Infection Control Policies

Preventable HAIs have become a national focus driven by advocates for patient safety, accreditation bodies, governmental agencies, insurance payers, and professional organizations. Effective prevention of HAIs requires a coordinated effort among infection control practitioners, the hospital epidemiologist, infectious diseases experts, microbiologists, and the ICU staff. All members of the ICU staff should become familiar with the hospital’s infection prevention and control policies to prevent, diagnose, and treat HAIs. These policies typically include contact isolation precautions for patients infected or colonized with certain antimicrobial-resistant organisms including methicillin-resistant Staphylococcus aureus (MRSA), vancomycin-resistant enterococci (VRE), and multidrug resistant gram-negative organisms in addition to patients infected with Clostridium difficile. Policies generally establish and standardize indications and basic practices for insertion, maintenance, monitoring, and discontinuation of catheters and other indwelling devices. Each institution should also standardize procedures for surgical antisepsis, instrument processing and sterilization, and timing and choice of perioperative antibiotic prophylaxis.

Policy development should be supplemented by HAI surveillance and direct observation of hand hygiene and other infection prevention practices. To be successful, infection prevention and control require a multidisciplinary collaboration to design and implement interventions in the ICU to decrease HAI rates.

The department of infection prevention and control should provide quantitative infection data to the ICU leadership team and staff so that HAIs can be investigated and appropriate prevention and quality improvement measures initiated.

Special Infection Risks in the Intensive Care Unit

Rates of HAIs are generally higher in ICUs because of higher utilization rates of indwelling devices and because of the severity of illness and complexity of ICU patients. The ongoing need for all indwelling devices should be assessed daily. Devices should be removed immediately when their presence is no longer indicated or in cases when the device is suspected to be the source of infection, such as in the case of a CRBSI or CAUTI.

Broad-spectrum empiric antibiotic use is also more common in ICUs (Chapter 18), predisposing patients to infection or superinfection with fungal organisms, multidrug-resistant organisms (MDROs), and Clostridium difficile. To minimize these risks, the need for antibiotics should be carefully evaluated prior to their initiation. If antibiotics are necessary, the narrowest effective antibiotic regimen should be selected based on most likely organisms. However, before initiating empirical antimicrobial therapy, all appropriate cultures based on suspected sources of infection should be obtained so that the antibiotic regimen can be tailored to the cultures and susceptibility results. The continuation of empiric antimicrobial therapy is an issue that should be addressed daily on rounds and should be justified by culture results and the patient’s clinical status (Chapter 18). Judicious antibiotic use should be viewed as a requisite for effective infection prevention and control.

Clinical Definitions and Surveillance Definitions

A CRBSI is an infection of an intravascular catheter or catheter site with associated bacteremia. As with other device-associated infections in the ICU, it is important to understand the difference between clinical definitions (those used by clinicians to make a diagnosis and guide management) and surveillance definitions (those used by infection prevention and control personnel to monitor and report HAIs). Each definition has different objectives and has different criteria designed to achieve these goals. Therefore, a particular case may meet the surveillance definition but may not meet the clinical definition or vice versa.

The Infectious Disease Society of America (IDSA) has published clinical definitions and management recommendations for CRBSI (see “Diagnosis” and “Management and Treatment”). The CDC’s National Healthcare Safety Network (NHSN) has developed the most widely used surveillance definitions. Unlike the IDSA definitions, the NHSN definition focuses only on infections caused by certain catheters (defined as central lines) and uses the term central line-associated bloodstream infection (CLABSI) to describe these infections. Central lines are catheters that terminate near the heart or in central blood vessels including aorta, pulmonary artery, superior vena cava, inferior vena cava, internal jugular, brachiocephalic, external iliac, common iliac, femoral, and, in neonates, umbilical vessels. Notably excluded from the CLABSI definition are extracorporeal membrane oxygenators (ECMOs), intraaortic balloon pumps (IABPs), femoral arterial catheters, and peripheral and midline intravenous (IV) catheters. On the other hand, the CRBSI definition includes all catheters. For the purpose of surveillance, CLABSI is defined by the presence of a catheter in one of the previously listed central blood vessels within 48 hours preceding the positive blood culture and the absence of another source to which the bacteremia can be attributed. For the diagnosis of CLABSI involving organisms that are commonly skin contaminants (e.g., coagulase-negative Staphylococcus), there must be two or more positive blood cultures plus systemic evidence of infection that cannot be attributed to another source.

Incidence Rates and Pathogens

The overall incidence of CLABSIs reported by NHSN is between 1.3 and 5.6/1000 catheter days (1 catheter day = 1 catheter in 1 patient for 1 day). Estimates are that the number of CLABSIs originating in ICUs has decreased by 58% from 2001 to 2009 at least in part because of an increased national focus on infection prevention and control. However, certain characteristics of ICU patients place them at higher risk (Box 14.1). Most CRBSIs are secondary to gram-positive organisms (see Table 14.1). Factors that increase the likelihood of gram-negative CRBSI include critical illness, neutropenia, prior antibiotic exposure, and femoral site. Factors that increase the likelihood of Candida CRBSI include parenteral nutrition, prolonged exposure to broad-spectrum antibiotics, hematologic malignancy, organ or bone marrow transplantation, femoral site, and presence of multiple sites of Candida colonization elsewhere in the patient (urine, respiratory, etc.).

As with other infections in the ICU, antibiotic resistance is a problem. Methicillin-resistant S. aureus (MRSA) represents over 50% of isolates of S. aureus in some regions. Vancomycin-resistant Enterococcus (VRE) is also becoming a more prevalent cause of CRBSI. Emergence of extended-spectrum beta-lactamases (ESBLs) and carbapenemases in gram-negative organisms has resulted in resistance to most available antibiotics. There is increasing resistance to fluconazole among non-albicans Candida species, and fluconazole resistance has also been reported in C. albicans.

Pathogenesis

Short-term catheters are most commonly contaminated by skin flora entering at the insertion site and migrating along the outside of the catheter. Contamination of catheter hubs or lumens by contact with hands, access devices, or contaminated fluids can also occur. Hematogenous spread of bacteria from other sites is a less common cause of CRBSI. Interactions between the catheter and infecting organisms facilitate adhesion to the catheter and production of a biofilm, which helps certain infecting organisms to evade host defenses.

Diagnosis

Except in cases when local inflammation is present around the catheter, the only signs of CRBSI are often fever or leukocytosis. Two sets of blood cultures should be obtained when signs and symptoms of infection are present, and positive blood cultures in the absence of another potential site of infection should raise the question of CRBSI. Blood cultures should be drawn before initiation of antimicrobial therapy, if possible, and from two peripheral sites (or one peripheral and one at the time of placing a new central line aseptically). If one sample is drawn from an existing central venous catheter, at least one other set should be obtained from a peripheral vein. If quantitative blood cultures are available, comparison of cultures drawn through the lumen of the suspected catheter to cultures from a peripheral site may aid in diagnosis. The threshold for diagnosis of CRBSI using this method is threefold or greater colony-forming units (CFU) in the blood culture drawn from the catheter compared to the blood culture drawn from another site. Automated blood culture systems, which continuously monitor growth and time to positivity, are commonly available. These systems allow for calculation of the difference in time between when peripheral cultures and when catheter cultures become positive, or differential time to positivity (DTP). The threshold for diagnosis of CRBSI by this method is positivity of the blood culture drawn through the catheter at least 120 minutes before the blood culture drawn from the other site.

If the catheter is removed due to suspected CRBSI, the catheter tip may be cultured as well. Appropriate catheter cultures include semiquantitative and quantitative catheter cultures. Semiquantitative cultures are performed in the laboratory by rolling 5 cm of the catheter tip on a culture plate. The threshold for diagnosis of CRBSI by this method is > 15 CFU. Quantitative cultures are obtained by sonication and vortexing of the tip of the catheter to dislodge surface and intraluminal material into the culture medium. The threshold for diagnosis of CRBSI by this method is > 10² CFU. A positive culture of the catheter tip or of blood drawn through a catheter alone does not define a CRBSI. The diagnosis of CRBSI requires that signs of systemic infection are present and that blood cultures drawn from another site are positive with the same organism.

Meticulous aseptic technique is required when collecting all specimens. Skin and catheter access points must be prepped with 2% chlorhexidine (CHD), not povidone-iodine. Blood cultures should have the same volume of blood in all specimens in order to compare time to positivity or quantity of organisms.

Prevention

Efforts at prevention of CLABSI have yielded impressive and sustained results, including reports from many ICUs of no CLABSIs for extended periods of time. One simple yet effective means of preventing CRBSI is becoming familiar with appropriate indications for central venous catheters, placing those catheters only when appropriate indications are present, and removing catheters when those indications are no longer present. Steps should also be taken to prevent contamination at the time of insertion, which are often implemented in the form of a bundle, such as that advocated by the Institute for Healthcare Improvement (IHI) (Box 14.2). Emergently placed central lines (i.e., not placed using these techniques) should be removed within 48 hours. Catheter exchanges over a guide wire are discouraged except in rare cases because of high infection rates. A chlorhexidine (CHD)-impregnated sponge applied at the time of placement can further decrease infection rates.

Other successful interventions include the establishment of a culture of safety in the ICU, development of checklists to help achieve full compliance with prevention techniques, teaching and monitoring of competencies, feedback to practitioners of the data collected regarding outcomes, and support from administration to resource these efforts.

Hubs, needleless connectors, and injection ports should be disinfected before accessing the catheter. Transparent dressings should be used to cover the catheter exit site and should be changed every 5 to 7 days. Wet or soiled dressings should be replaced promptly. The catheter site should be inspected daily and the catheter removed if there are signs of inflammation. The continuing need for the catheter should also be reviewed daily and unnecessary catheters should be removed promptly.

If CLABSI rates remain high despite these interventions, other strategies can be used, including daily CHD bathing of high-risk patients and antibiotic or antiseptic-impregnated catheters and caps.

Intravenous (IV) administration sets should be changed daily if blood products or fat emulsions are given and changed every 6 to 12 hours if propofol is infused. Otherwise IV administration sets should be changed every 72 to 96 hours. Needleless systems should not include positive pressure mechanical valves. Properly placed and maintained CVCs do not need to be routinely replaced at regular intervals.

Although there are few data showing the development of resistant organisms when antibiotic-coated devices are used, there is still concern about antibiotic resistance arising over longer periods of time with widespread usage.

Another preventative technique is antibiotic-lock therapy (ALT), which consists of allowing high concentration of an antibiotic or other antimicrobial agent to reside inside the catheter in between usage. Some studies have shown benefit in preventing CRBSIs in patients with long term indwelling catheters, such as dialysis catheters, with the use of particular regimens. Further study of ALT to prevent CLABSI needs to be performed before this practice can be recommended.

Hemodialysis (HD) catheters are associated with higher infection rates than dialysis access by fistula or graft. If an HD catheter is needed for more than 3 weeks, infection rates can be decreased by the use of a tunneled catheter.

Arterial lines in adults should be placed in the radial, brachial, or dorsalis pedis artery rather than femoral or axillary artery. A minimum of sterile gloves, mask, cap, and a small fenestrated drape are adequate precautions during insertion, except for femoral or axillary locations for which full barrier precautions are required. Transducers and other components of the system (other than the catheter) should be replaced at 96-hour intervals. Disposable transducers are preferred over reusable transducers for infection prevention. Dextrose containing flush solutions should be avoided. A closed flush system should be used, and strict aseptic technique should be maintained when accessing the system.

Management and Treatment

The catheter should be removed if there are local signs of infection or inflammation. If the patient has hypotension, hypoperfusion, or organ failure, and there is no other obvious source of sepsis, the catheter should also be removed immediately. In some patients (e.g., clinically stable with difficult or dangerous to access blood vessels and without a clearly infected catheter), it may be reasonable to manage the patient without immediate catheter removal. In such cases, it may be reasonable to obtain blood culture through the catheter and a peripheral site and to remove the catheter if the DTP is > 120 minutes or if the patient does not improve. After appropriate cultures have been obtained, empiric antibiotics should be started and should include coverage for methicillin-resistant S. aureus (MRSA) and gram-negative rods (GNRs). The choice of antibiotic coverage for GNRs should be based on local antibiotic susceptibility patterns and prior cultures from the patient (Chapter 18). Antipseudomonal coverage should be initiated in all patients with neutropenia, severe illness, or known colonization with Pseudomonas. Antifungal coverage for Candida species should be considered in patients on parenteral nutrition, prolonged broad-spectrum antibiotics, hematologic malignancy, organ or bone marrow transplant, or multisite Candida colonization.

In general, once the diagnosis of CRBSI is established, the catheter should be removed if not done already. If the patient is clinically stable, it may be reasonable to attempt to sterilize an infected catheter without removing it under certain circumstances. If the patient has limited options for IV access and depends on long-term IV access for survival, removing the catheter may offer more risks than benefits. However, if the patient has impaired immunity, exit site or tunnel infection, presence of other intravascular hardware (e.g., mechanical heart valve), or infectious complications such as endocarditis, septic emboli, osteomyelitis, or suppurative thrombophlebitis, cure is unlikely and the catheter should be removed or changed. For short-term catheters, salvage should be attempted only for uncomplicated infections with coagulase-negative Staphylococcus (other than S. lugdunensis) and Enterococcus species; however, it should be acknowledged that treatment failure is more likely than if the catheter were removed. If salvage is attempted, the catheter should be removed immediately if there is no clinical improvement or if blood cultures remain positive after 72 hours of appropriate antibiotic therapy. If catheter salvage is attempted, systemic antibiotics may be supplemented by intraluminal antibiotic lock therapy or continuous infusion. Success at eradicating a CRBSI without catheter removal is uncommon with S. aureus, Pseudomonas, fungi, and mycobacteria.

Definitions

Ventilator-associated pneumonia (VAP) refers to pneumonia that is newly acquired during endotracheal intubation or with a tracheostomy. VAP is not intended to include pneumonias that are acquired prior to intubation and progress to require mechanical ventilatory support. Generally, VAPs arise at least 48 hours after endotracheal intubation. However, the NHSN surveillance definition of VAP includes any pneumonia arising in a tracheally intubated patient regardless of how soon it arises after intubation and pneumonias that arise within 48 hours after extubation. Ventilator-associated tracheobronchitis (VAT) is defined as a lower respiratory infection that arises during tracheal intubation without radiographic evidence of pneumonia. The NHSN has proposed broader surveillance reporting of adverse events during mechanical ventilation, including ventilator-associated event (VAE), ventilator-associated condition (VAC), infection-related ventilator-associated condition (IVAC), possible pneumonia, and probable pneumonia.

VAP is a subcategory of hospital-acquired pneumonia (HAP), which is a pneumonia that is acquired in the hospital and is diagnosed at least 48 hours after admission. HAP is a subcategory of health care–associated pneumonia (HCAP), which is a pneumonia that develops in a patient with certain types of recent medical care. The importance of distinguishing these categories of pneumonia from community-acquired pneumonia (CAP) (Chapter 65) is that these infections are likely to be associated with pathogens that are more antibiotic resistant. Although the concept of VAP is relatively straightforward, there is no agreement on standards for clinical diagnosis (see “Diagnosis”).

The term ventilator-associated respiratory infection (VARI) includes both types of ventilators associated lower respiratory tract infections, VAP and VAT. HAP is a subcategory of health care-associated pneumonia (HCAP), which is defined in the ATS/IDSA guidelines (2005) as a pneumonia that occurs in a “patient who was hospitalized in an acute care hospital for 2 or more days within 90 days of the infection; resided in a nursing home or long-term care facility; received recent intravenous antibiotic therapy, chemotherapy, or wound care within the past 30 days of the current infection; or attended a hospital or hemodialysis clinic (within the past 30 days of the current infection).”