Validation of a Model for Outcome Prediction After Endovascular Treatment for Ischemic Stroke

Overview

Journal Interv Neuroradiol

Publisher Sage Publications

Specialty Neurology

Date 2024 Jul 25

PMID 39053025

Authors

Pedro Martins

Krishna Sambhu

Mohamed Tarek

Jaydevsinh Dolia

Aqueel Pabaney

Jonathan Grossberg

Raul Nogueira

Diogo Haussen

Affiliations

Soon will be listed here.

Abstract

Introduction: The recently developed MR-PREDICTS@24 h model showed excellent performance in the MR-CLEAN Registry cohort in patients presenting within 12 h from onset. However, its applicability to an U.S. population and to patients presenting beyond 12 h from last known normal are still undetermined. We aim to externally validate the MR-PREDICTS@24 h model in a new geographic setting and in the late window.

Methods: In this retrospective analysis of a prospectively collected database from a comprehensive stroke center in the United States, we included patients with intracranial carotid artery or middle cerebral artery M1 or M2 segment occlusions who underwent endovascular therapy and applied the MR-PREDICTS@24 h formula to estimate the probabilities of functional outcome at day 90. The primary endpoint was the modified Rankin Scale (mRS) at 90 days.

Results: We included 1246 patients, 879 in the early (<12 h) and 367 in the late (≥12 h) cohort. For both cohorts, calibration and discrimination of the model were accurate throughout mRS levels, with absolute differences between estimated and predicted proportions ranging from 1% to 5%. Calibration metrics and curve inspections showed good performance for estimating the probabilities of mRS ≤ 1 to mRS ≤ 5 for the early cohort. For the late cohort, predictions were reliable for the probabilities of mRS ≤ 1 to mRS ≤ 4.

Conclusion: The MR-PREDICTS@24 h was transferrable to a real-world U.S.-based cohort in the early window and showed consistently accurate predictions for patients presenting in the late window without need for updating.

References

Goyal M, Fargen K, Turk A, Mocco J, Liebeskind D, Frei D . 2C or not 2C: defining an improved revascularization grading scale and the need for standardization of angiography outcomes in stroke trials. J Neurointerv Surg. 2013; 6(2):83-6. PMC: 4156591. DOI: 10.1136/neurintsurg-2013-010665. View

Saver J, Goyal M, Bonafe A, Diener H, Levy E, Pereira V . Stent-retriever thrombectomy after intravenous t-PA vs. t-PA alone in stroke. N Engl J Med. 2015; 372(24):2285-95. DOI: 10.1056/NEJMoa1415061. View

Collins G, Reitsma J, Altman D, Moons K . Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD): the TRIPOD Statement. BMC Med. 2015; 13:1. PMC: 4284921. DOI: 10.1186/s12916-014-0241-z. View

Schomaker M, Heumann C . Bootstrap inference when using multiple imputation. Stat Med. 2018; 37(14):2252-2266. PMC: 5986623. DOI: 10.1002/sim.7654. View

Kremers F, Venema E, Duvekot M, Yo L, Bokkers R, Lycklama A Nijeholt G . Outcome Prediction Models for Endovascular Treatment of Ischemic Stroke: Systematic Review and External Validation. Stroke. 2021; 53(3):825-836. PMC: 8884132. DOI: 10.1161/STROKEAHA.120.033445. View

Van Calster B, McLernon D, van Smeden M, Wynants L, Steyerberg E . Calibration: the Achilles heel of predictive analytics. BMC Med. 2019; 17(1):230. PMC: 6912996. DOI: 10.1186/s12916-019-1466-7. View

Chalos V, Venema E, Mulder M, Roozenbeek B, Steyerberg E, Wermer M . Development and Validation of a Postprocedural Model to Predict Outcome After Endovascular Treatment for Ischemic Stroke. JAMA Neurol. 2023; . PMC: 10391355. DOI: 10.1001/jamaneurol.2023.2392. View

Karamchandani R, Satyanarayana S, Yang H, Defilipp G, Strong D, Rhoten J . The Charlotte large artery occlusion endovascular therapy outcome score predicts outcome after basilar artery thrombectomy. J Neuroimaging. 2022; 32(5):860-865. DOI: 10.1111/jon.13039. View

Vandenbroucke J, von Elm E, Altman D, Gotzsche P, Mulrow C, Pocock S . Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaboration. PLoS Med. 2007; 4(10):e297. PMC: 2020496. DOI: 10.1371/journal.pmed.0040297. View

10.

Kniep H, Meyer L, Bechstein M, Broocks G, Guerreiro H, van Horn N . How Much of the Thrombectomy Related Improvement in Functional Outcome Is Already Apparent at 24 Hours and at Hospital Discharge?. Stroke. 2022; 53(9):2828-2837. DOI: 10.1161/STROKEAHA.121.037888. View

11.

Feyen L, Rohde S, Weinzierl M, Katoh M, Haage P, Munnich N . Outcome prediction prior to thrombectomy of the posterior circulation with machine learning. Interv Neuroradiol. 2023; :15910199231168164. DOI: 10.1177/15910199231168164. View

12.

Tan I, Demchuk A, Hopyan J, Zhang L, Gladstone D, Wong K . CT angiography clot burden score and collateral score: correlation with clinical and radiologic outcomes in acute middle cerebral artery infarct. AJNR Am J Neuroradiol. 2009; 30(3):525-31. PMC: 7051470. DOI: 10.3174/ajnr.A1408. View

13.

Anadani M, Finitsis S, Pop R, Darcourt J, Clarencon F, Richard S . ASPECTS evolution after endovascular successful reperfusion in the early and extended time window. Interv Neuroradiol. 2023; :15910199231196953. DOI: 10.1177/15910199231196953. View

14.

Van Calster B, Nieboer D, Vergouwe Y, De Cock B, Pencina M, Steyerberg E . A calibration hierarchy for risk models was defined: from utopia to empirical data. J Clin Epidemiol. 2016; 74:167-76. DOI: 10.1016/j.jclinepi.2015.12.005. View

15.

Steyerberg E, Moons K, van der Windt D, Hayden J, Perel P, Schroter S . Prognosis Research Strategy (PROGRESS) 3: prognostic model research. PLoS Med. 2013; 10(2):e1001381. PMC: 3564751. DOI: 10.1371/journal.pmed.1001381. View

16.

Nieboer D, Vergouwe Y, Roobol M, Ankerst D, Kattan M, Vickers A . Nonlinear modeling was applied thoughtfully for risk prediction: the Prostate Biopsy Collaborative Group. J Clin Epidemiol. 2015; 68(4):426-34. PMC: 4474141. DOI: 10.1016/j.jclinepi.2014.11.022. View

17.

Hallen S, Hootsmans N, Blaisdell L, Gutheil C, Han P . Physicians' perceptions of the value of prognostic models: the benefits and risks of prognostic confidence. Health Expect. 2014; 18(6):2266-77. PMC: 5810722. DOI: 10.1111/hex.12196. View