Browse the corpus

Community-Acquired Pneumonia TABLE 8. Risk Factors for Pathogens Causing Community-Acquired Pneumonia Risk Factor Common Pathogens Heavy alcohol use Streptococcus pneu moniae, Haemo phil us nflue nzae, Enterobacteriaceae, i Kebsiella pneumoniae COPD H. influenzae, S. pneumoniae, Moraxella catarrhalis, Legionella pneumophila, Pse ud o mon as ae ru gi nosa Structu ra I I u n g d isease ( bronch iectasi s, cystic fi brosis) P ae ru gi no sa, Bu rkho I d e ri a ce p a ci a, Ste notro pho mo n as m alto phi lia, Sta p hy I o co ccu s a u re u s, atyp ica I myco ba cte ri a Aspiration (seizures, neurologic impairment, loss of S. p ne u moni ae, H. i nfl u e nzae, Enterobacteriaceae consciousness) Age >65 years lnfluenza virus, S. pneumoniae, rhinovirus Post-viralillness S. aureus, Streptococcus pyogenes, S. pneumoniae Animalexposure Birds Ch I a myd o p h i I a p siua ci, H i sto p I a s m a ca p s u I atu m, C y pto coccus n e of o r m a n s Dogs Bo rdetel I a bronchiseptica Cats Pasteurella multocida Farm animals or domesticated pregnant animals Coxiella burnetii, Brucel I a species Horses Rhodococcus equi Rodent droppings Hantavirus Rabbits Fra n ci sel I a tu I a re n si s

Farm animals or domesticated pregnant animals Coxiella burnetii, Brucel I a species Horses Rhodococcus equi Rodent droppings Hantavirus Rabbits Fra n ci sel I a tu I a re n si s Hot tu b exposu re Leg ionell a pne u m o ni ae, atypica I mycobacteria (ca usi ng a hype rsensitivity pneumonitis) Geographic Eastern United States H i sto p I a s m a c a p s u I atu m, B I a sto m y ce s d e r m atiti d i s Southwest United States Coccidioides Southeast Asia Burkholderia pseudomal/ei, SARS, avian influenza Middle East MERS Late fall, winter, early spring (Western hemisphere) I nfl uenza vi ru s, pa ra i nfl uenza vi rus, rh i novi rus MERS = Middle East respiratory syndrome; SARS = severe acute respiratory syndrome.

Southwest United States Coccidioides Southeast Asia Burkholderia pseudomal/ei, SARS, avian influenza Middle East MERS Late fall, winter, early spring (Western hemisphere) I nfl uenza vi ru s, pa ra i nfl uenza vi rus, rh i novi rus MERS = Middle East respiratory syndrome; SARS = severe acute respiratory syndrome. Streptococcus pAogenes. Despite extensive laboratory investi- aspiration pneumonia include dysphagia, decreased con- gation, no causative organism was identified in 62'7, of patients sciousness (alcohol or substance use, seizures, stroke), poor in the CDC-EPIC trial. dentition, gastroesophageal reflux, and vomiting. Anaerobes Atypical pneumonia refers to CAP caused by organisms not remain important pathogens in patients with lung abscess or culturable on standard bacterial media, including viruses and empyema. Zoonotic causes of CAP include Coxiellabumetiiand fastidious bacteria such as Legionella species, Mycoplasma Francisellatularensis. Mycobacterial or fungal causes should be pneumoniae, and Chlamydophila pneumoniae. Respiratory considered in patients with immunocompromising conditions, viruses account for nearly all viral pneumonias, but less com- epidemiologic risk factors for infection (such as incarceration, mon pathogens include varicella zoster virus or Hantavirus. certain hobbies or occupations, and pertinent regional or for Legionella pneumophilo is a recognized cause of severe disease. eign travel), or subacute presentations and in those who do not Legionellosis is associated with water exposure, including hot respond to conventional antibacterial treatment. tubs and air conditioning units; however, infection may occur xEY POttrS without an obvious source. A history of recent travel has been reported in approximately 10'1, of Legionella cases reported to o The definition of mmmunity-acquired pneumonia has the CDC. expandedto include some patients previously categorized as CAP caused by anaerobic bacteria is uncommon. having health care-associated pneumonia because the Historically, anaerobes have been thought to be important path- microbiologz and treatment of patients in long-term care ogens in aspiration pneumonia. Recent studies have questioned facilities or who were hospitalized in the preceding 3 months this, with S. pneumonioe and Haemophilus inJluenzoe isolated do not differ substantially ffom that of community-dwelling much more commonly than anaerobes in patients with patients with similar comorbidities. (Continued) community-acquired aspiration pneumonia. Risk factors for

Streptococcus pAogenes. Despite extensive laboratory investi- aspiration pneumonia include dysphagia, decreased con- gation, no causative organism was identified in 62'7, of patients sciousness (alcohol or substance use, seizures, stroke), poor in the CDC-EPIC trial. dentition, gastroesophageal reflux, and vomiting. Anaerobes Atypical pneumonia refers to CAP caused by organisms not remain important pathogens in patients with lung abscess or culturable on standard bacterial media, including viruses and empyema. Zoonotic causes of CAP include Coxiellabumetiiand fastidious bacteria such as Legionella species, Mycoplasma Francisellatularensis. Mycobacterial or fungal causes should be pneumoniae, and Chlamydophila pneumoniae. Respiratory considered in patients with immunocompromising conditions, viruses account for nearly all viral pneumonias, but less com- epidemiologic risk factors for infection (such as incarceration, mon pathogens include varicella zoster virus or Hantavirus. certain hobbies or occupations, and pertinent regional or for Legionella pneumophilo is a recognized cause of severe disease. eign travel), or subacute presentations and in those who do not Legionellosis is associated with water exposure, including hot respond to conventional antibacterial treatment. tubs and air conditioning units; however, infection may occur xEY POttrS without an obvious source. A history of recent travel has been reported in approximately 10'1, of Legionella cases reported to o The definition of mmmunity-acquired pneumonia has the CDC. expandedto include some patients previously categorized as CAP caused by anaerobic bacteria is uncommon. having health care-associated pneumonia because the Historically, anaerobes have been thought to be important path- microbiologz and treatment of patients in long-term care ogens in aspiration pneumonia. Recent studies have questioned facilities or who were hospitalized in the preceding 3 months this, with S. pneumonioe and Haemophilus inJluenzoe isolated do not differ substantially ffom that of community-dwelling much more commonly than anaerobes in patients with patients with similar comorbidities. (Continued) community-acquired aspiration pneumonia. Risk factors for 16

Community-Acquired Pneumonia Site of Treatment Patient or Epidemiologic Most Common Organisms Regimens(s) Considerations Outpatient Healthy patient without risk factors Stre ptococcu s p n e u m o n i a e Amoxicillin for MRSA or Pseudomonas aeruginosa Mycoplasma OR Chlamydophila Doxycycline H ae m o ph i I u s i nfl u e nzae OR Respiratory viruses Macrolide (if local pneumococcal resistance <25%) Comorbidities" Same as above Respiratory fluoroq u i noloneb OR

Site of Treatment Patient or Epidemiologic Most Common Organisms Regimens(s) Considerations Outpatient Healthy patient without risk factors Stre ptococcu s p n e u m o n i a e Amoxicillin for MRSA or Pseudomonas aeruginosa Mycoplasma OR Chlamydophila Doxycycline H ae m o ph i I u s i nfl u e nzae OR Respiratory viruses Macrolide (if local pneumococcal resistance <25%) Comorbidities" Same as above Respiratory fluoroq u i noloneb OR Oral p-lactam'plus either a macrolide or doxycycline lnpatient, non-lCU' S. pneumoniae lntravenous B-lactamd plus a macrolide Mycoplasma OR Chlamydophila Respi ratory fl uoroq u i no lone H. influenzae

Oral p-lactam'plus either a macrolide or doxycycline lnpatient, non-lCU' S. pneumoniae lntravenous B-lactamd plus a macrolide Mycoplasma OR Chlamydophila Respi ratory fl uoroq u i no lone H. influenzae Legionella Respiratory viruses ICU treatment S. pneumoniae lntravenous B-lactamd plus a macrolide; if contraindicated, p-lactam Staphylococcus aureus plus respiratory fluoroquinolone is an H. influenzae acceptable alternative Legionella Gram-negative bacilli Any Risk factor for Pseudomonas (see Antipseudomonal p-lactam plus either text) a macrolide or an antipseudomonal quinolone Any Risk factor for CA-MRSA (see text) Standard therapy PLUS vancomycin OR linezolid ATS = American Thoracic Society; CA-MRSA = community-acquired methicillin-resistant Staphylococcus aureus; IDSA = lnfectious Diseases Society of America.

Legionella Respiratory viruses ICU treatment S. pneumoniae lntravenous B-lactamd plus a macrolide; if contraindicated, p-lactam Staphylococcus aureus plus respiratory fluoroquinolone is an H. influenzae acceptable alternative Legionella Gram-negative bacilli Any Risk factor for Pseudomonas (see Antipseudomonal p-lactam plus either text) a macrolide or an antipseudomonal quinolone Any Risk factor for CA-MRSA (see text) Standard therapy PLUS vancomycin OR linezolid ATS = American Thoracic Society; CA-MRSA = community-acquired methicillin-resistant Staphylococcus aureus; IDSA = lnfectious Diseases Society of America. "Comorbidities include chronic heart, lung, liver, or kidney disease; diabetes mellitus; alcoholism; asplenia; malignancies; and immunosuppression.

Legionella Respiratory viruses ICU treatment S. pneumoniae lntravenous B-lactamd plus a macrolide; if contraindicated, p-lactam Staphylococcus aureus plus respiratory fluoroquinolone is an H. influenzae acceptable alternative Legionella Gram-negative bacilli Any Risk factor for Pseudomonas (see Antipseudomonal p-lactam plus either text) a macrolide or an antipseudomonal quinolone Any Risk factor for CA-MRSA (see text) Standard therapy PLUS vancomycin OR linezolid ATS = American Thoracic Society; CA-MRSA = community-acquired methicillin-resistant Staphylococcus aureus; IDSA = lnfectious Diseases Society of America. "Comorbidities include chronic heart, lung, liver, or kidney disease; diabetes mellitus; alcoholism; asplenia; malignancies; and immunosuppression. 'Recommended oral p-lactams include amoxicillin-clavulanate, cefuroxime, or cefpodoxime. dRecommended parenteral B-lactams include ampicillin-sulbactam, ceftriaxone, or ceftaroline.

"Comorbidities include chronic heart, lung, liver, or kidney disease; diabetes mellitus; alcoholism; asplenia; malignancies; and immunosuppression. 'Recommended oral p-lactams include amoxicillin-clavulanate, cefuroxime, or cefpodoxime. dRecommended parenteral B-lactams include ampicillin-sulbactam, ceftriaxone, or ceftaroline. Data from Metlay JB Waterer GW Long AC, et al. Diagnosis and treatment of adults with community-acquired pneumonia. An official clinical practice guideline of the American Thoracic Society and lnfectious Diseases Society of America. Am J Respir Crit Care Med. 2019;200:e45-e67. IPMID: 31573350] doi:10.1 164/rccm.20190B-1 5Bl ST KEY POItIS (clrdnwd) Diagnostic Evaluation o Streptococcus pneumonioe accounts for only 5% to 15% CAP should be suspected and chest imaging performed in any

Data from Metlay JB Waterer GW Long AC, et al. Diagnosis and treatment of adults with community-acquired pneumonia. An official clinical practice guideline of the American Thoracic Society and lnfectious Diseases Society of America. Am J Respir Crit Care Med. 2019;200:e45-e67. IPMID: 31573350] doi:10.1 164/rccm.20190B-1 5Bl ST KEY POItIS (clrdnwd) Diagnostic Evaluation o Streptococcus pneumonioe accounts for only 5% to 15% CAP should be suspected and chest imaging performed in any of community-acquired pneumonia (CAP) infections patient presenting with fever associated with cough, dyspnea, requiring hospitalization, whereas rates of CAP caused or chest pain. Symptoms may be subtle or absent in older by Stophy lococcus euretts, Pseudomonas oeruginoso, adults or immunosuppressed patients, and clinicians should and Enterobacteriaceae are rising, even among patients maintain a low threshold for pursuing radiographic studies in without identifiable health care exposure. these populations. Posteroanterior and lateral chest radiogra- phy is recommended, acknowledging that, compared with o The significance of viral detection in community- chest CT, the sensitivity is less than 50%. In addition to con- acquired pneumonia is unclear; however, an antecedent firming the diagnosis, radiographic patterns may provide clues mild respiratory viral infection may increase the risk for to particular causes (Table 1O), guide clinical decisions regard- a secondary bacterial infection. ing appropriate site of care, and inform need for thoracentesis. 17

Community-Acquired Pneumonia Legionella positive test result does not exclude a concomitant bacterial pneumophila Staphylococcus pathogen. 1"/" aureus Enterobacteriaceae Diagnostic studies to identiflz a causative organism are not Mycoplasma 1o/o routinely indicated in outpatients with CAP but should be pneumoniae considered in hospitalized patients, particularly when they are 2% Fungus/ mycobacteria at risk for infection with methicillin-resistant S. oureus Streptococcus 1o/" (MRSA) or P. aeruginoso. All patients with CAP who require pneumoniae 5o/o admission to the ICU should undergo diagnostic evaluation in Virus plus bacteria an attempt to confirm a microbial cause. Interpretation of 37o sputum Gram stain and culture is hampered by the presence of oropharyngeal colonization, and growth may reflect non- pathogenic organisms. A good-quality sputum culture obtained before antibiotic initiation is suggestive or diagnostic in up to 80% of pneumococcal infections; sensitivity decreases after antibiotic therapy. Sputum Gram stain and culture are appropriate for patients admitted to the ICU (ideally an endotracheal aspirate in intubated patients), patients who did not respond to outpatient antibiotic therapy, patients with t I G U R E 9. The chart depicts percentages of pathogens detected among hospitalized patients with community-acquired pneumonia in the CDC-EPIC Study. cavitary lung lesions, and patients with underlying structural Data from Jain S, Self WH, Wunderink RG, et al. Community-acquired pneumonia requiring hospitalization lung disease. In these cases, consideration of mycobacterial or among U.S. adults. N Engl J Med. 201 5;373:415-27 . fungal causes may also be necessary. Blood culture results are positive in2On/,,to25"/o of patients When plain radiographs are normal but suspicion for CAP with pneumococcal pneumonia, although cultures offer lower remains high, chest radiography may be repeated after yield in patients with other bacterial causes. Blood cultures 24 hours; for patients at high risk (febrile neutropenia, risk for should routinely be obtained in patients with severe pneumo anthrax, or acute respiratory distress syndrome requiring nia and those treated empirically for MRSA or P aeruginoso, intervention) with normal radiographs, chest CT should be because a negative result may allow de-escalation of antibiotic pursued. therapy (see Antimicrobial Therapy). Pneumococcal urinary Routine laboratory studies are indicated to ascertain the antigen testing is more thanTO"/,, sensitive, and results are not severity of infection, determine the optimal site of care, and a ffected by antibi otic admi ni stration. Le gionella urinary anti

Legionella positive test result does not exclude a concomitant bacterial pneumophila Staphylococcus pathogen. 1"/" aureus Enterobacteriaceae Diagnostic studies to identiflz a causative organism are not Mycoplasma 1o/o routinely indicated in outpatients with CAP but should be pneumoniae considered in hospitalized patients, particularly when they are 2% Fungus/ mycobacteria at risk for infection with methicillin-resistant S. oureus Streptococcus 1o/" (MRSA) or P. aeruginoso. All patients with CAP who require pneumoniae 5o/o admission to the ICU should undergo diagnostic evaluation in Virus plus bacteria an attempt to confirm a microbial cause. Interpretation of 37o sputum Gram stain and culture is hampered by the presence of oropharyngeal colonization, and growth may reflect non- pathogenic organisms. A good-quality sputum culture obtained before antibiotic initiation is suggestive or diagnostic in up to 80% of pneumococcal infections; sensitivity decreases after antibiotic therapy. Sputum Gram stain and culture are appropriate for patients admitted to the ICU (ideally an endotracheal aspirate in intubated patients), patients who did not respond to outpatient antibiotic therapy, patients with t I G U R E 9. The chart depicts percentages of pathogens detected among hospitalized patients with community-acquired pneumonia in the CDC-EPIC Study. cavitary lung lesions, and patients with underlying structural Data from Jain S, Self WH, Wunderink RG, et al. Community-acquired pneumonia requiring hospitalization lung disease. In these cases, consideration of mycobacterial or among U.S. adults. N Engl J Med. 201 5;373:415-27 . fungal causes may also be necessary. Blood culture results are positive in2On/,,to25"/o of patients When plain radiographs are normal but suspicion for CAP with pneumococcal pneumonia, although cultures offer lower remains high, chest radiography may be repeated after yield in patients with other bacterial causes. Blood cultures 24 hours; for patients at high risk (febrile neutropenia, risk for should routinely be obtained in patients with severe pneumo anthrax, or acute respiratory distress syndrome requiring nia and those treated empirically for MRSA or P aeruginoso, intervention) with normal radiographs, chest CT should be because a negative result may allow de-escalation of antibiotic pursued. therapy (see Antimicrobial Therapy). Pneumococcal urinary Routine laboratory studies are indicated to ascertain the antigen testing is more thanTO"/,, sensitive, and results are not severity of infection, determine the optimal site of care, and a ffected by antibi otic admi ni stration. Le gionella urinary anti ensure appropriate antimicrobial dosing. HIV testing should gen test results are positive in most patients with L. pneu- be performed if indicated; a positive result expands the spec- mophilaserotype 1 infection. However, the test does not detect trum of potentially causative organisms. Procalcitonin level is other strains, and results can remain positive for prolonged insufficiently sensitive or specific to guide decision making periods after infection. Urinary antigen tests should be regarding initiation of antimicrobial therapy. Rapid testing for reserved for use in patients with severe CAP or with exposure influenza virus during influenza activity may assist in identi- to a Legionello outbreak. fying patients who would benefit from oseltamivir and who To improve detection of influenza, rapid nucleic acid require droplet precautions at hospital admission, but a amplification tests should be used instead of rapid influenza diagnostic tests. Respiratory viral panel results using nucleic TA$LE 1CI" Radiographic Patterns Associated with Specific acid amplification are positive in up to 40% of patients hospi Causes of Community-Acquired Pneumonia talized with CAP; however, a positive result may reflect viral Radiographic Common Pathogens coinfection or antecedent predisposing infection rather than Appearance current clinical illness. Although these panels are less helpful Lobarpneumonia Streptococcuspneumoniae in guiding decisions about discontinuing antibiotic therapy, a Legionella (mu ltilobar) positive respiratory viral panel might have significant infection Lung abscess/ Oral a naerobes, Staphy lococcu s au reu s, control implications among patients admitted to the hospital. cavitary lesion Klebsiell a p n eu moni ae, Noca rd a, i

ensure appropriate antimicrobial dosing. HIV testing should gen test results are positive in most patients with L. pneu- be performed if indicated; a positive result expands the spec- mophilaserotype 1 infection. However, the test does not detect trum of potentially causative organisms. Procalcitonin level is other strains, and results can remain positive for prolonged insufficiently sensitive or specific to guide decision making periods after infection. Urinary antigen tests should be regarding initiation of antimicrobial therapy. Rapid testing for reserved for use in patients with severe CAP or with exposure influenza virus during influenza activity may assist in identi- to a Legionello outbreak. fying patients who would benefit from oseltamivir and who To improve detection of influenza, rapid nucleic acid require droplet precautions at hospital admission, but a amplification tests should be used instead of rapid influenza diagnostic tests. Respiratory viral panel results using nucleic TA$LE 1CI" Radiographic Patterns Associated with Specific acid amplification are positive in up to 40% of patients hospi Causes of Community-Acquired Pneumonia talized with CAP; however, a positive result may reflect viral Radiographic Common Pathogens coinfection or antecedent predisposing infection rather than Appearance current clinical illness. Although these panels are less helpful Lobarpneumonia Streptococcuspneumoniae in guiding decisions about discontinuing antibiotic therapy, a Legionella (mu ltilobar) positive respiratory viral panel might have significant infection Lung abscess/ Oral a naerobes, Staphy lococcu s au reu s, control implications among patients admitted to the hospital. cavitary lesion Klebsiell a p n eu moni ae, Noca rd a, i Acti n o myces, Rhodococcus, Additional testing is indicated in select patients based on mycobacteria, endemic fungi epidemiologic risk factors (see Table 8) or clinical or radio I nterstitia I infi ltrate Atypical pathogens (Legio nella, graphic findings (see Table 10). Fungal and acid-fast bacilli Myco pl as m a, Ch I a myd o ph i I a), vi ruses stains of sputum or fungal antigen testing can be performed. Pleural effusion/ Oral anaerobes, anginosus-constellatus Serologz for C. burnetii, F. tularensis, Legionella, Mycoplasma, empyema group streptococci, S. aureus, S. and Chlamydophila, using acute and convalescent sera, can pneumoniae document seroconversion or a fourfold increase in titers.

Acti n o myces, Rhodococcus, Additional testing is indicated in select patients based on mycobacteria, endemic fungi epidemiologic risk factors (see Table 8) or clinical or radio I nterstitia I infi ltrate Atypical pathogens (Legio nella, graphic findings (see Table 10). Fungal and acid-fast bacilli Myco pl as m a, Ch I a myd o ph i I a), vi ruses stains of sputum or fungal antigen testing can be performed. Pleural effusion/ Oral anaerobes, anginosus-constellatus Serologz for C. burnetii, F. tularensis, Legionella, Mycoplasma, empyema group streptococci, S. aureus, S. and Chlamydophila, using acute and convalescent sera, can pneumoniae document seroconversion or a fourfold increase in titers. 18

Community-Acquired Pneumonia Diagnostic thoracentesis of parapneumonic effusions TABLE 1 1. Community-Acquired Pneumonia Clinical should be considered to exclude concomitant empyema Decision Support Scoring Systems for Site of Care requiring drainage (see MKSAP 19 Pulmonary and Critical PSl" IDSA/ATSb Care Medicine). Bronchoscopy with transbronchial biopsy Age >50 years Major Criteria should be considered in patients who do not respond to empiric therapy. Comorbidities Need for mechanical ventilation Malignancy Septic shock requiring I(EY POIITS vasopressors Congestive heart failure HVC o Diagnostic studies to identi0r a causative organism are Minor Criteria Cerebrovascular not routinely indicated in outpatients with community- disease Respiration rate >30/min acquired pneumonia (CAP) but should be considered in Kidney disease Temperature <36.8 "C (98.2'F) non ICU hospitalized patients when this information Liver disease Hypotension requiring would change management; diagnostic studies should aggressive fluid resuscitation be performed in all patients admitted to the ICU with Vitalsigns Confusion or disorientation CAP. Pulse rate >125/min BUN >20 mg/dL(>7.1 mmol/L) . For patients with pneumococcal pneumonia, a good- Respiration rate >30/min Leu kocyte cou nt <4000/pL quality sputum culture obtained before antibiotic initia- Temperature <35'C (4x 10e/L) (95 'F) or >40 "C (1 04'F) tion is suggestive or diagnostic in up to 80% of patients; Platelet count <1 00,000/pL pneumococcal urinary antigen testing is more than 70% SBP <90 mm Hg (100 x 10e/L) sensitive, and results are not affected by antibiotic Physicalexamination Por/Fto, ratio <250 administration. Altered mentation M u ltiloba r i nfi ltrates . Blood cultures should routinely be obtained in patients ATS =,American Thoracic Society; BUN = blood urea nitrogen; IDSA = lnfectious with severe pneumonia and those treated empirically Diseases Society of America; PSI = Pneumonia Severity lndex; SBP = systolic blood for methicillin - resist ant Staphylo c o c cus aure us or p ressu re.

Diagnostic thoracentesis of parapneumonic effusions TABLE 1 1. Community-Acquired Pneumonia Clinical should be considered to exclude concomitant empyema Decision Support Scoring Systems for Site of Care requiring drainage (see MKSAP 19 Pulmonary and Critical PSl" IDSA/ATSb Care Medicine). Bronchoscopy with transbronchial biopsy Age >50 years Major Criteria should be considered in patients who do not respond to empiric therapy. Comorbidities Need for mechanical ventilation Malignancy Septic shock requiring I(EY POIITS vasopressors Congestive heart failure HVC o Diagnostic studies to identi0r a causative organism are Minor Criteria Cerebrovascular not routinely indicated in outpatients with community- disease Respiration rate >30/min acquired pneumonia (CAP) but should be considered in Kidney disease Temperature <36.8 "C (98.2'F) non ICU hospitalized patients when this information Liver disease Hypotension requiring would change management; diagnostic studies should aggressive fluid resuscitation be performed in all patients admitted to the ICU with Vitalsigns Confusion or disorientation CAP. Pulse rate >125/min BUN >20 mg/dL(>7.1 mmol/L) . For patients with pneumococcal pneumonia, a good- Respiration rate >30/min Leu kocyte cou nt <4000/pL quality sputum culture obtained before antibiotic initia- Temperature <35'C (4x 10e/L) (95 'F) or >40 "C (1 04'F) tion is suggestive or diagnostic in up to 80% of patients; Platelet count <1 00,000/pL pneumococcal urinary antigen testing is more than 70% SBP <90 mm Hg (100 x 10e/L) sensitive, and results are not affected by antibiotic Physicalexamination Por/Fto, ratio <250 administration. Altered mentation M u ltiloba r i nfi ltrates . Blood cultures should routinely be obtained in patients ATS =,American Thoracic Society; BUN = blood urea nitrogen; IDSA = lnfectious with severe pneumonia and those treated empirically Diseases Society of America; PSI = Pneumonia Severity lndex; SBP = systolic blood for methicillin - resist ant Staphylo c o c cus aure us or p ressu re. P se udomonas aeruginoso. "PSl Step I Screen: lf no factors are present, patient is assigned risk class I and can likely be managed as an outpatient. Patients with at least one risk factor are o Legionello urinary antigen test results are positive in stratified by a more complex Step 2 scale (not shown) into risk class ll-V. bIDSA/ATS Criteria for Defining Severe CAP: Patients with one major criterion have most patients with L. pneumophilo serotype l infection, severe CAP and are best managed in the lCU. Consideration of minor criteria plus but it does not detect other strains, and results can clinicaljudgment are recommended to guide the need for higher levels of care. remain positive for prolonged periods after infection.

P se udomonas aeruginoso. "PSl Step I Screen: lf no factors are present, patient is assigned risk class I and can likely be managed as an outpatient. Patients with at least one risk factor are o Legionello urinary antigen test results are positive in stratified by a more complex Step 2 scale (not shown) into risk class ll-V. bIDSA/ATS Criteria for Defining Severe CAP: Patients with one major criterion have most patients with L. pneumophilo serotype l infection, severe CAP and are best managed in the lCU. Consideration of minor criteria plus but it does not detect other strains, and results can clinicaljudgment are recommended to guide the need for higher levels of care. remain positive for prolonged periods after infection. KEY POII{I Management o A clinical prediction model (the Pneumonia Severity Site of Care Index or Infectious Diseases Society of Americai Ambulatory management is adequate for many patients with American Thoracic Society criteria) can be used to CAP. Clinical prediction models are available to identify identify patients with community-acquired pneumonia patients who can be safely managed in the outpatient setting. who may require hospital or ICU admission. The Pneumonia Severity Index (PSI) is a validated predictor of all cause mortality at 30 days. The initial assessment deter- Antimicrobial Therapy mines the presence of 11 variables associated with adverse IDSAiATS guidelines for the management of CAP were outcomes (TaUte 11). Patients with no risk factors (severity risk updated in 2Ol9 to address the challenges of increasingly class I) can typically be managed as outpatients. Although the resistant organisms causing CAP. These recommendations PSI may aid in site-of-care decisions, it should not supersede balance the need to effectively treat infection, often in the clinical judgment. The use of the PSI is preferred over the use absence of an identified pathogen, with the competing of the CURB-65 as it identifies a larger proportion of patients imperative for judicious antibiotic use. Treatment recom- at low risk and has greater discriminative power in predicting mendations are stratified by site of care which in turn is mortality. determined by infection severity and comorbidities (see Consensus guidelines by the Infectious Diseases Society Table 9 and Table 11). of American and the American Thoracic Society (IDSA/ATS) In otherwise healthy patients who do not require hospi provide criteria to identify patients with severe pneumonia talization, recommended regimens include monotherapy with who are best managed in the ICU (see Table 11). In addition to doxycycline or amoxicillin (1 g three times per day). Compared the need for mechanical ventilation or vasopressor support to with previous IDSA CAP guidelines, macrolides (azithromycin maintain blood pressure, the presence of three or more minor and clarithromycin) are only recommended if the regional criteria suggests a higher mortality rate for which ICU care is prevalence of pneumococcal resistance to this class is known recommended. to be less than 25'l..

KEY POII{I Management o A clinical prediction model (the Pneumonia Severity Site of Care Index or Infectious Diseases Society of Americai Ambulatory management is adequate for many patients with American Thoracic Society criteria) can be used to CAP. Clinical prediction models are available to identify identify patients with community-acquired pneumonia patients who can be safely managed in the outpatient setting. who may require hospital or ICU admission. The Pneumonia Severity Index (PSI) is a validated predictor of all cause mortality at 30 days. The initial assessment deter- Antimicrobial Therapy mines the presence of 11 variables associated with adverse IDSAiATS guidelines for the management of CAP were outcomes (TaUte 11). Patients with no risk factors (severity risk updated in 2Ol9 to address the challenges of increasingly class I) can typically be managed as outpatients. Although the resistant organisms causing CAP. These recommendations PSI may aid in site-of-care decisions, it should not supersede balance the need to effectively treat infection, often in the clinical judgment. The use of the PSI is preferred over the use absence of an identified pathogen, with the competing of the CURB-65 as it identifies a larger proportion of patients imperative for judicious antibiotic use. Treatment recom- at low risk and has greater discriminative power in predicting mendations are stratified by site of care which in turn is mortality. determined by infection severity and comorbidities (see Consensus guidelines by the Infectious Diseases Society Table 9 and Table 11). of American and the American Thoracic Society (IDSA/ATS) In otherwise healthy patients who do not require hospi provide criteria to identify patients with severe pneumonia talization, recommended regimens include monotherapy with who are best managed in the ICU (see Table 11). In addition to doxycycline or amoxicillin (1 g three times per day). Compared the need for mechanical ventilation or vasopressor support to with previous IDSA CAP guidelines, macrolides (azithromycin maintain blood pressure, the presence of three or more minor and clarithromycin) are only recommended if the regional criteria suggests a higher mortality rate for which ICU care is prevalence of pneumococcal resistance to this class is known recommended. to be less than 25'l.. 19

Community-Acquired Pneumonia For patients with significant comorbidities treated as out- fibrosis) or medical conditions requiring repeated courses of patients, a respiratory fluoroquinolone alone or a p-lactam antibiotics. Infection with this bacterium in the preceding year plus either a macrolide or doxycycline is recommended. or previous hospitalization or parenteral antibiotic administra- Options for oral p-lactams include amoxicillin-clavulanate or tion in the preceding 90 days are also recognized risk factors for a second-generation cephalosporin. Macrolides and quino- infection. When clinical concern for Pseudomonas is present, lones may prolong the QTc interval, and alternative agents treatment should include an antipseudomonal p-lactam should be considered in patients at risk for torsades de pointes, (piperacillin-tazobactam, cefepime, ceftazidime, aztreonam, including those with a history of a long QT interval, taking imipenem, or meropenem) in conjunction with an agent active other medications that can prolong the QT interval, and elec- against atypical organisms (either a macrolide or a respiratory trolyte abnormalities. Patients who have not responded to a fluoroquinolone). Treatment with a quinolone in the empiric previous CAP regimen who still meet criteria for outpatient regimen is relatively contraindicated in patients who did not therapy should be retreated with an agent from an alternative respond to a previous course of a drug in this class. class. Patients with CAP who test positive for influenza should Patients with more severe infections requiring hospitali- also be treated with an antiviral agent regardless of duration of zation may be infected with a broader spectrum of bacterial illness before diagnosis. Although respiratory viruses may pathogens, reflecting host susceptibility and organism viru- cause severe CAP, identification of a viral organism does not lence. Hospitalized patients with CAP are most commonly exclude a bacterial coinfection. S. oureus, S. pneumoniae, and infected with the organisms shown in Figure 9. Recommended Streptococcus pAogenes have all been associated with postin empiric regimens are shown in Table 9. Patients with previ- fluenza necrotizing CAP. Therefore, the IDSA/ATS guidelines ously isolated respiratory MRSA or P aeruginoso are at risk for recommend prescribing antibiotics for CAP even when a viral pathogen is identified. recurrent infections with these organisms; therefore, an anti- biotic active against the previously cultured pathogens should Anaerobic infections are uncommon causes of CAP. For patients in whom a concern exists for anaerobic infection be included. who can be treated in the ambulatory setting, amoxicillin or The microbioloSz and suggested treatment regimens amoxicillin-clavulanate are active against more common among patients with CAP requiring ICU care are shown in oral anaerobic bacteria. In hospitalized patients with CAP, the Table 9. In this population, combination therapy with a mac- addition of anaerobic coverage is only recommended if lung rolide has shown a survival benefit compared with combina- abscess or empyema is present. tion therapy with a respiratory fluoroquinolone, making the For patients with severe pneumonia in the ICU who are former the preferred treatment. responding to empiric therapy without culture evidence of Methicillin-resistant S. oureus (UnSn) is not adequately MRSA or Pseudomonas at 48 hours, the likelihood of infec- treated by the previously discussed empiric regimens, yet it is tion with one of these agents is low and antibiotics can be increasingly recognized as causing CAB particularly in criti- de escalated to a standard regimen for nonsevere CAP in cally ill patients. CAP caused by MRSA is associated with pre- hospitalized patients. For patients with uncomplicated CAP ceding influenza infection, hospitalization or parenteral who demonstrate rapid defervescence and clinical improve- antibiotics in the last 90 days, and injection drug use, although ment over the first 3 days, a S-day course of therapy is ade- it may present in patients without any identifiable risk factors. quate for cure. Exceptions include patients with cavitary Empiric therapy for MRSA should be considered in patients disease or lung abscess, empyema, concomitant bacteremia, with one of these risk factors, MRSA growth on a previous extrapulmonary infection, or ongoing instability, defined as respiratory culture, a suspicious Gram stain (gram-positive persistent fever, abnormal vital signs, or hypoxia. Many cocci in clusters), conventional therapy failure, pleural based authorities also recommend prolonged duration of antibiotics Iung nodules (suggesting septic pulmonary emboli), or cavi- (at least 14 days) for CAP caused by S. aureus or Pseudotfiohas; tary lung lesions. Optimal treatment for MRSA CAP has not fungal or mycobacterial lung infections require an extended been defined, but options include vancomycin and linezolid; course of treatment. ceftaroline is FDA approved for CAP treatment and has activity against MRSA, but it has not been specifically approved for I(EY POII{TS treatment of MRSA CAP. Notably, daptomycin binds to sur- o Outpatient empiric therapy for community-acquired factant, resulting in negligible alveolar levels, and is therefore pneumonia in otherwise healthy patients includes not effective in pulmonary infections. In patients with blood monotherapy with doxycycline or amoxicillin. cultures and respiratory cultures negative for MRSA, antibiotic o Recommended empiric regimens for hospitalized non- therapy can be de-escalated. ICU patients include a parenteral B-lactam agent (a third- P. aeruginoso is another pathogen not adequately treated generation cephalosporin or ampicillin-sulbactam) plus by standard empiric regimens. Pseudomonos should be con- a macrolide or monotherapy with a respiratory fluoro- sidered in immunocompromised patients and in patients with quinolone. (Continued) underlying structural lung disease (bronchiectasis or cystic

For patients with significant comorbidities treated as out- fibrosis) or medical conditions requiring repeated courses of patients, a respiratory fluoroquinolone alone or a p-lactam antibiotics. Infection with this bacterium in the preceding year plus either a macrolide or doxycycline is recommended. or previous hospitalization or parenteral antibiotic administra- Options for oral p-lactams include amoxicillin-clavulanate or tion in the preceding 90 days are also recognized risk factors for a second-generation cephalosporin. Macrolides and quino- infection. When clinical concern for Pseudomonas is present, lones may prolong the QTc interval, and alternative agents treatment should include an antipseudomonal p-lactam should be considered in patients at risk for torsades de pointes, (piperacillin-tazobactam, cefepime, ceftazidime, aztreonam, including those with a history of a long QT interval, taking imipenem, or meropenem) in conjunction with an agent active other medications that can prolong the QT interval, and elec- against atypical organisms (either a macrolide or a respiratory trolyte abnormalities. Patients who have not responded to a fluoroquinolone). Treatment with a quinolone in the empiric previous CAP regimen who still meet criteria for outpatient regimen is relatively contraindicated in patients who did not therapy should be retreated with an agent from an alternative respond to a previous course of a drug in this class. class. Patients with CAP who test positive for influenza should Patients with more severe infections requiring hospitali- also be treated with an antiviral agent regardless of duration of zation may be infected with a broader spectrum of bacterial illness before diagnosis. Although respiratory viruses may pathogens, reflecting host susceptibility and organism viru- cause severe CAP, identification of a viral organism does not lence. Hospitalized patients with CAP are most commonly exclude a bacterial coinfection. S. oureus, S. pneumoniae, and infected with the organisms shown in Figure 9. Recommended Streptococcus pAogenes have all been associated with postin empiric regimens are shown in Table 9. Patients with previ- fluenza necrotizing CAP. Therefore, the IDSA/ATS guidelines ously isolated respiratory MRSA or P aeruginoso are at risk for recommend prescribing antibiotics for CAP even when a viral pathogen is identified. recurrent infections with these organisms; therefore, an anti- biotic active against the previously cultured pathogens should Anaerobic infections are uncommon causes of CAP. For patients in whom a concern exists for anaerobic infection be included. who can be treated in the ambulatory setting, amoxicillin or The microbioloSz and suggested treatment regimens amoxicillin-clavulanate are active against more common among patients with CAP requiring ICU care are shown in oral anaerobic bacteria. In hospitalized patients with CAP, the Table 9. In this population, combination therapy with a mac- addition of anaerobic coverage is only recommended if lung rolide has shown a survival benefit compared with combina- abscess or empyema is present. tion therapy with a respiratory fluoroquinolone, making the For patients with severe pneumonia in the ICU who are former the preferred treatment. responding to empiric therapy without culture evidence of Methicillin-resistant S. oureus (UnSn) is not adequately MRSA or Pseudomonas at 48 hours, the likelihood of infec- treated by the previously discussed empiric regimens, yet it is tion with one of these agents is low and antibiotics can be increasingly recognized as causing CAB particularly in criti- de escalated to a standard regimen for nonsevere CAP in cally ill patients. CAP caused by MRSA is associated with pre- hospitalized patients. For patients with uncomplicated CAP ceding influenza infection, hospitalization or parenteral who demonstrate rapid defervescence and clinical improve- antibiotics in the last 90 days, and injection drug use, although ment over the first 3 days, a S-day course of therapy is ade- it may present in patients without any identifiable risk factors. quate for cure. Exceptions include patients with cavitary Empiric therapy for MRSA should be considered in patients disease or lung abscess, empyema, concomitant bacteremia, with one of these risk factors, MRSA growth on a previous extrapulmonary infection, or ongoing instability, defined as respiratory culture, a suspicious Gram stain (gram-positive persistent fever, abnormal vital signs, or hypoxia. Many cocci in clusters), conventional therapy failure, pleural based authorities also recommend prolonged duration of antibiotics Iung nodules (suggesting septic pulmonary emboli), or cavi- (at least 14 days) for CAP caused by S. aureus or Pseudotfiohas; tary lung lesions. Optimal treatment for MRSA CAP has not fungal or mycobacterial lung infections require an extended been defined, but options include vancomycin and linezolid; course of treatment. ceftaroline is FDA approved for CAP treatment and has activity against MRSA, but it has not been specifically approved for I(EY POII{TS treatment of MRSA CAP. Notably, daptomycin binds to sur- o Outpatient empiric therapy for community-acquired factant, resulting in negligible alveolar levels, and is therefore pneumonia in otherwise healthy patients includes not effective in pulmonary infections. In patients with blood monotherapy with doxycycline or amoxicillin. cultures and respiratory cultures negative for MRSA, antibiotic o Recommended empiric regimens for hospitalized non- therapy can be de-escalated. ICU patients include a parenteral B-lactam agent (a third- P. aeruginoso is another pathogen not adequately treated generation cephalosporin or ampicillin-sulbactam) plus by standard empiric regimens. Pseudomonos should be con- a macrolide or monotherapy with a respiratory fluoro- sidered in immunocompromised patients and in patients with quinolone. (Continued) underlying structural lung disease (bronchiectasis or cystic 20

Tick-Borne Diseases TAgLE 12. Complications of Community-Acquired Pneumonia ;o rgan System Syndrome Comments Pulmonary Nonresolving pneumonia Consi der resista nt i nfectio n s, no n i nfectious ca uses Lung abscess Prolonged course of antimicrobial treatment Empyema Chest tube drainage of infected pleural fluid ARDS Lung protective ventilation strategy indicated; glucocorticoids may decrease this complication Neurologic Delirium May reflect hypoxemia, hypercarbia, or ICU stay Hematologic Leukopenia May be related to sepsis, medication effect Thrombocytopenia May be related to sepsis, medication effect Cardiac Acute coronary syndrome Seen in 5"/"-10"/o of hospitalized patients Cardiac arrhythmias Most commonly atrial fibrillation Kidney Acute kidney injury May be related to hypoperfusion or medication effect Endocrine Adrenal insufficiency Waterhouse-Friderichsen syndrome (acute adrenal necrosis), occurring in the setti n g of overwhel m i n g bacteria I i nfection/septic shock ARDS = acute respiratory distress syndrome.

TAgLE 12. Complications of Community-Acquired Pneumonia ;o rgan System Syndrome Comments Pulmonary Nonresolving pneumonia Consi der resista nt i nfectio n s, no n i nfectious ca uses Lung abscess Prolonged course of antimicrobial treatment Empyema Chest tube drainage of infected pleural fluid ARDS Lung protective ventilation strategy indicated; glucocorticoids may decrease this complication Neurologic Delirium May reflect hypoxemia, hypercarbia, or ICU stay Hematologic Leukopenia May be related to sepsis, medication effect Thrombocytopenia May be related to sepsis, medication effect Cardiac Acute coronary syndrome Seen in 5"/"-10"/o of hospitalized patients Cardiac arrhythmias Most commonly atrial fibrillation Kidney Acute kidney injury May be related to hypoperfusion or medication effect Endocrine Adrenal insufficiency Waterhouse-Friderichsen syndrome (acute adrenal necrosis), occurring in the setti n g of overwhel m i n g bacteria I i nfection/septic shock ARDS = acute respiratory distress syndrome. l( E Y P 0l tl TS (continued) Follow-up . Treatment regimens for patients with community- In adults with CAP whose symptoms have resolved within 5 to acquired pneumonia requiring ICU care include coad- 7 days, routine follow-up chest imaging is typically not ministration of a parenteral B-lactam active against necessary. Streptococcus pneumoniae and a second agent active Readmission rates among hospitalized patients approach against Legionella species. 20%. This population should have close outpatient follow-up to ensure clinical stability after therapy completion.

l( E Y P 0l tl TS (continued) Follow-up . Treatment regimens for patients with community- In adults with CAP whose symptoms have resolved within 5 to acquired pneumonia requiring ICU care include coad- 7 days, routine follow-up chest imaging is typically not ministration of a parenteral B-lactam active against necessary. Streptococcus pneumoniae and a second agent active Readmission rates among hospitalized patients approach against Legionella species. 20%. This population should have close outpatient follow-up to ensure clinical stability after therapy completion. KEY POIT{T Complications o In adults with community-acquired pneumonia whose HVC CAP has a mortality rate of 10% to 12"1, among hospitalized symptoms have resolved within 5 to 7 days, routine follow- patients. Survivors may experience significant morbidity, up chest imaging is usually not necessary. including prolonged hospitalization, protracted convales- cence, and high rates of hospital readmission. Related compli- cations include localized lung inflammation, secondary spread of infection, and toxicity related to treatment (Table 12). Lack of response to antimicrobials raises consideration Tick-Borne Diseases of a resistant or atypical organism, loculated infection (such Lyme Disease as empyema), or an infection mimic (tumor, vasculitis, con- Lyme disease is the most common tick-borne infection in the gestive heart failure, pulmonary embolism). Patients with United States, with more than 30,000 new infections reported significant pleural fluid collections should undergo diag- annually. More than 95% of infections in the United States nostic thoracentesis; chest tube drainage is indicated for occur in the northeastern, mid-Atlantic, and upper Midwest empyema. regions (Figure 1o). These areas are endemic for the vector, Glucocorticoids are not routinely recommended and lxodes scapularis (the blacklegged deer tick; Figure [). The should be reserved for patients without documented adrenal causative spirochete, Borrelia burgdorferi, is transmitted insufficiency or refractory septic shock. intradermally when a tick ingests a blood meal. r(EY P0rr{Ts In highly select situations, prophylaxis with a single dose (ZOO mg) of doxycycline may decrease the risk of Lyme disease . Lack of response to antimicrobials in patients with community-acquired pneumonia raises consideration after a tick bite. Prophylactic doxycycline is only recom- of a resistant or atypical organism, loculated infection, mended if (r) the tick is reliably identified as a blacklegged or an infection mimic. deer tick; (2) attachment lasts 36 hours or longer; (3) antibiot ics can be started less than 72 hours after tick removal; and o Patients hospitalized with community-acquired pneu- (4) prevalence of B. burgdorferi infection of ticks in the region monia experience significant morbidity and are at high exceeds 20%. Otherwise, observation is recommended, with risk for readmission. treatment given if suggestive symptoms develop.

KEY POIT{T Complications o In adults with community-acquired pneumonia whose HVC CAP has a mortality rate of 10% to 12"1, among hospitalized symptoms have resolved within 5 to 7 days, routine follow- patients. Survivors may experience significant morbidity, up chest imaging is usually not necessary. including prolonged hospitalization, protracted convales- cence, and high rates of hospital readmission. Related compli- cations include localized lung inflammation, secondary spread of infection, and toxicity related to treatment (Table 12). Lack of response to antimicrobials raises consideration Tick-Borne Diseases of a resistant or atypical organism, loculated infection (such Lyme Disease as empyema), or an infection mimic (tumor, vasculitis, con- Lyme disease is the most common tick-borne infection in the gestive heart failure, pulmonary embolism). Patients with United States, with more than 30,000 new infections reported significant pleural fluid collections should undergo diag- annually. More than 95% of infections in the United States nostic thoracentesis; chest tube drainage is indicated for occur in the northeastern, mid-Atlantic, and upper Midwest empyema. regions (Figure 1o). These areas are endemic for the vector, Glucocorticoids are not routinely recommended and lxodes scapularis (the blacklegged deer tick; Figure [). The should be reserved for patients without documented adrenal causative spirochete, Borrelia burgdorferi, is transmitted insufficiency or refractory septic shock. intradermally when a tick ingests a blood meal. r(EY P0rr{Ts In highly select situations, prophylaxis with a single dose (ZOO mg) of doxycycline may decrease the risk of Lyme disease . Lack of response to antimicrobials in patients with community-acquired pneumonia raises consideration after a tick bite. Prophylactic doxycycline is only recom- of a resistant or atypical organism, loculated infection, mended if (r) the tick is reliably identified as a blacklegged or an infection mimic. deer tick; (2) attachment lasts 36 hours or longer; (3) antibiot ics can be started less than 72 hours after tick removal; and o Patients hospitalized with community-acquired pneu- (4) prevalence of B. burgdorferi infection of ticks in the region monia experience significant morbidity and are at high exceeds 20%. Otherwise, observation is recommended, with risk for readmission. treatment given if suggestive symptoms develop. 21