Diagnosis and Treatment of Community-Acquired Pneumonia M. NAWAL LUTFIYYA, PH.D., ERIC HENLEY, M.D., M.P.H., and LINDA F. CHANG, PHARM.D., M.P.H., B.C.P.S. University of Illinois College of Medicine at Rockford, Rockford, Illinois STEPHANIE WESSEL REYBURN, M.D., M.P.H., Mayo School of Graduate Medical Education, Rochester, Minnesota Patients with community-acquired pneumonia often present with cough, fever, chills, fatigue, dyspnea, rigors, and pleuritic chest pain. When a patient presents with suspected community-acquired pneu- monia, the physician should first assess the need for hospitalization using a mortality prediction tool, such as the Pneumonia Severity Index, combined with clinical judgment. Consensus guidelines from several organizations recommend empiric therapy with macrolides, fluoroquinolones, or doxycycline. Patients who are hospitalized should be switched from parenteral antibiotics to oral antibiotics after their symptoms improve, they are afebrile, and they are able to tolerate oral medications. Clinical pathways are important tools to improve care and maximize cost-effectiveness in hospitalized patients. (Am Fam Physician 2006;73:442-50. Copyright © 2006 American Academy of Family Physicians.) ILLUSTRATION BY MARK E. SCHULER Members of various ommunity-acquired pneumonia rates from CAP increase with the presence family medicine depart- (CAP) is defined as pneumonia of comorbidity and increased age; the condi- ments develop articles for “Practical Therapeutics.” not acquired in a hospital or a tion affects persons of any race or sex equally. This article is one in a long-term care facility. Despite the The decrease in death rates from pneumonia series coordinated by the C availability of potent new antimicrobials and and influenza are largely attributed to vac- Department of Family 1 Medicine at the University effective vaccines, an estimated 5.6 million cines for vulnerable populations (e.g., older of Illinois at Chicago cases of CAP occur annually in the United and immunocompromised persons). College of Medicine, States.2 The estimated total annual cost of Chicago, Ill. Guest editor health care for CAP in the United States is Clinical Presentation of the series is Eric Henley, 2 M.D., M.P.H. $8.4 billion. Table 1 presents an overview of Pneumonia is an inflammation or infection CAP including definition, signs and symp- of the lungs that causes them to function toms, etiology, and risk factors. abnormally. Pneumonia can be classified as typical or atypical, although the clini- Epidemiology cal presentations are often similar. Several The epidemiology of CAP is unclear because symptoms commonly present in patients few population-based statistics on the con- with pneumonia. dition alone are available. The Centers for Disease Control and Prevention (CDC) com- TYpeS OF cAP bines pneumonia with influenza when col- Typical pneumonia usually is caused by lecting data on morbidity and mortality, bacteria such as Streptococcus pneumoniae. although they do not combine them when Atypical pneumonia usually is caused by the collecting hospital discharge data. In 2001, influenza virus, mycoplasma, chlamydia, influenza and pneumonia combined were legionella, adenovirus, or other unidenti- the seventh leading causes of death in the fied microorganism. The patient’s age is the United States,3,4 down from sixth in previous main differentiating factor between typical years, and represented an age-adjusted death and atypical pneumonia; young adults are rate of 21.8 per 100,000 patients.3 Death more prone to atypical causes,5,6 and very Downloaded from the American Family Physician Web site at www.aafp.org/afp. Copyright© 2006 American Academy of Family Physicians. For the private, noncommercial use of one individual user of the Web site. All other rights reserved. Contact [email protected] for copyright questions and/or permission requests. SORT: KEY RECOMMENDATIONS FOR PRACTICE Evidence Clinical recommendation rating References Patients with suspected community-acquired pneumonia (CAP) should C 8 receive chest radiography. The Pneumonia Severity Index should be used to assist in decisions A 8, 9, 15, 16 regarding hospitalization of patients with CAP. The initial treatment of CAP is empiric, and macrolides or doxycycline C 8, 9, 29 (Vibramycin) should be used in most patients. Respiratory fluoroquinolones should be used when patients have failed C 8, 9, 28, 29 first-line regimens, have significant comorbidities, have had recent antibiotic therapy, are allergic to alternative agents, or have a documented infection with highly drug-resistant pneumococci. A = consistent, good-quality patient-oriented evidence; B = inconsistent or limited-quality patient-oriented evi- dence; C = consensus, disease-oriented evidence, usual practice, expert opinion, or case series. For information about the SORT evidence rating system, see page 374 or http://www.aafp.org/afpsort.xml. young and older persons are more predis- sentations may include headache and myalgia. posed to typical causes. Certain etiologies, such as legionella, also may produce gastrointestinal symptoms. SYMPTOMS Common clinical symptoms of CAP include Diagnosis cough, fever, chills, fatigue, dyspnea, rigors, PHYSICAL eXAMINATION and pleuritic chest pain. Depending on the Physical examination may reveal dullness pathogen, a patient’s cough may be persistent to percussion of the chest, crackles or rales and dry, or it may produce sputum. Other pre- on auscultation, bronchial breath sounds, table 1 Overview of Community-Acquired Pneumonia Definition Etiology Risk factors Lower respiratory tract infection in a Bacterial Age older than 65 years nonhospitalized person that is associated with Chlamydia species Human immunodeficiency virus or symptoms of acute infection with or without Haemophilus influenzae immunocompromised new infiltrate on chest radiographs Legionella species Recent antibiotic therapy or resistance Clinical presentation Moraxella catarrhalis to antibiotics Temperature greater than 38˚C (100.4˚F) Mycoplasma pneumoniae Comorbidities Cough with or without sputum, hemoptysis Staphylococcus aureus Asthma Pleuritic chest pain Streptococcus pneumoniae Cerebrovascular disease Myalgia Viral Chronic obstructive pulmonary Gastrointestinal symptoms disease Adenovirus Dyspnea Chronic renal failure Influenza A and B Malaise, fatigue Congestive heart failure Parainfluenza Rales, rhonchi, wheezing Diabetes Respiratory syncytial virus Egophony, bronchial breath sounds Liver disease Endemic fungi Dullness to percussion Neoplastic disease Blastomycosis Atypical symptoms in older patients Coccidioidomycosis Histoplasmosis February 1, 2006 ◆ Volume 73, Number 3 www.aafp.org/afp American Family Physician 443 Community-Acquired Pneumonia tactile fremitus, and egophony (“E” to “A” the latest American Thoracic Society (ATS) changes). The patient also may be tachy- guidelines for the diagnosis and treatment of pneic. A prospective study7 showed that adults with CAP, “all patients with suspected patients with typical pneumonia were more CAP should have a chest radiograph to estab- likely than not to present with dyspnea and lish the diagnosis and identify complications bronchial breath sounds on auscultation. (pleural effusion, multilobar disease).”8 Chest radiography may reveal a lobar consolidation, RADIOGRAPHY which is common in typical pneumonia; or Chest radiography (posteroanterior and lateral it could show bilateral, more diffuse infil- views) has been shown to be a critical compo- trates than those commonly seen in atypi- nent in diagnosing pneumonia.8 According to cal pneumonia. However, chest radiography performed early in the course of the disease could be negative. table 2 Sensitivity and Specificity of Diagnostic Tests for CAP LABORATORY TESTS Historically, common laboratory tests for Diagnostic tests by pathogen Sensitivity (%) Specificity (%) pneumonia have included leukocyte count, Chlamydia sputum Gram stain, two sets of blood cul- Rapid PCR (sputum, BAL fluid) 30 to 95 > 95 tures, and urine antigens. However, the Serology (fourfold rise in serum 10 to 100 — validity of these tests has recently been and convalescent titers) questioned after low positive culture rates Sputum culture 10 to 80 > 95 were found (e.g., culture isolates of S. pneu- Gram-negative rods moniae were present in only 40 to 50 percent 9 Sputum Gram stain 15 to 100 11 to 100 of cases). Such low positive culture rates Haemophilus influenzae, are likely due to problems with retrieving Moraxella catarrhalis, samples from the lower respiratory tract, Pneumoniae previous administration of antibiotics, con- Sputum culture Diagnostic yield Diagnostic yield tamination from the upper airways, faulty 20 to 79* 20 to 79* separation of sputum from saliva when Influenza streaking slides or plates,9 or viral etiology. Rapid DFA (sputum, BAL fluid) 22 to 75 90 Furthermore, sputum samples are adequate Legionella pneumophila in only 52.3 percent of patients with CAP, DFA (sputum, BAL fluid) 22 to 75 90 and only 44 percent of those samples contain PCR (sputum, BAL fluid) 83 to 100 > 95 pathogens.10 Nonetheless, initial therapy Serum acute titer 10 to 27 > 85 often is guided by the assumption that the Urinary antigen 55 to 90 > 95 presenting disease is caused by a common Mycoplasma pneumoniae bacterial pathogen. Antibiotic titers 75 to 95 > 90 Findings11 also cast doubt on the clinical Cold agglutinins 50 to 60 — utility of obtaining blood cultures from PCR (sputum, BAL fluid) 30 to 95 > 95 patients with suspected