Infectious complications of peripherally inserted central catheters in the hospital setting

Background: Reported rates of bloodstream infections (BSIs) related to the use of peripherally inserted central venous catheters (PICCs) in the outpatient setting have ranged from 0.4 to 0.8/1000 catheter-days. Results of studies of PICC line-related infections in the hospital however are limited.

Study: A prospective study was conducted to determine the risk of PICC-related infections in hospitalized patients. The following data were collected: patient demographics, the team placing the PICC (intravenous [IV] team or interventional radiology), PICC line duration of use in the hospital, the type of PICC (single or double lumen), and the patient unit when the PICC was inserted (the intensive care unit [ICU] or regular ward). All microbiology tests were reviewed for positive blood cultures in conjunction with the patient's chart to determine the source of infection.

Results: Between November 2006 and June 2007, 642 PICCS were inserted resulting in 5705 catheter-days. Three hundred eighty-seven (60.3%) were inserted by the IV team and 255 (39.7%) by interventional radiology; 17.9% of the PICC lines were single lumen, and 82.1% were double lumen. In 251 cases (39.1%), blood cultures were done, and, of those, 69 (10.7%) were positive for bacteremia; of those, 13 had PICC-related BSI. Of 333 PICCs that were removed prior to discharge, 22 catheter tips were cultured, and, in 10 of these cases, tip cultures matched the catheter-related (CR)-BSI organism. Catheter utilization rate was 8.5 for no CR-BSIs versus 27.9 for CR-BSIs (P < .001) (median in-hospital rate was 6 days). The CR-BSI rate was 2.3/1000 catheter-days; the mean days-to-BSI, 27.9 ± 22.7 days; and the median, 23 days. The overall infection rate was 2.3 CR-BSIs/1000 catheter-days. Multivariant analysis of age, sex, type of line, team placing line, unit location of patients where the line was inserted, and the duration of use showed that only 2 factors were significant: duration of line use (days) and female sex. Females developed CR-BSIs (OR, 7.26; 95% CI, 1.28-41.2; P = .007) more frequently.

Conclusion: The hospital-wide PICC-related BSI rate was 2.3/1000 catheter-days. The findings that CR-BSIs were associated with duration of catheter use and sex suggest that the risk of CR-BSIs with PICCs is related to insertion and utilization. (Am J Infect Control 2008;35:000-0.)