Preventing healthcare-acquired infections in intensive care units (ICUs) is a daily concern of physicians providing care for critically ill patients. Patients in ICUs are at increased risk for infection for multiple reasons, including their underlying illness, the use of medical devices for organ system support and hemodynamic monitoring, impaired nutritional status that contributes immune function compromise, and ongoing exposure to hospital antibiotic-resistant bacterial flora. The focus of this chapter is to review the general epidemiology of these infections, the factors contributing to their development, preventative strategies, and important characteristics of key healthcare-acquired pathogens.

Studies performed over the last two decades have found that infections in ICU patients are both common and significant. Work by Craven and colleagues in the early 1980s at Boston City Hospital in adult medical and surgical ICUs found that
overall 28% of the patients developed at least one nosocomial infection, and once infected patients had threefold increases in mortality [1]. Similarly, in 1995, European investigators assessed the prevalence of nosocomial infections in a multinational survey of 1,417 ICUs in 17 nations on one single day (the EPIC Study), and found an overall prevalence of ICU-acquired infection of 21% [2]. A more recent study in a single US medical ICU performed over 20 months in 2000 to 2001 found that 42% of patients requiring at least 48 hours of ICU care had a microbiologically confirmed infection, and patients with infection had a 1.9-fold increased risk of in-hospital mortality (p < 0.001) [3].

The types of infection seen in ICU patients have varied slightly over time and between types of ICU units, but several types of infections have predominated. In the Boston City Hospital study, the incidence of infections was higher in surgical ICU than in medical ICU patients; and pneumonia, surgical wound infections, urinary tract infections (UTIs), and bloodstream infections (BSIs) were the most frequent infections [1]. In the European Prevalence of Infection in Intensive Care (EPIC) study, the principal infections identified were ventilator-associated pneumonia (47% of infections), tracheobronchitis (18%), UTI (18%), and bacteremias (12%) [2]. Klevens and others, using a multistep approach, estimated 394,288 hospital-associated infections among adults and children in ICU in 2002 from US hospitals. The infection rate per 1,000 patient-days was 13.0: among all, UTI was the highest (3.38) followed by pneumonia (3.33) and BSI (2.71) [4]. The use of medical devices is a predominant cause of infection with the majority of episodes of nosocomial pneumonia associated with mechanical ventilation, healthcare-acquired infection (HAI) UTIs associated with urinary catheterization, and primary BSIs linked to central venous catheters.

Additional data through the National Nosocomial Infection Surveillance (NNIS) system of the Centers for Disease Control and Prevention (CDC) has assessed nosocomial infections in differing types of ICUs, and found that trauma/surgical and neurosurgical ICUs tend to have more nosocomial pneumonia than medical or coronary care ICUs; and that pediatric, trauma, and burn ICUs have more BSIs than medical ICUs [5,6]. The differences in infection rates noted are likely related to the size of the unit (small vs. large), the type of more predominant device use (urinary catheters, endotracheal tubes, and vascular catheters), the age group of the patients (pediatric vs. adults), and the most predominant illness of the patients (coronary, surgical, burn, medical, and pediatric).

Pediatric intensive care units differ from adult ICUs in many ways. First, they are typically combined units (medical and surgical). Second, their beds are not commonly physically separated as adults ICU beds. Third, pediatric patients usually have less comorbidity than adults. Data from the CDC NNIS system during the years 1992 through 1997 found a mean overall patient infection rate of 6.1%, with the principal infections being venous catheter–associated BSIs, followed by pneumonia and UTIs [6].

The predominant causes of ICU infections are a limited number of bacterial and fungal pathogens. In general, the pathogens that are seen can be characterized as those that survive well in a moist environment (e.g., Gram-negative bacteria including Enterobacter strains, Pseudomonas aeruginosa, and Acinetobacter species), those that colonize the skin and produce biofilm to allow adherence to catheters and other devices (e.g., Staphylococcus aureus and coagulase-negative staphylococci), and those which are resistant to commonly used antibiotics (e.g., methicillin-resistant S. aureus [MRSA], vancomycin-resistant enterococci [VRE], multidrug-resistant Gram-negative bacteria, and Candida species). In the EPIC study, the predominant pathogens were Enterobacteriaceae (34.4%), S. aureus (30.1%), and P. aeruginosa (28.7%) [2]. It was notable that 60% of the S. aureus isolates were MRSA, and that coagulase-negative staphylococci (19.1%) and fungi (17.1%) were common [2]. During the last 12 years, there has been an increasing trend toward highly antibiotic-resistant pathogens in the ICU setting. Data from the NNIS system on US ICUs comparing data from 1998 through 2003 has shown a progressive rise in the prevalence of MRSA to 60%, as well as dramatic increases in the prevalence of Klebsiella strains resistant to third-generation cephalosporins and P. aeruginosa resistant to cephalosporins and imipenem (Fig. 78.1) [5].

In the NNIS pediatric ICU study noted previously, for primary BSIs, coagulase-negative staphylococci were the most common pathogens (38%) followed by Gram-negative bacilli (25%) [6]. For nosocomial pneumonia, P. aeruginosa (22%) was the most frequent pathogen followed by S. aureus (17%), and for UTI, Gram-negative aerobic bacilli were the most frequent pathogens (57%) followed by fungi, most frequently Candida albicans (14%).

In addition to these predominant pathogens, there are several situations where other pathogens are a concern in the ICU. A number of institutions have noted the emergence of extended-spectrum β-lactamases (ESBL) producing Klebsiella and Escherichia coli strains [7,8]. In addition, in a few institutions that have used carbapenems extensively (often to try to treat ESBL-positive Gram-negative bacilli), the carbapenem-resistant Gram-negative pathogens Stenotrophomonas maltophilia, Klebsiella pneumoniae, Acinetobacter baumannii, and Burkholderia cepacia have emerged [9,10,11,12,13]. Finally, the fungal pathogens Candida parapsilosis and Malassezia furfur have been seen in patients receiving total parenteral nutrition, the latter being seen only with lipid supplementation [14,15].

The length of ICU stay is the predominant risk factor for nosocomial infection followed by the use of medical devices [2,3,6]. In the NNIS surveillance studies and subsequent studies by the CDC’s current National Healthcare Safety Network (NHSN), nosocomial infection rates for nosocomial pneumonia, BSIs, and UTIs have correlated strongly with device use [4,16]. Other risk factors include the patient’s underlying illness, selected medications, and the type of healthcare facility. In the EPIC study, seven risk factors were determined for ICU-acquired infection: increased length of stay (> 48 hours), mechanical ventilation, diagnosis of trauma, central venous, pulmonary artery, urinary catheterization, and stress ulcer prophylaxis [2]. Teaching hospitals with higher rates of device utilization have had higher device-associated infection rates [4,16]. As in adult ICUs, the most important risk factors for nosocomial infection in pediatric ICUs appears to be the length of ICU stay and rate of device utilization [2,6].

A potential risk factor undergoing intense study at this time is hyperglycemia. Hyperglycemia is common in the ICU setting due to underlying disease, physiologic stress, and parenteral nutritional support. In vitro investigations suggest that hyperglycemia can impair polymorphonuclear leukocyte and monocyte phagocytic and bactericidal activities [17]. A large randomized trial performed in a single surgical ICU found that tight control of blood glucose during the ICU stay (maintaining blood glucose 80 to 110 mg per dL) reduced overall mortality, the incidence of bacteremias, and the number of patients who required more than 10 days of antibiotic therapy [18].
However, a subsequent study of the impact of tight glycemic control on outcomes in a medical ICU did not find the same benefit, and further investigation of both the risk of infection with hyperglycemia and optimal treatment is needed [19].

A number of approaches have been found to help prevent ICU-associated infections. The comprehensive use of standard infection control practices as well as enhanced infection control precautions for selected pathogens, limiting the use of medical devices, and careful attention to architectural design are key components of strategies to prevent ICU infections. In addition, the implementation of targeted quality improvement programs for central vascular catheter infections and ventilator-associated pneumonia have been shown to be highly effective approaches to decreasing infection rates.