CLABSI rates and duration of central device


Intravenous literature: Yoshida, J., Ishimaru, T., Kikuchi, T., Matsubara, N. and Asano, I. (2011) Association between risk of bloodstream infection and duration of use of totally implantable access ports and central lines: A 24-month study. AJIC: American Journal of Infection Control. 39(7), p.e39-e43.


Background – Prolonged use of totally implantable access ports (APs) and central lines (CLs) has been known to carry a risk of bloodstream infection (BSI), but the safe cutoff day for discontinuing use remains unknown. We performed a receiver operating characteristic (ROC) curve analysis to determine this cutoff.

Methods – A retrospective 24-month study covered a total of 22,481 days of device use. For each day of use, the following findings were recorded: patient age and sex; presence or absence of diabetes mellitus, preexisting sepsis, and renal disease; and occurrence of device-associated BSI. BSI was defined in accordance with the Centers for Disease Control and Prevention’s definition of catheter-related infection.

Results – BSIs occurred in 81 patients with an AP, for a BSI rate of 2.81 cases per 1,000 days of use. Among the 896 patients with a CL, the BSI rate was 5.60 cases per 1,000 days of use. The ROC analysis found a cutoff time of 33 days for APs (median days of use, 48) and 10 days for CLs (median days of use, 20.5). For the total 22,481 days of use, the odds ratio between APs and CLs with respect to BSI was 0.556 (95% confidence interval [CI], 0.256-1.208; P = .138). Days of use beyond the cutoff had an odds ratio of 2.867 (95% CI, 1.823-4.507; P < .001). Among the risk factors, preexisting sepsis had an odds ratio of 7.843 (95% CI, 4.666-13.184; P < .001).

Conclusion – Use of an AP for more than 33 days and a CL for more than 10 days may carry an increased risk of device-associated BSI. These cutoff periods are longer than those expected at the time of device placement and indicate the importance of postplacement care.

Main page


Comments are closed.