Rescue therapy for vasospasm following aneurysmal subarachnoid hemorrhage: a propensity score–matched analysis with machine learning

View More View Less
  • 1 Department of Neurosurgery, Mount Sinai Health System, New York, New York;
  • | 2 Department of Anesthesiology, University of Minnesota, Minneapolis, Minnesota;
  • | 3 Departments of Epidemiology, Surgery, and Neurosurgery, College of Public Health and Carver College of Medicine, University of Iowa, Iowa City, Iowa;
  • | 4 Nuffield Department of Surgical Sciences, University of Oxford, United Kingdom;
  • | 5 Wayne State University School of Medicine, Detroit, Michigan;
  • | 6 Bassett HealthCare, Cooperstown, New York;
  • | 7 Brain Center, Department of Neurology and Neurosurgery, University Medical Center Utrecht, Utrecht University, Utrecht, The Netherlands;
  • | 8 Division of Neurosurgery, Prince of Wales Hospital and the Chinese University of Hong Kong, China;
  • | 9 University of Cambridge, Nuffield Health Cambridge Hospital, Cambridge, United Kingdom;
  • | 10 Department of Anaesthesia, Cheriton House, James Cook University Hospital, Middlesbrough, United Kingdom;
  • | 11 Department of Neurosurgery, Düsseldorf University Hospital, Heinrich-Heine-Universität, Düsseldorf, Germany;
  • | 12 Department of Neurosurgery, Medical Faculty Mannheim, University of Heidelberg, Mannheim, Germany;
  • | 13 Department of Critical Care, University of Groningen, University Medical Center Groningen, The Netherlands;
  • | 14 Divisions of Neurosurgery and
  • | 15 Neurology, St. Michael’s Hospital, Toronto, Ontario;
  • | 16 Neuroscience 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 Education and Public Health, St. Michael’s Hospital, University of Toronto, Keenan Research Centre and Li Ka Shing Knowledge Institute, Toronto, Ontario, Canada;
  • | 18 Departments of Anesthesiology and Critical Care Medicine, Neurology, and Neurosurgery, Johns Hopkins University School of Medicine, Baltimore, Maryland;
  • | 19 Department of Neurosurgery, Kurashiki Central Hospital, Kurashiki-city, Okayama, Japan;
  • | 20 Department of Neurosurgery, Lenox Hill Hospital, New York, New York;
  • | 21 Department of Critical Care, Hospital Israelita Albert Einstein and Hospital Alemao Oswaldo Cruz, São Paulo, Brazil;
  • | 22 Department of Neurosurgery, Radboud University Medical Center, Nijmegen, The Netherlands; and
  • | 23 University of California San Francisco, Fresno Campus, University Neurosciences Institutes, Fresno, California
Restricted access

Purchase Now

USD  $45.00

JNS + Pediatrics - 1 year subscription bundle (Individuals Only)

USD  $515.00

JNS + Pediatrics + Spine - 1 year subscription bundle (Individuals Only)

USD  $612.00
Print or Print + Online

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.

ABBREVIATIONS

ALISAH = Albumin in Subarachnoid Hemorrhage; aVSP = angiographic vasospasm; BP = blood pressure; CONSCIOUS-1 = Clazosentan to Overcome Neurological Ischemia and Infarction Occurring After Subarachnoid Hemorrhage; DCI = delayed cerebral ischemia; DSAT = Database of Subarachnoid Treatment; GCS = Glasgow Coma Scale; GOS = Glasgow Outcome Scale; HHU = Heinrich Heine University; ICA = internal carotid artery; ML = machine learning; NEWTON-1 = Nimodipine Microparticles to Enhance Recovery While Reducing Toxicity After Subarachnoid Hemorrhage; RCT = randomized controlled trial; ROC = receiver operating characteristic; SAH = subarachnoid hemorrhage; SAHIT = Subarachnoid Hemorrhage International Trialists; SHAP = Shapley Additive Explanation; TCD = transcranial Doppler; WFNS = World Federation of Neurosurgical Societies.

Supplementary Materials

    • Supplemental Data (PDF 595 KB)

Illustration from Schneider et al. (pp 205–214). Copyright Elyssa Siegel. Published with permission.

JNS + Pediatrics - 1 year subscription bundle (Individuals Only)

USD  $515.00

JNS + Pediatrics + Spine - 1 year subscription bundle (Individuals Only)

USD  $612.00
  • 1

    Connolly ES Jr, Rabinstein AA, Carhuapoma JR, et al. Guidelines for the management of aneurysmal subarachnoid hemorrhage: a guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke. 2012;43(6):17111737.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 2

    Diringer MN, Bleck TP, Claude Hemphill J III, et al. Critical care management of patients following aneurysmal subarachnoid hemorrhage: recommendations from the Neurocritical Care Society’s Multidisciplinary Consensus Conference. Neurocrit Care. 2011;15(2):211240.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 3

    Ray WZ, Moran CJ, Derdeyn CP, et al. Near-complete resolution of angiographic cerebral vasospasm after extreme elevation of mean arterial pressure: case report. Surg Neurol. 2009;72(4):347354.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 4

    Dankbaar JW, Slooter AJ, Rinkel GJ, Schaaf IC. Effect of different components of triple-H therapy on cerebral perfusion in patients with aneurysmal subarachnoid haemorrhage: a systematic review. Crit Care. 2010;14(1):R23.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 5

    Appelboom G, Strozyk D, Hwang BY, et al. Bedside use of a dual aortic balloon occlusion for the treatment of cerebral vasospasm. Neurocrit Care. 2010;13(3):385388.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 6

    Jun P, Ko NU, English JD, et al. Endovascular treatment of medically refractory cerebral vasospasm following aneurysmal subarachnoid hemorrhage. AJNR Am J Neuroradiol. 2010;31(10):19111916.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 7

    Shankar JJ, dos Santos MP, Deus-Silva L, Lum C. Angiographic evaluation of the effect of intra-arterial milrinone therapy in patients with vasospasm from aneurysmal subarachnoid hemorrhage. Neuroradiology. 2011;53(2):123128.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 8

    Shankar JJ, Tan IY, Krings T, et al. CT angiography for evaluation of cerebral vasospasm following acute subarachnoid haemorrhage. Neuroradiology. 2012;54(3):197203.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 9

    Smith WS, Dowd CF, Johnston SC, et al. Neurotoxicity of intra-arterial papaverine preserved with chlorobutanol used for the treatment of cerebral vasospasm after aneurysmal subarachnoid hemorrhage. Stroke. 2004;35(11):25182522.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 10

    Terry A, Zipfel G, Milner E, et al. Safety and technical efficacy of over-the-wire balloons for the treatment of subarachnoid hemorrhage-induced cerebral vasospasm. Neurosurg Focus. 2006;21(3):E14.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 11

    Lennihan L, Mayer SA, Fink ME, et al. Effect of hypervolemic therapy on cerebral blood flow after subarachnoid hemorrhage: a randomized controlled trial. Stroke. 2000;31(2):383391.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 12

    Zwienenberg-Lee M, Hartman J, Rudisill N, et al. Effect of prophylactic transluminal balloon angioplasty on cerebral vasospasm and outcome in patients with Fisher grade III subarachnoid hemorrhage: results of a phase II multicenter, randomized, clinical trial. Stroke. 2008;39(6):17591765.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 13

    Gathier CS, van den Bergh WM, van der Jagt M, et al. Induced hypertension for delayed cerebral ischemia after aneurysmal subarachnoid hemorrhage: a randomized clinical trial. Stroke. 2018;49(1):7683.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 14

    Macdonald RL, Cusimano MD, Etminan N, et al. Subarachnoid Hemorrhage International Trialists data repository (SAHIT). World Neurosurg. 2013;79(3-4):418422.

  • 15

    Macdonald RL, Kassell NF, Mayer S, et al. Clazosentan to overcome neurological ischemia and infarction occurring after subarachnoid hemorrhage (CONSCIOUS-1): randomized, double-blind, placebo-controlled phase 2 dose-finding trial. Stroke. 2008;39(11):30153021.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 16

    Suarez JI, Martin RH, Calvillo E, et al. The Albumin in Subarachnoid Hemorrhage (ALISAH) multicenter pilot clinical trial: safety and neurologic outcomes. Stroke. 2012;43(3):683690.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 17

    Le Roux PD, Elliott JP, Newell DW, et al. Predicting outcome in poor-grade patients with subarachnoid hemorrhage: a retrospective review of 159 aggressively managed cases. J Neurosurg. 1996;85(1):3949.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 18

    Etminan N, Beseoglu K, Eicker SO, et al. Prospective, randomized, open-label phase II trial on concomitant intraventricular fibrinolysis and low-frequency rotation after severe subarachnoid hemorrhage. Stroke. 2013;44(8):21622168.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 19

    Hänggi D, Etminan N, Aldrich F, et al. Randomized, open-label, phase 1/2a study to determine the maximum tolerated dose of intraventricular sustained release nimodipine for subarachnoid hemorrhage (NEWTON [Nimodipine Microparticles to Enhance Recovery While Reducing Toxicity After Subarachnoid Hemorrhage]). Stroke. 2017;48(1):145151.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 20

    Haley EC Jr, Kassell NF, Apperson-Hansen C, et al. A randomized, double-blind, vehicle-controlled trial of tirilazad mesylate in patients with aneurysmal subarachnoid hemorrhage: a cooperative study in North America. J Neurosurg. 1997;86(3):467474.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 21

    Kassell NF, Haley EC Jr, Apperson-Hansen C, Alves WM. Randomized, double-blind, vehicle-controlled trial of tirilazad mesylate in patients with aneurysmal subarachnoid hemorrhage: a cooperative study in Europe, Australia, and New Zealand. J Neurosurg. 1996;84(2):221228.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 22

    Lanzino G, Kassell NF. Double-blind, randomized, vehicle-controlled study of high-dose tirilazad mesylate in women with aneurysmal subarachnoid hemorrhage. Part II. A cooperative study in North America. J Neurosurg. 1999;90(6):10181024.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 23

    Vergouwen MD, Vermeulen M, van Gijn J, et al. Definition of delayed cerebral ischemia after aneurysmal subarachnoid hemorrhage as an outcome event in clinical trials and observational studies: proposal of a multidisciplinary research group. Stroke. 2010;41(10):23912395.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 24

    Report of World Federation of Neurological Surgeons committee on a universal subarachnoid hemorrhage grading scale. J Neurosurg. 1988;68(6):985986.

  • 25

    Frontera JA, Claassen J, Schmidt JM, et al. Prediction of symptomatic vasospasm after subarachnoid hemorrhage: the modified Fisher scale. Neurosurgery. 2006;59(1):2127.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 26

    McMillan T, Wilson L, Ponsford J, et al. The Glasgow Outcome Scale—40 years of application and refinement. Nat Rev Neurol. 2016;12(8):477485.

  • 27

    Zhang Z, Zhao Y, Canes A, et al. Predictive analytics with gradient boosting in clinical medicine. Ann Transl Med. 2019;7(7):152.

  • 28

    Lundberg S, Erion G, Chen H, et al. Explainable AI for trees: from local explanations to global understanding. arXiv. Preprint posted online May 11, 2010. https://arxiv.org/abs/1905.04610

    • Search Google Scholar
    • Export Citation
  • 29

    Lundberg SM, Nair B, Vavilala MS, et al. Explainable machine-learning predictions for the prevention of hypoxaemia during surgery. Nat Biomed Eng. 2018;2(10):749760.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 30

    Lundberg SM, Lee S-I. A unified approach to interpreting model predictions. Adv Neural Inf Process Syst. 2017;30:47654774.

  • 31

    Fujimoto K, Kojadinovic IK, Marichal JL. Axiomatic characterizations of probabilistic and cardinal-probabilistic interaction indices. Games Econ Behav. 2006;55(1):7299.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 32

    Jaja BNR, Saposnik G, Lingsma HF, et al. Development and validation of outcome prediction models for aneurysmal subarachnoid haemorrhage: the SAHIT multinational cohort study. BMJ. 2018;360:j5745.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 33

    Frontera JA, Fernandez A, Schmidt JM, et al. Defining vasospasm after subarachnoid hemorrhage: what is the most clinically relevant definition? Stroke. 2009;40(6):19631968.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 34

    Venkatraman A, Khawaja AM, Gupta S, et al. Intra-arterial vasodilators for vasospasm following aneurysmal subarachnoid hemorrhage: a meta-analysis. J Neurointerv Surg. 2018;10(4):380387.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 35

    Maher M, Schweizer TA, Macdonald RL. Treatment of spontaneous subarachnoid hemorrhage: guidelines and gaps. Stroke. 2020;51(4):13261332.

Metrics

All Time Past Year Past 30 Days
Abstract Views 93 93 93
Full Text Views 87 87 87
PDF Downloads 127 127 127
EPUB Downloads 0 0 0