RESUMO
Peripherally inserted central venous catheters (PICCs) have a potential advantage in preventing central line-associated bloodstream infection (CLABSI) compared with the centrally inserted ones (CICCs). However, due to a limited number of studies with insufficient statistical evaluation, the superiority of PICCs is difficult to be generalized in adult hematology unit. We conducted a single-center retrospective study and compared the risk of CLABSI between 472 CICCs and 557 PICCs inserted in adult patients with hematological disorders through conventional multivariate models and a propensity score-adjusted analysis. The overall CLABSI incidence in CICCs and PICCs was 5.11 and 3.29 per 1000 catheter days (P = 0.024). The multivariate Cox regression analysis (hazard ratio [HR]: 0.48; 95% confidence interval [CI]: 0.31-0.75; P = 0.001) and Fine-Gray subdistribution analysis (HR: 0.59; 95% CI: 0.37-0.93; P = 0.023) demonstrated that PICC was independently associated with a reduced risk of CLABSI. Moreover, the stabilized inverse probability of treatment weighting analysis, which further reduced the selection bias between CICCs and PICCs, showed that PICCs significantly prevented CLABSI (HR: 0.58; 95% CI: 0.35-0.94; P = 0.029). Microbiologically, PICCs showed a significant decrease in gram-positive cocci (P = 0.001) and an increase in gram-positive bacilli (P = 0.002) because of a remarkable reduction in Staphylococci and increase in Corynebacterium species responsible for CLABSI. Our study confirmed that PICC was a superior alternative to CICC in preventing CLABSI in the adult hematology unit, while it posed a microbiological shift in local epidemiology.