Early neurologic examination is not reliable for prognostication in post–cardiac arrest patients who undergo therapeutic hypothermia

Abstract

BackgroundRecent advances in post–cardiac arrest (CA) care including therapeutic hypothermia (TH) have improved survival and favorable neurologic outcomes for survivors of CA. Survivors often present with deep coma and lack of brainstem reflexes, which are generally associated with adverse outcomes in many disease processes. Little is known regarding the role of initial emergency department (ED) neurological examination and its potential for prognostication. ObjectivesThe purpose of this study is to determine if components of a standardized neurologic examination are reliable prognosticators in patients recently resuscitated from CA. We hypothesize that lack of neurologic function does not reliably predict an adverse outcome and, therefore, should not be used to determine eligibility for TH. MethodsA standardized neurologic examination was performed in the ED on a prospective, convenience cohort of post-CA patients presenting to a CA resuscitation center who would undergo a comprehensive postarrest care pathway that included TH. Data such as prior sedation or active neuromuscular blockade were documented to evaluate for the presence of possible confounders. Examination findings were then compared with hospital survival and neurologic outcome at discharge as defined by the cerebral performance category (CPC) score as documented in the institutional TH registry. ResultsForty-nine subjects were enrolled, most of whom presented comatose with a Glasgow Coma Scale of 3 (n=41, 83.7%). Nineteen subjects (38.8%) had absence of all examination findings, of which 4 of 19 (21.1%) survived to hospital discharge. Of those with at least 1 positive examination finding, 13 of 30 subjects (43.3%) survived to hospital discharge. Subgroup analysis showed that 9 of the 19 patients with absence of brainstem reflexes did not have evidence of active neuromuscular blockade at the time of the examination; 2 of 9 (22.1%) survived to hospital discharge. Eight of these subjects in this group had not received any prior sedation; 1 of 8 (12.5%) survived to hospital discharge. Only 1 of the 17 subjects who survived was discharged with poor neurologic function with a CPC score=3, whereas all others who survived had good neurologic function, CPC score=1. ConclusionIn this cohort of patients treated in a comprehensive postarrest care pathway that included TH, absence of neurologic function on initial ED presentation was not reliable for prognostication. Given these findings, clinicians should refrain from using the initial ED neurological examination to guide the aggressiveness of care or in counseling of family members regarding anticipated outcome.