Healthcare-Associated Infections

Leilani Paitoonpong

Chun Kwan Bonnie Wong

Trish M. Perl

INTRODUCTION

Healthcare-associated infections (HAIs) are infections that patients acquire during the course of receiving medical treatment for other conditions.¹ They are defined as localized or systemic conditions resulting from an adverse reaction to the presence of an infectious agent or its toxin(s) that has no evidence of being present or incubating at the time of admission to the care setting.¹ In general, infections are defined as associated with health care if they develop 48 hours after admission or receiving medical care or within 30 days of having a surgical procedure. They may be caused by a variety of infectious agents from endogenous and exogenous sources. Encapsulated in this general concept is the acquisition of organisms of epidemiologic significance that may not cause infection per se. In this case, the individual may become colonized with and harbor an organism that does not cause infection but could increase the risk of developing an infection. Such organisms include methicillin-resistant Staphylococcus aureus (MRSA), vancomycin-resistant enterococcus (VRE), and gram-negative rods (GNRs) that produce extended-spectrum beta-lactamases (ESBLs) or carbapenemases.

HAIs occur in all settings of care and across the continuum of care, including hospital acute care units, day procedure centers, ambulatory outpatient clinics, dialysis centers, long-term care facilities, nursing homes, and rehabilitation centers. To better reflect the diversity of the modern healthcare system that our patients encounter nowadays, the term HAIs has replaced older terms such as “nosocomial infections,” “hospital-acquired infections,” and “hospital-onset infections.”

HAIs occur for primarily four reasons:

Host factors: Individuals may have a compromised immune system due to underlying disease or due to disruption of mucosal and skin surfaces that increases their risk of developing infections or acquiring organisms that are either high- and lowvirulence organisms.
Environment: The hospital environment promotes the spread of microbial pathogens. The proximity to other patients, contamination of common equipment, exposure to water contaminated with microorganisms, presence of construction and renovation, tasks that healthcare providers (HCPs) perform, and the often-unwashed hands of HCPs contribute to creating the ideal conditions for transmission of infectious organisms.
Technology: Technology-based advances in health care provide sophisticated methods of monitoring and caring for patients. Unfortunately, these advances also provide new portals of entry for infection, alter normal host flora, and may increase antimicrobial resistance, thereby increasing the risk of HAIs.
Human factors: With the tremendous resource cuts that have occurred in the healthcare industry in recent years, the number and skill level of caregivers has decreased. Nontraditional and support staff now provide previously specialized nursing functions. HCPs are busier than ever, allowing them little time to observe simple infection prevention practices, such as hand washing. These changes may contribute substantially to healthcare-associated transmission of organisms and development of infections.

Bacteria cause approximately 90% of HAIs, with viruses, fungi, protozoa, and other classes of microorganisms causing the remaining infections. Based on surveillance data collected by the National Healthcare Safety Network (NHSN), the most common pathogens isolated from any HAIs are
coagulase-negative staphylococci (CoNS, 15.3%), Staphylococcus aureus (14.5%), Enterococcus species (12.1%), Escherichia coli (9.6%), and Pseudomonas aeruginosa (7.9%).² As is true for community-acquired infections, the most frequently isolated organism depends on its source. E. coli most commonly causes urinary tract infections (UTIs; 21.4%); S. aureus most commonly causes surgical site infections (SSIs; 30%); CoNS most commonly causes central-line associated bloodstream infection (CLA-BSI; 34.1%); and S. aureus and P. aeruginosa cause 24.4% and 14.3%, respectively, of ventilatorassociated pneumonia (VAP).²

THE MAGNITUDE OF PROBLEM OF HAIs

The problem of HAIs is a global issue. These infections are a major cause of morbidity and mortality in both pediatric and adult populations in the United States and worldwide. HAIs occasionally affect HCPs as well. The World Health Organization (WHO) estimates that 1.4 million people are affected with an HAI at any given moment (www.who.int/gpsc/country_work/summary_20100430_en.pdf). The prevalence of infection varies from 4.5% to 19.1%, with infection rates being higher in developing countries (Table 14-1). HAIs are becoming more of a challenge as modern medical care utilizes more invasive procedures and more sophisticated devices. Hence, much of the surveillance focus has been on infections associated with the placement of these devices. The emergence of antimicrobial-resistant pathogens and a lack of new antimicrobials in the pipeline has further fueled the problem. The increasing awareness of resistant pathogens has fostered a search for systems that can measure both colonization and infection. An aging population, the AIDS epidemic, the growth of chemotherapeutic options for cancer treatment, and a growing transplant population have also expanded the populations at risk for infection.³

In the United States, the Centers for Disease Control and Prevention (CDC) estimates that nearly 2 million patients experience an HAI each year, meaning 1 of every 10-20 patients hospitalized in the country develops such an infection.⁴ These infections cause almost 100,000 deaths and are associated with an extra $4.5 billion to $6.5 billion in costs.⁵ Based on these estimations, the number of deaths attributable to HAIs exceeds that for several of the top six leading causes of death in the United States. Of these infections, 36% occur in the urinary tract, 20% at the surgical site, 11% in the lung, and 11% in the bloodstream.⁴

Table 14-1 WHO Estimate of the Burden of Healthcare-Associated Infections Worldwide, 1995-2008

Country	Estimated Prevalence
Korea	3.7%
United States	4.5%
Norway	5.1%
Latvia	5.7%
France	6.7%
Lebanon	6.8%
Thailand	7.3%
United Kingdom and Ireland	7.6%
Cyprus	7.9%
Italy	8.3%
Finland	9.1%
Lithuania	9.2%
Greece	9.3%
Scotland	9.5%
Switzerland	10.1%
Canada	11.6%
Turkey	13.4%
Malaysia	13.9%
Brazil	14.0%
Tanzania	14.8%
Tunisia	17.8%
Morocco	17.8%
Mali	18.7%
Albania	19.1%
Data from World Health Organization (2011). The Burden of Health Care-Associated Infection Worldwide. www.who.int/gpsc/country_ work/summary_20100430_en.pdf. Accessed February 29, 2012.

Multiple studies have shown that HAIs prolong the duration of hospitalization, and increase morbidity and mortality to patients.⁶, ⁷, ⁸, ⁹, ¹⁰, ¹¹ and ¹² They are costly both to the patient and to the healthcare system. It is estimated that at least 20% of all such infections could be prevented by better hygiene and infection control procedures.¹³

Worldwide, HAIs are now recognized as an important public health problem and warrant attention from policymakers, public health authorities, hospital leadership, and frontline healthcare personnel. Preventing these infections has now become a national priority for many countries, with initiatives in this area being led by healthcare organizations, professional associations, government and accrediting agencies, legislators, regulators, payers, and consumer advocacy groups.¹⁴ Public reporting of HAIs occurs in many countries, including France, the United Kingdom, and the United States.

More than 4 million people in the European Union (EU) acquire a HAI annually, and approximately 37,000 die as a direct result of the infection. A point prevalence study conducted in 1417 intensive
care units (ICUs) in 17 European countries demonstrated that as many as 44.8% patients were infected, and half of these patients (20.6%) acquired their infections in the ICU. Pneumonia was most common infection in this cohort, followed by urinary tract and bloodstream infections. Moreover, many of the infections were associated with multidrug-resistant organisms.¹⁵

The burden of HAIs is faced not only by developed countries, but also by resource-limited countries. While reports from developed countries are able to provide a glimpse into the scale of the problem of HAIs, few data are available from the developing world to quantify its scope. Nonetheless, the available data suggest that the prevalence of these infections, and hence the burden of disease, is higher in the developing world (Table 14-1).

Data collected in the 1980s in a WHO cooperative study, which included 55 hospitals in 14 countries from four WHO regions, revealed a mean prevalence of 8.7% for HAIs. The highest frequency of infection was noted in ICUs and acute surgical and orthopedic units. A slightly different epidemiology in terms of type of HAIs was used in this study—namely, surgical site infections contributed the most (25.1%), followed by urinary tract infections (22%) and pneumonia (20.5%).¹⁶

A 6-year surveillance study from 2003-2008 involving 173 ICUs in Latin America, Asia, Africa, and Europe, using the NHSN definitions, revealed a markedly higher rates of CLA-BSI, VAP, and catheterassociated urinary tract infections (CA-UTIs) in these regions than in comparable U.S. ICUs. The crude unadjusted excess mortality for device-related infections was reported to range from 23.6% (CLA-BSI) to 29.3% (VAP). The survey also reported higher frequencies of MRSA, Klebsiella pneumoniae resistant to third-generation cephalosporins, Acinetobacter baumannii resistant to imipenem, and Pseudomonas aeruginosa resistant to pipericillin.¹⁷

A more recent study from WHO estimated that the prevalence of HAIs, as derived from pooled data from countries outside Europe and the United States, was 15.5 per 100 patients (95% confidence interval [CI]: 12.6-18.9).¹⁸ The pooled overall HAI density in adult ICUs was estimated at 47.9 per 1000 patientdays (95% CI: 36.7-59.1), at least three times as high as the density reported from the United States.¹⁸ The lack of national infection control guidelines, the lack of a framework to reinforce compliance with guidelines when they are in place, and a lack of administrative and financial support and trained personnel in the settings of these resource-limited countries all contribute to their alarmingly higher rates of HAIs.¹⁷

While the burden of these infections varies from study to study because of the various methodologies used by investigators, several key points can be made:

These infections are common.
They are associated with significant morbidity and mortality.
Multidrug-resistant organisms are commonly the cause of these infections and colonization.
HAIs affect healthcare systems and patients worldwide.

Keeping these facts in mind, infection prevention programs should be organized to reduce HAIs and the spread of resistant organisms, along with their associated morbidity and mortality. The ultimate goal of these programs is to improve patient care, reduce hospital stays, and reduce healthcare-related costs. It is estimated that approximately one-third of HAIs might be prevented by well-organized infection control programs. Unfortunately, less than one-tenth (6% to 9%) of infections are actually prevented.¹⁹, ²⁰ The following components of these programs are crucial in reducing HAIs:

A trained physician with expertise in infection prevention and control (hospital epidemiologist)
At least one infection control practitioner (ICP; now called an infection preventionist [IP]) per 250 beds²¹
A computerized surveillance system
A system of reporting HAI/colonization rates of hospitalized or at-risk patients to practicing physicians and surgeons¹⁹

HISTORY

The earliest advice regarding hospital hygiene was probably written in the fourth century B.C., in the Charaka-Samhita, a Sanskrit textbook of medicine, based on Indian Vedic medicine. The following excerpt demonstrates the early principles of hospital infection prevention and control:

In the first place a mansion must be constructed under the supervision of an engineer well conversant with the science of building mansions and houses. It shall be spacious and roomy…. One portion at least should be open to the current wind. It should not be exposed to smoke, or dust, or injurious sound or touch or taste or form or scent…. After this should be secured a body of attendants of good behavior, distinguished for purity and cleanliness of habits.²²

The role of the facility and the environment continued to be recognized in medieval and renaissance Europe, where hospitals were overcrowded. Pioneers such as Theodoric of Bologna and Casper Stromayr reformed hospital practice by bathing their patients or shaving the site of operation, yet met with little success.²³ Toward the end of the 1700s, Madam Necker proposed “nursing the sick in a single bed.” Prior to this time, as many as eight patients were nursed in a bed amid appalling squalor.²⁴

Simpson, a Scottish surgeon, demonstrated an enduring concept that mortality following amputation was proportional to the size of the hospital and the degree of overcrowding.²⁵, ²⁶ And in the mid-1800s, Florence Nightingale, a nurse, published a dramatic report after her Crimean War experiences at the military hospitals, which demonstrated that far more soldiers died of HAIs than died from the primary effects of battle injuries.²⁷ Furthermore, Nightingale, with the help of John Farr, demonstrated a direct relationship between sanitary conditions at a hospital and postoperative complications.²⁸ She proposed that ward sisters maintain records of hospital patients who developed infections and introduced broad hospital hygiene, thereby pioneering the concept of HAI surveillance.

During the same era, Dr. Ignaz Semmelweis, the head of obstetrical service at the Royal Lying-In Hospital of Vienna, investigated the high maternal mortality rate at his facility. In doing so, he probably undertook the first hospital-based epidemiologic study.²⁹ Semmelweis recognized that rates of puerperal fever were higher in the ward staffed by physicians as opposed to the ward tended by nurse-midwives. A key difference between the two types of providers was that the physicians participated in autopsies of women with puerperal sepsis. They would then return to the wards to care for women in labor. Semmelweis observed this variation in care and linked the increased rates to a lack of hand washing after performing autopsies. Once hand washing was instituted after performing autopsies as a control measure, the ward-specific excess rate of puerperal sepsis and mortality associated with physicians declined and was equal to that for the midwives.²⁹

Louis Pasteur continued to develop our understanding of microbiology and the role of organisms in infection when he demonstrated that air is contaminated with living germs and that growth of such organisms leads to spoilage.³⁰, ³¹ and ³² Lister then reasoned that microbes were responsible for wound suppuration and introduced antisepsis to surgery.³³ He proposed controlling suppuration by preventing contaminated air from coming in contact with a wound. Similarly, in 1895, George Emerson Brewer, an American surgeon, recognized the problem of infections after surgery and undertook an intensive surveillance project to estimate the frequency of surgical wound infections, now called surgical site infections (SSIs).³⁴ He reported that SSI rates among patients after clean surgical operation was not 5% or less, but actually 39%—findings that inspired a review of surgical techniques and environmental risk factors. In 1915, Brewer reported in a follow-up a steady decline of SSIs to 1.2% among patients undergoing clean surgical procedures.³⁵

In 1937, with the introduction of penicillin to treat serious staphylococcal and streptococcal infections, misconceptions developed that antimicrobial agents might be able to control and eventually eliminate all infectious diseases. Some physicians worried less about infections because nontoxic drugs could prevent serious complications and cure infections. In the 1940s, the concept of postsurgical antimicrobial prophylaxis was introduced, and widespread use of penicillin occurred. The increased use of penicillin resulted in the rapid emergence of resistant staphylococci. In the late 1950s, the first epidemics of healthcare-associated penicillinase-producing Staphylococcus aureus were reported in Europe and North America. In these outbreaks, patients—primarily neonates admitted to hospital nurseries without infections—subsequently developed staphylococcal sepsis.³⁶ These outbreaks generated greater interest in prevention practices within healthcare and regulatory agencies.

Over the ensuing decade, gram-positive organisms were the primary pathogens causing HAIs. Such infections increased in severity as treatment options shrank with the emergence of resistance to methicillin in S. aureus. By the 1970s, and perhaps in response to antibiotics with good gram-positive coverage, gramnegative organisms became most frequent cause of HAIs. However, the 1980s, new gram-positive organisms—specifically, CoNS, MRSA, and fungi, primarily Candida species—emerged as important HAI pathogens.³⁷ Since then, both bacteria and fungi causing HAIs have become increasingly resistant to antimicrobial agents.

For example, the prevalence of methicillin resistance among S. aureus isolates increased from 2.4% in 1975 to 29% in 1991.³⁸ Infections caused by MRSA are similar to those caused by methicillinsensitive S. aureus (MSSA), in that they are life threatening and, in fact, are associated with increased mortality.³⁹ Vancomycin is the most commonly used antimicrobial agent against these resistant organisms, and its frequent empiric use has contributed
to the development of vancomycin resistance among S. aureus and other organisms. Likewise, the incidence of VRE infection has also increased dramatically. Most concerning is the emergence of S. aureus strains with intermediate susceptibility to vancomycin or other glycopeptide antibiotics. Should glycopeptide-intermediate susceptibility (GISA) emerge as a major pathogen, we would essentially reenter the pre-antibiotic era with respect to their control.

In 1959, the first ICP was appointed in England to control hospital-acquired Staphylococcus infections.⁴⁰ It was not until 1963, however, that a full-time practitioner was hired to prevent HAIs in the United States.⁴¹ In 1964, Boston City Hospital conducted some of the first modern prevalence studies and reported that 15% of inpatients had HAIs.²⁰, ⁴² By 1968, the CDC was training ICPs in surveillance, prevention, and control of nosocomial infections, and in 1969, the Joint Commission for Accreditation of Healthcare Organizations (now The Joint Commission) mandated that all hospitals support an ICP. These trained professionals, who come from nursing, medical technology, and epidemiology backgrounds, are now called infection preventionists and are key to institutional program prevention and control efforts.

In 2003, the severe acute respiratory syndrome (SARS) epidemic catapulted the need for better infection control strategies and public health and institutional infrastructure into the public eye.⁴³, ⁴⁴ and ⁴⁵ This epidemic killed more than 800 people and infected more than 8000 people worldwide.⁴⁶ With increasing visibility in the press, the emerging problem with antimicrobial resistance, and the demand for greater transparency, both patients and regulators worldwide have demanded that HAI rates, and in some cases process measures, be publicly reported. This trend has inspired a greater focus on patient safety, as HAIs are the primary complication that puts patients at risk. Fortunately, this has led to a call for zero infections. While this goal may not be achievable, we can have zero tolerance for practices and processes that are not supported by evidence to reduce patient risk.

SURVEILLANCE FOR HAIs

Definitions

HAIs are defined using a set of standard criteria that have been developed over time. In 2012, the definitions were undergoing revision. The CDC took the lead and developed the National Nosocomial Infection Surveillance (NNIS) system in 1970, which was composed of non-randomly selected hospitals that voluntarily submitted their data collected using standard methods and definitions. In 2005, NNIS was restructured into the National Health Safety Network (NHSN), which has almost 4000 participating hospitals. Today, the NNIS/NHSN is the most important source of national data on HAIs in the United States.⁴⁷ This system has also served as the model for systems and programs in other countries designed to ameliorate HAIs.

The CDC/NHSN has assumed a primary role in developing HAI definitions for the purposes of surveillance.¹ These definitions have been tested and validated for surveillance purposes in many acute care settings. Separate definitions have been developed for long-term care facilities.⁴⁸ Use of uniform definitions are critical if investigators are to identify patterns within hospitals and to compare rates of HAIs between hospitals or data systems. While not perfect, these definitions’ persistence over the past 40 years has provided the infection prevention community with a standard way to identify complications that are associated with poor patient outcomes, to follow trends over time, and to measure the impact of prevention strategies. The definitions are often based on microbiologic, serologic, histologic results, although they may also use clinical criteria to differentiate a colonized body site versus one causing infections.

Measures of Frequency of HAIs

HAIs are evaluated by calculating rates, rather than the simple number of events or infections. For instance, the presence of six HAIs at a large institution (bed size > 500) has different implications than the same number of infections at a small institution (bed size < 25).

Crude Infection Rate

The most common measure of occurrence, the crude infection rate, is the ratio of infections per 100 admissions or discharges:

The crude infection rate can also be site-specific. For example:

Adjusted Infection Rate

To derive a rate that represents changes in the HAI rate from one period to another, the crude infection rate is often adjusted by patient-days or number of procedures. This step corrects for a patient being admitted in one month and discharged in another month. An example of an adjusted infection rate is the ratio of infections per 1000 patient-days or 100 surgical procedures.

The crude infection rate does not adjust for risk factors for HAIs that vary among patients. The riskadjusted infection rate adjusts the rate by the major risk factor for these device-related infections by the number of days that a medical device (e.g., central venous catheter, urinary catheter, ventilator) is used. A prospective study demonstrated that data collected for HAI rates should be adjusted for risk factors to correct for severity of illness.⁴⁹

Device-Associated Infection Rates

A device-associated infection rate is a specific example of a risk-adjusted infection rate. The device-associated rate is most commonly reported as infections per 1000 device-days:

Before calculating the risk-adjusted infection rate, certain steps should be followed:

Decide on the time period for the analysis (week, month, quarter, half-year, or year).
Select the patient population (ICU, surgical patients).
Choose the site of infection for which the rate is to be calculated (numerator).

The infection selected should be site specific and should occur in the selected population; the infection onset-days should occur during the time period selected. Once the device-days are determined, infections such as catheter-associated urinary tract infections would be calculated using the following formulation:

Calculation of the Number of Days That a Device Is in Place

In January 2011, on the first day of the month, suppose 20 patients had urinary catheters; 19 had catheters on day 2; 15 had catheters on day 3; 25 had catheters on day 4; 20 had catheters on day 5; 15 had catheters on day 6; and 15 had catheters on day 7. The number of patients with urinary catheters from days 1 to 7 is added (20 + 19 + 15 + 25 + 20 + 15 + 15), yielding 129 urinary catheter-days for the first week (Table 14-2). The total catheter-days for the entire month is the sum of the daily counts.

Device-Day Utilization

To determine the percentage of patient-days compared to device-days, the device utilization ratio can be calculated. The device-day utilization rate is specifically useful for measuring infection risk among patients in ICUs.

To calculate the number of patient-days, let us use another example. In January 2011, 20 patients were hospitalized at midnight on the unit on the first day; 20 on day 2; 18 on day 3; 25 on day 4; 24 on day 5; 20 on day 6; and 18 on day 7. To calculate the patient-days, add the number of patients in the unit from day 1 to day 7. The total number of patientdays for the week is 145 (Table 14-2). Remember that the patient-days are always calculated at the same time. Thus the patient-days are the total number of days that patients are in a unit during a selected time
period. We had earlier calculated the total urinary catheter-days to be 129. Thus

Table 14-2 Comparison of Denominators: Device Utilization Days versus Patient Days

Number of Day	Number of Device-Days	Patient-Days
1	20	20
2	19	20
3	15	18
4	25	25
5	20	24
6	15	20
7	15	18
Total	129	145

Eighty-nine percent of patient-days were also urinary catheter-days for the first week of the month.

Calculating the device-associated, device-day rate and utilization ratio helps IPs to evaluate how their hospital compares with the mean rates, such as those reported by the NHSN system. Some caveats apply. If the denominator is small (less than 50 device-days or patient-days), this ratio will not be a good estimate of the “true” device utilization. Therefore, a longer time period should be chosen. Also, not all hospitals are similar to the hospitals included in the comparison group. If huge variations in hospital infection rates are noted, reasons for these variations should be explored. Specialty care facilities, for example, may have higher rates of infection due to particularly vulnerable patients or types of surgery performed.

Incidence and Prevalence

To measure incidence and prevalence, surveillance surveys are performed. Incidence surveys determine the rate of new infections during a given time, whereas prevalence surveys determine the proportion of patients with infections at a given point in time. To calculate the incidence rate (I), the number of HAIs (during a given month) is divided by the number of patients discharged or admitted (during the same month), or by the number of patient-days⁴⁰, ⁵⁰:

In addition, incidence rates can be site or organism specific, with the number of infections at a given site being used in the numerator, and the denominator as described previously.

Prevalence surveys are designed to measure all current HAIs. Because patients who experience HAIs are likely to have longer hospital stays, infected patients are typically over-represented in the hospital population. The prevalence rate then overestimates infection rates, compared with incidence rates. Nevertheless, prevalence surveys are often used because they are relatively easy to perform and provide a picture at a single point in time.⁵¹ Prevalence surveys are useful in to determine the HAI rates in large populations of high-risk patients or to identify areas where targeted investigation may be useful.

During a prevalence survey, infection control personnel examine all patients’ medical records and interview clinical staff to identify HAIs. The prevalence rate (P) is calculated by dividing the number of active (current) HAIs present on a given day in hospitalized patients by the number of patients hospitalized on the same day:

Methods to Evaluate Risk Factors for HAIs and the Impact of Interventions

Both observational and interventional epidemiologic studies can be used to study HAIs. Observational studies, such as case-control studies and cohort studies, are used to identify risk factors in outbreak investigations or epidemiologic studies and to assess outcomes. Intervention studies, such as clinical trials, are used to evaluate whether a given intervention is effective. Surveillance and outbreak investigations are routinely performed in most infection control programs. Randomized clinical trials are usually performed in infection control programs with active research programs.

Surveillance, a key component of infection control programs, has been defined as “a dynamic process for gathering, managing, analyzing, and reporting data on events which occur in specific populations.”⁵² Surveillance relies on four building blocks:

Collecting relevant data systematically for a specified purpose and during a defined period of time
Managing and organizing the data
Analyzing and interpreting the data
Communicating the results to those empowered to make beneficial changes

Surveillance is also a method used to measure the rate of HAIs events. The purpose of measuring the event is to determine the baseline rate, thereby enabling the investigator to compare the rate after an intervention has been instituted. Hence, surveillance is a necessary component of an intervention study.

Methods of Surveillance

Each institution should determine which method is best suited to its facility. The most common surveillance methods include the following:

Hospital-wide surveillance
Prevalence surveys
Targeted surveillance
Periodic surveillance⁴⁷

Hospital-wide surveillance is the most comprehensive technique, but is also expensive and labor intensive. Prevalence surveys are inexpensive and can be applied to individual units or entire institutions.⁴⁷ Targeted surveillance focuses on selected areas of the hospital, selected patient populations, or selected organisms.³ When based on the highest-risk areas or patients, it has been preferred over hospital-wide surveillance because it represents the most effective use of resources.⁵³ However, surveillance for certain epidemiologically important organisms may need to be hospital-wide.⁵³ Periodic surveillance is used when surveillance methods are done only during specified time intervals.³

Various methods or mixtures of methods are used to identify patients with HAIs:

Total chart review
Selective medical record review (e.g., chart summaries—either handwritten, such as the nursing summary notes, or stored in hospital databases)
Reports of clinical symptoms from providers
Clinical ward rounds
Review of laboratory reports
Extraction of data from pharmacy records, such as antimicrobial use
Computer alerts and computer-based automated surveillance
Follow-up letters/calls to providers⁴⁷

The sensitivity and specificity vary for these methods of detecting HAIs when compared with the gold standard—total chart review. However, total chart review has not been shown to improve the sensitivity of detecting HAIs, compared with selective review of charts, based on review of laboratory reports and the nursing summary notes (74-94% versus 75-94%) and is very time intensive.⁵⁴ Casefinding methods must be applied systematically so that results are comparable over time. Most IPs choose their surveillance strategies based on the type of hospital, the patient population served, the resources available, and the “local” epidemiology and culture. New surveillance technologies using computer programs have been developed as well. For example, many institutions have computer programs or software that facilitate either surveillance or identification of patients colonized or infected with epidemiologically important organisms.⁴⁷, ⁵⁵ This software usually integrates the result from the microbiology, radiology, and pharmacy databases and clinical information from a variety of data sources.³ Automated surveillance with user-definable control charts is more efficient at identifying potential outbreaks than routine surveillance.⁵⁶

Outbreak investigations are the epidemiologic studies most closely identified with infection prevention programs, even though epidemic-related infections account for fewer than 5% of all HAIs.⁵⁷ Outbreak investigations begin with the identification of an unusual occurrence or an excessive rate of infections. Cases are defined and described in terms of place, person, and time. Organisms should be speciated and compared for similarity, using antimicrobial susceptibility and molecular technologies. If the cause of the outbreak is not overtly apparent or if a study is required to confirm the important exposures or causes, a case-control study is usually performed to determine risk factors. Steps for outbreak investigation are provided in Table 14-3.

Information about the risk factors for and the outcomes of HAIs is often obtained through different methods of investigation. Observational, casecontrol, and cohort studies are the most common study designs used. For example, knowledge of the risk factors for surgical site infections was obtained through a series of cohort studies in which surgical patients were followed postoperatively for SSIs.⁵⁸ Information on any potential risk factors for SSI was collected on all patients in the cohort, patients with
and without SSIs were compared, and relative risks for the risk factors were calculated. The costs of HAIs have also been measured using matched case-control methods and cohort studies. For example, patients with CA-UTI have been matched to controls with similar diagnoses and severity of illness, and then costs and lengths of stay compared.⁵⁹, ⁶⁰

Table 14-3 Steps to Investigate an Outbreak

Determine the nature, location, and severity of the problem.
Identify cases.
Document notes regarding the outbreak in an orderly fashion.
Conduct a literature search.
Create a preliminary questionnaire.
Review medical records to establish a case definition.
Save isolates from cases, as well as from suspected cases and/or source.
Summarize the data into an easy-to-read line listing.
Form a hypothesis (source, mode of transmission, cause).
Test the hypothesis by either a case-control or cohort study.
Create an epidemic curve.
Demonstrate biologic plausibility.
Inform the appropriate individuals and agencies.
Institute emergency control measures:

i. Eliminate the source, which could be the environment, a patient, or the HCP.

ii. Protect exposed individuals by administering chemoprophylaxis or immunization.
Evaluate the control measures.
Document and report the outbreak.

Randomized controlled trials are the best study method to prove causation. In the setting of population-based analysis, two strategies are employed: those that randomize individuals to receive one treatment/intervention or not, and those that randomize units (cluster randomized) to a treatment or not. Cluster randomized studies are becoming more common, as they simplify study implementation; that is, units (rather than individual patients) are randomized to one intervention, which reduces the need to track individual patients while allowing the investigators to retain control over potential confounders through randomization. Clinical trials have been used to show the effects of sterilization and disinfection, closed urinary drainage systems, intravascular catheter care, dressing techniques, and care of respiratory therapy equipment related to HAIs.⁶¹

In addition, clinical trials have been important in showing what does not work. For example, a randomized clinical trial of antimicrobial prophylaxis for patients with chronic urinary catheters demonstrated no difference in the rate of UTIs and febrile episodes, compared with the control group, whose members did not receive prophylaxis.⁶²

PATHOPHYSIOLOGY AND RISK FACTORS

Hospitalized patients, in general, are at high risk for infection, due to their underlying illness (immunosuppression, diabetes); hospitalization circumstances (trauma, burns); environmental, microbiologic, and virulence factors; procedure-related interventions, such as surgery or medical care (urinary catheter, vascular catheter, ventilators); and the process of care (patient/nurse ratios, inappropriate antibiotic use). Because patient characteristics are frequently beyond our control and difficult to alter, reduction in HAIs is best achieved by altering HCP behaviors, procedure-related techniques and conditions, or other processes of care.

Host Factors

Host factors that contribute to hospitalized patients developing infections include extremes of age, severity of underlying illnesses, immune dysfunction (T- or B-cell mediated), poor nutrition, genetic factors, and loss of the body’s normal protective functions (skin integrity, microbial imbalance). Extremes of age are a major risk factor for HAIs.⁴⁹, ⁶³, ⁶⁴ and ⁶⁵ In fact, data collated by the CDC indicated that 54% of all HAIs in adults occurred among those aged 65 years or older.³⁸ Thus, while those older than age 60 represented only 23% to 24% of discharges, they accounted for 37% to 64% of all HAIs.⁶⁶, ⁶⁷

Several studies have shown that HAIs are related to underlying illness. In particular, pulmonary, cutaneous, and hematologic diseases alter host defenses by changing or modifying normal flora, breaching normal anatomic barriers, suppressing inflammatory responses, and modifying the reticuloendothelial system.⁶⁸, ⁶⁹ and ⁷⁰ Underlying disease that causes immunosuppression, such as cancer or HIV infection, can make the host highly prone to HAIs.⁷¹ In a 2-year study conducted in an oncology ICU, the overall infection rate was 50 cases per 100 patients, or 91.7 cases per 1000 patient-days.⁷² Neutropenic patients have altered immunity, are exposed to many antimicrobial agents, and are hospitalized frequently for prolonged periods of time. These patients may be at a higher risk of developing HAIs due to bacterial colonization and catheter placement.⁷³, ⁷⁴ and ⁷⁵ Other underlying diseases put certain groups of patients at higher risk for HAIs. For example, obese patients are at higher risk of postoperative infection in various kinds of surgical procedures.⁷⁶, ⁷⁷, ⁷⁸, ⁷⁹, ⁸⁰, ⁸¹, ⁸² and ⁸³ Furthermore, obesity was found to be an independent risk factor for bloodstream catheter-related infection in ICU in a recent study.⁸⁴

Environment

Air, water, and the inanimate surfaces surrounding the patient are referred to as the environment. Contamination of the floor, walls, bed frames, chairs, water, and air of the patient’s environment may lead to HAIs.

Air

Malfunctioning or inadequate ventilation systems in healthcare facilities may not adequately filter air. Aspergillus species—one of the most invasive fungi—under appropriate environmental conditions can produce and disseminate several thousand spores per cubic meter of air.⁸⁵, ⁸⁶ and ⁸⁷ These spores can remain suspended in air for long periods, but eventually they settle and can contaminate surfaces. The spores remain viable for months and can become airborne when dust-generating activities are performed, such as construction or demolition. If walls or surfaces
become wet and are not replaced, molds can also grow and become sources of infection.

Water

Since the etiologic agent of Legionnaires’ disease was first identified in 1976, numerous outbreaks of healthcare-associated Legionnaires’ disease have been identified.⁸⁸, ⁸⁹ and ⁹⁰ Legionella can colonize the water systems of large buildings and hospitals.⁹¹ Hospital hotwater distribution systems and water-cooling towers for air conditioners have been implicated as sources of legionellosis outbreaks in patients.⁹², ⁹³, ⁹⁴, ⁹⁵, ⁹⁶, ⁹⁷, ⁹⁸, ⁹⁹, ¹⁰⁰, ¹⁰¹, ¹⁰², ¹⁰³, ¹⁰⁴, ¹⁰⁵, ¹⁰⁶, ¹⁰⁷, ¹⁰⁸ and ¹⁰⁹ Other water sources should also be considered as potential sources of contamination with pathogens, including ice machines, humidifiers, and areas where water accumulates and is maintained at the appropriate temperature.¹¹⁰, ¹¹¹, ¹¹², ¹¹³, ¹¹⁴, ¹¹⁵, ¹¹⁶, ¹¹⁷, ¹¹⁸, ¹¹⁹ and ¹²⁰ Besides Legionella, the hospital water system or attached equipment can be a source of either an outbreak or pseudo-outbreak involving non-tuberculous Mycobacterium, fungi, and other nonfermenter gram-negative organisms (e.g., Pseudomonas spp., Burkholderia spp., Stenotrophomonas spp., Acinetobacter spp., Ralstonia spp., Sphingomonas spp., Afipia felis).¹²¹, ¹²², ¹²³, ¹²⁴ and ¹²⁵ Whenever outbreaks of these organisms occur, the link to water reservoir should be on the priority list to investigate. An outbreak of Legionnaires’ disease in hospitals in the fall of 2011 resulted in the elimination of water features from hospitals. While soothing and attractive, the risk of legionella contamination of these water reservoirs was high.

Inanimate Objects and Environmental Surfaces

Spread of resistant organisms has been linked to a contaminated environment or a fomite.¹²⁶ For example, Clostridium difficile, Acinetobacter species, VRE, and S. aureus have been cultured in the environment and in some cases linked to subsequent infections in patients.¹²⁷, ¹²⁸, ¹²⁹ and ¹³⁰ Also, viral respiratory viruses have been found to contaminate fomites and have been a source of HAIs.¹³¹

Microbiologic Factors

The microbiologic factors that contribute to healthcare-associated transmission of infectious disease include virulence factors, the pathogen’s ability to survive in the hospital environment, and antimicrobial resistance. While S. aureus and Pseudomonas aeruginosa are highly virulent pathogens, pathogens of relatively low virulence, such as Acinetobacter spp. can also cause HAIs in immunocompromised or critically ill patients. Many of these bacteria have characteristics (e.g., adhesions, resistance to disinfectants, spore formation) that facilitate their survival in the hospital environment so they can be transmitted to patients via thermometers, ventilators, urinary and vascular catheters, and other fomites. For example, microorganisms such as Pseudomonas, Acinetobacter, Serratia, and Enterobacter species can survive in hospital environments and can be relatively resistant to disinfectants.¹³², ¹³³, ¹³⁴ and ¹³⁵ These pathogens live in water and soil, where they are exposed to antimicrobial substances and may develop an inherent resistance to common antimicrobial agents. In another example, coagulase-negative staphylococci adhere to prosthetic devices and vascular catheters; this has become a major cause of bloodstream infections in patients with foreign bodies, such as prosthetic joints, valves, or permanent central venous catheters.

Extrinsic Factors

The extrinsic factors that contribute to HAIs include medical treatment and interventions, such as placement of invasive devices and operative procedures. The use of chemotherapy may cause immunosuppression and mucosal disruption and allow organisms a port of entry into the host. Equipment, such as dialysis machines or ventilators, may or may not be cleaned or maintained properly, and it may incorporate complicated reservoirs, filters, or mechanisms to prevent backflow—all features that can malfunction and lead to mechanisms of entry for organisms. Once organisms enter the body, infection may ensue. Nasogastric and endotracheal tubes have been shown to increase the risk of acquiring healthcare-associated pneumonia.¹³⁶ Although all of the previously mentioned extrinsic elements contribute to HAIs, the most frequently implicated extrinsic factors are surgical operations and invasive devices.

Another increasingly important factor is the use of antimicrobial agents that can lead to imbalances in the normal symbiotic relationship of organisms in the gastrointestinal (GI) tract, on the skin and other bodily surfaces. When the normal state of human endogenous flora is altered, selective pressure favors antimicrobial-resistant organisms. In fact, patients who received norfloxacin or fluconazole for GI tract decontamination during episodes of neutropenia have been shown to develop resistant organisms as a result of this type of selective pressure.¹³⁷, ¹³⁸ Organisms such as VRE can proliferate when broad-spectrum antimicrobials have killed the normal GI tract gramnegative and anaerobic flora.¹³⁹ Most importantly, C. difficile is an organism that emerges as a result of antibiotics and its disruption of the normal gastrointestinal balance.

ETIOLOGY AND TRANSMISSION

Endogenous versus Exogenous Organisms

There are four potential sources that can transmit microorganisms and lead to infections. Three of these are exogenous sources—fixed structures or the facilities of the hospital, devices or instruments used at the hospital, and healthcare personnel—and one is endogenous—the (source) patients. Exogenous infections are a direct result of pathogenic or nonpathogenic organisms directly acquired from the environment. These infections can be transmitted via the airborne route, through fomites or direct contact with carriers, by ingesting contaminated foods, or by parenteral inoculation. Endogenous source infections are divided into primary or secondary infections. Organisms that are a part of a patient’s normal flora cause primary endogenous infections; organisms that become part of the patient’s flora during the hospital stay cause secondary endogenous infections.¹⁴⁰

Transmission of Microorganisms

Healthcare-associated transmission of organisms can occur by five routes: contact, droplet, airborne, common vehicle, and vector-borne.

Contact

The most frequent route that leads to the development of HAIs is direct or indirect contact.

Direct contact between body surfaces results in the transfer of microorganisms between a susceptible host and a colonized or infected individual. This type of transmission usually requires personal contact. It can occur between patient and HCP or between two patients, with one patient serving as the source of the infectious microorganisms and the other as a susceptible host. An example of such transmission would be transfer of organisms on the HCP’s hands to another patient. In addition, HCPs are significant reservoirs of microorganisms. They can carry potentially infectious organisms on their skin, which can colonize their hands or nails. For example, increases in SSIs due to nasal carriage of S. aureus have been traced to HCPs who carry the organism.¹⁴¹ Furthermore, outbreaks of S. aureus, P. aeruginosa, and Candida species infections have also occurred from colonized or infected HCPs.¹⁴¹, ¹⁴², ¹⁴³, ¹⁴⁴, ¹⁴⁵, ¹⁴⁶, ¹⁴⁷ and ¹⁴⁸ HCPs and family members have been the sources of influenza, group A streptococcus, pertussis, and other organisms that have been transmitted nosocomially.¹⁴⁹, ¹⁵⁰, ¹⁵¹, ¹⁵², ¹⁵³, ¹⁵⁴, ¹⁵⁵, ¹⁵⁶, ¹⁵⁷, ¹⁵⁸, ¹⁵⁹, ¹⁶⁰, ¹⁶¹, ¹⁶², ¹⁶³, ¹⁶⁴, ¹⁶⁵, ¹⁶⁶ and ¹⁶⁷

Indirect-contact transmission involves contact of a susceptible host with a contaminated, usually inanimate object or fomites. Fomites can be medical devices (e.g., resuscitation bags, endotracheal tubes, suction devices, ventilators, endoscopes), instruments (e.g., rectal thermometers, blood pressure cuffs, stethoscopes), supplies such as dressings (especially in burn units), and toys (especially stuffed animals).¹⁶⁸ For example, Hughes and colleagues demonstrated that stuffed teddy bears used in a hand washing promotional campaign were heavily contaminated and colonized by organisms.¹⁶⁹

Droplet

Organisms can be transmitted by respiratory droplets, especially respiratory viruses. Coughing, sneezing, and talking can produce droplets containing microorganisms that can be transmitted as far as 6 feet. Droplets are larger particles greater than 5 µm in size that are propelled a short distance through the air and can settle on the nasal mucosa, conjunctiva, or mouth of the host. Respiratory secretions may contain organisms that can be transmitted to HCPs or other patients while performing such tasks as suctioning, nebulization of medication, or bronchoscopy. Because these droplets are larger and heavier, they can contaminate inanimate surfaces as they land. Droplets transmit most respiratory viruses, pertussis, and meningococcus.⁵³

Airborne

Organisms are also transmitted via air (airborne), by the spread of either the evaporated airborne droplets (5µm or smaller in size) that contain microorganisms or dust particles to which the infectious agent is attached. These small particles can remain suspended in the air for longer periods of time and can be propelled for greater distances than droplets. Transmission can occur in healthcare facilities that experience disruption of ventilation systems or inadequate respiratory protection.¹⁷⁰ Mycobacterium tuberculosis and viruses such as varicella, smallpox, and measles are transmitted in this fashion.⁵³

Common Vehicle

Organisms can be transmitted extrinsically through common vehicles, such as food, water, medications, intravenous fluids, contaminated blood products, and medical equipment or devices. In health care, such infections can manifest as bloodstream infections or diarrhea. Contamination of medications mixed on wards is a common source of infection in the developing world; outbreaks of norovirus, for example, have been transmitted in this way.

Vector-Borne

Vector-borne transmission of microorganisms within hospitals or healthcare settings is unusual in developed countries. Vectors such as mosquitoes, flies, ticks, and others can transmit microorganisms if they are present in the facility.

MAJOR TYPES OF HAIs

HAIs can be classified according to the source and, more specifically, as being associated with either devices or procedures. In the United States, CDC/NHSN has grouped HAIs into 13 major categories to facilitate data analysis and comparison of data.¹ Four types of HAIs and their respective prevention guidelines, as issued by CDC, are discussed in further detail here, as they have been shown to cause the majority of HAIs.⁴, ¹⁷¹, ¹⁷² and ¹⁷³

Urinary Tract Infection

A urinary tract infection is an infection involving any part of the urinary system, including the urethra, bladder, ureters, and kidney. UTIs are the most common type of HAI reported to the NHSN, accounting for more than 30% of HAIs in acute care hospitals and perhaps even more in long-term care facilities.⁴ Among UTIs acquired in the hospitals, at least 75% are associated with a urinary catheter. Approximately 15% to 25% of hospitalized patients receive urinary catheters during their hospital stay. Urinary catheterization not only causes bacteriuria that commonly leads to unnecessary antimicrobial use, but the urinary drainage systems also serve as reservoirs for multidrug-resistant bacteria and as a source of transmission to other patients. Prolonged use of the urinary catheter is the most important risk factor for developing a catheterassociated urinary tract infection. Given this risk, catheters should be inserted only when clinically indicated and should be removed as soon as they are no longer needed. Owing to this practice, it is the rate of UTIs associated with indwelling urinary catheters that is commonly reported to national databases such as NHSN.¹

The pathogens most frequently associated with CA-UTI in hospitals reporting to NHSN between 2006 and 2007 were Escherichia coli (21.4%) and Candida spp. (21.0%), followed by Enterococcus spp. (14.9%), P. aeruginosa (10.0%), Klebsiella pneumoniae (7.7%), and Enterobacter spp. (4.1%).² An increasing proportion of these organisms are becoming resistant to multiple antimicrobial agents.¹⁷⁴, ¹⁷⁵ CA-UTIs have been shown to be associated with increased morbidity, mortality, hospital cost, and length of stay.¹⁷⁶, ¹⁷⁷ and ¹⁷⁸ Although morbidity and mortality from CA-UTIs are considered to be relatively low compared to other HAIs, such as central line-associated bloodstream infections and lower respiratory tract infections, the high prevalence of urinary catheter use leads to a large cumulative burden of infections. The CDC has estimated that as many as 139,000 hospital-onset, symptomatic CAUTIs occurred in 2007, resulting in as much as $131 million in excess direct medical costs.¹⁷⁹

In view of the magnitude of this problem, in the United States, the Department of Health and Human Services (DHHS) developed an action plan to prevent HAIs, with a 5-year national prevention target of a 25% decrease from baseline for CA-UTI rates as measured by the NHSN. The CDC updated its guideline for prevention of CA-UTI in 2009 to put greater emphasis on the following aspects of care:

Appropriate urinary catheter use
Proper techniques for urinary catheter insertion
Proper techniques for urinary catheter maintenance
Quality improvement programs
Administrative infrastructure
Surveillance

An estimated 17% to 69% of CA-UTIs may be preventable with these infection control measures.¹⁸⁰ In fact, with the introduction of more CA-UTI prevention efforts and improved general infection prevention strategies in recent years, a pervasive and sustained decline in the incidence rates of symptomatic UTI and asymptomatic bacteriuria was observed in all major adult ICU types reported under the prior National Nosocomial Infections Surveillance (NNIS) and current NHSN system.¹⁷⁹, ¹⁸⁰

Bloodstream Infections

Healthcare-associated bloodstream infections (BSIs) develop in more than 250,000 patients in U.S. hospitals every year, with an estimated attributable mortality of 12% to 25% for each infection.¹⁸¹, ¹⁸², ¹⁸³ and ¹⁸⁴ The marginal cost to the healthcare system is approximately $25,000 per episode.¹⁸⁵ Most BSIs result from a secondary source such as a postoperative wound or intra-abdominal infection, urinary tract infection, or pneumonia. Primary bacteremia occurs when the source is unrecognized; most of these
cases are related to exposure to intravascular devices (IVDs).¹⁸⁴, ¹⁸⁶ A vast variety of IVDs are employed in modern medicine. For example, a catheter can be designated by the type of vessel it occupies; its intended life span; its site of insertion; its pathway from skin to vessel; its physical length; its function (dialysis, pressure measurement); or some special character of the catheter, each of which is associated with a different infection risk.¹⁸³

The terminology used to define intravascular catheter-related infections can be confusing. The terms “catheter-related bloodstream infection” (CR-BSI) and “central line-associated bloodstream infection” (CLA-BSI) are most widely used in the literature, sometimes interchangeably. In fact, CR-BSI is a clinical definition that requires specific laboratory testing (e.g., quantitative blood cultures of differential time to positivity) to help identify the catheter as the source of the BSI. In contrast, the term CLA-BSI is typically used for surveillance purposes and is defined as a primary BSI in a patient who had a central line within the 48-hour period before the development of the BSI and is not bloodstream related to an infection at another site.¹

The most common organisms causing healthcare-associated BSIs are coagulase-negative staphylococci (CoNS; 31% of isolates), Staphylococcus aureus (20%), enterococci (9%), and Candida species (9%). CoNS, Pseudomonas spp., Enterobacter spp., Serratia spp., and Acinetobacter spp. are more likely to cause infections in patients in ICUs. Gram-negative bacilli accounted for one-fifth of CLA-BSIs reported to CDC and the Surveillance and Control of Pathogens of Epidemiological Importance (SCOPE) database. The proportion of S. aureus isolates with methicillin resistance increased from 22% in 1995 to 57% in 2001. Similar to the trend observed in CA-UTIs, the proportion of healthcare-associated BSIs due to multidrug-resistant organisms is increasing in U.S. hospitals.¹⁸⁷

With the implementation of different prevention strategies in a variety of clinical settings, studies have consistently shown a declining incidence rate of CLA-BSI in the United States.¹⁸⁸, ¹⁸⁹, ¹⁹⁰ and ¹⁹¹ In 2011, the CDC updated its guideline for prevention of intravascular catheter-related infections to emphasize performance improvement by implementing bundled prevention strategies¹⁸³:

Educating and training healthcare personnel who insert and maintain catheters
Using maximal sterile barrier precautions during central venous catheter insertion
Using a greater than 0.5% strength chlorhexidine skin preparation with alcohol for antisepsis
Avoiding routine replacement of central venous catheters as a strategy to prevent infection
Using antiseptic/antibiotic-impregnated short-term central venous catheters and chlorhexidine-impregnated sponge dressings if the rate of infection is not decreasing despite adherence to the other strategies

The CDC guideline also emphasizes the importance of documenting and reporting rates of compliance with all components of the bundle as benchmarks for quality and performance improvement.¹⁸³ Since its publication, other, less expensive strategies such as chlorhexidine bathing have emerged as promising approaches.¹⁹², ¹⁹³, ¹⁹⁴ and ¹⁹⁵

Healthcare-Associated Pneumonia and Ventilator-Associated Pneumonia

Healthcare-associated pneumonia (HCAP) is the leading cause of death among hospitalized patients.¹⁹⁶, ¹⁹⁷ Pneumonia is classified as HCAP when it occurs in any patient who (1) was hospitalized in an acute care hospital for 2 or more days within 90 days of the infection; (2) resides in a nursing home or long-term care facility; (3) received home intravenous antimicrobial therapy, chemotherapy, or home wound care within the past 30 days of the current infection; or (4) attended a hospital or hemodialysis clinic. HCAP may be further categorized as either early onset, in which pneumonia occurs during the first 4 days of hospitalization, and late onset, in which the disease occurs at least 4 days after admission.¹ HCAP typically increases the length of ICU stay from 5 to 7 days.¹⁹⁸

Ventilator-associated pneumonia (VAP) arises in patients who have a device to assist or control respiration continuously through endotracheal tube or tracheostomy. This device must be present within the 48-hour period before the onset of infection.¹ A meta-analysis demonstrated that the incidence of VAP among patients receiving mechanical ventilation was 9.7% (95% CI: 7.0 -12.5) and incidence among patients in medical ICUs was 17.0% (95% CI: 5.9-28.0).¹⁹⁸ The additional cost for treating VAP was as high as $13,467 (in the year 2003).¹⁹⁸

When microorganisms enter into and colonize the lower respiratory tract, HCAP can develop.¹⁹⁷ Under suitable circumstances, the pathogen can invade the mucosal and other host defenses, leading to disease.¹⁹⁷ Aspiration of oropharyngeal secretions
plays an important role for developing HCAP. Any conditions that precipitate, aggravate, or influence the volume or severity of aspiration can contribute higher risk for infection—for example, supine position, swallowing difficulty, gastric overextension, depressed sensorium, and enteral feeding.¹⁹⁷, ¹⁹⁹, ²⁰⁰ Proton pump inhibitors (PPIs), which are commonly used among hospitalized patients for stress ulcer prophylaxis, may increase the risk of pneumonia.²⁰¹, ²⁰², ²⁰³, ²⁰⁴, ²⁰⁵ and ²⁰⁶ One hypothesis suggests that this agent decreases gastric acidity, an innate aspect of immunity, and then produces bacterial overgrowth.²⁰⁷ Oropharyngeal colonization with multidrug-resistant organisms (MDROs) increases risk of pulmonary infection with these organisms. Risk factors for infection with MDROs have been extensively studied and include the following:

Receiving antimicrobial treatment within the preceding 90 days¹⁹⁷, ²⁰⁸, ²⁰⁹, ²¹⁰, ²¹¹ and ²¹²
Current hospitalization of at least 5 days
Previous hospitalization in the preceding 90 days for at least 2 days
Immunosuppressive state
High frequency of antimicrobial resistance in the community or in the specific unit
The presence of risk factors for HCAP
Residence in a nursing home or extended care facility
Home infusion therapy (including antibiotics)
Chronic dialysis within 30 days
Home wound care
A family member with multidrug-resistant pathogen¹⁹⁷

Quasi-experimental studies have correlated the initiation of treatment with an inadequate spectrum of antimicrobial agent with increased mortality in HCAP/VAP patients.²¹³, ²¹⁴, ²¹⁵ and ²¹⁶

Intubation is the single most important risk factor: it increases the risk of developing pneumonia and VAP at least sixfold and increases mortality by 55%.¹⁹⁷, ²¹⁷, ²¹⁸ Once the first line of host defenses is breached by intubation and placement of an endotracheal tube, patients tend to aspirate organisms that colonize the oropharynx or upper gastrointestinal tract. Risk factors for developing VAP may be divided into those that can be modified and those that cannot be modified.¹⁹⁷ Nonmodifiable risk factors include advanced age, male gender, preexisting pulmonary disease, and severe underlying illness.¹⁹⁷, ²¹⁹, ²²⁰ and ²²¹ Modifiable risk factors include contaminated ventilator circuits, an endotracheal cuff with low pressure, multiple patient transfers, supine position of the patient, and gastric overextension.¹⁹⁷, ²²⁰, ²²¹

HCAP and VAP are mostly commonly caused by bacteria (multiple organisms), but can also be caused by fungi and viruses.¹⁹⁷ The most likely causative organism can be altered by the host’s immune status.¹⁹⁷ The most common pathogens causing HCAP are S. aureus, Streptococcus pneumoniae, P. aeruginosa, and Haemophilus influenzae.²⁰⁸, ²⁰⁹, ²¹¹, ²²², ²²³ The pathogens causing early-onset HCAP resemble those organisms associated with community-acquired pneumonia, such as S. pneumoniae, Moraxella catarrhalis, and H. influenzae. Similarly, gram-negative bacilli or S. aureus, including MRSA, fungi, and Legionella spp., occur in late-onset HCAP.¹ Worldwide, the pathogens causing VAP are predominately gram negative, consisting of P. aeruginosa, Acinetobacter baumannii, and Klebsiella spp.²²⁴, ²²⁵, ²²⁶, ²²⁷, ²²⁸, ²²⁹, ²³⁰, ²³¹, ²³², ²³³ and ²³⁴ Compared with patients having HCAP, the prevalence of gram-negative pathogens is higher among VAP patients, while S. aureus is more common in patients with HCAP.²³⁵ Respiratory viruses can cause pneumonia at any time of hospitalization and are likely underappreciated as a cause of HCAP and VAP.¹ HCAP among immunocompromised hosts can be caused by inhalation of aerosols or droplets contaminated with Legionella species, Aspergillus species and other molds, respiratory syncytial virus (RSV), influenza virus, and other respiratory viruses.²³⁶, ²³⁷, ²³⁸, ²³⁹, ²⁴⁰ and ²⁴¹

Outbreaks of HCAP/VAP have been documented in numerous healthcare settings. Organisms associated with these outbreaks include influenza,²⁴², ²⁴³, ²⁴⁴ and ²⁴⁵ RSV,²⁴⁶, ²⁴⁷, ²⁴⁸ and ²⁴⁹ human metapneumovirus,²⁵⁰, ²⁵¹ adenovirus,²⁵² measles,²⁵³ herpes simplex virus (HSV),²⁵⁴ P. aeruginosa,²⁵⁵, ²⁵⁶, ²⁵⁷, ²⁵⁸ and ²⁵⁹ A. baummannii,²⁶⁰, ²⁶¹ and ²⁶² Enterobacter cloacae,²⁶³, ²⁶⁴ K. pneumoniae,²⁶⁵, ²⁶⁶ B. cepacia,²⁶⁷, ²⁶⁸ Serratia marcescens,²⁶⁹ M. catarrhalis,²⁷⁰ Neisseria meningitides,²⁷¹, ²⁷² Flavobacterium meningosepticum,²⁷³ S. aureus,²⁵⁷, ²⁷⁴, ²⁷⁵ and ²⁷⁶ S. pneumoniae,²⁷⁷, ²⁷⁸ Legionella spp.,⁹⁵, ⁹⁷, ¹⁰⁸, ¹¹⁵, ¹²⁰, ²⁷⁹, ²⁸⁰, ²⁸¹, ²⁸², ²⁸³, ²⁸⁴, ²⁸⁵ and ²⁸⁶ and Pneumocystis jerovecii.²⁸⁷, ²⁸⁸ and ²⁸⁹ Generally, in the setting of problems with ventilation or construction, Aspergillus spp. and Mycobacterium tuberculosis can also be associated with clusters in healthcare facilities (as discussed later in this chapter). The bacterial pathogens that cause HCAP or VAP may be multidrug resistant; in quasi-experimental studies, when an inadequate spectrum of antimicrobial agent was initiated as empiric therapy for HCAP/VAP, the mortality was increased.²¹³, ²¹⁴, ²¹⁵ and ²¹⁶

Many interventions to decrease HCAP/VAP have been investigated. In 2003, CDC published “Guidelines for Preventing Health Care-Associated Pneumonia.” These guidelines outline the importance of (1) environmental controls to prevented invasive pulmonary aspergillosis and Legionellosis,
(2) standard precautions, (3) immunizations, (4) measures to prevent aspiration, and (5) surveillance.²²¹

In 2008, Society for Hospital Epidemiology of America (SHEA) and Infectious Diseases Society of America (IDSA) published “Strategies to Prevent Ventilator-Associated Pneumonia in Acute Care Hospitals” in 2008.²⁹⁰ To prevent VAP, these guidelines recommended interventions to decrease aspiration of secretions, including raising the head of bed, avoiding gastric overdistension, avoiding unintended extubation and reintubation, using a cuffed endotracheal tube with in-line or subglottic suctioning, and maintaining an endotracheal cuff pressure of at least 20 cm H₂O.²⁹⁰ Furthermore, orotracheal intubation is preferable to nasotracheal intubation, as the latter may increase risk of sinusitis and VAP.²⁹⁰ Three metaanalyses demonstrated the benefits of performing oral decontamination with antiseptic solution and favored chlorhexidine as the antiseptic of choice, although the impact was seen primarily in patients in cardiosurgical intensive care units.²⁹¹, ²⁹² However, an impact on mortality has not been demonstrated from this intervention.²⁹¹, ²⁹³ It remains a standard of care to perform regular oral care with an antiseptic solution in mechanically ventilated patients.²⁹⁰

Surveillance for HCAP and VAP and process measures should be conducted in patients deemed at high risk for infection. The use of standard definitions is an important component of diagnosis of VAP, especially in ICU patients who have many risks for fever and pulmonary infiltrates. The VAP rate should be calculated by dividing the number of VAP cases by the number of ventilator-days.

Surgical Site Infection

Surgical site infection causes approximately onefourth of all HAIs, making it a major HAI especially among surgical patients. In the United States, SSI is the second or third most frequent HAI, responsible for 29% of all HAIs and 14% of all healthcare-associated adverse events.¹⁸⁰ As many as 5% of surgical patients develop SSIs.²⁹⁴ This rate may be higher in patients cared for in developing countries. SSIs cause significant morbidity and account for 55% of all the extra hospital days attributed to HAIs.²⁹⁵ Such infections increase length of hospital stay, readmission rates, and medical costs.²⁹⁶, ²⁹⁷, ²⁹⁸, ²⁹⁹, ³⁰⁰, ³⁰¹, ³⁰², ³⁰³, ³⁰⁴ and ³⁰⁵ Patients with an SSI have a 2 to 11 times higher risk of death, compared to surgical patients without an SSI.³⁰⁶ These numbers highlight the tremendous burden of these infections and the importance of preventing them. Because of the frequency, morbidity, mortality, and economic burden of these infections, SSIs are given the high priority for surveillance.

The definition of an SSI has been debated for years. For the purposes of surveillance, a reasonable definition should be highly sensitive but must balance specificity. It should be applied systematically and used to analyze rates, examine risk factors, develop prevention strategies, and assess trends over time. Ideally, the definition chosen should remain unchanged to facilitate comparisons of data through the years and to determine whether interventions implemented reduce these rates. In 1992, a consensus group that included CDC, SHEA, and the Surgical Infection Society (SIS) modified how SSI was defined and changed the name from “surgical wound infection” to “surgical site infection.”³⁰⁷, ³⁰⁸ The definition for NHSN is currently being revised.

SSIs are divided into incisional and organ-space SSIs. Incisional SSIs are further classified as involving only the skin and subcutaneous tissue (superficial incisional SSIs) or involving deep soft tissues of the incision (deep incisional SSIs). Superficial infections require that at least one of the following occur within 30 days of the operation³⁰⁹:

Pus appears from the incision
Organisms are isolated aseptically from cultured fluid or tissue
At least one of the following signs and symptoms: pain or tenderness, localized swelling, redness, or heat when the surgeon deliberately opens the surgical incision
The surgeon or attending physician diagnoses the infection

SSI secondary to prosthetic devices or foreign body can occur up to one year after the operation.

Only 33% to 67% of infected wounds are cultured, limiting the use of microbiologic case finding as a single case-finding strategy.³¹⁰ Direct inoculation is the most common pathway by which organisms cause infection. Endogenous flora can cause infections, especially those involving S. aureus (including MRSA). Carriers of S. aureus have a 2- to 14-fold increased risk of infection.³¹¹, ³¹² Approximately 85% of S. aureus infections are attributable to a carrier state, a fact that has led to the development of decolonization strategies including the use of intranasal mupirocin and chlorhexidine bathing.³¹², ³¹³ Additionally, SSIs may result from exogenous sources—for example, a carrier healthcare worker (e.g., S. aureus and group A streptococci) contaminated equipment, and a defect in the ventilation system of the operative theater.¹⁴⁹, ³¹⁴, ³¹⁵, ³¹⁶ and ³¹⁷ In those cases, the most common causes of SSI are S. aureus, enterococci, and coagulase-negative staphylococci. However, the infecting organisms vary according to the site of the surgical
incision.³⁰⁹, ³¹⁸ For operations that involve only skin or a clean wound, the most common pathogens are the skin flora (e.g., S. aureus, coagulase-negative staphylococci), while gram-negative and anaerobic organisms are often found in patients who undergo operations that involve the bowel, such as colon surgery.³⁰⁹ Table 14-4 shows the classification of surgical site infections.

Table 14-4 Surgical Site (Wound) Classification

Wound Class	Type of Wound	Definition	Example
I	Clean	Uninfected operative wound No inflammation is encountered No involvement of the respiratory, alimentary, genital, or uninfected urinary tract Closed and, if necessary, drained with closed drainage Operative incisional wounds that follow nonpenetrating (blunt) trauma (those meeting the criteria)	Breast surgery Vascular surgery (which is not infected) Non-infected eye surgery Splenectomy Most orthopedic surgery Lymph node biopsy/excision Elective cesarean section with no premature rupture of membranes
II	Clean-contaminated	Operative wound that enters the respiratory, alimentary, genital, or urinary tract under controlled conditions without unusual contamination No evidence of infection or major break in technique	GI surgery, including laparoscopy Cholecystectomy Vaginal hysterectomy Head and neck surgery
III	Contaminated	Open, fresh, accidental wounds Operations with major breaks in sterile technique	Open cardiac massage Gross spillage from the gastrointestinal tract Incisions involving acute inflammation
IV	Dirty/infected	Old traumatic wounds with retained devitalized tissue involving an existing clinical infection or perforated viscus	Debridement of infectious tissue
Adapted from Berard F, Gandon J. Postoperative wound infections: the influence of ultraviolet irradiation of the operating room and of various other factors. Ann Surg. 1964;160(suppl 2):1-192.

Outbreaks of SSIs are rarely associated with HCPs (Table 14-5). For example, S. pyogenes (group A streptococci) is occasionally reported to cause SSIs, most commonly spread from asymptomatic individuals including HCPs.¹⁴⁹, ¹⁵⁰, ¹⁵¹, ¹⁵², ¹⁵³, ¹⁵⁴, ¹⁵⁵, ¹⁵⁶, ¹⁵⁷ and ¹⁵⁸, ³²⁰, ³²¹ In this case a single case should prompt increased surveillance and some investigation. Similarly, HCPs can be carriers of S. aureus and an investigation of HCP is indicated when fingerprinting of strains reveals they are clonal and suggests a single source. Outbreaks of P. aeruginosa have been caused by surgical staffs with onychomycosis.¹⁴⁴, ¹⁴⁷

Risk factors that increase a patient’s risk of developing an SSI can be categorized as intrinsic (patient related) or extrinsic (operation-related).³⁰⁶ The patient’s risk may increase if he or she has more than one risk factor. Patient characteristics, including the underlying medical condition(s) and the severity of the underlying disease, can increase the risk of developing an SSI. To date, very few controlled studies have examined the relative importance of these risk factors. Those additional risk factors that definitely increase the risk of developing an SSI include morbid obesity, smoking, old age, diabetes mellitus, a prolonged preoperative hospital stay, or a preoperative infection.⁸³, ³⁰⁶, ³⁰⁹, ³²⁹, ³³⁰ and ³³¹ Several studies have shown that patients who are malnourished, have low albumin, have cancer, or are receiving immunosuppressive therapy are at increased risk of developing an SSI.³⁰⁶, ³³², ³³³ and ³³⁴ Unfortunately, it is frequently difficult to ameliorate these risks. If the patient’s surgery is not urgent, and the health condition, such as diabetes, can be controlled or the patient surgery can be delayed, then a delay until the patient’s health improves can improve his or her outcome.

Certain practices and procedures that occur in the operating room or during the perioperative period can increase the risk of SSI. In some cases, HCPs can alter their practices and potentially the patient’s outcome. Those surgical and environmental factors that may
alter the patient’s risk of developing an SSI include a wound classified as contaminated or dirty (Table 14-4), a surgical procedure involving the abdomen, a long operative time, hair removal by razor (especially the night before surgery, so that preoperative colonization occurs), multiple surgical procedures, poor hemostasis during the procedure with blood loss and the need for transfusions, prolonged presence of drains, a lessskilled or inexperienced surgeon, and a low intraoperative body temperature.³³⁵, ³³⁶, ³³⁷, ³³⁸, ³³⁹, ³⁴⁰ and ³⁴¹ Risk factors for SSIs are summarized in Table 14-6.

Table 14-5 Examples of Outbreaks of SSIs: Relationships Between Organisms and Sources

Organisms	Sources	Reference
S. pyogenes	HCP (throat, vaginal, rectal, and skin colonization)	149-158, 320, 321
P. aeruginosa	HCP with onychomycosis	144, 147
MRSA	HCP (nasal, sinus, skin colonization)	145, 146
S. epidemidis	HCP	322
R. bronchialis	HCP’s dog with fingernail contamination	323
P. mirabilis	Contaminated instruments	314
P. aeruginosa	Contaminated automated equipment	324
Aspergillus spp.	Defect in ventilation system of operating rooms	315
Nontuberculous mycobacterium	Contaminated instruments	316, 325
Nontuberculous mycobacterium	Contaminated skin marking	326, 327
Legionella spp.	Wound contaminated with tap water in postoperative period	328

One of the most important interventions—if not the most important single intervention—is the appropriate use and timing of perioperative antimicrobial prophylaxis.³⁵³ Pathogens that infect surgical sites during surgical procedures can be acquired from the patient, the hospital environment, or HCPs. The patient’s endogenous flora are responsible for most infections. In surgical procedures involving the gastrointestinal, respiratory, genital, and urinary tracts, antimicrobial prophylaxis has proved efficacious in reducing SSIs. For many cardiac, neurosurgical, and orthopedic procedures, prophylaxis decreases the risk of developing an SSI. For most clean surgical procedures, antimicrobial prophylaxis remains controversial, although experts suggest giving patients a prophylactic antimicrobial agent when a foreignbody implant is inserted or if an infection occurring in a clean procedure would be associated with severe or life-threatening consequences (e.g., valve replacement).³⁰⁶ Finally, if the perioperative antimicrobial agent is administered after the incision, the risk of an SSI increases sixfold above the risk when it is administered 30 minutes to 1 hour (2 hours is allowed
for vancomycin and fluoroquinolones) before the incision.³⁰⁶ Antimicrobial prophylaxis should be stopped within 24 hours after the surgery for most procedures.³⁰⁶

Table 14-6 Summary of Risk Factors for Surgical Site Infections⁸³, ³⁰⁶, ³⁰⁹, ³²⁹, ³³⁰, ³³¹, ³³², ³³³, ³³⁴, ³³⁵, ³³⁶, ³³⁷, ³³⁸, ³³⁹, ³⁴⁰, ³⁴¹, ³⁴², ³⁴³, ³⁴⁴, ³⁴⁵, ³⁴⁶, ³⁴⁷, ³⁴⁸, ³⁴⁹, ³⁵⁰ and ³⁵¹

Patient Related		Operation Related
Modifiable	Nonmodifiable
Diabetes, poor glucose control Obesity Smoking Immunosuppressive agents Malnutrition Low albumin Anemia HIV (CD4 < 200/mm³) S. aureus colonization Poor oral hygiene Preoperative hospital stay Preoperative infection Poor hemostasis	Age ASA^* score	Wound class Hair-removal Preoperative infections Operative time Open surgery (compared to laparoscopic surgery) Surgeon skill Hypoxia Hypothermia Blood transfusion Prolonged presence of a drain Intra-operative administered fraction of inspired oxygen³⁰⁹
^*ASA: American Society of Anesthesiologists.

The patient’s skin is a major source of intraoperative contamination and organisms. Resident flora such as coagulase-negative staphylococci can be found in the deeper layers of the dermis; hence, cleaning and disinfecting skin before operation by antiseptic reduces the risk of SSIs. A recent randomized study demonstrated the benefit of using a chlorhexidine-alcohol based preparation for skin antisepsis over povidone-iodine in patients undergoing clean-contaminated operations.³⁵⁴

Preoperative cleaning of the operative team’s hands is equally important and is performed by the operative team. Recent studies have demonstrated that the timeconsuming hand scrubbing with brushes for 10 minutes did not decrease SSI rate and that a shorter scrub or hand rubbing by aqueous alcohol solutions is adequate.³⁵⁵ Scrubbing with brushes may damage the HCP’s skin and result in increased shedding of bacteria.³⁵⁶

The operating facilities should be designed to promote best cleaning and disinfection practices, provide appropriate air handling, and assure that HCPs can follow best practices. The operating rooms should have a semi-sterile and sterile core, and the operating room itself must have higher pressure than the outside corridor (positive pressure). The doors to the operating room suite should remain closed at all times and the traffic in the room should be minimized.³⁵⁷ Pass-throughs should be designed to minimize the traffic even further. Most exogenous wound contamination occurs during the operation through contact or airborne transmission of organisms; hence, events occurring after the operation (e.g., ward dressings and isolation techniques) are less likely to contribute to SSIs.

Table 14-7 Percentage of SSIs Detected After Hospital Discharge

Author (Year)		Patient Population	SSIs Detected After Discharged (%)
Roy (1994)³⁶⁰; Avato (2002)³⁶⁴;		Coronary artery bypass surgery	52-95
	Noman (2011)³⁶⁵
Mannian (2011)³⁴⁷		Cardiothoracic	61
Wójkowska-Mach (2008)³⁶⁶		Total knee or hip arthroplasty	59
Huotari (2006)³⁶⁷		Orthopedic surgery	28-90
Huenger (2005)³⁶⁸		Hip arthroplasty	25
Friedman (2001)³⁶⁹		Total knee arthroplasty	72
Sands (1996)³⁵⁹		Non-obstetrical	84
Simchen (1992)³⁷⁰		Herniography	51
Hulton (1992)³⁷¹; Mitt (2005)³⁷²;		Cesarean section	42-95.4
	Cardoso del Monte (2010)³⁷³
Weigelt (1992)³⁷⁴		General surgery	35
Law (1990)³⁷⁵		Elective procedures	59
Reimer (1987)³⁷⁶; Manian (1990)²⁵⁶;		All procedures	2-73.7
	Olson (1990)³³⁴
Oliveira (2004)³⁷⁸; Prospero (2006)³⁷⁹;		Appendectomy	71
	Bilimoria (2010)³⁶³
Brown (1987)³⁶²		Major procedures	46

Finally, the importance of transparency and sharing SSI rates with surgeons, operating room staff, and leadership cannot be overemphasized. Reporting of surgeon-specific SSI rates can also have a role, although their use for credentialing purposes should be carefully considered. Programs that perform surveillance for SSIs may reduce these rates as much as 35% by reporting the rates associated with specific surgical teams.³⁰⁸ To save resources, some programs have adopted methods that group the entire surgical population into patients who are at a higher or lower risk of developing SSIs. In risk stratification, procedures are classified and reported using a risk index. The NNIS risk index—the most commonly used stratification process—groups procedures into three categories.³⁵⁸

One of the major challenges in infection prevention and control is deciding when to include postdischarge surveillance for detecting SSIs. This strategy is important, as these infections can develop after the patient has been discharged home. Routine surveillance processes may not capture such SSIs (Table 14-7).³¹⁰, ³⁵⁸, ³⁵⁹, ³⁶⁰, ³⁶¹, ³⁶² and ³⁶³ Thus surveying surgeons’ offices to monitor each patient for a
defined period after his or her surgery should be incorporated into surveillance strategies. Telephone surveys and questionnaires sent to patients and/or physicians have been used as well. Which method provides the most accurate data has not yet been determined.³⁰⁸ While more difficult to perform, surveys that cover the postdischarge time period are important, as those SSIs that develop after either clean or clean-contaminated procedures are the most likely to occur after hospital discharge.

EPIDEMIOLOGICALLY IMPORTANT PATHOGENS AND EMERGING PATHOGENS

Bacteria

Methicillin-Resistant Staphylococcus aureus

Methicillin-resistant Staphylococcus aureus is resistant to many beta-lactam antibiotics.³⁸¹ Resistance to methicillin and other semi-synthetic penicillins was identified soon after this class of antibiotic was introduced into clinical care in the 1960s.³⁸² The first outbreak of MRSA in the United States was documented in late 1960s⁴²; since then, MRSA organisms have become common gram-positive organisms, responsible for HAIs in many countries around the world.³⁸³, ³⁸⁴, ³⁸⁵, ³⁸⁶, ³⁸⁷ and ³⁸⁸ The prevalence of MRSA varies in different parts of the world, however, and reflects the approach to antimicrobial stewardship and infection prevention practices.³⁸⁹, ³⁹⁰ Furthermore, MRSA has emerged in several continents as a zoonotic infection, and persons who contact animals may at increased risk for MRSA acquisition as compared to the general population.³⁹¹, ³⁹², ³⁹³, ³⁹⁴ and ³⁹⁵

Infections with MRSA are associated with morbidity and mortality. Patients with MRSA bacteremia have higher mortality rate compare with patients with methicillin-sensitive S. aureus bacteremia.³⁹⁶, ³⁹⁷ However, this higher mortality rate may not result from increased virulence of resistant strains, but rather from other factors, including delays in the initiation of effective antimicrobial treatment, lesseffective antimicrobial treatment, and higher severity of underlying illness among persons with antimicrobial-resistant strains.³⁹⁷

Traditionally MRSA was considered to be a healthcare-associated organism, but in the 1990s it was reported to be transmitted in previously healthy patients without a history of hospitalization—essentially members of the community.³⁹⁸, ³⁹⁹ The so-called community-associated MRSA (CA-MRSA) strains were less frequently resistant to other antibiotics than healthcare-associated MRSA strains (HA-MRSA).³⁸² Both strains contain a gene coding for methicillin resistance, the mec-A gene. The mec-A gene is located on a mobile genetic element called the staphylococcal chromosome cassette (SCC mec); this cassette differs between CA-MRSA and HA-MRSA strains.³⁸² In addition, CA-MRSA strains commonly carry a gene coding for a toxin, with the most frequently discussed being the Panton-Valentine leukocidin (PVL) toxin that causes skin and soft-tissue necrosis.³⁸² Furthermore, phenotypically MRSA organisms are divided into eight distinct clusters identified by pulsed-field gel electrophoresis—that is, USA 100 through USA 800.³⁸² Strains USA 300 and USA 400 are classified as “community associated,” and the others are classified as HA-MRSA.⁴⁰⁰ The USA 300 strain was first reported in the United States and later spread to other parts of the world.⁴⁰¹, ⁴⁰² Recent epidemiologic studies indicate that USA 300 (HR-MRSA) strains have entered into healthcare facilities and are commonly transmitted to patients as an HAI.⁴⁰³, ⁴⁰⁴

Thus the original designation of healthcare- versus community-associated MRSA is no longer as useful for differentiating the source of acquisition. For the purpose of surveillance, SHEA has proposed definitions that use time-based elements instead.³⁹⁷ The terms “hospital-onset” and “community-onset” MRSA have been identified based on the time of specimen collection.³⁹⁷ If MRSA is identified from a specimen obtained after the third day of hospitalization, this isolate is classified as “hospital-onset MRSA.” Isolates identified before the third day of a patient’s hospitalization are classified as “community-onset MRSA.”³⁹⁷

S. aureus has a unique epidemiology that drives some of the prevention practices in healthcare settings. The nares and skin are the ecologic niches of the organism.⁴⁰⁵, ⁴⁰⁶ Controversy exists on whether the nares alone or the nares with additional sites (throat, perirectal, umbilical, or axilla/groin) are needed to determine colonization. Nonetheless, all agree that higher rates of nasal colonization with MRSA are seen in patients with diabetes mellitus, patients receiving hemodialysis, HIV-infected patients, and intravenous drug users.⁴⁰⁷ MRSA colonization is the most important risk factor for infection⁴⁰⁸, ⁴⁰⁹ and ⁴¹⁰; the risk of developing an infection within 18 months after detection of MRSA colonization is as high as 29%.⁴¹¹ Other risk factors include antimicrobial use, prolonged hospitalization, recent hospitalization or surgery, presence of foreign bodies, residence in a long-term facility, and frequent contact with the healthcare system or HCPs.³⁹⁷, ⁴¹², ⁴¹³, ⁴¹⁴, ⁴¹⁵, ⁴¹⁶ and ⁴¹⁷ Colonization pressure or the amount of a resistant organism on the unit is also an independent risk factor for MRSA acquisition.⁴¹⁸, ⁴¹⁹ and ⁴²⁰

Transmission of MRSA is associated with a complex web of events. First, multiple sites harbor the organism; thus, in a person with nasal colonization, the hands, fingers, and skin are also commonly colonized with the same MRSA strain.⁴²¹, ⁴²² Second, MRSA can spread by the contaminated HCP’s hands after being acquired by direct contact with the MRSA-colonized patient or the MRSA-contaminated environment.⁴²³, ⁴²⁴ Third, HCPs are rarely the source of MRSA. A meta-analysis demonstrated prevalence of nasal colonization among HCPs was 4.6% (95% CI: 1.0-8.2%),⁴²⁵ yet the number of outbreaks reported where HCPs were the source of MRSA is very small.⁴²⁵, ⁴²⁶ and ⁴²⁷ Several CA-MRSA outbreaks in healthcare settings have been reported in which healthy HCPs have been sickened.⁴²⁸, ⁴²⁹, ⁴³⁰, ⁴³¹, ⁴³², ⁴³³, ⁴³⁴, ⁴³⁵, ⁴³⁶, ⁴³⁷, ⁴³⁸, ⁴³⁹, ⁴⁴⁰, ⁴⁴¹ and ⁴⁴² Most of these outbreaks occurred in neonatal and maternal units.

The increased use of vancomycin to treat MRSA has contributed to the emergence of vancomycin-resistant organisms such as VRE and vancomycin (or glycopeptide)-intermediate S. aureus (VISA/GISA). GISA is thought to result from prolonged use of a glycopeptide antibiotic that leads to changes in cellwall synthesis and decreases affinity for agents such as vancomycin. In contrast, vancomycin-resistant S. aureus (VRSA) strains have resulted from the acquisition of genetic material from enterococci, not antimicrobial pressure.⁴⁴³, ⁴⁴⁴ and ⁴⁴⁵ In the seven cases of VRSA were identified in the United States between 2002 and 2006, all of the patients had a history of prior MRSA and enterococcal infection/colonization.⁴⁴⁶ Risk factors for VRSA include prior MRSA infection/colonization and antecedent vancomycin use.⁴⁴⁷, ⁴⁴⁸ and ⁴⁴⁹

MRSA is a major problem globally and in most healthcare systems in North America.⁴⁵⁰ Several approaches for its prevention have been developed and have proved successful, yet many issues remain unresolved.³¹² Basic infection prevention measures such as hand hygiene, transmission-based precautions, and environmental cleaning should be applied to patients with MRSA colonization or infection. These patients should also be placed on contact precautions.⁵³ The duration of contact precautions necessary for patients with MRSA is disputed.³⁹⁷ The patient’s environment should have adequate cleaning and disinfection with U.S. Environmental Protection Agency (EPA)-registered hospital disinfectants. A hand hygiene compliance program and antimicrobial control program should be implemented.

Active surveillance screening for MRSA in highrisk (or sometimes all) patients is advocated by some sources.⁴⁵¹, ⁴⁵² Others advocate universal screening of asymptomatic patients and isolation of those who grow the organism, an approach known as the “search and isolate” strategy. This strategy seems to be more useful in an outbreak setting and for patients within certain high-risk groups.⁴⁵³, ⁴⁵⁴, ⁴⁵⁵, ⁴⁵⁶, ⁴⁵⁷, ⁴⁵⁸ and ⁴⁵⁹ The role of universal screening is one of the most controversial issues and has not yet been determined.³⁹⁷, ⁴⁵¹, ⁴⁵², ⁴⁶⁰, ⁴⁶¹ and ⁴⁶² However, several recent U.S. studies do support the widespread use of active screening and decolonization with chlorhexidine in areas where MRSA is endemic.⁴⁵¹, ⁴⁵² In North America, screening of HCPs for MRSA should be considered in an outbreak setting if they are epidemiologically linked to a cluster of MRSA infections.³⁹⁷

Europeans have adapted a variety of approaches to preventing MRSA. The French have demonstrated the impact of a countrywide hand hygiene campaign on MRSA nationally and shown dramatic decreases in infection rates.⁴⁶³ The Netherlands, a country with a low incidence of MRSA, has successfully controlled MRSA transmission by implementing a national “search-and-destroy” policy.⁴⁶⁴ This policy moves beyond patient screening and contact isolation until the cultures are MRSA negative, and includes HCP screening, patient and HCP decolonization, ward closure, and intensive environmental cleaning.⁴⁶⁴, ⁴⁶⁵ and ⁴⁶⁶ The benefit of this policy in other countries with higher prevalence and different settings has not demonstrated.

HCPs should be educated about MRSA and preventive measures. Furthermore, patients and their family should be advised how to decrease spreading of the resistant organisms.

Vancomycin-Resistant Enterococcus

VRE is another emerging pathogen that is causing an increasing proportion of healthcare-associated enterococcal infections. Two of these two species are primary human pathogens: E. faecalis and E. faecium (most VRE are E. faecium).⁴⁶⁷

Enterococci are normal inhabitants of the GI tract and cause healthcare-associated urinary tract, bloodstream, wound, and intra-abdominal infections. Although less virulent than S. aureus, these bacteria are intrinsically resistant to multiple antimicrobial agents and can cause morbidity and mortality, especially in immunocompromised patients.⁴⁶⁸, ⁴⁶⁹ and ⁴⁷⁰ VRE is the most common cause of bacteremia in bone marrow and stem cell recipients in early post-transplant period (days 4-10).⁴⁷⁰ This organism is associated with increased morbidity in such patients, and a meta-analysis demonstrated that VRE bacteremia increases mortality compared to bacteremia caused by vancomycin-sensitive enterococci.⁴⁷¹

VRE strains were first described in the 1980s and became a significant healthcare problem in the 1990s.⁴⁷², ⁴⁷³ and ⁴⁷⁴ Like many resistant organisms, these strains have become disseminated or emerged in a variety of settings.⁴⁷⁵, ⁴⁷⁶, ⁴⁷⁷, ⁴⁷⁸, ⁴⁷⁹, ⁴⁸⁰, ⁴⁸¹, ⁴⁸², ⁴⁸³, ⁴⁸⁴, ⁴⁸⁵, ⁴⁸⁶, ⁴⁸⁷, ⁴⁸⁸, ⁴⁸⁹, ⁴⁹⁰, ⁴⁹¹, ⁴⁹², ⁴⁹³, ⁴⁹⁴, ⁴⁹⁵, ⁴⁹⁶, ⁴⁹⁷, ⁴⁹⁸, ⁴⁹⁹, ⁵⁰⁰, ⁵⁰¹, ⁵⁰², ⁵⁰³, ⁵⁰⁴, ⁵⁰⁵, ⁵⁰⁶, ⁵⁰⁷, ⁵⁰⁸ and ⁵⁰⁹ The Clinical and Laboratory Standards Institute has defined the minimal inhibitory concentration (MIC) of vancomycin in any enterococci strains of 32 µg/mL as resistant to vancomycin.⁵¹⁰ Patients with serious underlying illnesses who have received multiple antimicrobial agents, including vancomycin, cephalosporins, and antianaerobic drugs, and who are colonized with VRE are at risk for developing a clinically significant VRE infection.⁴⁷⁵, ⁵¹¹, ⁵¹², ⁵¹³ and ⁵¹⁴ Colonization pressure is another important risk factor, defined as the proportion of VRE-colonized patients within a time-period; however, the definition of colonization pressure has varied among studies.⁵¹⁵, ⁵¹⁶, ⁵¹⁷, ⁵¹⁸ and ⁵¹⁹ Other risk factors for infection include having an operation or a central line, characteristics of the patient’s hospitalization (duration, ICU, transplant or oncology service admission, nursery), and contamination of the environment (proximity to a VRE-colonized patient, contaminated equipment, or supplies).⁴⁷⁵, ⁴⁷⁶ and ⁴⁷⁷, ⁴⁸⁷, ⁵²⁰, ⁵²¹, ⁵²² and ⁵²³

Colonization with VRE may persist for long periods of time.⁵²⁴ The organism has a predilection to contaminate the environment, so transmission of VRE can occur via direct contact with colonized patients or by indirect contact with HCPs, environmental surfaces, or inanimate objects.⁵²⁵, ⁵²⁶ and ⁵²⁷ Enterococci can persist on dry inanimate surfaces for as long as 4 months.⁵²⁸

One of the primary reasons that VRE is worrisome is that VRE can transfer vancomycin- or glycopeptide-resistance genes to other bacteria.⁴⁴³, ⁴⁶⁷ In 1992, VRSA was shown to develop in vitro by transfer of van A gene from E. faecalis to S. aureus, and the first clinical infection with VRSA was reported in 2002.⁵²⁹, ⁵³⁰

Prevention measures are directed to the known sources of the organism and include the use and compliance with hand hygiene. Contact precautions should be applied for patients with VRE colonization/infection. Other preventive measures include cohorting of colonized patients, antimicrobial stewardship, and cleaning if contaminated environment and supplies.⁵¹³, ⁵³¹ Active surveillance culture may be performed in high-risk patients, with stool, rectal, or perirectal swabs generally being used for VRE detection.⁵³²

Clostridium difficile

Clostridium difficile infection (CDI) is a major healthcare-associated disease that has emerged internationally.⁵³³, ⁵³⁴ C. difficile bacteria colonize the intestinal tract in 1% to 3% of the normal population and cause disease by disrupting normal flora subsequent to antimicrobial therapy.⁵³⁴, ⁵³⁵ C. difficile is a spore-forming gram-positive bacterium that produces two important toxins: toxin A (enterotoxin) and toxin B (cytotoxin).⁵³⁶, ⁵³⁷ Although most pathogenic strains of C. difficile produce both toxins, 7% to 10% produce only toxin B—the toxin that is essential for virulence.⁵³⁶, ⁵³⁷ The primary mode of transmission is person-to-person spread through the fecal-oral route, although contamination of the environment is important for indirect transmission within hospitals and other healthcare settings.⁵³⁸ The incubation period of CDI remains uncertain but has been estimated to be a median of 2-3 days after exposure to C. difficile.⁵³⁸, ⁵³⁹

C. difficile was first identified as a pathogen associated with antimicrobial-associated diarrhea in 1978.⁵⁴⁰ In the early years, its association with antibiotics was established and the common culprit agents were clindamycin, followed by ampicillin.⁵⁴⁰ In the 1980s, cephalosporins superseded these agents as the most commonly implicated antibiotics linked to CDI.⁵⁴⁰ Until 2003, C. difficile was viewed as an “annoyance”; in that year, however, the hypervirulence strain BI/NAP1/027 emerged and became associated with disease that was more severe, more refractory to standard therapy, and more likely to relapse.⁵³³, ⁵³⁴, ⁵⁴⁰, ⁵⁴¹ The presumed virulence factor results from a deletion within the gene responsible for downregulation of toxin production.⁵³³, ⁵³⁴, ⁵⁴⁰, ⁵⁴¹ This strain is strongly associated with use of fluoroquinolones.⁵⁴¹, ⁵⁴² and ⁵⁴³ The designations of this strain are based on different methods for strain typing: by its pulsedfield gel electrophoresis (PFGE) pattern, NAP1 (for North American PFGE type 1); by its restriction endonuclease analysis pattern, BI; and by its PCR ribotype designation, 027 strain.⁵³⁴, ⁵³⁸ This strain was first reported in North America, but has since spread to Europe and the rest of the world.⁵⁴¹, ⁵⁴⁴, ⁵⁴⁵, ⁵⁴⁶, ⁵⁴⁷, ⁵⁴⁸, ⁵⁴⁹, ⁵⁵⁰, ⁵⁵¹, ⁵⁵², ⁵⁵³, ⁵⁵⁴, ⁵⁵⁵, ⁵⁵⁶, ⁵⁵⁷ and ⁵⁵⁸ In a frightening twist, this epidemic strain has emerged in low-risk populations living in the community without extensive antibiotic or healthcare exposure.⁵⁵⁹

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