Chapter 35: Management of Pneumonias

Although the symptoms and signs of infectious pneumonia were known and described by both patients and physicians in ancient times, the pathophysiology of pneumonia has only recently become clear. Consequently, effective treatment for bacterial pneumonia has been available for <100 years. The presence of bacteria in the respiratory tissue of patients with pneumonia was first identified in the latter half of the nineteenth century. However, without effective antimicrobial medications, the morbidity and mortality from pneumonia remained high during the first half of the twentieth century. Thus, prior to the availability of penicillin in the 1940s, 80% of patients in the United States with bacteremic pneumococcal pneumonia died. The mortality rate dropped to approximately 20% after the clinical introduction of penicillin, and the current mortality rate for community-acquired pneumonia is approximately 4%. Even so, pneumonia remains the most common cause of infectious death in the United States. Also of concern, many of the once highly effective antimicrobial medications for the treatment of pneumonia have been rendered ineffective with the emergence of resistant organisms.

In clinical practice, infectious pneumonia is divided into two main categories: (1) community-acquired pneumonia (CAP); and (2) hospital-acquired pneumonia (HAP). Community-acquired pneumonia can be defined as any infectious pneumonia in patients who are not hospitalized, or who have been hospitalized for <48 hours and not admitted to a hospital or extended health care facility in the past 90 days. By contrast, hospital-acquired pneumonia includes pneumonias that develop in patients who have been hospitalized for >48 hours, or who have been admitted to hospitals or extended health care facilities in the previous 90 days. Categorizing pneumonia as either CAP or HAP is useful because the spectrum of likely infectious pathogens differs significantly between the two groups. Management decisions can thereby be tailored to the most likely pathogens in each setting, even before the potential isolation of a specific pathogen. Immunosuppressed persons are at special risk of developing pneumonia due to any of the organisms commonly associated with CAP or HAP, as well as to opportunistic pathogens that usually do not cause disease among immunocompetent patients.

The first step in management of CAP is prevention. Vaccines against three common CAP pathogens are available: influenza, Streptococcus pneumonia (pneumococcus), and Haemophilus influenzae. Seasonal influenza vaccination is recommended for children who are 6-23 months old, adults > 50 years of age, healthcare providers, and certain other high-risk individuals including those with significant medical comorbidities. Likewise, pneumococcal vaccination is recommended for adults ≥65 years of age and for certain other high risk individuals, again including those with significant medical comorbidities.

Evaluation of a patient with suspected CAP starts with a comprehensive history including the presence or absence of fever, chills, malaise, cough, sputum production, dyspnea, and chest pain among other symptoms. Physical findings compatible with pneumonia include tachypnea, increased tactile fremitus, dullness to percussion, and inspiratory crackles or egophony on chest auscultation (Chap. 14). Opacities or parenchymal consolidation on thoracic imaging studies support the clinical diagnosis (Figs. 35.1, 35.2; see also Chap. 15).

Because CAP by definition involves patients who develop pneumonia outside of a health care setting, one of the first steps in management is to decide the most optimal setting to provide safe treatment for the patient. For example, will outpatient treatment suffice, or does the patient need to be admitted to the hospital and if so, is intensive care unit (ICU) admission warranted? These questions are best answered by the initial assessment of severity of a patient's pneumonia. In this regard, a helpful tool is the CURB-65 score which is a refinement of an instrument originally set forth by the British Thoracic Society. CURB-65 is an acronym for the five components of the score: Confusion, Blood urea nitrogen level, Respiratory rate, Blood pressure, and age of 65 years or older (Table 35.1).

The risk of mortality increases with increasing values of a CURB-65 score. Thus, patients with a CURB-65 score of 0 or 1 are considered at low risk for complications and can be usually considered for ambulatory treatment. In this context, however, other clinical factors such as the patient's living environment and social support, as well as comorbidities need to be taken into account to assure the safety and efficacy of treatment. Admission to the hospital is recommended when the patient's CURB-65 score is ≥2 and many patients with scores of ≥3 will require admission to the ICU. The American Thoracic Society (ATS) and Infectious Disease Society of America (IDSA) have also developed consensus-related criteria for Severe CAP (Table 35.2). By these criteria, patients with ≥ one major criteria or ≥ three minor criteria require ICU admission.

Diagnostic tests to determine the specific organism responsible for the patient's CAP who will be treated as out-patients are optional. These tests include blood cultures, sputum Gram stain and culture, and urinary antigen tests for Legionella spp. and S. pneumonia (Chap. 19). Patients with a history of unusual environmental exposures, international travel within the past 2 weeks, pleural effusion, and severe underlying comorbidities such as active alcohol abuse, chronic liver disease, chronic obstructive lung disease, and asplenia, represent subpopulations in which such further testing for a specific organism is indicated. For patients with CAP who are admitted to the hospital, pretreatment blood cultures and sputum Gram stain and cultures are indicated if the patient has severe CAP or comorbidities. Patients requiring endotracheal intubation should have specimens obtained from the lower respiratory tract in the form of endotracheal aspirates or bronchoalveolar lavage (Chap. 18).

The choice of initial antibiotic(s) in CAP is guided by the anticipated susceptibility of the most likely pathogens. In some cases a pathogen will be identified via sputum smears, cultures, serologic or other diagnostic tests. Unfortunately, such results are rarely available when treatment needs to be initiated. At best, the identity of a pathogen is known within the first 24 hours, although the identification process usually takes longer. In many instances, a pathogen will never be identified. Clinically, the likely pathogens are determined by both the severity of the pneumonia and the specific comorbidities of the patient (Table 35.3). Across all severities of CAP, Streptococcus pneumoniae is the most commonly identified pathogen, whereas Mycoplasma pneumoniae, Chlamydia pneumoniae, and respiratory viruses are more common pathogens in patients who are less severely ill. In contrast, Staphylococcus aureus, Legionella spp., and gram-negative bacilli are more often found in patients with severe pneumonia. Specific comorbidities and the likely pathogens in CAP are shown in Table 35.4. Less common pathogens and associated clinical scenarios are listed in Table 35.5.

Physician task forces within the ATS and IDSA have developed evidence-based guidelines which suggest that initial antibiotic therapy for CAP be based on both the severity of pneumonia and the individual patient's specific comorbidities. In subjects who are to be treated as outpatients, a macrolide antibiotic is preferred, unless the patient has significant comorbidities in which case a respiratory fluoroquinolone is recommended. A a-lactam antibiotic plus a macrolide may be used in place of a respiratory fluoroquinolone. If the severity of pneumonia requires inpatient treatment, the preferred therapy is either a respiratory fluoroquinolone or a combination of a β-lactam antibiotic and a macrolide. Finally, for the severity ill patient with CAP that requires ICU admission, a β-lactam antibiotic or ampicillin-sulbactam in addition to a respiratory fluoroquinolone or macrolide is recommended. If a specific pathogen is identified, the antimicrobial regimen is subsequently tailored to the antimicrobial sensitivity of that organism.

Hospital-acquired pneumonia (HAP) is defined as a pneumonia that occurs 48 hours or more after admission to the hospital. HAP includes pneumonias that develop after a patient is intubated and placed on mechanical ventilation (ie, ventilator-associated pneumonia (VAP); Chap. 28). It also includes pneumonia which develops in patients who have resided in an extended care facility or hospital in the previous 90 days (eg, a healthcare-associated pneumonia). Moreover, pneumonia in patients receiving hemodialysis, chemotherapy, or wound care outside the home is also considered a healthcare-associated pneumonia. This definition encompasses these several groups of patients because each is at risk for infection with a similar spectrum of pathogens including multidrug-resistant (MDR) organisms. Current data suggest that HAP occurs in up to 1% of all hospitalized patients and in perhaps one-third of patients who are being mechanically ventilated (Chaps. 28 and 30). The clinical importance of HAP is underscored by the fact that as a disease category it has been shown to have an attributable mortality rate of up to 50%, despite optimal supportive care.

As with CAP, the management of HAP begins with prevention. Avoidance of endotracheal intubation, if possible by utilizing non-invasive mechanical ventilation with a facial mask, reduces the risk for development of HAP, as do processes that tend to reduce the duration of intubation. Maintenance of a semi-recumbent position with the head elevated 30°-45° above the supine position decreases the risk of aspiration of oral or gastric substances into the lungs, thereby reducing the incidence of HAP in ICU patients. Finally, appropriate hand hygiene by all medical staff and appropriate use of antibiotics are important in the prevention of HAP.

The development of fever, leukocytosis or leukopenia, purulent sputum, and respiratory distress generally herald the development of HAP. In addition, new or progressive pulmonary infiltrate(s) on chest imaging studies is generally required for the diagnosis of HAP (Fig. 35.3).

In this setting, a positive culture of sputum or tracheal aspirates corroborates the diagnosis of HAP and provides valuable guidance regarding the initial choice of antimicrobial treatment. In addition to imaging the chest and culturing of the respiratory tract, preferably via endotracheal aspirate or bronchoscopy, patients with suspected HAP require evaluation of their oxygenation via pulse oximetry or arterial blood gas sampling (Chaps. 17, 18, 19). Blood cultures also should be obtained. It is recommended that cultures of both the lower respiratory tract and blood be collected before starting or changing antibiotics.

Prompt initiation of appropriate antibiotic therapy is indicated in all patients suspected of having HAP, since mortality increases with delay of antibiotic therapy or with the initiation of inappropriate antibiotic therapy. The initial antibiotic regimen is chosen empirically based upon risk factors for specific organisms and local patterns of organism prevalence and antimicrobial resistance. The first consideration when choosing empiric therapy for HAP is to determine the risk of MDR pathogens. Patients are considered to be at low risk for MDR pathogens if HAP develops within the first 4 days of hospitalization and if they do not have other risk factors for MDR pathogens (Table 35.4).

Likely pathogens in these patients include Streptococcus pneumoniae, Hemophilus influenzae, methicillin-sensitive Staphylococcus aureus, and antibiotic-sensitive enteric gram-negative bacilli (Table 35.5). Accordingly, appropriate antibiotic choices include ceftriaxone, ampicillin/sulbactam, ertapenem, and those quinolones that show good activity against S. pneumoniae.

If patients are at risk for an MDR pathogen, or if HAP develops on or after the fifth hospital day or with a described risk factor (Table 35.4), then potential pathogens include Pseudomonas aeruginosa, extended spectrum beta-lactamase (ESBL) producing Klebsiella pneumoniae, Acinetobacter spp., or methicillin-resistant Staphylococcus aureus (MRSA). Correspondingly, appropriate initial antibiotic choices therefore include an anti-pseudomonal cephalosporin, a carbapenem, or a β-lactam/β-lactamase inhibitor combination plus an aminoglycoside or anti-pseudomonal fluoroquinolone plus vancomycin or linezolid (Table 35.6). These choices include "double antibiotic coverage" of gram-negative organisms and coverage of MRSA. Double gram-negative antibacterial coverage increases the likelihood that the organism will be susceptible to at least one of the antibiotics. Local antimicrobial resistance patterns should further be taken into account when choosing antimicrobial combinations.

After 48-72 hours, it is important that the patient with suspected HAP be clinically reevaluated. If the patient has improved and cultures identify a specific pathogen, the antibiotic regimen may be "de-escalated" to include a single agent to which the pathogen is sensitive. If the patient is clinically improved but cultures are negative, discontinuation of antibiotic coverage for HAP should be considered, especially if blood cultures are negative and the lower respiratory tract specimen is without a significant number of inflammatory cells.