A Study to Determine the Predictive Value of Intraoperative Neuromonitoring during Resection of Posterior Fossa Tumors at a Tertiary Facility in Kenya

Background: Intraoperative neuromonitoring involves the continuous evaluation of one of the neural pathways in an anesthetized patient. Multimodal intraoperative neuromonitoring involves the utility of somatosensory and motor-induced potentials as well as electromyography to monitor the integrity of the nervous system during surgery. Multimodal intraoperative neuromonitoring has been used in both adult and pediatric neurosurgery. However, there still remains institutional variation in protocols depending on technology and alert criteria used. Additional studies that have demonstrated potential benefits of utilizing checklists for addressing neuromonitoring alerts during surgery still recommend large-scale multicenter trial to demonstrate clinical efficacy. The intent of this study is to assess the efficacy of the multimodal techniques in detecting and preventing neurologic injury in patients with posterior fossa tumors undergoing surgical resection.

Broad Objective: This article aims to determine predictive value of multimodal intraoperative neuromonitoring during resection of posterior fossa tumors at a tertiary facility in Kenya.

Methods: This was a prospective cohort study. Thirty patients with brain tumor resected under intraoperative neuromonitoring were followed up to the time of discharge or death. A structured questionnaire was used to collect both the demographic and basic clinical data. The data collected included the age, sex, Karnofsky score, neurological deficits, type of brain tumor, and location of brain tumor.

Data Analysis: The data collected using the standard questionnaire was coded and entered into a password protected excel sheet. Cleaning and editing to ensure accuracy and completeness was done before transfer to SPSS version 26 for analysis. Descriptive analysis of the categorical data including age, sex, type of brain tumor was done using frequencies and percentages. Data were presented in bar graphs and pie charts.

Results: Significant intraoperative coMEP alerts occurred in 27.8% of patients, with 33.3% experiencing adverse intraoperative events. Immediate postoperative complications were present in 44.4% of patients, predominantly cranial nerve palsies. The correlation between intraoperative events and coMEP alerts was not significant (Pearson’s correlation = 0.326). However, coMEP alerts significantly predicted immediate (Chi-square = 0.003) and 24-hour postoperative deficits (Chi-square = 0.001), with logistic regression indicating a strong association with cranial nerve palsy (p = 0.011, OR = 48). ROC analysis showed high predictive value for 24-hour (AUC = 0.962) and 1-month postoperative deficits (AUC = 0.917).

Conclusion: coMEP alerts are significant predictors of immediate and short-term postoperative neurological deficits but not long-term outcomes. Tumor location and histopathological type did not significantly influence intraoperative alerts or postoperative deficits. These findings highlight the importance of comprehensive intraoperative monitoring to improve surgical outcomes.

Publikationsverlauf

Artikel online veröffentlicht:
07. Februar 2025