EPIC study: Duration of line use predicts infection

Remove lines promptly when no longer needed

Time is the greatest predictor of infection when it comes to central lines, with risk proceeding inexorably upward as lines remain in place, an epidemiologist recently reported in Toronto at the Society for Healthcare Epidemiology of America (SHEA).

"We had a sharp rise in infection rate as the line remained in place," reported Bryan Simmons, MD, health care epidemiologist at Methodist Health Systems in Memphis, TN. "This was the overwhelming risk factor for infection."

Simmons presented some of the first data from the ongoing Evaluation of Processes and Indicators in Infection Control (EPIC) study, which is being sponsored in part by the Centers for Disease Control and Prevention.¹ The study was conducted in 50 intensive care units (ICUs) worldwide over a 13-month period. The ICUs represent 20 states and 13 other countries.

EPIC researchers are looking at patient and process factors that contribute to central-line-associated primary blood stream infections (BSIs), a costly infection in terms of both dollars and lives. Volunteers at each site assessed a random sample of line insertions. In all, 116 BSIs (3.9%) were observed in the 2,970 sampled patients. Univariate analysis showed significantly increased risk of infection with increasing line days. Patients with lines in for more than 13 days were at 14 times the risk of those with lines in for less than five days. The finding underscores the importance of removing central lines as soon as medically possible, he emphasized.

"In the first one or two days, there was actually very low risk," Simmons said. "But after that it is somewhat linear."

It is impossible to specify a cut-point — generalized for all patients — when central lines should be left or taken out, he said. Other risk factors for infection were male gender, subclavian insertion (compared to jugular), insertion in the study ICU (compared to in the OR), the use of the line for total parenteral nutrition, patient being post-transplant or post-surgery, low albumin levels, unusual occurrences at the time of insertion, insertion attempts at multiple sites, and lack of appropriate supplies at the time of insertion.

"We also looked at insertion practice," he said. "This information was obtained by questioning the insertor within 24 hours."

Reference

1. Simmons B, Kritchevsky S, Braun B, et al. Risk factors for central line associated BSIs in 50 ICUs worldwide: The EPIC study. Abstract 103. Presented at the Society for Healthcare Epidemiology of America. Toronto; April 2001. n