An Experimental Model to Measure Irrigation-Induced Cerebrospinal Fluid Alterations During Interlaminar Spine Endoscopy

Endoscopic spine surgery is a rapidly advancing minimally invasive technique, generally associated with low complication rates when performed by experienced surgeons. However, cerebral complications such as seizures have been reported during the procedure. The underlying mechanisms behind these complications remain poorly understood. We hypothesize the composition of cerebrospinal fluid (CSF), specifically glucose levels, might be altered due to an irrigation caused wash-out, potentially contributing to cerebral events observed during endoscopic spine surgeries. The study was performed in 4 human cadavers. After a sacral approach to the dural sac, it was filled with 20% glucose solution. An intradural catheter was introduced into the cisterna magna through a percutaneous C0/C1 approach and an external ventricular drain was placed in the side ventricle. Catheters' positioning was confirmed through contrast-enhanced X-ray imaging and intracranial pressure measurements, ensuring correct placement and sufficient filling of the dural sac. We then conducted interlaminar spine endoscopy with irrigation pressure set at 40 mmHg and Ringer's solution as irrigation fluid. 1ml of CSF was repeatedly obtained from the cisterna magna during 40 minutes as a baseline measurement. A 5×5 mm dural tear was created at the endoscopy level, and in the first two specimens 1ml of CSF was directly collected from the cisterna magna at intervals of 1,3,5,10,15,20, and 30 minutes. In the second two specimens, measurements were conducted in the same intervals but re-filling of the dural sac with glucose was performed after previous endoscopy. Glucose concentrations were quantified through laboratory chemical analysis and visually assessed using reagent strips. Intradural glucose levels remained stable for the observed duration of 40 minutes during endoscopy. Irrigation in the presence of a dura lesion caused an immediate decrease in glucose concentrations in the cisterna magna in each specimen from 589mmol/l, 549mmol/l, 28mmol/l and 64mmol/l, respectively, to 307mmol/l, 87mmol/l, 3mmol/l, 6mmol/l, respectively, after 30min. This reduction was consistent with visual readings from the reagent strips. The study observed a notable reduction of intradural glucose concentration during interlaminar thoracic spine endoscopy in the presence of a dural lesion. Over the 30-minute irrigation period, glucose levels decreased approx. 2 to 10 times compared to initial measurements. This suggests that alterations in the CSF composition, including glucose concentration, could play a role in cerebral complications during endoscopic spine surgery. These findings warrant further investigation into the pathophysiology of cerebral events related to changes in CSF during endoscopic procedures.