There is a persistent debate on the optimal target temperature to use during cooling procedures in cardiac arrest survivors. A large randomized clinical trial (RCT) including more than 900 patients showed that targeted temperature management (TTM) at 33 °C had similar mortality and unfavorable neurological outcome (UO) rates as TTM at 36 °C in out-of-hospital cardiac arrest patients with any initial rhythm. Since then, several observational studies have been published on the effects of changes in target temperature (i.e., from 33 to 36 °C) on patients' outcome. We performed a systematic literature search from 1 January 2014 to 4 December 2020 and identified nine retrospective studies (very low levels of certainty; high risk of bias), including 3799 patients, that evaluated TTM at 33 °C vs. TTM at 36 °C on the occurrence of UO (n = seven studies) and mortality (n = nine studies). TTM at 33 °C was associated with a lower risk of UO when studies assessing neurological outcome with the Cerebral Performance Categories were analyzed (OR 0.80 [95% CIs 0.65-0.99]; p = 0.04). No differences in mortality were observed within the two TTM strategies. These results suggest that an inappropriate translation of TTM protocols from large well-conducted randomized trials into clinical management may result in unexpected effects on patients' outcome. As for all newly commercialized drugs, epidemiological studies and surveillance programs with an adequate follow-up on large databases are necessary to understand how RCTs are implemented into medical practice.
Keywords: 33 °C; 36 °C; cardiac arrest; dose; outcome; targeted temperature management.