BACKGROUND AND PURPOSE: Delayed cerebral ischemia (DCI) after aneurysmal subarachnoid hemorrhage negatively impacts long-term recovery but is often detected too late to prevent damage. We aim to develop hourly risk scores using routinely collected clinical data to detect DCI. METHODS: A DCI classification model was trained using vital sign measurements (heart rate, blood pressure, respiratory rate, and oxygen saturation) and demographics routinely collected for clinical care. Twenty-two time-varying physiological measures were computed including mean, SD, and cross-correlation of heart rate time series with each of the other vitals. Classification was achieved using an ensemble approach with L2-regularized logistic regression, random forest, and support vector machines models. Classifier performance was determined by area under the receiver operating characteristic curves and confusion matrices. Hourly DCI risk scores were generated as the posterior probability at time t using the Ensemble classifier on cohorts recruited at 2 external institutions (n=38 and 40). RESULTS: Three hundred ten patients were included in the training model (median, 54 years old [interquartile range, 45-65]; 80.2% women, 28.4% Hunt and Hess scale 4-5, 38.7% Modified Fisher Scale 3-4); 101 (33%) developed DCI with a median onset day 6 (interquartile range, 5-8). Classification accuracy before DCI onset was 0.83 (interquartile range, 0.76-0.83) area under the receiver operating characteristic curve. Risk scores applied to external institution datasets correctly predicted 64% and 91% of DCI events as early as 12 hours before clinical detection, with 2.7 and 1.6 true alerts for every false alert. CONCLUSIONS: An hourly risk score for DCI derived from routine vital signs may have the potential to alert clinicians to DCI, which could reduce neurological injury.
BACKGROUND AND PURPOSE: Delayed cerebral ischemia (DCI) after aneurysmal subarachnoid hemorrhage negatively impacts long-term recovery but is often detected too late to prevent damage. We aim to develop hourly risk scores using routinely collected clinical data to detect DCI. METHODS: A DCI classification model was trained using vital sign measurements (heart rate, blood pressure, respiratory rate, and oxygen saturation) and demographics routinely collected for clinical care. Twenty-two time-varying physiological measures were computed including mean, SD, and cross-correlation of heart rate time series with each of the other vitals. Classification was achieved using an ensemble approach with L2-regularized logistic regression, random forest, and support vector machines models. Classifier performance was determined by area under the receiver operating characteristic curves and confusion matrices. Hourly DCI risk scores were generated as the posterior probability at time t using the Ensemble classifier on cohorts recruited at 2 external institutions (n=38 and 40). RESULTS: Three hundred ten patients were included in the training model (median, 54 years old [interquartile range, 45-65]; 80.2% women, 28.4% Hunt and Hess scale 4-5, 38.7% Modified Fisher Scale 3-4); 101 (33%) developed DCI with a median onset day 6 (interquartile range, 5-8). Classification accuracy before DCI onset was 0.83 (interquartile range, 0.76-0.83) area under the receiver operating characteristic curve. Risk scores applied to external institution datasets correctly predicted 64% and 91% of DCI events as early as 12 hours before clinical detection, with 2.7 and 1.6 true alerts for every false alert. CONCLUSIONS: An hourly risk score for DCI derived from routine vital signs may have the potential to alert clinicians to DCI, which could reduce neurological injury.
Authors: Mervyn D I Vergouwen; Marinus Vermeulen; Jan van Gijn; Gabriel J E Rinkel; Eelco F Wijdicks; J Paul Muizelaar; A David Mendelow; Seppo Juvela; Howard Yonas; Karel G Terbrugge; R Loch Macdonald; Michael N Diringer; Joseph P Broderick; Jens P Dreier; Yvo B W E M Roos Journal: Stroke Date: 2010-08-26 Impact factor: 7.914
Authors: J Michael Schmidt; Katja E Wartenberg; Andres Fernandez; Jan Claassen; Fred Rincon; Noeleen D Ostapkovich; Neeraj Badjatia; Augusto Parra; E Sander Connolly; Stephan A Mayer Journal: J Neurosurg Date: 2008-12 Impact factor: 5.115
Authors: Michael Veldeman; Walid Albanna; Miriam Weiss; Catharina Conzen; Tobias Philip Schmidt; Henna Schulze-Steinen; Martin Wiesmann; Hans Clusmann; Gerrit Alexander Schubert Journal: J Neurosurg Date: 2020-05-15 Impact factor: 5.115
Authors: Jose I Suarez; Muniza K Sheikh; R Loch Macdonald; Sepideh Amin-Hanjani; Robert D Brown; Airton Leonardo de Oliveira Manoel; Colin P Derdeyn; Nima Etminan; Emanuela Keller; Peter D Leroux; Stephan A Mayer; Akio Morita; Gabriel Rinkel; Daniel Rufennacht; Martin N Stienen; James Torner; Mervyn D I Vergouwen; George K C Wong Journal: Neurocrit Care Date: 2019-06 Impact factor: 3.210
Authors: Harshal Deshmukh; Matthew Hinkley; Louise Dulhanty; Hiren C Patel; J P Galea Journal: Acta Neurochir (Wien) Date: 2016-03-01 Impact factor: 2.216
Authors: Brandon P Lucke-Wold; Aric F Logsdon; Branavan Manoranjan; Ryan C Turner; Evan McConnell; George Edward Vates; Jason D Huber; Charles L Rosen; J Marc Simard Journal: Int J Mol Sci Date: 2016-04-02 Impact factor: 5.923
Authors: Jens Witsch; Valérie Spalart; Kimberly Martinod; Hauke Schneider; Joachim Oertel; Jürgen Geisel; Philipp Hendrix; Sina Hemmer Journal: Crit Care Explor Date: 2022-05-17
Authors: István Csók; Jürgen Grauvogel; Christian Scheiwe; Jürgen Bardutzky; Thomas Wehrum; Jürgen Beck; Peter C Reinacher; Roland Roelz Journal: Front Neurol Date: 2022-03-02 Impact factor: 4.003