Introduction:
Direct paired meta-analyses and network meta-analysis were conducted to compare the incidence of Catheter-associated bloodstream infections (CRBSIs) in different types of central venous access devices (CVADs).

Material and methods:
The PubMed, EMBASE, Web of Science, Cochrane databases, CNKI and CBM were systematically searched from inception to May 31, 2024 for randomized controlled trials (RCTs) comparing the incidence of CRBSIs across various types of CVADs. Literature screening, data extraction, and risk bias evaluation were all independently conducted by two individuals. Direct paired meta-analyses and network me-ta-analysis were performed using RevMan5.3 and Stata14.0 software, respectively.

Results:
A total of five literatures were included. Paired meta-analyses revealed that the incidence of CRBSIs was lower in the peripherally inserted central catheters (PICCs) group compared to the central venous catheters(CVCs) group(RR=0.23, 95%CI(0.13-0.43), P<0.00001). The incidence of CRBSIs in PICCs group was observed to be lower compared to that in totally implantable venous access ports(TIVAPs) group (RR=0.45, 95%CI(0.23-0.87), P=0.02). Descriptive analysis revealed a higher incidence of CRBSIs in CVCs group compared to the TIVAP group (RR=2.97, 95%CI(1.65-5.17), P=0.0002). The network meta-analysis revealed a significantly lower incidence of CRBSIs in the PICCs group compared to the CVCs group. However, no statistically significant differences were observed in other comparisons. Based on the cumulative ranking curve test, the incidence of CRBSIs in various CVADs was ranked as follows: PICCs(97.20%)> TIVAPs(50.00%)>CVCs(2.80%).

Conclusions:
The available evidence suggests that PICCs exhibit the lowest incidence of CVADs, followed by TIVAPs. Therefore, PICCs should be prioritized when selecting CVADs.