Fig. 3: Trial Sequential Analysis of all-cause mortality.

The Trial Sequential Analysis of all-cause mortality showed that the required information size to confirm or reject that adding cerebral NIRS monitoring reduced the relative risk of death by 20% was 14,509 participants, using a 7.2% mortality in the control group, an alpha of 2.5%, a beta of 10%, and the diversity of 0% found in the meta-analysis. The accrued information size was 1489 participants, which was compatible with a reduced risk of death by 84% or an increased risk of death by 263% (TSA-adjusted CI 0.16–3.63). Thus, the trial sequential monitoring boundaries for benefit, harm or futility were not crossed. DARIS diversity-adjusted required information size, Pc proportion in control group, RR relative risk.