Current neurocritical care relies on fixed, population-based intracranial pressure (ICP) thresholds, which fail to account for inter-individual variability or the dynamic nature of cerebral physiology. Individualized ICP (iICP) thresholds have emerged as a potential personalized alternative but have largely relied on complete recording periods, limiting their derivation to post-hoc calculations and preventing continuous real-time estimation. Recently, a novel algorithm capable of continuously deriving iICP was developed; however, this algorithm has not yet been validated beyond a single-center cohort, and its prognostic utility remains uncertain. Therefore, the goals of this study were to externally validate the performance of this continuous iICP algorithm and to evaluate the prognostic utility of continuously derived iICP compared with current guideline-based ICP thresholds. A retrospective multicenter study was conducted using high-resolution physiologic datasets from the CAHR-TBI collaborative. iICP was derived using the recently developed continuous iICP algorithm. Algorithm performance was externally validated by assessing derivation yields and associated quality metrics and comparing with those reported in the original study that introduced the algorithm. Outcome associations were evaluated using mean hourly dose of ICP above iICP and guideline-based thresholds. Logistic regression models were used to assess and compare prognostic performance, with discrimination quantified using optimism-corrected area under the receiver operating characteristic curve and added predictive value assessed using Nagelkerke's R2. Finally, intensity-duration heatmaps were generated to characterize the relationship between iICP-based insult burden and clinical outcomes. The continuous iICP algorithm demonstrated consistent performance across parameter settings compared with the original study, supporting its external validity. Select hybrid iICP configurations, defaulting to 20 mmHg or 22 mmHg during periods where iICP could not be identified, outperformed guideline-based ICP thresholds in prognostic models. Heatmap analyses demonstrated a graded relationship between iICP-defined insult burden and outcome, with greater burden of ICP above iICP associated with poorer outcomes. Overall, iICP represents a promising adjunct to conventional guideline-based ICP management in neurocritical care. The findings of this study suggest potential clinical benefit from iICP-supplemented ICP management; however, prospective randomized controlled trials will be required to determine whether such an approach translates into meaningful improvements in patient outcomes.

Glioblastoma is the most aggressive primary tumor of the central nervous system, with particularly poor prognosis in elderly patients. Individuals aged ≥ 75 years remain underrepresented in clinical trials, resulting in limited evidence to guide treatment decisions. This study aimed to analyze treatment patterns and survival outcomes in patients aged 75 years or older with newly diagnosed IDH-wildtype glioblastoma. We conducted a single-center retrospective study of 108 consecutive patients aged ≥ 75 years who underwent surgical intervention between 2016 and 2022. Patients were stratified by age group (75–79 vs. ≥80 years), surgical modality (biopsy vs. resection), extent of resection (maximal vs. submaximal), MGMT promoter methylation status, and adjuvant therapy. Overall survival (OS) and progression-free survival (PFS) were assessed using Kaplan–Meier analysis and Cox proportional hazards models. The median age was 79 years, and 46.3

Abstract Obsessive-compulsive disorder (OCD) is a chronic and debilitating condition, often resistant to conventional treatments. Deep brain stimulation (DBS) emerges as a promising intervention, but its efficacy varies among patients, underscoring the need to understand the predictive factors influencing its outcomes. To evaluate and compare disease and patient characteristics in their ability to predict response to DBS in OCD patients. All major databases were searched for original studies. This study differentiated responders, partial responders, and non-responders based on relative post-treatment Y-BOCS score changes and pre-defined Y-BOCS cut-off scores. Clinical predictors were assessed using qualitative synthesis, univariate analysis, stepwise and regularisation-tuned multivariate linear and logistic regression analyses. The meta-analysis, comprising 28 studies with a pooled sample of 296 patients, found that higher baseline Y-BOCS scores significantly predicted favourable long-term DBS response in the multivariate regression analysis (p = 0.0075), whereas antipsychotic use was identified as a significant predictor of non-response (p = 0.0138). In the univariate analysis, antidepressant use was negatively associated with DBS response (p = 0.027), while anxiolytic use was positively associated with short-term improvement (p = 0.025). Symmetry, hoarding, and perfectionism symptomology at baseline predicted reduced short-term improvement (p = 0.027). Adjusting for stimulation target, aggression, and intrusive thoughts, baseline symptomology emerged as a positive predictor of DBS response in the multivariate model (p = 0.0475). In the multivariate analysis, excluding studies with high risk of bias, the symptom combination anxiety, avoidance, and fear was associated with DBS response (p = 0.024). Disease duration, gender, anatomical target location, and age did not predict DBS response (p > 0.05). This study found that a higher baseline severity of OCD symptoms significantly predicted a greater likelihood of long-term response to DBS. In contrast, the use of antipsychotics and antidepressants was associated with poorer outcomes, while anxiolytic use appeared to support short-term improvement. Compulsive symptom profiles characterised by symmetry, hoarding, and perfectionism were associated with non-response, whereas the presence of aggression and intrusive thoughts, as well as anxiety, fear and avoidance symptomology, was linked to a positive response. Further research with homogenous methodology and outcome reporting, as well as randomised control trials, are required to further elucidate these phenotypes and allow for more personalised and, thus, likely more effective DBS treatment strategies for all OCD patients.

Surgical site infections (SSIs) are a serious complication of spinal surgery, leading to prolonged hospital stays, increased healthcare costs, and worse patient outcomes. This review aimed to synthesise current evidence on modifiable risk factors for SSIs in spinal surgery to identify targets for intervention, perioperative optimisation, and improve patient care. A systematic search of MEDLINE, Embase, Scopus, Academic Search Ultimate, and CINAHL was conducted on 10 March 2024. Risk of bias was assessed using the Newcastle-Ottawa Scale, ROBINS-E, and the Oxford Centre for Evidence-Based Medicine: Levels of Evidence. A random-effects meta-analysis generated pooled odds ratios (ORs) with 95

Normal pressure hydrocephalus (NPH) is a cause of neurological impairment, with programmable shunt valves serving as the standard treatment. However, magnetic interference can result in altered pressure settings, affecting valve functionality. This systematic review investigates the effects of magnetic resonance imaging (MRI) on programmable shunt valves in NPH patients. Conducted in adherence to PRISMA guidelines, an extensive literature search across PubMed and Embase databases from January 2000 to January 2025, was performed. Studies evaluating the impact of MRI at 1.5T and 3T field strengths on valve settings were included. Valve reprogrammability and outcomes were analysed, with statistical comparisons across MRI strengths and valve types using Mann-Whitney U and ANOVA tests. The pooled analysis included 475 NPH patients across 11 studies. Valve setting alteration was required in 44.2

Purpose In comatose patients at risk of death by neurologic criteria (DNC), spinal-mediated movements (SMM) and movements of unclear neuroanatomic origin (MUO) are occasionally challenging to discriminate from cerebral-mediated movements. Our objectives were to assess the respective prevalence and semiology of SMM and of MUO in this population and to estimate the associations between these movements with cerebral blood flow and perfusion.Methods In this prospective cohort study conducted in 15 intensive care units across Canada, we enrolled consecutive, brain-injured adults with an unconfounded Glasgow Coma Scale score of 3. Physicians conducted standardized DNC clinical evaluation, and participants underwent a brain CT-perfusion scan with CT-angiography reconstructions within a 2-h delay. We assessed the prevalence and semiology of SMM and MUO with descriptive statistics. We estimated the associations between SMM and MUO with cerebral blood flow and brain perfusion using generalized linear mixed models with a logit link function, age and sex as covariates, and random intercepts for study sites.Results We included 282 participants with a median [IQR] age of 60 [47-69] years. The respective prevalence of SMM and MUO were 27% (95% CI: 22-32%) and 12% (95% CI: 9-16%). SMM and MUO were not associated with the presence of cerebral blood flow on CT-angiography (aOR for SMM: 1.14, 95% CI: 0.63-2.05; aOR for MUO: 1.36, 95% CI: 0.61-3.01) or brain perfusion on CT-perfusion (aOR for SMM: 1.44, 95% CI: 0.77-2.68; aOR for MUO: 1.75, 95% CI: 0.77-3.97). Findings were similar in the subgroup of 204 patients fulfilling clinical criteria for DNC.Conclusions SMM and MUO are common among comatose patients at risk of DNC. Their prevalence is similar among alive comatose patients and patients fulfilling clinical criteria for DNC. These movements are not associated with cerebral blood flow on CT-angiography or brain perfusion on CT-perfusion.Trial registration: Registered on ClinicalTrials.gov: NCT03098511 on March 27, 2017.

Study Design Retrospective single-center cohort study. Objective To evaluate associations between clinical, radiographic, and surgical factors and the risk of implant failure and postoperative complications following spinal stabilization for metastatic disease. Summary of Background Data Spinal metastases frequently cause pain, neurological deficits, and instability, often requiring surgical decompression and stabilization. As survival improves and the use of spinal instrumentation increases, understanding surgery-associated complication risks has become increasingly important. While postoperative complications in metastatic spine surgery have been reported across multiple cohorts, factors related to construct design and anchorage, including implant failure, wound infection, and postoperative bleeding, remain incompletely characterized. This study examines the incidence of these complications and their associated factors in patients undergoing stabilization for metastatic spinal disease. Methods This retrospective study retrospectively analyzed 149 patients who underwent stabilization surgery for spinal metastases between 2010 and 2020. Instability was assessed using the Spinal Instability Neoplastic Score (SINS), functional status using the Karnofsky Performance Score, and neurological status using the Frankel classification. Associations between patient-, disease-, and surgery-related variables and postoperative outcomes were examined primarily using univariate logistic regression analyses. Results In univariate analyses, implant failure was significantly associated with instrumentation anchored in metastatically involved adjacent vertebrae (OR 9.22), reoperation for tumor progression (OR 8.25), preoperative Frankel Score E (OR 5.25), active smoking (OR 4.05), and postoperative wound infection (OR 4.88). Postoperative wound infection was associated with construct anchorage in metastatically involved adjacent vertebrae (OR 4.78) and a preoperative ACE-27 comorbidity score of 1 (OR 3.54). Postoperative bleeding or hematoma was associated with wound infection (OR 5.21), advanced age at surgery (OR 1.09), impaired preoperative mobility (OR 3.78), and lower postoperative Karnofsky Performance Score (OR 0.96). The Spinal Instability Neoplastic Score was associated with implant failure in univariate analysis but did not account for construct-end anchorage. Conclusions Anchoring spinal constructs in vertebrae affected by metastases was consistently associated with increased risks of implant failure and postoperative infection following stabilization surgery. These findings highlight the mechanical and biological vulnerability of metastatic bone and suggest that, when feasible, fixation into structurally uninvolved vertebrae may reduce postoperative complications.

Objective: Spinal infection (SI) is associated with substantial morbidity and mortality, and optimal treatment strategies remain debated, particularly in medically fragile patients. The Mortality in Spinal Infection (MSI-20) score was developed as the first dedicated prognostic tool for mortality risk estimation, but external validation in large cohorts has been lacking. This study aimed to externally validate the MSI-20 in a large multicenter registry cohort and assess its performance, clinical thresholds, and generalizability.Methods: This retrospective multi-institutional registry study included 1,122 adult patients with clinically, radiologically, and laboratory-confirmed SI treated at 6 tertiary referral centers between 2010 and 2023. The multicenter registry design enabled robust analysis of this relatively rare outcome. MSI-20 scores were calculated according to the original definition. Predictive performance was evaluated using receiver operating characteristic (ROC) analysis with 95% confidence intervals. Mortality across score strata and center-wise performance were analyzed, with exploratory assessment of additional baseline predictors.Results: Mean age was 66.4±13.1 years, and overall mortality was 18.3%. The MSI-20 demonstrated fair discriminative ability (area under the curve [AUC], 0.68; 95% confidence interval [CI], 64–72). The optimal ROC threshold was 3.5; a cutoff ≥4 yielded sensitivity 0.49 and specificity 0.74 (F1 score, 0.38). Mortality increased progressively, approaching 50% (47.4%) at scores ≥9. Performance was consistent across centers (AUC, 0.60–0.75). Exploratory univariable analyses did not identify additional baseline variables with consistent strong associations beyond the MSI-20 components.Conclusion: This large multicenter registry study provides the first external validation of the MSI-20, confirming reproducibility and generalizability. The score demonstrated fair discrimination and a clear risk gradient across increasing score categories. Given its simplicity and high negative predictive value, the MSI-20 may serve as an adjunctive tool to support risk communication, multidisciplinary discussion, patient counseling, and preoperative optimization in spinal infection.

In moderate-to-severe traumatic brain injury (TBI), anemia may exacerbate secondary cerebral injury, and patients with multiple trauma may be especially vulnerable to cerebral hypoxia. We conducted a secondary analysis of the HEMOTION trial (liberal transfusion strategy > 10 g/dL vs. restrictive > 7 g/dL), to assess whether multiple trauma modifies the effect of transfusion strategy after moderate-to-severe TBI. We included all HEMOTION trial participants (n = 742) and defined multiple trauma using three definitions: (1) extracranial injury with Injury Severity Score (ISS) > 15; (2) extracranial injury requiring emergency extracranial surgery; (3) spinal injury with neurological deficit. The primary outcome was the 6-month Glasgow Outcome Scale Extended (GOS-E). We tested interactions between transfusion strategy and multiple trauma status using sliding dichotomy and hierarchical Poisson regression, with sensitivity analyses using classical dichotomy (GOS-E ≤ 4) and proportional odds models. Secondary outcomes included mortality, quality of life (EQ-5D-5L, QOLIBRI), functional independence (FIM), and depression (PHQ-9). We found no interaction between multiple trauma status and transfusion strategies on the 6-month GOS-E across all three definitions. The adjusted relative risk (RR) of an unfavourable outcome with the liberal strategy was 0.87 (95

Controlled donation after circulatory determination of death (DCD) is feasible only if circulatory arrest occurs soon after withdrawal of life-sustaining measures (WLSM). When organ recovery cannot proceed because this time interval is too long, there are potential negative implications, including perceptions of “secondary loss” for patients’ families and significant resource consumption. The DCD-N score is a validated clinical tool for predicting rapid death following WLSM. We hypothesized that neuroimaging evidence of effaced perimesencephalic cisterns improves prediction of time to death compared with the DCD-N score alone. In a retrospective population-based cohort study, DCD-N scores were prospectively determined in patients for whom consent for DCD had been obtained. Perimesencephalic cisterns on last available neuroimaging were assessed in duplicate and classified as normal, partially effaced, or completely effaced. Multivariable logistic regression assessed the capacity of DCD-N score and effaced cisterns to predict death within 1, 2, or 3 h of WLSM. Of 164 consecutive patients, 49 (30

BACKGROUND:Odontoid fractures of the second cervical vertebra commonly affect elderly patients due to osteoporosis and low-energy trauma. Treatment is controversial, with prolonged cervical collar immobilization risking non-union and complications, and surgical C1-C2 stabilization involving higher upfront surgical risks. High-level evidence from randomized controlled trials to guide optimal treatment decisions is lacking. The SCORE study aims to determine whether surgical stabilization is non-inferior to conservative collar management in maintaining functional independence for elderly patients with unstable odontoid fractures. METHODS:SCORE is a multicenter, parallel-group, randomized controlled non-inferiority trial enrolling 322 patients aged ≥70 years with acute (≤2 weeks) unstable odontoid Type II, III, or atypical fractures. Participants will be randomized 1:1, stratified by center, to receive surgical stabilization via posterior C1-C2 fixation or conservative management with a rigid cervical collar. The primary outcome measure is the change in Barthel Index (BI) from baseline to 12 weeks. Secondary outcomes include quality of life (EQ-5D), neck pain (Visual Analog Scale, VAS), neck disability (Neck Disability Index, NDI), radiographic fusion, treatment compliance, cross-over rates to surgery, and incidence of adverse and serious adverse events up to 6 months. Follow-ups will take place at 12 weeks and 6 months post-injury, with an additional visit at approximately 2 weeks post-surgery for surgical patients. Analysis will use mixed models for repeated measures, targeting 90% power to detect non-inferiority within a 5-point margin on the BI (one-sided α = 0.025), accounting for 15% attrition. DISCUSSION:This trial addresses a critical evidence gap by directly comparing surgical and conservative treatments, aiming to guide clinical decision-making and improve functional outcomes and quality of life in elderly patients. TRIAL REGISTRATION:ClinicalTrials.gov, ID: NCT06961578.

BACKGROUND:Brain injury related to hypoxic-ischemic insults post-cardiac arrest is a highly morbid and often fatal condition for which neuroprognostication remains challenging. There has been a significant increase in studies assessing the accuracy of multimodal approaches in predicting poor neurological outcomes post-cardiac arrest, and contemporary guidelines recommend this approach. We conducted a systematic review to assess multimodal versus unimodal approaches in neuroprognostication for predicting a poor neurological outcome for adult post-cardiac arrest patients at hospital discharge or beyond. METHODS:PRISMA methodological standards were followed. MEDLINE, EMBASE and CINAHL were searched from inception until January 18, 2024, with no restrictions. Abstract and full-text review was completed in duplicate. Original studies assessing the prognostic accuracy (specificity and false positive rate [FPR]) of multimodal compared with unimodal approaches were included. The risk of bias was assessed using the QUIPS tool. Data were extracted in duplicate. RESULTS:Of 791 abstracts, 12 studies were included. The FPR in predicting poor neurological outcomes ranged from 0% to 5% using a multimodal approach compared to 0% to 31% with a unimodal test. The risk of bias was moderate to high for most components. CONCLUSIONS:A multimodal approach may improve the FPR in predicting poor neurological outcomes of post-cardiac arrest patients.

Spinal melanocytic tumors are rare, with limited data on their clinical course and aggressiveness. Since intraoperative dark pigmentation and infiltrative margins can lead to misclassification of MC as MM—especially in the absence of a known primary—this multicenter study characterizes melanocytomas (MC) and contrasts them with malignant melanoma (MM). We retrospectively analyzed 56 patients with spinal intradural melanocytic tumors (2010–2024) from seven European neurosurgical centers. Clinical, radiological, histological, and surgical features were analyzed. Univariate and multivariate analyses were performed to identify prognostic risk factors. The study included 22 patients with spinal MC and 34 patients with MM. Median age was comparable (61 vs. 58 years, p = 0.09), but MC patients had a longer symptom history (13 vs. 1.3 months, p = 0.0001) and more often intramedullary tumors (72.7

BACKGROUND CONTEXT:Percutaneous screw osteosynthesis is the gold standard for managing sacral fragility fractures in geriatric patients with immobilizing pain. However, comparative evidence regarding the optimal type, length, or insertion position of sacroiliac screws remains limited. PURPOSE:This study aimed to compare outcomes between long transsacral screws bridging both sacroiliac joints and short sacroiliac screws. STUDY DESIGN/SETTING:Retrospective cohort single-center study. PATIENT SAMPLE:Geriatric patients treated with percutaneous sacroiliac screws for sacral fragility fractures. OUTCOME MEASURES:Primary outcome: screw loosening at 3-, 6-, and 12-month follow-ups. SECONDARY OUTCOMES:surgical duration, postoperative pain, mobility improvement, and hospital stay length. METHODS:Data from 122 patients (median age 81, 84 % female) treated between 2018 and 2021 were analyzed. Patients were categorized into three groups [1]: two long transsacral screws [2], a combination of one long and two short screws, and[3] four short sacroiliac screws. Fracture characteristics, FFP classification, and risk factors for screw loosening were evaluated. RESULTS:Fractures were bilateral in 73 %, with FFP classifications of type 2 (48 %), type 3 (12 %), and type 4 (40 %). Anterior pelvic fractures were present in 63 %, comminuted fractures in 34 %, and H-type fractures in 29 %. Loosening rates were 17 % in the long-screw group, 6 % in the combination group, and 4 % in the short-screw group. Surgical duration was shortest for long screws (mean 52.6 min) compared to the combination (61.8 min) and short-screw (83.4 min) groups. Pain scores decreased below 5 in 88 % of patients at 3 months and 92 % at 12 months. Screw length was a significant risk factor for loosening (p = 0.04). CONCLUSIONS:Long transsacral screws offer minimally invasive fixation with reduced surgical duration but higher loosening rates. Osteosynthesis with four short sacroiliac screws demonstrates superior long-term stability, making it a promising option for sacral fragility fractures.

BACKGROUND CONTEXT: Spondylodiscitis management presents significant clinical challenges, particularly in critically ill patients, where the risks and benefits of surgical intervention must be carefully balanced. The optimal timing of surgery in this context remains a subject of debate. PURPOSE: This study aims to evaluate the effectiveness of early surgery versus delayed surgery or conservative management in critically ill patients with de novo pyogenic spondylodiscitis. STUDY DESIGN/SETTING: This is an international, multicenter retrospective cohort study involving 24 centers, primarily in Europe. PATIENT SAMPLE: The study included 192 critically ill patients (65.63% male) with a median age of 69 years, all severely affected by pyogenic spondylodiscitis characterized by an initial CRP level >200 mg/l or the presence of two out of four Systemic Inflammatory Response Syndrome criteria upon admission. OUTCOME MEASURES: The primary outcome was 30-day mortality. Secondary outcomes included length of ICU stay, length of hospital stay, and relapse rates of spondylodiscitis. METHODS: Patients were divided into three groups: early surgery (within three days of admission), delayed surgery (after three days of admission), and conservative therapy. Propensity score matching and multivariate regression analyses were performed to adjust for baseline differences and assess the impact of treatment modalities on mortality and other clinical outcomes. RESULTS: Delayed surgery was associated with significantly lower 30-day mortality (4.05%) compared to early surgery (27.85%) and conservative therapy (27.78%) (p<.001). Delayed surgery also resulted in shorter hospital stays (42.76 days) compared to conservative therapy (55.53 days) and early surgery (26.33 days) (p<.001), and shorter ICU stays (4.52 days) compared to conservative therapy (16.48 days) and early surgery (7.92 days) (p<.001). The optimal window for surgery, minimizing mortality, was identified as ten to fourteen days postadmission (p=.02). Risk factors for increased mortality included age (p<.05), multiple organ failure (p<.05), and vertebral body destruction (p<.05), whereas delayed surgery (p<.05) and the presence of an epidural abscess were associated with reduced mortality (p<.05). CONCLUSIONS: Delayed surgery, optimally between 10 to 14 days postadmission, was associated with lower mortality in critically ill spondylodiscitis patients. These findings highlight the potential benefits of considering surgical timing to improve patient outcomes. (c) 2024 The Author(s). Published by Elsevier Inc. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/)

INTRODUCTION:Sacral fractures are an increasingly recognized clinical entity, particularly among older adults with osteoporosis. However, national-level data on long-term trends in incidence, diagnostic imaging, treatment strategies, and the recognition of underlying bone fragility remain limited. This study aimed to evaluate nationwide trends in sacral fracture care in Germany over a 19-year period. METHODS:We conducted a retrospective analysis of all inpatient cases with a primary diagnosis of sacral fracture (ICD-10-GM: S32.1) recorded in the German Federal Statistical Office database from 2005 to 2023. Fragility fractures were defined as cases in patients aged ≥65 years. Outcomes included use of CT and MRI (OPS codes), surgical versus conservative treatment, and coded diagnoses of osteoporosis (ICD-10: M80-M82). Time trends were analyzed using linear regression; group comparisons were conducted with t-tests and chi-square tests (p < 0.05). RESULTS:A total of 162,116 sacral fractures were identified. Annual cases increased from 1,861 in 2005 to 7,695 in 2023. Fragility fractures in women aged ≥65 years rose significantly, from 985 to 12,901 cases (p < 0.0001). CT use increased by 241% and MRI by 175%, with a significant shift toward CT as the preferred modality (p < 0.0001). Despite increased access to minimally invasive options, surgical treatment rates remained stable at approximately 20% (p = 0.15). Osteoporosis was documented in only 1.5% of cases. CONCLUSIONS:The incidence of sacral fractures in Germany has risen markedly, driven by an aging population and under-recognized bone fragility. While cross-sectional imaging use has expanded, surgical treatment remains underutilized, and osteoporosis continues to be grossly underdiagnosed. These findings underscore a systemic gap in secondary prevention and highlight the need for integrated, bone-focused trauma care models.

Background Artificial intelligence (AI) models have shown potential for diagnosing and prognosticating traumatic spinal cord injury (tSCI), but their clinical utility remains uncertain. Method ology: The primary aim was to evaluate the performance of AI algorithms in diagnosing and prognosticating tSCI. Subsequent systematic searching of seven databases identified studies evaluating AI models. PROBAST and TRIPOD tools were used to assess the quality and reporting of included studies (PROSPERO: CRD42023464722). Fourteen studies, comprising 20 models and 280,817 pooled imaging datasets, were included. Analysis was conducted in line with the SWiM guidelines. Results For prognostication, 11 studies predicted outcomes including AIS improvement (30%), mortality and ambulatory ability (20% each), and discharge or length of stay (10%). The mean AUC was 0.770 (range: 0.682–0.902), indicating moderate predictive performance. Diagnostic models utilising DTI, CT, and T2-weighted MRI with CNN-based segmentation achieved a weighted mean accuracy of 0.898 (range: 0.813–0.938), outperforming prognostic models. Conclusion AI demonstrates strong diagnostic accuracy (mean accuracy: 0.898) and moderate prognostic capability (mean AUC: 0.770) for tSCI. However, the lack of standardised frameworks and external validation limits clinical applicability. Future models should integrate multimodal data, including imaging, patient characteristics, and clinician judgment, to improve utility and alignment with clinical practice.

Introduction:In cases of space-occupying cerebellar ischemic strokes, guidelines recommend suboccipital decompressive surgery (SDC). While in supratentorial hemispheric stroke, the size of the bone flap has been the subject of many studies and ample debate, no studies have been conducted to determine the optimal size of the bone flap to be removed in SDC. Research question:To determine the optimal size of SDC in ischemic cerebellar stroke. Methods:This is a multicentric retrospective study of patients undergoing SDC for ischemic cerebellar stroke. SDC size was determined in two perpendicular planes on early postoperative CT scans: (a) maximal lateral extension (L) and (b) maximal craniocaudal extension (CC) in cm. The primary endpoint was functional outcome according to modified Rankin Scale (mRS) at three months. Secondary outcome was mortality at three months, as well as surgical complications. Results:A total of 88 patients were included in the final analysis. The mean L diameter of the SDC analyzed was 7 cm (SD 1.5), whereas the mean CC diameter was 4.4 cm (SD .8). When dichotomizing patients based on a threshold of L ≥ 6.5 cm, favorable outcome was more likely in the group with L ≥ 6.5 cm (OR = 3.23, 95%CI 1.02-10.56, p = .045). No statistically significant differences were observed in mortality at three months (OR = .66, 95%CI .24-1.78, p = .40). Conclusions:In ischemic cerebellar stroke, a suboccipital craniectomy with a maximum lateral diameter of ≥6.5 cm appears to yield better functional outcomes than smaller ones. Prospective studies are needed to confirm these results.