Hospital-acquired pneumonia (HAP) develops at least 48 h after hospital admission. The most common pathogens are gram-negative bacilli and Staphylococcus aureus; drug-resistant organisms are an important concern. Symptoms and signs are the same as those for community-acquired pneumonia, but in ventilated patients, pneumonia may also manifest as worsening oxygenation and increased tracheal secretions. Diagnosis is suspected on the basis of clinical presentation and chest x-ray and is confirmed by blood culture or bronchoscopic sampling of the lower respiratory tract. Treatment is with antibiotics. Overall prognosis is poor, due in part to comorbidities.

HAP includes ventilator-associated pneumonia (VAP), postoperative pneumonia, and pneumonia that develops in unventilated but otherwise moderately or critically ill hospitalized inpatients. It also includes the new category healthcare-associated pneumonia (HCAP), which refers to pneumonia acquired by patients in healthcare facilities such as chronic care facilities, dialysis centers, and infusion centers.

Etiology

The most common cause is microaspiration of bacteria that colonize the oropharynx and upper airways in seriously ill patients.

Risk factors

Endotracheal intubation with mechanical ventilation poses the greatest overall risk; VAP constitutes > 85% of all cases, with pneumonia occurring in 17 to 23% of ventilated patients. Endotracheal intubation breaches airway defenses, impairs cough and mucociliary clearance, and facilitates microaspiration of bacteria-laden secretions that pool above the inflated endotracheal tube cuff. In addition, bacteria form a biofilm on and within the endotracheal tube that protects them from antibiotics and host defenses.

In nonintubated patients, risk factors include previous antibiotic treatment, high gastric pH (due to stress ulcer prophylaxis or therapy), and coexisting cardiac, pulmonary, hepatic, or renal insufficiency. Major risk factors for postoperative pneumonia are age > 70, abdominal or thoracic surgery, and dependent functional status.

Pathogens

Pathogens and antibiotic resistance patterns vary significantly among institutions and can vary within institutions over short periods (eg, month to month). In general, the most important pathogen is Pseudomonas aeruginosa, which is especially common in pneumonias acquired in intensive care settings and in patients with cystic fibrosis, neutropenia, advanced AIDS, and bronchiectasis. Other important pathogens include enteric gram-negative bacteria (mainly Enterobacter sp, Klebsiella pneumoniae, Escherichia coli, Serratia marcescens, Proteus sp, and Acinetobacter sp) and both methicillin-sensitive and methicillin-resistant Staphylococcus aureus.

S. aureus, Streptococcus pneumoniae, and Haemophilus influenzae are most commonly implicated when pneumonia develops within 4 to 7 days of hospitalization, whereas enteric gram-negative organisms become more common with increasing duration of intubation. Patients with HAP due to S. aureus or gram-negative bacilli tend to be elderly or have serious circumstances, such as needing a ventilator, undergoing chemotherapy for cancer, or having chronic pulmonary disease.

Prior antibiotic treatment greatly increases the likelihood of polymicrobial infection, resistant organisms, particularly methicillin-resistant S. aureus, and Pseudomonas infection. Infection with a resistant organism markedly worsens mortality and morbidity.

High-dose corticosteroids increase the risk of Legionella and Pseudomonas infections.

Symptoms and Signs

Symptoms and signs in nonintubated patients are generally the same as those for community-acquired pneumonia (see Pneumonia: Symptoms and Signs). Pneumonia in critically ill, mechanically ventilated patients more typically causes fever and increased respiratory rate or heart rate or changes in respiratory parameters, such as an increase in purulent secretions or worsening hypoxemia.

Diagnosis

• Chest x-ray and clinical criteria (limited accuracy)
• Sometimes bronchoscopy, blood cultures

Diagnosis is imperfect. In practice, HAP is often suspected on the basis of the appearance of a new infiltrate on a chest x-ray that is taken for evaluation of new symptoms or signs or of leukocytosis. However, no symptom, sign, or x-ray finding is sensitive or specific for the diagnosis, because all can be caused by atelectasis, pulmonary embolism, or pulmonary edema and may be part of the clinical findings in acute respiratory distress syndrome. Alternative diagnoses should be sought, particularly in patients who have a pneumonia risk score < 6 (see Table 5: Pneumonia: Hospital-Acquired Pneumonia Risk Index).

Table 5
PrintOpen table
Hospital-Acquired Pneumonia Risk Index Factor Points Temperature (°C) ≥ 36.5 and ≤ 38.4 0 ≥ 38.5 and ≤ 38.9 1 ≥ 39 and ≤ 36 2 Blood leukocytes (μL) ≥ 4,000 and ≤ 11,000 0 < 4,000 or > 11,000 1 Band forms ≥ 50% 1 Tracheal secretions None 0 Nonpurulent 1 Purulent 2 Oxygenation: Pao2/Fio2, mm Hg > 240 or ARDS 0 ≤ 240 and no ARDS 2 Pulmonary radiography No infiltrate 0 Diffuse (or patchy) infiltrate 1 Localized infiltrate 2 Progression of infiltrate* None 0 Progression (heart failure and ARDS excluded) 2 Growth of pathogenic bacteria on tracheal aspirate culture* No, rare, or light growth 0 Moderate or heavy growth 1 Same bacteria as on Gram stain 1 *Criteria applicable 72 h after initial diagnosis. Score ≥ 6 suggests hospital-acquired pneumonia. Score < 6 suggests alternative process. Pao2/Fio2 = ratio of arterial O2 pressure to fraction of inspired O2; ARDS = acute respiratory distress syndrome. Adapted from Singh N, Rogers P, Atwood CW, et al: Short-course empiric antibiotic therapy for patients with pulmonary infiltrates in the intensive care unit. American Journal of Respiratory and Critical Care Medicine 162:505–511, 2000. Hospital-Acquired Pneumonia Risk Index Factor Points Temperature (°C) ≥ 36.5 and ≤ 38.4 0 ≥ 38.5 and ≤ 38.9 1 ≥ 39 and ≤ 36 2 Blood leukocytes (μL) ≥ 4,000 and ≤ 11,000 0 < 4,000 or > 11,000 1 Band forms ≥ 50% 1 Tracheal secretions None 0 Nonpurulent 1 Purulent 2 Oxygenation: Pao2/Fio2, mm Hg > 240 or ARDS 0 ≤ 240 and no ARDS 2 Pulmonary radiography No infiltrate 0 Diffuse (or patchy) infiltrate 1 Localized infiltrate 2 Progression of infiltrate* None 0 Progression (heart failure and ARDS excluded) 2 Growth of pathogenic bacteria on tracheal aspirate culture* No, rare, or light growth 0 Moderate or heavy growth 1 Same bacteria as on Gram stain 1 *Criteria applicable 72 h after initial diagnosis. Score ≥ 6 suggests hospital-acquired pneumonia. Score < 6 suggests alternative process. Pao2/Fio2 = ratio of arterial O2 pressure to fraction of inspired O2; ARDS = acute respiratory distress syndrome. Adapted from Singh N, Rogers P, Atwood CW, et al: Short-course empiric antibiotic therapy for patients with pulmonary infiltrates in the intensive care unit. American Journal of Respiratory and Critical Care Medicine 162:505–511, 2000.

Gram stain and culture of endotracheal aspirates are of uncertain benefit, because specimens are likely to be contaminated with bacteria that are colonizers as well as pathogens, and a positive culture may or may not indicate infection. Bronchoscopic sampling of lower airway secretions for quantitative culture seems to yield more reliable specimens, but the effect of this approach on outcomes is undetermined. Measurement of inflammatory mediators in bronchoalveolar lavage fluid may play a future role in diagnosis; eg, a concentration of soluble triggering receptor expressed on myeloid cells (a protein expressed and shed by immune cells during infection) > 5 pg/mL may help distinguish bacterial and fungal pneumonia from noninfectious causes of clinical and radiographic changes in ventilated patients. However, this approach requires further investigation. The only finding that reliably identifies both pneumonia and the responsible organism is a pleural fluid culture that is positive for a respiratory pathogen. Blood cultures are relatively specific if a respiratory pathogen is identified but are insensitive.

Prognosis

The mortality associated with HAP due to gram-negative infection is about 25 to 50% despite the availability of effective antibiotics. Whether death is due to underlying illness or to the pneumonia itself is uncertain. Women may be at greater risk of death. The mortality rate with S. aureus pneumonia is 10 to 40%, in part due to the serious circumstances with which it is associated.

Treatment

• Empirically chosen antibiotics active against resistant gram-negative and gram-positive organisms

If the diagnosis is suspected, treatment is with antibiotics that are chosen empirically based on local sensitivity patterns, specific patient risk factors, and the conditions noted in Table 2: Pneumonia: Community-Acquired Pneumonia in Adults .

Indiscriminate use of antibiotics is a major contributor to development of antimicrobial resistance. Therefore, treatment may begin with initial use of broad-spectrum drugs, which are replaced by the most specific drug available for the pathogens identified by culture. Alternative strategies for limiting resistance that have not proved effective include stopping antibiotics after 72 h in patients whose pulmonary infection scores (see Table 5: Pneumonia: Hospital-Acquired Pneumonia Risk Index) improve to < 6 and regularly rotating empirically chosen antibiotics (eg, q 3 to 6 mo).

Multiple regimens exist, but all should include antibiotics that are effective against both resistant gram-negative and gram-positive organisms. Options include

• A carbapenem (imipenem/cilastatin 500 mg IV q 6 h or 1 g q 8 h or meropenem 1 g IV q 8 h), monobactam (aztreonam 1 to 2 g IV q 8 h), or piperacillin/tazobactam 4.5 g q 6 h
• Ceftazidime 2 g IV q 8 h or cefepime 1 to 2 g q 8 to 12 h
• These drugs are given alone or combined with vancomycin 15 mg/kg q 12 h

Linezolid 600 mg IV q 12 h may be used for some pulmonary infections involving methicillin-resistant S. aureus. Daptomycin should not be used for pulmonary infections.

Prevention

Most measures focus on preventing VAP. Semiupright or upright positioning reduces risk of aspiration and pneumonia compared with recumbent positioning and is the simplest and most effective preventive method. Noninvasive ventilation using continuous positive airway pressure (CPAP) or bilevel positive airway pressure (BiPAP) prevents the breach in airway defense that occurs with endotracheal intubation and eliminates the need for intubation in some patients.

Continuous aspiration of subglottic secretions using a specially designed endotracheal tube attached to a suction device seems to reduce the risk of aspiration.

Selective decontamination of the oropharynx (using topical gentamicin, colistin, chlorhexidine, vancomycin cream, or a combination) or of the entire GI tract (using polymyxin, an aminoglycoside or quinolone, and either nystatin or amphotericin B) is controversial because of concerns about resistant strains and because decontamination, although it decreases incidence of HAP, has not been shown to decrease mortality.

Surveillance cultures and routinely changing ventilator circuits or endotracheal tubes have not been shown to decrease VAP.

Incentive spirometry is recommended to help prevent postoperative pneumonia.

Last full review/revision May 2008 by John G. Bartlett, MD