Michael L Martini1, Sean N Neifert1, William H Shuman1, Emily K Chapman1, Alexander J Schüpper1, Eric K Oermann1, J Mocco1, Michael Todd2, James C Torner3, Andrew Molyneux4, Stephan Mayer5, Peter Le Roux6, Mervyn D I Vergouwen7, Gabriel J E Rinkel7, George K C Wong8, Peter Kirkpatrick9, Audrey Quinn10, Daniel Hänggi11, Nima Etminan12, Walter M van den Bergh13, Blessing N R Jaja14,15,16, Michael Cusimano17, Tom A Schweizer16, Jose I Suarez18, Hitoshi Fukuda19, Sen Yamagata19, Benjamin Lo20, Airton Leonardo de Oliveira Manoel21, Hieronymus D Boogaarts22, R Loch Macdonald23. 1. 1Department of Neurosurgery, Mount Sinai Health System, New York, New York. 2. 2Department of Anesthesiology, University of Minnesota, Minneapolis, Minnesota. 3. 3Departments of Epidemiology, Surgery, and Neurosurgery, College of Public Health and Carver College of Medicine, University of Iowa, Iowa City, Iowa. 4. 4Nuffield Department of Surgical Sciences, University of Oxford, United Kingdom. 5. 5Wayne State University School of Medicine, Detroit, Michigan. 6. 6Bassett HealthCare, Cooperstown, New York. 7. 7Brain Center, Department of Neurology and Neurosurgery, University Medical Center Utrecht, Utrecht University, Utrecht, The Netherlands. 8. 8Division of Neurosurgery, Prince of Wales Hospital and the Chinese University of Hong Kong, China. 9. 9University of Cambridge, Nuffield Health Cambridge Hospital, Cambridge, United Kingdom. 10. 10Department of Anaesthesia, Cheriton House, James Cook University Hospital, Middlesbrough, United Kingdom. 11. 11Department of Neurosurgery, Düsseldorf University Hospital, Heinrich-Heine-Universität, Düsseldorf, Germany. 12. 12Department of Neurosurgery, Medical Faculty Mannheim, University of Heidelberg, Mannheim, Germany. 13. 13Department of Critical Care, University of Groningen, University Medical Center Groningen, The Netherlands. 14. Divisions of14Neurosurgery and. 15. 15Neurology, St. Michael's Hospital, Toronto, Ontario. 16. 16Neuroscience Research Program, Keenan Research Centre for Biomedical Science and Li Ka Shing Knowledge Institute, St. Michael's Hospital, Department of Surgery, University of Toronto, Ontario. 17. 17Education and Public Health, St. Michael's Hospital, University of Toronto, Keenan Research Centre and Li Ka Shing Knowledge Institute, Toronto, Ontario, Canada. 18. 18Departments of Anesthesiology and Critical Care Medicine, Neurology, and Neurosurgery, Johns Hopkins University School of Medicine, Baltimore, Maryland. 19. 19Department of Neurosurgery, Kurashiki Central Hospital, Kurashiki-city, Okayama, Japan. 20. 20Department of Neurosurgery, Lenox Hill Hospital, New York, New York. 21. 21Department of Critical Care, Hospital Israelita Albert Einstein and Hospital Alemao Oswaldo Cruz, São Paulo, Brazil. 22. 22Department of Neurosurgery, Radboud University Medical Center, Nijmegen, The Netherlands; and. 23. 23University of California San Francisco, Fresno Campus, University Neurosciences Institutes, Fresno, California.
Abstract
OBJECTIVE: Rescue therapies have been recommended for patients with angiographic vasospasm (aVSP) and delayed cerebral ischemia (DCI) following subarachnoid hemorrhage (SAH). However, there is little evidence from randomized clinical trials that these therapies are safe and effective. The primary aim of this study was to apply game theory-based methods in explainable machine learning (ML) and propensity score matching to determine if rescue therapy was associated with better 3-month outcomes following post-SAH aVSP and DCI. The authors also sought to use these explainable ML methods to identify patient populations that were more likely to receive rescue therapy and factors associated with better outcomes after rescue therapy. METHODS: Data for patients with aVSP or DCI after SAH were obtained from 8 clinical trials and 1 observational study in the Subarachnoid Hemorrhage International Trialists repository. Gradient boosting ML models were constructed for each patient to predict the probability of receiving rescue therapy and the 3-month Glasgow Outcome Scale (GOS) score. Favorable outcome was defined as a 3-month GOS score of 4 or 5. Shapley Additive Explanation (SHAP) values were calculated for each patient-derived model to quantify feature importance and interaction effects. Variables with high SHAP importance in predicting rescue therapy administration were used in a propensity score-matched analysis of rescue therapy and 3-month GOS scores. RESULTS: The authors identified 1532 patients with aVSP or DCI. Predictive, explainable ML models revealed that aneurysm characteristics and neurological complications, but not admission neurological scores, carried the highest relative importance rankings in predicting whether rescue therapy was administered. Younger age and absence of cerebral ischemia/infarction were invariably linked to better rescue outcomes, whereas the other important predictors of outcome varied by rescue type (interventional or noninterventional). In a propensity score-matched analysis guided by SHAP-based variable selection, rescue therapy was associated with higher odds of 3-month GOS scores of 4-5 (OR 1.63, 95% CI 1.22-2.17). CONCLUSIONS: Rescue therapy may increase the odds of good outcome in patients with aVSP or DCI after SAH. Given the strong association between cerebral ischemia/infarction and poor outcome, trials focusing on preventative or therapeutic interventions in these patients may be most able to demonstrate improvements in clinical outcomes. Insights developed from these models may be helpful for improving patient selection and trial design.
OBJECTIVE: Rescue therapies have been recommended for patients with angiographic vasospasm (aVSP) and delayed cerebral ischemia (DCI) following subarachnoid hemorrhage (SAH). However, there is little evidence from randomized clinical trials that these therapies are safe and effective. The primary aim of this study was to apply game theory-based methods in explainable machine learning (ML) and propensity score matching to determine if rescue therapy was associated with better 3-month outcomes following post-SAH aVSP and DCI. The authors also sought to use these explainable ML methods to identify patient populations that were more likely to receive rescue therapy and factors associated with better outcomes after rescue therapy. METHODS: Data for patients with aVSP or DCI after SAH were obtained from 8 clinical trials and 1 observational study in the Subarachnoid Hemorrhage International Trialists repository. Gradient boosting ML models were constructed for each patient to predict the probability of receiving rescue therapy and the 3-month Glasgow Outcome Scale (GOS) score. Favorable outcome was defined as a 3-month GOS score of 4 or 5. Shapley Additive Explanation (SHAP) values were calculated for each patient-derived model to quantify feature importance and interaction effects. Variables with high SHAP importance in predicting rescue therapy administration were used in a propensity score-matched analysis of rescue therapy and 3-month GOS scores. RESULTS: The authors identified 1532 patients with aVSP or DCI. Predictive, explainable ML models revealed that aneurysm characteristics and neurological complications, but not admission neurological scores, carried the highest relative importance rankings in predicting whether rescue therapy was administered. Younger age and absence of cerebral ischemia/infarction were invariably linked to better rescue outcomes, whereas the other important predictors of outcome varied by rescue type (interventional or noninterventional). In a propensity score-matched analysis guided by SHAP-based variable selection, rescue therapy was associated with higher odds of 3-month GOS scores of 4-5 (OR 1.63, 95% CI 1.22-2.17). CONCLUSIONS: Rescue therapy may increase the odds of good outcome in patients with aVSP or DCI after SAH. Given the strong association between cerebral ischemia/infarction and poor outcome, trials focusing on preventative or therapeutic interventions in these patients may be most able to demonstrate improvements in clinical outcomes. Insights developed from these models may be helpful for improving patient selection and trial design.