Faculty Bibliography

OBJECTIVE:Ruptured, large pediatric arteriovenous malformations (AVMs) pose a significant management challenge due to high rerupture risk and the morbidity of conventional treatments. Volume-staged stereotactic radiosurgery (VS-SRS) has emerged as an alternative; however, evidence for its use in this specific population is limited. This study aimed to evaluate the long-term efficacy and safety of VS-SRS for ruptured, large AVMs in a dedicated pediatric cohort. METHODS:This international, multicenter, retrospective cohort study from 21 centers analyzed outcomes for 42 pediatric patients (age < 18 years) with previously ruptured, large AVMs treated with VS-SRS. The primary outcome was complete AVM obliteration, and secondary outcomes included post-SRS hemorrhage, radiation-induced changes (RICs), and favorable outcome. Favorable outcome in this study was defined as obliteration without post-SRS hemorrhage or permanent RIC. RESULTS:At the initial SRS, the median patient age was 14.5 years, and the median AVM volume was 15.0 cm3; most AVMs were high grade (Spetzler-Martin grades IV and V). With a median follow-up of 41.5 months, complete AVM obliteration was achieved in 21 patients (50.0%). The cumulative obliteration rate was 37% at 5 years and 56% at 10 years. Patients with a nidus volume of ≤ 10 cm3 had significantly higher cumulative obliteration rates than those with a nidus volume > 10 cm3 (log-rank test, p = 0.021). Similarly, patients treated with a prescription dose > 17 Gy showed significantly higher cumulative obliteration rates compared to those treated with a dose ≤ 17 Gy (log-rank test, p = 0.012). The cumulative 5-year incidences of hemorrhage and RICs following VS-SRS were 19% and 7%, respectively. In multivariable analysis, only larger total AVM volume was an independent predictor of a lower likelihood of favorable outcome (hazard ratio 0.89, 95% CI 0.80-0.99; p = 0.036). CONCLUSIONS:In this multicenter pediatric cohort, VS-SRS for ruptured, large AVMs provides a reasonable chance of long-term obliteration, but generally acceptable risks of post-SRS hemorrhage and RICs exist. VS-SRS should be considered as an option for pediatric patients with large, ruptured AVMs.

Deep venous drainage (DVD) is considered a negative prognostic factor in AVM surgery, yet its effect on postoperative functional decline remains incompletely defined. This study evaluates whether DVD predicts worsened functional status after surgical resection of Spetzler-Martin Grade II-III AVMs. This retrospective multicenter study analyzed 129 patients with Spetzler-Martin Grade II-III AVMs across nine centers in North America and Europe who underwent primary surgical resection. We excluded cases with prior endovascular or stereotactic interventions. The primary outcome measured was poor functional status, defined as modified Rankin Scale (mRS) score 3-6 at last follow up. Among 129 patients with Spetzler-Martin Grade II-III AVMs, 38 (29.5%) exhibited deep venous drainage (DVD). Poor functional outcome (mRS ≥ 3) at last follow-up occurred in 14 patients (10.9%). This occurred in 6 of 38 patients with DVD (15.8%) compared with 8 of 91 without DVD (8.8%; Fisher's exact p = 0.244). On univariate Firth-penalized logistic regression, DVD was not significantly associated with poor outcome (OR 1.96, 95% CI 0.65-5.89; p = 0.228). In the primary reduced Firth model adjusting for age and pre-existing functional disability, DVD was independently associated with poor outcome (OR 6.87, 95% CI 1.07-44.20; p = 0.042). Increasing age (OR 1.08 per year, 95% CI 1.02-1.13; p = 0.004) and pre-existing functional disability (OR 6.53, 95% CI 1.63-26.22; p = 0.008) were also independently associated with poor outcome. DVD is associated with functional decline following surgical resection of Spetzler-Martin Grade II-III AVMs after adjustment for age and pre-existing functional disability.

BACKGROUND AND OBJECTIVES/OBJECTIVE:General purpose vision-language models (VLMs) demonstrate impressive capabilities, but their opaque training on uncurated internet data poses critical limitations for high-stakes decision making, such as in neurosurgery. We present CNS-Obsidian, a neurosurgical VLM trained on peer-reviewed neurosurgical literature, and demonstrate its clinical utility compared with GPT-4o in a real-world setting. METHODS:We compiled 23 984 articles from Neurosurgery Publications journals, yielding 78 853 figures and captions. Using GPT-4o and Claude Sonnet-3.5, we converted these image-text pairs into 263 064 training samples across 3 formats: instruction fine-tuning, multiple-choice questions, and differential diagnosis. We trained CNS-Obsidian, a fine-tune of the 34-billion parameter Large Language and Visual Assistant-Next model. In a blinded, randomized deployment trial at NYU Langone Health (August 30-November 30, 2024), neurosurgeons were assigned to use either CNS-Obsidian or a Health Insurance Portability and Accountability Act-compliant GPT-4o end point as a diagnostic copilot after patient consultations. Primary outcomes were diagnostic helpfulness and accuracy, assessed through user ratings and presence of the correct diagnosis within the VLM-provided differential, respectively. RESULTS:CNS-Obsidian matched GPT-4o on synthetic questions (76.13% vs 77.54%, P = .235), but only achieved 46.81% accuracy on human-generated questions vs GPT-4o's 65.70% (P < 10-15). In the randomized trial, 70 consultations were evaluated (32 CNS-Obsidian, 38 GPT-4o) from 959 total consults (7.3% utilization). CNS-Obsidian received positive ratings in 40.62% of cases vs 57.89% for GPT-4o (P = .230). Both models included correct diagnosis in approximately 60% of cases (59.38% vs 65.79%, P = .626). CONCLUSION/CONCLUSIONS:Domain-specific VLMs trained on curated scientific literature can approach frontier model performance in specialized medical domains despite being orders of magnitude smaller and less expensive to train. This establishes a transparent framework for scientific communities to build specialized artificial intelligence models. However, low clinical utilization suggests chatbot interfaces may not align with specialist workflows, indicating need for alternative artificial intelligence integration strategies.

OBJECTIVE:The aim of this study was to evaluate freedom from tumor progression and clinical outcomes in older adults with small- to medium-sized vestibular schwannoma (VS) managed by observation versus stereotactic radiosurgery (SRS) to better inform optimal management in this patient population. METHODS:In this international multicenter study, patients aged ≥ 60 years with Koos grade I or II VS managed by observation or SRS were retrospectively reviewed. Propensity score matching was conducted using patient characteristics, tumor size, and hearing assessments. Outcome measures of freedom from tumor progression, serviceable hearing preservation (SHP), and neurological function (tinnitus, vestibulopathy, House-Brackmann grade, and trigeminal nerve function) were assessed for both groups. RESULTS:The observation and SRS groups each comprised 51 matched patients (median age 68 years for both). Ipsilateral serviceable hearing was observed at presentation for 35 patients in each group. The median follow-up duration was 39 months in the observation group versus 27 months in the SRS group (p = 0.5). Tumor progression was significantly lower with SRS than with observation (2% vs 52.9%, p < 0.001). The rate of 5-year freedom from tumor progression was 100% in the SRS group versus 43% (95% CI 29%-64%) in the observation group. The rate of 10-year freedom from tumor progression was 90% (95% CI 73%-100%) in the SRS group versus 20% (95% CI 8.5%-49%) in the observation group. At the last follow-up, hearing loss occurred in 42.9% of the observation group and 51.4% of those who underwent SRS (p = 0.5). The 3-year SHP rate was 68% (95% CI 53%-88%) versus 65% (95% CI 49%-85%) (p = 0.8), and the 5-year SHP rate was 53% in both groups (95% CI 36%-78% in the observation group and 95% CI 35%-79% in the SRS group, p = 0.8). The composite endpoint of tumor progression and/or worsened neurological outcome, including hearing loss, tinnitus, vestibulopathy, facial nerve dysfunction (House-Brackmann grade), or trigeminal dysfunction, demonstrated a significantly lower rate in the SRS group (17.6%) compared with the observation group (66.7%) (p < 0.001). CONCLUSIONS:SRS in older patients for management of Koos grade I or II VS resulted in significantly superior rates of freedom from tumor progression, comparable hearing preservation rates, and significantly higher rates of favorable overall radiographic and neurological outcomes than observation alone. Compared with observation, SRS might be the preferred management option in this patient population.

BACKGROUND AND OBJECTIVES/OBJECTIVE:articles into graphical abstracts using Cascade Styling Sheets (CSS) templates and iterative prompting of a frontier vision language model and to conduct a human evaluation of this pipeline. METHODS:We developed an automated pipeline to convert extracted manuscript content into standardized graphical abstracts. The pipeline implements a custom CSS profile designed to match existing journal standards. Using Claude Sonnet-3.5, we generated structured hypertext markup language summaries organized into 6 sections: Objectives, Background, Methods, Results, Discussion, and Conclusion. The model selected up to 2 representative figures per manuscript based on caption analysis. We evaluated performance using 100 randomly selected articles published between 2020 and 2024 (95 from Neurosurgery, 4 from Operative Neurosurgery, 1 from Neurosurgery Practice). Three Editorial Review Board members independently assessed abstracts using 3 binary criteria: (1) proper formatting, (2) factual accuracy, and (3) visual appeal. RESULTS:Generated graphical abstracts achieved proper formatting in 85% of cases (95% CI: 76.7%-90.7%), factual accuracy in 99% (95% CI: 94.4%-99.9%), and visual appropriateness in 82% (95% CI: 73.3%-88.3%). Overall, 70% of abstracts (95% CI: 60.5%-78.1%) met all 3 criteria and were deemed "publication ready" without manual intervention. Error analysis revealed poor figure selection (40.0%) as the most common failure mode, followed by title replacement errors from PDF extraction (26.7%). CONCLUSION/CONCLUSIONS:Our artificial intelligence-CSS pipeline demonstrates the feasibility of automating graphical abstract generation for neurosurgical manuscripts, achieving publication-ready quality in 70% of cases with 99% factual accuracy. This technology offers a scalable augmentation tool that can reduce the design burden for authors, enhancing visual scientific communication in neurosurgical publishing while complementing human expertise.

BACKGROUND AND OBJECTIVES/OBJECTIVE:Developing precise metrics for clinical use is vital to translating outcomes data to practice. Risk probabilities, such as the complication rate of surgery or the lifetime risk of an aneurysm rupture, are widely used for counseling patients, but their comparison may be misleading if risks are spread over different time horizons. This study evaluates a new risk-assessment approach called risk-weighted impact (RWI) that applies event probabilities to estimate the average number of years of life impacted by event occurrence. METHODS:Decision-making policies based on RWI and cumulative lifetime event risk were applied to determine management in a simplified model of incidental cerebral aneurysms through Monte Carlo simulation (1000 iterations of 10 000 synthetic patients). In addition, a web-based application was created to simplify risk-assessment calculations and comparisons. RESULTS:When treatment of incidental cerebral aneurysms was simulated using both risk assessment methods, there was disagreement in 25.2% (95% CI: 24.4%-26.1%) of cases, with the RWI policy preferring observation, while event-risk policy preferred intervention. In these patients, the number of poor outcomes was nearly the same, 110 (95% CI: 91-129) in RWI policy and 111 (95% CI: 90-132) in event-risk policy, but the RWI policy resulted in 874.8 fewer quality-adjusted life year lost (95% CI: 299.5-1466.6) due to adverse events occurring an average of 11.3 years later (95% CI: 8.2-14.1 years). CONCLUSION/CONCLUSIONS:Only using cumulative lifetime event risks may understate the impact of an up-front treatment given that a larger proportion of risk is assumed at an earlier age, when more years of life are in jeopardy. RWI offers an alternative approach to thinking about risk, using the same inputs (event probabilities and life expectancy) to compare estimated patient impact. RWI is more aligned with clinical objectives and is a valuable metric for risk assessment and decision making.

BACKGROUND AND OBJECTIVES/OBJECTIVE:The initial management of symptomatic cavernous sinus meningioma (CSM) remains a matter of debate. This study compares long-term tumor control and neurological outcomes in patients who underwent either primary stereotactic radiosurgery (SRS) or initial surgery followed by adjuvant SRS for both residual and progressive disease. METHODS:We identified 292 patients with CSM who underwent Gamma Knife radiosurgery between 1987 and 2024 (median follow-up, 77 months). The initial management included 167 patients who underwent primary SRS and 125 patients who had adjuvant SRS after surgical resection. Propensity score matching (1:1) was performed to control for potential selection bias differences, resulting in 150 matched patients. The primary outcomes included tumor control, neurological improvement, and neurological worsening. Kaplan-Meier analysis and logistic regression were performed. RESULTS:Compared with adjuvant SRS, primary SRS was associated with enhanced tumor control (log-rank P = .049) at 5 years (96.5% vs 90.5%) and 10 years (89.5% vs 77.8%). Neurological improvement rates were greater in the primary SRS group (10-year: 39.3% vs 25.2%, log-rank P = .008; odds ratio = 2.84, 95% CI: 1.33-6.07, P = .01), often related to improved abducens nerve function (log-rank P = .025) in primary SRS patients. The risk of neurological worsening was no different between primary and adjuvant SRS patients (P = .995). CONCLUSION/CONCLUSIONS:This study provides further evidence that primary SRS is a better option than initial surgery followed by adjuvant SRS for the initial management of patients with symptomatic CSM.

BACKGROUND AND OBJECTIVES/OBJECTIVE:Traditional medical board examinations present clinical information in static vignettes with multiple-choices (MC), fundamentally different from how physicians gather and integrate data in practice. Recent advances in large language models (LLMs) offer promising approaches to creating more realistic clinical interactive conversations. However, these approaches are limited in neurosurgery, where patient communication capacity varies significantly and diagnosis heavily relies on objective data such as imaging and neurological examinations. We aimed to develop and evaluate a multi-artificial intelligence (AI) agent conversation framework for neurosurgical case assessment that enables realistic clinical interactions through simulated patients and structured access to objective clinical data. METHODS:We developed a framework to convert 608 Self-Assessment in Neurological Surgery first-order diagnosis questions into conversation sessions using 3 specialized AI agents: patient AI for subjective information, system AI for objective data, and clinical AI for diagnostic reasoning. We evaluated generative pretrained transformer 4o's (GPT-4o's) diagnostic accuracy across traditional vignettes, patient-only conversations, and patient + system AI interactions, with human benchmark testing from 10 neurosurgery residents. RESULTS:= .0030) using fewer interactions and reported high educational value of the interactive format. CONCLUSION/CONCLUSIONS:This multi-AI agent framework provides both a more challenging evaluation method for LLMs and an engaging educational tool for neurosurgical training. The significant performance drops in conversational formats suggest that traditional MC testing may overestimate LLMs' clinical reasoning capabilities, while the framework's interactive nature offers promising applications for enhancing medical education.