We aimed to evaluate whether subcutaneous tunnelling in peripherally inserted central catheter (PICC) placement could reduce the occurrence of central-line-associated bloodstream infection (CLABSI).

We conducted an open-label, multi-centre, randomized, controlled trial in five tertiary hospitals. Adult hospitalized patients requiring a PICC were randomized in a one-to-one ratio to conventional (cPICC) or tunnelled PICC (tPICC) arms using a centralized web-based computer-generated stratified randomization. CLABSI rates between groups were compared in a modified intention-to-treat population. Safety including the incidence of exit-site infection or haemorrhage-associated catheter removal were also compared. This trial was registered with Clinical Research Information Service of Republic of Korea (KCT0005521).

From November 2020 to March 2023, 1324 participants were enrolled and randomly assigned to tPICC (N = 662) and cPICC (N = 662). This study was terminated early due to the cohort CLABSI rate being lower than estimated, therefore, the original sample size of 1694 would render the study underpowered to detect a difference in CLABSI rates. In the tPICC, CLABSI occurred in 13 of 651 participants over 11,071 catheter-days (1.2/1000 catheter-days), compared with 20 among 650 patients with cPICC over 11,141 catheter-days (1.8/1000 catheter-days, rate ratio 0.65, 95% confidence interval 0.30–1.38, P=0.30). The incidence of exit-site infection (29 tPICC, 36 cPICC, P=0.5) and haemorrhage-associated catheter removal (11 tPICC, 11 cPICC, P=0.99) did not show a difference between the two groups.

Due to insufficient sample size, this study could not demonstrate a statistically significant CLABSI risk reduction in the tPICC group compared with the cPICC group. Both groups had similar rates of exit site infection and bleeding.