Objective: Catheter colonization is commonly used as an end point in studies on central venous catheter (CVC) infections. This study aimed at comparing the rates of catheter colonization in patients according to the administration of antibiotics at removal.
Design: Prospective study over a 5-year period. All patients with a CVC removed without suspicion of CVC infection were included.
Setting: An adult medical/surgical intensive care unit at a university teaching hospital.
Patients: A total of 472 patients with CVC inserted for a mean CVC duration of 6.0 +/- 3.7 days were included.
Methods: Antibiotics at removal was defined as antibiotics administered within 48[Symbol: see text]h before CVC removal, and CVC colonization as catheter tip culture yielding > or = 10(3) colony-forming units per milliliter.
Interventions: None.
Measurements and results: Of 472 patients, 302 (64%) were receiving antibiotics at removal. The number of CVC colonizations per number of CVCs inserted was 4.7% (22 of 472) and the number of CVC colonizations per 1000 CVC days was 8.0. Administration of antibiotics at removal was associated with a lower risk of CVC colonization. Irrespective of CVC duration, CVC colonization rate differed between patients with or without antibiotics at removal (Kaplan-Meier test, p=0.04).
Conclusions: The proportion of patients with antibiotics at time of removal should be taken into account when interpreting the results of surveys and trials using the quantitative CVC tip culture to define CVC colonization as an end point.