1368: Quantifying Hypotension Burden Using Intermittent Map Measurements After Pediatric Cardiac Arrest

Introduction: Hypotension burden below the 20th percentile-for-age calculated from high-resolution, continuous mean arterial pressure (MAP) data is associated with unfavorable outcome after pediatric cardiac arrest. Continuous MAP data are often unavailable, but intermittent MAPs from the electronic health record (EHR) are readily accessible. We aimed to determine if hypotension burden calculated from intermittent MAP was associated with unfavorable outcome and whether this method agreed with continuous MAP-derived burden. We hypothesized that greater intermittent hypotension burden would be associated with unfavorable outcome and show acceptable agreement with continuous burden by Bland-Altman analysis. Methods: In this single-center retrospective study of children who received post-arrest care with an invasive arterial line, we calculated hypotension burden (combining magnitude and duration of hypotension) below the 20th percentile-for-age during the first 24 hours post-arrest using 1) intermittent MAP documented in the EHR, derived from both the arterial line and oscillometry, and 2) high-resolution continuously sampled MAP. Unfavorable outcome was defined as Pediatric Cerebral Performance Category score ≥3 with an increase from baseline. We evaluated the association of intermittent burden with outcome using Mann-Whitney tests and univariable logistic regression. We assessed agreement between intermittent and continuous burden using Bland-Altman analysis. Results: We analyzed 139 patients (median age 4.5 IQR 0.9,12 years). 63% of patients had an unfavorable outcome. Intermittent MAPs were recorded every 30 10,60 minutes in the EHR. Hypotension burden below the 20th percentile-for-age using intermittent MAPs was greater for patients with unfavorable compared to favorable outcome (0.51 0.05,1.21 vs. 0.02 0,0.60, p=0.018). Greater intermittent burden was associated with increased odds of unfavorable outcome (OR 1.50, 95%CI 1.02-2.21, p=0.039). Bland-Altman analysis did not reveal significant systematic bias between measurement methods (mean difference 0.21±1.34 mmHg-hrs, p=0.07). Conclusions: Hypotension burden below the 20th percentile-for-age using EHR-derived intermittent MAP values was associated with unfavorable outcome and agreed with hypotension burden calculated with continuous data.