Evaluating the efficacy of intraoperative NIRS cutoff values in detecting spinal cord ischemia during surgery
摘要
Purpose: Paralysis is a serious complication of surgeries that interferes with the blood supply of the anterior spinal cord, with rates of spinal cord injury (SCI) from approximately 1% in general spine surgeries to 4–40% following thoracoabdominal aortic aneurysm (TAAA) repair. Near-infrared spectroscopy (NIRS) provides a non-invasive, real-time method for monitoring tissue oxygenation, largely unaffected by anesthetics. Given the heightened risk of neurologic injury during TAAA repair, this procedure is used to evaluate the effectiveness of standard regional spinal oxygen saturation (rSpO₂) cutoff values in predicting neurological outcomes. Methods: This retrospective study analyzed 25 patients undergoing open TAAA repair. NIRS data were recorded at the ischemic site and a reference location throughout surgery. Neurological outcomes were assessed postoperatively based on paralysis, hemiparesis, or extremity weakness. After excluding eight patients due to poor signal quality, 17 patients were included. NIRS values at six key time points were compared between outcome groups. Bayesian statistics assessed the relationship between significant NIRS “drops” (< 80% of baseline) and neurological outcomes. Results: Seven patients exhibited new neurological deficits (4 temporary). No credible association was found between intraoperative NIRS drops and postoperative neurological outcomes at any analyzed time point. A moderate effect was observed at the end of surgery (Hedges’ g = − 1.21), suggesting a potential difference between groups, although the Bayesian credible interval included zero (posterior mean = − 0.82, 94% HDI [− 1.8, 0.18]). Conclusions: In this limited cohort, intraoperative NIRS cutoff values did not significantly correlate with postoperative neurological deficits following TAAA repair. Postoperative NIRS monitoring may be more informative for detecting spinal cord ischemia and preventing paralysis.