Related Experiment Video
Updated: Feb 21, 2026
09:14
Pre-Chiasmatic, Single Injection of Autologous Blood to Induce Experimental Subarachnoid Hemorrhage in a Rat Model
Published on: June 18, 2021
2.9K
Electronic Health Data Predict Outcomes After Aneurysmal Subarachnoid Hemorrhage
Sahar F Zafar1, Eva N Postma2, Siddharth Biswal2
1Department of Neurology, Lunder 6 Neurosciences Intensive Care Unit, Massachusetts General Hospital, 55 Fruit Street, Boston, MA, 02114, USA. sfzafar@mgh.harvard.edu.
Neurocritical Care
|October 7, 2017
Summary
Predicting neurologic outcomes after aneurysmal subarachnoid hemorrhage (aSAH) is possible using clinical and physiological data. Key predictors include APACHE II score, Glasgow Coma Scale (GCS), and intracranial pressure (ICP) variance.
Area of Science:
- Neurology
- Critical Care Medicine
- Health Informatics
Background:
- Aneurysmal subarachnoid hemorrhage (aSAH) is a critical neurological condition.
- Identifying predictors of neurologic outcomes is crucial for patient management.
- Electronic health records offer a rich source of data for such predictions.
Purpose of the Study:
- To identify clinical and physiological parameters predicting neurologic outcomes after aSAH.
- To develop a predictive model for patient outcomes using electronic health data.
- To assess the utility of early physiological data variance in outcome prediction.
Main Methods:
- Retrospective cohort study of 153 patients with aSAH (2011-2016).
- Evaluation of 473 predictor variables, including laboratory and physiological data (min, max, median, variance for first 3 days).
- Development of penalized logistic regression and multivariate multilevel models to predict Glasgow Outcome Scale (GOS) outcomes.
Main Results:
- Multivariate analysis identified APACHE II score, Glasgow Coma Scale (GCS), white blood cell (WBC) count, mean arterial pressure, serum glucose variance, intracranial pressure (ICP) variance, and serum sodium as predictors of mortality.
- Predictors of death/dependence versus independence (GOS 4-5) included levetiracetam, mechanical ventilation, WBC count, heart rate, ICP variance, GCS, APACHE II, and epileptiform discharges.
- A multiclass prediction model achieved >80% accuracy for poor/good outcomes and >70% for intermediate outcomes, using GCS, APACHE II, periodic discharges, lacosamide, and rebleeding.
Conclusions:
- Early physiological data variance significantly impacts patient outcomes after aSAH.
- Intracranial pressure (ICP), glucose levels, and electroencephalography patterns are valuable for disease severity and risk stratification.
- These electronically retrievable features can guide early goal-directed therapy.
