Introduction: The authors introduced a symptomatic cerebral vasospasm (SCV) prediction model built with freeware based on a 91-patient dataset. In a prospective test group of 22 patients at the same hospital, this model outperformed logistic regression models in vasospasm prediction. One of the model's limitations was a question of reproducibility in other centers. In this report, the authors describe their experience with the prospective use of the model and discuss its utility to predict SCV following aneurysmal subarachnoid hemorrhage.

Methods: Patient data of 25 consecutive cases of aneurysm rupture were prospectively assessed by the model to predict SCV. The prediction was then compared with actual outcome. For the purpose of this report, SCV is defined as a delayed focal decline in neurological examination correlated with an area of radiographic vasospasm. This serves as the primary end point of the predictive model. Each case prediction is reported, along with strength of prediction, which is built into the model. The model's positive predictive value, negative predictive value, and sensitivity and specificity are reported.

Results: Twenty-five patients were included in the analysis. Six patients (24%) were diagnosed clinically and radiographically with SCV. The model predicted that 9 of those patients would have SCV (positive predictive value of 67%). The model predicted that 16 patients would not have SCV and was correct in all cases (negative predictive value of 100%). The sensitivity of the model was 100%, and the specificity of the was 84%.

Conclusions: Our neural network model of SCV takes into account several evidence-based factors and provides a useful tool with high accuracy to predict symptomatic vasospasm following aneurysm subarachnoid hemorrhage

Patient Care: Our research provides a predictive model for symptomatic cerebral vasospasm which can improve patient care by identifying patients at a high and low risk of SCV which when treated expeditiously, results in improved outcomes.

Learning Objectives: By the end of this session, participants should be able to: 1) Understand the role of neural networks to analyze complex, multi-factorial events, such as cerebral vasospasm, 2) Recognize and discuss the variables used to create our neural network model of SCV, 3) Appreciate the predictive value of the neural network model and identify potential limitations

References: Dumont TM, Rughani AI, Tranmer BI. Prediction of symptomatic cerebral vasospasm after aneurysmal subarachnoid hemorrhage with an artificial neural network: feasibility and comparison with logistic regression models. World Neurosurg. 2011;75:57-63.