A healthcare-associated infection (HAI), or nosocomial infection, is defined as a localized or systemic condition resulting from the presence of an infectious agent(s) or its toxin(s) that develop in a hospital or other healthcare facility that were not present or incubating at the time of admission.¹ Healthcare-associated infections increase healthcare costs and contribute to extended intensive care unit (ICU) length of stay and increased morbidities. Mortality rates associated with healthcare-associated infections are significantly higher than those associated with community-acquired infections. While reports vary, recent data suggests that on a given day, 1 of every 25 inpatients at US acute care hospitals had at least 1 HAI.² Recent estimates suggest that the annual direct cost of HAIs to the healthcare system is approximately $6.65 billion.³ Risk factors for development of an HAI include length of hospital stay, the presence of a central catheter, presence in a critical care unit, and mechanical ventilation. The leading causes of HAIs include pneumonias, surgical-site infections (SSIs), gastrointestinal infections (including those caused by Clostridium difficile), catheter-associated urinary tract infections, and bloodstream infections.²

The mode of infection that develops in the hospital is thought to be due to either (1) autoinfection or an infection that was present with the patient on admission but without signs or symptoms of infection or (2) cross-contamination or the patient acquires an infective agent in the hospital and subsequently becomes infected. Controlling the infection also requires separating the source and cutting the modes of transmission. There are standard, droplet, airborne, contact, neutropenic, and bone marrow transplant precautions.

Standard precautions include the use of handwashing, gloves, mask, eye protection and/or face shield, gown, patient care equipment, environmental control, and occupational health to prevent injuries from needles, scalpels, and other sharp devices. Controlling the infection also requires separating the source and cutting the modes of transmission.

Droplet precautions prevent the transmission of respiratory particles larger than 5 microns in size, and can travel up to 3 feet away.⁴ Transmission can occur when healthcare workers’ (HCWs) mucosal membranes (eyes, nose, and mouth) come in contact with the patient’s respiratory secretions. Patients should be in private rooms, although cohorting of patients with the same infection is appropriate. Surgical masks are advised for healthcare workers who come in direct contact with the patient. Because of the closed circuit and ventilator filters, transmission of organisms from patients who are intubated is unlikely. However, a surgical mask should still be worn, particularly when the patient is being transported, because the endotracheal tube can get disconnected.

Airborne precautions address respiratory particles smaller than 5 microns that can remain suspended in the air. Patients should be in private rooms with negative pressure and a minimum of 6 to 12 air changes per hour.⁴ Healthcare workers must wear respirators with 95% filtering efficiency. In order to function reliably, these devices must not allow leakage. Therefore, HCWs must be fit-tested to ensure reliable protection. Transport of patients should be minimized; when necessary, the patient should be masked.

Patients on contact precautions warrant private rooms, and transmission occurs via direct contact or indirect contact with a contaminated object or person. Healthcare workers must perform hand hygiene and wear gloves, and if contact with the patient is anticipated or the patient has diarrhea, they should wear gowns. The gowns must be disposed of after leaving the room and healthcare workers must perform hand hygiene again.

Neutropenic precautions aim to prevent a neutropenic patient from acquiring an infection from the environment. The definition of neutropenia is institution dependent but generally includes an absolute neutrophil count (ANC) of less than 500 cells/µL. Neutropenic precautions include hand washing, nonspecific protective gear, avoiding plants or dried flowers, avoiding rectal temperatures, and avoiding suppositories.

Bone marrow transplant patients should be in a room with high-efficiency particulate arrestance (HEPA) filtration, which removes particles larger than 0.3 µ in diameter; directed, positive pressure airflow; and 12 air changes per hour to prevent invasive fungal infections.⁴ Hematopoietic cell transplantation (HCT) recipients with multidermatomal varicella zoster virus (VZV) should be placed in contact and airborne precautions. The contact precautions should be enforced until all skin lesions are crusted. The airborne precautions should be enforced from 8 days after exposure to VZV until 21 days from last exposure, or 28 days postexposure if the patient receives VZV immune globulin (VZIG) because patients are infectious before rash appears.⁴

Appropriate precautions for specific pathogens are summarized in Table 19-1.^4,5

Decolonization is used to prevent hospital acquired infections and target the nose, skin, and gastrointestinal tract. It is most effective for patients who are at risk of infection for a short period of time, such as surgical patients and ICU patients. Highest level of evidence is preventing surgical site infections.⁶ There are two approaches: horizontal approach which targets a broad range of microbes and vertical approach which targets a specific microbe. There is also universal versus targeted decolonization.⁶ Universal decolonization is for any high risk patient regardless of colonization status and targeted decolonization requires screening prior to decolonization. For 1 study, universal decolonization via intranasal mupirocin for 5 days and daily chlorhexidine impregnated cloths reduced MRSA clinical isolates and blood stream infections from any pathogens more effectively than targeted decolonization.⁷ Mupirocin is produced by Pseudomonas fluorescens and is affective against staphylococci, streptococci, and gram-negative organisms such as N. gonorrhoeae, H. influenza, and M. catarrhalis.⁶ Chlorhexidine has good coverage against gram-positive and gram-negative bacteria and yeast.⁶

Nasal agents for decolonization include mupirocin, bacitracin, retapamulin, povidone-iodine, and investigational agents (tea tree oil, photodynamic therapy, omiganan pentahydrochloride, lysostaphin).⁶ Topical agents include chlorhexidine gluconate, hexachlorophene, povidone-iodine, triclosan, and sodium hypochlorite.⁶ Although currently not recommended, oral agents that have been used for decolonization include rifampin, quinolones, trimethoprim-sulfamethoxazole, novobiocin, clindamycin, doxycycline, and minocycline.⁶

Hospital-acquired pneumonia (HAP) is pneumonia that develops 48 hours or more after admission and did not appear to be incubating at the time of admission. It was previously thought that patients with healthcare associated pneumonia (HCAP)—defined as pneumonia in patients from nursing homes, hemodialysis centers, or clinics, or who were hospitalized in the last 3 months—were also at high risk for multidrug-resistant organisms (MDRO). However, the Infectious Disease Society of America (IDSA) 2016 guidelines suggest that this risk has been overstated and that treating MDROs empirically as HAP does not improve outcomes.^8-10

Multidrug resistance, particularly with Gram-negative bacilli, is resistance to 2 or more antibiotics. Extensive drug-resistant Gram-negative bacilli are resistant to all antibiotics except for colistin, tigecycline, and aminoglycosides. Panresistance is resistance to all systemic antibiotics. Table 19-2 lists the risk factors.^8-10

In the 2016 IDSA guidelines, ventilator associated pneumonia (VAP) was noted to be a distinct entity from hospital-acquired pneumonia rather than a subset. Reported incidence varies, ranging from 10% to 25% with an all-cause mortality of 25% to 50%.^8,11 Ventilator-acquired pneumonia develops after more than 48 hours of mechanical ventilation, but diagnosis may be difficult. At present, the consensus is that VAP is suspected when there is clinical deterioration defined as an increase in positive end-expiratory pressure (PEEP) by 3 cm H₂O or an increase in the fraction of inspired oxygen (FIO₂) by 0.20 (a ventilator-associated condition) in association with aberration in temperature or white blood cell count. Possible VAP is present when these findings are found in concert with purulent secretions or isolation of a potential pathogen from respiratory cultures.^8,11

Prevention should focus on avoiding intubation when possible, minimizing sedation and avoiding benzodiazepines, maintaining and improving physical conditioning with early mobilization and exercise, minimizing pooling of secretions above the endotracheal tube cuff by means of endotracheal tubes with subglottic secretion drainage ports, and elevating the head of the bed to 30 degree to 45 degree.¹² Endotracheal tubes are now available with subglottic secretion drainage ports. The use of subglottic drainage may reduce the incidence of VAP by 55% and may further reduce antibiotic utilization. Current evidence supports the use of oral care with chlorhexidine in intubated patients, particularly for preventing postoperative respiratory infections in cardiac surgery patients. At this time, data suggests that these oral antiseptics may lower VAP rates but insufficient data is available to determine the impact on duration of mechanical ventilation or mortality.¹²

Initiation of antibiotics should be dependent on clinical judgment and not on biomarkers such as procalcitonin, C-reactive protein (CRP), and soluble triggering receptor expressed on myeloid cells-1 (sTREM-1).⁸ Typical pathogens include aerobic Gram-negative bacilli (eg, Escherichia coli, Klebsiella pneumoniae, Enterobacter species, Pseudomonas aeruginosa, Acinetobacter species) and Gram-positive cocci (eg, Staphylococcus aureus, including methicillin-resistant S aureus [MRSA], Streptococcus species). Nosocomial pneumonia due to viruses or fungi is significantly less common, except in the immunocompromised patient.¹³

Current guidelines recommend empiric therapy of HAP and VAP with antimicrobials providing coverage for S aureus, P aeruginosa, and other Gram-negative bacilli. MRSA should be covered with vancomycin or linezolid if MDR risk factors are present—particularly in patients who have received antibiotics within the last 90 days—or in hospital units with a high prevalence, greater than 10% to 20% of MRSA. Monotherapy for confirmed pseudomonal infections is recommended unless the patient is high risk for death, in septic shock, or has a structural lung disease such as bronchiectasis and cystic fibrosis. Dual therapy is then recommended. Aminoglycosides are generally avoided due to their high mortality risk, possibly greater than 25%.⁸ Inhaled antibiotics are reserved for Gram-negative bacilli that are only susceptible to aminoglycosides or polymyxins as a last resort. Tables 19-3 and 19-4 list the recommended empiric antibiotics for HAP and VAP.⁸

Treatment should be narrowed once culture results are available. Patients should be reassessed 72 hours after initiation of therapy for the possibility of narrowing the antibiotic regimen based on cultures and the clinical picture. The decision to de-escalate antibiotics can be particularly difficult in culture-negative HAP. However, current literature has found that patients with culture-negative HAP tend to have less severe disease. Serum procalcitonin may help support decisions to discontinue antibiotics but may have limited benefit since recent data suggests that treatment can be safely limited to 7 days regardless of pathogen.

Recent literature supports the use of respiratory cultures in guiding VAP therapy, particularly as a guide in the need for MDR coverage. The role of bronchoalveolar lavage (BAL) in collection of respiratory cultures is controversial. The use of bronchoscopic sampling does not improve mortality, duration of mechanical ventilation, or length of stay but may lead to more rapid de-escalation of antimicrobial therapy, potentially reducing antibiotic resistance. Current guidelines recommend noninvasive sampling (endotracheal aspiration [ETA]) rather than invasive sampling (eg, BAL); the yield of sputum culture via ETA is approximately 33% to 83%.^14,15 Bronchoalveolar lavage and ETA samples showing fewer than 10 epithelial cells and more than 25 polymorphonuclear leukocytes per low-power field are adequate for analysis. In the event of performing invasive methods, such as BAL or protected specimen brush (PSB), a BAL less than 10⁴ colony-forming units (CFU)/mL and a PSB less than 10³ CFU/mL are used as thresholds to withhold antibiotics to avoid unnecessary treatment, cost, and complications.³ Respiratory secretions growing Candida often indicate colonization and rarely require antifungal therapy.¹⁶ Currently, no evidence supports repeating sputum to ensure adequate treatment.