Abstract TP550: Clinical and Demographic Predictors of Cerebral Vasopasm Occurrence Following Sub-arachnoid Hemorrhages

Abstract

Introduction: Cerebral vasospasm (CV) often follows aneurysmal subarachnoid hemorrhage (SAH) and, when severe, may lead to delayed cerebral ischemia (DCI) that can contribute substantially to morbidity and mortality. It is important to identify patients at high risk for DCI in order to allow for closer monitoring and timely treatment. This study analyzes the association of DCI with co-morbidities and demographic variables. Hypothesis: Diseases of the circulatory system may be predictive of DCI risk in SAH patients. Methods: Data from a total of 20039 SAH events between 2013 and 2018 were collected from 178 US hospitals. The occurrence of DCI during the SAH hospitalization was determined with ICD 9/10 diagnosis codes. Patient age was compared between the DCI and non-DCI SAH cases using t-tests. Odds ratios and Fisher exact tests were used to evaluate the relative incidence of DCI according to patient sex and the presence/absence of comorbidities and other clinical covariates (only those that occurred in at least 5% of SAH patients). Statistical significance of co-morbidities associated with DCI occurrence was evaluated following adjustment for false discovery rates. Results: In our cohort, 874 (4.36%) SAH patients experienced an episode of DCI. SAH patients with DCI were significantly younger than non-DCI SAH patients (DCI mean 60.40, non-DCI 66.95, p << 0.001), and there was a significantly higher relative incidence of DCI in female vs male patients (OR = 1.967, p << 0.001). Several co-morbidities were strongly associated with DCI occurrence, most notably hypoosmolality (OR=2.67), anemia (OR=2.15), hyperglycemia (OR=2.15), disorders of phosphorus metabolism (OR =2.55) and hyperlipidemia (OR=1.52), all with p << 0.001. In contrast, conditions such as hyperosmolality (OR=0.233), acidosis (OR=0.354), sepsis (0.393), diabetes (0.581) and kidney failure (OR=0.483) have a significant (p <<0.001) negative association with DCI. Hypertension does not appear to be significantly predictive of DCI or its absence. Conclusions: The data in this study suggest that age and sex are strongly predictive of DCI risk, and that several facets of an SAH patient’s metabolite and ion balance may either protect or predispose him/her to CV and DCI incidence.