How Do I Diagnose and Manage Catheter-Related Bloodstream Infections?

Incidence

Annually in the United States alone, more than 5 million central venous catheters (CVCs) are inserted, and patients are exposed to more than 15 million catheter days in the intensive care unit (ICU). Approximately 250,000 bloodstream infections are reported in hospitals, 80,000 of which are in critical care units. The reported incidence density in the literature is highly variable; in a review by Maki, the reported variance was from 0.1 to 2.7 cases/1000 catheter days. Although some studies have questioned whether catheter-related bloodstream infections (CRBSIs) are associated with mortality, others have reported up to 25% directly attributable deaths. Length of hospital stay and health-care costs are significantly increased by an episode of CRBSI. As reported by Shah, on average, affected patients will stay an extra 10 to 20 days in the hospital, with their health-care costs increasing by an additional $4000 to $56,000 per episode of CRBSI. Overall, it is estimated that CRBSI accounts for 11% of health-care–associated infections (HAIs) in the United States. Given the significant impact on patient outcomes, reduction in the incidence of CRBSI has become a priority for health-care providers. However, it is possible to achieve dramatic reductions when institutions have introduced educational programs and policies focused on minimizing their incidence.

Diagnosis

The diagnosis of CRBSI may be established by criteria where the catheter is left in situ and separate criteria where the catheter has been removed. In the clinical context, where a patient manifests signs of sepsis, a CVC is in situ and there is no other focus of sepsis identified, then the likelihood of the catheter being the source is increased. Where the catheter is not removed, the quantitative method to establish the diagnosis is more than 100 colony-forming units (CFUs)/mL of blood drawn through the CVC. It is recommended to pair the sample obtained from the CVC with peripherally obtained cultures. Confirmatory evidence of line infection is indicated when the same species is identified in both CVC and peripherally drawn samples, there is a differential time to positivity of more than 2 hours for CVC drawn samples, and the culture yield is more than fivefold higher for the blood obtained through the catheter. The criteria for diagnosis where the catheter has been removed are established by a positive culture of a catheter segment; this may be semiquantitative (>15 CFU) or quantitative (>1000 CFU).

Pathogenesis

Infection of an in-dwelling catheter occurs by a number of mechanisms. Organisms that have colonized the patient’s skin may track along the catheter path and infect the catheter tip. This is the most likely portal in the short-term (<10 days in situ). Infection of the catheter hub tips may also occur as a result of handling by health-care personnel; this appears to be the leading etiology when the catheter has been in situ for a prolonged period. Rarely, CRBSI may result because of hematogenous seeding of the catheter from a remote source of sepsis, such as pneumonia. Finally, contaminated infusions have been implicated in rare instances.

Organisms

Epidemiologic data on organisms frequently identified are compiled in the United States by the National Healthcare Safety Network (NHSN) of the Centers for Disease Control and Prevention (CDC). The most common isolates remain coagulase-negative staphylococci (31%), but Staphylococcus aureus (20%) and enterococci (9%) are also frequent isolates. Fungi have become more prevalent, and Candida species are increasingly implicated as the pathogen involved (currently 9%). Gram-negative organisms now appear to account for approximately 20% of cases, with Escherichia coli and Klebsiella subspecies accounting for 6% and 5%, respectively. Antimicrobial resistance is also increasing; cases of gram-negative organisms resistant to third-generation cephalosporins and carbapenems are becoming more prevalent. This is also the case with Candida infections where fluconazole resistance is becoming more common. However, methicillin-resistant Staphylococcus aureus (MRSA) infections appear to be decreasing. However, isolation of Staphylococcus infections should prompt a thorough evaluation for endocarditis, including echocardiography.

Risk Factors

Several risk factors have been identified, including the following:

1.

Inexperience of the physician performing the procedure.
2.

Failure to adhere to maximum sterile barrier precautions; this requires thorough antisepsis of the insertion site, all health-care–associated personnel in the vicinity to wear protective clothing, and draping of the patient’s whole body with sterile covers.
3.

Density of flora on the patient’s skin surface.
4.

Duration of catheter insertion; the risk is magnified fourfold when the catheter is in situ more than 7 days and fivefold when the catheter is in more than 15 days.
5.

In the ICU, a high nurse/patient ratio.
6.

Patient factors, such as immune status, nutritional state, steroid therapy, and coincidental sepsis.

Other considerations include the antiseptic solution used, the material used to manufacture the catheter, and the pathogenicity of the infecting organism. Chlorhexidine (at least 0.5% in alcohol; ideally 2%) has become the standard antiseptic solution; data from studies suggest that its use may be associated with a 1.6% decrease in CRBSI and 0.23% mortality improvement, with associated cost benefits. However, on repeated exposure, patients may become sensitized, and rarely, severe reactions, including anaphylaxis, have been described. Povidone-iodine and 70% alcohol are acceptable alternatives in these cases. In relation to catheter material, devices manufactured with polyvinyl chloride or polyethylene appear to have a higher rate of colonization and CRBSI than those manufactured with polytetrafluoroethylene or polyurethane because they may be intrinsically less resistant to biofilm formation.

Prevention

Prevention of HAI has become a priority for health-care providers. In relation to CRBSI, Pronovost demonstrated in a large study of 103 ICUs (representing > 375,000 CVC days) a mean reduction in incidence density from 7.7 to 1.4/1000 catheter days at 18 months after the adoption of a specific protocol. The interventions were five straightforward practices: meticulous hand hygiene, sterile barrier precautions, chlorhexidine antisepsis, avoidance of the femoral site, and removal of the catheter when no longer clinically indicated ( Table 45-1 ). Comprehensive guidelines have been published to assist institutions in devising evidence-based programs to reduce their rate of CRBSI.

Table 45-1

Central Line “Bundle” Shown to Reduce CRBSI

1.

Handwashing
2.

Full-barrier precautions (during the insertion of CVC)
3.

Chlorhexidine (2%) to clean the skin (allow to dry before insertion)
4.

Avoiding the femoral site if possible
5.

Removing unnecessary catheters

CRBSI , catheter-related bloodstream infection; CVC , central venous catheter.

Meticulous attention to hygiene is a fundamental objective that must be achieved. Investment in education and training in performing CVC insertion is imperative and is required on an institution-wide basis given the prevalence of these infections in the non-ICU environment. Handling of the line after insertion demands strict observation of hand hygiene practices and correct management of infusates, lines, and dressings. Use of two-dimensional (2-D) ultrasound has not definitively demonstrated a reduction in the incidence of CRBSI; however, there is a reduction in technical complications (such as carotid arterial puncture), time to insertion, and possibly reduced colonization at the internal jugular site. Consequently, if the technology is unavailable, ultrasound-guided CVC insertion is recommended. Where possible, a nonsuture-based anchoring device should be used. In addition to a chlorhexidine-based antiseptic agent, use of a chlorhexidine-impregnated sponge device placed at the line insertion site has demonstrated efficacy at reducing CRBSI even where the baseline rate of infection was low. Similarly, daily cleansing of the catheter insertion site with a 2% chlorhexidine wash is also beneficial.

The subclavian vein is the recommended site for routine line placement, for example, for total parenteral nutrition (TPN). This is, however, controversial. Studies have shown that, compared with the subclavian site, there are higher rates of colonization of both the internal jugular and the femoral sites (in particular with obese patients). Surprisingly, this does not appear to translate to higher rates of infection. The subclavian site is associated with a higher rate of complications—inadvertent arterial puncture or pneumothorax—and is technically more difficult to perform with ultrasound guidance. Therefore, while the subclavian site may appear to be the preferred option, the clinician must consider other factors when deciding on the site for line placement, such as respiratory reserve in the event of pneumothorax or coagulopathy. It is clear, however, that attempting to reduce the burden of skin colonization with prophylactic antibiotics does not reduce the incidence of CRBSI, and their use for this purpose is not indicated. Similarly, although there appears to be a linear relationship between the duration of line insertion and the CRBSI, scheduled line removal and reinsertion expose the patient to the technical hazards of the procedures without the benefit of reducing the CRBSI rate. Instead, a transparent dressing should be applied enabling daily inspection of the site, with prompt removal of the line if any symptoms of sepsis develop in the absence of another focus.

Care of the management of infusion is critical. For nonblood- or nonlipid-containing preparations where the infusions are administered without interruption, it is recommended that the administration sets are changed between 4 and 7 days. For lipid-based preparations, this frequency is typically increased, and on average it is advised to change these sets every 24 hours. Propofol administration sets should be changed every 6 to 12 hours.

As discussed previously, the material used to manufacture the catheter can influence the development of CRBSI. As a further development, catheters coated with antiseptics (typically chlorhexidine/silver sulfadiazine) or antibiotics (minocycline/rifampicin) have been developed. The current second generation of antiseptic-coated catheters differs from the earlier version by having triple the amount of antiseptic applied. They are also coated on internal and external lumina, as opposed to the eternal lumen only in the first-generation device. These second-generation antibiotic-coated catheters demonstrated superiority in reducing CRBSI when compared with the first-generation antiseptic-coated lines. A Cochrane Collaboration meta-analysis of studies comparing impregnated with “plain” central lines demonstrated outcome benefit in intensive care (relative risk [RR], 0.68; 95% confidence interval [CI], 0.59 to 0.78) but not in hematology or oncology units and not in long-term TPN patients. It is recommended that impregnated catheters be considered for use in instances where a catheter is expected to be left in situ for more than 5 days and the institutional CRBSI rate remains above an acceptable threshold despite instituting a comprehensive training and education program and adoption of best practices.

Author’s Recommendations

Only gold members can continue reading. Log In or Register to continue