64

Acute Pneumonia
GERALD R. DONOWITZ tion. Changes in fibronectin secretion and in binding characteristics of epithelium for various lectins occur as a response to underlying diseases. This may help to explain why colonization occurs in some clinical settings and not in others. Particles larger than 10 m are efficiently filtered by the hair in the anterior nares or impact onto mucosal surfaces because of the configuration of the upper airways and the nasal turbinates. The cough and epiglottic reflexes also keep large particulate matter from reaching the central airways. The trachea and conducting airways of the transbronchial tree are usually effective in entrapping particles from 2 to 10 m in size. The sharp angles at which the central airways branch cause particles to impact on mucosal surfaces, where they are entrapped by endobronchial mucus. Once entrapped, particles are removed by ciliated epithelium to the oropharynx. Epithelial cells, which line the conducting airways, submucosal glands, and alveoli, produce airway surface liquida complex mixture of proteins and peptides mixed with plasma transudate. Airway surface liquid contains lysozyme, lactoferrin, and secretory leukocyte proteinase inhibitor, all of which possess microbicidal activity. Respiratory epithelial cells produce other potent antimicrobial peptides including cathelicidins and -defensins. These peptides possess individual antimicrobial activity as well as synergistic antimicrobial activity with each other. In addition, the -defensins may act as chemokines for memory T cells and dendritic cells, thereby serving as a link between the innate and adaptive immune systems. Most bacteria are 0.5 to 2 m in size. This size particle may reach the terminal airways and alveoli. No mucociliary apparatus exists at this level, yet a variety of humoral and cell-mediated host defenses function here. The alveolar-lining fluid contains surfactant, fibronectin, IgG, and complement, all of which are effective opsonins. Surfactant is composed of several components (SP-A, SP-B, SP-C, SP-D) that serve to increase the microbicidal capacity of macrophages. These compounds may also affect free radical production and lymphocyte activity.5 SP-A and SP-D are collectinsa family of collagenous carbohydrate-binding proteins. These proteins bind a variety of organisms, including viruses, gram-negative and gram-positive bacteria, mycobacteria, and fungi, which may decrease their virulence or enhance phagocytosis by neutrophils and alveolar macrophages.6 Free fatty acids, lysozyme, iron-binding proteins, and defensins are also present and may be directly microbicidal. Phagocytic cells, including macrophages and neutrophils, play a major role in pulmonary host defense. Four distinct populations of macrophages exist in the lung and vary in their location and function.7 The alveolar macrophage is located in the alveolar lining fluid at the interphase between air and lung tissue. It serves as the resident phagocytic cell in the lower airway and is the first phagocyte encountered by inert particles and potential pathogens entering the lung via inspired air. Alveolar macrophages play several critical roles.3 As phagocytic cells, they can eliminate certain organisms. If the numbers of organisms increase beyond the macrophages capability to handle them or if the organisms involved are particularly virulent (e.g., Pseudomonas aeruginosa), the macrophage becomes a mediator of an inflammatory response by producing cytokines that recruit neutrophils into the lung.8 Interstitial macrophages are located in the lung connective tissue and serve both as phagocytic cells and antigen-processing cells. Dendritic cells derive from monocytes and are located within the epithelium of the trachea, conducting airways, terminal airways, alveolar septa, pulmonary vasculature, and visceral pleura. These cells are therefore positioned to interact with antigens in inhaled air. Dendritic

n 1901, Sir William Osler1 noted in the fourth edition of his book, The Principles and Practice of Medicine, that the most widespread and fatal of all acute diseases, pneumonia, is now Captain of the Men of Death. Over a century later, the prominence of pneumonia as a clinical entity remains. It is among the top ten most common causes of death among all age groups in the United States, the sixth leading cause of death in those 65 years or older, and the single most common cause of infection-related mortality.2 The clinical challenge of treating community-acquired pneumonia is the large number of microbial agents that can cause disease (Tables 64-1A-D), the difficulty in making a clinical and etiologic diagnosis, and the fact that no single antimicrobial regimen can cover all the possible etiologies. Because a specific etiologic diagnosis is often not possible at the time initial treatment is begun, the clinician must decide what empiric therapy is most appropriate. The increasing prevalence of antibiotic resistance among many of the most common pathogens has made this challenge more difficult. An understanding of the pathogenesis of the disease, evaluation of relevant data from a careful history and physical examination, recognition of common clinical patterns of infection, and information from the microbiology laboratory all aid in narrowing down the possible etiologic agents of pneumonia, thereby allowing reasonable therapy to be selected empirically.

Host Defenses and Pathogenesis

The lung is constantly exposed to the mixture of gases, particulate material, and microbes that constitute inspired air. In addition, organisms from oral secretions frequently seep down from the upper airways as a consequence of microaspiration. Yet, the lower airways usually remain sterile because of the defense mechanisms of the respiratory tract. The development of acute pulmonary infection indicates either a defect in host defenses, exposure to a particularly virulent microorganism, or an overwhelming inoculum. Infectious agents gain entry to the lower respiratory tract through aspiration of upper airway resident flora, inhalation of aerosolized material, and less frequently, metastatic seeding of the lung from blood. PULMONARY DEFENSE SYSTEMS The pulmonary defense system involves both innate and adaptive immunity, including anatomic and mechanical barriers, humoral immunity, cell-mediated immunity, and phagocyte activity (Table 64-2).3,4 The upper airways, including the nasopharynx, oropharynx, and larynx, are the sites first exposed to inhaled microorganisms. The nasal mucosa contains ciliated epithelium and mucus-producing cells. Mechanical clearance of entrapped organisms occurs through the nasopharynx via expulsion or swallowing. In the oropharynx, the flow of saliva, sloughing of epithelial cells, local production of complement, and bacterial interference from resident flora serve as important factors in local host defense. Secretory immunoglobulin A (IgA) is the major immunoglobulin produced in the upper airways and accounts for 10% of the total protein of nasal secretions. It possesses antibacterial and antiviral activity despite being a relatively poor opsonin. Despite some controversy, low IgA levels are probably not associated with increased bacterial infection. IgG and IgM enter the airways predominantly via transudation from the blood. Their roles in bacterial opsonization, complement activation, agglutination, and neutralization activity are similar to those noted in serum. Adherence of microorganisms to epithelial surfaces of the upper airways is a critical initial step in colonization and subsequent infec-

891

892

PART II

Major Clinical Syndromes

64-1A

TABLE

Causative Agents of Acute PneumoniaBacteria Uncommon Acinetobacter var. anitratus Actinomyces and Arachnia spp. Bacillus spp. Moraxella catarrhalis Campylobacter fetus Eikenella corrodens Francisella tularensis Neisseria meningitidis Nocardia spp. Pasteurella multocida Proteus spp. Pseudomonas pseudomallei Salmonella spp. Enterococcus faecalis Streptococcus pyogenes

64-1C

TABLE

Causative Agents of Acute PneumoniaFungi

Common Streptococcus pneumoniae Staphylococcus aureus Haemophilus influenzae Mixed anaerobic bacteria (aspiration) Bacteroides spp. Fusobacterium spp. Peptostreptococcus spp. Peptococcus spp. Prevotella spp. Enterobacteriaceae Escherichia coli Klebsiella pneumoniae Enterobacter spp. Serratia spp. Pseudomonas aeruginosa Legionella spp. (including L. pneumophila and L. micdadei)

Common Histoplasma capsulatum Coccidioides immitis Agents of Mucormycosis Rhizopus spp. Absidia spp. Mucor spp. Cunninghamella spp. Uncommon Aspergillus spp. Candida spp.

cells possess an enhanced capacity to capture, process, and present class II antigens. They can migrate to lymphoid tissue, where they can stimulate T-cell immune responses. Dendritic cells can also produce a variety of cytokines and chemokines, including interleukin (IL)-12, which serves to stimulate B-cell immune function.9 The intravascular macrophage is located in the capillary endothelial cells. These cells are actively phagocytic and remove foreign or damaged material entering the lungs via the bloodstream. Neutrophil recruitment is crucial for the inflammatory response in the lung. The mechanisms involved in the initial detection of organisms in the lung and the generation and subsequent resolution of a response to them are now being more clearly delineated.10-15 Microorganisms express molecular recognition patterns that are unique and different from those of the host. Pattern recognition receptor families such as Toll-like receptors are present on epithelioid cells, alveolar macrophages, dendritic cells as well as other cells that are located in strategic areas of the lung and either individually or in groups serve to recognize molecular patterns of invading organisms. This recognition leads to the generation of early response cytokines like tumor necrosis
TABLE

64-1B

Causative Agents of Acute PneumoniaViruses Adults Common Influenza A virus Influenza B virus Respiratory syncytial virus Human metapneumovirus Adenovirus types 4 and 7 (in military recruits) Uncommon Rhinovirus Enteroviruses Echovirus Coxsackievirus Epstein-Barr virus Cytomegalovirus Varicella-zoster virus Parainfluenza virus Measles virus Herpes simplex virus Hantavirus Human herpesvirus 6 Coronavirus (SARS)

Children Common Respiratory syncytial virus Parainfluenza virus types 1, 2, 3 Influenza A virus Uncommon Adenovirus types 1, 2, 3, 5 Influenza B virus Rhinovirus Coxsackievirus Echovirus Measles virus Hantavirus

factor- (TNF-) and IL-I that then activate transcription factors such as mitogen-activated protein kinase, phosphoinositide 3-kinase, nuclear factor kappa B (NF-B), and interferon-regulatory factors. These transcription factors serve as a common pathway for pattern recognition receptors and orchestrate the development of the inflammatory response by mediating the transcription of chemokines, adhesin molecules, and other cytokines. This signal cascade serves two purposes. The first is to generate and maintain the inflammatory response to recruit neutrophils into areas of microbial invasion. The other goal is to activate anti-inflammatory response mediators, which leads to the shedding of receptors, neutralization of cytokines, and inhibition of macrophage recruitment. These all serve to ensure that the inflammatory response is held in check and that noninvolved areas of lung are not injured. It is this balance of proinflammatory and antiinflammatory cytokines and effector molecules that allows for sterilization of an infected area of lung without gross destruction of lung tissue. Other lung parenchymal cells may also help regulate the inflammatory response.16 In addition to epithelial cells, interstitial macrophages, and dendritic cells, endothelial cells, pulmonary smooth muscle cells, and fibroblasts produce both proinflammatory (e.g., colonystimulating factors, chemokines) and anti-inflammatory (e.g., IL-10) factors. Cell-mediated immunity is central to adaptive immune responses in the lung and is especially important against certain pathogens, including viruses and intracellular organisms, that can survive within pulmonary macrophages (e.g., Mycobacterium, Legionella). The lung has lymphoid tissue where homing and differentiation of previously uncommitted cells to memory T and B cells occur.17 Most of the organized lymphoid tissue in the lung is located in follicles along the bronchial tree in bronchus-associated lymphoid tissues (BALT). The lymphoid centers are morphologically similar to Peyers patches in the intestine and are similarly associated with mucosal epithelium. Inhaled
TABLE

64-1D

Causative Agents of Acute Pneumonia Other Agents Nontuberculous Mycobacteria Rickettsia M. abscessus Coxiella burnetii M. avium complex Rickettsia rickettsiae M. kansasii Mycoplasma and Chlamydia M. chelonae Mycoplasma pneumoniae M. fortuitum Chlamydophila psittaci M. xenopi Chlamydia trachomatis M. simiae Chlamydophila pneumoniae (TWAR) M. scrofulaceum Mycobacteria M. malmoense Mycobacterium tuberculosis Parasites Ascaris lumbricoides Pneumocystis jirovecii Strongyloides stercoralis Toxoplasma gondii Paragonimus westermani

64 Acute Pneumonia

893

TABLE

64-2

Pulmonary Host Defenses Host Defense Mechanism* Nasal hair Turbinates Anatomy of upper airways Mucociliary apparatus IgA secretion Saliva Sloughing of epithelial cells Cough Bacterial interference Complement production Cough, epiglottic reflexes Sharp-angled branching of airways Mucociliary apparatus Airway surface liquid (lysozyme, lactoferrin, secretory leukocyte proteinase inhibitor) Dendritic cells Antigen processing and presentation Bronchus-associated lymphoid tissue (BALT)} stimulation of memory and effector T cells and B cells Immunoglobulin production (IgG, IgM, IgA)

Location Upper Airways Nasopharynx

Oropharynx

Conducting Airways Trachea, bronchi

Lower Respiratory Tract Terminal airways, alveoli

Alveolar lining fluid (surfactant, fibronectin, immunoglobulin, complement, free fatty acid, iron-binding proteins) Alveolar macrophages Neutrophil recruitment (pattern recognition receptorstranscription factor stimulationproinflammatory and anti-inflammatory cytokine and chemokine production) Dendritic cells Antigen processing and presentation Bronchus-associated lymphoid tissue (BALT)} stimulation of memory and effector T cells and B cells

*Aspects of native and adaptive immunity play a role throughout the respiratory tract. Major component of adaptive immunity and important in response to vaccines and prior infections. Major component of innate immunity.

antigens therefore are able to cross the epithelial surface and immediately encounter cells involved with antigen processing. Once these antigens are processed and presented, it is in the BALT that B and T lymphocytes localize and are stimulated to become memory cells and effector cells. Normal lung parenchyma usually contains few lymphoid cells, which represent only 5% to 10% of the total cell population. The lymphocytes present are memory cells located in the submucosa and lamina propria; effector cells located between epithelial cells and in the interstitium; and cells thought to be preactivated, awaiting stimulation by inhaled antigens in the alveolus. The majority of lymphocytes are T cells, with 35% to 45% representing a CD4 (helper, inducer) phenotype, and 18% to 32% representing a CD8 (suppressor) phenotype.18 Antigens inhaled into the alveolus and captured by antigen-presenting cells subsequently activate intra-alveolar lymphoid cells. These cells can stimulate the migration of memory lymphocytes into the area, leading to a localized accumulation of antigen-specific T and B lymphocytes, many of which possess effector cell function. As is true in other anatomic areas, binding of T cells to endothelium is a critical first step in the inflammatory process and is mediated by the interaction of leukocyte function-associated antigen (LFA)-1 integrins on the lymphocyte cell surface, with ligands exposed by endothelium in areas of inflammation (intercellular adhesion molecules 1 and 2 and vascular cell adhesion molecule 1). Expression of these ligands on pulmonary endothelium is upregulated by inflammatory mediators, such as IL-1, interferon-, and TNF-, and by bacterial lipopolysaccharides. Pulmonary lymphocytes are thought to shuttle between two functionally distinct lymphoid areas, the BALT, which can be viewed as the afferent limb where antigens first stimulate an immune response, and the lung parenchyma, where differentiated T and B cells participate in the inflammatory response, which can be viewed as the efferent limb. The increase in numbers of memory T cells after antigenic stimulation

may occur as a result of local proliferation or via migration of cells from BALT. Lymphocytes in the lung have three major roles: (1) antibody production, (2) cytotoxic activity, and (3) inflammatory mediator production. The lung contains a variety of cytotoxic T cells including natural killer cells (antigen nonrestricted), antibody-dependent cytotoxic cells, and antigen-restricted cytotoxic cells. Pulmonary T cells produce a large number of cytokines. Mouse models suggest that unstimulated T cells produce mainly IL-2. After stimulation and conversion to memory T cells, two distinct groupings of cytokines are produced. The helper T cell 1 (Th1) and helper T cell 2 (Th2) pattern of cytokine production noted in murine models occurs in humans, although it appears to be less restrictive. Th1 cells produce interferon-, IL-2, IL-6, and IL-10 and contribute to cell-mediated immunity, whereas Th2 cells produce IL-4, IL-5, and IL-10 and contribute to humoral immune function. Furthermore, IL-3, TNF-, granulocytemacrophage colony-stimulating factor, and chemokines are secreted by both Th1 and Th2 phenotypes. Th1 cells are involved in cellmediated inflammatory reactions, whereas Th2 cells stimulate antibody production, especially IgE, and stimulate eosinophil activity. IMPAIRMENT OF PULMONARY DEFENSES The defenses of the lung, when they are functioning normally, are extremely efficient in maintaining sterility of the lower airways. However, a number of factors are known to interfere with these defenses and predispose the host to infection. Alterations in the level of consciousness from any cause (stroke, seizures, drug intoxication, anesthesia, alcohol abuse, and even normal sleep) can compromise epiglottic closure and lead to aspiration of oropharyngeal flora into the lower respiratory tract.19 Cigarette smoke, perhaps the most common agent involved in compromising natural pulmonary defense mechanisms, disrupts both mucociliary function and macrophage activity.20

894

PART II

Major Clinical Syndromes

Alcohol not only impairs the cough and epiglottic reflexes but also has been associated with increased colonization of the oropharynx with aerobic gram-negative bacilli, decreased mobilization of neutrophils,21 abnormal phagocyte oxidative metabolism, and abnormal chemotaxis.22 Alcohol effectively blocks the TNF response to endotoxin, with decreased recruitment of neutrophils to the lung. Furthermore, alcohol enhances monocyte production of IL-10, a cytokine with antiinflammatory properties.23 Infections with Mycoplasma pneumoniae or Haemophilus influenzae may interfere with normal ciliary function.24 Viruses may actually destroy respiratory epithelium and may disrupt normal ciliary activity. Neutrophil function, including chemotaxis, phagocytosis, and stimulation of oxidative metabolism and alveolar macrophage function, may also be inhibited by certain viral infections.25 Sepsis associated with extrapulmonary infections may undermine lung defense mechanisms. In animal models, exposure to lipopolysaccharide or endotoxin decreases lung clearance of a bacterial challenge.26 Infection with human immunodeficiency virus (HIV) compromises many of the components of pulmonary host defense. Quantitative defects involve the naive CD4 T cells initially, with the memory CD4 T cells depleted more rapidly later in infection. Functional defects caused by the virus include impaired response to remote recall antigens, inhibited response to soluble antigen followed in time by decreased T-cell response to alloantigens and mitogens, impaired IL-2 and interferon- production, and decreased immunoglobulin production.27 In BALT, destruction of dendritic cells and degeneration of lymphoid follicles have been noted. Defective antigen presentation by dendritic cells has also been observed. Abnormal chemotaxis, phagocytosis, and oxidative metabolism in neutrophils of patients with acquired immunodeficiency syndrome (AIDS) have been described. Iatrogenic manipulations that bypass or interfere with the usual host defenses of the upper airways (endotracheal tubes, nasogastric tubes, and respiratory therapy machinery) all predispose to infection.28 A variety of commonly prescribed drugs have been shown to inhibit host defenses in vitro or in models, but the clinical significance of this is uncertain. These agents include aspirin,29 erythromycin,30 aminophylline,31 and proton pump inhibitors.32 A recent study documented that the rates of pneumonia increased in hospitalized patients if protonpump inhibitors or histamine type 2 (H2) receptor antagonists were used compared with hospitalized patients in whom these acidsuppressive medications were not used (4.9% and 2.0%, respectively).32a Other factors that impair pulmonary host defenses include hypoxemia, acidosis, toxic inhalations,33 pulmonary edema,34 uremia, malnutrition, immunosuppressive agents,35 and mechanical obstruction. Older adults are at increased risk for the development of pneumonia (see Chapter 314). Although numerous factors play an important role in this regard, including an increased number and increased severity of underlying diseases and an increased number of hospitalizations, there are age-related impairments in host defenses.36,37 Less effective mucociliary clearance and abnormal elastic recoil may lead to less effective coughing and clearing of the upper airways. Some populations of elderly patients have an increased incidence of microaspiration. Changes in humoral immunity and cell-mediated immune function have been documented in older persons, though their role in the development of infection remains unclear. Immune dysregulation has been shown to occur in the elderly, so that low-grade inflammation occurs in the lung in the absence of clinically detectable infection. Recurrent episodes of bacterial pneumonia suggest the presence of specific predisposing factors.37 In children and young adults, recurrent pneumonia is associated with defects in host defenses, including leukocyte function38 and immunoglobulin production.38-40 Congenital defects in ciliary activity, including the immotile cilia syndrome,41 Kartagener syndrome (ciliary dysfunction, situs inversus, sinusitis, bronchiectasis),42 Youngs syndrome (azoospermia, sinusitis, pneumonia), and cystic fibrosis, are other clinical entities associated with recurrent pneumonia in young persons. Structural lung abnormalities such as bronchiectasis and pulmonary sequestration are also important predisposing factors for both younger and older patient populations. As more has become known about the molecular basis of the

inflammatory response, it has become clear that a variety of genetic polymorphisms exist that are associated with predisposition for the development of pneumonia. It is important to recognize that these defects may be associated with a narrow range of potential pathogens which may aid in the identification of the defect.12,15 Although most congenital defects in host defenses appear in childhood, common variable hypogammaglobulinemia may first appear in adulthood with recurrent pneumonia. Acquired host defense defects are more varied and include malignancies (lymphoma, chronic lymphocytic leukemia, myeloma), infection (AIDS), and iatrogenic causes (immune suppression associated with solid organ or marrow transplantation or cancer chemotherapy). Underlying respiratory tract disorders, such as chronic obstructive pulmonary disease (COPD), bronchiectasis, adult-onset cystic fibrosis, bronchopulmonary sequestration, and tracheobronchomegaly, may occur with pneumonia. Bronchial obstruction due to intrinsic compression (adenocarcinoma) or extrinsic compression (lymphadenopathy due to sarcoidosis or malignancy) has also been associated with recurrent episodes of pneumonia. Underlying diseases that predispose to aspiration lead to an increased incidence of pneumonia. These may be associated with gastrointestinal diseases (tracheoesophageal fistula, esophageal diverticula, esophageal reflux, esophageal stricture), neuromuscular disorders (myasthenia gravis, dementia, amyotrophic lateral sclerosis), and cancer of the head and neck. Some systemic illnesses, including WeberChristian disease, chronic renal failure, diabetes, and sickle cell disease, have been associated with pneumonia.

Clinical Evaluation
HISTORY The history should define (1) symptoms consistent with the diagnosis of pneumonia, (2) the clinical setting in which the pneumonia takes place, (3) defects in host defense that could predispose to the development of pneumonia, and (4) possible exposures to specific pathogens. Respiratory symptoms are commonly encountered in primary care practices but usually are not caused by pneumonia. Of over 10 million visits reported associated with a chief complaint of cough, only 4% to 6% actually involved pneumonia.43 Although the prevalence of pneumonia may vary depending on the age of the patient population and the presence of comorbid diseases, pneumonia remains only one of many possible explanations for respiratory complaints. Therefore, a serious effort should be made to differentiate pneumonia from other clinical entities with which it may be confused. Although the clinical findings of pneumonia related to the respiratory tract should be sought, including cough, sputum production, dyspnea, and fever, it should be recognized that nonrespiratory symptoms are commonly present. These include fatigue, sweats, headache, nausea, and myalgia.44,45 With increasing age, both respiratory and nonrespiratory symptoms of pneumonia become less frequent. Unfortunately, symptoms at presentation elucidated by a careful history may not always be able to distinguish pneumonia from other respiratory problems.46 Specific etiologic agents of pneumonia have been associated with certain underlying diseases and patient populations. Mycoplasmal pneumonia occurs more often in younger people, but it may be a cause of pneumonia in older patients and require hospitalization.47 Gramnegative pneumonia tends to occur in older adults, especially those who are debilitated with comorbid diseases.48 Tuberculosis should be suspected in the homeless, those infected with HIV, those who come from developing countries where tuberculosis is prevalent, and those who have been exposed to others with the disease. Staphylococcal pneumonia classically has been noted during epidemics of influenza. Over the past 20 years, methicillin-resistant Staphylococcus aureus (MRSA) has grown in importance as an etiology of ventilatorassociated pneumonia (VAP). In addition, since 2002, a strain of MRSA which carries the staphylococcal cassette chromosome (SCC) IV and contains the gene coding for Panton-Valentine leukocidin, has emerged as an important cause of community-acquired pneumonia. Known as community-acquired MRSA or CA-MRSA, the organism

64 Acute Pneumonia

895

causes severe pneumonia with high mortality in relatively young populations without underlying comorbidities.49,50 Severe acute respiratory syndrome (SARS) is a pneumonia syndrome caused by a coronavirus, associated with travel to China, Taiwan, Hong Kong, Singapore, Vietnam, and Toronto, Canada.51 Pneumonia has been noted to occur with increased frequency in patients with a variety of underlying disorders, such as congestive heart failure, diabetes, alcoholism, and COPD.48,52 In one series of 292 patients with pneumonia, only 18% were found to have no underlying disease.48 Certain lifestyle factors have also been associated with an increased risk of pneumonia. These include cigarette smoking, alcohol use (especially in males), contact with children and pets, and living in a household with more than 10 people.53 A history of antecedent upper respiratory tract infection has been elicited in 36% to 50% of patients with acute pneumonia, especially in those with pneumococcal disease.54 Recent dental manipulations, sedative overdoses, seizures, alcoholism, or loss of consciousness for any reason should raise the suspicion of anaerobic infection caused by aspiration of oral contents.19 Special note should be made of the relationship between pneumonia and patients with COPD.55 Although well-controlled studies are lacking, it does appear that patients with COPD have an increased incidence of pneumonia. However, because the tracheobronchial tree is often colonized with Streptococcus pneumoniae and H. influenzae, it has been difficult to distinguish clearly between colonization and infection in many studies. Although these organisms play an important role as etiologic agents of pneumonia in this patient population, most of the clinical studies were carried out before it was recognized that other less common pathogens also play a significant role in causing disease. The roles of Moraxella catarrhalis, Legionella, Chlamydophila, and aerobic gram-negative rods including P. aeruginosa have been established.55-57 Cystic fibrosis is commonly associated with Pseudomonas and staphylococcal pulmonary infections.58 Burkholderia spp. and Stenotrophomonas spp. are also important pulmonary pathogens in this setting. Pulmonary alveolar proteinosis is associated with Nocardia infection. Patients infected with HIV are at high risk for the development of pulmonary infections.59-61 Although Pneumocystis pneumonia remains an important clinical entity in this patient population, prophylaxis has reduced its incidence and importance as a cause of death.62 In considering the nature of pulmonary infection in patients infected with HIV, geographic exposures, demographic characteristics of the patient, and the degree of immune suppression should be considered.59,60 Although in the past, Pneumocystis infection has been reported to be relatively uncommon in patients with AIDS in Africa, more recent data have suggested that the incidence is increasing. Overall, however, it is a more common complication of HIV infection in Europe and the United States. In many centers worldwide, Mycobacterium tuberculosis is now viewed as the major pulmonary pathogen in patients with AIDS.63 Though use of highly active antiretroviral therapy has led to a decreased incidence of disease, its overall importance in AIDS as a pulmonary pathogen remains. Because fungal infections play a major role in this patient population, exposure in endemic areas of histoplasmosis, blastomycosis, and coccidioidomycosis should be considered. The greatest risk appears to be associated with intravenous and inhaled drug use.59,64 Common pulmonary pathogens such as S. pneumoniae, H. influenzae, and S. aureus play an important role.60 The incidence of invasive pneumococcal pneumonia is 100 to 300 times higher in HIVinfected patients, and the incidence of H. influenzae infection may be up to 100-fold higher in this population. P. aeruginosa pneumonia is often associated with bacteremia or cavitary lung disease and is a later complication of HIV infection, occurring with CD4 counts of less than 25 cells/L. The infection is associated with the presence of central venous catheters and urinary catheters and the use of steroids. Other agents, such as Legionella, Nocardia, aerobic gram-negative bacilli, and Rhodococcus equi, are also important causes of pneumonia. In patients infected with HIV, the relationship between the degree of immune suppression using the CD4 count as a marker and the specific etiology of pneumonia deserves emphasis. Bacterial pneumonia and tuberculosis usually occur when the CD4 count is less than

400/L with increased risk when the count falls below 200 cells/L.64,65 Pneumocystis and disseminated tuberculosis are associated with CD4 counts below 200 cells/L, and disseminated nontuberculous mycobacterial infections and disseminated fungal infection occur with CD4 counts less than 50 to 100 cells/L.65 Pulmonary infections in HIVinfected patients are discussed in more detail in Chapter 122. Pneumonia developing in hospitalized patients often involves Enterobacteriaceae, P. aeruginosa, and S. aureus, organisms that are unusual in community-acquired disease.66 Pneumonia in older adults, especially those who are bedridden or who have chronic diseases, is more often associated with gram-negative bacilli than is pneumonia in younger populations.67 However, lack of clear definition between colonization and true infection has made the actual role of gramnegative organisms unclear. In general, S. pneumoniae, nontypeable strains of H. influenzae, and M. catarrhalis are important pathogens in older adults. Recently, it has been recognized that patients with outpatient contact with the health care system develop pneumonia with etiologic agents that may be seen in both community-acquired and nosocomial pneumonia.68-70 Increased importance of MRSA, aerobic gram-negative rods including Pseudomonas aeruginosa, and mixed aerobic/anaerobic organisms due to aspiration are associated with this new syndrome of health careassociated pneumonia (see discussion in later section, Pneumonia Syndromes). Important aspects of a patients history that may suggest specific potential infectious agents include occupation, exposure to animals, travel, and sexual history (Table 64-3). The presence of noninfectious pulmonary disease, such as tumors or pulmonary emboli, which may masquerade as pneumonia, may also be suggested by a careful history.

TABLE

64-3

Pneumonia: Etiology Suggested by Environmental History Environmental History Exposure to cattle, swine, horses, goat hair, raw wool, animal hides Possible agent of bioterrorism Exposure to cattle, goats, pigs; ingestion of unpasteurized dairy products; employment as abattoir worker or veterinarian Travel to West Indies, Australia, Guam, Southeast Asia, South and Central America Exposure to ground squirrels, chipmunks, rabbits, prairie dogs, rats Possible agent of bioterrorism Exposure to tissue or body fluids of infected animals during trapping, hunting, or skinning (rabbits, hares, foxes, squirrels) or to bites of an infected arthropod (flies, ticks) Handling or ingesting poorly cooked meat from an infected animal Possible agent of bioterrorism Exposure to birds (parrots, budgerigars, cockatoos, pigeons, turkeys) Exposure to wild rodents, dogs, cats, pigs, cattle, or horses, or exposure to water contaminated with animal urine Residence in or travel to San Joaquin Valley, southern California, southwestern Texas, southern Arizona, New Mexico Exposure to bat droppings or dust from soil enriched with bird droppings Exposure to infected goats, cattle, sheep, domestic animals, and their secretions (milk, amniotic fluid, placenta, feces) Exposure to contaminated aerosols (e.g., air coolers, hospital water supply) Exposure to infected dogs and cats Exposure to rodent droppings, urine, saliva Travel to area of outbreaks

Infectious Agent Anthrax

Brucellosis Melioidosis Plague Tularemia

Psittacosis Leptospirosis Coccidioidomycosis Histoplasmosis Q fever Legionnaires disease Pasteurella multocida Hantavirus SARS

SARS, severe acute respiratory syndrome.

896

PART II

Major Clinical Syndromes

PHYSICAL EXAMINATION Fever is reported to be present in 65% to 90% of patients with pneumonia. It may be sustained, remittent, or at times hectic. Fever patterns per se, however, are not useful for establishing a specific diagnosis. The temperature should be taken rectally to reduce error caused by rapid mouth breathing. Recording of postural changes in blood pressure and pulse rate is useful in assessing hydration and intravascular fluid volume. The pulse usually increases by 10 beats per minute for every degree (centigrade) of temperature elevation. A pulsetemperature deficit (e.g., a relative bradycardia for the amount of fever) should suggest viral infection, mycoplasmal infection, chlamydial infection, tularemia, or infection with Legionella. Cyanosis, a rapid respiratory rate, the use of accessory muscles of respiration, sternal retraction, and nasal flaring suggest serious respiratory compromise. Furuncles are rarely secondary to staphylococcal pneumonia acquired by the respiratory route, but they may signal a source of bacteremia with subsequent pneumonia via hematogenous spread, although this is uncommon. Herpes labialis is seen in up to 40% of patients with pneumococcal pneumonia.71 Bullous myringitis is an infrequent but significant finding in mycoplasmal pneumonia. The presence of poor dentition should suggest a mixed infection due to aspiration of anaerobes and aerobes that colonize the oropharynx. Although edentulous patients may develop anaerobic pneumonia as a result of aspiration, it is uncommon.72 Examination of the thorax may reveal splinting, or an inspiratory lag on the side of the lesion, that is suggestive of bacterial pneumonia. Early in the disease process, definite signs of pulmonary involvement may be lacking or may be manifest only as fine rales. Chest examination may reveal these early signs of pneumonia even though the chest film is normal. Evidence of consolidation (dullness on percussion, bronchial breath sounds, and E to A changes) is highly suggestive of bacterial infection, but may be absent in two thirds of patients ill enough to be hospitalized and may be absent more often in patients treated as outpatients.73 Patients with mycoplasmal or viral infection may exhibit few abnormalities on physical examination, despite the presence of impressive infiltrates on the chest film. The overall usefulness of the history and physical examination to detect the presence of pneumonia has been questioned.46,74 A great deal of interobserver variation exists in detecting the signs and symptoms.46 In one series, three examiners seeing the same patients could not consistently agree on the physical examination findings. The diagnosis of pneumonia could be made with a sensitivity of only 47% to 69% and with a specificity of 50% to 75%.75 Rare findings such as egobronchophony and asymmetrical chest movements have a high predictive value for pneumonia. Other findings are usually not helpful. The absence of any vital sign abnormalities (i.e., respiratory rate greater than 20 breaths/min, heart rate greater than 100 beats/min, and temperature greater than 37.8 C) has been associated with a less than 1% chance of a patients having pneumonia, assuming a pneumonia prevalence of 5% in the population under study.74 Others have questioned the importance of any of these specific findings on detecting the presence of pneumonia.46 The probability of detecting pneumonia varies with the patient population, the prevalence of pneumonia in that population, the threshold values for defining a vital sign as abnormal, and the ability of the clinician to detect abnormal physical findings. No single physical finding is particularly helpful in making a definite diagnosis. However, a constellation of cough, fever, tachycardia, and crackles raises the possibility of pneumonia being present to 18% to 42%. Therefore, although variable and nondefinitive, a complete history and physical examination may be extremely helpful in guiding the workup of pneumonia. DIAGNOSTIC TESTING The presence of pneumonia is suggested by clinical features derived from a careful history and physical examination and confirmed by radiographic imaging of the chest that shows an infiltrate. The role of

microbiologic tests to identify the specific etiology is an important though controversial element of care. Most empirical antibiotic regimens are successful in the therapy of community-acquired pneumonia, especially mild to moderates cases. Studies comparing empirical therapy to laboratory-guided, pathogen-directed care have shown no differences in efficacy, although increased side effects were noted in the patients receiving empirical therapy.76 Efforts to determine the specific etiology of community-acquired pneumonia are justified by the fact that they (1) may enable the clinician to narrow the antibiotic spectrum, and use fewer agents, thereby decreasing exposure of the patient to potential side effects and potentially reducing the development of resistance; (2) may aid in the specific antibiotic choice for an individual patient, depending on the specific epidemiology of infection and the specific resistance patterns of the locale; and (3) may identify pathogens not usually suspected and therefore not usually covered by empirical therapy. On a broader scale, identifying specific etiologies may help define new agents, trends in antibiotic resistance in established agents, and epidemiology of infectious outbreaks. Though even aggressive efforts to diagnose the cause of pneumonia using some of the most advanced techniques will yield positive results only 58% of the time, diagnostic yields are higher in sicker patients, where the information is most helpful.77 The most recent guidelines from the Infectious Diseases Society of America/American Thoracic Society (IDSA/ATS) have suggested diagnostic testing whenever the result is likely to change individual antibiotic management, or in patients where the diagnostic yield is thought to be greatest.78 EXAMINATION OF SPUTUM AND OTHER RESPIRATORY TRACT SAMPLES Microscopic examination and culture of expectorated sputum remain the mainstays of the laboratory evaluation of pneumonia despite ongoing controversy concerning their sensitivity and specificity. Of patients admitted to the hospital with community-acquired pneumonia, 40% to 60% will not be able to produce sputum. Of those that do, 45% to 50% of samples may be judged to be inadequate for further study because of oropharyngeal contamination.79-81 Many patients have received antibiotics before the studies are carried out, which drastically reduces the diagnostic yield. A variety of organisms cannot be detected by Gram stain, including species of Legionella, Mycoplasma, Chlamydia, and Chlamydophila. However, in patients who produce sputum of adequate quality to be examined (minimal or no oropharyngeal contamination) and who have not received prior antibiotics, diagnostic yields of 80% for sputum Gram stain have been reported in patients with bacteremic S. pneumoniae pneumonia.80 Despite its pitfalls, sputum Gram stain is noninvasive, can be carried out at no risk to the patient, and under the right circumstances may aid in the diagnosis and choice of empirical therapy in patients with community-acquired pneumonia.82 Examination of the sputum should include observation of the color, amount, consistency, and odor. Mucopurulent sputum is most commonly found with bacterial pneumonia or bronchitis. However, sputum of a similar nature has been described in one third to one half of patients with mycoplasma or adenovirus infections.83 Scant or watery sputum is more often noted with these and other atypical pneumonias. Rusty sputum suggests alveolar involvement and has been most commonly (although not solely) associated with pneumococcal pneumonia.84 Dark red, mucoid sputum (currant jelly sputum) suggests Friedlnders pneumonia caused by encapsulated Klebsiella pneumoniae (Fig. 64-1).85 Foul-smelling sputum is associated with mixed anaerobic infections most commonly seen with aspiration.72 To maximize the diagnostic yield of the sputum examination, only samples with minimal oropharyngeal contamination should be reviewed. As a guide, the number of neutrophils and epithelial cells should be quantitated under low power (100), with further examination reserved for samples containing 25 or more neutrophils and 10 or fewer epithelial cells. Samples with more epithelial cells and fewer neutrophils are usually nondiagnostic and should be discarded. The

64 Acute Pneumonia

897

Figure 64-1 Currant jelly sputum associated with Klebsiella pneumoniae pneumonia.

Figure 64-3 Expectorated sputum demonstrating a positive quellung reaction in a patient with pneumococcal pneumonia.

morphologic and staining characteristics of any bacteria seen should be recorded and an estimate made of the predominant organisms (Figs. 64-2 to 64-6). When no bacterial predominance exists, this should be noted as well. In the appropriate clinical setting, a predominance of gram-positive, lancet-shaped diplococci should suggest pneumococcal infection (see Fig. 64-2). When strict criteria for Gram stain positivity are used (predominant flora or more than 10 gram-positive, lancet-shaped diplococci per oil immersion field [1000], or both), the specificity of the Gram stain for identifying pneumococci has been shown to be 85%, with a sensitivity of 62%.86 The diagnostic yield of the sputum examination for pneumococci can be maximized by the use of the quellung reaction (see Fig. 64-3), although it is rarely used. Anticapsular antiserum reacts with capsular polysaccharide, and this may be seen as a distinctly outlined capsule. Because pneumococci may be part of the nasopharyngeal flora in 10% to 50% of healthy adults and often colonize the lower airways in patients with chronic bronchitis, identification of the organism does not mean that it is the cause of disease.87 However, it is our experience that the large number of pneumococci necessary to produce a positive Gram stain or quellung reaction is unusual in carriers. The Gram stain may reveal large numbers of organisms in patients with bronchitis. The sputum Gram stain is helpful to identify organisms other than pneumococci. Small gram-negative coccobacillary organisms are characteristic of H. influenzae (see Fig. 64-4). The sensitivity of the sputum Gram stain for detecting H. influenzae is usually less than that for S.

pneumoniae but has been reported to be up to 80%. Staphylococci appear as gram-positive cocci in tetrads and grapelike clusters (see Fig. 64-5). Organisms of mixed morphology are characteristic of anaerobic infection. Few bacteria are seen with legionnaires disease, mycoplasma pneumonia, and viral pneumonia. When the criteria for minimal oropharyngeal contamination is met and when a predominant morphology is observed, the sensitivity of the sputum Gram stain in matching organisms found in the blood was reported to be 85% in patients with community-acquired pneumonia. In contrast, sputum cultures positive for pneumococci are found in only 50% to 60% of patients with pneumonia and pneumococcal bacteremia.88 In some centers, sputum examination has been a useful means of diagnosing Pneumocystis pneumonia in patients with AIDS. The use of commercially available monoclonal antibodies or Giemsas, Gomoris methenamine silver, or toluidine blue O stain has led to a diagnosis in up to 50% of cases, making more aggressive diagnostic procedures unnecessary.89 Special sputum staining techniques are important in identifying other organisms such as mycobacteria (Fig. 64-7). Sputum culture as a means of diagnosing pneumonia is as controversial as the sputum Gram stain. Not all patients with pneumonia will produce sputum, and when they do, patients with bacteremic pneumococcal pneumonia have been reported to have negative sputum cultures in 45% to 50% of cases, even when large numbers of organisms have been noted on a Gram stain.88 Similarly, 34% to 47% of sputum cultures are negative with proven H. influenzae pneumo-

Figure 64-2 Expectorated sputum with gram-positive, lancetshaped diplococci from a patient with pneumococcal pneumonia.

Figure 64-4 Expectorated sputum with gram-negative coccobacillary forms from a patient with Haemophilus influenzae pneumonia.

898

PART II

Major Clinical Syndromes

Figure 64-5 Expectorated sputum with clusters of gram-positive cocci in a patient with Staphylococcus aureus pneumonia.

Figure 64-7 Expectorated sputum with acid-fast bacilli in a patient with Mycobacterium tuberculosis.

nia.90,91 Furthermore, sputum cultures have frequently been shown to yield more bacterial species than more invasive methods of obtaining respiratory tract secretions.92 Contamination with gram-negative bacilli from the oropharynx has been noted in 32% of sputum cultures. A lack of correlation between findings from sputum culture and findings from blood cultures and serologic studies has been observed. Several key parameters have been identified in efforts to maximize the diagnostic yield from sputum culture. Procurement of adequate sputum samples is an essential first step. With increasing numbers of epithelial cells and decreasing numbers of neutrophils, an increased amount of oropharyngeal contamination is present, as indicated by the isolation of more bacterial species. The presence of alveolar macrophages does not alter the bacteriologic findings when substantial numbers of epithelial cells are present, indicating that otherwise adequate samples of sputum can be contaminated with oropharyngeal contents and thereby rendered nondiagnostic. This type of initial screening has proved helpful in differentiating adequate sputum samples from saliva, thereby increasing the diagnostic yield of sputum culture. When organisms such as M. tuberculosis (see Fig. 64-7), Legionella, and Pneumocystis are found in sputum, clinical infection is indicated regardless of the sputum quality because these organisms are not normal flora. When culture of sputum is delayed, the isolation of pneumococci is less likely because of overgrowth by oropharyngeal flora. Rapid processing of samples is therefore another important factor leading to

Figure 64-6 Expectorated sputum with gram-negative rods in a patient with Klebsiella pneumoniae pneumonia.

higher diagnostic yield. Some reports suggest that with adequate sputum samples and prompt culture of specimens, the diagnostic yield of the sputum culture may be improved. Antigen detection in respiratory secretions has been used for over 2 decades to try to maximize the diagnostic yield of sputum, especially for infections caused by S. pneumoniae, Pneumocystis, Legionella pneumophila, and a variety of respiratory viruses. The direct fluorescent antibody assays for L. pneumophila and Pneumocystis jirovecii are the most commonly used, with sensitivities of 25% to 75% for Legionella and 80% for Pneumocystis. Specificities of 90% have been reported in the assays for each pathogen. Strains of Legionella other than L. pneumophila may be missed in these assays. For other organisms, such as Chlamydophila, problems with colonization versus infection, varying sensitivities, and cross-reactivity with nonpathogens have limited the usefulness of the study. Although direct fluorescent antibody tests have been used to detect Chlamydia trachomatis, the assay is insufficiently sensitive (varying between 20% and 60%) for detection of Chlamydophila pneumoniae.93 On the other hand, direct fluorescent antibody assays are over 80% sensitive and 90% specific when used for the detection of Pneumocystis. Detection of microbial nucleic acid in respiratory tract secretions remains an area of ongoing study.94-97 Nucleic acid amplification assays, especially polymerase chain reaction (PCR) are particularly attractive because they have the capability of detecting minute amounts of material from potential pathogens, do not appear to be influenced by prior antibiotic therapy, and can be performed quickly. Although a variety of PCR techniques have been described, FDA licensed assays exist for only M. tuberculosis and Legionella species. Assays for more commonly encountered organisms such as S. pneumoniae, H. influenzae, and Chlamydophila have been developed, but lack of standardization and difficulty in determining true infection from colonization remain problematic. False-negative results have been reported because of the presence of natural inhibitors. Although PCR has been used to detect P. jirovecii, published studies have detected positive results in the setting of negative cultures and absence of clinical features of infection. Individual PCR assays as well as multiplex assays have been used to detect viruses. It is unclear if positive results indicate upper rather than lower respiratory tract infection, colonization, or true infection of the lung. PCR has become a mainstay in the diagnosis of SARS.96 PCR techniques have been used to identify DNA from M. tuberculosis in both sputum and lavage fluid. Sensitivities of 83.5% and specificities as high as 99% have been noted. Sensitivities of 63% have been reported in patients who were smear negative but culture positive. However, PCR has been reported to be positive in 70% of people with prior exposure to tuberculosis but no active disease. This problem of false-positive results remains.

64 Acute Pneumonia

899

FIBER-OPTIC BRONCHOSCOPY Although the sputum examination should always be included in the initial evaluation of patients with pneumonia, it may be inadequate for a presumptive diagnosis. In cases in which (1) no sputum is produced, (2) no clear predominance of a potential pathogen exists on sputum Gram stain or culture, (3) there has been a poor response to antibiotics chosen on the basis of expectorated sputum, (4) gram-negative rods or yeast forms are found in the sputum, or (5) the possibility of superinfection exists, a more direct method of obtaining lower respiratory tract secretions may be necessary. Fiber-optic bronchoscopy is usually not performed in patients with community-acquired pneumonia unless severe pneumonia, unresolving pneumonia, or a clear failure of antibiotic therapy is encountered.98 Use of the protected brush catheter and quantitative culturing of material obtained from the procedure have both minimized the problem of oropharyngeal contamination and helped to differentiate colonization from true infection. Approximately 106 to 108 organisms per milliliter (mL) are present in lung tissue involved with pneumonia. Accounting for dilution of samples, a bacterial count of more than 103 to 104 has been used as a breakpoint for determining the clinical significance of an isolate. Although sensitivities of 70% to 97% and specificities of 95% to 100% have been reported in the diagnosis of community-acquired pneumonia, bronchoscopy has proven to be less useful in some patient groups.99 These include patients who have already received antibiotics, patients with purulent bronchitis in whom bacterial counts greater than 103 are noted, and patients with underlying structural disease in whom over 50% of bronchoscopic specimens yield significant numbers of organisms even in the absence of pneumonia.99 Gram stain of specimens obtained from fiber-optic bronchoscopy has been used as a guide to empirical therapy while cultures are pending. A positive Gram stain predicts growth of more than 103 colony-forming units (CFU)/mL with up to 78% sensitivity. When studied prospectively early in the course of community-acquired pneumonia, bronchoscopy has yielded a diagnosis in approximately 50% of patients.100 Bronchoscopy with a protected specimen brush has been shown to have sensitivities as high as 82% to 100% and as low as 36% with specificities as high as 60% to 77% and as low as 50%.101-103 Differences in exclusion and inclusion criteria, different definitions of pneumonia, and the acceptance or rejection of patients with recent antibiotic changes may explain the different results.104 The use of antibiotics markedly diminishes the diagnostic yield of the procedure.105 Most bacterial species initially found by a protected specimen brush are undetectable after 72 hours of antibiotic therapy, and the majority of organisms found are resistant to the antibiotics given. These may have no role in the infection. However, in a patient with ongoing pneumonia despite antibiotic therapy, bronchoscopy with a protected specimen brush should pick up resistant organisms that may be playing a role in infection.101 False-negative findings are seen in up to 30% to 40% of patients, which may reflect the fact that bacterial counts may differ by fiftyfold in areas of infected lung versus noninfected adjacent areas, making the sampling site an important consideration. Other possible explanations include prior antibiotic use, technique problems, and in some cases an early stage of pneumonia where bacterial numbers are not yet high enough to reach the breakpoint of the procedure. Bronchoalveolar lavage (BAL), in which a segment of the lung is washed with sterile fluid, is useful in establishing the etiologic agent of pneumonia. Approximately 100 million alveoli are sampled, and consequently a larger area of lung is evaluated than with the protected specimen brush. A diagnostic threshold of 104 CFU/mL is used with lavage because the procedure recovers 5 to 10 times more organisms than brushing. The most consistent results have been seen in the diagnosis of Pneumocystis pneumonia in patients with AIDS. Diagnostic yields of 89% to 98% have been reported.106 Excellent yields have been noted in detecting cytomegalovirus in patients with AIDS as well as in bone marrow and solid organ transplant recipients.107 Although the

isolation of the organism does not prove that it is the cause of the pneumonia. BAL has proved especially useful for diagnosing pneumonia caused by M. tuberculosis.108 Culture of BAL material has a sensitivity of 85%, even in the presence of negative cultures of expectorated sputum and gastric aspirate. In patients with miliary tuberculosis in whom sputum culture yields are low (25%), culture of M. tuberculosis from BAL fluid approximates 100%. In addition to culture and staining, adenosine deaminase levels and enzyme-linked immunosorbent assay (ELISA) for antibodies to M. tuberculosis have been studied.109,110 BAL has also been used for the diagnosis of atypical pneumonias, including those caused by Legionella species and M. pneumoniae. Both bronchoscopy and BAL have been used widely in patients with ventilator-associated pneumonia (VAP).111,112 The bacteriologic strategy used in attempting to make an etiologic diagnosis in patients with VAP uses bronchoscopy, BAL, and endobronchial aspiration to establish the presence or absence of pulmonary infection as well as to determine the specific etiology.69 Although the risks of bronchoscopy are relatively small, hypoxia occurs in 13% to 28% of patients on ventilators undergoing BAL. In patients with gram-negative pneumonia, the procedure may be followed by a sepsis-like picture with increased temperature and decreased mean arterial pressure. OTHER TECHNIQUES A variety of less invasive techniques have been used in attempts to determine the cause of pneumonia without resorting to bronchoscopy. Blind endotracheal suctioning with quantitative cultures has compared favorably with bronchoscopic procedures in some studies.113 With a threshold of greater than 105 CFU/mL, the sensitivity for predicting ventilator-associated pneumonia was comparable to that of lavage or protected brush procedures, although the specificity was somewhat lower.113 Furthermore, no differences in mortality, length of stay in the intensive care unit, or duration of mechanical ventilation were noted when quantitative endotracheal cultures were used as the sole means of diagnosis compared to bronchoalveolar lavage and to protected specimen brush. Others have reported false-negative rates of over 30% and many more organisms isolated by endotracheal suctioning than by brushing.114 At present, this procedure is best used when bronchoscopy cannot be done. The blind protected specimen brush has compared favorably with bronchoscopically guided procedures, with 86% agreement.115 Nonbronchoscopic bronchoalveolar lavage has been obtained in some cases by a protected catheter to minimize contamination. Sensitivities of 70% to 80% and specificities of 66% to 96% have been noted.116 Mortality from ventilator-associated pneumonia is unchanged independent of whether bronchoscopic or nonbronchoscopic procedures are used for diagnosis.117 LUNG BIOPSY Direct means of obtaining diagnostic material in patients with pneumonia include percutaneous lung aspiration, transbronchial lung biopsy, thoracoscopy, and open lung biopsy.118 These procedures are usually reserved for cases of pneumonia in impaired hosts and in pediatric populations, in whom sputum is not routinely available. Biopsy procedures are rarely indicated in the previously well patient with acute pneumonia. The indications and usefulness of these invasive procedures remain controversial. Lung aspiration has provided a diagnostic yield of 30% to 82% in adults and children, although falsenegative rates of up to 18% have been reported.119 Bleeding and pneumothorax have been reported as major complications in 5% to 39% of procedures.119 The use of transbronchial biopsy in the diagnosis of pneumonia has been reviewed, revealing similar diagnostic yields although somewhat lower complication rates.120 Thoracoscopy, in which the pleura and underlying lung are visualized before biopsy, has been used in several series of children and

900

PART II

Major Clinical Syndromes

adults with pneumonia. Despite a diagnostic yield of over 90% and low complication rates,121,122 there has not been extensive experience with this procedure. Open lung biopsy remains the definitive invasive procedure for making an etiologic diagnosis of pneumonia in immunosuppressed patients, with diagnostic yields of 60% to 100%.123 The incidence of pneumothorax and bleeding is usually less than 10%, even in patients who are thrombocytopenic.120 However, some have questioned whether open lung biopsy provides meaningful information that significantly affects a patients clinical outcome. EXAMINATION OF PLEURAL EFFUSIONS The characteristics of pleural effusions and their importance in the differential diagnosis of pulmonary disease are discussed in Chapter 65. Pleural effusion or parapneumonic effusion will occur in 20% to 40% of hospitalized patients with pneumonia.124 The incidence of pleural effusions associated with pneumonia varies with the etiologic agent, from approximately 40% to 57% with pneumococci, to 50% to 70% with gram-negative bacilli, to up to 95% with group A streptococci.84,125 Pleural fluid cultures, when positive, are specific for the organism causing the underlying pneumonia. Furthermore, analysis of pleural fluid may play a major role in determining when drainage is necessary as well as differentiating other causes of pulmonary infiltrates that may mimic bacterial pneumonia, including tuberculosis, tumors, pulmonary emboli, and collagen vascular diseases. If neutrophils are not the predominant cell type seen in the pleural space, a diagnosis other than bacterial pneumonia should be sought. Pleural biopsy specimens from patients with acute bacterial pneumonia are nonspecific and are therefore of little use in the differential diagnosis. Parapneumonic effusions can be divided into three stages.124,125 The first stage or exudative stage is culture negative, has a pH greater than 7.2, glucose level greater than 60 mg/dL, and an LDH level that is less than three times normal. This stage is due to pulmonary interstitial fluid entering the pleural space and increased permeability of the capillaries in the pleura. These uncomplicated pleural effusions usually resolve with therapy for the underlying disease. Without appropriate therapy, pleural effusions become infected with the organisms causing the underlying pneumonia and develop into the second stage or fibropurulent stage. This is associated with positive microbial cultures, pH less than 7.2, glucose level less than 60 mg/dL, and LDH greater than three times normal. This is now a complicated pleural effusion which requires drainage. The most sensitive finding in determining if a pleural effusion needs drainage is a pleural fluid pH less than 7.2. This usually occurs before the other chemical parameters associated with complicated pleural effusions develop.124 If pH is used to determine whether an effusion is to be drained, it must be measured with a blood gas machine and not a pH meter or pH indicator strip, which are inaccurate. If left untreated, fibropurulent pleural effusions will develop into stage three effusions where a thick pleural rind is formed, restricting normal lung expansion. Empyema is defined as pus in the pleural space and represents a late manifestation of complicated pleural effusions. The presence of empyema mandates draining the pleural space. Complicated pleural effusions will yield positive culture results over 50% of the time, making thoracentesis and culture of fluid a valuable means of making an etiologic diagnosis of the underlying pneumonia.126 Other diagnostic tools have proven useful in identifying organisms associated with pleural effusions. PCR technology has been useful in detecting M. tuberculosis in effusions with a sensitivity of approximately 70% and specificity of 100%.127 Adenosine deaminase, an enzyme associated with lymphocytes, may also be used to detect M. tuberculosis, with sensitivity and specificity of 93%.128,129 Detection of the lymphocyte-related cytokine soluble IL-2 receptor may also be useful.

BLOOD CULTURE, SEROLOGIC STUDIES, AND URINE STUDIES, INCLUDING ANTIGEN DETECTION Blood cultures are positive in 4 to 18 of patients hospitalized with community-acquired pneumonia.130-134 Recent studies have suggested that positive blood cultures add little to the management of patients hospitalized with community-acquired pneumonia and are not predictive of increased mortality. Although some studies have suggested that the presence or absence of bacteremia cannot be predicted, others have shown that the presence of liver disease, hypotension, hyperthermia or hypothermia, tachycardia, elevated BUN, hyponatremia, and increased or decreased WBC are independent predictors of bacteremia and can be used to define a high-risk group of patients.130-132 The presence of positive blood cultures is highly specific, may be helpful in narrowing antibiotic use, and may identify the presence of unusual organisms that would not be adequately covered by routine empirical antibiotic coverage.133,135 Therefore, blood cultures should be obtained in all patients suspected of having bacterial pneumonia who are ill enough to be hospitalized. Further, because the etiology of pneumonia is not always found, assessment of clinical response to initial therapy is important. Blood cultures may be of help in patients not responding to antibiotic therapy.133 Serologic assays have been used for decades to try to identify potential etiologies of pneumonia. Success has been limited, although recent improvements have been made. Serum antibody assays for diagnosis of M. pneumoniae and C. pneumoniae infection have been widely used. The Centers for Disease Control and Prevention (CDC) and the Laboratory Centre for Disease Control (LCDC) have established diagnostic standards. Microimmunofluorescence (MIF) for serum chlamydophilal antigens has been recommended, though enzyme immunoassays are also available and may be more sensitive and specific.136 An IgM titer of greater than 1 : 16 or a fourfold rise in IgG value is used to define positivity. Use of a single IgG value is not viewed as a definitive test. Because the present assays show day-to-day variation, it has been suggested that acute and convalescent titers be assayed at the same time.137 A fourfold rise in IgG rather than a single clinical titer is accepted as a positive test for M. pneumoniae. Although an elevated IgM titer suggests a recent infection, reinfection with mycoplasma occurs frequently and a rise of IgM may not always be seen.138 Cold agglutinins may be elevated in infections with M. pneumoniae. Titers greater than or equal to 1 : 4 are suggestive of M. pneumoniae infection. For both Mycoplasma and Chlamydophila infections, nucleic amplification technologies are being examined as alternative diagnostic modalities. S. pneumoniae produces a variety of antigens and surface markers that are type- or species-specific.139 Although both antigen and antibody detection methods in serum have been studied, none has become clinically significant. PCR techniques have been applied to whole blood for the detection of pneumococci, but the assays remain experimental.140 A variety of assays have been used to detect pathogens that have been difficult to isolate using routine culture techniques. Serologic assays have been used to diagnose infections caused by Legionella species, M. pneumoniae, Chlamydophila species, and Coxiella burnetii.141 The sensitivity and specificity of the assays vary, and their overall usefulness in making a rapid diagnosis is limited. PCR has also been used to detect DNA from Pneumocystis in blood from patients with AIDS and in BAL fluid.142 PCR techniques have also been used for detection of Legionella and can be designed to detect all the Legionella species. DNA probes have been used to successfully identify M. tuberculosis, Mycobacterium avium intracellulare, Mycobacterium kansasii, and Mycobacterium gordonae from colonies growing in solid and liquid medium, which has reduced the time of identification from 59 days to between 17 and 31 days. A variety of cytokines are released into the circulation as a result of infection.143-146 Evidence suggests that these biomarkers may be useful adjuncts in diagnosing pneumonia and predicting severity of

64 Acute Pneumonia

901

disease.143-146 The calcitonin family of gene products, especially procalcitonin, C-reactive protein (CRP), and soluble triggering receptor expressed on myeloid cells (STREP-1), have been the markers most often associated with pneumonia. Procalcitonin appears to be the earliest marker to appear during the course of infection. Elevated levels (greater than 0.25 to 0.5 g/L) have been used in some studies to decide which patients should be treated with antibiotics and, based on whether levels fall, how long antibiotics should be continued.147,148 Procalcitonin has also been used as a gauge of pneumonia-related mortality.149 C-reactive protein is an acute-phase reactant produced in the liver as a response to a variety of stimuli, including infection. Normal values of less than 10 mg/L are unusual in patients with pneumonia and can be used to exclude the diagnosis. Levels of 100 mg/L or greater suggests the diagnosis of pneumonia and has been associated with an increased 30-day mortality and a greater likelihood of need for ventilator or vasopressor support, all associated with severe pneumonia.146 In comparative trials with procalcitonin, CRP appears to have the best diagnostic capability although definitive studies are lacking.145 Other cytokines studied include IL-6 and TNF-, but their correlations with pneumonia appear less consistent. Cortisol levels have also been shown to predict the severity of pneumonia and the chance of survival.150,151 Large-scale, randomized studies using biomarkers are lacking, and therefore their role in diagnosis and severity assessment in pneumonia has not been clearly defined. Some have felt that these biomarkers offer little more than the severity assessment tools already used clinically.152 Antigen detection in urine rather than blood or sputum has become a successful means of detecting some important pulmonary pathogens. Soluble L. pneumophila antigen can be detected in urine using a commercially available enzyme immunoassay (EIA). Although it is useful for detecting only L. pneumophila serogroup 1, this assay offers the advantage of being rapid and noninvasive, and it has a sensitivity of 80% to 95% and a specificity estimated to be 99%.153 A relative problem with this method is that antigenuria may persist for weeks to months after therapy.153 An immunochromatographic membrane test has been developed to detect the C polysaccharide cell wall antigen found in all S. pneumoniae in urine of patients with pneumonia (Binax NOW; Binax, Inc., Scarborough, ME).154 This has been reported to be an extremely useful means of diagnosing pneumococcal pneumonia. Using a variety of standard diagnostic tests as controls, overall sensitivities of 65.5% to

100%, specificities of approximately 94% to 100%, and positive predictive values of 62% have been noted.155-159 Sensitivities have, in general, been high in bacteremia episodes, with the yields increased slightly by concentrating the urine. The test is not affected by the prior use of antibiotics. Potential problems with the urinary antigen assay include weakly positive results caused by non-pneumococcal organisms, false-positive results in children with nasopharyngeal carriage rather than true infection, and positive results lasting for weeks after the infection has resolved.156,160 A shortfall of the test is that no organism is isolated and no antibiotic susceptibilities can be carried out. Despite these problems, the urinary antigen test for S. pneumoniae has become an accepted diagnostic modality and has been recommended in the IDSA/ATS. Recently, an assay for detection of pneumolysin in urine has been developed. Although studies are limited, sensitivity appears lower than that of the Binax NOW assay for C polysaccharide, but specificity appears comparable.161 RADIOLOGIC EXAMINATION Chest radiography plays a critical role in the diagnosis of pneumonia, and for many it represents the gold standard of making a clinical diagnosis. The differential diagnosis of respiratory complaints and abnormal physical findings includes upper and lower respiratory tract infection as well as an array of noninfectious entities. Demonstration of an abnormal chest radiograph consistent with pneumonia differentiates a patient population that may benefit from antibiotic therapy from the populations that will not. Because overuse of antibiotics for therapy of upper respiratory infections has been documented and may contribute to the growing problem of antibiotic resistance, identifying patients who really should be receiving antibiotic therapy is clearly of importance. The chest radiograph is readily available, is reasonably reliable (despite interobserver variability),162,163 and should be obtained in most patients suspected of having pneumonia.78,162-164 The extent and nature of radiographic abnormalities may define patients who are more seriously ill and may need close monitoring. The chest film usually does not show an infiltrate pattern that is very helpful in making a specific etiologic diagnosis (Fig. 64-8). However, certain features may be of some diagnostic aid. Lobar consolidation, cavitation, and large pleural effusions support a bacterial cause (Figs. 64-9 and 64-10). Most lobar pneumonias are pneumococcal, whereas most pneumococcal pneumonias are not lobar. When bilateral diffuse

Figure 64-8 A, Normal chest radiograph. B, Patchy infiltrate representing bronchopneumonia in a patient with Streptococcus pneumoniae infection.

902

PART II

Major Clinical Syndromes

Figure 64-9 A, Chest radiograph showing dense left lower consolidation consistent with bacterial pneumonia, in this case caused by Streptococcus pneumoniae. B, Lateral radiograph of a patient with left lower lobe pneumococcal pneumonia.

involvement is noted, Pneumocystis pneumonia, Legionella pneumonia, or a primary viral pneumonia should be suspected. Staphylococcal pneumonia may result from infection metastasizing from a primary focus unrelated to the lung. In these cases, multiple nodular infiltrates throughout the lung may be seen. Staphylococci may cause marked necrosis of lung tissue with ill-defined thin-walled cavities (pneuma-

toceles), bronchopleural fistulas, and empyema, especially in children (Fig. 64-11). Although pneumatoceles are diagnostically significant findings in staphylococcal pneumonia, they may be seen in pneumonias with other causes, including K. pneumoniae, H. influenzae, S. pneumoniae, and, more rarely, Pneumocystis. Staphylococcus aureus producing the Panton-Valentine leukocidin, whether methicillin-

A
Figure 64-10

B
Chest radiographs showing a large left pleural effusion in a patient with Klebsiella pneumoniae pneumonia.

64 Acute Pneumonia

903

Figure 64-11 Pneumatocele formation in the left upper lobe of a patient with staphylococcal pneumonia.

resistant or not, is associated with necrotizing pneumonia with multilobar cavitary lesions and is frequently associated with pleural effusions and empyema.165,166 Pulmonary infections due to Pseudomonas may cavitate. Pseudomonas and other gram-negative bacilli most commonly cause lower lobe pneumonia. Aspiration pneumonia should be considered along with gramnegative and staphylococcal pneumonias as a source of necrotizing pneumonia, cavitation, and empyema. Aspiration pneumonia commonly involves either the superior segment or the basilar segment of either lower lobe, or the posterior segment of the upper lobes, depending on whether aspiration occurred in the dependent or the upright position. Chronic aspiration most commonly results in bilateral lower lobe pneumonia, although it may involve one side more than the other. Viral infection of the lower airway involves respiratory epithelium and parenchyma adjacent to terminal respiratory bronchioles. Diffuse hemorrhagic congestion of alveolar septa may occur as well.167 The radiographic concomitants of these pathologic findings usually involve patchy areas of peribronchial ground-glass opacity, air-space consolidation, and poorly defined small nodules. Diffuse and localized involvement with both interstitial and alveolar patterns has been noted (Fig. 64-12).167 There is little radiologic distinction between the various viral etiologies of pneumonia. Influenza pneumonia is associated with poorly defined, patchy air-space consolidation with rapid confluence. Varicella pneumonia usually involves peribronchial involvement with nodular infiltrates. Adenovirus, which is more common in immunecompromised hosts, may be associated with diffuse bilateral bronchopneumonia, areas of overinflation, atelectasis, and nodular opacities.168 Lobar or subsegmental consolidation mimicking bacterial pneumonia may also be seen. People with hantavirus pneumonia usually present with interstitial edema, which may progress to consolidation representing a pulmonary capillary leak syndrome. Bilateral involvement and pleural effusion are common.167 In the majority of cases of SARS, caused by a coronavirus, bilateral basilar infiltrates primarily involving an interstitial pattern are present. Progression with symmetrical airspace disease has been commonly described.169,170 A recently defined viral pulmonary pathogen is the human metapneumovirus. Although most cases involve upper respiratory tract infections in children, pneumonia in adults has been described.171,172 Multilobe infiltrates have been noted in 50% of cases, though most disease is unilateral. Mycoplasma pneumonia often manifests with an interstitial pattern in a peribronchial and perivascular distribution. Consolidation is noted in approximately 38% of patients, usually in the lower lobe. Once this consolidation stage is reached, radiologic differentiation between bacterial and mycoplasmal pneumonia is difficult. Cavitation is rare, although pleural effusions may be seen in approximately 20% of cases.173

People with legionnaires disease may initially present with a radiographic picture similar to that of mycoplasma pneumonia. A patchy interstitial or finely nodular pattern is seen in the lower lobe.174 However, unlike the situation with mycoplasmal pneumonia, pneumonia with more than two-lobe involvement is commonly seen. Rapid progression and pleural effusions are also common. Pulmonary nodules, either single or multiple and with segmental infiltrates, may occur with pneumonia caused by Legionella micdadei. As in pneumonia caused by L. pneumophila, rapid radiologic progression of the disease is characteristic.175 Chest radiographs are most helpful in conjunction with the clinical history and physical examination, but, as noted previously, are often not helpful in making a specific etiologic diagnosis. Without clinical information, readings of radiographs correctly identify pneumonia as bacterial only 67% of the time and as viral only 65% of the time. Mycoplasmal pneumonia is usually incorrectly identified as bacterial. The usefulness of computed tomography (CT) in managing chest infections, including pneumonia, has been reviewed.176,177 In the immunocompetent host, chest CT is most helpful in evaluating recurrent pneumonia or infections unresponsive to therapy. Pneumonia developing behind an obstruction caused by tumors or other masses and lung abscess may also be better defined by CT than by routine chest radiographs. High-resolution CT has been shown to improve radiographic characterization of lung infection.176 Compared with a routine chest radiograph, high-resolution CT detects lung abnormalities more often and does a better job in defining disease in the upper and lower lobes and in the lingula. However, exposure to more radiation (the radiation from one CT scan equals that from six to seven chest radiographs) and the increased expense (approximately seven times the cost of a chest radiograph) has limited its use as the initial radiographic procedure. Furthermore, it is unclear if all abnormalities found on the chest CT scan truly represent pneumonia.177 In the immunocompromised host in whom infection is only one of the possible causes of abnormal chest radiographs, chest CT or one of its variations, such as spiral CT or high-resolution CT, may aid in better defining a questionable chest radiograph and may be helpful in localizing involved areas of lung as a guide to biopsy procedures. Certain infections, such as those caused by Aspergillus, M. tuberculosis,

Figure 64-12 Bilateral involvement with a mixed interstitial-alveolar pattern in a patient with viral pneumonia.

904

PART II

Major Clinical Syndromes

and Pneumocystis, have characteristic appearances on CT that in the correct clinical setting may make invasive procedures unnecessary. Techniques such as perfusion magnetic resonance imagery have been shown to be able to differentiate pneumonia from COPD and pulmonary emboli. The clinical utility of these techniques remains to be defined. Nuclear medicine procedures have been used to detect pneumonia. These procedures include gallium-67 citrate scans, indium-111 labeled granulocyte scans, and technetium-99 diethylenetriamine penta-acetic acid aerosol clearance.178 In general, these procedures have been used in patients with AIDS to define the presence of lung infection in the absence of abnormal chest radiographs. In patients with AIDS, diffuse uptake of gallium is usually seen with Pneumocystis infection but may also be seen with infection caused by Mycobacterium avium complex, cytomegalovirus, and Cryptococcus neoformans, and in patients with lymphoma. Localized uptake may be associated with bacterial disease. Focal uptake corresponding to lymph node areas has been associated with M. avium complex, M. tuberculosis, and lymphoma.

Pneumonia Syndromes
ACUTE COMMUNITY-ACQUIRED PNEUMONIA A long list of bacterial, fungal, viral, and protozoal agents may cause pneumonia. Because initial evaluation rarely results in a specific etiologic diagnosis, antibiotic therapy is usually begun empirically. Defining pneumonia syndromes on the basis of clinical, epidemiologic, radiographic, and laboratory parameters, with a limited number of organisms commonly associated with each syndrome, has helped the clinician to select rational empirical therapy for the most likely organisms involved. Many of the syndromes have overlapping signs and symptoms, which at times makes clear identification of a specific syndrome in an individual impossible.179,180 Furthermore, the characteristics of the syndrome of acute community-acquired pneumonia as defined almost 30 years ago are changing.48,180,181 Increases in the numbers of patients living longer, more and varied comorbidities, and expanded contact with various aspects of the health care system have led to a wider array of presentations, etiologic agents, and strategies for empiric therapy. Newly described microbial agents are being recognized as potential causes of community-acquired pneumonia. Subgrouping syndromes under the general description of communityacquired pneumonia may be made based on patient age, severity of illness, comorbidities, need for hospitalization, and epidemiologic setting. Patients with acute community-acquired pneumonia are usually in their mid 50s to late 60s. Peak incidences of disease in midwinter and early spring have been described, but there is no pneumonia season, and disease takes place throughout the year. Most patients (58% to 89%) have one or more chronic underlying diseases. Immunosuppression related to malignancy, neutropenia, the chronic use of steroids or myelosuppressive agents, or HIV infection may be present in up to 57% of patients.180,182 Classically, community-acquired pneumonia presents with a sudden onset of a chill followed by fever, pleuritic chest pain, and cough that produces mucopurulent sputum. The signs, symptoms, and physical findings vary according to the age of the patient, therapy with antibiotics before presentation, and the severity of illness. These classic findings in some combination are present in approximately 81% of patients with community-acquired pneumonia. Patients usually present after having been ill for a mean of 6 days. Cough is noted in greater than 80% to 90% of patients and is productive in 60% to 80%.45,180,181 Chest pain is present in 30% to 46% of cases, chills in 40% to 70%, and true rigors in 15%.44,45,180 A variety of nonrespiratory symptoms are associated with pneumonia, including fatigue (91%), anorexia (71%), sweats (69%), and nausea (41%).44 Both respiratory and nonrespiratory findings occur less frequently in older age groups.44

Physical examination reveals fever in 68% to 78% of patients but may be seen less commonly in older populations. Tachypnea (respiratory rate greater than 24 to 30 breaths per minute) is noted in 45% to 69% of patients and may be more frequently seen in older age groups.44 Tachycardia (pulse rate greater than 100 beats/min) is noted in 45%. Rales are noted in 78% of patients, and signs of consolidation are noted in 29%. Most commonly, the white blood cell count is in the range of 15,000 to 35,000/mm3, and the differential cell count reveals an increased number of juvenile forms. Leukopenia may be noted and is a poor prognostic sign.183 The hematocrit and the red blood cell indices are usually normal. Sputum is thick and purulent and may be rust colored. The sputum Gram stain reveals numerous neutrophils and bacteria, often with a single organism predominating. Chest films show areas of parenchymal involvement, usually with an alveolar-filling process. There is moderate hypoxemia due to ventilation perfusion abnormalities. Even with rigorous laboratory evaluation and using definitions of definite, probable, and possible causes, a microbiologic diagnosis may be made in only 20% to 70% of cases of community-acquired pneumonia.47,82,180 In the past, 50% to 90% of cases of acute community-acquired pneumonia were caused by S. pneumoniae.43 More recently, some published reports have indicated that the relative importance of the pneumococcus has varied, showing this range to be from 16% to 60%.65,181 However, pneumococcus remains the leading cause of the syndrome of acute community-acquired pneumonia in most series. Severe pneumococcal infections, including pneumonia, have been associated with prior splenectomy due either to trauma or to staging for Hodgkins disease,184-186 abnormal immunoglobulin responses (myeloma, lymphoma, HIV infection),187 and functional asplenia due to systemic lupus erythematosus or marrow transplant. An estimated 3% to 38% of cases of acute community-acquired pneumonia are caused by H. influenzae.48,188 The true incidence of this organism is obscured by the difficulty of isolating it from sputum and identifying it in sputum Gram stain, and by the difficulty of distinguishing colonization from infection. The age of patients, presence of underlying disease, and presentation are all similar to those of pneumococcal disease. Although the use of the H. influenzae conjugate vaccine has decreased the incidence of invasive disease caused by H. influenzae type B, there is a strikingly increased incidence of invasive disease including pneumonia caused by nontypeable strains. In one series, over 50% of isolates from patients with invasive H. influenzae disease were nontypeable.189 Classically, S. aureus has accounted for 2% to 5% of acute community-acquired pneumonia cases190 and takes on increased importance as a cause of pneumonia in older adults and in those with influenza. Patients who develop postinfluenza pneumonia may be younger and have less underlying disease than other patients with communityacquired pneumonia. Clinical signs and symptoms of influenza are present but appear to resolve over several days. After a variable period of time ranging from 2 to 14 days, symptoms suddenly reappear, with the onset of shaking chills, pleuritic chest pain, and cough that produces purulent sputum. An elevated white blood cell count with a shift to the left, physical signs of pulmonary consolidation, and radiographic evidence of focal parenchymal disease appear. The sputum Gram stain is consistent with bacterial pneumonia. Although pneumococcus still represents the most common etiologic agent, staphylococcal disease occurs with a higher frequency than that noted in noninfluenza-related, community-acquired pneumonia. As noted previously, in 2002 many cases of community-acquired pneumonia caused by S. aureus were described in France.49,191 Patients were young, had few if any comorbidities, usually presented after a flulike illness with high fevers, leukopenia, tachycardia, tachypnea, and multilobar disease on x-ray rapidly evolving into bilateral disease. Acute respiratory distress syndrome (ARDS) was a frequent complication of infection and mortality rates were higher than in patients with pneumonia caused by S. aureus without the Panton-Valentine leukocidin. Similar disease patterns have been described in the United States

64 Acute Pneumonia

905

with these strains of community-acquired of MRSA, or CA-MRSA.50,192 These CA-MRSA pneumonias are associated with severe pneumonia with mortality rates of 29% to 60%. Flulike illness precedes the onset of disease in 75% of patients with documented influenza noted in 71% of cases. CA-MRSA pneumonia has therefore become an important entity to consider in the right clinical setting. Aerobic gram-negative bacteria, exclusive of H. influenzae, and mixed aerobic and anaerobic infections cause most of the remaining cases of acute community-acquired pneumonia. Gram-negative rods may cause anywhere from 7% to 18% of pneumonia cases. K. pneumoniae, P. aeruginosa, and Enterobacter species are the most often isolated organisms.193 Gram-negative bacilli are particularly important pathogens in older adults, especially those with chronic underlying disease and those who are bedridden and recently hospitalized. Pseudomonas infection should be suspected in patients with pulmonary comorbidities and recent hospital stays. Legionella species are the most important water-related pulmonary pathogens in the United States with regard to mortality and morbidity. The importance of Legionella species in causing pneumonia has varied greatly in different geographic areas, with incidences ranging from 2% to 30%.180,194 Since 2003, an increased incidence of legionellosis has been observed in the United States, especially on the East coast.195 Although infection may occur at any age, those aged 45 to 64 now appear to be at greatest risk. No clinical features reliably distinguish Legionella species pneumonia from that caused by other bacteria.196 However, the presence of a high fever (greater than 40 C), male sex, previous -lactam therapy, multilobar involvement, rapid progression of radiographic abnormalities, a need for intensive care, gastrointestinal and neurologic abnormalities, elevated liver enzyme levels, and increased creatinine levels have all been associated with Legionella pneumonia.180,194,196 Moraxella catarrhalis has also been identified as a cause of pneumonia.197 The overall incidence of disease caused by this bacterium is low, but it is an important pathogen in older adults with COPD and various forms of immunosuppression. COMMUNITY-ACQUIRED PNEUMONIA IN THE OLDER ADULT Pneumonia in the elderly has become an increasingly important clinical entity as the worlds population has aged.198 Pneumonia is the second or third most common reason for hospitalization in those 65 and older and represents a major cause of morbidity and mortality. In some series, pneumonia represents the leading cause of death in this population (see Chapter 314). For those older than 60 years, pneumonia is a predictor of increased mortality after the specific episode has resolved and for several years thereafter.199 The clinical presentation of pneumonia in older adults (especially those older than 80) may be more subtle than in younger populations, with more gradual onset of symptoms and fever and the classic signs of pneumonia.44,180,200 Fever occurs less commonly in older adults, and temperature elevation is muted.44,180,200 The classic findings of cough, fever, and dyspnea may be absent in over half of older adults.201 Chills and rigors may be less frequently seen as well.44,180 Tachypnea (respiratory rate of greater than 24 to 30 breaths per minute) is a more frequent finding in older adults and has been observed in up to 69% of patients. Rales are common and are noted in 78% to 84% of patients, although signs of true consolidation are found in only 29%. Nonrespiratory symptoms may be the major presenting feature. The initial presentation of older adults with pneumonia may include decline in functional status, weakness, subtle changes in mental status, and anorexia or abdominal pain. It has been suggested that the nonspecific presentation of pneumonia in older adults may result in great part from the prevalence of dementia in this population.201 Bacteremia, metastatic foci of infection, and death are more frequent in older populations.48,67,183 Specific etiologic diagnoses are made less frequently in older adults, with approximately 20% to 50% of patients having an etiologic agent

defined.200 The absence of productive cough and common prior use of antibiotics may explain this observation. Etiologies have varied in different series depending on the means of diagnosis, the patient population studied (outpatient versus institutionalized older adults), and the geographic location. In general, the cause of community-acquired pneumonia in the older population follows the general trend of infection in younger populations. S. pneumoniae remains the predominant organism, accounting for 20% to 60% of cases. H. influenzae, usually a nontypeable strain, is frequently the second most common agent, accounting for 7% to 11% of episodes.202 The importance of other aerobic gram-negative bacilli in causing pneumonia in older adults remains a question in part because the criteria for diagnosis of true pneumonia versus colonization vary. In recent studies, 1% to 3% of pneumonia may be caused by non-Haemophilus gram-negative bacilli.200 Although increased oropharyngeal colonization with aerobic gram-negative bacilli has been documented in the older population and is thought to be a predisposition to development of pneumonia caused by these organisms, colonization appears to be related to debility of the patient rather than age.203 Other factors associated with increasing colonization with gram-negative organisms include prior use of antibiotics, decreased activity, diabetes, alcoholism, and incontinence. Older adults are at greater risk for infection with group B streptococci, M. catarrhalis, and Legionella species, although the overall incidence of these agents in the older population is relatively low.197 Legionella has been described as a cause of severe pneumonia (see later definition) in the elderly. Polymicrobial infections and pneumonia due to aspiration have both been noted to occur more frequently in older adults.67,204 It is unclear which agents cause atypical pneumonia in the older population. Most series suggest that M. pneumoniae pneumonia is unusual, although it has been documented by other investigators to be a significant cause of pneumonia leading to hospitalization in older adults.47,205,206 In one series, over 15% of the documented cases of mycoplasmal pneumonia were in patients 60 years or older.205 Chlamydophila infections appear commonly in the older population and may cause up to 32% of pneumonias.207 Viral agents play an important role as etiologies of pneumonia in the elderly although historically their role has been underestimated given the difficulty in culturing them and the relative insensitivity of serologic tests.208,209 With the development of sensitive nucleic acid amplification tests like the reverse transcriptase polymerase chain reaction, their role as a cause of pneumonia has begun to be more clearly defined.208-210 Recent studies have suggested a viral etiology in 15% to 29% of all patients hospitalized with pneumonia with significantly more viral infections noted in the older age groups (median age of 76).210, 211 Influenza is consistently the most commonly isolated virus usually followed by respiratory syncytial virus (RSV), human metapneumovirus, parainfluenza virus, enterovirus, and coronavirus. Approximately 16% to 27% of viral pneumonias are mixed infections with bacteria.210,211 Increased adherence of S. pneumoniae to human tracheal epithelial cells in the presence of rhinovirus has been reported with a similar relationship observed with enteroviruses.212 This relationship raises the question as to whether some viruses play a role as facilitators for bacterial infection rather than roles as true pulmonary pathogens. Clinically, viral pneumonia in the elderly cannot be differentiated from bacterial pneumonia by clinical, routine laboratory, or radiologic parameters. As with bacterial disease, the signs of viral pneumonia may be more subtle and may only involve fever and altered mental status.209 Dyspnea, wheezing, and productive cough are commonly observed. Myalgia, though commonly found with most viral etiologies, is most often seen with influenza. Bronchospasm and wheezing are seen more commonly with RSV.208 NURSING HOME PNEUMONIA Residents of nursing homes represent an important subpopulation of older adults at risk for pneumonia.213 Pneumonia is the second most

906

PART II

Major Clinical Syndromes

frequent infection in this setting and carries the highest mortality of any infection.214 In addition, it is a common reason for the elderly to be hospitalized.215 Silent aspiration is a major risk factor, as are poor functional status, poor oral hygiene, nasogastric feeding, swallowing difficulties, confusion, the presence of obstructive lung disease, the presence of a tracheostomy, and advancing age.214,216 The subtle presentation noted in other older adult populations occurs in those in a nursing home setting. S. pneumoniae remains the predominant etiology, followed by nontypeable strains of H. influenzae and M. catarrhalis. Other organisms such as Chlamydophila, M. pneumoniae, Legionella, and aerobic gram-negative bacilli including P. aeruginosa, are infrequent causes. Outbreaks of pneumonia have occurred in nursing homes and have involved Legionella, Chlamydophila, influenza, parainfluenza, respiratory syncytial virus, and rhinovirus.217 COMMUNITY-ACQUIRED PNEUMONIA IN PATIENTS WITH AIDS The cause of community-acquired pneumonia in patients with AIDS has changed significantly since the beginning of the epidemic. With the development of highly active antiretroviral therapy (HAART) and effective prophylactic strategies, the incidence of P. jirovecii (carinii) pneumonia has been halved with associated mortality rates reduced by 60%. Even with prophylaxis, however, Pneumocystis infection may develop in the setting of severe immune deficiency.218 Pneumocystis remains the most frequent AIDS-defining opportunistic pneumonia in the United States. Bacterial pneumonia has recently been shown to have a ten times higher incidence in HIV-positive versus HIV-negative cohorts.218 The rate of invasive pneumococcal disease may be as high as 100 to 300 times greater in HIV-infected patients than in nonHIVinfected controls.219 H. influenzae and S. aureus are also important pathogens.60,219-222 A variety of other bacteria have been implicated, including R. equi, Escherichia coli, Serratia species, and P. aeruginosa. Infections with Pseudomonas are associated with late stages of disease, the presence of central venous catheters, the presence of urinary catheters, and the use of steroids. Mycobacterium tuberculosis, nontuberculous mycobacteria, C. neoformans, and cytomegalovirus also play important roles as etiologic agents.60 The incidence of pneumonia reported to be caused by atypical agents is low. Although influenza and rhinovirus have been shown to be common causes of febrile respiratory illnesses in HIV-infected patients, pneumonia is unusual.223 Even with careful study, up to 75% of HIV-infected patients with pneumonia may have an infection without a proven etiology. SEVERE COMMUNITY-ACQUIRED PNEUMONIA Approximately 10% of patients with community-acquired pneumonia will develop severe disease, as defined by admission to an intensive care unit due to the presence of shock requiring vasopressors or respiratory failure requiring mechanical ventilation.224 Early identification of patients who are at higher risk for developing severe pneumonia is important because these patients have a higher mortality rate and require more supportive care. Further, patients with severe pneumonia are infected with a different spectrum of etiologic agents and would therefore benefit from different empirical antibiotic strategies than patients with less severe disease. Advanced age, presence of significant comorbidities, inadequate or delayed antibiotic therapy, and genetic predisposition have all been thought to be associated with the development of severe community-acquired pneumonia.225 Approximately one third of patients with severe pneumonia would have been previously healthy. Although shock or respiratory failure are usually evident and serve as major criteria for defining severe pneumonia, patients without these findings may also benefit from intensive care unit monitoring. A variety of prediction scores have been developed to assess severity in patients with pneumonia, including the modified British Thoracic Society severity score, the pneumonia Patient Outcome Research Team (PORT) score (also known as the pneumonia severity index

[PSI]),226 Confusion, Urea, Respiratory rate, low Blood pressure (CURB) score,227 CURB plus age older than 65 (CURB-65) score,228 and CURB-65 without the urea level (CRB-65) score.229 (Further use of these scoring systems will be discussed later in Management and Therapy of Pneumonia.) When compared as a means of predicting mortality and need for intensive care unit admission, sensitivities and specificities have been variable, with the best predictors being the PORT score, modified British Thoracic Society severity score, and the CURB systems.226 Recently, the American Thoracic Society and Infectious Diseases Society of America combined elements of these previously devised systems to define severe pneumonia.78 Shock requiring vasopressors and need for mechanical ventilation were major criteria, either one of which defined severe disease. In addition, a series of minor criteria were listed, any three of which defined severe disease. These minor criteria included respiratory rate higher than 30 breaths/minute, Pao2/Fio2 less than 250, multilobar infiltrates, new-onset disorientation, BUN greater than 20 mg/dL, leukopenia (WBC less than 4000 cells/mm3), thrombocytopenia (platelet count less than 100,000 cells/mm3), core temperature lower than 36 C, and hypotension requiring fluid resuscitation. These remain suggested but not validated criteria. The most recent prediction tool is the SMART-COP developed by the Australian community-acquired pneumonia study.230 Using a scoring system based on vital sign parameters, including oxygen saturation, mental status, albumin level, extent of radiographic abnormalities, and arterial pH, the schema identified 92% of patients needing ventilator or vasopressor support, including those not initially admitted to an ICU. S. pneumoniae and L. pneumophila are the organisms most commonly involved in cases of severe pneumonia. Gram-negative bacilli, especially Klebsiella species, must be considered in patients who have significant underlying disease, such as COPD, diabetes, and alcohol abuse.224 In some series, M. pneumoniae is involved in up to 11% of patients with community-acquired pneumonia requiring intensive care. Mortality rates in patients with severe pneumonia have recently been reported as high as 28% versus the 4% to 8% noted in patients admitted to non-ICU services.47,224 Tachypnea (more than 30 breaths/ minute), diastolic blood pressure less than 60 mm Hg, and blood urea nitrogen levels above 7 mmol/L have been shown to be independently associated with death from pneumonia.231,232 Other parameters identified with increased mortality include severe underlying disease, underlying neoplastic disease, age older than 60 years, absence of pleuritic chest pain, a change in mental status, acute respiratory failure requiring ventilator support, bilateral pulmonary involvement, bacteremia, a neutrophil count under 3500 mm3, a total serum protein level less than 45 g/L, a serum creatinine level greater than 15 mg/L, the presence of shock, inadequate initial antibiotic therapy, radiographic progression in the first 48 hours, and pneumonia caused by S. aureus or gram-negative bacilli.232-234 HEALTH CAREASSOCIATED PNEUMONIA In the past, a basic distinction in the epidemiology of pneumonia has been whether the infection developed in the community or in the hospital (see Chapter 303). The distinction was clinically relevant because the importance of various etiologic agents differed as did antibiotic susceptibilities. Consequently, the guidelines for empirical antibiotic therapy differed depending on where the infection developed. Since an increased amount of health care delivery has been shifted to the outpatient setting, even complex medical conditions may be handled without hospitalization. Subsequently, a growing number of patients develop pneumonia after extensive outpatient contact with various aspects of the health care system. This has led to a blurring of the distinction between community-acquired and nosocomial pneumonia. Recently it has been recognized that health careassociated pneumonia represents a new syndrome, which is a hybrid of community-acquired pneumonia and hospital-associated pneumonia.68-70 Patients who have been hospitalized within 90 days of developing

64 Acute Pneumonia

907

pneumonia; patients attending hemodialysis clinics; patients receiving intravenous therapy, wound care, or chemotherapy at home; and residents of long-term care facilities or nursing homes are the most likely involved. S. aureus including MRSA, aerobic gram-negative bacilli including P. aeruginosa, and mixed aerobic-anaerobic pathogens associated with aspiration have been most commonly reported. The role of S. pneumoniae has been variable, but in general it appears to play a lesser role than in patients with classic community-acquired pneumonia. Overall mortality appears to be higher in patients with health careassociated pneumonia (10.3%) than in community-acquired pneumonia (4.3%).70 It is not clear whether this is due to increased comorbidities in patients, more virulent organisms causing infection, or an increased incidence of inappropriate antibiotic usage in the first 48 hours of care. ATYPICAL PNEUMONIA SYNDROME By the late 1930s, most of the main bacterial causes of pneumonia had been defined. In 1938, Hobart Reimann235 described a small number of patients with a clinical picture that was atypical in that episodes began as a mild respiratory tract illness that was followed by pneumonia with dyspnea and cough without sputum. M. pneumoniae, C. pneumoniae (formerly known as Chlamydia pneumoniae), L. pneumophila, and respiratory viruses are the most significant causes of atypical pneumonia. Other agents such as Chlamydophila psittaci (formerly Chlamydia psittaci), Francisella tularensis, M. tuberculosis, and C. burnetii may also cause atypical pneumonia. In patients with AIDS, Pneumocystis and nontuberculous mycobacteria should also be included. Although some series report that almost 50% of patients with community-acquired pneumonia demonstrate serologic evidence of mycoplasmal or chlamydial pneumonia, or both, other series suggest an incidence of 7% to 28%.47,179,236 Historically, the epidemiology and clinical features of the atypical pneumonias were thought to be distinct enough to differentiate them clearly from other causes of community-acquired pneumonia. It is now clear that differentiation between atypical agents and typical bacterial causes of community-acquired pneumonia is imprecise.180 The difficulty in identifying the various etiologic agents associated with atypical pneumonia has made consistent estimates of incidence difficult. Although a variety of nucleic acid amplification tests have been used to diagnose the major etiologic agents involved in atypical pneumonia, lack of standardized assays makes comparison of infection rates in different studies difficult.237,238 M. pneumoniae may account for 10% to 30% of cases of community-acquired pneumonia, with the highest percentage noted in patients well enough to be treated as outpatients.205 It is most likely to occur in the older child (older than 5 years), the adolescent, and the young adult. The majority of cases occur in those younger than 40 years, although 15% of patients hospitalized with mycoplasmal pneumonia were older than 60.47,179,205 An increased incidence of disease and true epidemics has been documented in relatively enclosed populations of young adults at military bases, colleges, and boarding schools. Mycoplasmal infection occurs throughout the year, although a relative increase in incidence is noted in the late summer and fall. The course of M. pneumoniae is characterized by up to 10 days of symptoms before presentation, as is true with many of the other agents involved in atypical pneumonia. In its classic form, mycoplasmal infection presents with constitutional symptoms and a progression from the upper to the lower respiratory tract. Sore throat is often the initial finding. Bullous myringitis is seen in only about 5% of cases but when present is suggestive of mycoplasmal infection. Fever, malaise, coryza, headache, and cough represent the major clinical findings. Pleuritic chest pain, splinting, and respiratory distress are not usually seen. Moist or crepitant rales may be heard. Sputum production is variable, and the sputum is purulent in one third to one half of the cases. Gram stain and culture of sputum usually reveal mouth flora. White blood cell counts greater than 10,000/mm3 are uncommon, occurring in approximately 20% of the patients.83 An elevated sedimentation rate is

noted in about 25% of the cases. Pulmonary involvement seen on radiographs is commonly more extensive than the physical examination would indicate. Unilateral or bilateral patchy infiltrates in one or more segments, usually in the lower lobes, are noted in a bronchial or peribronchial distribution. Upper lobe involvement and pleural effusions are rare. Progression of the radiographic picture, despite a stable clinical picture, may be seen. The overall clinical course in most cases is benign. Disappearance of constitutional symptoms is usually noted in the first and second weeks, although cough and radiographic changes may persist for several weeks. Occasionally, M. pneumoniae infection presents as severe community-acquired pneumonia requiring intensive care. A number of extrapulmonary manifestations may occur with M. pneumoniae, including involvement of skin, central nervous system, blood, and kidneys (see Chapter 184). C. pneumoniae has emerged as an important cause of atypical pneumonia and may account for approximately 6% to 20% of communityacquired pneumonia cases.47,179,239-242 It has also been postulated to be an important co-pathogen, most often associated with S. pneumoniae. Although disease is uncommon in those younger than 5 years, serologic evidence of infection has been noted in over 50% of adults, and more recent studies suggest an important role for Chlamydophila in community-acquired pneumonia in those older than 65 years.242 Disease usually occurs sporadically, although epidemics have been well documented. The majority of infections are either asymptomatic or produce mild symptoms. As with mycoplasmal infection, sore throat and hoarseness herald the onset of pneumonia, although the progression of symptoms appears slower than that noted with mycoplasma or viral pneumonia. Cough may begin after several days to weeks, suggesting a biphasic illness. Hoarseness and sinus tenderness appear more commonly than in patients infected with Mycoplasma or viruses. The white blood cell count is rarely elevated. Pneumonia with C. pneumoniae is usually mild, although complete recovery may be slow. Cough and malaise may persist for weeks to months. Reinfection occurs and appears to be milder than primary infection and is usually not associated with pneumonia. Chronic and latent infections have also been described. Infection with C. pneumoniae has been associated with exacerbations of COPD and asthma. In general, few features distinguish chlamydial pneumonia from infection caused by other atypical agents or other bacteria. C. pneumoniae infections have been associated with extrapulmonary manifestations, including otitis, sinusitis, pericarditis, myocarditis, and endocarditis. It has also been associated with coronary artery disease, although the definite relationship remains unclear (see Chapter 182). C. trachomatis may be a pulmonary pathogen in immunocompromised and in healthy hosts, including newborns.243 Productive cough, myalgias, and fever associated with diffuse nonsegmental infiltrates appear most commonly. The agent has also been associated with chronic pneumonia in neonates and infants. Onset occurs at 2 to 3 weeks of age and is associated with tachypnea, a staccato cough with periods of cyanosis and emesis, a lack of fever, and diffuse interstitial and patchy alveolar infiltrates on chest radiographs. Elevated IgG and IgM levels and absolute eosinophilia have also been noted (see Chapter 180). Of the viral agents associated with atypical pneumonia in adults, influenza A and B, adenovirus types 3, 4, and 7 (especially in military recruits), human metapneumovirus,244,245 respiratory syncytial virus (especially in older adult and immune-suppressed patients), and parainfluenza virus are the most common.210,246-248 Data suggest that respiratory syncytial virus may cause pneumonia in 1% to 5% of immunocompetent adults. Reports of other viral agents causing pneumonia are scant but have included enterovirus, coronavirus, the herpesviruses, and hantavirus. A coronavirus has been shown to be the agent involved in SARS (see Chapter 155). Rhinovirus has been isolated in nasopharyngeal secretions and BAL fluid from immunocompetent hosts with pneumonia but may play a more important role as a cause of pneumonia in the immune-compromised patient.209,210,249 Elderly patients, especially those with comorbidities, are frequently the population at greatest risk for viral pneumonias.

908

PART II

Major Clinical Syndromes

Legionella is now recognized as an important cause of the atypical pneumonia syndrome, although patients infected with Legionella may also present with the syndrome of acute bacterial community-acquired pneumonia. It accounts for 2% to 8% of cases involving hospitalization.47 Legionella species are among the top three to four organisms causing pneumonia that requires intensive care unit monitoring.250,251 L. pneumophila causes over 80% of cases of Legionella pneumonia, with approximately 50% of cases caused by serogroup 1.236 Inhalation of aerosolized organisms after exposure to environmental reservoirs, such as fresh water and moist soil, has been the usual means of acquiring the organism, although aspiration is now thought to be an alternate route of infection.252 An increased incidence during summer months has been observed. Cigarette smoking, chronic lung disease, and immunosuppression are consistently noted risk factors for the development of disease. Although early symptoms of malaise, muscle aches, headaches, and nonproductive cough resemble the onset of a viral syndrome, the rapid progression of pulmonary symptoms and relatively high fever, often exceeding 40 C, is noteworthy.252 L. pneumophila pneumonia is associated with a variety of extrapulmonary findings and laboratory abnormalities, including mental status changes, abdominal complaints (loose stools or diarrhea), headache, bradycardia, elevation of hepatic enzyme levels, hypophosphatemia, hyponatremia, elevated serum lactate dehydrogenase levels, and elevated serum creatinine levels. No single finding or laboratory test can distinguish L. pneumophila pneumonia from pneumonias of other causes. Scoring systems have been developed that may help identify Legionella, but the systems have not been validated in large prospective series.196 Extrapulmonary infection is unusual, but when it does occur, it usually involves the heart with myocarditis, pericarditis, and postcardiotomy-like syndrome.252 Unfortunately, none of these findings distinguishes between L. pneumophila pneumonia, pneumonia caused by other atypical agents, and pneumonia caused by more typical bacterial pathogens. Similarly, radiographic manifestations do not distinguish Legionella infections from those of other causes. Patchy interstitial infiltrates, or nodular infiltrates that may progress rapidly even with adequate therapy, are characteristic. Pleural effusions may be noted in up to one third of patients. PNEUMONIA IN THE SETTING OF ASPIRATION The clinical setting in which aspiration occurs includes any disease state in which consciousness is altered and the normal gag and swallowing reflexes are abnormal. Older adult patients; patients in chronic care facilities, especially those who are neurologically impaired; patients during the acute phase of stroke; bedridden patients receiving tube feedings; and patients with dementia fit into the category of individuals susceptible to aspiration.253,254 The pathogenesis of lung injury due to acid aspiration has been delineated.255 The presence of acidic contents in the lung induces the release of proinflammatory cytokines including TNF- and IL-8. These and other cytokines recruit neutrophils into the lung. Activated neutrophils appear to be the key mediators of acute lung injury after acid aspiration, although a role for complement has also been demonstrated.256 Although aspiration may be a witnessed event, the majority of episodes are silent and are brought to medical attention by their sequelae. Three major syndromes are recognized as a consequence of aspiration: chemical pneumonitis, bronchial obstruction secondary to aspiration of particulate matter, and bacterial aspiration pneumonia.257 Aspiration may be associated with the acute respiratory distress syndrome, atelectasis, bronchial hyperreactivity, and fibrosis. Bacterial aspiration pneumonia occurs in more than 60% of cases of chemical aspiration.204 Although chemical pneumonitis and mechanical obstruction usually cause acute symptoms, aspiration pneumonia is more insidious, with symptoms usually occurring gradually several days after the initial episode of aspiration. Pneumonitis, necrotizing pneumonia, abscess, and empyema are common. Symptoms often include fever, weight

loss, and productive cough. Putrid sputum is produced in 50% of the cases.72 Anemia and an elevated white blood cell count are frequently associated findings. The bacteriologic findings in aspiration pneumonia reflect the flora of the oropharynx, and the importance of periodontal disease in this regard has been noted. Studies have documented anaerobic involvement alone in 45% to 58% of cases258 or in combination with aerobes in 22% to 46% of cases.72,259 Bacteroides species, Porphyromonas species, Prevotella melaninogenica, Fusobacterium species, and anaerobic gram-positive cocci are the predominant anaerobes isolated. In community-acquired aspiration pneumonia, Streptococcus species and H. influenzae are the most common aerobic isolates.258,260 M. catarrhalis and Eikenella corrodens may also be involved. In contrast, gram-negative bacilli (including P. aeruginosa) and S. aureus are the most commonly isolated aerobes from nosocomial aspiration pneumonia including ventilator-associated pneumonia.258,260 PULMONARY INFILTRATES WITH EOSINOPHILIA Pulmonary infiltrates with eosinophilia (PIE) is a syndrome associated with a variety of clinical entities, only some of which have an infectious cause.261 Pulmonary eosinophilia with transient, peripheral pulmonary infiltrates and minimal symptoms has been associated with Ascaris and Strongyloides infections. Ascaris is probably the leading parasitic cause of the syndrome worldwide. Prolonged pulmonary eosinophilia associated with weight loss, fever, cough, and dyspnea may be due to tuberculosis, brucellosis, psittacosis, coccidioidomycosis, histoplasmosis, and parasitic infections including ascariasis, strongyloidiasis, paragonimiasis, echinococcosis, visceral larva migrans, cutaneous larva migrans, and infections with Schistosoma, Dirofilaria immitis, and Ancylostoma species. Noninfectious causes include drug allergy, sarcoidosis, eosinophilic leukemia, Hodgkins disease, and hypersensitivity pneumonitis (e.g., pigeon breeders disease).262 A PIE syndrome has been associated with Pneumocystis pneumonia in AIDS patients. Acute eosinophilic pneumonia is a distinct clinical entity occurring in younger (20 to 30 years) otherwise healthy individuals. It is marked by the acute onset of dyspnea, nonproductive cough, fever, hypoxia, and chest pain. Although leukocytosis is common, peripheral eosinophilia is usually absent. Bilateral, diffuse pulmonary infiltrates are commonly seen. Radiographic abnormalities usually begin as interstitial infiltrates that progress to alveolar infiltrates. Chest CT reveals a ground-glass opacification with interlobular septal thickening. BAL yields marked (25% to 62%) eosinophilia, which is the diagnostic feature of the disease. Although most patients have received antibiotics, rapid stabilization occurs with steroid use. It has been suggested that chronic eosinophilic pneumonia may represent a unique clinical entity that is a form of collagen-vascular disease or an infection in a hyperimmune patient. A subacute onset of cough, dyspnea, fever, and weight loss associated with peripheral eosinophilia are the common features. Unlike the situation in acute eosinophilic pneumonia, respiratory failure is rare. Peripheral infiltrates are usually seen on radiographs. Focal interstitial fibrosis, bronchiolitis obliterans, microabscesses, and sarcoid-like granulomas are characteristic pathologic features. A rapid response to steroids has been reported. Tropical eosinophilia consists of myalgia, fatigue, weight loss, and anorexia associated with cough, frequently with nocturnal exacerbations, wheezing, dyspnea, and marked peripheral eosinophilia in patients who have lived in or visited the tropics. Radiographic changes are distinctive and include increased interstitial markings with 2- to 4-mm nodules throughout the lungs with preferential involvement of the bases. Most cases are thought to represent immunologic hyperresponsiveness to microfilarial infection and can be treated with diethylcarbamazine. Other causes of PIE syndrome include bronchopulmonary Aspergillus, which should be suspected when a patient with PIE presents with asthma and pulmonary vasculitis. Patients with Churg-Strauss syndrome frequently have eosinophilia along with allergic angiitis and

64 Acute Pneumonia

909

granulomatosis and present with asthma, diffuse pulmonary infiltrates, and multiorgan involvement. Hypereosinophilic syndrome, eosinophilic granuloma (also known as primary pulmonary Langerhans cell histiocytosis), bronchiolitis obliterans organizing pneumonia (BOOP), Sjgrens syndrome, and postradiation pneumonitis are unusual cases of pulmonary infiltrates with eosinophilia. NOSOCOMIAL PNEUMONIA AND PNEUMONIA IN THE IMMUNOSUPPRESSED HOST Nosocomial pneumonia is the second leading type of nosocomial infection and accounts for 13% to 18% of all such infections. It is the leading cause of infection-related deaths in hospitalized patients with an attributable mortality of 33% to 50%.263 Higher mortality rates have been observed when patients are bacteremic or have pneumonia caused by P. aeruginosa or Acinetobacter species. The morbidity associated with nosocomial pneumonia includes longer hospital stays (an average of 7 to 9 days) and higher costs for health care (an estimated $2 billion annually).264 Risk factors for the development of nosocomial pneumonia have been categorized as patient related, infection-control related, or intervention related. Patient-related risk factors include age greater than 70 years, severe underlying disease, malnutrition, coma, metabolic acidosis, and the presence of any of a number of comorbid illnesses (COPD, alcoholism, azotemia, central nervous system dysfunction). Risk factors related to infection control include a lack of hand hygiene and glove-use practices and the use of contaminated respiratory equipment. Intervention-related risk factors involve those procedures and therapies that undermine normal host defenses or allow the host to be exposed to large inocula of bacteria. Sedatives and narcotics may lead to aspiration; corticosteroids and cytotoxic agents blunt the normal host response to infection; and the prolonged use of antibiotics engenders resistance. Surgical procedures, especially involving the chest and abdomen, are associated with changes in host defenses that predispose to pneumonia. The use of ventilator support is perhaps the greatest risk factor for the development of nosocomial pneumonia, presenting a risk over 20 times that of unventilated patients.265 Data suggest that there is a 1% to 3% per day risk for developing pneumonia while on a ventilator with a higher risk during the first 5 days of intubation.265 Because of this preeminence as a risk factor, the term ventilatorassociated pneumonia (VAP) has been accepted as an important subcategory of nosocomial pneumonia. The use of antacids and histamine type 2 blockers that raise the gastric pH has been shown to increase stomach colonization with aerobic gram-negative rods. Whether this leads to an increase in nosocomial pneumonia remains controversial.266,267 As noted, acidsuppressive medications have recently been shown to increase the incidence of pneumonia in hospitalized patients, although the exact mechanism involved remains unclear.32a The percentage of patients with VAP caused by organisms initially found in the stomach ranges from 0% to 55%.268 Approximately 60% of cases of nosocomial pneumonia are caused by aerobic gram-negative bacilli, with members of the family Enterobacteriaceae (K. pneumoniae, E. coli, Serratia marcescens, Acinetobacter species, Enterobacter species) and Pseudomonas species accounting for the majority of these. S. aureus causes 13% to 40% of nosocomial pneumonia and appears to be more common in burn units, in patients with wound infections, and in patients recently ventilated after neurosurgery or head trauma.69,269 MRSA and multidrug-resistant (MDR) organisms now play major etiologic roles. In contrast to its prominent role in community-acquired pneumonia, S. pneumoniae causes only 3% to 20% of nosocomial pneumonias in most studies and is associated with infection developing early in the hospital course.66,270 Anaerobic bacteria have been isolated in up to 35% of cases of nosocomial pneumonia, although usually less than 5% of infections are thought to be caused by these organisms. They play a role when aspiration is likely to have occurred. Pneumonia caused by Legionella species may occur sporadically or as part of outbreaks.

Recent consensus guidelines have been established concerning the risks, etiologies, diagnostic work-up, and therapies for nosocomial pneumonia and VAP.69 A more in-depth review can be found in Chapter 303. Pneumonia in the immunocompromised host is perhaps the most complex of all the pneumonia syndromes, because it represents the interaction of host defense defects engendered by the underlying disease as well as the chemotherapy of that disease, exposure to potential pathogens in the community and within the hospital setting, and reactivation of infectious processes that had previously been dormant. Community-acquired pneumonia, atypical pneumonia, aspiration pneumonia, and nosocomial pneumonia all take place in the compromised host. A large number of bacterial, fungal, viral, and noninfectious etiologies must be considered.271 Reviews of the topic are found in Chapter 303 and Chapters 308 to 312.

Management and Therapy of Pneumonia

The first decision confronting the clinician is whether the patient presenting with respiratory symptoms in fact has pneumonia. The difficulties in establishing a diagnosis on clinical grounds and the potential problem of overprescribing empirical antibiotics for all patients with respiratory findings have been reviewed. A chest radiograph is usually necessary to establish a definitive diagnosis of pneumonia. The next decision is whether the patient is to be hospitalized, as will be the case 18% to 30% of the time.272-274 Numerous studies have examined prognostic features and predictors of the clinical outcome for patients with community-acquired pneumonia that have been adapted to help make this decision. Although early assessment tools used a combination of clinical, epidemiologic, laboratory, and radiographic parameters, more recently developed tools have relied on clinical parameters alone that can be evaluated at the bedside. 225-228,230 One of the earliest developed and most widely used assessment tool is the PORT score, also known as the pneumonia severity index (PSI).225 This system uses 20 clinical parameters in categories of age, presence of comorbidities, vital sign abnormalities, and laboratory and radiologic findings. Based on a point system, five prognostic groups (I-V) were defined. The lowest scores (Group I) are associated with low mortality (0.1%), and the highest scores (Group V) are associated with the highest mortality (27%). As a guideline for hospitalization, patients in Groups I and II are usually treated as outpatients, patients in Group III are in a borderline group, and patients in Groups IV and V are admitted to either a routine ward or ICU. The PORT score or PSI has been validated and widely endorsed.275,276 A problem with this assessment tool, however, has been the need to obtain a variety of laboratory and radiologic results before being able to numerically assess the patient. Alternative systems, such as the CURB and CURB-65 scores, use parameters that are more readily obtained.227,228 The CURB index was formulated from the British Thoracic Society study and uses four clinical parameters, which include new onset of confusion, urea level greater than 7 mmol/L, respiratory rate greater than 30 breaths/ minute, and systolic blood pressure less than 90 mm Hg or diastolic blood pressure less than 60 mm Hg. The presence of two or more criteria suggested an increased mortality and defined severe pneumonia.227 The CURB-65 system, which was developed later, added age older than 65 to the system, with the presence of greater than three parameters leading to prediction of increased mortality.228 In both the CURB and CURB-65 systems, patients with fewer abnormal parameters had lower mortality predicted and could be treated as outpatients. The CRB-65 score removed the requirement for a urea level, making the system laboratory-free with patient assessment done completely at the bedside.229 There have been few comparative trials of the various assessment systems. The CURB-65 and CRB-65 tools appear comparable in predicting high and low mortality groupings.277 A recent comparison of the PORT or PSI system with the CURB-65 system suggested that the PSI system predicted mortality more accurately.278

910

PART II

Major Clinical Syndromes

It is important to recognize that any severity assessment tool serves only as a guideline, not as an absolute. Clinical judgment regarding presence of hypoxia, stability of the home situation, ability to take oral medications, reliability in taking medication, likelihood of returning for follow-up, and likelihood of calling for help when needed all play a role in deciding whether a patient can be treated at home or in a hospital.279 Further, it has been shown that patients with low PSI or CURB-65 scores may need hospitalization and ICU monitoring and that severity assessment scoring systems may underestimate mortality depending on the type of clinical setting in which they are used.280 It has been suggested that a three-step appraisal be used to decide where a patient should be treated. The first step identifies any contraindications to care outside the hospital (e.g., hypoxia, social instability, emotional or psychological instability). The second step involves calculating the pneumonia severity index (although other scoring systems could be used) and assigning the patient to a risk group. The final step is using clinical judgment to be sure the decision to treat a patient at home is consistent with the information gleaned from the history and physical examination.281 Strict reliance on severity indices has failed to predict complex outcomes in up to 39% of patients thought to be well enough to be treated as outpatients.282 EMPIRICAL THERAPY OF COMMUNITYACQUIRED PNEUMONIA The next problem is determining the most likely cause of pneumonia. If diagnostic studies, as described previously, yield a likely cause, then specific therapy can be initiated. For most patients, a specific diagnosis cannot be established with certainty before the onset of therapy. Selecting an empirical antibiotic regimen is a continuing clinical challenge. Recent controversies and questions have included when antibiotics should be started, how to determine the most appropriate antibiotics to use, and how long therapy should continue. Timing of Antibiotics In 1997, a retrospective review of over 14,000 Medicare patient hospitalizations suggested that antibiotic therapy given within 8 hours of presentation was associated with a decreased mortality.283 A second retrospective study of similar design in 2004 showed that antibiotics given within 4 hours of presentation would result in lower mortality.284 Neither study corrected for pneumonia etiology or antibiotics used. Despite the lack of a prospective randomized study, advising and regulatory agencies including the Joint Commission and the Centers for Medicare and Medicaid Services used the 4-hour rule as a core quality measure. Subsequent meta-analysis failed to substantiate these findings, although other series suggested that in critically ill patients, earlier appropriate antibiotic therapy had a mortality advantage compared to delayed or inappropriate antibiotic use.285,286 Misdiagnosis, delays in making the correct diagnosis, overuse of antibiotics, and an increase in Clostridium difficile colitis in patients, many of whom should not have received antibiotics at all, occurred as a result of this rule.287-289 In 2007, the Joint Commission suggested that antibiotics be started within 6 hours, even though no other data had been presented. The joint IDSA/ATS guidelines have suggested a more commonsense approach that antibiotic treatment for pneumonia be started as soon as possible after the diagnosis is considered likely.78 Empirical Antibiotic Therapy It has recently become evident that an understanding of pharmacokinetics and pharmacodynamics is important in selecting appropriate empirical antibiotic therapy for patients with pneumonia. This is especially true when resistant S. pneumoniae is a consideration.290-293 In the past, in vitro susceptibilities were reported independent of body sites. Therefore, pneumococcal resistance was designed to be true for meningitis as well as pneumonia. Because all -lactam compounds, including penicillin, are time-dependent killers, active drug levels need to be above the minimal inhibitory concentration (MIC) of the organism being treated for 30% to 35% of the dosing interval for a successful

outcome when penicillin is used, 40% to 50% of the dosing interval when a cephalosporin is used, and 20% to 25% of the dosing interval when a carbapenem is being used. This principle explained why patients could be cured with penicillin even when the pneumococcus being treated was shown to be nonsusceptible to penicillin. Based upon this understanding, and with the acknowledgment that in vitro susceptibility thresholds should be appropriate for the site of infection, the Clinical and Laboratory Standards Institute changed the susceptibility breakpoints for penicillin against S. pneumoniae to now be defined as susceptible at less than or equal to 2 g/mL, intermediate at 4 g/mL, and resistant at equal to or greater than 8 g/mL.294,295 Using these new standards, approximately 92% of S. pneumoniae strains are susceptible to penicillin.295 Risk factors for the presence of penicillin resistance appear to be age, residence in a long-term care facility, and perhaps most important, exposure to -lactam antibiotics within the previous 3 months.296 Similar relationships between prior use of other antibiotic classes and subsequent resistance developing to members of that class of antibiotics have also been established.296 Increased MICs to penicillin have been associated with parallel resistance to other agents frequently used in the therapy of pneumonia. Using the old susceptibility standards, 25.5% of pneumococci resistant to penicillin were shown to be resistant to other antibiotics. Data concerning multidrug-resistant strains of pneumococci using the new susceptibility thresholds are lacking. Although overall rates may certainly decrease, it is still important to recognize general patterns of resistance as shown in Table 64-4. It remains unclear as to what level of resistance will yield clinical failures. It is important to note that as with penicillins, in vitro susceptibility to other agents does not always predict clinical efficacy. Resistance of S. pneumoniae to macrolides and azalides (azithromycin, clarithromycin) occurs either because of a blockage of the ribosomal binding area encoded by the erm (B) gene or because of an efflux pump mechanism encoded by the nef (A) gene that removes drug from the cells interior. The latter usually leads to low-level resistance that may be overcome by higher dosages. The former mechanism is usually associated with high-level resistance that cannot be overcome by higher dosages. Clinical failure has been described even with low-level resistance, making the empirical use of azalides and macrolides problematic.297,298 A number of respiratory tract quinolones (levofloxacin, moxifloxacin, gemifloxacin) are licensed for use in the United States as therapy for respiratory infections. These agents possess activity against the majority of bacterial and atypical agents involved in lower respiratory infections, and their potency is not significantly affected by the presence of penicillin resistance. Interestingly, when these agents are compared clinically with -lactam antibiotics with various activities against nonsusceptible strains of pneumococci and no real activity

TABLE

64-4

S. pneumoniae Resistance to Commonly Used Antibiotics % Resistance 10 20 5 35 15-25 18-25 29 5 5 1-5 1 1

Antibiotic Penicillin, amoxicillin-clavulanic acid Cefuroxime axetil Ceftriaxone Trimethoprim-sulfamethoxazole Azithromycin Clarithromycin Doxycycline Imipenem Meropenem Respiratory tract quinolones (levofloxacin, moxifloxacin, etc.) Quinupristin/dalfopristin Linezolid

Adapted from Musher DM. Streptococcus pneumonia. In Mandell GM, Bennett JB, Dolin RD, eds. Mandell, Douglas and Bennetts Principles and Practice of Infectious Diseases. 7th ed. Elsevier; 2010; Jacobs MR. Clinical significance of antimicrobial resistance in Streptococcus pneumoniae. S Afr Med J. 2007;97:1133-1140.

64 Acute Pneumonia

911

against atypical agents, the overall outcomes have, in general, been equivalent. No clear clinical superiority for any respiratory tract quinolone has been consistently demonstrated. Resistance to the respiratory tract quinolones by pneumococcal strains has already been reported. Overall resistance rates are less than 1%, but rates may be significantly higher in certain geographic areas, and patient populations with resistance rates as high as 5% noted in patients in long-term care facilities.299 Failure of quinolone therapy for S. pneumoniae infection due to resistance is very uncommon.300 The need for antibiotic combinations in the empirical therapy of pneumonia remains an ongoing question. Combination therapy with -lactam antibiotics and macrolides, especially azithromycin, has been associated in some studies with decreased mortality and decreased length of hospital stay. Older patients with community-acquired pneumonia and patients with proven pneumococcal pneumonia, including those with bacteremia, were the patients involved in these studies.301-303 Many of the studies were retrospective, could not identify the etiology of pneumonia in up to 60% to 90% of cases, and did not always control

for confounding variables. Smaller subsequent studies, one of which was prospective, could not substantiate these findings.304,305 Although combination therapy may lower mortality in patients with severe pneumonia, it remains unclear if this occurs in hospitalized patients who are not severely ill. It would appear that effects, when observed, are related to the use of macrolides rather than other classes of agents like quinolones.306 The anti-inflammatory effect of macrolide compounds may be a critical aspect of their interaction with other antimicrobial agents.307 Guidelines have been recently developed by the American Thoracic Society (ATS) and the Infectious Diseases Society of America (IDSA).78 In general, patients are stratified into outpatient versus inpatient treatment based on PSI or CURB-65 scoring systems. Recognition of the most likely etiologic agent in any given clinical situation and recognition of the organisms most likely to cause morbidity and mortality are emphasized. Finally, prevalence of common antibiotic resistance patterns and risks of acquisition are recognized. Empirical antibiotic therapy is reviewed in Table 64-5.

TABLE

64-5

Guide to Empirical Choice of Antimicrobial Agent for Treating Patients with Community-Acquired Pneumonia (CAP) Preferred Treatment Options

Oral-based -lactam, macrolide,* or doxycycline A respiratory fluoroquinolone alone, an advanced macrolide plus high-dose amoxicillin, or an advanced macrolide plus high-dose amoxicillin-clavulanate Comorbidities (COPD, Diabetes, Renal Failure or Congestive Heart Failure, or Malignancy) No recent antibiotic therapy An advanced macrolide plus -lactam or a respiratory fluoroquinolone Recent antibiotic therapy A respiratory fluoroquinolone alone or an advanced macrolide plus a -lactam** Suspected aspiration with infection Amoxicillin-clavulanate or clindamycin Influenza with bacterial superinfection Vancomycin, linezolid, or other coverage for MRSA or CA-MRSA Inpatient Medical Ward No recent antibiotic therapy A respiratory fluoroquinolone alone or an advanced macrolide plus a -lactam Recent antibiotic therapy An advanced macrolide plus a -lactam, or a respiratory fluoroquinolone alone (regimen selected will depend on nature of recent antibiotic therapy) Intensive Care Unit (ICU) Pseudomonas infection is not an issue A -lactam plus either an advanced macrolide or a respiratory fluoroquinolone Pseudomonas infection is not an issue but patient has a -lactam allergy A respiratory fluoroquinolone, with or without clindamycin Pseudomonas infection is an issue (cystic fibrosis, impaired host Either (1) an antipseudomonal -lactam plus ciprofloxacin, or (2) an antipseudomonal defenses) agent plus an aminoglycoside plus a respiratory fluoroquinolone or a macrolide Pseudomonas infection is an issue but the patient has a -lactam allergy Aztreonam plus aminoglycoside plus levofloxacin or other respiratory quinolone Anti-Pseudomonas cephalosporin, carbapenem (not ertapenem) or -lactam/-lactamase Health careassociated exposure inhibitor with anti-Pseudomonas activity plus vancomycin (for MRSA coverage) quinolone or aminoglycoside Nursing Home Receiving treatment in nursing home A respiratory fluoroquinolone alone or vancomycin (for S. aureus including MRSA) plus a -lactam (cefepime or piperacillin/tazobactam if Pseudomonas is suspected; ceftriaxone if Pseudomonas is not suspected) Hospitalized Same as for medical ward and ICU
*Azithromycin, or clarithromycin. That is, the patient was given a course of antibiotic(s) for treatment of any infection within the past 3 months, excluding the current episode of infection. Such treatment is a risk factor for drug-resistant Streptococcus pneumoniae and possibly for infection with gram-negative bacilli. Depending on the class of antibiotics recently given, one or another of the suggested options may be selected. Recent use of a fluoroquinolone should dictate selection of a nonfluoroquinolone regimen, and vice versa. Moxifloxacin, levofloxacin, or gemifloxacin. Azithromycin or clarithromycin. Dosage, 1 g PO three times/day. Dosage, 2 g PO two times/day. **High-dose amoxicillin (1 g, 3 times/day), high-dose amoxicillin-clavulanate (2 g, 2 times/day), cefpodoxime, cefprozil, or cefuroxime. Cefotaxime, ceftriaxone, ampicillin-sulbactam, or ertapenem. The antipseudomonal agents chosen reflect this concern. Risk factors for Pseudomonas infection include severe structural lung disease (e.g., bronchiectasis) and recent antibiotic therapy, health careassociated exposures or stay in hospital (especially in the ICU). For patients with CAP in the ICU, coverage for S. pneumoniae and Legionella species must always be considered. Piperacillin-tazobactam, imipenem, meropenem, and cefepime are excellent -lactams and are adequate for most S. pneumoniae and H. influenzae infections. They may be preferred when there is concern for relatively unusual CAP pathogens, such as P. aeruginosa, Klebsiella species, and other gram-negative bacteria. Piperacillin, piperacillin-tazobactam, imipenem, meropenem, or cefepime. Data suggest that older adults receiving aminoglycosides have worse outcomes. Dosage for hospitalized patients, 750 mg/daily. COPD, chronic obstructive pulmonary disease. Data from Mandell LA, Wunderink RG, Anzueto A, et al. Infectious Diseases Society of America/American Thoracic Society consensus guidelines of community-acquired pneumonia in adults. Clin Infect Dis. 2007;44:S27-S72.

Patient Characteristics Outpatient Previously Healthy No recent antibiotic therapy Recent antibiotic therapy

912

PART II

Major Clinical Syndromes

For a patient who does not require hospitalization and for whom no clear distinction between typical (e.g., pneumococcal) and atypical (mycoplasmal, chlamydial) pneumonia can be made, both types of organisms should be covered. Risks for the presence of drug-resistant S. pneumoniae should be assessed. Previous use of antibiotics, especially a -lactam or macrolide, is most predictive of the presence of resistance. Age alone, in general, should not influence drug choice. Where risk of drug-resistant S. pneumoniae is low, oral -lactam agents (amoxicillin-clavulanic acid, cefuroxime axetil), azalides/macrolides (azithromycin, clarithromycin, or erythromycin), or respiratory tract quinolones (levofloxacin, gemifloxacin, moxifloxacin) are all adequate choices. Increased resistance in strains of pneumococci to the azalide/ macrolide agents is a problem because therapeutic failures have been noted. Although it has been suggested that these agents may be used as long as the resistance rate is less than 25%, recent analysis suggests that resistance level is too high and will be associated with increased morbidity and mortality.308 Doxycycline (a preferred treatment option in the IDSA guideline) and trimethoprim-sulfamethoxazole may be used but are less favored because of decreasing activity against strains of pneumococci. For patients with an increased risk for poor outcome because of age or underlying disease, or where the risk for infection with resistant pneumococci exists because of prior antibiotic use, the respiratory tract quinolones are the agents most likely to be effective. They have activity against all strains of S. pneumoniae, including penicillin-resistant strains, and they have the added benefit of activity against atypical agents. Although resistance is a potential problem of increased use of quinolones, it has not yet emerged as a significant problem. A -lactam plus a macrolide is a comparable regimen. Regardless of the initial choice of antibiotic, once an organism is isolated, coverage should be narrowed down, if possible, on the basis of susceptibility. Patients who are ill enough to require hospitalization should be treated with parenteral agents that cover the likely pathogens. Our choice would be a -lactam (ceftriaxone or cefotaxime) plus azithromycin. A carbapenem may be used as the -lactam component of this regimen, although overall experience with it in this setting is limited. A respiratory tract quinolone would be a comparable regimen. If there are factors that suggest a specific etiology, or a Gram stain is revealing, specific antibiotic coverage should be used. Although these regimens represent the basic course of therapy, specific clinical circumstances may warrant variation. For example, S. aureus pneumonia, including CA-MRSA, should be considered during an influenza outbreak even though S. pneumoniae is still the major etiologic agent. Agents with activity against methicillin-resistant S. aureus (MRSA) should be used if there is reason to suspect its presence as the etiology of pneumonia. Vancomycin, linezolid, and quinupristin-dalfopristin are considerations. Where aspiration pneumonia is a possibility, agents with activity against oral anaerobes are needed, including ampicillin-sulbactam or clindamycin. Seven percent to 18% of community-acquired pneumonia cases are caused by aerobic gramnegative bacilli, including P. aeruginosa. Risk factors previously noted for gram-negative pneumonia should therefore be sought. Where gram-negative bacilli are suspected, infection with P. aeruginosa should be a concern, and therapy with an antipseudomonal -lactam compound (e.g., cefepime, piperacillin-tazobactam, imipenem, or meropenem) is a reasonable choice. Where Pseudomonas involvement can be excluded, agents such as cefotaxime, ceftriaxone, or a carbapenem could be considered. Debate exists as to whether agents such as aminoglycosides or quinolones need to be added to a -lactam therapy for gram-negative pneumonia. Data exist to support both sides of the controversy.309-311 We favor combination therapy for patients who are severely ill, at least until culture results from sputum and blood are available. In patients who are allergic to penicillin, aztreonam with a respiratory tract quinolone, with or without an aminoglycoside, could be used. In the patient admitted to an intensive care unit, therapy should be directed against S. pneumoniae, penicillin-resistant strains, Legionella

species, gram-negative rods, and M. pneumoniae. If infection with P. aeruginosa is unlikely (no recent hospitalization, no recent antibiotic use, no pulmonary comorbidities, no gram-negative rods), a -lactam plus either an azalide/macrolide or a respiratory tract quinolone would be therapies of first choice. Ceftriaxone or cefotaxime would be reasonable choices for the -lactam. Where Pseudomonas infection cannot be excluded, an antipseudomonal -lactam (cefepime, imipenem, meropenem, doripenem, or piperacillin-tazobactam) plus a respiratory tract quinolone or azalide/macrolide could be used. We favor cefepime or piperacillin-tazobactam plus a respiratory tract quinolone. An aminoglycoside could be added as a third agent for synergy against Pseudomonas. Evidence in the literature favoring one regimen over any other is lacking. The effect of initial antibiotic selection on outcomes in patients with pneumonia remains an area of ongoing controversy and study.312,313 Most of the evidence involves the use of nonexperimental cohort studies where the role of unmeasured variables is hard to determine. Therefore, until prospective, well-controlled studies are carried out, the real effect of initial antibiotic choice on pneumonia outcomes can only be estimated. DURATION OF THERAPY Until recently, the duration of antibiotic therapy for pneumonia has been based on anecdotal patterns of behavior. The classic 10 to 14 days of care is unsupported by evidence.314 Recent data suggest that clinical stability occurs more quickly, and therefore antibiotic therapy may be safely discontinued earlier.315-317 Clinical stability is defined as normalization of previously abnormal physiologic parameters, including heart rate, respiratory rate, oxygenation, blood pressure, mental state, and ability to care for oneself.315 Most physiologic abnormalities will correct in 2 to 3 days. Normalization of all physiologic abnormalities may take 5 to 7 days. Patients who score higher on the severity scale take longer to stabilize than patients who score lower. By day 3, relapses of previously corrected physiologic abnormalities occur only 4% to 6% of the time. Although meta-analysis has shown that there are few studies that are prospective, well controlled, use the same antibiotic and dosing schedule, and only vary the duration of therapy, those few studies have suggested that less than 7 days and as short as 3 days are just as effective as any longer durations of therapy316,317 for mild to moderate pneumonia. With age, presence of underlying comorbidities including immune compromise, and more virulent pathogens, clinical stability may be delayed, and therefore duration of antibiotic therapy may be lengthened. Oral antibiotic therapy is safe after clinical stability has been reached.318-320 There is no clear usefulness of observing a patient within the hospital after a switch to oral therapy.321 However, it is important to recognize that discharging patients before stability has been reached may lead to increased rehospitalization and mortality.322 Using the same definitions of clinical stability, it has been shown that the greater the number of factors remaining abnormal at discharge, the greater the chance of readmission or death. Once discharged, outpatient follow-up may be required, because greater than 64% of patients with community-acquired pneumonia will have some related residual symptoms, including fever, cough, shortness of breath, chest pain, sputum production, fatigue, or gastrointestinal symptoms. In younger patients without underlying pulmonary disease, these findings are more likely to resolve earlier. Aspiration pneumonia and lung abscess are discussed in Chapter 66. Nosocomial pneumonia is discussed in Chapter 303. ADJUNCTIVE THERAPY With the recognition that the inflammatory response is a balance between proinflammatory and anti-inflammatory mediators, the effects of agents with anti-inflammatory properties on the treatment of pneumonia is being studied. The effects of the macrolides in this

64 Acute Pneumonia

913

regard has been discussed earlier. Critical illness-related corticosteroid insufficiency (CIRCI) has been associated with community-acquired pneumonia. Theoretically, steroid use would help correct the deficient performance of the hypothalamic-pituitary-adrenal axis. Though several studies have examined the role of steroids in this setting, results have been contradictory and no clear role has been defined.323-325 Statins also possess anti-inflammatory properties. Recent observational studies have reported decreased mortality in patients who had been using statins before admission for pneumonia. The studies were not randomized or controlled for other potentially important variables such as underlying health, or socioeconomic status.326,327 Although other adjunctive therapies have been described, they have not yet been shown to have a significant role in therapy.328

Pneumonia Prevention
Vaccination against influenza and perhaps S. pneumoniae are important interventions in preventing pneumonia. In older adults, influenza vaccine may decrease the incidence of pneumonia by 53%.329,330 Influ-

enza vaccine is suggested for any person 6 months of age or older, who, because of age or underlying disease, is at risk for influenza-related complications. This includes persons older than 50 years; nursing home residents; people with chronic pulmonary or cardiac disease, or with chronic diseases such as diabetes, renal failure, or hematologic disorders; patients who are immunosuppressed; those taking chronic salicylate therapy; and women in their second or third trimester of pregnancy. Health care workers, workers in nursing homes, and those who provide care to older adults or debilitated persons should also be targeted for influenza vaccination.331 The role of pneumococcal polysaccharide vaccine continues to be somewhat controversial. The incidence of bacteremia may be reduced, but the incidence of pneumonia appears unchanged.332 The currently available protein polysaccharide vaccine is for pediatric use only. Although initial data showed that disease in adults was reduced by the introduction of the conjugate vaccine in children, it is clear that nonvaccine strains, including those resistant to penicillin, are emerging as important causes of pneumonia. Pneumococcal vaccine is recommended for patients older than 65 and those who have recovered from community-acquired pneumonia (see Chapter 200).

REFERENCES
1. Osler W. The Principles and Practice of Medicine. 4th ed. New York: Appleton; 1901:108. 2. Minino A, Heron AP, Smith BL. Death, preliminary data for 2004. National Vital Statistics Report. 2006;54(19):1-52. 3. Sibille Y, Reynolds HY. Macrophage and polymorphonuclear neutrophils in lung defense and injury. Am Rev Respir Dis. 1990;141:471-501. 4. Reynolds HY. Pulmonary host defenses. Chest. 1989;95:223S230S. 5. Wright JR. Immunomodulatory functions of surfactant. Physiol Rev. 1997;77:931-962. 6. Zhang P, Summer WR, Bagby GJ, et al. Innate immunity and pulmonary host defense. Immunol Rev. 2000;173:39-51. 7. Lohmann-Matthes ML, Steinmller C, Franke-Ullman G. Pulmonary macrophages. Eur Respir J. 1994;7:1678-1689. 8. MacNee W, Selby C. Neutrophil kinetics in the lung. Clin Sci. 1990;79:97-107. 9. Reynolds HY. Advances in understanding pulmonary host defense mechanisms: Dendritic cell function and immunomodulation. Curr Opin Pulm Med. 2000;6:209-216. 10. Strieter RM, Standiford TJ, Huffnagle GB, et al. The good, the bad, and the ugly: The role of chemokines in models of human disease. J Immunol. 1996;196:3583-3586. 11. Prince AS, Mizgerd JP, Wiener-Kronish J, et al. Cell signaling underlying the pathophysiology of pneumonia. Am J Physiol Lung Cell Mol Physiol. 2006;291:297-300. 12. Chaudhuri N, Whyte M, Sabroe I. Reducing the toll of inflammatory lung disease. Chest. 2007;131:1550-1556. 13. Puren A, Feldman C, Savage N, et al. Patterns of cytokine expression in community-acquired pneumonia. Chest. 1995; 107:1342-1349. 14. Strieter RM, Belperio JA, Keane MP. Host innate defenses in the lung: The role of cytokines. Curr Opin Infect Dis. 2003;16:193198. 15. Mizgerd P. Acute lower respiratory tract infection. N Engl J Med. 2008;358:716-727. 16. Strieter RM, Kunkel S, Showell H, et al. Endothelial cell gene expression of a neutrophil chemotactic factor by TNF-, LPS, and IL-. Science. 1989;243:1467-1469. 17. Agostini C, Chilosi M, Zambello R, et al. Pulmonary immune cells in health and disease. Lymphocytes. 1993;6:1378-1401. 18. Fishman AP, Reynolds HY, Elias JA, et al. Pulmonary defense mechanisms against infection. In: Fishman AP, ed. Fishmans Pulmonary Diseases and Disorders. 3rd ed. New York: McGraw Hill; 1998;265-274. 19. Huxley EJ, Viroslav J, Gray WR, et al. Pharyngeal aspiration in normal adults and patients with depressed consciousness. Am J Med. 1978;64:564-568. 20. Green GM, Carolin D. The depressant effect of cigarette smoke on the in vitro antibacterial activity of alveolar macrophages. N Engl J Med. 1967;276:421-427. 21. MacGregor RR. Alcohol and immune defense. JAMA. 1986;256: 1474-1479. 22. Schopf RE, Trompter M, Bork K, et al. Effects of ethanol and acetaldehyde on phagocytic function. Arch Dermatol Res. 1985;277:131-137. 23. Zhang P, Bagby GJ, Happel KI, et al. Pulmonary host defenses and alcohol. Front Biosci. 2002;7:1314-1330. 24. Nelson S, Mason LM, Kolls J, et al. Pathophysiology of pneumonia. Clin Chest Med. 1995;16:1-12. 25. Warshauer D, Goldstein E, Akers T, et al. Effect of influenza viral infection on the ingestion and killing of bacteria by alveolar macrophages. Am Rev Respir Dis. 1977;115:269-277. 26. Nelson S, Chidiac C, Bagby G, et al. Endotoxin-induced suppression of lung host defenses. J Med. 1990;21:85-103. 27. Phair J, Palella FJ Jr. Host impairments associated with human immunodeficiency virus infection. In: Niederman MS, Sarosi GA, Glassroth J, eds. Respiratory Infections. 2nd ed. Philadelphia: Lippincott Williams & Wilkins; 2001:59-66. 28. Mason CM. The pathogenesis and presentation of nosocomial pneumonia. Crit Care Rev. 1991;2:145-155. 29. Espesito AL. Aspirin impairs antibacterial mechanisms in experimental pneumococcal pneumonia. Am Rev Respir Dis. 1984; 130:857-862. 30. Nelson S, Summer WR, Terry PB, et al. Erythromycin-induced suppression of pulmonary antibacterial defenses: A potential mechanism of superinfection in the lung. Am Rev Respir Dis. 1987;136:1207-1212. 31. Nelson S, Summer WR, Jakab EJ. Aminophylline-induced suppression of pulmonary antibacterial defenses. Am Rev Respir Dis. 1985;131:923-927. 32. Gulmez SE, Holm E, Frederiksen H, et al. Use of proton pump inhibitors and the risk of community-acquired pneumonia. Arch Intern Med. 2007;167:950-956. 32a. Herzig HJ, Howell MD, Ngo LH, et al. Acid-suppressive medication use and the risk of hospital-acquired pneumonia. JAMA. 2009;301:2120-2128. 33. Ehrlich R, Henry MC. Chronic toxicity of nitrogen dioxide: 1. Effect on resistance to bacterial pneumonia. Arch Environ Health. 1968;17:860-865. 34. LaForce FM, Mullane JF, Boehme RF, et al. The effect of pulmonary edema on antibacterial defenses of the lung. J Lab Clin Med. 1973;82:634-648. 35. Huber GL, LaForce FM, Mason RJ, et al. Impairment of pulmonary bacterial defense mechanisms by immunosuppressive agents. Surg Forum. 1970;21:285-286. 36. Gyetko MR, Toews GB. Immunology of the aging lung. Clin Chest Med. 1993;14:379-391. 37. Geppert EF. Chronic recurrent pneumonia. Semin Respir Infect. 1992;7:282-288. 38. Donowitz GR, Mandell GL. Clinical presentation and unusual infections. In: Gallin JI, Fauci AS, eds. Advances in Host Defense Mechanisms, vol 3. New York: Raven Press; 1983:55-75. 39. Donabedian H, Gallin JI. The hyperimmunoglobulin E recurrent infection (Jobs) syndrome. Medicine (Baltimore). 1983;62: 195-208. 40. Beck S, Heiner DC. Selective immunoglobulin G4 deficiency and recurrent infections of the respiratory tract. Am Rev Respir Dis. 1981;124:94-96. 41. Eliasson R, Mossberg B, Camner P, et al. The immotile-cilia syndrome. N Engl J Med. 1977;297:1-6. 42. Kartagener M. Zur Pathologie der Bronchiektasien: Bronkiectasien bei Situs Inversus. Beitr Klin Tuberk. 1933;83:489-501. 43. Metlay JP, Stafford RS, Singer DE. National trends in the use of antibiotics by primary care physicians for adult patients with cough. Arch Intern Med. 1998;158:1813-1818. 44. Metlay JP, Schulz R, Li YH, et al. Influence of age on symptoms at presentation in patients with community-acquired pneumonia. Arch Intern Med. 1997;157:1453-1459. 45. Halm EA, Teirstein AS. Management of community-acquired pneumonia. N Engl J Med. 2002;347:2039-2045. 46. Murray HW, Masur H, Senterfit L, et al. The protean manifestations of Mycoplasma pneumoniae infection in adults. Am J Med. 1975;58:229-242. 47. Marston BJ, Plouffe JF, File TM, et al. Incidence of communityacquired pneumonia requiring hospitalization: Results of a 48. 49. population-based active surveillance study in Ohio. Arch Intern Med. 1997;157:1709-1718. Braun MM, Cot TR, Rabkin CS. Trends in death with tuberculosis during the AIDS era. JAMA. 1993;269:2865-2868. Gillet Y, Issartel B, Vanhems P, et al. Association between Staphylococcus aureus strains carrying gene for Panton-Valentine leukocidins and highly lethal necrotizing pneumonia in young immunocompetent patients. Lancet. 2002;359:753-759. Monaco M, Antonucci R, Palange P, et al. Methicillin-resistant Staphyloccus aureus necrotizing pneumonia. Emerg Infect Dis. 2005;11:1647-1648. Franklin B. SARS: 1918 Revisited? The urgent need for global collaboration in public health. Mayo Clinic Proc. 2003;78:813816. Sullivan RJ, Dowdle WR, Marine WM, et al. Adult pneumonia in a general hospital: Etiology and host risk factors. Arch Intern Med. 1972;129:935-942. Almirall J, Bolbor I, Serra-Prat M, et al. New evidence of risk factors for community-acquired pneumonia: A populationbased study. Eur Respir J. 2008;31:1274-1284. Lepow ML, Balassanian N, Emmerich J, et al. Interrelationships of viral, mycoplasmal and bacterial agents in uncomplicated pneumonia. Am Rev Respir Dis. 1968;97:533-545. Griffith DE, Mazurek GH. Pneumonia in chronic obstructive lung disease. Infect Dis Clin North Am. 1991;5:467-484. Torres A, Dorca J, Zalacain R, et al. Community-acquired pneumonia in chronic obstructive pulmonary disease: A Spanish multicenter study. Am J Respir Crit Care Med. 1996;154:14561461. Pietila MP, Thomas CF. Inflammation and infection in exacerbations of chronic obstructive pulmonary disease. Semin Respir Infect. 2003;18:9-16. Hoiby N. Epidemiological investigations of the respiratory tract bacteriology in patients with cystic fibrosis. Acta Pathol Microbiol Scand B. 1974;82:541-550. Wallace JM, Rao AV, Glassroth J, et al. Respiratory illness in person with acquired immunodeficiency virus infection. Am Rev Respir Dis. 1993;148:1523-1529. Rosen MJ. Overview of pulmonary complications. Clin Chest Med. 1996;17:621-631. Davis J, Fei M, Huang L. Respiratory infection complicating HIV infection. Curr Opin Infect Dis. 2008;21:184-190. Hoover DR, Saah AJ, Bacellar H, et al. Clinical manifestations of AIDS in the era of pneumocystis prophylaxis. Multicenter AIDS Cohort Study. N Engl J Med. 1993;329:1922-1926. Hull MW, Phillips P, Montaner SG. Changing global epidemiology of pulmonary manifestations of HIV/AIDS. Chest. 2008; 134:1287-1298. Hirschtick RE, Glassroth J, Jordan MC, et al. Bacterial pneumonia in persons infected with the human immunodeficiency virus. N Engl J Med. 1995;333:845-851. Hanson DL, Chusy SY, Farizo KM, et al. Distribution of CD4 T lymphocytes at diagnosis of acquired immunodeficiency syndrome-defining and other human immunodeficiency virusrelated illness. Arch Intern Med. 1995;155:1537-1542. Gross PA. Epidemiology of hospital-acquired pneumonia. Semin Respir Infect. 1987;2:2-7. Marrie TJ, Haldane EV, Faulkner RS, et al. Community acquired pneumonia requiring hospitalization: Is it different in the elderly? J Am Geriatr Soc. 1985;33:671-680. Carratala J, Garcia-Vidal G. What is healthcare-associated pneumonia and how it is managed? Curr Opin Infect Dis. 2008;21:168173.

50. 51. 52. 53. 54. 55. 56.

57. 58. 59. 60. 61. 62. 63. 64. 65.

66. 67. 68.

914

PART II

Major Clinical Syndromes

69. American Thoracic Society, Infectious Diseases Society of America. Guidelines for the management of adults with hospital-acquired, ventilator-associated, and healthcare associatedpneumonia. Am J Respir Crit Care Med. 2007;171:388-416. 70. Carratala J, Mykietiuk A, Fernandez-Sabe N, et al. Health careassociated pneumonia requiring hospital admission. Arch Intern Med. 2007;167:1393-1399. 71. Heffron R. Pneumonia. New York: Commonwealth Fund; 1939:505. 72. Bartlett JG, Finegold SM. Anaerobic infections of the lung and pleural space. Am Rev Respir Dis. 1974;110:56-77. 73. Woodhead M. Management of pneumonia in the outpatient setting. Semin Respir Infect. 1998;3:8-16. 74. Metlay JP, Kapoor WN, Fine MJ. Does this patient have community-acquired pneumonia? JAMA. 1997;278:1440-1445. 75. Wipf JE, Lipsky BA, Hirschmann JV, et al. Diagnosing pneumonia by physical examination: Relevant or relic? Arch Intern Med. 1999;159:1082-1087. 76. van der Eerden MM, Vlaspolder F, de Graaff CS, et al. Comparison between pathogen directed antibiotic treatment and empirical broad spectrum antibiotic treatment in patients with community acquired pneumonia: A prospective randomised study. Thorax. 2005;60:672-678. 77. Hohenthal U, Vainionpaa R, Meurman O, et al. Aetiological diagnosis of community acquired pneumonia: Utility of rapid microbiological methods with respect to disease severity. J Infect Dis. 2008;40:131-138. 78. Mandell LA, Wunderink RG, Anzueto A, et al. Infectious Disease Society of American/American Thoracic Society consensus guidelines of community-acquired pneumonia in adults. Clin Infect Dis. 2007;44:S27-S72. 79. Roson B, Carratala J, Verdaguer R, et al. Prospective study of the usefulness of sputum Gram stain in the initial approach of community-acquired pneumonia requiring hospitalization. Clin Infect Dis. 2000;31:869-874. 80. Musher DM, Montoy AR, Wanahita A. Diagnostic value of microscopic examination of Gram-stained sputum and sputum cultures in patients with bacteremic pneumococcal pneumonia. Clin Infect Dis. 2004;30:165-169. 81. Miyashita N, Shimizu H, Ouchi K, et al. Assessment of the usefulness of sputum gram stain and culture for diagnosis of community-acquired pneumonia requiring hospitalization. Med Sci Monit. 2008;14:171-176. 82. Bartlett JG. Decline in microbial studies for patients with pulmonary infection. Clin Infect Dis. 2004;39:170-172. 83. George RB, Ziskind MM, Rasch JR, et al. Mycoplasma and adenovirus pneumonias: Comparison with other atypical pneumonias in a military population. Ann Intern Med. 1966;65: 931-942. 84. Reimann H. The Pneumonias. Philadelphia: Saunders; 1938:67. 85. Solomon S. Primary Friedlander pneumonia. JAMA. 1937;108: 937-947. 86. Rein MF, Gwaltney JM, OBrien WM, et al. Accuracy of the Grams stain in identifying pneumococci in sputum. JAMA. 1978;239:2671-2673. 87. Hendley JO, Sande MA, Stewart PM, et al. Spread of Streptococcus pneumoniae in families: I. Carriage rates and distribution of types. J Infect Dis. 1975;132:55-61. 88. Barrett-Connor E. The non-value of sputum culture in the diagnosis of pneumococcal pneumonia. Am Rev Respir Dis. 1971; 103:845-848. 89. Zaman MK, Wooten OH, Suprahmonya B, et al. Rapid noninvasive diagnosing of Pneumocystis carinii from induced liquefied sputum. Ann Intern Med. 1988;109:7-10. 90. Wallace RJ, Musher DM, Martin RR. Haemophilus influenzae pneumonia in adults. Am J Med. 1978;64:87-93. 91. Levin D, Schwarz M, Matthay R, et al. Bacteremic Haemophilus influenzae pneumonia in adults: A report of 24 cases and a review of the literature. Am J Med. 1977;62:219-224. 92. Davidson M, Tempest B, Palmer DL. Bacteriologic diagnosis of acute pneumonia: Comparison of sputum, transtracheal aspirates, and lung aspirates. JAMA. 1976;235:158-163. 93. Skerrett SJ. Diagnostic testing to establish a microbial cause is helpful in the management of community-acquired pneumonia. Semin Respir Infect. 1997;12:308-321. 94. Strlin K. Usefulness of aetiological tests for guiding antibiotic therapy in community acquired pneumonia. Int J Antimicrob Agents. 2008;31:3-11. 95. Diederen BMW, Kluytmans JA, Vanderbroucke-Grauls CM. Utility of real-time PCR for diagnosis of Legionnaires disease in routine clinical practice. J Clin Microbiol. 2008;2:671-677. 96. Murdoch DR. Molecular genetic methods in the diagnosis of lower respiratory tract infections. APMIS. 2004;112:713-727. 97. Nolte FS. Molecular diagnostics for detection of bacterial and viral pathogens in community-acquired pneumonia. Clin Infect Dis. 2008;47:S123-126. 98. Baselski VS, Wunderink RG. Bronchoscopic diagnosis of pneumonia. Clin Microbiol Rev. 1994;7:533-558. 99. Pollock HM, Hawkins EL, Bonner JR, et al. Diagnosis of bacterial pulmonary infections with quantitative protected catheter cultures obtained during bronchoscopy. J Clin Microbiol. 1983; 17:255-259. 100. van der Eerden MM, Vlaspolder F, de Graaff, CS, et al. Value of intensive diagnostic microbiological investigation in low- and high-risk patients with community-acquired pneumonia. Eur J Clin Microbiol Infect Dis. 2005;24:241-249.

101. Chastre J, Fagon J, Barnet-Lesco M, et al. Evaluation of bronchoscopic techniques for the diagnosis of nosocomial pneumonia. Am J Respir Crit Care Med. 1995;152:231-240. 102. Broughton WA, Middleton RM, Kirkpatrick MB, et al. Bronchoscopic protected specimen brush and bronchoalveolar lavage in the diagnosis of bacterial pneumonia. Infect Dis Clin North Am. 1991;5:432-452. 103. Marquette CH, Copin M, Wallet F, et al. Diagnostic tests for pneumonia in ventilated patients: Prospective evaluation of diagnostic accuracy using histology as a diagnostic gold standard. Am J Respir Crit Care Med. 1995;151:1878-1888. 104. Michand S, Suzuki S, Harbarth S. Effects of design-related bias in studies of diagnostic tests for ventilator-associated pneumonia. Am J Respir Crit Care Med. 2002;166:1320-1325 105. Timsit J, Misset B, Renard B, et al. Effect of previous antimicrobial therapy on the accuracy of the main procedure used to diagnose nosocomial pneumonia in patients who are using ventilation. Chest. 1995;108:1036-1040. 106. Broaddus C, Dake MD, Stulburg MS, et al. Bronchoalveolar lavage and transbronchial biopsy for the diagnosis of pulmonary infections in the acquired immunodeficiency syndrome. Ann Intern Med. 1986;102:747-752. 107. Pisani RJ, Wright AJ. Clinical utility of bronchoalveolar lavage in immunocompromised hosts. Mayo Clin Proc. 1992;67:221227. 108. Baughman RP, Dohn MN, Loudon RG, et al. Bronchoscopy with bronchoalveolar lavage in tuberculosis and fungal infection. Chest. 1991;99:92-97. 109. Bovornkitti S, Pushpakom R. Adenosine deaminase in bronchoalveolar lavage fluid. Chest. 1988;94:1113. 110. Raja A, Baughman RP, Daniel TM. The detection by immune assay of antibody to mycobacterial antigens in bronchoalveolar lavage fluid from patients with tuberculosis and control subjects. Chest. 1988;94:133-137. 111. Gerbeaux P, Ledoray V, Boussuges A, et al. Diagnosis of nosocomial pneumonia in mechanically ventilated patients. Am J Respir Crit Care Med. 1998;157:76-80. 112. Jourdain B, Joly-Guillou ML, Dombret MC, et al. Usefulness of quantitative cultures of BAL fluid for diagnosing nosocomial pneumonia in ventilated patients. Chest. 1997;111:411-418. 113. El-Ebiary M, Torres A, Gonzalez J, et al. Quantitative cultures of endotracheal aspirates for the diagnosis of ventilatorassociated pneumonia. Am Rev Respir Dis. 1993;148:15521557. 114. Jourdain B, Novora A, Joly-Guillou ML, et al. Role of quantitative values of endotracheal aspirate in the diagnosis of nosocomial pneumonia. Am J Respir Crit Care Med. 1995;152:241246. 115. Marik PE, Brown WJ. A comparison of bronchoscopic vs blind protected specimen brush sampling in patients with suspected ventilator-associated pneumonia. Chest. 1995;108:203-207. 116. Kollef MH, Bock KR, Richards RD, et al. The safety and diagnostic accuracy of minibronchoalveolar lavage in patients with suspected ventilator-associated pneumonia. Ann Intern Med. 1995;122:743-748. 117. Bregeon F, Papazian L, Visconti A, et al. Relationship of microbiologic diagnostic criteria to morbidity and mortality in patients with ventilator-associated pneumonia. JAMA. 1997; 277:655-662. 118. Busk MF, Rosenow EC III, Wilson WR. Invasive procedures in the diagnosis of pneumonia. Semin Respir Infect. 1988;3:113122. 119. Torres A, Jimenez P, de la Bellacasa JP, et al. Diagnostic value of nonfluoroscopic percutaneous lung needle aspiration in patients with pneumonia. Chest. 1990;98:840-844. 120. Cockerill FR III, Wilson WR, Carpenter HA, et al. Open lung biopsy in immunocompromised patients. Arch Intern Med. 1985;145:1398-1404. 121. Dijkman JH, van der Meer JWM, Bakker W, et al. Transpleural lung biopsy by the thoracoscopic route in patients with diffuse interstitial pulmonary disease. Chest. 1982;82:76-83. 122. Rodgers BM. Thoracoscopy in children. Poumon Coeur. 1981;37:301-306. 123. Springmeyer SC, Silvestri RC, Sale GE, et al. The role of transbronchial biopsy for the diagnosis of diffuse pneumonias in immunocompromised marrow transplant recipients. Am Rev Respir Dis. 1982;116:763-765. 124. Light RW. Parapneumonic effusions and empyema. Proc Am Thorac Soc. 2006;3:75-80. 125. Sahn S. Diagnosis and management of parapneumonic effusions and empyema. Clin Infect Dis. 2007;45:1480-1486. 126. Maskell NA, Davies CWH, Nunn AJ, et al. U.K. Controlled trial of intrapleural streptokinase for pleural infection. N Engl J Med. 2005;352:865-874. 127. Nagesh BS, Sehgol S, Jindd SK, et al. Evaluation of polymerase chain reaction for detection of Mycobacterium tuberculosis in pleural fluid. Chest. 2001;119:1737-1741. 128. Greco S, Girardi E, Masciangelo R, et al. Adenosine deaminase and interferon gamma measurements for the diagnosis of tuberculous pleurisy: A meta-analysis. Int J Tuberc Lung Dis. 2003;7: 777-786. 129. Liang QL, Shi HZ, Wang K, et al. Diagnostic accuracy of adenosin deaminase in tuberculous pleurisy: A meta-analysis. Respir Med. 2008;102:744-754. 130. Metersky ML, Ma A, Bratzler DW, et al. Predicting bacteremia in patients with community-acquired pneumonia. Am J Respir Crit Care Med. 2004;169:342-347.

131. Campbell SG, Marrie TJ, Anstey R, et al. The contribution of blood cultures to the clinical management of adult patients admitted to the hospital with community-acquired pneumonia: A prospective observational study. Chest. 2004;123:1142-1150. 132. Bordon J, Peyrani P, Brock GN, et al. The presence of pneumococcal bacteremia dose not influence clinical outcomes in patients with community-acquired pneumonia. Chest. 2008;133: 618-624. 133. Craven D. Blood cultures for community-acquired pneumonia piecing together a mosaic for doing less. Am J Respir Crit Care Med. 2004;169:327-335. 134. Marrie TJ. Blood cultures in ambulatory patients who are discharged from emergency with community-acquired pneumonia. Can J Infect Dis. 2004;15:21-24. 135. Berk SL. Justifying the use of blood cultures when diagnosing community-acquired pneumonia. Chest. 1995;108:891-892. 136. Miyashita N, Ouchi K, Kawasaki K, et al. Comparison of serological tests for detection of immunoglobulin M antibodies to Chlamydophila pneumoniae. Respirology. 2008;13:427-431. 137. Dowell SF, Peeling RW, Boman J, et al. Standardizing Chlamydia pneumoniae assays: Recommendations from the Centers for Disease Control and Prevention (USA) and the Laboratory Centre for Disease Control (Canada). Clin Infect Dis. 2001;33:492503. 138. Sillis M. The limitation of IgM assays in the serologic diagnosis of Mycoplasma pneumoniae infections. J Med Microbiol. 1990;33: 253-258. 139. Venkatesan P, MacFarlane JT. Role of pneumococcal antigen in the diagnosis of pneumococcal pneumonia [editorial]. Thorax. 1992;47:329-331. 140. Kee C, Palladino S, Kay I, et al. Feasibility of real-time polymerase chain reaction in whole blood to identify Streptococcus pneumoniae in patients with community-acquired pneumonia. Diagn Microbiol Infect Dis. 2008;61:72-75. 141. Campbell JF, Spika JS. The serodiagnosis of nonpneumococcal bacterial pneumonia. Semin Respir Infect. 1988;3:123-130. 142. Schlager N, Godwin K, Sepkowitz D, et al. Application of DNA amplification to pneumocystosis: Presence of serum Pneumocystis carinii DNA during human and experimentally induced Pneumocystis carinii pneumonia. J Exp Med. 1992;176:13271333. 143. Niederman MS. Biological markers to determine eligibility in trials for community-acquired pneumonia: A focus on procalcitonin. Clin Infect Dis. 2008;47:S127-S132. 144. Christ-Crain M, Buller B. Biomarkers in respiratory tract infections: Diagnostic guides to antibiotic prescription, prognostic markers and mediators. Eur Respir J. 2007;30:556-573. 145. Povoa P. Serum markers in community-acquired pneumonia and ventilator-associated pneumonia. Curr Opin Infect Dis. 2008;21:157-162. 146. Chalmers JD, Singanayagam A, Hill AT. C-Reactive protein is an independent predictor of severity in community-acquired pneumonia. Am J Med. 2008;121:219-255. 147. Crist-Crain M, Jaccard-Stolz D, Bingisser R, et al. Effect of procalcitonin-guided treatment on antibiotic use and outcome in lower respiratory tract infections: Cluster randomized, singleblinded intervention trial. Lancet. 2006;363:600-607. 148. Christ-Crain M, Stolz D, Bingisser R, et al. Procalcitonin guidance of antibiotic therapy in community-acquired pneumonia: A randomized trial. Am J Respir Crit Care Med. 2006;174:8493. 149. Jensen JU, Heslet L, Jensen TH, et al. Procalcitonin increases in early identification of critically ill patients at high risk of mortality. Crit Care Med. 2006;34:2596-2602. 150. Salluh JI, Bozza FA, Soares M, et al. Adrenal response in severe community-acquired pneumonia: Impact on outcomes and disease severity. Chest. 2008;134:947-954. 151. Christ-Crain M, Stolz D, Jutla S, et al. Free and total cortisol levels as predictors of severity and outcome in communityacquired pneumonia. Am J Respir Crit Care Med. 2007;176:913920. 152. Brivet FG, Jacobs FM, Prat D, et al. Sophisticated biomarkers for community-acquired pneumonia severity assessment: Gadgets or useful tools? Intensive Care Med. 2008;34:975-976. 153. Ruf B, Schurmann D, Horbach I, et al. Frequency and diagnosis of Legionella pneumophila: A 3-year prospective study with emphasis on application of urinary antigen detection. J Infect Dis. 1990;162:1341-1347. 154. Domnguez J, Gal N, Blanco S, et al. Detection of Streptococcus pneumoniae antigen by a rapid immunochromatographic assay in urine samples. Chest. 2001;119:243-249. 155. Rosn B, Fernndez-Sab N, Carratal J, et al. Contribution of a urinary antigen assay (Binax NOW) to the early diagnosis of pneumococcal pneumonia. Clin Infect Dis. 2004;38:222-226. 156. Ishida T, Hashimoto T, Arita M, et al. A 2-year prospective study of a urinary antigen-detection test for Streptococcus pneumoniae in community-acquired pneumonia: Utility and clinical impact on the reported etiology. J Infect Chemother. 2004;10:359363. 157. Marcos MA, Jimenez de Anta MT, de la Bellacasa JP, et al. Rapid urinary antigen test for diagnosis of pneumococcal communityacquired pneumonia in adults. Eur Respir J. 2003;21:209-214. 158. Murdoch DR, Laing RTR, Mills GD, et al. Evaluation of a rapid immunochromatographic test for detection of Streptococcus pneumoniae antigen in urine samples from adults with community-acquired pneumonia. J Clin Microbiol. 2001;39:34953498.

64 Acute Pneumonia

915

159. Gutirrez F, Masi M, Rodrguez JC, et al. Evaluation of the immunochromatographic Binax NOW assay for detection of Streptococcus pneumoniae urinary antigen in a prospective study of community-acquired pneumonia in Spain. Clin Infect Dis. 2003;36:286-292. 160. Dowell SF, Garman RL, Liu G, et al. Evaluation of Binax NOW, an assay for the detection of pneumococcal antigen in urine samples performed among pediatric patients. Clin Infect Dis. 2001;32:824-825. 161. del Mar Garcia-Suarez M, Cima-Cabal MD, Villaverde R, et al. Performance of a pneumolysin enzyme-linked immunosorbent assay for diagnosis of pneumococcal infections. J Clin Microbiol. 2007;45:3549-3554. 162. Albaum MN, Hill LC, Murphy M, et al. Interobserver reliability of the chest radiograph in community-acquired pneumonia. Chest. 1996;110:343-350. 163. Boersma WM, Daniels JMA, Lowenberg A, et al. Reliability of radiographic findings and the relation to etiologic agents in community-acquired pneumonia. Respir Med. 2006;100:926932. 164. Gonzales R, Steiner JF, Sande MA. Antibiotic prescribing for adults with colds, upper respiratory tract infections and bronchitis by ambulatory care physicians. JAMA. 1997;278:901904. 165. Rubinstein E, Kollef MH, Nathwani D. Pneumonia caused by methicillin-resistant Staphylococcus aureus. Clin Infect Dis. 2008;46:S378-S385. 166. Gillet E, Etienne J, Lina G, et al. Association of necrotizing pneumonia with panton-valentine leukocidin-producing Staphlococcus aureus, regardless of methicillin resistance. Clin Infect Dis. 2008;47:985-986. 167. Kim EA, Soo Lee K, Primack SL, et al. Viral pneumonias in adults: Radiologic and pathologic findings. Radiographics. 2002;22:S137-149. 168. Chong S, Lee KS, Kim TS, et al. Adenovirus pneumonia in adults: Radiographic and high-resolution CT findings in five patients. AJR Am J Roentgenol. 2006;186:1288-1293. 169. Poutanen SM, Low DE, Henry B, et al. Identification of severe acute respiratory syndrome in Canada. N Engl J Med. 2003; 348:1995-2005. 170. Wong KT, Antonio GE, Hui DS, et al. Severe acute respiratory syndrome: Radiographic appearances and pattern of progression in 138 patients. Radiology. 2003;228:401-406. 171. Johnstone J, Majumdar SR, Fox JD, et al. Human metapneumovirus pneumonia in adults: Results of a prospective study. Clin Infect Dis. 2008;46:571-574. 172. Hamelin ME, Cote S, Laforge J, et al. Human metapneumovirus infection in adults with community-acquired pneumonia and exacerbation of chronic obstructive pulmonary disease. Clin Infect Dis. 2005;41:498-502. 173. Fine NL, Smith LR, Sheedy PF. Frequency of pleural effusions of mycoplasma and viral pneumonias. N Engl J Med. 1970;283:790-793. 174. Fraser DW, Tsai TR, Orenstein W, et al. Legionnaires disease: Description of an epidemic of pneumonia. N Engl J Med. 1977;297:1189-1197. 175. Pope TL Jr, Armstrong P, Thompson R, et al. Pittsburgh pneumonia agent chest film manifestations. AJR Am J Roentgenol. 1982;138:237-241. 176. Syrjl H, Broas M, Suramo I, et al. High resolution computed tomography for the diagnosis of community-acquired pneumonia. Clin Infect Dis. 1998;27:358-363. 177. Hayden G, Wrenn K. Chest radiograph vs computed tomography in the evaluation for pneumonia. J Emerg Med. 2009;36: 266-270. 178. Kramer EL, Chaitanya RD. Pulmonary applications of nuclear medicine. Clin Chest Med. 1991;12:55-75. 179. Marrie TJ, Peeling RW, Fine MJ, et al. Ambulatory patients with community-acquired pneumonia: The frequency of atypical agents and clinical course. Am J Med. 1996;101:508-515. 180. Fang GD, Fine M, Orloff J, et al. New and emerging etiologies for community-acquired pneumonia with implications for therapy: A prospective multicenter study of 359 cases. Medicine (Baltimore). 1990;69:307-316. 181. Fine M, Smith D, Singer DE. Hospitalization decision in patients with community-acquired pneumonia: A prospective chart review. Am J Med. 1990;89:713-714. 182. Mundy LM, Auwaerter PG, Oldach D, et al. Communityacquired pneumonia: Impact of immune status. Am J Respir Crit Care Med. 1995;152:1309-1315. 183. Austrian R, Gold J. Pneumococcal bacteremia with especial reference to bacteremic pneumococcal pneumonia. Ann Intern Med. 1964;60:759-776. 184. Lipsky BA, Boyko EJ, Inui TS, et al. Risk factors for acquiring pneumococcal infections. Arch Intern Med. 1986;146:21792185. 185. Rosner F, Zarrabi MH. Late infections following splenectomy in Hodgkins disease. Cancer Invest. 1983;1:57-65. 186. Zarrabi MH, Rosner F. Serious infections in adults following splenectomy for trauma. Arch Intern Med. 1984;144:14211424. 187. Musher DM. Infections caused by Streptococcus pneumoniae: Clinical spectrum, pathogenesis, immunity and treatment. Clin Infect Dis. 1992;14:801-809. 188. Hirschmann JV, Everett ED. Haemophilus influenzae infections in adults: Report of nine cases and a review of literature. Medicine (Baltimore). 1979;58:80-94.

189. Dworkin MS, Park L, Borchardt SM. The changing epidemiology of invasive Haemophilus influenzae disease, especially in persons > or = 65 years old. Clin Infect Dis. 2007;44:810-816. 190. Hausmann W, Karlish AJ. Staphylococcal pneumonia in adults. Br Med J. 1956;2:845-847. 191. Dufour P, Gillet Y, Bes M, et al. Community-acquired methicillin-resistant Staphyloccus aureus infection in France: emergence of a single clone that produces Panton-Valentine leukocidin. Clin Infect Dis. 2002;35:819-824. 192. Hageman JC, Uyeki TM, Francis JS, et al. Severe communityacquired pneumonia due to Staphylococcus aureus, 2003-04 influenza season. Emerg Infect Dis. 2006;6:894-899. 193. Kang CI, Song JH, Oh WS, et al. Clinical outcomes and risk factors for community-acquired pneumonia caused by gramnegative bacilli. Eur J Clin Micro Inf Dis. 2008;27:657-661. 194. Yu VL, Kroboth FJ, Shonnard J, et al. Legionnaires disease: New clinical perspective from a prospective pneumonia study. Am J Med. 1982;73:357-361. 195. Neil K, Berkelman R. Increasing incidence of legionellosis in the United States. 1900-2005: Changing epidemiologic trends. Clin Infect Dis. 2008;47:591-599. 196. Cunha BA. Clinical features of legionnaires disease. Semin Respir Infect. 1998;13:116-127. 197. Nicotra B, Rivera M, Luman I, et al. Branhamella catarrhalis as a lower respiratory tract pathogen in patients with chronic lung disease. Arch Intern Med. 1986;146:890-893. 198. Donowitz GR, Cox HL. Bacterial community-acquired pneumonia in the elderly. Clin Geriatr Med. 2007;3:515-534. 199. Koivula I, Stn M, Mkel PH. Prognosis after communityacquired pneumonia in the elderly. Arch Intern Med. 1999;159: 1550-1555. 200. Fernndez-Sab N, Carratal J, Rosn B, et al. Communityacquired pneumonia in very elderly patients: Causative organisms, clinical characteristics, and outcomes. Medicine (Baltimore). 2003;82:159-169. 201. Johnson JC, Jayadevappa R, Baccash PD, et al. Nonspecific presentation of pneumonia in hospitalized older people: Age effect or dementia? J Am Geriatr Soc. 2000;48:1316-1320. 202. Berk SL, Holtsclaw SA, Wiener SL, et al. Nontypeable Haemophilus influenzae in the elderly. Arch Intern Med. 1982;142: 532-539. 203. Valenti WM, Trudell RG, Bentley DW. Factors predisposing to oropharyngeal colonization with gram-negative bacilli in the aged. N Engl J Med. 1978;298:1108-1111. 204. Kidd D, Lawson J, Nesbitt R, et al. The natural history and clinical consequences of aspiration in acute stroke. Q J Med. 1995;88:409-413. 205. Miyashita N, Ouchi K, Kawasaki K, et al. Mycoplasma pneumoniae pneumonia in the elderly. Med Sci Monit. 2008;14:CR387CR391. 206. Miyashita N, Obase Y, Ouchi K, et al. Clinical features of severe Mycoplasma pneumoniae pneumonia in adults admitted to an intensive care unit. J Med Microbiol. 2007;56: 1625-1629. 207. Maruyama T, Niederman Ms, Kobyashi T, et al. A prospective comparison of nursing home acquired pneumonia with hospital-acquired pneumonia in non-intubated elderly. Respir Med. 2008;102:1287-1295. 208. Falsey AR. Community-acquired viral pneumonia. Clin Geriatr Med. 2007;23:535-552. 209. Falsey AR, Walsh E. Viral pneumonia in older adults. Clin Infect Dis. 2006;42:518-524. 210. Johnstone J, Majumdar SR, Fox JD, et al. Viral infection in adults hospitalized with community acquired pneumonia: Prevalance, pathogens, and presentation. Chest. 2008;134:11411148. 211. Jennings LC, Anderson TP, Beynon KA, et al. Incidence and characteristics of viral community-acquired pneumonia in adults. Thorax. 2008;63:42-48. 212. Hohenthal U, Vainionp R, Nikoskelainen J, et al. The role of rhinoviruses in community acquired pneumonia in adults. Thorax. 2008;63:658-659. 213. Mylotte JM. Nursing home-acquired pneumonia. Clin Infect Dis. 2002;35:1205-1211. 214. Tsan L, Davis C, Langberg R, et al. Prevalance of nursing home infections in the department of veterans affairs nursing home care units. Am J Infect Control. 2008;36:173-179. 215. Chan Carusone SB, Walter SD, Brazil K, et al. Pneumonia and lower respiratory infections in nursing home residents: predictors of hospitalization and mortality. J Am Geriatr Soc. 2007;55:414-419. 216. Quagliarello V, Ginter S, Han L, et al. Modifiable risk factors for nursing home-acquired pneumonia. Clin Infect Dis. 2005;40: 1-6. 217. Hicks LA, Shepard CW, Britz PH, et al. Two outbreaks of severe respiratory disease in nursing homes associated with rhinovirus. J Am Geriatr Soc. 2006;54:284-289. 218. Davis LJ, Fei M, Huang L. Respiratory infection complicating HIV infection. Curr Opin Infect Dis. 2008;21:184-190. 219. Baril L, Astagneau P, Nguyen J, et al. Pyogenic bacterial pneumonia in human immunodeficiency virus-infected inpatients: A clinical, radiological, microbiological, and epidemiological study. Clin Infect Dis. 1998;26:964-971. 220. Pulmonary Complications of HIV Infection Study Group. Design of a prospective study of the pulmonary complications of human immunodeficiency virus infection. J Clin Epidemiol. 1993;46:497-507.

221. Cordero E, Pachon J, Rivero A, et al. Community-acquired bacterial pneumonia in human immunodeficiency virus-infected patients. Am J Respir Crit Care Med. 2000;162:2063-2068. 222. Madeddu G, Porqueddu FA, Cambosu F, et al. Bacterial community-acquired pneumonia in HIV-infected inpatients in the highly active antiretroviral therapy era. Infection. 2008;36:231236. 223. Klein Mb, Lu Y, Delbalso L, et al. Influenzavirus infection is a primary cause of febrile respiratory illness in HIV-infected adults, despite vaccinations. Clin Infect Dis. 2007;45:234240. 224. Pachon J, Prados D, Capote F, et al. Severe community-acquired pneumonia: Etiology, prognosis, treatment. Am Rev Respir Dis. 1990;142:369-373. 225. Fine MJ, Auble TE, Yealy DM, et al. A prediction rule to identify low-risk patients with community-acquired pneumonia. N Engl J Med. 1997;336:243-250. 226. Buising KL, Thursky KA, Black JF, et al. A prospective comparison of severity scores for identifying patients with severe community acquired pneumonia: Reconsidering what is meant by severe pneumonia. Thorax. 2006;61:419-424. 227. Neill AM, Martin IR, Weir R, et al. Community acquired pneumonia: aetiology and usefulness of severity criteria on admission. Thorax. 1996;51:1010-1016. 228. Lim WS, van der Eerden MM, Laing R, et al. Defining community acquired pneumonia severity on presentation to hospital: An international derivation and validation study. Thorax. 2003;58:377-382. 229. Capelastegui A, Espana PP, Quintana JM, et al. Validation of a predictive rule for the management of community-acquired pneumonia. Eur Respir J. 2006;27:151-157. 230. Charles PG, Wolfe R, Whitby M, et al. SMART-COP: A tool for predicting the need for intensive respiratory or vasopressor support in community-acquired pneumonia. Clin Infect Dis. 2008;47:375-384. 231. The British Thoracic Society Research Committee and the Public Health Laboratory Service. Community-acquired pneumonia in adults in British hospitals in 1982-1983: A survey of aetiology, mortality, prognostic factors and outcome. Q J Med. 1987;62:195-220. 232. Fine MJ, Orloff JJ, Arisumi D, et al. Prognosis of patients hospitalized with community-acquired pneumonia. Am J Med. 1990;88:S1N-S8N. 233. Leroy O, Santre L, Beuscart L. A five-year old study of severe community-acquired pneumonia with emphasis on prognosis in patients admitted to an intensive care unit. Intensive Care Med. 1995;21:24-31. 234. Lisboa T, Blot S, Waterer GW, et al. Radiological progression of pulmonary infiltrates predicts a worse prognosis in severe community-acquired pneumonia than bacteremia. Chest. 2009;135: 165-172. 235. Reimann HA. An acute infection of the respiratory tract with atypical pneumonia. JAMA. 1938;111:2377-2384. 236. File TM, Tan JS, Plouffe JF. The role of atypical pathogens: Mycoplasma pneumoniae, Chlamydia pneumoniae and Legionella pneumophila in respiratory infection. Infect Dis Clin North Am. 1998;12:509-592. 237. Loens K, Beck T, Ursi D, et al. Evaluation of different nucleic acid amplification techniques for the detection of M. pneumoniae, C. pneumoniae and Legionella spp. in respiratory specimens from patients with community-acquired pneumonia. J Microbiol Methods. 2008;73:257-262. 238. Loens K, Ursi D, Goossen H, et al. Evaluation of the NucliSens miniMAG RNA extraction and real-time NASBA applications for the detection of Mycoplasma pneumoniae and Chlamydophila pneumoniae in throat swabs. J Microbiol Methods. 2008;72:217219. 239. Grayston JT, Diwan VK, Cooney M, et al. Community and hospital acquired pneumonia associated with chlamydia TWAR infection demonstrated serologically. Arch Intern Med. 1989; 149:169-173. 240. Thorn DH, Grayston JT. Infections with Chlamydia pneumoniae strain TWAR. Clin Chest Med. 1991;12:245-256. 241. Blasi F, Tarsia P, Aliberti S, et al. Chlamydia pneumoniae and Mycoplasma pneumoniae. Semin Respir Crit Care Med. 2005;26:617-624. 242. Vila-Corcoles A, Ochoa-Gondar O, Rodriguez-Blanco T, et al. Epidemiology of community-acquired pneumonia in older adults: A population-based study. Respir Med. 2008.08.06. 243. Komaroff AL, Aronson MD, Schachter J. Chlamydia trachomatis infections in adults with community acquired pneumonia. JAMA. 1981;245:1319-1322. 244. Johnstone J, Majumdar SR, Fox JD, et al. Human metapneumovirus pneumonia in adults: results of a prospective study. Clin Infect Dis. 2008;46:571-574. 245. Hamelin ME, Cotes S, Laforge J, et al. Human metapneumovirus infection in adults with community-acquired pneumonia and exacerbation of chronic obstructive pulmonary disease. Clin Infect Dis. 2005;41:498-502. 246. Wenzel RP, McCormick DP, Beam WE Jr. Parainfluenza pneumonia in adults. JAMA. 1972;221:294-295. 247. Falsey AR, Cunningham CK, Baber WH, et al. Respiratory syncytial virus and influenza A infections in the hospitalized elderly. J Infect Dis. 1995;172:389-394. 248. Englund JA, Sullivan CJ, Jordan MC, et al. Respiratory syncytial virus infections in immunocompromised adults. Ann Intern Med. 1988;109:203-208.

916

PART II

Major Clinical Syndromes

249. Serra MC, Cervera C, Pumarola T, et al. Virological diagnosis in community-acquired pneumonia in immunocompromised patients. Eur Respir J. 2008;31:618-624. 250. Rello J, Quintana E, Ausiner V, et al. A three-year study of severe community-acquired pneumonia with emphasis on outcome. Chest. 1993;103:232-235. 251. Leeper KV Jr, Torres A. Community-acquired pneumonia in the intensive care unit. Clin Chest Med. 1995;16:155-171. 252. Stout JE, Yu VL. Legionellosis. N Engl J Med. 1997;337:682687. 253. Nakajoh K, Nakagawa T, Sekizawa K, et al. Relation between incidence of pneumonia and protective reflexes in post-stroke patients with oral or tube feeding. J Intern Med. 2000;247:3942. 254. Steinhagen V, Grossman A, Benecke R, et al. Swallowing disturbance pattern relates to brain lesion location in acute stroke patients. Stroke. 2009;40:1903-1906. 255. Matthay MA, Rosen GD. Acid aspiration induced lung injury: New insights and therapeutic options. Am J Respir Crit Care Med. 1996;154:277-278. 256. Robinovici R, Neville LF, Abdullah F, et al. Aspiration-induced lung injury: Role of complement. Crit Care Med. 1995;23:14051411. 257. Bartlett JG, Gorbach SL. The triple threat of aspiration pneumonia. Chest. 1979;68:560-566. 258. Marik PE, Careau P. The role of anaerobes in patients with ventilator-associated pneumonia and aspiration pneumonia. Chest. 1999;115:178-183. 259. Hammond JMJ, Potgieter PD, Hanslo D, et al. The etiology and antimicrobial susceptibility patterns of microorganisms in acute community-acquired lung abscess. Chest. 1995;108:937-941. 260. Lorber B, Swenson RM. Bacteriology of aspiration pneumonia: A prospective study of community and hospital-acquired cases. Ann Intern Med. 1974;81:329-331. 261. Katz, U, Shoenfeld Y. Pulmonary eosinophilia. Clinic Rev Allergy Immunol. 2008;34:367-371. 262. Schatz M, Wasserman S, Patterson R. Eosinophils and immunologic lung disease. Med Clin North Am. 1981;65:1055-1071. 263. Haley RW, Culver DH, White JW, et al. The nationwide nosocomial infection rate: A new need for vital statistics. Am J Epidemiol. 1985;121:159-167. 264. Wenzel RP. Hospital-acquired pneumonia: An overview of the current state of the art for prevention and control. Eur J Clin Microbiol Infect Dis. 1989;8:56-60. 265. George DL. Epidemiology of nosocomial pneumonia in intensive care unit patients. Clin Chest Med. 1995;16:29-44. 266. Donowitz LG, Page MC, Mileur BL, et al. Alteration of normal gastric flora in critical care patients receiving antacid and cimetidine therapy. Infect Control. 1986;7:23-26. 267. Simms HH, DeMoria E, McDonald L, et al. Role of gastric colonization in the development of pneumonia in critically ill trauma patients: Results of a prospective randomized trail. J Trauma. 1991;31:531-537. 268. Bonten MJM, Gaillard CA, deLeeuw PW, et al. Role of colonization of the upper intestinal tract in the pathogenesis of ventilator-associated pneumonia. Clin Infect Dis. 1997;24:309-319. 269. Rello J, Ausina V, Castella J, et al. Incidence, etiology and outcome of nosocomial pneumonia in mechanically ventilated patients. Chest. 1991;100:439-444. 270. Septimus EJ. Nosocomial bacterial pneumonia. Semin Respir Infect. 1989;4:245-252. 271. Cunha BA. Pneumonia in the compromised host. Infect Dis Clin North Am. 2001;15:591-612. 272. Niederman MS, Peters SP. Update in pulmonary medicine. Ann Intern Med. 1998;128:208-215. 273. Black ER, Mushlin AI, Griner PF, et al. Predicting the need for hospitalization of ambulatory patients with pneumonia. J Gen Intern Med. 1991;6:394-400. 274. Fine MJ, Smith DN, Singer DE. Hospitalization decision on patients with community-acquired pneumonia: A prospective cohort study. Am J Med. 1990;89:713-721. 275. Marrie TJ, Lau CY, Wheeler SL, et al. A controlled trial of a critical pathway for treatment of community-acquired pneumonia. JAMA. 2000;283:749-755. 276. Atlas SJ, Benzer TI, Borowsky LH, et al. Safely increasing the proportion of patients with community-acquired pneumonia treated as outpatients: An interventional trial. Arch Intern Med. 1998;158:1350-1356. 277. Zuberi FF, Khan JA. Prospective comparison of prediction rules of mortality risk for CAP in a developing country. Int J Tuberc Lung Dis. 2008;12:447-452. 278. Ananda-Rajah MR, Charles PGP, Melvani S, et al. Comparing the pneumonia severity index with CURB-65 in patients admitted with community acquired pneumonia. Scand J Inf Dis. 2008;40:293-300. 279. Seymann G, Barger K, Choo S, et al. Clinical judgment versus the pneumonia severity index in making the admission decision. J Emerg Med. 2007;34:261-268. 280. Schuetz P, Koller M, Christ-Crain E, et al. Predicting mortality with pneumonia severity scores: Importance of model

recalibration to local setting. Epidemiol Infect. 2008;136:1628-1637. 281. Mandell LA, Barlett JG, Dowell SF, et al. Update of practice guidelines for the management of community-acquired pneumonia in immunocompetent adults. Clin Infect Dis. 2003;37: 1405-1433. 282. Porath A, Schlaeffer F, Lieberman D. Appropriateness of hospitalization of patients with community-acquired pneumonia. Ann Emerg Med. 1996;27:176-183. 283. Meehan TP, Fine MJ, Krumholz HM, et al. Quality care, process, and outcomes in elderly patients with pneumonia. JAMA. 1997; 278:2080-2084. 284. Houck PM, Bratzler DW, Nsa W, et al. Timing of antibiotic administration and outcomes for Medicare patients hospitalized with community-acquired pneumonia. Arch Intern Med. 2004; 164:637-644. 285. Wachter RM, Flanders SA, Fee C, et al. Public reporting of antibiotic timing in patients with pneumonia: Lessons from a flawed performance measure. Ann Intern Med. 2008;149:29-32. 286. Tu KT, Wyer PC. Evidence behind the 4-hour rule for initiation of antibiotic therapy in community acquired pneumonia. Ann Emerg Med. 2008;20:1-12. 287. Welker JA, Huston M, McCue JD. Antibiotic timing and errors in diagnosing pneumonia. Arch Intern Med. 2008;168:351356. 288. Kanwar M, Brar N, Khatib R, et al. Misdiagnosis of communityacquired pneumonia and inappropriate utilization of antibiotics. Chest. 2007;131(6):1865-1867. 289. Polgreen PH, Chen YY, Cavenaugh JE, et al. An outbreak of severe Clostridium difficile-associated disease possibly related to inappropriate antimicrobial therapy for community-acquired pneumonia. Infect Control Hosp Epidemiol. 2007;28:212-214. 290. Klugman PK. Pneumococcal resistance to antibiotics. Clin Microbiol Rev. 1990;3:171-176. 291. Spika JS, Facklam RR, Rlikaytis BD, et al. Antimicrobial resistance to Streptococcus pneumoniae in the United States: 19791987. J Infect Dis. 1991;163:1273-1278. 292. Doern GV, Heilmann KP, Huynh HK, et al. Antimicrobial resistance among clinical isolates of Streptococcus pneumoniae in the United States during 1999-2000 including a comparison of resistance rates since 1994-1995. Antimicrob Agents Chemother. 2001;45:1721-1729. 293. Karlowsky JA, Thornsberry C, Jones ME, et al. Factors associated with relative rates of antimicrobial resistance among Streptococcus pneumoniae in the United States: Results from the TRUST Surveillance Program (1998-2002). Clin Infect Dis. 2003;36:963-970. 294. Clinical and Laboratory Standards Institute. Performance Standards for Antimicrobial Susceptibility Testing; Eighteenth International Supplement. Wayne, Pa: Clinical and Laboratory Standards Institute; 2008. 295. Jacobs MR. Clinical significance of antimicrobial resistance in Streptococcus pneumoniae. S Afr Med J. 2007;97:1133-1140. 296. Vanderkooi OG, Low DE, Green K, et al. Predicting antimicrobial resistance in invasive pneumococcal infections. Clin Infect Dis. 2005;40:1288-1297 297. Lonks JR, Garau J, Gomez L, et al. Failure of macrolide antibiotic treatment in patients with bacteremia due to erythromycinresistant Streptococcus pneumoniae. Clin Infect Dis. 2002;35: 556-564. 298. Daneman N, McGeer A, Green K, et al. Macrolide resistance in bacteremic pneumococcal disease: Implications for patient management. Clin Infect Dis. 2006;46:432-438. 299. Kupronis BA, Richards CL, Whitney CG, et al. Invasive pneumococcal disease in older adults residing in long-term care facilities and in the community. J Am Geriatr Soc. 2003;51: 1520-1525. 300. Davidson R, Cavalconti R, Brunton JL, et al. Resistance to levofloxacin and failure of treatment of pneumococcal pneumonia. N Engl J Med. 2002;346:747-750. 301. Waterer GW, Somes GW, Wunderink RG. Monotherapy may be suboptimal for severe bacteremic pneumococcal pneumonia. Arch Intern Med. 2001;161:1837-1842. 302. Gleason PP, Meehan TP, Fine JM, et al. Associations between initial antimicrobial therapy and medical outcomes for hospitalized elderly patients with pneumonia. Arch Intern Med. 1999;159:2562-2572. 303. Stahl JE, Barza M, DesJardin J, et al. Effect of macrolides as part of initial empiric therapy on length of stay in patients hospitalized with community-acquired pneumonia. Arch Intern Med. 1999;159:2576-2580. 304. Paul M, Nielson AD, Gafter-Gvili A, et al. The need for macrolides in hospitalised community-acquired pneumonia; Propensity analysis. Eur Respir J. 2007;30:525-531. 305. Dwyer R, Ortqvist A, Aufwerber E, et al. Addition of a macrolide to a -lactam in bacteremic pneumococcal pneumonia. Eur J Clin Microbiol Infect Dis. 2006;25:518-521. 306. Metersky ML, Ma A, Houck PM, et al. Antibiotics for bacteremic pneumonia: improved outcomes with macrolides but not fluoroquinolones. Chest. 2007;131:446-473.

307. Amsden GW. Anti-inflammatory effects of macrolidesan underappreciated benefit in the treatment of communityacquired respiratory tract infections and chronic inflammatory pulmonary conditions. J Antimicrob Chemother. 2005;55:1021. 308. Daneman N, Low DE, McGeer A, et al. At the threshold: Defining clinically meaningful resistance thresholds for antibiotic choice in community-acquired pneumonia. Clin Infect Dis. 2008;46:1131-1138. 309. Craig WA, Andes D. Aminoglycosides are useful for severe respiratory tract infections. Semin Respir Infect. 1997;12:271277. 310. Schentag JJ, Birmingham MC, Paladins JA, et al. Nosocomial pneumonia, optimizing antibiotics other than aminoglycosides is a more important determinant of successful clinical outcome and a better means of avoiding resistance. Semin Respir Infect. 1997;12:278-293. 311. Arancibia F, Bauer TT, Ewig S, et al. Community-acquired pneumonia due to gram-negative bacteria and Pseudomonas aeruginosa: Incidence, risk, and prognosis. Arch Intern Med. 2002;162:1849-1858. 312. Bratzler DW, Ma H, Nsa, W. Initial antibiotic selection and patient outcomes: Observations from the National Pneumonia Project. Clin Infect Dis. 2008;47:S193-S201. 313. Brown RB, Iannini P, Gross P, et al. Impact of initial antibiotic choice on clinical outcomes in community-acquired pneumonia. Chest. 2003;123:1503-1511. 314. Rice, LB. The Maxwell Finland Lecture: For the duration-rational antibiotic administration in the era of antimicrobial resistance and Clostridium difficile. Clin Inf Dis. 2008;46:491-496. 315. Halm EA, Fine MJ, Marrie TJ, et al. Time to clinical stability in patients hospitalized with community-acquired pneumonia: Implication for practice guidelines. JAMA. 1998;279:14521457. 316. Li JZ, Winston LG, Moore DH, et al. Efficacy of short-course antibiotic regimens for community-acquired pneumonia: a meta-analysis. Am J Med. 2007;120:783-790. 317. el Moussaoui R, de Borgie CAJM, van den Broek P, et al. Effectiveness of discontinuing antibiotic treatment after three days versus eight days in mild to moderate-severe community acquired pneumonia: Randomised, double blind study. BMJ. 2006;332:1355. 318. Castro-Guardiola A, Viejo-Rodrguez AL, Soler-Simon S, et al. Efficacy and safety of oral and early-switch therapy for community-acquired pneumonia: A randomized controlled trial. Am J Med. 2001;111:367-374. 319. Ramirez JA. Managing antiinfective therapy of communityacquired pneumonia in the hospital setting: Focus on switch therapy. Pharmacotherapy. 2001;21:79S-82S. 320. Hitt CM, Nightingale CH, Quintiliani R, et al. Streamlining antimicrobial therapy for lower respiratory tract infections. Clin Infect Dis. 1997;24:S231-5237. 321. Beaumont M, Shuster MG. Is an observation period necessary after intravenous antibiotics are changed to oral administration? Am J Med. 1999;106:114-116. 322. Halm EA, Fine MJ, Kapoor WN, et al. Instability on hospital discharge and the risk of adverse outcomes in patients with pneumonia. Arch Intern Med. 2002;162:1278-1284. 323. Annane D, Meduri GU, Marik P. Critical illness-related corticosteroid insufficiency and community-acquired pneumonia: Back to the future! Eur Respir J. 2008;31:1150-1152. 324. Confalonieri M, Urbino R, Potena A, et al. Hydrocortisone infusion for severe community-acquired pneumonia. Am J Respir Crit Care Med. 2005;171:242-248. 325. Sibila O, Agusti C, Torres A. Corticosteroids in severe pneumonia. Eur Respir J. 2008;32:259-264. 326. Thomsen RW, Riss A, Kornum JB, et al. Preadmission use of statins and outcomes after hospitalization with pneumonia. Arch Intern Med. 2008;168:2081-2087. 327. Mortensen EM, Pugh MJ, Copeland LA, et al. Impact of statins and angiotensin-converting enzyme inhibitors on mortality of subjects hospitalised with pneumonia. Eur Respir J. 2008;31: 611-617. 328. Siempos II, Vardakas Z, Kopterides P, et al. Adjunctive therapies for community-acquired pneumonia: A systematic review. J Antimicrob Chemother. 2008;62:661-668. 329. Gross PA, Hermogenes AW, Sacks HS, et al. The efficacy of influenza vaccine in elderly patients: A meta-analysis and review of the literature. Ann Intern Med. 1995;123:518-527. 330. Nichol KL, Nordin J, Mullooly J, et al. Influenza vaccination and reduction in hospitalizations for cardiac disease and stroke among the elderly. N Engl J Med. 2003;348:1322-1332. 331. Advisory Committee on Immunization Practices (ACIP) on the use of influenza vaccine and antiviral agents. MMWR Morb Mortal Wkly Rep. 2003;52:1-44. 332. Jackson LA, Neuzil KM, Yu O, et al. Effectiveness of pneumococcal polysaccharide vaccine in older adults. N Engl J Med. 2003;348:1747-1755.