns, where the yield justifies the stroke risk, and restraint in classic perimesencephalic bleeds — while still watching those patients for hydrocephalus rather than discharging them as low-risk.

The Journal of Neurosurgery also published the largest series to date of Cognard type five dural arteriovenous fistulas, drawn from the multinational CONDOR registry across 16 academic centres [3]. Only 37 of 1,077 fistula patients, about three percent, had perimedullary venous drainage. Most presented with non-hemorrhagic neurological deficits, and the foramen magnum was the commonest location. Embolization was the primary treatment in two-thirds, achieving complete obliteration in 71 percent; upfront microsurgery, used in about a fifth, obliterated 88 percent, while radiosurgery succeeded in only 40 percent. Complications were low — transient in three patients and permanent in one — and at a mean of about two and a half years, roughly seven in ten patients had a modified Rankin score of two or better, with no deaths. Notably, the mRS declines that did occur were attributed to causes other than the fistula. This supports aggressive but tailored multimodal treatment, and it argues against radiosurgery as a primary strategy for these lesions.

Rounding out the vascular theme, World Neurosurgery reported on 180 consecutive patients with primary middle cerebral artery M2 occlusion treated with thrombectomy, comparing contact aspiration first with stent retriever first [4]. Procedure-related subarachnoid hemorrhage — defined broadly to include asymptomatic minor blood on 24-hour CT — occurred in 29 percent overall, and was more than twice as common with a stent retriever first approach, 39 percent versus 18 percent. Smaller vessel diameter, more device passes, and the stent retriever first strategy were each independently associated with hemorrhage, and the finding held across several sensitivity analyses. Importantly, reperfusion success and functional outcomes were similar between strategies, and the hemorrhage itself was not independently associated with worse outcome once complexity and baseline severity were accounted for. So this is a safety signal rather than an efficacy signal — reasonable grounds to favour aspiration first in small-calibre M2 branches without expecting a change in ninety-day function.

Turning to skull base and neuro-oncology, three papers address how much intervention is enough. An international study across 27 institutions, published in the Journal of Neurosurgery, examined single-fraction stereotactic radiosurgery for vestibular schwannoma in 309 octogenarians [5]. Tumor control reached about 95 percent at both three and five years, with progression-free survival of 71 percent at five years and overall survival of 78 percent. The competing risk analysis is the number to remember: by five years, only about five percent of patients had tumor progression, while nearly twenty percent had died without progression. Communicating hydrocephalus developed in four percent and was managed with cerebrospinal fluid diversion. Age, sex, tumor volume, Koos grade, and margin dose did not predict survival or progression. In other words, radiosurgery reliably controls the tumor in this age group, and competing mortality — not the schwannoma — dominates the outlook, which should temper how aggressively we treat small, asymptomatic lesions in very elderly patients.

Also in the Journal of Neurosurgery, a contemporary series of 67 facial nerve schwannoma surgeries in 61 patients tested a function-centred strategy: nerve-preserving subtotal removal with bony decompression for moderate palsy, and total removal with reconstruction for severe or complete palsy, guided by intraoperative electromyography and a five-category location-based classification [7]. Nerve preservation was chosen in about seventy percent of cases and left facial function improved or stable in 87 percent of those. Mean House-Brackmann grade improved from 3.5 to 3.0 with preservation and from 4.6 to 4.0 with reconstruction; overall, function improved in about half of cases and worsened in 13 percent. The trade-off is durability — over a mean follow-up of just over seven years, more than a quarter of the preservation group needed additional treatment. Tumors confined to the temporal bone, younger patients, and favourable electromyography responses did best.

Neurosurgery contributes a pragmatic answer to a recurring ward question: do preoperative steroids sabotage the biopsy in suspected primary central nervous system lymphoma [9]. In 50 histologically confirmed cases over a decade, 34 had received steroids before biopsy. Only one patient in the steroid group required repeat biopsy, and none in the non-steroid group did, with no statistically significant differences between groups. The single non-diagnostic case involved prolonged steroid exposure. The reasonable reading is that a short course of steroids for symptomatic mass effect is unlikely to cost you the diagnosis, while prolonged exposure — particularly in patients who could have a steroid-responsive inflammatory mimic — remains a genuine hazard. This is a small retrospective series, so it argues for pragmatism rather than complacency: get the tissue promptly and sample adequately.

Finally, spine, pain, and functional neurosurgery. In the Journal of Neurosurgery: Spine, a series of 114 patients with severe degenerative sagittal imbalance treated with anterior column realignment reported acute proximal junctional failure in nearly a third of patients within six months, with revision surgery in 14 patients — about 39 percent of failures and 12 percent of the whole cohort — at a mean of about five months [6]. Bony failure drove the great majority of revisions. Independent risk factors were a cranially directed screw angle at the upper instrumented vertebra, which carried by far the largest effect, greater preoperative thoracic kyphosis, and a larger postoperative proximal junctional angle. The actionable point is technical: screw trajectory at the upper instrumented vertebra is modifiable in the operating room, and in patients with substantial preoperative thoracic kyphosis, it deserves particular attention.

On the perioperative side, the Journal of Clinical Neuroscience published a network meta-analysis of 13 randomized trials and 879 adults undergoing instrumented spinal fusion, comparing ketamine and pregabalin as opioid-sparing adjuncts [8]. Against pregabalin 150 milligrams as reference, ketamine was associated with significantly lower opioid consumption in the first 24 hours — a difference of about 57 milligrams of morphine equivalents — while pregabalin 300 milligrams and control did not differ significantly from the reference. For 24-hour pain scores, only control differed, with modestly higher pain than pregabalin 150. Adverse events were infrequent and similar across arms. But only six trials contributed opioid data, and these are indirect comparisons with wide confidence intervals, so this is hypothesis-generating support for ketamine within an enhanced recovery protocol, not proof of superiority.

And in the Journal of Neurosurgery: Pediatrics, the CHILD-DBS registry reports on 12 children under seven years old — below the age threshold of the FDA Humanitarian Device Exemption — who underwent pallidal deep brain stimulation for dystonia across five North American centres [10]. Half had prior intensive care admissions for status dystonicus and nearly two-thirds of the operations were urgent. The Burke-Fahn-Marsden movement subscale fell by about 40 percent at six months and the benefit persisted at a year. Complications were not trivial: three children had stereotactic pin transgression beyond the inner table and four required surgical management of wound complications, though no child suffered neurological sequelae and infected hardware was successfully reimplanted. Three children died of their underlying conditions more than six months after surgery. The signal is that stimulation works in very young children with severe refractory dystonia, at the cost of higher surgical risk — relevant to both referral practice and the regulatory conversation.

If you only have time for one paper this week, make it the Neurosurgery study of catheter angiogram-negative subarachnoid hemorrhage [2]. It directly reshapes a common admission pathway — telling you which bleeding patterns justify repeat angiography and its real stroke risk, and reminding you that these patients still develop hydrocephalus.

Here are the key takeaways from this week in Neurosurgery. First, angiogram-negative subarachnoid hemorrhage is not uniformly benign; repeat angiography earns its keep in diffuse or focal cortical bleeding patterns, and hydrocephalus occurs in about one in ten. Second, for cavernous malformations, brainstem location, prior hemorrhage, size, an associated venous anomaly, and Zabramski type one define a genuinely high-risk group worth quantifying with the validated nomogram. Third, in M2 thrombectomy, a stent retriever first strategy roughly doubled procedure-related subarachnoid hemorrhage without any difference in reperfusion or functional outcome. Fourth, in octogenarians with vestibular schwannoma, radiosurgery controls the tumor in about 95 percent, but competing mortality far exceeds progression — treat the patient, not just the scan. Fifth, in adult deformity surgery using anterior column realignment, upper instrumented vertebra screw trajectory is the dominant modifiable predictor of revision for proximal junctional failure. And finally, a short course of preoperative steroids is unlikely to compromise lymphoma diagnosis, but prolonged exposure still can.

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.