Month: January 2026

Posted on

Customized Titanium Implants for Cranial Fibrous Dysplasia: A Personalized Surgical Approach and Single-Center Experience

Operative Neurosurgery 30:100–108, 2026

This single-center clinical series reports on 33 patients with monostotic cranial fibrous dysplasia treated with computer-assisted resection and reconstruction using patient-specific, 3D‑printed titanium cranioplasty implants. Methods detail high-resolution CT planning, surgeon‑engineer collaboration for CAD/CAM templates, intraoperative guidance, standardized perioperative care, and long-term radiological and FACE-Q patient‑reported outcome follow-up.

Results show complete tumor resections, favorable cosmetic outcomes, no postoperative complications or recurrences across extended follow-up, and high patient satisfaction; discussion addresses material selection, cost (≈$3000 per implant), production time, accessibility in low-resource settings, and the need for comparative studies.

Patient-Specific Titanium Implants: Customized, computer-aided design (CAD) and manufacturing (CAM) titanium implants enable precise resection and effective reconstruction for cranial fibrous dysplasia (CFD), resulting in excellent cosmetic outcomes and no postoperative complications or recurrences in a 33-patient series.

Surgical Process: Preoperative high-resolution CT scans and 3D modeling are used for surgical planning, including mirrored contralateral anatomy for symmetry; intraoperative templates guide tumor resection, and implants are secured with miniscrews.

Clinical Outcomes: All patients achieved complete tumor removal, satisfactory cosmetic results confirmed by postoperative CT, and no surgical site infections or cerebrospinal fluid leaks during a mean follow-up of 76–108 months.

Patient Satisfaction: High patient-reported satisfaction was observed, with FACE-Q scores indicating 84.8% of patients highly satisfied (score ≥85) and no patients dissatisfied with surgical outcomes.

Material Advantages: Titanium implants are favored for their biocompatibility, mechanical strength, and low infection rates compared to other materials like PEEK and PMMA, making them especially suitable for CFD reconstruction.

Cost and Accessibility: Average total cost per implant was $3000, with a 4-week production time; while feasible in middle-income settings, access remains limited in low-resource environments due to financial and infrastructural barriers.

Limitations: The study’s retrospective, single-center design, absence of a comparative cohort, and variable follow-up duration limit generalizability and long-term conclusions.

Future Directions: Broader adoption of CAD/CAM and patient-specific implants in craniofacial surgery may improve outcomes, but further comparative and cost-effectiveness studies are needed, especially in resource-constrained settings.

Posted on

Spinal versus general anesthesia in robotic minimally invasive transforaminal lumbar interbody fusion: a comparative study on surgical outcomes

J Neurosurg Spine 44:99–107, 2026

This clinical study compares spinal anesthesia (SA) versus general anesthesia (GA) for robot-assisted minimally invasive transforaminal lumbar interbody fusion (RA‑MIS TLIF), reporting retrospective outcomes from 209 patients treated 2018–2024. Primary findings show SA patients had significantly shorter operative times, lower immediate postoperative pain scores, reduced estimated blood loss, and shorter hospital length of stay after propensity score matching and regression adjustment.

The authors contextualize results within advances in robotic spinal surgery and awake spine techniques, discuss safety and potential cost and opioid‑reduction benefits, and acknowledge limitations including retrospective design, single‑center data, and reduced matched cohort size. Conclusions support SA as a safe, efficient approach for RA‑MIS TLIF with calls for larger prospective studies and formal patient‑selection guidelines.

Spinal Anesthesia (SA) vs General Anesthesia (GA): In robot-assisted minimally invasive transforaminal lumbar interbody fusion (RA-MIS TLIF), SA significantly reduces operative times, postoperative pain, and hospital length of stay compared to GA, with no increase in complications or adverse outcomes.

Robotic Assistance Benefits: Robotic technology in spine surgery improves pedicle screw placement accuracy, reduces radiation exposure, and is associated with lower complication and revision rates, enhancing surgical safety and efficiency.

Study Design: A retrospective analysis of 209 patients (31 SA, 178 GA) from 2018–2024, with propensity score matching applied to control for confounders, allowing fair comparison between SA and GA cohorts for single-level procedures.

Key Outcomes (After Matching): SA cohort had shorter median total OR time (159 vs 283 min), procedure time (115 vs 201 min), lower intraoperative blood loss (25 vs 50 mL), lower first postoperative pain scores (median VAS 0 vs 5), and reduced mean length of stay (0.90 vs 2.64 days) compared to GA.

Patient Selection: The choice between SA and GA was based on patient preference and eligibility, with all SA cases being single-level procedures and comparable baseline demographics after matching.

Safety Profile: No increase in intraoperative or postoperative complications was observed with SA; screw placement accuracy remained high with robotic assistance.

Implications for Practice: Combining SA with RA-MIS TLIF offers a safe, efficient, and patient-centered approach that may lower healthcare costs and opioid requirements by reducing pain and hospitalization.

Limitations: Single-center, retrospective design with a relatively small matched cohort may limit generalizability; further prospective, multicenter studies are needed to validate these findings.

Posted on

Erector spinae plane block during standalone anterior lumbar surgery: impact on early ambulation, length of stay, and inpatient opioid use

J Neurosurg Spine 44:90–98, 2026

This clinical retrospective study assesses the impact of erector spinae plane block (ESPB) as an adjunct to multimodal analgesia in standalone anterior lumbar procedures (ALIF and lumbar TDR). Outcomes compared between ESPB and non-ESPB cohorts include in-hospital pain scores, opioid consumption (MME), time to ambulation, length of stay (LOS), and opioid-related complications.

Results show ESPB associated with lower day-of-surgery pain scores, earlier ambulation, and shorter LOS, with reduced in-hospital oral MME in univariate analysis; preoperative opioid use predicted higher perioperative opioid consumption and urinary retention despite ESPB. The authors emphasize ESPB as a component of enhanced recovery protocols and note limitations of retrospective design and sample size.

Erector Spinae Plane Block (ESPB): ESPB is a regional analgesic technique used as an adjunct in anterior-only lumbar surgeries, such as ALIF and total disc replacement, aiming to improve perioperative pain control and recovery outcomes.

Reduced Pain and Opioid Use: ESPB significantly lowers pain scores on the day of surgery and reduces in-hospital opioid use, especially oral morphine milligram equivalents (MMEs), compared to patients not receiving ESPB.

Shorter Hospital Stay: Patients receiving ESPB experience a significantly shorter hospital length of stay (LOS) and are more likely to be discharged earlier, including same-day discharge, than those without ESPB.

Faster Ambulation: ESPB is associated with a significantly shorter time to first ambulation after surgery, facilitating earlier rehabilitation.

Predictors of Opioid Use: Baseline (preoperative) opioid use is the strongest predictor of higher perioperative opioid requirements and is also linked to a higher incidence of postoperative urinary retention, regardless of ESPB administration.

Subgroup Benefits: Among ESPB patients, those with a shorter LOS (<2 days) had earlier ambulation, lower opioid use, and lower pain scores on postoperative day 1, indicating enhanced early recovery.

Multimodal Pain Management: ESPB should be considered as one component within a comprehensive multimodal pain management strategy (such as ERAS protocols), rather than as a standalone intervention.

Study Limitations: The findings are limited by the retrospective design, potential selection bias, lack of randomization, and incomplete data on preoperative opioid use, highlighting the need for larger, prospective studies

Posted on

Predicting Intracranial Pressure Levels: A Deep Learning Approach Using Computed Tomography Brain Scans

Neurosurgery 98:256–268, 2026

This clinical study evaluates deep learning models that predict whether intracranial pressure (ICP) exceeds 15 mm Hg from brain CT scans, integrating demographic and Glasgow Coma Scale data into image inputs. Four 3D architectures—including MobileNetV2 3D and DenseNet201 3D—were trained on 578 paired CT–ICP cases with preprocessing, augmentation, and explainability via class activation maps.

Results show MobileNetV2 3D achieved the best generalization (AUC 0.883, recall 81.8%), with demographic embedding improving performance; limitations include single-center data, class imbalance, and lack of external validation, and authors recommend multicenter expansion and refined region-specific feature extraction before clinical deployment.

Intracranial Pressure (ICP) Risk: Elevated ICP is a critical, potentially fatal condition requiring rapid diagnosis and intervention, but current gold-standard invasive monitoring methods carry risks and are not always feasible in emergency settings.

Noninvasive ICP Assessment Challenge: Existing noninvasive methods (e.g., CT-based qualitative markers) lack sufficient accuracy and reliability for routine emergency use, highlighting the need for improved approaches.

Deep Learning Solution: Four deep learning models were trained on a custom dataset of 578 paired brain CT scans, demographic information, and Glasgow Coma Scale (GCS) scores to classify whether ICP exceeds 15 mm Hg, addressing the gap in noninvasive, rapid ICP estimation.

Data Integration Innovation: Demographic and GCS data were embedded and merged with CT imaging, creating a multimodal input that improved model performance compared to imaging-only approaches.

Best Model Performance: The MobileNetV2 3D model with demographic data achieved the highest test AUC of 88.3% and recall of 81.8%, outperforming other architectures and showing promise for high-sensitivity emergency applications.

Explainability: Class Activation Maps (CAMs) were used to visualize which regions of the brain CT scans influenced model predictions, enhancing transparency and interpretability of the AI system.

Limitations: The study’s main limitations include a relatively small, single-center dataset with class imbalance, lack of external/multicenter validation, and potential inconsistencies due to timing mismatches between CT and ICP measurements.

Clinical Impact & Future Directions: This AI approach could reduce reliance on invasive monitoring and accelerate ICP triage in neurocritical care; further multicenter studies, prospective validation, and expansion to multiclass classification are needed for clinical deployment.

Posted on

Determinants of survival after re-resection for recurrent glioblastoma: a meta-analysis

Acta Neurochirurgica (2026) 168:11

This systematic review and meta-analysis examines prognostic factors affecting survival after re-resection for recurrent glioblastoma, synthesizing data from 30 studies (1,741 pooled patients). Key findings identify gross total resection and MGMT promoter methylation as strong positive predictors, while age and low preoperative KPS associate with poorer outcomes; adjuvant therapies and time to re-resection showed inconsistent effects.

The paper details search methods, risk-of-bias assessment, statistical approaches, sensitivity analyses for IDH status, and study heterogeneity limitations. Conclusions emphasize patient selection for re-resection based on functional status and molecular markers and call for prospective, standardized trials and individual-patient data analyses to refine management of recurrent glioblastoma.

Gross Total Resection (GTR): Achieving GTR at re-resection for recurrent glioblastoma is significantly associated with improved survival compared to subtotal resection (pooled HR ~0.52–0.70, p < 0.001).

MGMT Promoter Methylation: Patients with methylated MGMT promoter status at recurrence have significantly better survival following re-resection (multivariate HR = 0.45, 95% CI: 0.27–0.76, p < 0.01).

Preoperative Karnofsky Performance Status (KPS): A KPS score <70 before re-resection is strongly associated with poorer survival outcomes (HR = 2.25, 95% CI: 1.59–3.19, p < 0.001).

Age: Older age is modestly associated with worse survival after re-resection, but the effect size is small (HR = 1.02, 95% CI: 1.01–1.03, p < 0.001); age alone should not preclude aggressive treatment.

Adjuvant Chemotherapy and Radiotherapy: No significant survival benefit was found for adjuvant chemotherapy (HR = 0.69, p = 0.33), radiotherapy (HR = 0.62, p = 0.50), or combined chemoradiotherapy after re-resection.

Time to Re-resection: Longer time intervals between initial surgery and re-resection did not show a statistically significant association with improved survival (HR = 0.69, p = 0.16).

Personalized Approach: Selection for re-resection should prioritize patients with good performance status, favorable tumor characteristics, and methylated MGMT promoter, with GTR as a key goal.

Evidence Limitations: Most included studies were retrospective with heterogeneity in definitions and reporting; high-quality prospective trials are needed to refine prognostic assessments and treatment strategies.

Posted on

Electrophysiological monitoring of trigeminal nerve sensory root using sensory-masseter response for microvascular decompression in trigeminal neuralgia

Acta Neurochirurgica (2026) 168:13

This study presents a novel intraoperative electrophysiological technique—the sensory-masseter response (SMR)—for localizing neurovascular compression (NVC) during microvascular decompression (MVD) for trigeminal neuralgia. In 34 patients, SMR was recorded in 82.4% and showed strong spatial correlation with intraoperatively identified NVC, with significantly lower stimulation thresholds at compression sites (median 0.3 mA).

The authors describe methodology, threshold analyses, mechanistic hypotheses linking focal demyelination and sensory-motor anastomoses to SMR, illustrative cases, safety considerations, and limitations. They conclude SMR offers high spatial specificity as a potential real-time adjunct for MVD, but recommend larger studies to validate clinical utility.

Sensory-Masseter Response (SMR) Technique: SMR is a novel intraoperative electrophysiological monitoring method that applies microcurrent stimulation to the intracranial trigeminal sensory root and records compound muscle action potentials (CMAPs) from the masseter muscle to localize neurovascular compression (NVC) during microvascular decompression (MVD) for trigeminal neuralgia (TN).

High Spatial Specificity and Correlation with NVC: SMR was successfully recorded in 82.4% of cases (28/34), with a strong spatial correlation between SMR-positive sites and intraoperatively identified NVC (p < 0.001).

Lower Stimulation Threshold at NVC Sites: The stimulation threshold required to elicit SMR was significantly lower at NVC sites (median 0.3 mA, IQR 0.2–0.4 mA) compared to distal, central, or non-compressed segments, indicating localized neural hyperexcitability likely due to demyelination.

Mechanistic Basis—Sensory-Motor Root Anastomoses: SMR is likely mediated by abnormal anastomotic pathways between the trigeminal sensory and motor roots at the NVC site, enabling direct excitation of the masseter muscle and bypassing the central reflex arc.

Clinical Utility and Limitations: SMR provides real-time functional feedback for intraoperative localization of NVC, but its clinical value and influence on surgical outcomes require further validation in larger, controlled studies.

Potential Pathogenic Insight: The ability to elicit SMR at NVC sites supports the hypothesis that focal demyelination and abnormal sensory-motor conduction contribute to TN pathogenesis and characteristic symptoms like chewing-triggered pain.

Safety Considerations: SMR stimulation can cause transient blood pressure elevation due to sympathetic activation; precautions include deepening anesthesia and monitoring hemodynamics during stimulation.

Study Limitations: The main limitations include small sample size, absence of SMR testing in non-TN or healthy controls, possible technical constraints in nerve exposure, and restriction of recordings to the masseter muscle only.

Posted on

Does segmental alignment matter? A novel understanding of segmental compensation and reciprocal change following single-level lumbar reconstruction

J Neurosurg Spine 44:72–79, 2026

This clinical study analyzes segmental compensation and reciprocal changes after single-level L5–S1 anterior lumbar interbody fusion in 100 adults with normal preoperative lumbar lordosis. Using intradiscal angle, motion segment angle, and disc heights, the authors show significant index-level lordosis restoration and correlated increases in overall lumbar lordosis at 1 month and 1 year.

The paper documents consistent reciprocal decreases in lordosis at adjacent L4–5 and L3–4 levels, provides predictive linear equations for adjacent-level change based on index-level IDA change, and argues that restoring segmental lordosis may prevent maladaptive compensation and reduce adjacent segment disease risk, while noting limitations and need for longer follow-up.

Segmental Compensation: Adjacent lumbar spine segments compensate for loss of lordosis at a pathologic segment by increasing their own segmental lordosis to maintain overall lumbar lordosis (LL) within the normal range.

Reciprocal Change After Fusion: Restoration of lordosis at the L5–S1 level via anterior lumbar interbody fusion (ALIF) leads to a significant, measurable decrease in segmental lordosis at adjacent levels (L4–5 and L3–4), demonstrating a reciprocal relationship.

Predictive Model: The degree of reciprocal loss of lordosis at adjacent levels can be predicted using linear equations based on the lordosis gain at the fused segment (e.g., 1-year adjacent level decrease in IDA = −0.195 × [index level 1-year IDA change] + 0.332).

Global Alignment Maintenance: Despite significant increases in lordosis at the surgical level, total lumbar lordosis does not increase by the same amount due to compensatory decreases at adjacent levels, maintaining global alignment.

Clinical Implication: Proper restoration of segmental lordosis at the surgical level may help prevent negative consequences of persistent compensation at adjacent segments, potentially reducing the risk of adjacent segment disease.

Study Population: Findings are based on 100 adults with normal preoperative LL (PI–LL < 10°) who underwent single-level L5–S1 ALIF and achieved ≥5° increase in segmental lordosis postoperatively.

Correlation With Outcomes: Changes in lordosis at the surgical segment positively correlate with changes in overall lumbar lordosis at both 1 month and 1 year postoperatively.

Long-Term Impact: The persistence of reciprocal changes at adjacent segments up to 1 year suggests that segmental compensatory mechanisms are durable and may influence long-term spinal health and surgical planning.

Posted on

Minimally invasive lumbar decompression versus open decompression for lumbar spinal stenosis: a propensity score–matched analysis

J Neurosurg Spine 44:55–61, 2026

This propensity score–matched retrospective study compares minimally invasive lumbar decompression (mild) with open decompression for lumbar spinal stenosis at a tertiary multisite center from 2005–2024. Primary outcomes included pain change (NRS), reoperation, and perioperative complications, with MCID defined as 30% NRS improvement.

Results show open decompression yielded greater pain improvement, higher MCID attainment, and lower overall reoperation rates, while mild had fewer durotomies but more neurological deficits. The authors recommend independent prospective studies to validate comparative efficacy and cost-effectiveness and note limitations including retrospective design and incomplete radiographic severity data.

Open decompression is more effective than the mild (minimally invasive lumbar decompression) procedure for achieving clinically significant pain improvement in patients with symptomatic lumbar spinal stenosis (43.1% vs 22.2% reached MCID; p < 0.001).

Reoperation rates are higher after the mild procedure compared to open decompression (46.2% vs 29.3%; p = 0.008), indicating less durable symptom relief with mild.

Pain outcomes favor open decompression, with patients reporting lower pain scores at last follow-up (mean NRS 2.3 vs 5.4; p < 0.001) and greater overall improvement from baseline.

Complication profiles are similar overall, but mild is associated with higher rates of postoperative neurological deficits (6.3% vs 0.6%; p = 0.003), while open decompression has a higher rate of durotomy (2.9% vs 0%; p = 0.024).

Procedure characteristics: The mild procedure is performed percutaneously through a small incision, typically by pain medicine physicians, and targets debulking the ligamentum flavum without muscle dissection.

Study limitations include retrospective design, incomplete matching for stenosis severity and baseline pain, and missing data on some outcomes, suggesting a need for prospective studies.

Existing literature on mild is often industry-sponsored and lacks direct comparison to open decompression; independent studies show mild may be less effective and more likely to require further surgery.

Clinical recommendation: Open decompression remains the gold standard for patients with symptomatic lumbar spinal stenosis who fail conservative management, while the mild procedure may offer less pain relief and higher reoperation risk.

Posted on

Comparative Effectiveness of Conservative Management Versus Surgical Fixation in Acute Type II Odontoid Fractures

Neurosurgery 98:61–72, 2026

This retrospective single-center study compares radiological and functional outcomes of conservative versus surgical management for acute type II odontoid fractures in an elderly cohort treated from 2015–2023. Results show surgical fixation markedly increases radiographic union rates, while conservative bracing predominated and delivered superior early pain control with similar long-term functional independence.

Multivariable modeling identified surgery as the strongest positive predictor of union, whereas age ≥80, smoking, osteoporosis, higher frailty, and angulation ≥11° reduced union odds and lowered likelihood of surgical selection. Complication rates and hospital stays were higher after surgery, supporting individualized treatment decisions balancing union probability, pain, frailty, and perioperative risk.

Surgical fixation increases the odds of fracture union in acute type II odontoid fractures (adjusted OR = 6.6), but does not improve long-term functional independence or survival compared to conservative management.

Conservative management (mainly rigid collars) is preferred for elderly or frail patients and achieves similar functional outcomes (modified Rankin Scale ≤2: 75% vs 73%) and better early pain control (75% vs 47%) than surgery, despite lower union rates.

Fracture union rates are significantly higher after surgery (44% vs 10% with conservative treatment, P < .01), but most conservatively managed patients achieve stable nonunion without adverse functional impact.

Advanced age (≥80 years), frailty, smoking, osteoporosis, and fracture angulation ≥11° are all strong negative predictors of fracture union, regardless of treatment approach.

Surgical patients are typically younger, less frail, and have greater fracture displacement or posterior displacement, reflecting selection bias in surgical decision-making.

Complication rates are higher with surgery (40% vs 2% for conservative), and surgical patients have longer hospital stays (median 11 vs 3 days), but most complications do not require reoperation.

Mortality rates at five years are similar between conservative and surgical groups (52% vs 53%), indicating no survival advantage with operative intervention.

Conservative treatment is a safe, effective option for most elderly or frail patients with type II odontoid fractures, reserving surgery for younger, medically fit individuals or those with unstable fracture patterns

Posted on

Long-Term Outcomes of Surgical Clipping of Woven EndoBridge-Eligible Middle Cerebral Artery Bifurcation Aneurysms

Operative Neurosurgery 30:18–25, 2026

This clinical study evaluates long-term outcomes of microsurgical clipping for middle cerebral artery (MCA) wide-neck bifurcation aneurysms that meet radiographic criteria for treatment with the Woven EndoBridge (WEB) device. Using a retrospective review of a prospectively maintained registry, the authors compare functional outcomes, complication rates, recurrence, and retreatment against published WEB and clipping cohorts.

Results show high surgical efficacy: 99% treatment success, 94.1% of unruptured cases achieving final mRS ≤2, a 4.9% morbidity rate, and 0% mortality, with lower retreatment and higher complete occlusion rates than reported WEB series. The authors conclude microsurgical clipping remains a reliable option for MCA WNBAs and recommend location-specific subgroup analyses in future WEB studies.

Microsurgical Clipping Outcomes: Surgical clipping of WEB-eligible middle cerebral artery (MCA) wide-necked bifurcation aneurysms (WNBAs) achieved a 99% treatment success rate, 4.9% morbidity, 0% mortality, and 94.1% good functional outcomes (mRS ≤2) in unruptured cases, with low recurrence and retreatment rates.

Comparison to WEB Device: Clipping resulted in higher complete occlusion rates, lower retreatment rates, and lower thromboembolic complication rates compared to endovascular treatment with the Woven EndoBridge (WEB) device for MCA WNBAs.

Functional Outcomes: Both ruptured and unruptured aneurysms treated surgically showed excellent or improved functional status at follow-up, with 91.8% of unruptured and 94.1% of ruptured cases experiencing stable or improved mRS scores.

Complication Profile: Thromboembolic complications occurred in 5.9% of unruptured cases, with permanent neurological deficits in 4.9% and no surgical mortality; intraoperative rupture was rare (2.0%).

Limitations of WEB Evidence: Most WEB studies pooled multiple aneurysm locations, making it difficult to assess location-specific efficacy; at the MCA, WEB showed higher retreatment and thromboembolic rates than clipping, and lower rates of complete occlusion.

Treatment Selection: MCA aneurysms are often suitable for surgical clipping due to their accessible location and favorable anatomy, especially at experienced centers, supporting a “clip-first” approach for many MCA WNBAs.

Study Limitations: Findings are based on a retrospective, single-center cohort at a high-volume academic institution, limiting generalizability and direct comparison to randomized or multicenter WEB trials.

Clinical Recommendation: Both clipping and WEB are effective for MCA WNBAs, but surgical clipping may offer superior durability and occlusion; treatment choice should be individualized through shared decision-making, not solely based on minimally invasive appeal.

Posted on

Artificial intelligence–based deep learning model for evaluating procedural consistency in microvascular anastomosis

J Neurosurg 144:1–10, 2026

This study presents an LSTM-based deep learning model that objectively evaluates microvascular anastomosis performance by predicting hand-motion trajectories from MediaPipe-derived hand landmarks. It quantifies consistency using Kullback-Leibler divergence and validates complementary metrics—economy and flow of motion—comparing two expert neurosurgeons (repeat sessions) and one trainee in simulated end-to-side anastomoses.

Results show low KL divergence for experts versus higher divergence for the trainee, reflecting greater consistency and efficiency. The authors discuss methodology, model architecture choices, limitations in generalizability, and potential integration into microsurgical training workflows for objective skill assessment.

Deep Learning Model: An LSTM-based neural network was developed to objectively assess consistency and precision in microvascular anastomosis by predicting and comparing suturing hand movements using video-based hand landmark tracking, eliminating the need for physical sensors.

Hand Tracking Technology: The model utilized MediaPipe Hand Landmarker, a CNN-based system that detects 21 hand landmarks from standard video, enabling detailed, sensor-free motion analysis during microsurgical simulation.

Performance Metrics: Three primary metrics were used: Kullback-Leibler (KL) divergence for consistency, economy of motion (mean Euclidean distance of hand movement), and flow of motion (median time per suture), providing quantitative, objective evaluation of surgical skill.

Experimental Setup: Two expert neurosurgeons performed microanastomosis simulations (interrupted and continuous suturing) in two sessions one year apart, and a trainee performed the same task for comparison; all sessions were recorded and analyzed using the AI pipeline.

Results and Interpretation: Experts showed low KL divergence (high consistency) and efficient, rhythmic motion, while the trainee had higher KL divergence, longer suture intervals, and more variable motion, reflecting less developed skill.

Model Application: The approach enables rapid, automated assessment of multiple trainees using standard video equipment, supporting objective tracking of skill progression and facilitating feedback in training environments.

Model Rationale: LSTM architecture was chosen for its ability to model long-term temporal dependencies in sequential hand movement data, making it suitable for predicting surgical motion patterns over extended timeframes.

Limitations and Future Directions: Current findings are based on a small sample of experts and one trainee in a simulated environment; broader validation, metric standardization (especially for KL divergence), and extension to real operative settings are needed for generalizability.

Posted on

Spatial patterns of fat within the deep multifidus as a biomarker for chronic low back pain

The Spine Journal 26 (2026) 106−118

This clinical study maps spatial fat infiltration (FI) within the lumbar multifidus (MF) of 230 chronic low back pain (cLBP) patients using 3T IDEAL MRI and statistical parametric mapping to identify level- and region-specific FI patterns. The authors introduce a novel “fat-map” method and define a deep15 FI% (deepest 15% of MF) to compare regional FI with overall whole‑muscle FI across L1L2–L5S1.

Key results show elevated FI in the deep MF at L4L5 and L5S1 is associated with higher pain (PEG scores) and adjacent disc degeneration independent of age, sex, and BMI, while overall FI% is more strongly linked to demographic factors. The deep15 FI% emerges as a potential biomarker for cLBP, suggesting regional MF degeneration may better explain pain mechanisms than whole‑muscle summary measures.

Deep multifidus fat infiltration (FI): Elevated fat content in the deepest regions (deepest 15%) of the multifidus muscle at the lower lumbar spine (L4L5, L5S1) is more strongly associated with chronic low back pain (cLBP) symptoms and adjacent disc degeneration than overall muscle FI%.

Regional specificity: FI in the deep multifidus at lower lumbar levels is less influenced by age, sex, and BMI compared to overall FI%, making it a more specific biomarker for cLBP.

Novel biomarker (deep15 FI%): The “deep15 FI%” (mean fat fraction in the deepest 15% of the multifidus at L4L5/L5S1) is a regionally specific muscle quality measure, showing a stronger association with pain and disc degeneration than traditional whole-muscle FI% measures.

Demographic factors: Older age and female sex are associated with increased FI throughout the multifidus, while higher BMI mainly affects FI in the superficial 60% of the muscle, not the deep region at lower lumbar levels.

Pain association: Higher deep15 FI% at the lower lumbar levels is significantly associated with higher pain and pain interference scores (PEG survey), while overall FI% is not.

Disc degeneration link: Adjacent disc degeneration is the only degenerative spine feature consistently associated with increased deep multifidus FI at lower lumbar levels; other features (e.g., Modic changes, facet osteoarthritis) do not show this association when controlling for age, sex, and BMI.

Clinical implications: Deep multifidus FI could serve as a more precise imaging biomarker for cLBP and may guide future targeted interventions, though further longitudinal and interventional studies are needed to confirm causality and reversibility.

Limitations: The study’s spatial FI analysis is limited to radial (deep-to-superficial) direction and cross-sectional design; further research is needed to clarify causal mechanisms and intervention effects.

Posted on

Segmental Lordosis After Open Transforaminal Lumbar Interbody Fusion Using Expandable Oblique Versus Static Anterior Banana Cages

Operative Neurosurgery 30:78–89, 2026

This clinical research article compares segmental and lumbar lordosis outcomes after open transforaminal lumbar interbody fusion (TLIF) with posterior column osteotomy using either static anterior “banana” cages or obliquely placed expandable cages. In a single-surgeon retrospective cohort of 210 patients (327 segments), expandable cages produced a significantly greater median change in segmental lordosis (ΔSL) by 2.0° at six months, persisting after multivariate adjustment and propensity matching.

The study also reports inconsistent effects of cage type on overall lumbar lordosis (ΔLL), with subgroup and matched analyses yielding differing results, and found no difference in subsidence or complication rates. Findings emphasize the influence of preoperative segmental lordosis, segment level (notably L5–S1), and construct length on achieved correction, while noting limitations including retrospective design and lack of clinical outcomes.

Expandable Cages: Expandable obliquely placed cages in open transforaminal lumbar interbody fusion (TLIF) with posterior column osteotomy (PCO) produce a significantly greater median increase in segmental lordosis (ΔSL) of 2.0° compared to static anteriorly placed banana cages, representing a 50% increase.

Statistical Robustness: The greater segmental lordosis achieved with expandable cages remained significant after multivariate regression analysis and propensity score matching, confirming the reliability of the finding.

Overall Lumbar Lordosis (ΔLL): No clear advantage was found between cage types regarding the change in overall lumbar lordosis, with results varying depending on the statistical method used.

Surgical Technique Consistency: All surgeries were performed open with a full PCO using a consistent technique, minimizing confounding variables related to surgical approach or technique.

Preoperative Segmental Lordosis Impact: Segments with lower preoperative segmental lordosis (<15°) experienced the greatest increase in lordosis postoperatively, regardless of cage type.

L5-S1 Segment Benefit: Expandable cages were especially favorable at the L5-S1 segment, likely due to anatomical constraints that make insertion of large static cages more challenging at this level.

Complication and Subsidence Rates: No significant differences were observed between cage types in rates of complications, cage subsidence, or spondylolisthesis correction.

Clinical Outcomes Unclear: The study did not assess clinical outcomes, so the impact of the observed radiographic differences on patient-reported outcomes remains unknown.

Posted on

Fully automated image updating for brain shift compensation after dural opening

J Neurosurg 144:206–216, 2026

This study presents a fully automated intraoperative image-updating system that compensates for brain shift after dural opening by assimilating intraoperative stereovision (iSV) data into a biomechanical finite element model to deform preoperative MR (pMR) images into updated MR (uMR) images. The pipeline integrates FastSAM segmentation, vessel/sulcus feature registration, and FEM-based whole-brain deformation to produce clinically usable uMRs.

In fifteen open cranial cases, automated updates reduced mean target registration error from 6.2 mm (pMR) to 1.9 mm (uMR) and completed without user intervention in 3.9 ± 0.6 minutes. Results demonstrate robust segmentation (DSC 0.93), submillimeter iSV reconstruction accuracy, and potential for broader adoption, while noting limitations in small or featureless openings and need for deeper-structure validation.

Fully automated image updating: Developed and validated a fully automated system to update preoperative MRI images for brain shift compensation after dural opening in open cranial surgery, eliminating user intervention and expertise requirements.

Intraoperative stereovision (iSV) integration: Utilized iSV images to capture high-resolution surface deformation data, which was processed by deep learning-based segmentation (FastSAM) and registered with preoperative MRI using vessel and sulcus features.

Two-step registration process: Employed translation-only cross-correlation for global alignment and Demons deformable registration for local nonrigid deformation between iSV and preoperative MRI surfaces.

Biomechanical modeling: Assimilated extracted nonrigid cortical displacements into a finite element model to estimate whole-brain deformation and generate updated MR images (uMR).

Accuracy improvement: Achieved significant reduction in target registration error (TRE) from 6.2 ± 1.2 mm (pMR) to 1.9 ± 1.0 mm (uMR), with overall mean computational time of 3.9 ± 0.6 minutes and no user intervention.

Robustness and efficiency: System was robust across a range of surgical conditions (lesion type, craniotomy size, brain shift magnitude), and performance was not significantly affected by these variables.

Limitations: Current system requires clear iSV images (free of instruments/blood), is limited to initial post-dural opening updates, and surface accuracy was primarily evaluated; further development is needed for autonomous updates during resection and deeper structure validation.

Potential for broad adoption: Elimination of user dependency and minimal workflow interruption suggest strong potential for integration into routine open cranial 

Posted on

Clinical Outcomes of Decompressive Spine Surgery for Painless Cervical Myelopathy

Neurosurgery 98:161–173, 2026

This multicenter retrospective study analyzes 407 patients from the Michigan Spine Surgery Improvement Collaborative who underwent decompressive surgery for cervical spondylotic myelopathy presenting without neck or arm pain. Patient-reported outcomes (mJOA, PROMIS PF, EQ-5D) and clinical endpoints were assessed at baseline, 90 days, 1 year, and 2 years to quantify functional and quality-of-life changes after surgery.

Results show modest but clinically meaningful improvements in function and quality of life for a subset of patients, with peak MCID rates at one year and sustained patient satisfaction above 80% at two years. Severe preoperative myelopathy predicted worse immediate disposition and higher readmission, although some severe cases still achieved early functional gains.

Painless Cervical Myelopathy (CSM) Surgery: Decompressive spine surgery in patients with CSM but without neck or arm pain led to clinically significant improvements in myelopathic symptoms and physical function for a modest proportion of patients.

Quality of Life Gains: Surgery resulted in increased quality of life, with mean EQ-5D scores higher at all postoperative time points compared to baseline, and over 80% of patients reported sustained satisfaction up to two years after surgery.

Severity-Dependent Outcomes: Patients with severe myelopathy had worse immediate surgical outcomes (lower rates of discharge to home, higher 90-day readmission) and lower quality of life at two years compared to those with mild myelopathy.

Incidence of Postoperative Pain: Despite presenting without pain, 20% of patients developed persistent postoperative neck pain and 14% developed arm pain at two years, rates comparable to those seen in typical CSM surgery cohorts.

Functional Improvement Metrics: At one year, 49% achieved clinically meaningful improvement in physical function (PROMIS PF), 36% in mJOA, and 42% in quality of life (EQ-5D); these improvements were most pronounced at one year and declined by two years for some metrics.

Surgical Approach Differences: Anterior approaches were associated with higher postoperative satisfaction and lower rates of complications like dysphagia compared to posterior or combined approaches.

Study Limitations: Limitations include lack of radiological data, possible misclassification of myelopathy severity, and significant loss to follow-up, affecting long-term outcome interpretation.

Clinical Implication: Surgery may halt progression and provide functional and quality of life benefits even in painless CSM, but patient selection remains challenging and further objective, randomized studies are needed.

Posted on

Lumbar Multifidus Intramuscular Fat Concentrations are Associated With Recovery Following Decompressive Surgery for Lumbar Spinal Stenosis

Spine 2026;51:25–33

This longitudinal cohort study investigates whether preoperative intramuscular fat (IMF) in lumbar paraspinal muscles predicts five-year recovery and surgical success after decompressive surgery for lumbar spinal stenosis causing intermittent neurogenic claudication. Using automated MRI segmentation and quantitative IMF thresholds, outcomes included global perceived effect, Zurich Claudication Questionnaire-based surgical success, pain VAS, and disability scores.

Key findings show lower preoperative multifidus IMF (but not erector spinae IMF) was associated with higher rates of perceived recovery, surgical success, and reduced long-term disability over five years. No consistent relationship was found between IMF and leg or back pain trajectories; results were adjusted for age, BMI, sex, smoking, reoperation, and surgical technique.

Lumbar multifidus intramuscular fat (IMF): Lower preoperative IMF in the lumbar multifidus muscle is associated with higher rates of perceived recovery and surgical success after decompression surgery for lumbar spinal stenosis (LSS) over a five-year period.

Erector spinae IMF: Preoperative IMF levels in the erector spinae muscle are not significantly associated with recovery or surgical success following LSS surgery.

Disability outcomes: Patients with nonsevere IMF in the right lumbar multifidus experience less postoperative disability for up to five years compared to those with severe IMF.

Pain outcomes: No significant association exists between preoperative paraspinal IMF (either muscle) and the clinical course of leg or back pain intensity after surgery.

IMF quantification method: Automated MRI-based quantification and categorization of IMF (<50% = nonsevere, ≥50% = severe) using computer vision models enables objective assessment of paraspinal muscle health.

Prognostic value: Assessing lumbar multifidus IMF preoperatively can improve prediction of which patients are at risk for poor recovery and help tailor individual interventions.

Study limitations: Missing data (up to 32% at five years), dichotomized IMF classification, and limited field-of-view for some muscles may affect precision; more detailed and larger studies are needed.

Clinical implication: Routine IMF assessment may enhance clinical decision-making and rehabilitation strategies for patients undergoing lumbar decompressive surgery.

Posted on

Skull Base Anatomy Presented in 360° Photogrammetry 3-Dimensional Models

Operative Neurosurgery 30:124–136, 2026

This article presents the creation of seven photorealistic 360° photogrammetric 3D models of the central skull base derived from stepwise dissections of a formalin-fixed, vessel-injected cadaveric head. The workflow—dissection stages, smartphone-based image capture, cloud photogrammetry, refinement in Blender, and VR/MR upload—enables immersive visualization of cranial nerves, ICA/vertebral segments, and regional anatomy.

The models offer progressive exocranial-to-endocranial perspectives for education and preoperative planning, highlighting cavernous sinus, infratemporal and pterygopalatine fossae, petrous bone, and foraminal relationships. Limitations include a single-specimen dataset, color variations from fixation, and resolution constraints in deep cavities; nevertheless, the freely accessible VR models complement traditional dissection and anatomical atlases.

360° Photogrammetric 3D Models: Realistic, photogrammetry-based 3D models of the central skull base were created from cadaveric dissections, offering immersive 360° visualization of complex neuroanatomical structures for enhanced spatial understanding.

Stepwise Dissection and Scanning: Seven progressive anatomical models were generated by systematically dissecting and scanning a formalin-fixed, vessel-injected head specimen, documenting both exocranial and endocranial perspectives.

Key Structures Visualized: The models detail the courses of cranial nerves, major vessels (including all internal carotid artery segments), skull base foramina, infratemporal and pterygopalatine fossae, paranasal sinuses, and deep neck spaces.

Technical Workflow: High-resolution images were captured using a smartphone multi-camera system, processed via cloud-based photogrammetry, refined in 3D software, and made accessible through web, VR, and MR platforms.

Educational Value: The interactive models allow customizable, layered exploration of anatomy, overcoming limitations of traditional 2D images and static atlases, and are freely accessible for educational and preoperative planning purposes.

Limitations: The study used a single specimen, which may not represent anatomical variants; image quality in deep/narrow regions could be further improved with advanced imaging and fixation techniques.

Broad Accessibility: Smartphone-based and cloud photogrammetry methods make high-resolution anatomical modeling more accessible and less resource-intensive, facilitating widespread dissemination.

Conclusions: 360° photorealistic 3D models significantly enhance comprehension of skull base anatomy and are a valuable adjunct to traditional teaching, with potential to improve neurosurgical training and patient outcomes.

Posted on

A pilot randomized control trial comparing posterior paramedian versus midline incisions for interbody fusions of the lumbar spine

The Spine Journal 26 (2026) 85−93

This single-center pilot randomized controlled trial compares posterior midline versus posterior paramedian (Wiltse/minimally invasive) approaches for 1–2 level lumbar interbody fusions, assessing feasibility, infection rates, and reoperation. One hundred-one patients were randomized, with reported differences: higher deep infection (9.6% vs 4.1%) and greater reoperation rates (25% vs 6.1%) in the midline group, alongside improvements in PROMs for both groups.

Methods, perioperative care, and outcome measures are detailed, including CDC-defined deep infection, PROMs (ODI, SF-12, EQ-5D, VAS), and feasibility metrics for a multicenter RCT. The authors conclude the pilot supports a larger trial, recommend an interim analysis for reoperation differences, and outline limitations related to single-surgeon data, recruitment disruptions, and clustering effects.

Pilot RCT Findings: A single-center pilot randomized controlled trial compared posterior paramedian and midline incisions for lumbar interbody fusion, finding potential differences in deep infection and reoperation rates, with both groups showing improvement in patient-reported outcomes.

Infection Rates: Deep infection occurred in 9.6% of midline patients versus 4.1% in the paramedian group within 3 months post-surgery.

Reoperation Rates: 25% of midline patients required revision surgery compared to 6.1% in the paramedian group during follow-up.

Patient Outcomes: Both surgical approaches led to improvements in disability, pain, and quality of life scores at 6–18 weeks and 1 year postoperatively, with slightly greater improvements noted in the paramedian group.

Intraoperative Metrics: The paramedian group had lower average blood loss (268ml vs. 313ml) and a shorter average hospital stay (3.3 vs. 4.2 days) compared to the midline group.

Feasibility: The trial demonstrated high protocol adherence, low loss to follow-up, and good participant acceptance, supporting the feasibility of a larger multicenter RCT.

Limitations: The pilot was limited by single-center design, small sample size, protocol violations, and short follow-up; results may not be generalizable until confirmed by a larger trial.

Next Steps: The pilot data will inform sample size calculations and study design for a multicentered RCT to clarify whether paramedian approaches offer significant clinical advantages over midline incisions.

Posted on

Long-Term Mortality of Patients With Head Injuries—A 10-Year Follow-up Study With Population Controls Study Performed at Tampere University Hospital

Neurosurgery 98:105–114, 2026

This study reports a 10-year follow-up comparing survival and causes of death between 1,930 patients treated for head injuries at a Finnish university hospital and 9,605 matched population controls. After excluding deaths within the first year, patients had a hazard ratio of 1.84 for mortality, with excess deaths concentrated in the first five years and elevated unintentional/traumatic and alcohol-related causes.

Multivariable analysis found age, male sex, pre-existing conditions, chronic alcohol use, and substance abuse—rather than injury severity or CT-positivity—were independently associated with reduced survival, though greater TBI severity remained linked to additional mortality risk when comparing patient severity subgroups to their matched controls.

Long-term Mortality: Patients with head injuries have significantly reduced long-term survival compared to matched population controls, with a hazard ratio (HR) for death of 1.84 after excluding those who died in the first year post-injury.

Acute vs. Long-term Risk: Death rates are notably higher among patients with head injuries for up to 5 years after injury, then approach control levels thereafter.

Causes of Death: Unintentional and traumatic causes (9.6% vs 4.4%) and alcohol-related causes (8.4% vs 1.9%) are significantly more common among head injury patients than controls.

Patient Characteristics: Age, male sex, pre-existing conditions, chronic alcohol use, and regular substance abuse are independently associated with decreased survival, whereas injury severity and CT findings are not significant predictors in multivariate analysis.

TBI Severity: Even patients with no documented traumatic brain injury (TBI) have reduced survival compared to controls; increasing TBI severity is associated with additional mortality risk.

Lifestyle and Pre-existing Factors: Much of the reduced survival is linked to patient characteristics and lifestyle factors (such as substance abuse), not solely to injury-related factors.

Immediate Causes of Death: Aspiration pneumonia and epileptic causes are significantly overrepresented as immediate causes of death in head injury patients who survive more than one year.

Control Group Limitations: Use of population controls and lack of detailed control data may overestimate associations between head injury and mortality due to unmeasured confounders.

Posted on

Functional status in long-term survivors after mapping-guided surgery for diffuse low-grade glioma

J Neurosurg 144:139–150, 2026

This clinical study reports long-term functional outcomes in 103 consecutive patients with diffuse low-grade glioma who underwent mapping-guided resections and were followed for at least 15 years. Key findings include high overall survival (83.5%), mean postoperative KPS of 94.8, and 90% return-to-work rate, with low permanent neurological morbidity across 205 resections.

Comparative analysis shows patients who sustained employment had smaller pre/postoperative tumor volumes, greater extent of resection (including more supratotal resections), and less exposure to radiotherapy. The data support early maximal safe resection and postponement of radiotherapy to preserve long-term functional status and professional activity.

Functional Preservation: Long-term survivors of diffuse low-grade glioma (LGG) surgery had high rates of preserved functional status, with 90.7% of surviving patients maintaining a Karnofsky Performance Scale (KPS) score ≥ 80 after an average of 18.2 years follow-up.

Return to Work: 90% of patients were able to return to work after mapping-guided resection, and maintaining professional activity was strongly associated with higher preoperative KPS and greater extent of resection (EOR).

Extent of Resection (EOR): Greater EOR, particularly supratotal or total resections, correlated with better long-term functional outcomes and higher rates of continued employment.

Radiation Therapy Impact: Early or any radiotherapy (RT) was linked to lower rates of return to work and a reduced proportion of patients with KPS ≥ 80 at last follow-up, while chemotherapy did not show this negative association.

Timing of Surgery: Early surgery at diagnosis, especially in patients with higher KPS and smaller tumor volume, increased the chance for maximal resection and long-term preservation of functional status.

Low Neurological Morbidity: Permanent postoperative neurological deficits were rare (1.5% after 205 resections), supporting the safety of maximal resection with intraoperative mapping.

Malignant Transformation: The risk of malignant transformation was lower in patients with greater EOR and those who continued to work, suggesting oncological benefit from radical resection.

Adjuvant Therapy Strategy: A wait-and-watch strategy after maximal safe resection, postponing adjuvant treatments unless necessary, helped preserve long-term quality of life and autonomy.

Proudly powered by WordPress