Retrospective analysis of the prognostic value of electroencephalography patterns obtained in pediatric in-hospital cardiac arrest survivors during three years

Objective: To test the hypothesis that electroencephalography has prognostic value in children after in-hospital cardiac arrest. Design: Single-center, retrospective chart review. Setting: Urban tertiary care children's hospital, January 2001 to July 2004. Patients: Thirty-four consecutive children were identified from a registry of all patients resuscitated for cardiac arrest. Inclusion criteria were age >44 wks postmenstrual age to 18 yrs, survival after in-hospital cardiac arrest >24 hrs, and undergoing electroencephalography within 7 days after the cardiac arrest. Interventions: None. Measurements and Main Results: Prearrest, event, cardiopulmonary resuscitation, and postresuscitation variables were collected. Neurologic outcome was assessed by Pediatric Cerebral Performance Category (PCPC). Prearrest PCPC was estimated from chart review. Change in PCPC >1 or death between prearrest and time of hospital discharge was defined as poor neurologic outcome. Experts blinded to PCPC outcomes scored electroencephalographs from 1 (normal) to 5 (isoelectric). Sensitivity/ specificity analysis and receiver operating characteristic curve were developed with each electroencephalography grade cutoff. Of 184 consecutive patients who had cardiac arrests in our registry, 107 survived >24 hrs, and 83 met study criteria. Thirty-four patients had electroencephalography within the first 7 days after the cardiac arrest. Among those, 16 (47%) patients had good neurologic outcome defined as no change in PCPC, and 18 (53%) died or had poor neurologic outcome. Univariate analysis showed that a higher electroencephalography score performed within 7 days was associated with poor neurologic outcome (p <.05). Positive predictive value of electroencephalography grade 4-5 for poor neurologic outcome was 90%, and negative predictive value of electroencephalography grade 1-2 for poor neurologic outcome was 91%. Conclusions: In a single-center consecutive case series, electroencephalography background patterns during the initial 7 days after in-hospital cardiac arrest were associated with neurologic outcome in children.