Neurosurgery
-
Monitoring of electrocorticography for the purpose of detecting spreading depolarization (SD) events is becoming increasingly used both for research and clinical applications. Although such monitoring bears many similarities to standard long-term epilepsy monitoring, there are a number of differences that neurosurgeons need to be aware of when initiating such a program. In addition, most of the focus in SD monitoring has been on traumatic and vascular conditions, where invasive monitoring is used commonly, but electrocorticography is not commonly used. ⋯ It is also recognized that this is a rapidly evolving field and that new advances may disrupt these approaches and that there is a diversity of opinion on these topics, even among SD experts. Nonetheless, the general approach to SD monitoring has now been in use for >15 years and is the basis for several active and proposed clinical trials (NCT05337618, NCT04966546), so an understanding from a neurosurgical perspective of the rationale and approach to monitoring is warranted. In this review, we will consider the potential indications for SD monitoring in clinical trials or clinical care, the methodology for recording and interpreting, and finally some potential therapeutic approaches that are being considered in patients with clinically detected SD.
-
Recent studies have proposed computed tomography (CT) criteria for posterior ligamentous complex (PLC) injury: disrupted if ≥2 CT findings, indeterminate if single finding, and intact if 0 CT findings. The study aims to validate the CT criteria for PLC injury externally. ⋯ This study externally validates the previously proposed CT criteria for PLC injury. A total of ≥2 positive CT findings or 0 CT findings can be used as criteria for a disrupted PLC (B-type injury) or intact PLC (A-type injuries), respectively, without added MRI. A single CT finding implies indeterminate PLC status and the need for further MRI assessment. The CT criteria will potentially guide MRI indications and treatment decisions for neurologically intact thoracolumbar burst fractures.
-
Diagnosing ventriculostomy-related infection (VRI), a common complication after external ventricular drainage (EVD), is challenging and often associated with delayed initiation of antibiotic therapy. We aimed to develop a stewardship score to help in the decision of antibiotic therapy initiation when VRI is suspected. ⋯ The VERI score is a robust, predictive tool for assessing the risk of VRI in patients with EVD, potentially guiding more judicious use of antibiotic therapy in the intensive care unit setting.
-
Classical biomedical data science models are trained on a single modality and aimed at one specific task. However, the exponential increase in the size and capabilities of the foundation models inside and outside medicine shows a shift toward task-agnostic models using large-scale, often internet-based, data. Recent research into smaller foundation models trained on specific literature, such as programming textbooks, demonstrated that they can display capabilities similar to or superior to large generalist models, suggesting a potential middle ground between small task-specific and large foundation models. This study attempts to introduce a domain-specific multimodal model, Congress of Neurological Surgeons (CNS)-Contrastive Language-Image Pretraining (CLIP), developed for neurosurgical applications, leveraging data exclusively from Neurosurgery Publications. ⋯ This study presents a pioneering effort in building a domain-specific multimodal model using data from a medical society publication. The results indicate that domain-specific models, while less globally versatile, can offer advantages in specialized contexts. This emphasizes the importance of using tailored data and domain-focused development in training foundation models in neurosurgery and general medicine.