Positive blood cultures in a patient who has an indwelling vascular catheter and who has no other source of infection raise the possibility of catheter-associated infection. A variety of approaches have been devised to help differentiate whether a positive blood culture represents catheter-associated infection or has arisen from another source. Quantitative cultures of blood taken simultaneously from the catheter and from peripheral blood that demonstrate a difference of more than threefold microorganisms from the catheter are probably the most accurate method to determine if catheter-associated infection is present without removing the catheter [1,2]. However, few, if any, clinical laboratories routinely perform quantitative blood cultures.

Differential time to positivity of blood cultures taken from a central line compared with those taken from a peripheral vein is another diagnostic method. Blood cultures obtained from an infected central catheter may turn positive a least 2 hours sooner than blood drawn simultaneously from a peripheral vein [2].

Another method that does not require the removal of the catheter involves culture of peripheral blood as well as the insertion site and hub. Growth of more than 15 colonies of the same organism from all three sites suggests short-term catheter-related infection [4,9].

It is important to minimize the possibility of contamination when obtaining blood for culture by having specifically trained personnel obtain the samples. Disinfection of the skin and hub using alcohol, tincture of iodine, or alcoholic chlorhexidine, but not povidone-iodine, is recommended. Blood samples taken from a peripheral vein are preferred as they are less likely to be contaminated than blood samples obtained from catheter hubs. However, all of the techniques listed above require sampling from the catheter as well as from a peripheral vein.

Although very helpful in the diagnosis of catheter-associated infection, culture of the catheter necessitates removal of the catheter before the diagnosis can be made. For optimum culture, the catheter tip, or the introducer tip for pulmonary artery catheters, should be cultured. Quantitative cultures obtained
by vortexing or sonicating the catheter tip most accurately determine the numbers of microorganisms present on both the internal and external surfaces of the catheter. However, this method is not practical in the clinical setting [2].

Rolling the distal segment of the catheter on an agar plate yields semiquantitative results that compare favorably with quantitative methods and has gained the greatest acceptance. The presence of ≥ 15 bacterial colonies on an agar plate correlates significantly with the presence of local inflammation and signs and symptoms of bloodstream infection. No similar cutoff has been established when yeasts are grown from the catheter tip. Some patients with catheter-associated infection will have fewer colonies, and catheter tips from asymptomatic patients will sometimes yield ≥ 15 colonies. A drawback of this technique is that only the external portion of the catheter is cultured, not the lumen, which may be the primary site of infection in long-term catheters. The roll-plate technique is the recommended method for diagnosis of presumed infection in short-term catheters after they have been removed [2,4].

Regardless of the type of catheter inserted, the major risk factor for the development of catheter-associated infection is the breach of a major host defense against infection—the skin. Catheter-associated infections are usually due to normal skin flora, particularly Gram-positive cocci, such as coagulase-negative staphylococci and Staphylococcus aureus. However, the distribution of microorganisms on the skin varies. For example, Gram-negative bacilli, Candida species, and anaerobes are increased in the groin area and on the lower extremities [10].

The ecology of normal human flora is further altered by illness, hospitalization, and the presence of foreign bodies. The use of antimicrobial agents inhibits the growth of normal flora and contributes to the emergence of resistant Gram-negative bacilli, S. aureus, vancomycin-resistant enterococci, and yeasts. Patients who have a productive cough or a tracheostomy can easily contaminate their skin with organisms from their respiratory tract. The hands of healthcare personnel may facilitate the transfer of potential pathogens from patient to patient [6].

The site of catheter insertion influences the risk of infection. Central venous catheters inserted in the internal jugular vein become infected more often than those in the subclavian vein, perhaps because of difficulties in dressing the area and contamination with respiratory secretions [5,11]. Catheter insertion in the lower extremities should be avoided in adults because of increased risk of phlebitis and infection in this area of poor blood flow [5,6]. Placement of femoral lines should be a last resort in emergent situations or when no other vascular access is available, and these lines should be removed as soon as possible [5].

Catheter-associated phlebitis and infection are more likely to occur when catheters are inserted by inexperienced personnel rather than personnel who are trained in these techniques. Prospective, randomized trials have shown that strict adherence to sterile technique (i.e., mask, cap, and large sterile drape, gloves, and gown) is beneficial in preventing central venous catheter infections, and also highly cost effective [5,6]. The importance of sterile techniques using maximal barrier precautions for short-term central catheters cannot be overemphasized and should become a part of house staff training [5,6]. Use of a catheter checklist to ensure and document adherence to infection prevention practices at the time of insertion is recommended [6,12].

Ultrasound guidance for the insertion of central vascular catheters, especially internal jugular catheters, has been shown in a meta-analysis to decrease the risk of mechanical complications associated with placement [5]. A biodegradable collagen cuff impregnated with silver ions is commercially available to attach to short-term central venous catheters before insertion. Two initial randomized controlled trials showed protection against catheter-associated colonization and bacteremia, but subsequent trials have failed to show a decrease in infection rates [5,8]. Currently, it is recommended that these cuffs not be used [5].

Several different antiseptics, 70% alcohol, chlorhexidine, and iodine-based solutions, have been found to reduce microbial contamination at the insertion site of the catheter [4,5]. Several studies and a meta-analysis have found that chlorhexidine-based aqueous or alcoholic solutions are superior to povidone-iodine solutions in reducing colonization at the catheter insertion site and catheter-associated bacteremia [5,13,14]. Current recommendations are to use 2% chlorhexidine gluconate for antisepsis of the insertion site, allowing it to dry before catheter insertion [5,6]. Chlorhexidine products have not been approved for children less than 2 months of age.

Antimicrobial ointments have been shown to increase the risk of infection with Candida and antibiotic-resistant bacteria and may affect the integrity of some catheters. With the exception of povidone-iodine ointment for some hemodialysis
catheters, routine use of ointments at the catheter insertion site is discouraged [5,6,15].

A chlorhexidine-impregnated patch (BIOPATCH, Ethicon, Somerville, NJ) has proved efficacious in reducing colonization at the catheter site, and in reducing bloodstream infections [8,16,17]. There currently are no firm recommendations as to whether this device should be routinely used with short-term central venous catheters [5]. However, a recent guideline suggests considering use of this patch when the rates of catheter-associated infection remain above target rates despite consistent use of evidence-based prevention bundles [6]. Use of these patches also should be considered in patients who have limited access and a history of recurrent catheter-associated infections and in those who have a heightened risk of severe sequelae if infection should occur, such as patients with recently implanted intravascular devices [6]. The use of systemic antibiotics as prophylaxis before the placement of central venous devices is strongly discouraged because selection for antibiotic-resistant microorganisms is highly likely [5,6].

These catheters are inserted into the subclavian vein, the internal jugular vein, and rarely the femoral vein. They can be single or multilumen, depending on the specific needs of each patient. Some studies, but not all, have shown that multilumen catheters are associated with a higher rate of colonization and infection than single-lumen catheters, particularly when used for an extended period of time [18,19,20,21,22].

Increased risk for infection, especially with multilumen catheters, occurs with the frequent manipulations that are required in the care of critically ill patients. In one study, only one of three lumens was used in most of the multilumen catheters that were inserted [21]. Therefore, it is recommended that a central venous catheter be chosen with the minimum number of ports or lumens required for the care of the patient [5]. It is recommended that the care of these multilumen lines be limited to a few well-trained personnel and that the catheter be changed to a single-lumen catheter if all the ports are no longer needed [5]. These catheters should be used predominantly in patients in the intensive care setting.

Numerous antimicrobial agents (tetracyclines, rifamycins, glycopeptides, β-lactams, micafungin), antiseptics (benzalkonium chloride, chlorhexidine, tridodecylmethylammonium chloride, iodine, gentian violet, and silver molecules), and antithrombotic agents (heparin, ethylenediaminetetraacetate [EDTA]) alone or in various combinations have been bound to polymer material or used to coat the surfaces of catheters in the hope of reducing colonization, thrombosis, and subsequent infection [2,8,23,24]. Heparin-coated catheters should not be used because of concerns for developing heparin-induced thrombocytopenia.

Minocycline/rifampin-coated catheters (Cook Medical, Bloomington, IN), chlorhexidine/silver sulfadiazine-coated catheters (Arrow International, Reading, PA), and a silver-platinum-carbon–impregnated catheter (Vantex CVC with Oligon, Edwards Life Sciences, Irvine, CA) are currently available. All of these catheters have been shown to reduce catheter-associated colonization [2,5,8,23,25]. In some controlled trials, the rates of catheter-associated bacteremia were sufficient to demonstrate significant reductions in infection rates when antimicrobial catheters were compared with standard catheters [2,5,8,26]. One study showed that a minocycline/rifampin-containing device was more effective at reducing both catheter colonization and catheter-associated bloodstream infections than a chlorhexidine/silver sulfadiazine-coated catheter [27]. However, that study was performed using the first-generation chlorhexidine/silver sulfadiazine catheter (ArrowGard) that was coated only on the external surface. Subsequent studies assessing the second-generation catheter that is coated on both external and internal surfaces (ArrowGard Plus) against the minocycline/rifampin catheter have not been performed.