#98783: Healthcare-Associated Infections

Droplets containing micro-organisms can be transmitted in the air, causing infection in patients either directly or indirectly (through contamination of devices or equipment). Cleaning activities, such as sweeping, dry mopping, dusting, or shaking linen, can contribute to the transmission of airborne micro-organisms. Bacteria in the air primarily consist of gram-positive cocci from the skin, and they can be eliminated with appropriate ventilation and circulation of air [89]. Many airborne viruses, such as influenza and other respiratory viruses and measles, do not carry far from the source; others, such as tuberculosis and varicella zoster, may be spread over long distances [16]. The most common fungal spore to be transmitted through air is Aspergillus, which is carried through dust particles, can survive for long periods, and is easily inhaled [90]. Under normal circumstances, the level of contamination with this airborne fungus' spores is not high enough to cause disease in otherwise healthy individuals. However, in the healthcare setting, the fungus causes respiratory infection, primarily pneumonia, in susceptible hosts.

The prevalence of infection with Aspergillus within a healthcare setting has been strongly associated with Aspergillus spore counts. Consequently, air conditioning systems with high-efficiency particulate air (HEPA) filters are needed to minimize contamination [91]. HEPA filters are especially needed to prevent infection with Aspergillus in patients at high risk for infection due to a suppressed immune system [92]. In one study, the risk of transplant-related mortality and overall mortality in the first 100 days after transplantation were significantly lower among patients treated in rooms with HEPA and/or laminar flow units than among patients treated in conventional isolation units [93]. In these units, the air exchange rate should be high (more than 15 exchanges per hour), rooms should be tightly sealed, and the air pressure in the rooms should be positive in relation to the hallway [91,94,95]. HEPA filters are also used in the hoods in microbiology laboratories and pharmacies, laminar flow units in ICUs, and unidirectional flow units in operating room suites [16].

Air in the Operating Room

Maintaining a high quality of air in operating rooms is an essential factor in preventing postoperative infection. The number and movement of staff within the operating room create the primary sources of airborne bacteria. Other factors influencing airborne contamination include the type of surgery, the rate of air exchange, the initial quality of the air, the quality of the staff clothing and cleaning processes, and the level of compliance with infection control practices [16].

The CDC makes several suggestions about ventilation in the operating room in its guidelines for prevention of surgical site infections [29]. The Level I recommendations include:

Air During Construction

Special care must be taken to protect patients during repair or renovation of a healthcare facility, as construction work can facilitate the spread of airborne organisms such as Aspergillus species [16]. Some construction issues that contribute to the spread of infection include water-damaged building materials, disruption of duct work, open windows, and improper setting of fans or installation of filters [96].

The Joint Commission requires an inspection process for construction on a facility, and a risk assessment is part of that process [96]. Risk factors to consider include the patient population, the extent and duration of the project, the impact of the project on mechanical systems, and whether space with construction will be occupied [96]. A representative from a facility's infection control program should review any plans for construction to ensure that barriers are used as appropriate and patients, especially those with compromised immune systems, are moved to an area away from construction [89].

Water is a reservoir for several types of micro-organisms, including bacteria, fungi, and viruses, with viruses accounting for only a small percentage [87,97]. The quality of water within a healthcare setting must meet standards that vary according to use. Tap water must be safe to drink and use for baths (for hygiene and therapy) according to criteria dictated by local regulations and public health standards. The water supply to the healthcare facility can be disinfected by several methods, including chlorination, thermal eradication, ultraviolet light, and metal ionization [91].

The most common pathogen identified in tap water is P. aeruginosa [79]. In one study, researchers evaluated the association between tap water from faucets in a surgical ICU and patients with colonization or infection with P. aeruginosa [98]. The pathogen was found in 58% of water samples taken from individual faucets but was not identified in the main water supply. The genotypes of the micro-organism in 21 of the 45 patients were identical to those found in the tap water from the sink in the patient's room (15 patients) or in the adjacent room (6 patients). According to epidemiologic analysis, transmission of the pathogen had occurred from faucet to patient as well as from patient to faucet. P. aeruginosa is also the primary bacterial pathogen found in bath water [99]. The effect of infection with P. aeruginosa may be mild, as in folliculitis and external otitis, but wound infection may be more severe. Greater morbidity is associated with infection in individuals who have a compromised immune system or who have another health condition, such as diabetes [16].

Legionella, which causes infection of the respiratory tract, is another micro-organism commonly found in tap water and bath water. The highest concentrations of Legionella are found in areas of water distribution systems (hot water storage, cooling towers, condensers), where it colonizes [91]. Infection with Legionella is transmitted only through water, not through person-to-person contact. Inhalation of contaminated water droplets from shower heads or faucet aerators may cause disease [87]. In addition, high humidity levels in a room (through mists produced by respiratory equipment, for example) may promote the growth of Legionella and molds [93].

The WHO suggests that there is potential risk for HAIs if tap water is used for such purposes as ice machines or devices for washing eyes or ears, or for cleaning equipment [16]. Point-of-use filtration may help to reduce the risk of HAIs related to water [97]. Ducts, humidifiers, dehumidifiers, and other areas of a ventilation system should be kept clean and dry, as micro-organisms can colonize in water that accumulates in these areas [93]. Patients at high risk for infection should not be exposed to hospital water and sterile water should be used instead [88].

Another factor in the transmission of infection in the healthcare setting is the architectural design of the facility, and the Agency for Healthcare Research and Quality lists "good hospital design principles" as one of its 10 patient safety tips for hospitals [100]. When the American Institute of Architects and the Facility Guidelines Institute (FGI) updated its Guidelines for Design and Construction of Hospital and Health Care Facilities in 2006, they set single-bed private rooms as the minimum standard for new hospital construction [19]. This new standard was based on a literature review that showed, in part, that private rooms have been associated with lower rates of HAIs [101]. Among the benefits of single-patient rooms compared with multibed rooms are decreased risk of infection through contaminated surfaces (e.g., blood pressure cuffs, privacy curtains); availability of private bathrooms; greater ease of cleaning and decontamination; increased likelihood of appropriate hand hygiene between rooms (rather than between beds within a single room); and decreased risk of prolonged hospital stays and patient transfers, all of which are risk factors for HAIs [101]. Draft 2022 FGI guidelines for Design and Construction of Hospitals maintain the single-bed per room standard, but allow for two beds when the necessity of this arrangement has been demonstrated and approved by the authority having jurisdiction [102].

The WHO guidelines on infection control refer to "architectural segregation" according to risk [16]. Four areas of a healthcare facility are defined, with administrative sections considered as low-risk areas; regular patient wards as moderate-risk areas; ICUs, burn units, or isolation units as high-risk areas; and operating rooms as very high-risk areas. The WHO and others have recommended that traffic flow should be limited in higher risk areas [16].

The type of sink and the placement of sinks throughout a healthcare facility have been of critical concern because of the substantial role of handwashing in reducing the transmission of infection. As a result, sinks have been placed within easy access in each patient room. However, it is unclear that such placement promotes better hand hygiene, with no long-term clinically significant improvement in handwashing found when sinks are placed near points of clinical activity [103].

With the advent of alcohol-based handrub solutions as more effective hand hygiene, the placement of handrub dispensers has become more important than the placement of sinks [27]. The CDC guidelines on hand hygiene recommend placing dispensers in convenient locations, such as at the entrance of each patient room or at the bedside.

HAIs have been associated with several types of devices and equipment unique to healthcare facilities. The Spaulding classification, developed in 1968, is widely used to categorize devices according to their associated risk of infection [104]. The system includes three categories:

Most HAIs can be attributed to devices in the critical and semicritical categories, including intravascular catheters, surgical drains, urinary catheters, and endoscopic instruments [89]. Discussion here is limited to endoscopic instruments, as infections related to the other devices are addressed in detail later. In general, the transmission of pathogens on endoscopic devices has been attributed to noncompliance with appropriate reprocessing (cleaning, disinfection, sterilization, and drying) [17,32,105,106]. In particular, appropriate drying has been overlooked as an integral component of reprocessing, and guidelines have been inconsistent in recommendations on drying [107].

Bronchoscopes and gastrointestinal endoscopes are the primary diagnostic scopes used in healthcare settings. Both types of devices are associated with a low risk of infection transmission. Approximately 500,000 flexible bronchoscopies are done in the United States each year [17,108]. Few studies, however, have been carried out to evaluate the risk of infection; nosocomial infection related to bronchoscopy is difficult to detect and is likely under-recognized and under-reported [109]. In 2003, there were two reports of multiple pseudoinfections and true infections, primarily with P. aeruginosa, associated with bronchoscopes that had been reprocessed according to current standards [110,111]. However, in both reports, loose fittings over the valve stem for the working channel of the bronchoscope were thought to have prevented effective mechanical cleaning and disinfection [109]. Overall, the pathogens associated with bronchoscopy-related infection have been P. aeruginosa, Serratia marcescens, nontuberculous mycobacteria, and environmental fungi [109]. In 2014, the U.S. Food and Drug Administration (FDA) received 50 medical device reports that mentioned infection or device contamination associated with reprocessed flexible bronchoscopes [108]. During the course of investigating these reports, the FDA identified two recurrent themes that contributed to device contamination or device-associated infection: failure to meticulously follow the manufacturer's instructions for reprocessing (e.g., failure to perform thorough manual cleaning before high-level disinfection), and continued use of devices, despite integrity, maintenance, and mechanical issues (e.g., persistent channel kinks or bends).

More studies have evaluated the risk of infection associated with gastrointestinal endoscopy, which is performed on approximately 10 to 20 million people each year [112]. The American Society for Gastrointestinal Endoscopy (ASGE) estimates that infectious organisms are transmitted in 1 of 1.8 million gastrointestinal endoscopies [105]. Furthermore, all instances of infection during endoscopy have been the result of noncompliance with established guidelines for reprocessing of endoscopy equipment, highlighting the importance of adhering to these recommendations [32,112,113,114].

As with bronchoscopy, the pathogen with the highest rate of transmission associated with gastrointestinal endoscopy is P. aeruginosa [112,114]. As is true for other pathogens associated with endoscopy, infection with P. aeruginosa has resulted from nonadherence to reprocessing guidelines; however, this pathogen differs from the others because of its predilection for a moist environment. Many cases of infection with P. aeruginosa have been linked to the water supply to the endoscope and to failure to completely dry the endoscope channels with a 70% alcohol solution and forced air [107,112,114]. Salmonella species have also been associated with endoscopy, but no cases have been reported since the publication of the 1988 guidelines for standardized cleaning and disinfection of the devices [112,114]. Infection with Helicobacter pylori has also been related to suboptimal cleaning and disinfection [112]. Low rates of hepatitis B and C virus transmission have been reported, and most cases of infection with hepatitis C were found to be related to the inappropriate use of multiple-dose vials and/or syringes rather than to the endoscope itself [32,112].

Noncritical devices are often overlooked by healthcare workers as vectors for infection. These devices include diagnostic equipment, stethoscopes, and other commonplace items. A systematic review of 23 studies found bacterial contamination of 87% of sampled healthcare equipment, primarily stethoscope membranes, as well as diagnostic ultrasound equipment, otoscopes, and auriscopes [115]. The organisms found on healthcare equipment have primarily been S. aureus, including MRSA; Pseudomonas spp.; Acinetobacter spp.; and Pasteurella spp. [116]. Washing stethoscopes with either an ethanol-based cleanser or isopropyl alcohol pads significantly reduces bacterial growth, even that of MRSA [117,118,119].

Contamination of therapeutic ultrasound transducer heads and ultrasound gels were evaluated in another study, and the rate of contamination was 27% for the heads and 28% for the gels [120]. The transducer heads had low levels of contamination, and most of the micro-organisms were normal flora; however, high levels of contamination were found in the gels, and the micro-organisms included such pathogens as Stenotrophomonas maltophilia, S. aureus, Acinetobacter baumannii, and Rhodotorula mucilaginosa. In two other studies, contamination of dermatoscopes was evaluated. One study indicated colonization of only nonpathogenic bacteria and the other showed that use of an alcohol-based antibacterial gel as immersion fluid yielded no bacterial growth [121,122].

Ward-based computer terminals have also been shown to have low levels of contamination. In a study of two hospitals, MRSA was found on one of 13 computer terminals in one hospital and on five of 12 in another hospital. The rate of MRSA transmission was significantly higher at the hospital with the greater number of contaminated computers [123].

In summary, a high rate of micro-organism colonization has been found on equipment within the hospital setting, but contamination is usually at low levels and the risk of direct infection is low. In general, the findings of studies have suggested that adequate cleaning of equipment can prevent as many as one-third of HAIs [115].

Surgically implanted devices are a major source of HAI, and the development and use of intracardiac devices, orthopedic implants (prostheses and fixation devices), neurosurgical devices, cochlear implants, and breast and penile implants have increased over the past several years. The most common complication with all of these devices is infection [124,125,126]. The prevalence of infection associated with these devices varies, with the prevalence highest for left ventricular assist devices (Table 5) [124]. Orthopedic implants, such as joint prostheses and fracture fixation devices, are associated with the lowest rate of infection, but reported mortality rates have been as high as 18% [124].

Many implanted device-related infections are caused by contamination during insertion, but these infections are not always the result of micro-organisms transmitted in the healthcare setting. Rather, bacteria (and sometimes fungi) colonize by adhering to the surface of the implant through the development of a biofilm [124]. Biofilms present another challenge in managing infection; biofilms provide bacteria with an extremely high level of resistance to antimicrobial agents. In fact, biofilms can tolerate antibiotic concentrations of 10 to 1,000 times of that needed to destroy free-floating (planktonic) bacteria [124]. Many bacteria, including P. aeruginosa, are capable of existing in a planktonic state [114].

The CDC defines implanted device-related HAIs as those occurring within one year after implantation of a device, and the typical interval between implantation and infection varies according to the type of implant [127]. For some implants, early and late infections differ with respect to etiology and the causative micro-organisms [124].

The treatment of infections related to these devices depends on the severity of the infection and the patient's underlying condition. A multidisciplinary approach involving antibiotic therapy and surgical intervention (either debridement or removal of the device) can have a substantial impact on morbidity and mortality. For example, prosthetic valve endocarditis is associated with mortality rates of 42% to 100%, but the rate can be decreased 20-fold through an approach that combines medical and surgical therapy rather than medical therapy alone [124]. Empiric antibiotic therapy is usually appropriate once specimens have been obtained for culture, with the antibiotic agent chosen on the basis of the most common micro-organisms.

The duration of catheterization is the most important risk factor for infection [30]. Studies have shown that catheterization for more than 2 days is a significant risk factor for urinary tract infection as well as increased 30-day mortality [147]. Other risk factors for catheter-related urinary tract infection include no treatment with systemic antimicrobial agents, positive results on culture of the urethral meatus, microbial colonization of the urinary drainage bag, insertion of the catheter outside the operating room, and nonadherence to guidelines for appropriate catheter care. Patient-related risk factors include female gender, older age, diabetes, and an elevated level of serum creatinine at the time of catheterization [30].

Urinary tract infections can be caused by both endogenous and exogenous transmission. Normal flora from the gastrointestinal tract can spread to the urinary tract, or pathogens can be transmitted by caregivers carrying out tasks related to the catheter or drainage bag. Occasionally, pathogens are transmitted through urologic equipment that has not been adequately disinfected. Extraluminal ascension of bacteria along the catheter-urethral mucosa interface is the most common pathway of infection, accounting for approximately two-thirds of infections [30].

According to NHSN data for 2015–2017, the pathogens most commonly isolated from catheter-related urinary tract infections are E. coli (42%), followed by Klebsiella spp. (18%), P. aeruginosa (16%), and Enterococcus species (15%) [144]. Infections related to short-term catheterization is usually caused by a single agent, whereas infections related to long-term catheterization (30 days or more) is typically caused by multiple pathogens. P. mirabilis, Morganella morganii, and P. stuartii are additional common pathogens in infections related to long-term catheterization [30].

With regard to antimicrobial-resistant pathogens, the percentage of S. aureus infections resistant to oxacillins decreased in 2015–2017 (41% vs. 59% in 2009–2010) [144]. The percentage of Enterococcus faecium resistant to vancomycin increased, to 82% (from 80% in 2007–2008), as did the percentage of vancomycin-resistant E. faecalis (approximately 12% vs 6% in 2007–2008). The prevalence of multidrug-resistance among E. coli, Klebsiella, and Enterobacter spp., previously reported as having increased from 3% in 2007–2008 increased 10%, 13%, and 7%, respectively in 2015–2017 [144].

The principles of care required for prevention of catheter-associated urinary tract infection are well established: appropriate use, sterile catheter placement, maintenance of a closed drainage system, avoidance of back-flow, and minimal duration of catheter insertion. In addition to these technical aspects, a systems approach that standardizes care, educates, and fosters an interprofessional culture of attentiveness is also necessary for optimal preventive care [21,148].

The evidence-based guidelines for prevention of catheter-associated urinary tract infections were published by the CDC in 2009 (Table 8). The IDSA has published evidence-based clinical practice guidelines on the prevention, diagnosis, and treatment of catheter-associated urinary tract infections in 2010 [21,30]. According to both guidelines, the most important principles for prevention are:

As with prevention of all HAIs, handwashing is an essential element of aseptic technique and care of patients with catheters. In addition, healthcare staff should be educated and trained in proper techniques of catheter insertion and care.

Alternatives to indwelling catheters have been evaluated as an approach to preventing catheter-related urinary tract infections. The IDSA guidelines note that a suprapubic catheter may be considered as an alternative to short-term catheterization (level III, C), but use of this type of catheter is limited because an invasive procedure is needed for insertion [30]. Intermittent catheterization may also be considered as an alternative to short-term (level I, C) or long-term (level III, A) catheterization, and, for men who have minimal postvoid residual urine, condom catheterization can be considered as an alternative to short-term (level II, A) or long-term (level II, B) catheterization in those who are not cognitively impaired [30].

The use of catheters coated with an antimicrobial surface has been evaluated, especially those coated with silver, a highly effective antibacterial substance. In one study, the addition of silver did not reduce the incidence of bacteriuria when compared with silicone-based, hydrogel-coated urinary catheters with and without silver impregnation [149]. One meta-analysis (12 trials; 13,392 patients or catheters) showed that antimicrobial-coated catheters prevented or delayed the onset of bacteriuria in select patients, but the magnitude of the effect varied substantially according to several variables, including catheter type and publication year [150]. A subsequent systematic review (eight studies) found a favorable trend toward a lower rate of infection with silver alloy (vs uncoated) catheters, but the quality of some studies was poor and there was significant heterogeneity among the studies [151]. The 2009 CDC guidelines state that antimicrobial/antiseptic-impregnated catheters can be considered if the rate of catheter-associated urinary tract infections does not decrease after a comprehensive prevention strategy has been implemented [21]. The guidelines add that further research is needed to determine the effect of these catheters in reducing the risk of symptomatic infection, their inclusion among the primary interventions, and the patient populations most likely to benefit from them [21].

Similarly, the IDSA states that catheters coated with an antimicrobial surface may be considered to reduce the risk of infection for patients who are to have short-term catheterization (level II, B), but notes that the data on the effectiveness of this strategy are insufficient [30].

The IDSA guidelines note several prevention strategies that should not be used routinely, primarily because of insufficient data [30]:

The CDC and the IDSA classify catheter-associated urinary tract infections as symptomatic urinary tract infection, asymptomatic bacteriuria, or other infection of the urinary tract. Urine samples for urinalysis and quantitative urine culture (using a clean catch technique or catheterization) are necessary for accurate diagnosis [30,127].

According to the Infectious Diseases Society of America (IDSA) and the Society for Healthcare Epidemiology of America (SHEA), signs and symptoms compatible with catheter associated-urinary tract infection include new onset or worsening of fever, rigors, altered mental status, malaise, or lethargy with no other identified cause; flank pain; costovertebral angle tenderness; acute hematuria; pelvic discomfort; and in those whose catheters have been removed, dysuria, urgent or frequent urination, or suprapubic pain or tenderness.

(https://academic.oup.com/cid/article/50/5/625/324341 Last Accessed: January 20, 2022)

Strength of Recommendation/Level of Evidence: AIII (Good evidence from opinions of respected authorities based on clinical experience, descriptive studies, or reports of expert committees)

The IDSA notes that catheter-associated urinary tract infection is defined by the presence of signs or symptoms compatible with a urinary tract infection with no other identified source, and at least one bacterial species at a count of ≥103 cfu/mL in one urine specimen (level III, A) [30]. This threshold differs from that defined by the CDC (≥105 cfu/mL), which is intended for infection control surveillance rather than detection of infection in an individual patient [127]. In addition, the 2009 CDC guidelines define the criteria for symptomatic catheter-associated urinary tract infection as symptoms plus a positive urine culture at the threshold of ≥105 cfu/mL; if the urine culture result is between ≥103 and ≤105 cfu/mL, a positive urinalysis is needed to meet the diagnostic criteria [21].

Urine specimens for culture should not be obtained from the drainage bag; instead, a sample should be taken through the catheter port with use of aseptic technique [30]. If there is no port, a needle and syringe can be used to puncture the catheter tubing and collect the specimen [30]. For patients with a long-term indwelling catheter, the IDSA recommends replacing the catheter and collecting the specimen from the newly placed catheter [30].

Signs and symptoms suggestive of catheter-associated urinary tract infections include fever (new or worsening), flank pain, hematuria, pelvic discomfort, altered mental status, and malaise or lethargy not attributable to another cause; among patients without a current indwelling catheter, dysuria, urgency, and frequent urination are other symptoms [30]. However, studies have shown that the classic symptoms of urinary tract infection are uncommon among patients with a catheter-associated infection [30]. Pyuria is not diagnostic of catheter-associated urinary tract infection in patients with an indwelling catheter [30].

Asymptomatic bacteriuria is defined as the presence of significant bacteriuria with no signs or symptoms referable to the urinary tract [30].

The presence of bacteriuria (as defined by a positive urine culture) in a patient with a chronic indwelling catheter should be interpreted with caution. Often, this represents poor culture technique or may reflect an asymptomatic infection that need not be treated until the catheter can be removed or unless there is fever, flank pain, or worsening renal function. The IDSA recommends that in the absence of such signs these infections not be treated, as treatment has not been found to be beneficial [30,152,153]. However, for symptomatic infection, the results of urine culture and antibiotic sensitivities are an essential guide to effective treatment [30]. For patients with a catheter in place, the catheter should be discontinued, if possible, and the urine specimen should be one that is voided midstream (level III, A) [30]. If the catheter has been in place for more than 2 weeks and continued catheterization is necessary, the catheter should be replaced (level I, A) and the urine specimen should be collected from the newly placed catheter (level II, A) [30].

The IDSA guidelines do not specify which antimicrobial to use for treatment but do recommend duration of treatment. Treatment for 7 days is recommended for patients in whom symptoms resolve promptly, and treatment for 10 to 14 days is recommended for patients in whom symptom response is delayed (level III, A) [30]. The guidelines note that a five-day regimen of levofloxacin may be considered for patients who are not severely ill (level III, B) [30].

It has been suggested that catheter-associated urinary tract infection is the most preventable HAI [49]. Some studies have noted an increase in adherence to strategies to prevent catheter-associated urinary tract infection [53]. The greatest adherence has been to guidelines for wearing gloves (97%), using appropriate hand hygiene (89%), and maintaining a sterile barrier (81%) [154]. However, adherence to appropriate catheterization (in terms of both initial indication and duration) has been suboptimal, with studies demonstrating the following [30,147,148,155,156]:

Use of alternatives to catheterization is also low, ranging from 14% to 20% [155,156].

The use of provider reminder systems—alone or in combination with base strategies—has a moderate strength of evidence [1]. In one study, the combination of prompts in a computerized order-entry system and handheld bladder scanners led to an 81% decrease in the use of catheters and a 73% reduction in HAIs [157]. For hospitals without order-entry systems, a handwritten reminder that the patient has a catheter has been effective in reducing the rate of infection [158]. However, the use of reminder systems has been reported to be 9% to 12% [155,156].

The use of bundled interventions (e.g., staff education, electronic daily checklist, a nurse-driven removal protocol for indwelling urinary catheters) for reducing the rate of catheter-associated urinary tract infections was examined in a population of critically ill patients 18 years of age and older who were admitted to the ICU of one hospital [159]. The hospital had previously reported 13 catheter-associated urinary tract infections during one year (6 in the ICU), which exceeded the institution's goal of four or fewer such events annually. Researchers set objectives of a 30% reduction in reported catheter-associated urinary tract infections, a 20% reduction in urinary catheter days, and a 75% compliance rating in catheter-related documentation in the ICU. During the intervention phase, no catheter-associated urinary tract infections were reported, which reduced the rate by 1.33 per 1,000 catheter days. Documentation compliance increased significantly from 50.0% before intervention to 83.3% during intervention. The increase in catheter days (10.5%) was not statistically significant [159].

Despite the lack of adherence to prevention guidelines, some progress has been made in reducing this HAI; the rate of symptomatic catheter-associated urinary tract infections decreased 19% to 67% between 2000 and 2007 in ICUs [160]. Compliance with a nurse-driven evidence-based checklist led to a decrease in infections from 2.88/1,000 catheter days to 1.46/1,000 catheter days [161].

The CDC guidelines note that the following are effective elements of a quality improvement program [21]:

The Institute for Clinical Systems Improvement recommends that a basic nutritional assessment should be considered on all patients undergoing surgery as malnutrition is a known risk factor for decreased wound healing and increased surgical site infections.

(https://www.icsi.org/wp-content/uploads/2019/01/Periop.pdf Last Accessed: January 20, 2022)

Level of Evidence: Consensus Statement/Expert Opinion

Several patient-related and surgery-related factors increase the risk for surgical site infection. Patient-related factors include [29,136,140,170]:

Among the most common surgery-related factors are anesthesia score, duration of the operation, the use of drains, and inadequate aseptic technique [89]. In a study to determine the influence of risk factors on complications after colorectal surgery, body mass index, duration of the operation, and the surgeon who performed the operation were the three most important factors influencing surgical site infections [171].

Surgical site infections arise from both endogenous and exogenous transmission. The microbial sources of surgical site infections vary according to the type of surgery, and the micro-organism reported as being the most common in 2015–2017 was S. aureus, which accounted for nearly half of all reported orthopedic surgical site infections [144]. Other common causative pathogens were coagulase-negative staphylococci, Enterococcus faecalis, and E. coli [144].

Among the facilities reporting data on surgical site infections to the NHSN in 2015–2017, the percentage of resistant phenotypes was low for pathogens most often associated with surgical site infections [144]. The percentage resistance for most pathogens decreased compared with data for 2007–2008; for example, 42% of S. aureus infections were resistant to oxacillins, a significant decrease from 48% in 2007–2008. The percentage of vancomycin-resistant E. faecium decreased from 62% in 2009–2010 to 56% in 2015–2017, and the percentage of vancomycin-resistant E. faecalis decreased from 6.2% to 3.4% in 2015–2017 [144]. Rates of resistance of E. coli to extended-spectrum cephalosporins (18%) and of P. aeruginosa to aminoglycosides (6%) increased [144].

The CDC guideline for preventing surgical site infections, published in 1999, addresses a wide variety of issues, including preoperative preparation of the patient, antisepsis of the surgical team, management of surgical personnel with colonization or infection, antimicrobial prophylaxis, ventilation, cleaning and disinfection of environmental surfaces, microbiologic sampling, sterilization of surgical instruments, surgical attire and drapes, asepsis and surgical technique, postoperative incision care, and surveillance. The CDC's current guideline for prevention of surgical site infection was published in 2017 [29,172]. The guidelines recommend a combination of key components as a strategy to prevent surgical site infection (Table 10).

Before surgery, patients should be advised to shower or bathe (full body) with soap (antibacterial or non-antibacterial) or an antiseptic agent on at least the night before the operative day [29]. Antimicrobial prophylaxis should be administered only when indicated based on published clinical practice guidelines and timed such that a bactericidal concentration of the agents is established in the serum and tissues when the incision is made [29]. Antibiotic prophylaxis need not be maintained longer than a few hours after the incision has been closed. Additional guidance is provided in reference to specific surgical procedures and specialty operations (e.g., prosthetic joint arthroplasty) [29].

The CDC guidelines reference appropriate antibiotics on the basis of the type of surgery [29,172]. Meta-analyses have demonstrated lower rates of infection with a single-dose (long-acting) antibiotic and broader spectrum antibiotics, such as third-generation cephalosporins [173]. However, a complication of perioperative antibiotic prophylaxis is an increased frequency of adverse events, the most serious of which is infection with C. difficile, and this risk may be higher in association with a broad-spectrum antibiotic [173].

Systematic reviews reported in the mid-2000s showed that two previously recommended measures for preventing postoperative infection have no effect on the rate of surgical site infection. One review involved six trials in which preoperative washing was evaluated in a total of 10,007 patients. There was no significant difference in the rate of surgical site infections when 4% chlorhexidine gluconate was compared with placebo or no washing [174]. The other review involved 11 randomized controlled trials in which preoperative hair removal practices were evaluated. Comparison of razor, depilatory cream, or no hair removal showed no significant difference in the rate of surgical site infection [175,176]. However, when shaving was compared with clipping, there were significantly more surgical site infections after shaving. No difference was found in the rate of surgical site infections between clipping 1 day preoperatively or on the day of surgery. On the basis of these findings, preoperative antiseptic washing and shaving are no longer recommended.

Attention should also be directed at strategies to prevent surgical site infections with MRSA. The use of an MRSA prevention bundle—adherence to the guidelines for hand hygiene, decontamination of environment and equipment, active surveillance cultures, and contact precautions for patients with MRSA infection or colonization—led to significant decreases in the overall rate of surgical site infections in one study, with a 1% decrease in surgical site infections after cardiac surgery and a 65% decrease after orthopedic surgeries [177].

IDSA guidelines on the diagnosis and management of skin and soft tissue infections include a section on surgical site infections [164,178]. The guidelines note that the most reliable diagnostic information is the physical appearance of the site; local signs of infection include pain, swelling, erythema, and purulent drainage [164,178]. Clinical manifestations of a surgical site infection do not occur for at least 5 days postoperatively, with many infections not becoming apparent for as long as 2 weeks [164,178]. The IDSA notes that most postoperative fevers are not associated with a surgical site infection [164,178].

Surgical site infections are classified as superficial incisional, deep incisional, and organ/space infections. Strict criteria and standardized definitions are used in reporting infections and in surveillance programs [29,173]. The CDC described the criteria for each type of infection in its guidelines for the prevention of surgical site infections and defined the infections in the NHSN system according to this classification [29,127,179].

Superficial Incisional Classification

Infection occurs within 30 days after the operative procedure and involves only skin and subcutaneous tissue of the incision and at least one of the following:

Deep Incisional Classification

Infection occurs within 30 days after the operative procedure if no implant is left in place or within 1 year if implant is in place and the infection appears to be related to the operative procedure and involves deep soft tissues (e.g., fascial and muscle layers) of the incision and at least one of the following:

Organ/Space Classification

Infection occurs within 30 days after the operative procedure if no implant is left in place or within 1 year if the implant is in place and the infection appears to be related to the operative procedure and infection involves any part of the body, excluding the skin incision, fascia, or muscle layers, that is opened or manipulated during the operative procedure and at least one of the following:

Based on expert opinion, the IDSA recommends opening an infected surgical site, removing the infected material, and continuing dressing changes until the wound heals by secondary intention [164]. Although treatment with antibiotics is commonly started when a surgical site infection is diagnosed, the IDSA notes that little evidence has supported this approach [164]. A short course (24 to 48 hours) of antibiotics may be indicated for patients with a temperature higher than 38.5 degrees Centigrade or a pulse rate of more than 100 beats/min [164]. The guidelines add that treatment is usually empirical but may be selected according to results of wound culture [164]. IDSA offers guidance on the selection of antibiotics according to the operative site [164].

Intestinal or Genital Tract

Single agents: ticarcillin/clavulanate, piperacillin/tazobactam, imipenem/cilastatin, meropenem, ertapenem

Combination agents: ceftriaxone/metronidazole, ciprofloxacin/metronidazole, levofloxacin/metronidazole, ampicillin-sulbactam/gentamicin ampicillin-sulbactam/tobramycin

Trunk and Extremities (away from axilla or perineum)

Oxacillin, nafcillin, cefazolin, cephalexin, SMX-TMP, vancomycin

Axilla or Perineum

Metronidazole/ciprofloxacin, levofloxacin, ceftriaxone

For surgical site infections after implantation of a joint prosthesis, the approach depends on the duration of infection, stability of the implant, antimicrobial susceptibility of the pathogen, and condition of the surrounding soft tissue [180]. In this setting, rifampin has shown excellent activity against adherent staphylococci and may be useful in combination with beta-lactams, glycopeptides, fluoroquinolones, minocycline, trimethoprim, or fusidic acid [181].

The effect of prevention strategies—primarily the appropriate use of prophylactic antibiotics—on the rate of surgical site infection has been evaluated in several studies, and the results have been conflicting. For example, 56% adherence to this measure significantly reduced the incidence of surgical site infections in one study of patients who had colorectal surgery, with the rate decreasing from 22% to 3.5% [182]. There was no significant difference in the rate of surgical site infection between compliant and noncompliant prophylactic antibiotics [182]. In a retrospective review of 605 patients who had colorectal surgery with intestinal anastomosis showed that early administration of antibiotic prophylaxis and a nonstandard antibiotic were significantly associated with a greater risk of surgical site infection [183]. However, the timely administration of prophylactic antibiotics did not improve the rate of surgical site infections among nearly 9,200 elective major surgeries (all types); the rate was 5% for patients who received timely antibiotics compared with 6% for patients who received antibiotic prophylaxis outside of the recommended time [51].

More recent studies have analyzed adherence to the Surgical Care Improvement Project (SCIP) quality measures, with analysis of individual measures as well as groups of measures. Some studies have indicated that an increase in compliance with SCIP measures leads to a decrease in the rate of surgical site infection; the rate decreased from nearly 26% to 16% in one study and in another study, increasing compliance with SCIP measures from 38% to 92% led to a decrease in the rate of superficial surgical site infections from 13% to 8% [50,184]. Among patients who had laparotomy related to trauma (gunshot wound, stab wound, or blunt trauma), adherence to SCIP measures related to antibiotic prophylaxis resulted in a significantly lower rate of surgical site infection (17% vs 33%) as well as a shorter hospital stay (14 vs 19 days), even after controlling for several factors [185].

However, these data have not been consistent. In a study of nearly 500 patients who had colorectal surgery, compliance with all SCIP measures improved—from 40% to 68%—but the rate of surgical site infections remained essentially the same (approximately 19%) [52]. In addition, a retrospective review of 60,853 surgeries done over a five-year period at 112 Veterans Administration (VA) hospitals showed that improving adherence to five SCIP measures did not significantly lower the odds of surgical site infection [169].

Many have been critical of the SCIP strategy, noting that the evidence has not indicated that adherence to these quality measures alone has had an effect on reducing the overall risk of surgical site infection [186]. The quality measures do not factor in the skill of the surgeon, and other factors, such as state-of-the-art skin antisepsis and innovative antimicrobial technology, should be included in quality improvement programs [186,187].

Ways to help increase adherence to effective prevention strategies are the use of computerized standard orders for antibiotics, reminders and checklists, and auditing of the rates for individual physicians, with feedback [29]. A study demonstrated that point-of-care prompts increased compliance with timely antibiotic prophylaxis from 62% to 92%, with a corresponding decrease in the incidence of surgical site infections, from 1.1% to 0.7% [188]. In an extensive systematic review, there was moderate strength of evidence for the use of audit and feedback, with or without provider reminder systems, for improving adherence to appropriate timing of prophylactic antibiotics [1]. The IHI how-to guide for the prevention of surgical site infection outlines practical steps to help healthcare professionals ensure that prevention strategies are carried out (Table 11) [189].

Hospital-Acquired Pneumonia

In a systematic review, the American College of Physicians found several patient-related and surgery-related factors that increased the risk of postoperative pulmonary complications. The most common patient-related factors were the presence of COPD and an age older than 60 years [141]. Other significant factors were an American Society of Anesthesiologists (ASA) class 2 (defined as a patient with mild systemic disease) or higher, functional dependence, and congestive heart failure. Cigarette use was associated with a modest increase in risk, and obesity and mild or moderate asthma were not found to increase risk [141]. Use of a PPI or histamine-2 receptor antagonist is also thought to be a risk factor [142]. Surgery-related factors included prolonged duration of surgery (more than three to four hours), emergency surgery, and surgical site, with abdominal surgery, thoracic surgery, neurosurgery, head and neck surgery, vascular surgery, and aortic aneurysm repair being associated with the greatest risks [141].

Ventilator-Associated Pneumonia

The risk of ventilator-associated pneumonia correlates with the duration of intubation; the risk has been estimated to be 3% per day during the five-day period after intubation, decreasing to 2% per day for days 5 through 10 and to 1% per day for longer durations [193]. Nearly half of all cases of ventilator-associated pneumonia develop within the first four days of mechanical ventilation [190]. In addition to duration of ventilation, several other risk factors among adults have been identified, including a supine head position; use of a nasogastric tube, paralytic agents, or PPI or histamine-2 receptor antagonists; patient age; chronic lung disease; and head trauma [20,142]. In one study, ventilator-associated pneumonia was most frequently associated with ICU admission diagnoses of postoperative care, neurologic conditions, sepsis, and cardiac complications [194].

Gram-negative enteric bacilli and Pseudomonas spp. rarely colonize the upper respiratory tract of healthy individuals, but often do so in persons with an underlying disease, such as alcohol use disorder, and in those who are hospitalized or reside in nursing homes. Most cases of pneumonia that develop in a healthcare facility are caused by aspiration of oropharyngeal or gastric secretions colonized with hospital bacterial flora. Consequently, the prevalent causation as well as the antibiotic sensitivity pattern of resident pathogens will vary from region to region in relation to the type of facility and burden of antimicrobial usage. The selection of initial antibiotic therapy in these cases is based on the patient's risk factors for infection with a multidrug-resistant organism, such as MRSA, P. aeruginosa, K. pneumoniae, or Acinetobacter. The infectious disease and pulmonary specialty societies (IDSA and American Thoracic Society [ATS]) list the following risk factors for multidrug-resistant pathogens in patients presenting with hospital-acquired or ventilator-associated pneumonia [18]:

Approximately 50% of all cases of healthcare-associated pneumonia develop following surgical procedures, of which cardiac, abdominal, and orthopedic surgery confer the greatest risk. Viral and fungal pathogens are rare causes of hospital-acquired and ventilator-associated pneumonia in immunocompetent adults. Outbreaks of viral pneumonia may occur during influenza season, and influenza, parainfluenza, adenovirus, and respiratory syncytial virus (RSV) are involved in about 70% of those cases [18]. During the coronavirus disease (COVID-19) pandemic, SARS-CoV-2 has superseded the usual viral respiratory pathogens. Candida spp. and Aspergillus fumigatus may cause pneumonia in patients who have had organ transplantation or who have a compromised immune system and neutropenia.

Hospital-Acquired Pneumonia

Among adults with no previous antibiotic exposure, the most common bacterial causes of hospital-acquired pneumonia are S. aureus, S. pneumoniae, H. influenzae, E. coli, and K. pneumoniae [18,190,191,195]. Gram-negative bacilli resistant to first-generation cephalosporins also frequently develop in late-onset hospital-acquired pneumonia. When patients who have previously received antibiotics develop late-onset hospital-acquired pneumonia, the likelihood of causation by a multidrug-resistant pathogen, such as P. aeruginosa, A. baumannii, or MRSA, approaches 40% [191]. In a study of more than 3,600 patients admitted to an ICU, Pseudomonas spp. was the cause of pneumonia in 25% of patients; MRSA in 18%; and Acinetobacter spp. in 6% [195]. Other studies have shown that S. aureus is common among patients who are in a coma or have diabetes or renal failure; P. aeruginosa is common among patients who have had a prolonged stay in the ICU, have received prior antibiotics or corticosteroids, or who have structural lung disease; and Legionella is usually found in patients who have compromised immune systems [195].

Ventilator-Associated Pneumonia

In 2015–2017, the most common pathogens reported with ventilator-associated pneumonia in adults were S. aureus (29%) and P. aeruginosa (13%), followed by K. pneumonia/oxytoca (10%), Enterobacter spp. (8%), and H. influenzae (6%) [144]. Almost half of all cases of ventilator-associated pneumonia are caused by infection with more than one pathogen [190]. As with other forms of HAI, the percentage of S. aureus resistant to methicillin has decreased in recent years [3,163]. The percentage of vancomycin-resistant E. faecium has remained stable, but the percentage of vancomycin-resistant E. faecalis decreased from 10% in 2009–2010 to 7% in 2015–2017 [144]. In 2015–2017, the rates of resistance among Klebsiella spp. for extended-spectrum cephalosporins, carbapenems, multidrug were 21%, 7%, and 13%, respectively, and the rate of multidrug-resistant E. coli increased to nearly 10% [144].

The CDC has published guidelines for the prevention of hospital-acquired and ventilator-associated pneumonia, with a focus on strategies to decrease or eliminate modifiable risk factors [25]. These strategies are related to preoperative and postoperative care and measures to reduce the risk of transmission of etiologic pathogens. In addition, steps to prevent the spread of influenza virus are essential, especially during influenza season.

Hospital-Acquired Pneumonia

For many years, preventing postoperative pneumonia has been a part of initiatives to decrease complications among patients who have surgery. The Respiratory Risk Index was developed to classify patients as being at low, medium, or high risk for postoperative respiratory failure [191]. The factors in the index include the complexity of the surgery, the ASA status, and comorbidities.

Smoking triples the risk for pulmonary complications after surgery, and smoking cessation for at least 8 weeks before surgery, when possible, is recommended for current smokers [191]. The risk for complications in patients with respiratory disease or congestive heart failure can be ameliorated by optimum treatment before surgery (e.g., treatment with steroids for patients with COPD or asthma) [191].

Effective pain management after surgery can also help decrease the risk of pulmonary complications. For postoperative patients who are not mechanically intubated, the ability to cough and clear secretions is important for preventing pulmonary complications [191]. The use of incentive spirometry and deep breathing exercises are recommended, especially for people at high risk for pulmonary complications, as are frequent coughing and early movement (in bed and/or walking) [25,141,191]. Fair evidence supports the selective (rather than routine) use of a nasogastric tube after abdominal surgery [141].

Ventilator-Associated Pneumonia

Two guidelines were developed to focus specifically on the prevention of ventilator-associated pneumonia; one was jointly developed by the SHEA and IDSA, and the other was jointly developed by the Canadian Critical Care Trials Group and the Canadian Critical Care Society [20,36]. In addition, prevention of ventilator-associated pneumonia is addressed in the CDC's guidelines for preventing healthcare-associated pneumonia and in the IDSA/ATS guidelines on the management of healthcare-associated pneumonias [18; 25]. All of these agencies suggest a multicomponent strategy for prevention of pneumonia. Compliance with guidelines, however, has been slow; nursing surveys demonstrate rates of adherence to specific preventive measures ranging from 15% to 50% [192,196]. All of these agencies suggest a multicomponent strategy for prevention of pneumonia. Compliance with guidelines, however, has been slow; nursing surveys demonstrate rates of adherence to specific preventive measures ranging from 15% to 50% [192,196]. Education is beneficial, and training sessions are a proven means to enhance knowledge and practice among healthcare professionals caring for intubated patients [197].

The Institute for Healthcare Improvement (IHI) found that implementation of its ventilator bundle, a collection of five prevention strategies drawn from these guidelines, led to a 45% reduction in the incidence of ventilator-associated pneumonia [198]. The bundle includes the following interventions [198]:

Assessment of Readiness to Extubate

Because of the increasing risk of infection as the duration of ventilation increases, the primary goal is to extubate patients as early as possible. Thus, assessment of the readiness for extubation and weaning protocols are key aspects in the preventive approach [18,195]. Daily interruption of sedation until the patient is awake has been shown to significantly decrease the number of days on mechanical ventilation, from 7.3 days to 4.9 days in one study [199]. There are risks to this approach, such as the potential for increased pain, anxiety, and desaturation. However, the use of sedation interruption has been further demonstrated to reduce the complications of prolonged mechanical ventilation [200]. The SHEA/IDSA guidelines recommend daily assessment of the readiness to wean and the use of weaning protocols [36].

Elevation of the Head of the Bed

Reducing the risk of aspiration and contamination with gastric secretions also helps to prevent the development of ventilator-associated pneumonia. The risk of aspiration has been significantly reduced by positioning the patient with the head of the bed at an angle of 30 to 45 degrees [20,201,202]. In one randomized controlled trial, there were 18% fewer cases of ventilator-associated pneumonia among intubated patients in the group assigned to the recumbent position (45 degrees) compared with the group assigned to the supine position [202]. In another study, elevation of the head of the bed at 30 degrees was the most effective measure among a group of preventive interventions, resulting in a 52% variance in the rate of ventilator-associated pneumonia [203]. Both the ATS/IDSA and SHEA/IDSA guidelines recommend maintaining the head of the bed at a 30- to 45-degree angle [18,36].

Daily Oral Care with Chlorhexidine

Oral care interventions have been suggested by some, in part because of an association between a high level of dental plaque and a high rate of colonization with aerobic pathogens, including S. aureus, gram-negative bacilli, and P. aeruginosa [204]. Research has shown that oral decontamination with chlorhexidine leads to a significant reduction in the colonization of pathogens in the oropharynx; in most studies, the intervention has not had a significant effect on the rate of ventilator-associated pneumonia or associated mortality, but more recent studies have shown significant decreases in the rate of ventilator-associated pneumonia [205,206,207]. Including tooth brushing with chlorhexidine does not seem to add benefit [208,209]. Regular oral care with an antiseptic solution or chlorhexidine is recommended in the ATS/IDSA and SHEA/IDSA guidelines [18,36].

Prophylaxis of Peptic Ulcer Disease

Prophylaxis of peptic ulcer disease has evolved with some conflicting views. Antacids, histamine-2 antagonists, and sucralfate have been traditionally given to patients receiving mechanical ventilation to prevent the formation of ulcers. However, reducing the amount of gastric acid can increase the risk of colonization of gram-negative bacilli in the stomach. As a result, WHO recommended avoiding the use of these agents [16]. The CDC noted that there was insufficient evidence on the use of peptic ulcer prophylaxis and included no recommendations in this regard in its guidelines [25]. The ATS/IDSA guidelines stated that the risks and benefits of prophylaxis should be weighed carefully [18]. The most recent guidelines, developed by SHEA/IDSA, notes that histamine-2 receptor antagonists and PPIs should be avoided in patients who are not at high risk for developing a stress ulcer or stress gastritis [36].

Prophylaxis of Deep-Vein Thrombosis

There is no clear relation between prophylaxis of deep-vein thrombosis and ventilator-associated pneumonia, but the American College of Chest Physicians reported a decrease in the rate of ventilator-associated pneumonia when such prophylaxis was implemented as part of a package of interventions and included this measure in its clinical practice guidelines [210].

Other Measures

In addition to these interventions, other measures have been recommended to help prevent ventilator-associated pneumonia. One such measure is selective decontamination of the digestive tract, which involves the use of either topical antiseptic, oral antibiotics, or a brief course of systemic antibiotics [191]. A meta-analysis (28 studies) showed that selective decontamination of the digestive or respiratory tract with use of topical antiseptic or antimicrobial agents helped reduce the frequency of ventilator-associated pneumonia in the ICU [211]. The estimate of efficacy in prevention was 27% for antiseptics and 36% for antibiotics. Neither had an effect on mortality. This intervention is recommended in the SHEA/IDSA guidelines [36].

Other preventive measures are targeted primarily to the care and use of ventilator equipment and practices in direct patient care. Meticulous attention to aseptic care of the equipment is necessary, and all reusable components, such as nebulizers, should be disinfected or sterilized. Tubing circuits should be replaced after more than 48 hours, or earlier if there are signs of malfunction or contamination [25]. Changes in the design of the endotracheal tube have also been evaluated; for example, a tube with a suction port above the cuff allows for continuous aspiration of subglottic secretions. Use of this specially designed endotracheal tube has led to significantly lower rates of ventilator-associated pneumonia, as well as shorter durations of ventilation and shorter stays in the ICU [212,213]. Among patients who had major cardiac surgery, the greatest benefit was found for patients who received ventilation for more than 48 hours [213]. The cost of the tube is higher than traditional tubes but is offset by overall cost savings in preventing ventilator-associated pneumonia [212]. In one meta-analysis, subglottic secretion drainage was significantly associated with a decreased incidence of ventilator-associated pneumonia, shorter time on mechanical ventilation, and longer time to the development of ventilator-associated pneumonia [214]. The CDC, the ATS/IDSA, and the SHEA/IDSA guidelines recommend subglottic secretion drainage with this tube when possible [18,25,36].

The use of noninvasive ventilation is another measure that has reduced the incidence of ventilator-associated pneumonia [25,215,216]. In one study, the incidence decreased from 20% to 8% when noninvasive ventilation was used routinely for critically ill patients with acute exacerbation of chronic obstructive pulmonary disease or severe cardiogenic pulmonary edema [217]. Again, the CDC, the ATS/IDSA, and the SHEA/IDSA guidelines recommend the use of noninvasive ventilation when possible [18,25,36].

The difficulty in diagnosing hospital-acquired or ventilator-associated pneumonia has been well established [18,193,218]. The clinical signs can resemble those of other, noninfectious conditions, and the specificity of clinical criteria is low [190]. According to the CDC definition, the diagnosis in adults is made on the basis of clinical signs and symptoms and results of laboratory testing or imaging and must meet one of two criteria [219,220].

Criterion 1

For any patient, at least one of the following:

AND at least two of the following:

Criterion 2

Two or more serial chest radiographs showing at least one of the following:

In patients without underlying pulmonary or cardiac disease (e.g., respiratory distress syndrome, pulmonary edema, chronic obstructive pulmonary disease), one definitive chest radiograph is acceptable.

There are no compelling data to recommend a specific approach to diagnosing hospital-acquired or ventilator-associated pneumonia. For patients who are not receiving mechanical ventilation, collection of a sputum specimen should be attempted before antibiotic therapy is begun [195,221]. Specimens for culture can be obtained by bronchoscopy with a protected specimen brush to limit contamination or by bronchoalveolar lavage. The latter method has been found to lead to higher rates of treatment than that based on the CDC definition, and one study showed that preferential sampling of the right lung improved the diagnostic accuracy of bronchoalveolar lavage [195,222,223]. However, the invasive procedure has disadvantages, including high cost, need for technical expertise, and potential for false-negative results [195,222].

The 2016 IDSA/ATS guidelines recommend collecting specimens from the lower respiratory tract for culture, preferably by noninvasive techniques and reliance on semiquantitative culture technique [18]. Noninvasive methods to obtain respiratory samples in patients with hospital-acquired (but not ventilator-associated) include spontaneous expectoration, sputum induction, nasotracheal suctioning (in a patient unable to produce a sample), and endotracheal aspiration in a patient who subsequently requires mechanical ventilation [18]. A 2012 meta-analysis (and a 2014 update) found no evidence that the use of quantitative cultures of respiratory secretions resulted in decreased mortality, reduced time in ICU and on mechanical ventilation, or higher rates of antibiotic change compared with qualitative cultures in patients with ventilator-associated pneumonia [224,225]. In addition, there was no difference in mortality, whether invasive or noninvasive methods were used to obtain specimens.

Treatment is complicated by two divergent needs: the need for empiric therapy with a broad-spectrum antibiotic, to aid in reducing mortality rates, and the need to avoid the indiscriminate use of antibiotics, to avoid the development of resistance. To address this complex issue, the strategy of de-escalation therapy was developed. With this treatment approach, a broad-spectrum antibiotic targeted to likely pathogens is administered, and the antibiotic regimen is altered, if necessary, after the results of cultures are known [226,227]. This strategy has reduced the mortality rate while achieving an overall objective of a more judicious use of antibiotics [226,228,229]. In one study, de-escalation therapy led to a significantly lower mortality rate compared with either escalation therapy or therapy that was neither escalated nor de-escalated (17% compared with 43% and 24%, respectively) [194].

It has been emphasized that this approach, and empiric treatment of healthcare-acquired pneumonia in general, calls for knowledge of the infection history of the healthcare facility and of individual patient units [190,195,230]. Microbiology laboratory reports can provide such details, and physicians should prescribe initial antibiotics that are likely to be active against these pathogens.

In patients with suspected ventilator-associated pneumonia, the IDSA and the American Thoracic Society recommend including coverage for S. aureus, Pseudomonas aeruginosa, and other gram-negative bacilli in all empiric regimens.

(https://www.idsociety.org/practice-guideline/hap_vap Last Accessed: January 20, 2022)

Strength of Recommendation/Level of Evidence: Strong recommendation, low-quality evidence

The IDSA/ATS guidelines provide several recommendations for the management of both hospital-acquired and ventilator-associated pneumonia [18]:

Specific treatment depends on the timing of onset and the presence or absence of risk factors for infection with multidrug-resistant organisms. For early-onset pneumonia and/or patients with no such risk factors, limited-spectrum antibiotic therapy is recommended (Table 12) [18]. For late-onset pneumonia and/or patients at increased risk for multidrug-resistant bacteria, a broad-spectrum antibiotic therapy is recommended.

For patients with ventilator-associated pneumonia, the IDSA and the American Thoracic Society recommend a seven-day course of antimicrobial therapy rather than a longer duration. There exist situations in which a shorter or longer duration of antibiotics may be indicated, depending upon the rate of improvement of clinical, radiologic, and laboratory parameters.

(https://www.idsociety.org/practice-guideline/hap_vap Last Accessed: January 20, 2022)

Strength of Recommendation/Level of Evidence: Strong recommendation, moderate-quality evidence

Ventilator-associated pneumonia is often caused by MRSA and gram-negative bacilli such as Acinetobacter spp. and Pseudomonas. Vancomycin has been considered the first choice for treatment of MRSA infections [226]. However, the ATS/IDSA guidelines note that linezolid may have advantages over vancomycin for ventilator-associated pneumonia caused by MRSA [18]. Linezolid has been compared with vancomycin for the treatment of pneumonia caused by MRSA in many studies and has been found to improve survival and to be more cost-effective [231,232,233,234]. In one study, the rate of early microbiologic cure was not significantly higher for linezolid than for vancomycin, although there were trends favoring linezolid in several secondary clinical outcomes, such as clinical cure; duration of ventilation, hospitalization, and stay in ICU; survival time not on a ventilator; and overall survival [235]. The findings led the authors to suggest that the benefit of linezolid may be related to factors other than bacterial clearance.

According to a meta-analysis, a short fixed-course (7 or 8 days) of antibiotic therapy may be more appropriate than a prolonged course (10 to 15 days) for patients with ventilator-associated pneumonia not caused by non-fermenting gram-negative bacilli [236]. The short course reduced the recurrence rate of ventilator-associated pneumonia caused by multiresistant organisms without adversely affecting other outcomes. Among patients with non-fermenting gram-negative bacilli, recurrence was greater after the short course. The authors confirmed these findings in a follow-up study published in 2015 [237].

For cases of ventilator-associated pneumonia caused by gram-negative bacilli that are susceptible only to aminoglycosides or polymyxins the suggestion is to use both inhaled and systemic antibiotics, rather than systemic antibiotics alone [18]. It is also reasonable to consider adjunctive inhaled antibiotic treatment as a last resort for patients who are not responding to intravenous antibiotics alone, whether the infecting organism is or is not multidrug resistant.

Adherence to guidelines for the prevention of ventilator-associated pneumonia is low, with surveys of nurses demonstrating rates of adherence to specific preventive measures ranging from 15% to 50% [192,196]. Adherence to a bundle of prevention strategies (head-of-bed elevation, oral chlorhexidine gel, sedation holds, and a weaning protocol), with 70% compliance, led to a significant reduction in ventilator-associated pneumonia, from 32 cases per 1,000 ventilator-days to 12 cases per 1,000 ventilator-days [54]. The IHI how-to guide on preventing ventilator-associated pneumonia provides several practical recommendations, and posting compliance with the ventilator bundle in a prominent place in the ICU can encourage and motivate staff (Table 13) [198].

The use of physician-led multidisciplinary rounds with team decision making, checklists, and a focus on the ventilator bundle has led to significant reductions in ventilator-associated pneumonia [238,239,240]. Moderate strength evidence has shown that the use of audit and feedback and reminder systems improve adherence to an overall ventilator-associated pneumonia bundle as well as reduce infection rates [1]. Education sessions have also led to enhanced knowledge and practice among healthcare professionals caring for intubated patients [197].

The lack of adherence to guideline-directed treatment of pneumonia cases associated with healthcare facilities is evidenced by wide variations in practice. For example, one study showed that more than 100 different antibiotic regimens had been prescribed as initial treatment and that de-escalation therapy was used for only 22% of patients [194]. Adherence rates for treatment of pneumonia associated with healthcare facilities have been reported to be lower than rates of adherence to guidelines for treatment of community-acquired pneumonia. In one survey, guideline-recommended antibiotics were used 9% of the time for healthcare-associated pneumonia compared with 78% of the time for community-associated pneumonia [241]. This lack of adherence was not due to unfamiliarity or disagreement with the guidelines; 71% of the survey respondents said they were aware of the guidelines, and 79% said they agreed with and practiced according to them. It is reasonable to expect that strategies used to enhance adherence to guidelines in the community-acquired pneumonia setting would also be effective in the setting of hospital-acquired and ventilator-associated pneumonia. Such strategies include feedback on performance, reminder systems, standardized order sets, and education emphasizing outcomes and cost-effectiveness.

There are several types of intravascular catheters, and the risk of intravascular device-related bloodstream infections varies according to type. These catheters include:

The nontunneled central venous catheter accounts for the majority of all intravascular device-related bloodstream infections [28]. Peripheral catheters (arterial and venous) are rarely associated with bloodstream infections, and totally implantable catheters are associated with the lowest risk [28]. A systematic review of 200 prospective studies of intravascular device-related bloodstream infections indicated that the level of risk associated with various types of devices can vary substantially depending on whether risk is expressed as the number of infections per 100 intravascular device-days or 1,000 intravascular device-days [244]. The risks associated with peripheral intravenous catheters were much higher when expressed over 1,000 intravascular device-days, pointing to the need for prevention strategies targeted to all types of devices [244].

Other risk factors are the length of time the catheter is in place and factors related to the patient's health status (severity of illness, presence of burns or surgical wounds, compromised immune system, nutritional status) [28,35].

Intravascular device-related bloodstream infections are transmitted by both endogenous and exogenous routes. The most common cause of infection related to short-term catheters is migration of skin organisms at the site of insertion, with the organisms traveling along the surface of the catheter and colonization at the catheter tip [28]. Direct contamination of the catheter or catheter hub by contact with hands or contaminated fluids or devices is another cause [28]. Hematogenous seeding from another focus of infection is a less common cause, and contamination of infusion fluid is rare [28].

The most commonly reported pathogens for intravascular device-related bloodstream infections in 2015-2017 were S. aureus (16%), Enterococcus spp. (13%), Candida spp. (13%), coagulase-negative staphylococci (11%), Klebsiella spp. (10%), E. coli (7%), and Candida glabrata (5%) [144].

The percentage of resistant S. aureus was 48% in 2015–2017 compared with 55% in 2009–2010 [144]. Among the drug-resistant pathogens with the greatest increases compared with data for 2007–2008 were multidrug-resistant E. coli (10% vs. 4%), extended spectrum cephalosporin-resistant E. coli (21% vs. 19%), and aminoglycoside-resistant P. aeruginosa (14% vs 10%) [144].

The CDC guidelines on the prevention of intravascular device-related bloodstream infections were published in 2002 and updated in 2011 and 2017 [28]. The most recent guidelines emphasize the following points [28]:

The CDC guidelines define maximal sterile barrier precautions as the use of a cap, mask, sterile gloves, sterile gown, and a sterile full-body drape during insertion of an intravascular device (level IB) [28]. A sterile sleeve should also be used to protect pulmonary artery catheters during insertion (level IB) [28].

The CDC guidelines recommend use of an antiseptic of 70% alcohol, tincture of iodine, or chlorhexidine gluconate solution with alcohol before insertion of peripheral venous catheters (level IB) and a >0.5% chlorhexidine preparation with alcohol before insertion of central venous catheters or peripheral artery catheters and during dressing changes (category IA) [28]. The guidelines note that chlorhexidine preparations with alcohol have not been compared with povidone iodine in alcohol and thus no recommendation can be made in this regard [28]. In a meta-analysis of eight studies (4,143 catheters, primarily central line catheters), the chlorhexidine solution was found to reduce the risk for bloodstream infection by 49% [245]. In a subsequent study, use of this solution led to a 1.6% decrease in the rate of bloodstream infection, a 0.23% decrease in the incidence of death, and a cost savings of $113 per catheter used compared with povidone-iodine solutions [246].

Most intravascular device-related bloodstream infections develop at the site of insertion, due to the density of skin flora [28]. Rates of infection vary according to insertion site, with catheters in the internal jugular vein being associated with a greater risk of infection than catheters in the subclavian vein [28,247,248]. A 2005 study indicated that the site of insertion was not a risk factor for infection when experienced or trained healthcare workers inserted the catheters [249]. However, such experience will not always be the norm, and the subclavian vein has been recommended by the CDC as the preferred site when possible [28].

Another strategy to prevent infection has been the development of central venous catheters with antimicrobial coatings. These coatings have included a combination of chlorhexidine and silver sulfadiazine and a combination of minocycline and rifampin [250]. Both types of catheters are associated with a significantly lower rate of infection than that associated with standard catheters. When compared with each other, catheters impregnated with minocycline and rifampin were 12 times less likely to cause bloodstream infections than those coated with chlorhexidine and silver sulfadiazine [251]. The chlorhexidine-silver sulfadiazine coating has since been enhanced, and these second-generation catheters have significantly reduced bacterial colonization, with a trend toward fewer bloodstream infections [252,253,254]. Each coating adds to the cost of the catheter, and cost-effective analyses are necessary.

On the basis of studies of intravascular device-related bloodstream infections, a bundle consisting of five preventive measures has been recommended:

As defined by the CDC, bloodstream infections fall into two categories: laboratory-confirmed infection and clinical sepsis. Clinical sepsis is no longer used in reporting on adults and children and is restricted to use for neonates and infants [127]. For a diagnosis of laboratory-confirmed bloodstream infection, one of the two following criteria must be met [219].

Criterion 1

Recognized pathogen found on one or more blood cultures and organism cultured from blood is not related to an infection at another site

Criterion 2

At least one of the following signs or symptoms:

and signs and symptoms and positive laboratory results are not related to an infection at another site and common skin contaminant (e.g., diphtheroids, Bacillus spp., Propionibacterium spp., coagulase-negative staphylococci, viridans group streptococcus, Aerococcus spp., or Micrococcus spp.) is cultured from two or more blood cultures drawn on separate occasions. Criterion elements must occur within a timeframe that does not exceed a gap of one calendar day [255].

There are several approaches to diagnosing an intravascular device-related bloodstream infection. A meta-analysis of 51 studies published between 1966 and 2004 was designed to identify which method was the most accurate [256]. The studies had involved the eight most commonly used diagnostic methods: culture (qualitative, semiquantitative, or quantitative) of a catheter segment; culture (qualitative or quantitative) of blood obtained through the catheter; paired quantitative cultures (blood obtained through the catheter as well as from a peripheral site); differential time to positivity (monitoring of cultures of blood obtained through the catheter and from a peripheral site); and acridine orange leukocyte cytospin. The paired cultures method was the most accurate, with a pooled specificity of 99%, followed by qualitative culture of blood drawn through the catheter and acridine orange leukocyte cytospin [256].

The management of an intravascular device-related bloodstream infection does not always include removal of the device. Authors of consensus-based treatment guidelines advise that the decision to remove a tunneled catheter or implanted device suspected to be the source of bacteremia or fungemia should be based on the following factors [257]:

The Centers for Disease Control and Prevention recommends using a chlorhexidine/silver sulfadiazine- or minocycline/rifampin-impregnated central venous catheter in patients whose catheter is expected to remain in place longer than five days if, after successful implementation of a comprehensive strategy to reduce rates, the central line-associated bloodstream infection rate is not decreasing.

(https://www.cdc.gov/hai/pdfs/bsi-guidelines-2011.pdf Last Accessed: January 20, 2022)

Strength of Recommendation/Level of Evidence: IA (Strongly recommended for implementation and strongly supported by well-designed experimental, clinical, or epidemiologic studies)

Nontunneled central venous catheters should be removed in most cases of bacteremia or fungemia [257]. Antibiotic therapy alone has resolved 80% of infections caused by coagulase-negative staphylococcal bacteria, but in cases of infection with S. aureus or Candida, infection has persisted when the catheter has been maintained [257,258].

One strategy was developed in an attempt to retain the catheter. With so-called antibiotic lock therapy, antibiotics are instilled through the catheter after injection of an anticoagulant, locking a high concentration of the antibiotic in the lumen [258]. This approach is used in combination with systemic antibiotic therapy, and the antibiotics used have included vancomycin, cefazolin, and clindamycin. Fluconazole and amphotericin B have been used occasionally for infection with Candida spp., and another flush solution (low concentrations of minocycline and EDTA) has demonstrated activity against staphylococci, gram-negative bacilli, and Candida spp. [258]. Early empiric antifungal therapy is important if infection with Candida is suspected, as delayed treatment has been associated with higher mortality [259].

The guidelines for treatment of intravascular device-related bloodstream infection suggest antimicrobial selection according to the known or suspected pathogen (Table 14) [257]. Empiric antibiotic therapy should be selected according to clinical features and context, with due consideration for the local prevalence of resistant pathogens. Vancomycin is recommended as initial treatment for coagulase-negative staphylococci and for S. aureus until the sensitivity to methicillin is determined, at which point treatment should be changed to semisynthetic penicillin if the identified pathogen is sensitive [257]. Vancomycin should not be used as a front-line treatment for infections with methicillin-sensitive S. aureus.

If fever or other signs of infection persist after removal of the catheter, the patient should be evaluated for complications such as septic phlebitis, distant abscess, and endocarditis. Bacterial endocarditis has been found in 25% of patients with intravascular device-related bloodstream infection caused by S. aureus[258]. The findings of one study suggested that patients with MRSA bacteremia and underlying chronic liver disease were at higher risk for endocarditis [260].

As is the case with guidelines for other HAIs, adherence to prevention guidelines is suboptimal. A survey of more than 500 hospitals showed that, while adherence to the two most strongly recommended prevention strategies—maximal sterile barrier precautions and antisepsis with chlorhexidine gluconate—was good at VA hospitals (84% and 91%, respectively), the rates were lower at non-VA hospitals (71% and 69%) [261]. Adherence to a combination of maximal sterile barrier precautions, chlorhexidine gluconate, and avoidance of central line changes was even lower: 62% and 44%, respectively [261]. In another study, central venous catheters were routinely changed to prevent infection in about 15% of hospitals [262].

Implementing a prevention bundle has significantly reduced intravascular device-related bloodstream infections: the decrease was from 5.9 per 1,000 catheter-days to 3.1 per 1,000 in one study and from 7.7 per 1,000 catheter-days to 1.4 per 1,000 in another [57,238]. A how-to guide developed by the IHI provides practical suggestions for implementing the bundle (Table 15) [263]. A checklist should be developed for use when inserting a catheter to ensure adherence to all prevention strategies [35].

Adherence to appropriate postinsertion care has been the focus of some studies. In one study, there were breaches in postinsertion care in 45% of cases [264]. The primary breaches were non-intact dressing (158 breaches per 1,000 catheter-days) and incorrectly placed caps and taps (156 breaches per 1,000 catheter-days) [264]. The rate of intravascular device-related bloodstream infection during the study period was 5.5 per 1,000 catheter-days [264]. In another study, nursing staff used a postinsertion care bundle that consisted of the following: daily inspection of the insertion site; site care if the dressing was wet, soiled, or had not been changed for 7 days; documentation of ongoing need for the catheter; proper application of a chlorhexidine gluconate-impregnated sponge at the insertion site; appropriate hand hygiene before handling the intravenous system; and application of an alcohol scrub to the infusion hub for 15 seconds before each entry [265]. Adherence to this bundle led to a significant decrease in intravascular device-related bloodstream infections, from 5.7 per 1,000 catheter-days to 1.1 per 1,000 catheter-days [265].

The availability of policies regarding prevention strategies is also lacking. Although 80% of 25 ICUs (10 hospitals) had written policies for insertion of central venous catheters, only 28% had a policy requiring maximal sterile barrier precautions, and 36% and 60% of the units required hand hygiene before accessing a central venous catheter or treating the exit site, respectively [262]. A formal educational program on catheter insertion was in place at 52% of the units [262]. Education, in the form of self-study modules with pretest and post-test, along with didactic lectures and integration of evidence-based guidelines have been associated with increases in adherence to recommended practices and decreases in bloodstream infections [266]. One systematic review included 27 interventional studies of central line insertion or maintenance or both in adult ICU settings with documentation of central line-associated bloodstream infection incidence per 1,000 catheter days [267]. Statistical significance was found in 26 of the 27 studies in terms of infection reduction, despite large variations in the length or type of educational intervention. The authors suggest that providing continuing education on infection prevention measures may improve post-insertion outcomes [267].

A systematic review demonstrated moderate strength of evidence for audit and feedback and provider reminder systems, along with base strategies [1].

The primary risk factors for infection with C. difficile are antibiotic use, older age, and hospitalization [33]. Exposure to antibiotic agents is the most modifiable risk factor, an association reported in more than 96% of hospitalized patients in one study [280]. Antibiotics increase the risk by suppressing or altering normal bowel microflora, thereby facilitating overgrowth of relatively dormant C. difficile organisms. Many antibiotics have been implicated, but fluoroquinolones, cephalosporins, carbapenems, and clindamycin have been found to confer high risk [33]. The likelihood of infection increases with longer hospitalizations, with a 15% to 45% risk of colonization among patients hospitalized for one to three weeks [280].

C. difficile is an exogenous infection that is transmitted person-to-person through the fecal-oral route and possibly via contact with contaminated environmental surfaces (e.g., bedding, commodes, bath tubs).

Guidelines developed by SHEA/IDSA in 2010, and updated in 2017 and 2021, offer recommendations for prevention, diagnosis, and management of C. difficile[33,281]. (The scope of the 2021 focused update is restricted to adults and includes new data for fidaxomicin and for bezlotoxumab, a monoclonal antibody targeting toxin B produced by C. difficile [281].)

Control measures include restriction of antibiotic use; isolation precautions for healthcare workers, patients, and visitors; and environmental cleaning and disinfection (Table 16) [33]. The guidelines note that the use of antibiotics should be minimized and that an antibiotic stewardship program should be developed and implemented by all hospitals [33]. Appropriate hand hygiene is essential, and soap and water should be used rather than alcohol-based handrubs, as alcohol is not effective at killing C. difficile spores [33]. Gowns, gloves, and contact precautions for the duration of diarrhea are also recommended. The guidelines suggest that removing environmental sources of C. difficile, such as replacing rectal thermometers with disposable ones, can help reduce the incidence of C. difficile infection. The guidelines also note that the following are not recommended: routine environmental screening for C. difficile (level III, C); routine identification of asymptomatic carriers for infection control purposes (level III, A); and use of probiotics to prevent infection (level I, B) [33].

According to the Infectious Diseases Society of America (IDSA) and the Society for Healthcare Epidemiology of America (SHEA), patients with at least three unexplained and new-onset unformed stools in 24 hours are the preferred target population for testing for Clostridioides difficile infection (CDI).

(https://www.idsociety.org/practice-guideline/clostridium-difficile Last Accessed: January 20, 2022)

Strength of Recommendation/Level of Evidence: Weak recommendation, very low quality of evidence

Infection with C. difficile is diagnosed on the basis of clinical findings and the results of laboratory testing [33]. C. difficile infection is defined as (1) the presence of diarrhea (passage of three or more unformed stools in up to 24 consecutive hours) and (2) positive results on stool testing for the presence of toxigenic C. difficile or its toxins or findings of pseudomembranous colitis on colonoscopy or histopathologic evaluation [33]. Diarrhea may be absent in up to 20% of patients with fulminant colitis or postoperative ileus [280]. Other symptoms include fever, nausea, vomiting, abdominal pain or tenderness, and loss of appetite, but these symptoms are found in about half of patients with the infection [33,268].

Diagnostic stool testing should be done only on unformed stool (level II, B), and testing for asymptomatic patients is not useful (level III, B) [33]. Repeat testing during the same episode of diarrhea is discouraged, as it does not provide clinically useful information (level II, B) [33].

Several diagnostic tests are available to detect C. difficile, and they vary in terms of sensitivity, specificity, and turnaround times. Stool culture is the most sensitive test, but it is not clinically practical because of the slow turnaround time [33]. The sensitivity of cell cytotoxicity assay has been reported to range from 67% to 100%, whereas enzyme immunoassay testing for toxins A and B has a sensitivity ranging from 63% to 94% and a specificity of 75% to 100% [33]. Enzyme immunoassay testing is rapid and less expensive than other tests but it is a suboptimal choice compared with cell cytotoxicity assay (level II, B) [33]. The SHEA/IDSA guidelines note that polymerase chain reaction testing appears to be rapid, sensitive, and specific, but more data on its usefulness are needed before it can be recommended for routine use (level II, B) [33].

According to the IDSA and the SHEA, fidaxomicin is recommended over vancomycin for an initial episode of C. difficile infection. Vancomycin remains an acceptable alternative.

(https://www.idsociety.org/practice-guideline/clostridioides-difficile-2021-focused-update Last Accessed: January 20, 2022)

Level of Evidence: Consensus Statement/Expert Opinion

The most important step in treating C. difficile-associated diarrhea is to discontinue the inciting antibiotic as soon as possible [33]. This approach alone will lead to resolution of diarrhea in approximately 15% to 25% of patients with mild infection [268,280]. Antibiotic treatment of the diarrhea should not begin until the culture or toxin assay results are known, as approximately 30% of hospitalized patients with antibiotic-associated diarrhea will have C. difficile infection [33]. However, if severe or complicated C. difficile infection is suspected, empirical treatment should be started as soon as the diagnosis is suspected (level III, C) [33]. The SHEA/IDSA guidelines recommend fidaxomicin rather than a standard course of vancomycin for an initial episode of C. difficile gastrointestinal infection, whether mild or moderately severe. Implementation of this recommendation depends upon available resources. Vancomycin remains an acceptable alternative [281]. For an initial episode of C. difficile, a dosage of fidaxomicin 200 mg orally twice daily for 10 days is recommended. Vancomycin 125 mg orally four times per day for 10 days is the recommended alternative regimen [281]. Table 17 outlines the guideline recommendations for treatment according to severity of illness [281].

For patients with a recurrent episode of C. difficile (within the last six months), the SHEA/IDSA panel conditionally recommends using bezlotoxumab as a co-intervention with standard-of-care antibiotics, rather than antibiotics alone [281]. Bezlotoxumab has been shown to be effective in preventing C. difficile infection when administered at any time before ending antibacterial treatment [282]. Patients with a primary episode and other risk factors for recurrence (e.g., 65 years of age and older, immunocompromised, severe C. difficile, C. difficile infection in the last six months) may benefit from receiving bezlotoxumab [281]. Bezlotoxumab was approved by the FDA in 2016 and was the first humanized monoclonal antibody effective against C. difficile toxin B. It is approved for the prevention of recurrent C. difficile infection in high-risk adults [281,283]. Bezlotoxumab is administered as a one-time infusion at a recommended dose of 10 mg/kg over 60 minutes [281,283]. In patients with a history of congestive heart failure, the FDA warns that bezlotoxumab should be reserved for use when the benefit outweighs the risk [281,283].

Surgical management may be necessary for severe, fulminant cases of C. difficile colitis. In such cases, subtotal colectomy with preservation of the rectum is advisable; alternatively, a diverting loop ileostomy with colonic lavage followed by antegrade vancomycin flushes may lead to improved outcomes [33].

In addition to infection-directed antibiotics, treatment of C. difficile infection also includes fluid replacement, and electrolyte normalization [284]. The use of antiperistaltic agents should be avoided, as they may obscure symptoms, lead to retained toxin, and precipitate toxic megacolon (level III, C) [33]. The use of probiotics has been suggested as an adjunct to antibiotic treatment, but a systematic review found insufficient data to support probiotics as adjunct therapy and no evidence to support its use alone [285].

The principles underlying the judicious use of antibiotics are the limitation of unnecessary antibiotics, obtaining timely culture and sensitivity data, selecting the most appropriate treatment, and prescribing the appropriate dose [312]. In addition, studies have shown that antimicrobial use can be decreased by using explicit criteria to identify patients with HAIs as well as those at highest risk for infection [313]. Antibiotic stewardship programs provide guidance to clinicians and have been shown to improve patient outcomes, reduce the burden of antibiotic resistance, and save healthcare dollars [2,314].

In their "Guidelines for Developing an Institutional Program to Enhance Antimicrobial Stewardship," SHEA and IDSA recommend that two core strategies of such a program are a prospective audit with intervention and feedback and formulary restriction and preauthorization [72]. Other elements of an effective antimicrobial stewardship program include [72]:

The goal of phase four of the CDC national HAI Action Plan is to "slow the emergence of resistant bacteria and prevent the spread of resistant infection" (Table 20) [2]. As previously stated, the HHS is working to update the plan with new indicator targets and data, new research and intervention efforts, and a review of the impact of the COVID-19 public health emergency on HAIs [2]. The CDC is coordinating stewardship activities among federal agencies, public health, and healthcare systems [2]:

The Agency for Healthcare Research and Quality (AHRQ) Safety Program for Improving Antibiotic Use engaged in establishing or improving antimicrobial stewardship programs in 389 ambulatory care practices throughout the United States [340]. This program involves webinars, audio presentations, educational tools, and office hours to engage stewardship leaders and clinical staff to address attitudes and cultures that challenge judicious antibiotic prescribing and incorporate best practices for the management of common infections. Over the period of one year (December 2019 through November 2020), antibiotic prescribing at participating clinics was cut overall by nearly 48%, while prescribing for acute respiratory infections was reduced by 37% [340].

The following case outlines an investigative process and illustrates that the source of an outbreak may be unusual [336].

A cardiac surgeon noticed a cluster of cases of sternal wound dehiscence among his patients who had had surgery. Specimens from the wounds were obtained for culture. Microbiologic evaluation indicated that the infections were predominantly caused by Enterobacter cloacae, and molecular typing and serotyping demonstrated that the isolates were similar. No infections had developed after operations the surgeon had performed at other hospitals. No breach in aseptic technique was identified. All of the infected patients had been operated on in the same operating room, and the environment was screened. No source was found. Further questioning of the surgeon's operative practice revealed one difference from other cardiac surgeons: he used semi-frozen sodium lactate solution to achieve cardioplegia. Swabbing of the freezer used for the solution identified E. cloacae of the same typing as that found in the wound infections. The hypothesis was that contamination of the freezer led to contamination of the ice/slush solution, and the micro-organism was transmitted to the patients. The freezer was replaced, a rigorous cleaning schedule was instituted, and no further cases have occurred.

The following are overviews of selected potential outbreaks. Identification and early action in the case of any of these outbreaks will limit the adverse effects.

Group A Streptococci

Most outbreaks of group A streptococci involve surgical wounds, and the source can usually be traced to an asymptomatic carrier in the operating room or on the wound care team [89,337]. Standard Precautions are sufficient if the wound is minor; if it is major, Contact Precautions should be instituted and followed for 24 hours after initiation of effective therapy [23]. The healthcare worker should receive antimicrobial therapy as appropriate and leave the setting until completion of therapy.

Pulmonary Tuberculosis

Dealing with pulmonary tuberculosis involves prompt identification of the disease and determining the susceptible individuals who were exposed to the patient before isolation [89]. Airborne Precautions should be instituted and remain in place until the patient is receiving effective therapy, is improving clinically, and the culture results for three consecutive sputum specimens, collected on different days, are negative. Comprehensive information is available in the CDC guidelines for preventing the transmission of tuberculosis in healthcare facilities [338].

Legionella

The source of HAI with Legionella pneumonia is usually contaminated water [89]. Implementation of Standard Precautions for the patient is sufficient [23]. Laboratory-based surveillance for nosocomial Legionella should be performed, and samples of tap water should be obtained for culture. If the culture is positive, it is best to obtain cultures from patients who have healthcare-associated pneumonia. There are more than 40 known types of Legionella species, but most outbreaks are caused by Legionella pneumophila serotypes 1 and 6.

Antibiotic-Resistant Micro-Organisms

Outbreaks of antibiotic resistance have involved MRSA, VRE, and, most recently, vancomycin-resistant S. aureus [339]. In such outbreaks, it is important to identify patients with colonization or infection early and isolate them or cohort them. Contact Precautions should be implemented and carried out until antibiotic therapy has been completed and cultures are negative [23]. The importance of adhering to proper hand hygiene and other elements of Contact Precautions should be emphasized. Healthcare workers who were involved with patients before isolation should be evaluated for colonization and infection and treated appropriately.

Other Outbreaks

The potential for other outbreaks or epidemics vary, and the CDC website, http://www.bt.cdc.gov, offers resources on emergency preparedness for outbreaks or epidemics caused by potential agents of bioterrorism, including anthrax and viral hemorrhagic fever. A Bioterrorism Readiness Plan template is also available (https://emergency.cdc.gov/bioterrorism/pdf/13apr99APIC-CDCBioterrorism.pdf). Many aspects should be considered when planning for bioterrorism preparedness, and each department of a healthcare facility can play an important role.