Skip to main content
. 2022 Aug 17;9:927642. doi: 10.3389/fcvm.2022.927642

FIGURE 7.

FIGURE 7

Subgroup meta-analysis by study type. Both the random-effects model (D + L) and fixed-effects model (I–V) have been presented in the Forrest plot. Significant heterogeneity existed in studies reported in the abstract, while there was no obvious heterogeneity in a subgroup analysis of original article studies. Therefore, the fixed-effects model should be performed in a meta-analysis of post-operative PPI in three original articles, and the random-effects model should be used in subgroup analysis in abstract studies. However, the COP group had a significantly lower risk of PPI in either subgroup meta-analysis.