Machine Learning Predicts Cerebral Vasospasm in Subarachnoid Hemorrhage Patients

TL;DR

Machine learning models accurately predict cerebral vasospasm in subarachnoid hemorrhage patients up to a week in advance, helping improve ICU care and resource use.

Contribution

The study introduces a novel machine learning model for early prediction of verapamil-requiring cerebral vasospasm in SAH patients using multi-center data.

Findings

Machine learning models achieved an AUC of 0.88 in predicting verapamil-requiring cerebral vasospasm up to a week in advance.

The models successfully ruled out 8% of non-verapamil patients with zero false negatives.

External validation confirmed similar performance at a separate institution.

Abstract

Cerebral vasospasm (CV) is a feared complication occurring in 20–40% of patients following subarachnoid hemorrhage (SAH) and is known to contribute to delayed cerebral ischemia. It is standard practice to admit SAH patients to intensive care for an extended period of vigilant, resource-intensive, clinical monitoring. We used machine learning to predict CV requiring verapamil (CVRV) in the largest and only multi-center study to date.

SAH patients admitted to UCLA from 2013–2022 and a validation cohort from VUMC from 2018–2023 were included. For each patient, 172 unique intensive care unit (ICU) variables were extracted through the primary endpoint, namely first verapamil administration or ICU downgrade. At each institution, a light gradient boosting machine (LightGBM) was trained using five- fold cross validation to predict the primary endpoint at various timepoints during hospital admission. Receiver-operator curves (ROC) and precision-recall (PR) curves were generated.

A total of 1,750 patients were included from UCLA, 125 receiving verapamil. LightGBM achieved an area under the ROC (AUC) of 0.88 an average of over one week in advance, and successfully ruled out 8% of non-verapamil patients with zero false negatives. Minimum leukocyte count, maximum platelet count, and maximum intracranial pressure were the variables with highest predictive accuracy. Our models predicted “no CVRV” vs “CVRV within three days” vs “CVRV after three days” with AUCs=0.88, 0.83, and 0.88, respectively. For external validation at VUMC, 1,654 patients were included, 75 receiving verapamil. Predictive models at VUMC performed very similarly to those at UCLA, averaging 0.01 AUC points lower.

We present an accurate (AUC=0.88) and early (>1 week prior) predictor of CVRV using machine learning over two large cohorts of subarachnoid hemorrhage patients at separate institutions. This represents a significant step towards optimized clinical management and improved resource allocation in the intensive care setting of subarachnoid hemorrhage patients.

Figures