An accumulating body of work indicates that suppression helps to control intrusive memories. It keeps the unwanted memory out of awareness and can eventually cause forgetting. The material used to study suppression, however, has largely been emotionally neutral and much simpler than memories that people typically would want to avoid. Küpper et al. (2014) introduced a procedure based on complex aversive memories that enhances the ecological validity of this research while maintaining high internal validity. In a close replication study, we first confirm that suppression impairs the recall of details and also of the gist of aversive scenes. In a meta-analysis, we then synthesize the evidence from the yet limited number of studies. Across all variants of the procedure, this analysis yielded a small to moderate overall effect for suppression-induced forgetting (hedge’s g = 0.30, 95% CI = [0.08 0.53]), though with a rather wide prediction interval ([-0.24 0.84]). This effect remained significant following adjustment for publication bias. Notably, the size of suppression-induced forgetting was larger when disregarding studies that changed the procedure for neuroimaging purposes in ways that arguably compromise their ability to observe the behavioral effect. Together, these findings corroborate that the examined procedure is a viable option for the more naturalistic study of memory suppression.