Jonathan Elmer1, Bobby L Jones2, Daniel S Nagin3. 1. Departments of Emergency Medicine, Critical Care Medicine and Neurology, University of Pittsburgh School of Medicine, Pittsburgh, PA, USA. Electronic address: elmerjp@upmc.edu. 2. Western Psychiatric Institute and Clinic of UPMC, Pittsburgh, PA, USA. 3. Heinz College, Carnegie Mellon University, Pittsburgh, PA, USA.
Abstract
INTRODUCTION: Predicting outcome after cardiac arrest is challenging. We previously tested group-based trajectory modeling (GBTM) for prognostication based on baseline characteristics and quantitative electroencephalographic (EEG) trajectories. Here, we describe implementation of this method in a freely available software package and test its performance against alternative options. METHODS: We included comatose patients admitted to a single center after resuscitation from cardiac arrest from April 2010 to April 2019 who underwent ≥6 h of EEG monitoring. We abstracted clinical information from our prospective registry and summarized suppression ratio in 48 hourly epochs. We tested three classes of longitudinal models: frequentist, statistically based GBTMs; non-parametric (i.e. machine learning) k-means models; and Bayesian regression. Our primary outcome of interest was discharge CPC 1-3 (vs unconsciousness or death). We compared sensitivity for detecting poor outcome at a false positive rate (FPR) <1%. RESULTS: Of 1,010 included subjects, 250 (25%) were awake and alive at hospital discharge. GBTM and k-means derived trajectories, group sizes and group-specific outcomes were comparable. Conditional on an FPR < 1%, GBTMs yielded optimal sensitivity (38%) over 48 h. More sensitive methods had 2-3 % FPRs. CONCLUSION: We explored fundamentally different tools for patient-level predictions based on longitudinal and time-invariant patient data. Of the evaluated methods, GBTM resulted in optimal sensitivity while maintaining a false positive rate <1%. The provided code and software of this method provides an easy-to-use implementation for outcome prediction based on GBTMs.
INTRODUCTION: Predicting outcome after cardiac arrest is challenging. We previously tested group-based trajectory modeling (GBTM) for prognostication based on baseline characteristics and quantitative electroencephalographic (EEG) trajectories. Here, we describe implementation of this method in a freely available software package and test its performance against alternative options. METHODS: We included comatosepatients admitted to a single center after resuscitation from cardiac arrest from April 2010 to April 2019 who underwent ≥6 h of EEG monitoring. We abstracted clinical information from our prospective registry and summarized suppression ratio in 48 hourly epochs. We tested three classes of longitudinal models: frequentist, statistically based GBTMs; non-parametric (i.e. machine learning) k-means models; and Bayesian regression. Our primary outcome of interest was discharge CPC 1-3 (vs unconsciousness or death). We compared sensitivity for detecting poor outcome at a false positive rate (FPR) <1%. RESULTS: Of 1,010 included subjects, 250 (25%) were awake and alive at hospital discharge. GBTM and k-means derived trajectories, group sizes and group-specific outcomes were comparable. Conditional on an FPR < 1%, GBTMs yielded optimal sensitivity (38%) over 48 h. More sensitive methods had 2-3 % FPRs. CONCLUSION: We explored fundamentally different tools for patient-level predictions based on longitudinal and time-invariant patient data. Of the evaluated methods, GBTM resulted in optimal sensitivity while maintaining a false positive rate <1%. The provided code and software of this method provides an easy-to-use implementation for outcome prediction based on GBTMs.
Authors: Sunil B Nagaraj; Marleen C Tjepkema-Cloostermans; Barry J Ruijter; Jeannette Hofmeijer; Michel J A M van Putten Journal: Clin Neurophysiol Date: 2018-10-27 Impact factor: 3.708
Authors: Patrick J Coppler; Jonathan Elmer; Luis Calderon; Alexa Sabedra; Ankur A Doshi; Clifton W Callaway; Jon C Rittenberger; Cameron Dezfulian Journal: Resuscitation Date: 2015-01-28 Impact factor: 5.262
Authors: Clifton W Callaway; Michael W Donnino; Ericka L Fink; Romergryko G Geocadin; Eyal Golan; Karl B Kern; Marion Leary; William J Meurer; Mary Ann Peberdy; Trevonne M Thompson; Janice L Zimmerman Journal: Circulation Date: 2015-11-03 Impact factor: 29.690
Authors: Jonathan Elmer; Bobby L Jones; Vladimir I Zadorozhny; Juan Carlos Puyana; Kate L Flickinger; Clifton W Callaway; Daniel Nagin Journal: Resuscitation Date: 2019-02-27 Impact factor: 5.262
Authors: Mohammad M Ghassemi; Edilberto Amorim; Tuka Alhanai; Jong W Lee; Susan T Herman; Adithya Sivaraju; Nicolas Gaspard; Lawrence J Hirsch; Benjamin M Scirica; Siddharth Biswal; Valdery Moura Junior; Sydney S Cash; Emery N Brown; Roger G Mark; M Brandon Westover Journal: Crit Care Med Date: 2019-10 Impact factor: 7.598
Authors: Jonathan Elmer; Jon C Rittenberger; Patrick J Coppler; Francis X Guyette; Ankur A Doshi; Clifton W Callaway Journal: Resuscitation Date: 2016-09-17 Impact factor: 5.262
Authors: Jonathan Elmer; Cesar Torres; Tom P Aufderheide; Michael A Austin; Clifton W Callaway; Eyal Golan; Heather Herren; Jamie Jasti; Peter J Kudenchuk; Damon C Scales; Dion Stub; Derek K Richardson; Dana M Zive Journal: Resuscitation Date: 2016-02-03 Impact factor: 5.262
Authors: Claudio Sandroni; Alain Cariou; Fabio Cavallaro; Tobias Cronberg; Hans Friberg; Cornelia Hoedemaekers; Janneke Horn; Jerry P Nolan; Andrea O Rossetti; Jasmeet Soar Journal: Intensive Care Med Date: 2014-11-15 Impact factor: 17.440
Authors: Jonathan Elmer; Patrick J Coppler; Bobby L Jones; Daniel S Nagin; Clifton W Callaway Journal: Neurology Date: 2022-07-05 Impact factor: 11.800
Authors: Mark C Hwang; MinJae Lee; Lianne S Gensler; Matthew A Brown; Amirali Tahanan; Mohammad H Rahbar; Theresa Hunter; Mingyan Shan; Mariko L Ishimori; John D Reveille; Michael H Weisman; Thomas J Learch Journal: Rheumatology (Oxford) Date: 2022-05-05 Impact factor: 7.046