24 vs. 72 hours of hypothermia for pediatric cardiac arrest: A pilot, randomized controlled trial

Abstract

Aim: Children surviving cardiac arrest (CA) lack proven neuroprotective therapies. The role of biomarkers in assessing response to interventions is unknown. We hypothesized that 72 versus 24 h of hypothermia (HT) would produce more favorable biomarker profiles after pediatric CA.

Methods: This single center pilot randomized trial tested HT (33 ± 1 °C) for 24 vs. 72 h in 34 children with CA. Children comatose after return of circulation aged 1 week to 17 years and treated with HT by their physician were eligible. Serum was collected twice daily on days 1-4 and once on day 7. Mortality was assessed at 6 months.

Results: Patient characteristics, baseline biomarker concentrations, and adverse events were similar between groups. Eight (47%) and 4 (24%) children died in the 24 h and 72 h groups, p = .3. Serum neuron specific enolase (NSE) concentration was increased in the 24 vs. 72 h group at 84 h-96 h (median [interquartile range] 47.7 [3.9, 79.9] vs. 1.4 [0.0, 11.1] ng/ml, p = .02) and on day 7 (18.2 [3.2, 74.0] vs. 2.6 [0.0, 12.8] ng/ml, p = .047). Serum S100b was increased in the 24 h vs. 72 h group at 12 h-24 h, 36 h-84 h, and on day 7, all p < 0.05. HT duration was associated with S100b (but not NSE or MBP) concentration on day 7 in multivariate analyses.

Conclusion: Serum biomarkers show promise as theragnostic tools in pediatric CA. Our biomarker and safety data also suggest that 72 h duration after pediatric CA warrants additional exploration.

Keywords: Biomarker; Cardiac arrest; Critical care; Neurologic disorders; Outcome; Pediatric.

Figure 1

CONSORT diagram for enrollment and outcomes.

Figure 2

Patient hourly temperature curves by randomized group. Graph begins once patient has reached the target temperature (32-34°C). Data are presented as mean ± standard deviation.

Figure 3

a-c. Serum NSE, S100b, and MBP biomarker concentrations by 12 hour intervals post-ROSC and randomized group. *p < 0.05, randomized group difference, P-values for continuous variables are based on Wilcoxon rank tests. Data presented as median (95% confidence interval).

Figure 3

a-c. Serum NSE, S100b, and MBP biomarker concentrations by 12 hour intervals post-ROSC and randomized group. *p < 0.05, randomized group difference, P-values for continuous variables are based on Wilcoxon rank tests. Data presented as median (95% confidence interval).