LAUSR

Out-of-hospital cardiac arrest (OHCA) in children continues to have a poor prognosis. In a recent study of 12 877 OHCAs in children,1 30-day survival was 9.1%, and survival with favorable neurologic outcome (cerebral performance category of 1 or 2) was 2.5%. After 20 and 30 minutes of emergency medical services–initiated cardiopulmonary resuscitation (CPR), survival was 2.4% and 0.66%, respectively, and survival with favorable neurologic outcome was 0.43% and 0.10%, respectively.1 In the multicenter randomized Therapeutic Hypothermia After Pediatric Cardiac Arrest (THAPCA) study of OHCA,2 1-year survival was 33%, and 1-year survival with favorable neurologic outcome (a Vineland Adaptive Behavior Scale score ≥70) was 16%. The THAPCA randomized clinical trial tested hypothermia vs normothermia treatment after return of spontaneous circulation and found no difference in outcome. Hypothermia can be added to the hundreds of neuroprotective interventions found to be beneficial in nonhuman animal experiments that have not translated to improved outcomes for human adults or children, suggesting that different paradigms of resuscitation research are necessary.3,4 The use of one such paradigm, outcomes research, is described by Topjian et al5 in this issue of JAMA Pediatrics; the authors use the THAPCA OHCA trial database to explore potentially modifiable predictors of adverse outcomes. In this post hoc secondary analysis of data on 292 children and adolescents, early hypotension (in the first 6 hours of targeted temperature management, occurring in 78 patients [26.7%]) had an independent statistically significant association with mortality by hospital discharge. The survival to hospital discharge was 25.6% (20 of 78 patients) among patients with early hypotension, and 43.5% (93 of 214 patients) among patients without early hypotension, for an odds ratio (adjusted for witnessed status, duration of CPR, and the presence of preexisting conditions) of 0.39 (95% CI, 0.20-0.74). The burden of early hypotension (the percentage of hourly recorded blood pressure measurements in the first 6 hours of targeted temperature management that indicated hypotension) was independently associated with mortality, with an adjusted odds ratio for survival of 0.80 (95% CI, 0.69-0.93) for each 10% increase in burden, suggesting a dose-response relationship. Later hypotension (in the first 72 hours of targeted temperature management) was not associated with outcome. This study is important for several reasons. First, early hypotension is a potentially modifiable risk factor for adverse outcome. Although patients with early hypotension were more likely to receive a vasoactive agent, 18 of these (23.1%) received no vasoactive agents. The doses of these agents and intravenous volume given were not reported. Hypotension likely could have been avoided and treated more aggressively, and this management procedure may in the future translate to better survival. Second, this study draws attention to the importance of obtaining data on the association of blood pressure with outcomes. Although keeping systolic blood pressure (SBP) above a threshold of the fifth percentile of population norms is often recommended, no trial data support this suggestion.6 This study is an important contribution because it suggests that targeting an SBP above the fifth percentile may be associated with improved outcomes after OHCA. Third, the study raises important questions that warrant further research. For example, would outcomes be even better if SBP was above the 10th or the 25th percentile? Would targeting the mean arterial pressure (MAP) in addition to or instead of the SBP be associated with better outcomes? The answers to these questions are not obvious. In adult patients with critical illness and hypotension who require vasopressor support, trial data suggest against the use of higher blood pressure targets (MAP, 75-85 mm Hg) compared with lower ones (MAP, 60-70 mm Hg).7 Several trials have found benefit from abnormal, more permissive resuscitation targets in the critically ill; these interventions include oxygen therapy, feeding, mechanical ventilation, blood transfusion, and insulin therapy.8 Even the evidence supporting fluid resuscitation remains conflicted owing to potential harms of fluid overload and the difficulty in predicting fluid responsiveness.9 The results of the Fluid Expansion as Supportive Therapy (FEAST) trial in African children with severe febrile illnesses10 revealed that fluid boluses improved perfusion by 1 hour but increased mortality, usually owing to cardiovascular collapse that peaked at 2 to 11 hours after the bolus. Fourth, the study used the most accurate data to define the normative fifth percentile of SBP.11 Different guidelines give different definitions for the fifth percentile of SBP.6 Pediatric advanced life support guidelines suggest calculating this as SBP = 70 mm Hg + (age)(2).12 Different formulas were suggested using actual data for the 50th and 95th percentile blood pressure, and assuming that blood pressure is normally distributed (ie, the difference between the 50th and 95th percentile is the same as the difference between fifth and 50th percentile) and that one-third of the cardiac Related article Opinion