• Background Infected central venous catheters cause morbidity and mortality. • Objective To compare the risk for colonization of central venous catheters used for total parenteral nutrition with that of catheters used for other purposes. • Methods Retrospective review of prospectively acquired data on 260 patients with a stay in a surgical intensive care unit longer than 3 days. Single-lumen catheters used solely for total parenteral nutrition were inserted into the subclavian vein and cared for by a dedicated team. Catheters for other purposes were placed and cared for by other staff. Catheters were cultured if clinical findings suggested infection. • Results Of 854 central venous catheters, 61 (7%) were used for total parenteral nutrition. During 4712 catheter days of observation, 89 catheters of all types were colonized. Risk factors for colonization included duration of catheterization (P < .001), having 3 or more lumens (hazard ratio, 1.7; 95% CI, 1.1–2.6), pulmonary artery catheterization (hazard ratio, 1.7; 95% CI, 1.1–2.7), and placement in the internal jugular vein (hazard ratio, 1.6; 95% CI, 1.1–2.5). Catheters used for total parenteral nutrition (hazard ratio, 0.14; 95% CI, 0.04–0.57) and those in the subclavian vein (hazard ratio, 0.51; 95% CI, 0.3–0.8) were at lower risk of colonization. In a multivariate Cox model, the only significant factor was a 5-fold lower risk of infection for catheters used for total parenteral nutrition (hazard ratio, 0.19; 95% CI, 0.04–0.83). • Conclusion Rates of colonization were lowest for catheters used solely for total parenteral nutrition, suggesting that a team approach improves patients’ care.