Current issues of ACP Journal Club are published in Annals of Internal Medicine


Triple lumen central catheters reduced the need for vascular access without increased rates of infection

ACP J Club. 1993 Jan-Feb;118:11. doi:10.7326/ACPJC-1993-118-1-011

Source Citation

Farkas JC, Liu N, Bleriot JP, et al. Single- versus triple-lumen central catheter-related sepsis: a prospective randomized study in a critically ill population. Am J Med. 1992 Sep;93:277-82.



To compare single- and triple-lumen catheters for rates of catheter-related sepsis and need for peripheral vascular access in patients in an intensive care unit (ICU).


Randomized controlled trial with follow-up to ICU discharge.


8-bed medical and surgical ICU in France.


108 consecutive patients > 18 years, who needed central venous access for > 48 hours. 17 catheters were excluded because they could not be inserted (n = 2), they were removed without culture (n = 7), or patients were discharged from the ICU with catheters in place (n = 8). 91 patients (mean age 64.5 y) and their 129 catheters were analyzed.


Physicians chose either internal jugular or subclavian insertion sites, and patients were then randomized to either single-lumen (one 14-gauge channel) or triple-lumen catheters (one 16- and two 18-gauge channels). Using Seldinger techniques, catheters were inserted percutaneously without subsequent tunneling, covered with sterile occlusive adhesive transparent dressings, and checked every 48 hours. Catheters were removed for suspicion of catheter-related sepsis, uselessness of central venous access, catheterization > 21 days, or discharge from the ICU. Recatheterization or complementary peripheral vascular access was done as required.

Main outcome measures

Duration of insertion, infections based on clinical findings and catheter tip culture, and need for peripheral vascular access.

Main results

68 single-lumen (27 subclavian, 41 internal jugular) and 61 triple-lumen (23 subclavian, 38 internal jugular) catheters were used. Triple-lumen catheters were inserted for a shorter time than single-lumen catheters (9.1 vs 11.6 d, P = 0.04) because they were considered to be useless earlier (P = 0.02). The groups did not differ for catheter-related sepsis rates defined by clinical signs and qualitative tip culture or by quantitative tip culture, for time between insertion and sepsis, or for colonization rate. 37% of patients in the single-lumen group and 2% of patients in the triple-lumen group (P for the difference < 0.001) needed peripheral vascular access.


In an adult intensive care unit, the use of triple-lumen catheters was associated with decreased need for peripheral vascular access. The incidence of catheter-related sepsis did not differ for single- and triple-lumen catheters.

Source of funding: Not stated.

For article reprint: Dr. J.C. Farkas, Unité de Réanimation, Clinique de la Défense, 16 bd Emile Zola, 92000 Nanterre, France. FAX 33-1-46696591.


Since being introduced a decade and a half ago, multilumen central venous catheters have been increasingly used to allow several simultaneous fluid and drug infusions. In most ICUs single-lumen catheters are now used less than multilumen devices. Infection is the most frequent complication of central venous catheterization, and early reports implicated bacterial contamination through the catheter hub and lumen as the major cause. With the advent of multilumen devices, it was logical to conclude that the risk for infection might be proportional to the number of lumens.

Although some investigations have concluded that multilumen catheters pose a greater risk for infection than their single-lumen counterparts, most studies have shown no difference. Sample sizes of these studies are too small to detect clinically important differences in outcome, however. Discrepancies among some of these reports are probably because of differences in study design, such as the criteria used to define catheter-related sepsis, the frequency of catheter exchange or replacement, and, in particular, the likelihood of selection bias. The randomized trial by Farkas and colleagues rigorously addresses the issue of the effect of single compared with multilumen catheters. Offsetting this advantage is the significantly shorter duration of catheterization in patients with triple-lumen catheters compared with those who had single-lumen devices. Nevertheless, their study corroborates other studies that have shown that the risks for infection from single and multilumen catheters are comparable.

Occasionally, patients in the ICU who have a single-lumen central catheter will subsequently require a second central access, subjecting the patient to additional insertion-related risks. Thus, initial use of a multilumen device may decrease the risk for noninfectious complications in such patients. The study was not designed to investigate this problem.

The report by Farkas and colleagues supports the safety and use of multilumen catheters in critically ill patients who require central venous access. Obviously careful attention to other strategies to minimize device-related bacteremia are necessary.

James A. Kruse, MD
Wayne State UniversityDetroit, Michigan, USA