Predictors of mortality at 3 months in patients with skull base tumor resections in a low-income setting

Globally, skull base tumors are among the most challenging tumors to treat and are known for their significant morbidity and mortality. Hence, this study aimed to identify robust associated factors that contribute to mortality of patients following surgical resection for a variety of skull base tumors at the 3-month follow-up period. This in turn helps devise an evidence-based meticulous treatment strategy and baseline input for quality improvement work.

A retrospective cohort study of patients undergoing skull base tumor resection was conducted at two large-volume neurosurgery centers in Ethiopia. The categorical variables were expressed in frequencies and percentages. Normal distribution of continuous data was checked by histogram and the Shapiro-Wilk test. Median with interquartile range (IQR) was calculated for skewed data, while mean with standard deviation (SD) was used for normally distributed data. Odds ratio and adjusted odds ratio (AOR) were used to express the result of univariate and multivariate binary logistic analyses, respectively. A p-value <0.005 was considered statistically significant at 95% confidence interval (CI).

The study involved 266 patients. Of this, women accounted for 63.5% of patients. The median age of patients was 37 (±IQR = 17) years while the median size of the tumor in this study was 4.9 (±IQR 1.5) cm. The mean duration of symptoms at time of presentation was 17.3 (±SD = 11.1) months. Meningioma, pituitary adenoma, and craniopharyngioma contributed to 68.4%, 19.2%, and 9% of the skull-based tumors, respectively. Mortality following skull base tumor resection was 21.1%. On multivariable binary logistic regression analysis, intraoperative iatrogenic vascular insult (AOR = 28.76, 95% CI: 6.12-135.08, p = 0.000), intraventricular hemorrhage (AOR = 6.32, 95% CI: 1.19-33.63, p = 0.031), hospital-associated infection (AOR = 6.96, 95% CI: 2.04-23.67, p = 0.002), and extubation time exceeding 24 h (AOR = 12.89, 95% CI: 4.89-40.34, p = 0.000) were statistically significant with 3-month mortality.

Mortality from skull base tumor resection remains high in our setting. Holistic pre-operative surgical planning, meticulous intraoperative execution of procedures, and post-operative dedicated follow-up of patients in a neurointensive care unit alongside quality improvement works on identified risks of mortality are strongly recommended to improve patient outcomes. The urgent need for setup improvement and further training of neurosurgeons is also underscored.