Volume 1 Supplement 1

17th International Symposium on Intensive Care and Emergency Medicine

Open Access

Positive cultures of central venous catheter (CVC) in intensive care unit (ICU) patients: results from a prospective survey

  • B Souweine1,
  • O Traore1,
  • L Badrikian1,
  • L Bret1,
  • N Gazuy1 and
  • H Laveran1
Critical Care19971(Suppl 1):P037

DOI: 10.1186/cc43

Published: 1 March 1997

A 1-year prospective survey was conducted to determine the incidence rate of positive catheter culture (PCC) in ICU patients.


All consecutive CVC including dialysis catheters were studied. All dialysis catheters were single lumen CVC. The sites of CVC insertion, and the number of lumen of non-dialysis CVC were at the discretion of the attending physician. After CVC removal a quantitative culture of the CVC distal tip was performed. A 103 cfu/ml CVC quantitative culture threshold defined PCC. The risk factors studied for PCC were type of CVC, site of insertion, number of lumen, duration of catheterization, fever, and inflammatory local signs at the time of CVC removal.

The introducers for pulmonary artery catheters were studied separately.


During that time 420 CVC were placed in 173 patients. Among them, 153 were dialysis catheters. The sites of insertion were internal jugular vein (n = 250), femoral vein (n = 120), subclavian vein (n = 34), and axillary vein (n = 16). Twenty-nine CVC were tunneled included the 16 CVC axillary sites. The number of single lumen, double lumen, and triple lumen CVC were 170, 191, and 59, respectively. The median duration of catheterization was 5 days. The incidence rate of PCC was 13.7%. It corresponded to 2.2 PCC/100 CVC-day. The type of CVC, the site of insertion, the number of CVC lumen, and fever at the time of CVC removal are not associated with PCC (P = 0.15, P = 0.35, P = 0.73, and P = 0.57, respectively). The risk factors for PCC were duration of catheterization > 5 days (P = 0.017), and the presence of local inflammatory signs (P < 0.001). Erythema was the only inflammatory sign associated with PCC in 12/13 cases, and a CVC site infection defined as presence of pus occurred in one case. The PCC yielded coagulase negative staphylococci (n = 26), Enterobacter spp (n = 11), Staphylococcus aureus (n = 8), Pseudomonas aeruginosa (n = 7), and others (n = 9). Bacteremia related to CVC with PCC occurred in five cases (1.19%), none of them were Gram-negative bacteria.

Twenty-two introducers were placed for a median duration of 4 days. The sites of insertion were internal jugular vein (n = 15), subclavian vein (n = 7). None of the introducers were associated to local inflammatory signs. Only one introducer culture yielded ≥ 103 cfu/ml (coagulase negative Staphylococcus).


In this study Enterobacter spp emerged as the second PCC etiologic agent. However, none of the Enterobacter spp were causative agents of CVC-related bacteremia. Physicians should be aware of PCC risk in cases of local erythema, and consider removing CVC.

Authors’ Affiliations

Service de réanimation polyvalente, Hôpital G Montpied


© Current Science Ltd 1997