All Content

Fewer Catheter-Related Complications with Central Access in ICU Patients


Clinical question

What type of venous access -- central or peripheral -- is better for patients in the intensive care unit who have no absolute indication for central access?

Bottom line

As compared with central venous catheters (CVCs), the initial use of peripheral venous catheters (PVCs) in patients in the intensive care unit (ICU) leads to more complications, primarily related to the difficulty in inserting of these catheters. Furthermore, the majority of patients in the PVC group may eventually require a CVC because of an increase in the rate of venotoxic drug infusions or because they have difficulty maintaining the PVC. (LOE = 1b-)


Ricard J, Salomon L, Boyer A, et al. Central or peripheral catheters for initial venous access of ICU patients. Crit Care Med 2013;41(9):2108-2115.

Study design

Randomized controlled trial (nonblinded)

Funding source





Inpatient (ICU only)


Using concealed allocation, these investigators randomized ICU patients requiring venous access to receive either a CVC or PVC. Only patients without any absolute indications for CVCs were included. CVCs could be inserted into jugular, subclavian, or femoral sites at the discretion of the clinician. PVCs were 18-gauge or 20-gauge short catheters. Patients with a predefined increase in the rate of venotoxic drug infusions or those with difficulty maintaining a PVC were able to cross-over to the CVC group. More than half the patients in the PVC group ultimately received a CVC. Baseline characteristics were comparable between the 2 groups, with a mean age of 64 years, similar predicted mortality scores, and the majority of patients requiring mechanical ventilation. The primary outcome was the number of catheter-related complications defined as major mechanical, maintenance-related, infectious, or thrombotic complications. The determination of what constituted a major or minor complication was made a priori by the investigators. They also used a validated classification system of adverse events that rated the complications from grade 1 (minimal symptoms) to grade 5 (death). For example, one major mechanical CVC complication was the need to change insertion site, whereas a major mechanical PVC complication was needing more than 5 attempts to place a PVC. Overall, the PVC group had a greater number of major complications (133 vs 87; P = .02), the majority of which were PVC insertion difficulties (n = 56), erythema at insertion site (n = 20), and subcutaneous diffusion (n = 19). None were life threatening. When the complications were categorized using the grading classification, there were again more complications per patient in the PVC group (1.54 vs 0.89; P = .0001), mainly due to a larger number of grade 1 and grade 2 complications. Finally, more time was spent by doctors and nurses in managing catheters in the PVC group.

Dr. Kulkarni is an assistant professor of hospital medicine at Northwestern University in Chicago.

Next Article:

   Comments ()