Intensity of Vascular Access Device Use in Critical Care and its Impact on Catheter Associated Bloodstream Infection Rates

Background: The presence of a central venous catheter (CVC) is a risk for hospital acquired bloodstream infection (HABSI) in critically ill patients.  The rate of HABSI attributable to CVCs is defined as catheter associated bloodstream infections (CLABSI) over CVC days.  This rate does not risk adjust for patients whose care requires multiple CVCs and an arterial catheter. 

Objective:

We performed a pilot study to evaluate the intensity of line use in our critical care units (ICUs) and its potential impact on CLABSI rates. 

Methods:

The number of CVCs and arterial lines were counted Monday though Friday for all patients admitted to one of 209 adult ICU beds at the Cleveland Clinic between June 15^th and August 5^th by 2 of the investigators. Six different ICU groups were observed: cardiovascular and thoracic surgery, cardiac, heart failure, medical, surgical, and neurosciences.  Line days were calculated and defined as traditional line days (one or more central venous catheter per patient = 1 line day), actual line days (1 line day for each central venous catheter present), and total vascular access days (actual line days plus arterial line days).  Surveillance for HABSI including CLABSIs as defined by NHSN criteria was ongoing during the observation period.   Infections are prospectively recorded in an infection control database.  Patient days were calculated based on investigator observations.  CLABSI rates and device utilization ratios were calculated by using all three line day denominators and were compared.

Results:

The observation cohort was comprised of 1,636 patients and 5,940 patient days.   Of the 1,636 patients, 428 were not observed to have a CVC.  247 (15%) were observed to have more than one CVC and these patients were responsible for 19% of the total patient days.  There were 4,666 traditional line days resulting in a device utilization ratio of 0.786.  There were 5,936 actual line days count resulting in a device utilization ratio of 1.272.  There were 3,853 arterial line days.  Total vascular access days were 9,789 resulting in a device utilization ratio of 1.648.

During the observation period there were 18 CLABSIs.  The CLABSI rate was 3.86 per 1,000 traditional device days, 3.03 per 1,000 actual line days, and 1.84 per total vascular access days (p= 0.03 RR0.47[0.25-0.93] for the comparison of traditional vs. total vascular access days).

Conclusions:

Our pilot study demonstrated that traditional methodology for tracking device utilization underestimated the intensity of line use in our ICUs.  Using total vascular access days as the denominator resulted in a significant difference in CLABSI rates as compared with NHSN definition.  Further work is needed to explore the relationship between intensity of lines and the risk of CLABSI and HABSI.