Association of hypotension burden calculated from intermittently documented blood pressures and outcome after pediatric cardiac arrest.

Abstract

Hypotension burden measured using continuous high-resolution blood pressures (BPs) is associated with outcome following pediatric cardiac arrest. It is unclear whether this holds when using EHR-documented blood pressures, which are more available but measured differently. Our objective was to determine the association of hypotension burden calculated using EHR-documented BPs with outcome after pediatric cardiac arrest. Secondarily, we compared hypotension burden calculated using EHR-documented BPs vs. continuous high-resolution BPs. In this single-center retrospective study, we quantified hypotension burden using EHR-documented BPs for 24 hours following return of spontaneous circulation. We determined association of hypotension burden below the 10^th percentile for age with unfavorable outcome (Pediatric Cerebral Performance Category Score ≥3 with change from pre-arrest) using logistic regression controlling for age, sex, arrest location, and lactate. Hypotension burden using EHR-documented BPs vs. continuous high-resolution BPs was compared with Bland-Altman analysis. 139 patients were analyzed (median age 54 months, 60.7% male); 63% had unfavorable outcome. Using EHR-documented BPs, increasing burden of hypotension below 10^th percentile-for-age was associated with greater odds of unfavorable outcome (odds ratio 4.64, 95% CI [1.21,17.6], p=0.025). Hypotension burden was similar when calculated using EHR-documented BPs vs. continuous high-resolution BPs (0.03 [0,0.26] vs 0.04 [0,0.38] mmHg-hrs, p=0.165). There was no significant bias toward either measurement modality on Bland-Altman analysis (mean difference 0.22±0.95, p= 0.096). Hypotension burden calculated using EHR-documented BPs was associated with unfavorable outcome after pediatric cardiac arrest and closely approximated hypotension burden from continuous high-resolution BPs.