We searched MEDLINE, Embase, and electronic databases of Cochrane Central Register of Controlled Trials. We included randomized controlled trials comparing endoscopy and fluoroscopy placement of PNTs in critically ill patients. Two reviewers assessed the quality of each study and collected data independently. We performed the meta-analysis with Cochrane Collaboration RevMan 5.3.
Three randomized controlled trials involving 243 patients were included. There were no significant differences in the placement success rate (RR, 0.99; 95% CI, 0.93, 1.06; z = 0.20, P = .84,) or procedure time (standardized mean difference, − 0.08; 95% CI, − 6.93, 6.77; z = 0.02, P = .98) between the 2 groups. No severe complications (digestive tract hemorrhage, perforation, respiratory problems, hemodynamic instability, or death) were noted in the three studies. There was a slight difference in the incidence of minor complications (RR, 8.12; 95% CI, 1.07, 61.53; z = 2.03, P = .04) between the 2 groups.
Endoscopy and fluoroscopy placement of PNTs can be accurately and safely performed in critically ill patients. Endoscopy may be at least equally as safe as fluoroscopy for the placement of PNTs.