Although NIs are broadly defined as any infection that is not present on hospital admission, the CDC has proposed precise definitions of NIs, which are now generally considered the gold standard (a detailed list is available online at www.cdc.gov/ncidod/dhqp/nnis-pubs.html).¹³,¹⁴ In general, an infection is considered nosocomial if it develops 48 hours after hospital admission with no evidence of prior incubation. Additionally, infections that develop up to 3 days after the patient is discharged from the hospital or within 30 days of asurgical procedure are attributed to the hospital or surgical procedure, respectively.

The CDC National Nosocomial Infection Surveillance (NNIS) system¹⁵ collects data from numerous hospitals around the country for epidemiologic purposes. This data has demonstrated that infection rates vary widely depending on the type of ICU and the population of patients served (e.g., medical, surgical, or pediatric). These variations in infection rates suggest that the epidemiology for NIs differs with the type of units. Because of such demographic and geographic variations in infection rates, it is therefore recommended that each type of unit compare its own locally collected data with that collected by the NNIS for units of similar patient populations and alter infection control measures accordingly.

Pneumonia, urinary tract infections (UTIs), and bloodstream infections (BSIs) make up most of the reported
NIs.¹⁶,¹⁷ A report that looked at NIs from 205 combine medical-surgical ICUs from 1992 to 1998 found that infections from these three sites represented 68% of all reported infections with nosocomial pneumonia being the most frequently reported (31%), followed by UTIs (23%), and bloodstream infections (14%).¹⁷ Most infections were associated with medical devices: 83% of episodes of nosocomial pneumonia were associated with mechanical ventilation, 97% of UTIs occurred in patients with urinary catheters, and 87% of BSIs occurred with central line usage.¹⁷

Independent risk factors for the development of NIs have been identified. These include the severity of the underlying illness as assessed by the Applied Physiology and Chronic Health Evaluation (APACHE) score, prolonged hospitalization, mechanical ventilation, vascular access catheterization, and use of parenteral nutrition.¹⁸ Changes in hospital economics may also contribute to the risk of developing NIs. Understaffing of ICUs, as well as patient overcrowding, has been reported to be a risk factor for the development of NIs.¹⁹,²⁰ Patientto-nurse ratios were found to be a major independent risk factor with significantly increased relative risks associated with higher ratios.²¹ Reduced numbers of nursing and hospital staff with increasingly greater patient-care responsibilities for care givers not only has the potential to increase the risk of NIs, but also potentially impacts medical error rates and mortality. Risk factors for site-specific infections are discussed in each individual section given in the subsequent text.

The pathophysiology of NIs is believed to be dependent on two key factors: colonization of the host by potentially pathogenic microorganisms and impaired host defenses.²² Critically ill patients often have impaired host defenses due to the underlying disease process. The release of various cytokines and anti-inflammatory mediators can suppress the patient’s immune system, thereby increasing the risk of infection. Patients’ local immune defense mechanisms can also be impaired by invasive medical devices. For example, coughing and sneezing are impaired by endotracheal intubation and sedation leading to increased risk of respiratory infections. These devices can also act as a nidus for infection.

Colonization of the host can arise from endogenous or exogenous sources. Patients in the ICU are susceptible to bacterial colonization due to decreased host defense mechanisms, prolonged hospital stays, the use of indwelling medical devices, and the use of antibiotics. Antibiotics can exert selective pressure on the patient’s endogenous flora, potentially leading to colonization by pathogenic microorganisms. Common host reservoirs for NIs include the skin, oropharynx, urinary tract, and the gastrointestinal (GI) tract.²² The hands of health care workers are a well-documented and common source of exogenous colonization. Other sources include droplet or aerosol spread or contaminated medical equipment. Understanding potential sources of colonization is important to help identify routes of transmission and to assist in taking preventative infection control measures.

Any microorganism can be responsible for NIs in critically ill patients. Many infections in this subset of patients are polymicrobial in nature. There has been a recent change in the pattern of organisms with increased gram-positive infections.²¹ Recent evidence suggests that fungal infections are increasing in prevalence as well.¹⁶ The European Prevalence of Infection in Intensive Care (EPIC) study²³ identified the most common reported pathogens responsible for NIs. Staphylococcus aureus and Pseudomonas aeruginosa were identified most frequently (30% and 29%, respectively), with the remaining microorganisms including coagulase-negative staphylococci (19%), yeast (17%), Escherichia coli (13%), enterococci spp (12%), Acinetobacter spp (9%), and Klebsiella spp.²³,²⁴

It is important to understand that the exact patterns of microorganisms responsible for NIs vary by patient population, local infection control policies, antibiotic usage, type of ICU, and even geographic locations. Once again, the physician caring for critically ill patients must have a solid knowledge of local microorganisms and resistance patterns to best institute infection control practices and appropriate antibiotic usage. This is best achieved by careful coordination between health care workers, ICU pharmacists, the infection control team, and microbiology.

One of the most effective strategies against NIs is prevention. The simplest and most effective preventative strategy appears to be basic hygiene and care with the insertion and maintenance of indwelling medical devices (i.e., central lines). The transmission of microorganisms from the hands of health care workers is well documented in the literature. Hand washing before and after patient contact remains one of the simplest and most effective infection control procedures, but compliance is often poor.²⁵,²⁶,²⁷,²⁸ Numerous factors may play a role in the low compliance with hand washing including lack of time and inconvenient placement of hand washing facilities. Additionally, health care workers who participate in frequent hand washing run the risk of skin damage including allergy and intolerance to hand cleaning solutions. Skin damage results in potential shedding of more organisms into the surrounding environment, thereby increasing the risk of infection.²⁹

There is growing evidence that hand washing with alcohol-based gel solutions not only reduces hand contamination more than hand washing alone but may also in fact improve compliance.³⁰,³¹ Updated guidelines for hand hygiene in the health care setting are available from the CDC website (www.cdc.gov/handhygiene/). The use of universal precautions should also be stressed. Gloves should be worn in all situations where contact with blood, mucous membranes, secretions, or a wound is anticipated. Hand washing must be undertaken after glove removal because hands may be contaminated during the procedure or after glove removal. Masks should be worn when contamination from splashes or sprays may occur.