Strengths and limitations
The results of this network meta-analysis must be interpreted in light
of its limitations. Several important limitations arise from the use of
observational studies in a meta-analysis. The presence of unmeasured
confounders and possible treatment allocation bias cannot be excluded.
Ideally, a network meta-analysis is performed solely with the inclusion
of randomized populations to guarantee that the relative treatment
effects deriving from the network analysis remain unbiased21. However, the randomized evidence regarding the
topic of hypothermic strategies in aortic arch surgery is scarce,
leading to a relatively large amount of non-randomized studies in our
network meta-analysis. Hence, incorporating both types of data allows
assessments of larger sample sizes and multiple treatments
simultaneously. Furthermore, there may be variability in surgeon and
center expertise, technical variabilities and postoperative protocols.
Additionally, heterogeneity may arise due to different definitions for
stroke and postoperative AKI.