A Central Venous Line Protocol Decreases Bloodstream Infections and Length of Stay in a Trauma Intensive Care Unit Population

Abstract

We evaluated the benefit of a central venous line (CVL) protocol on bloodstream infections (BSIs) and outcome in a trauma intensive care unit (ICU) population. We prospectively compared three groups: Group 1 (January 2003 to June 2004) preprotocol; Group 2 (July 2004 to June 2005) after the start of the protocol that included minimizing CVL use and strict universal precautions; and Group 3 (July 2005 to December 2006) after the addition of a line supply cart and nursing checklist. There were 1622 trauma patients admitted to the trauma ICU during the study period of whom 542 had a CVL. Group 3 had a higher Injury Severity Score (ISS) compared with both Groups 2 and 1 (28.3 ± 13.0 vs 23.5 ± 11.7 vs 22.8 ± 12.0, P = 0.0002) but had a lower BSI rate/1000 line days (Group 1:16.5; Group 2:15.0; Group 3: 7.7). Adjusting for ISS group, three had shorter ICU length of stay (LOS) compared with Group 1 (12.11 ± 1.46 vs 18.16 ± 1.51, P = 0.01). Logistic regression showed ISS (P = 0.04; OR, 1.025; CI, 1.001-1.050) and a lack of CVL protocol (P = 0.01; OR, 0.31; CI, 0.13-0.76) to be independent predictors of BSI. CVL protocols decrease both BSI and LOS in trauma patients. Strict enforcement by a nurse preserves the integrity of the protocol.