BACKGROUND: The most widely used classifications of adverse events (AEs) in neurosurgery define their severity according to the therapy used to treat them. This concept has substantial shortcomings because it does not reflect the severity of AEs that are not treated, such as new neurological deficits. OBJECTIVE: To present a novel multidimensional and patient-centered classification of the severity of AE in neurosurgery and evaluate its applicability. METHODS: The Therapy-Disability-Neurology (TDN) grading system classifies AEs depending on the associated therapy, disability, and neurological deficits. We conducted a 2-center retrospective observational study on 6071 interventions covering the whole neurosurgical spectrum with data prospectively recorded between 2013 and 2019 at 2 institutions from 2 countries. RESULTS: Using the first patient cohort (4680 interventions), a positive correlation was found between severity of AE and LOS as well as treatment cost. Each grade was associated with a greater deterioration of the Karnofsky Performance Status Scale (KPS) at discharge and at follow-up. When using the same methods on the external validation cohort (1391 interventions), correlations between the grades of AE, LOS, and KPS at discharge were even more pronounced. CONCLUSION: Our results suggest that the TDN grade is consistent with clinical and economic repercussions of AE and thus reflects AE severity. It is easily interpreted and enables comparison between different medical centers. The standardized report of the severity of AE in the scientific literature could constitute an important step forward toward a more critical, patient-centered, and evidence-based decision-making in neurosurgery.

Spinal arachnoid web (SAW) is a rare condition characterized by focal thickening of the arachnoid membrane causing displacement and compression of the spinal cord with progressive symptoms and neurological deficits. Recent reports and clinical experience suggest that SAW is a distinct entity with specific radiological findings and treatment strategies distinguishable from other arachnopathies and potential differential diagnoses. To better define the diagnostic and clinical features, treatment options and outcomes of surgically treated SAW. Multicentric retrospective cohort study. Twelve cases of SAW surgically treated at three different centers. Self-reported and neurological outcome measurements (pain, sensory-motor deficits, vegetative dysfunctions) were assessed at follow-up timepoints. Retrospective review of prospectively collected data on all patients surgically treated for SAW from three participating neurosurgical centers between 2014 and 2020. Clinicopathological data, including neurological presentation, radiological and histological findings and outcome data were analyzed. Twelve radiologically and surgically confirmed cases of SAW were analyzed. Mean patient age was 54.7 [±12.7], 67% were male. All SAWs were located in the posterior thoracic dural sac. On magnetic resonance imaging (MRI), the "scalpel sign" - a characteristic focal dorsal indentation of the spinal cord resembling a scalpel blade - was identified in all patients. A focal intramedullary syrinx was present in 83%. Preoperative clinical symptoms included signs of myelopathy, pain, weakness and sensory loss, most commonly affecting the trunk/upper back or lower extremities. Laminectomy or laminoplasty with intradural excision of the SAW was the surgical treatment of choice in all cases. Intraoperative ultrasound was valuable to visualize the cerebrospinal fluid (CSF) flow obstruction, confirm the SAW location before dura incision and to control adequacy of resection. After surgery, sensory loss and weakness in particular showed significant improvement. The present study comprises the largest series of surgically treated SAW, underscoring the unique clinical, radiographic, histopathological and surgical findings. We want to emphasize SAW being a distinct entity of spinal arachnopathy with a favorable long-term outcome if diagnosed correctly and treated surgically. Intraoperative ultrasound aids visualizing the SAW before dural incision, as well as verifying restored CSF flow after resection.

Outcome assessments after surgery for degenerative lumbar disorders (DLDs) rely on subjective patient-reported outcomes (PROMs). New objective functional capacity tests, like the smartphone-based 6-min walking test (6WT), have been introduced but presumably also do not reflect the patient's real-life functional performance. Pilot study to analyze changes in smartphone-based real-life activity data for physical performance outcome in patients undergoing surgery for DLD. Prospective observational study of DLD patients. Objective functional capacity and subjective outcomes were measured using 6WT and PROMs. Real-life physical performance data were acquired retrospectively using Apple iPhone Health data and compared against objective capacity and subjective outcomes. Eight patients (mean 46 years, 62% male) provided 286.858 smartphone mile counts. PROMs and physical capacity (6WT) significantly improved postoperatively. 6WT results increased from 352m pre-to 555/567m at 6/12 weeks postoperatively (p ​= ​0.03). For physical performance a linear mixed effect models showed an increase in daily distance in the first 4 months after surgery (slope +0.178; p ​< ​0.001). However, those increases reversed from 4 until 12 months postoperatively (negative slope estimate of −0.076; p ​< ​0.001). Smartphone-derived physical performance measures showed a positive correlation with corresponding physical capacity in the 6WT (R ​= ​0.57,p ​= ​0.004) and negative correlations with PROMs (COMI: R ​= ​−0.62p ​= ​0.001; ZCQ-Physical-Function: R ​= ​−0.68,p ​< ​0.001; ZCQ-Symptom-Severity: R ​= ​−0.52,p ​= ​0.009). Smartphone-based real-life activity data allows for longitudinal physical performance assessment. Physical performance correlated with physical capacity and patient's subjective perception of disability. However, physical performance may be more resistant to postoperative longtime change which should consult a more cautious use as objective outcome measure.

Recent therapeutic advances in metastatic melanoma have led to improved overall survival (OS) rates, with consequently an increased incidence of brain metastases (BM). The role of BM resection in the era of targeted and immunotherapy should be reassessed. In the current study we analysed the role of residual intracranial tumour load in a cohort of melanoma BM patients. Retrospective single-centre analysis of a prospective registry of resected melanoma BM from 2013 to 2021. Correlations of residual tumour volume and outcome were determined with respect to patient, tumour and treatment regimens characteristics. 121 individual patients (66% male, mean age 59.9 years) were identified and included in the study. Pre- and postoperative systemic treatments included BRAF/MEK inhibitors, as well as combination or monotherapy of immune-checkpoint inhibitors (ICIs). Median OS of the entire cohort was 20 months. Cox proportional-hazard analysis revealed postoperative anti-CTLA4+anti-PD-1 therapy (HR 0.07, p = .01) and postoperative residual intracranial tumour burden (HR 1.4, p = .027) as significant predictors for OS. Further analysis revealed that ICI-naïve patients with residual tumour volume ≤3.5 cm³ and postoperative ICI showed significantly prolonged OS compared to patients with residual volume >3.5 cm³ (p < .0001). Subgroup analysis of ICI-naïve patients showed steroid intake postoperatively to be negatively associated with OS, however residual tumour volume ≤3.5 cm³ remained independently correlated with superior OS (HR 0.14, p < .001). Besides known predictive factors like postoperative ICI, a maximal intracranial tumour burden reduction seems to be beneficial, especially in ICI-naïve patients. This highlights the importance of local CNS control and the need to further investigating the role of initial surgical tumour load reduction in randomised clinical trials.

This paper analyzes the regulatory frameworks for artificial intelligence/machine learning AI/ML-enabled medical devices in the European Union (EU), the United States (US), and the Republic of Korea, with a focus on applications in spine surgery. The aim is to provide guidance for developers and researchers navigating regulatory pathways. A review of current literature, regulatory documents, and legislative frameworks was conducted. Key differences in regulatory bodies, risk classification, submission requirements, and approval pathways for AI/ML medical devices were examined in the EU, US, and Republic of Korea. The EU AI Act (2024) establishes a risk-based framework, requiring regulatory review based on device risk, with high-risk devices subject to stricter oversight. The US applies a more flexible approach, allowing multiple submission pathways and incorporating a focus on continuous learning. The Republic of Korea emphasizes possibilities of streamlined approval and with growing use of real-world data to support validation. Developers must ensure regulatory alignment early in the development process, focusing on key aspects like dataset quality, transparency, and continuous monitoring. Across all regions, the need for technical documentation, quality management systems, and bias mitigation are essential for approval. Developers are encouraged to adopt adaptable strategies to comply with evolving regulatory standards, ensuring models remain transparent, fair, and reliable. The EU’s comprehensive AI Act enforces stricter oversight, while the US and Korea offer more flexible pathways. Developers of spine surgery AI/ML devices must tailor development strategies to align with regional regulations, emphasizing transparent development, quality assurance, and postmarket monitoring to ensure approval success.