The n e w e ng l a n d j o u r na l of m e dic i n e

clinical practice
Caren G. Solomon, M.D., M.P.H., Editor

Community-Acquired Pneumonia
Richard G. Wunderink, M.D., and Grant W. Waterer, M.B., B.S., Ph.D.

This Journal feature begins with a case vignette highlighting a common clinical problem.
Evidence supporting various strategies is then presented, followed by a review of formal guidelines,
when they exist. The article ends with the authors’ clinical recommendations.

A 67-year-old woman with mild Alzheimer’s disease who has a 2-day history of pro- From the Division of Pulmonary and Crit-
ductive cough, fever, and increased confusion is transferred from a nursing home to ical Care Medicine, Northwestern Uni-
versity Feinberg School of Medicine, Chi-
the emergency department. According to the transfer records, she has had no recent cago (R.G.W., G.W.W.); and the University
hospitalizations or recent use of antibiotic agents. Her temperature is 38.4°C (101°F), of Western Australia, Perth (G.W.W.). Ad-
the blood pressure is 145/85 mm Hg, the respiratory rate is 30 breaths per minute, the dress reprint requests to Dr. Wunderink
at r-wunderink@northwestern.edu.
heart rate is 120 beats per minute, and the oxygen saturation is 91% while she is
breathing ambient air. Crackles are heard in both lower lung fields. She is oriented to N Engl J Med 2014;370:543-51.
DOI: 10.1056/NEJMcp1214869
person only. The white-cell count is 4000 per cubic millimeter, the serum sodium level Copyright © 2014 Massachusetts Medical Society
is 130 mmol per liter, and the blood urea nitrogen is 25 mg per deciliter (9.0 mmol per
liter). A radiograph of the chest shows infiltrates in both lower lobes. How and where
should this patient be treated?

THE CL INIC A L PROBL EM

Pneumonia is sometimes referred to as the forgotten killer. The World Health Or-
ganization estimates that lower respiratory tract infection is the most common
infectious cause of death in the world (the third most common cause overall), with
almost 3.5 million deaths yearly.1 Together, pneumonia and influenza constitute
the ninth leading cause of death in the United States, resulting in 50,000 estimated An audio version
of this article is
deaths in 2010.2 This number is probably underestimated, since deaths from sepsis available at
(for which pneumonia is the most common source)3 and deaths attributed to other NEJM.org
conditions (e.g., cancer and Alzheimer’s disease) for which pneumonia is the termi-
nal event are coded separately.
Community-acquired pneumonia that is severe enough to require hospitaliza-
tion is associated with excess mortality over the subsequent years among survi-
vors,4-6 even among young people without underlying disease.5 Admission to the
hospital for community-acquired pneumonia is also costly, especially if care in an
intensive care unit (ICU) is required.7
Because of the economic cost, associated mortality, and heterogeneity of manage-
ment, community-acquired pneumonia has been a focus of Centers for Medicare and
Medicaid Services (CMS) and the Joint Commission (TJC) quality-improvement
efforts, public reporting of outcomes, and possible pay-for-performance initiatives.8
This article focuses on management strategies for community-acquired pneumo-
nia, with particular emphasis on interventions to reduce mortality and costs.

S T R ATEGIE S A ND E V IDENCE

DIAGNOSIS
The diagnosis of community-acquired pneumonia is not difficult in patients who
do not have underlying cardiopulmonary disease. A triad of evidence of infection

n engl j med 370;6 nejm.org february 6, 2014 543
The New England Journal of Medicine
Downloaded from nejm.org by amalia ro on February 9, 2014. For personal use only. No other uses without permission.
Copyright © 2014 Massachusetts Medical Society. All rights reserved.

creased the rates of pneumococcal pneumonia. chlamydophila. Confusion may be the only presenting symp. . No other uses without permission. particular episode among the large number of trates in up to 15% of cases. for community-acquired pneumonia. widespread vaccination of infants and children sis. the ex. or For outpatients. doxycycline. and a new or changed infiltrate as them to other antibiotics previously approved observed on radiography usually accurately iden. for young adults. especially munity-acquired pneumonia can be very difficult.g. pulmonary fibro.org february 6. or abnormal pulmonary ex. basis of randomized. The key tifies a patient with community-acquired pneu. (fever or chills and leukocytosis).. underly- ing cardiopulmonary disease and atypical presentation in elderly persons can delay recognition. • The current criteria for health care–associated pneumonia result in excessive use of broad-spectrum antibiotic agents. All rights reserved. the coverage of atypical bac- congestive heart failure. the toxin-mediated. and fluoroquinolones are the most appropriate INITIAL MANAGEMENT agents for the atypical bacterial pathogens. (moxifloxacin at a dose of 400 mg per day or ment (home. and two radiologists approved oral antibiotics are recent antibiotic reading the same chest radiograph disagree in use (which may be associated with a risk of class 10% of cases. and In patients with lung cancer. community-acquired. Macrolides.6 nejm. Numerous antibiotics are approved for the toms localized to the respiratory system (cough. inpatient floor. • Patients with three or more minor criteria for severe community-acquired pneumonia (e. For personal use only. • Alternative antibiotic treatment should be based on the presence of multiple risk factors for health care–associated pneumonia. • Although the diagnosis of community-acquired pneumonia is straightforward in most cases. Copyright © 2014 Massachusetts Medical Society. for whom herd immunity from Atypical presentations also complicate diagno. and a high respiratory rate) should receive extensive intervention in the emer- gency department and be considered for admission to the intensive care unit. the diagnosis of com. or uniquely characteristic syndromes (e.g. unit. controlled trials comparing amination). levofloxacin at a dose of 750 mg per day) or the 544 n engl j med 370. treatment of community-acquired pneumonia increased sputum production. The presence of multiple pneumonia-specific alternative risk factors may allow focused diagnostic testing and treatment. 2014. methicillin-resistant Staphylococcus aureus syndrome). to appropriate therapy is adequate coverage of monia. elevated blood urea nitrogen. terial pathogens is most important. and the appropriate location of treat. Table 1 reviews the differential diagnosis Streptococcus pneumoniae and the atypical bacte- of community-acquired pneumonia. structural lung disease).10 resistance12) and cost.org by amalia ro on February 9. Choice of Antibiotic Therapy For patients admitted to a regular hospital Three interrelated decisions must be made al. sis or other chronic infiltrative lung disease.11 agnosis. with a conjugate pneumococcal vaccine has de- tom in elderly patients. shortness of by the Food and Drug Administration on the breath. 2014 The New England Journal of Medicine Downloaded from nejm.9 Infiltrates on radiographs may also be The primary factors in the choice of agent for a subtle: an individual radiologist may miss infil. guidelines from the Infectious Diseases most simultaneously when a patient first pre­ Society of America and the American Thoracic sents — the choice of antibiotic therapy. Society (IDSA–ATS) recommend first-line treat- tent of testing to determine the cause of the ment with either a respiratory fluoroquinolone pneumonia. signs or symp. leading to a delay in di. rial pathogens (mycoplasma. or ICU). specific risks (e.. The n e w e ng l a n d j o u r na l of m e dic i n e key Clinical points Community-acquired pneumonia • Community-acquired pneumonia remains a leading cause of death in the United States and around the world.g. legionella).. confusion. chest pain. • The majority of hospitalized patients with community-acquired pneumonia can be treated with either a respiratory fluoroquinolone or a combination of cephalosporin and a macrolide.

17 Acute exacerbation of bronchiectasis Although S. . clinical pr actice combination of a second-generation or third-gen- Table 1. No other uses without permission. Duration of Antibiotic Treatment biotic dose was associated with increased The currently recommended duration of antibi- in-hospital mortality. unless they are immuno- do not have community-acquired pneumonia.8 observational study involving patients with sep- This cutoff was modified from retrospective tic shock showed a decrease in survival rates of analyses of large Medicare databases20. Pulmonary infarction mended antibiotics increases.14.6 nejm. Differential Diagnosis of Community-Acquired eration cephalosporin and a macrolide. 2014. An biotic dose within 6 hours after presentation.13 These Pneumonia. antibiotics should be given within the monia or even ventilator-associated pneumonia. pneumonia.. Acute eosinophilic pneumonia mon cause of severe community-acquired pneu- Hypersensitivity pneumonitis monia requiring ICU admission. However. prompt fluid resuscitation.19 Since fluoroquinolones disease have essentially the same antibacterial spectrum Influenza as macrolides. All rights reserved. mortality and length Acute exacerbation of pulmonary fibrosis of hospital stay decrease. compromised. the better outcome with macro. 8% for each hour of delay. who are at increased risk for bacteria resistant to ognition of and early intervention for incipient these empirical antibiotic regimens. Copyright © 2014 Massachusetts Medical Society.21 show. efforts to de. combination therapy consisting of a cephalosporin with either Pulmonary vasculitis a fluoroquinolone or a macrolide is recommend. physicians need to identify patients ment. n engl j med 370. mon among these are patients with risk factors proved patient outcomes.25.28 The current IDSA–ATS guidelines do not Health care–associated pneumonia has been cat- recommend a specific time to the administra.26 for health care–associated pneumonia (Table 2).15.org by amalia ro on February 9. 2014 545 The New England Journal of Medicine Downloaded from nejm.13 An exception is made for patients in usually associated with hospital-acquired pneu- shock.13 Observational evidence suggests that the Normal chest radiograph macrolide combination may be associated with Acute exacerbation of chronic obstructive pulmonary better outcomes.16. otic therapy for community-acquired pneumonia crease the time to the first administration of is 5 to 7 days.15 Quality-improvement projects also con.22 and have not resulted in corre. even in in inappropriate antibiotic use in patients who severely ill patients.24 A shorter ORGANISMS time to antibiotic administration may simply be Although the above recommendations apply to a marker of multiple beneficial care patterns the majority of patients with community-acquired (e. less crowding in the emergency depart.13 There is no evidence that pro- antibiotic therapy have resulted in an increase longed courses lead to better outcomes. with adverse consequences such as Clostridium difficile colitis. Cocaine-induced lung injury (“crack lung”) ed. Most com- respiratory failure) that are associated with im. Asthma with associated viral syndrome to explain infec- tious symptoms Timing of Initiation of Therapy A CMS–TJC quality metric for community-acquired pneumonia is administration of the first anti­ first hour after the onset of hypotension. TREATMENT OF PATIENTS AT RISK FOR RESISTANT sponding decreases in mortality. identifying patients with pneumonia that devel- courage treatment as soon as the diagnosis is ops outside the hospital yet is caused by pathogens made.org february 6. Pertussis such as immunomodulation. egorized as a discrete entity. For personal use only.g.23.18. with the goal of tion of the first antibiotic dose but instead en. Acute bronchitis lides may be explained by nonbactericidal effects.27 ing that an interval of more than 4 hours be- tween the initial presentation and the first anti. recommendations are based primarily on large inpatient administrative databases that show Abnormal chest radiograph reduced mortality with recommended antibiotics Congestive heart failure with associated viral syndrome to explain infectious symptoms as compared with other antibiotics or combina- tions. pneumoniae remains the most com. Aspiration pneumonitis sistently show that as adherence to these recom. and the rec.

these regimens have antibiotic therapy. Clinical Features Suggesting Community- Acquired MRSA Pneumonia. well as by the methicillin-sensitive variant) results In reports of data from tertiary care centers. Copyright © 2014 Massachusetts Medical Society. previously healthy patient Pneumonia-specific criteria‡ Severe pneumonia during summer months Hospitalization for ≥2 days during the previous 90 days * MRSA denotes methicillin-resistant Staphylococcus Antibiotic use during the previous 90 days aureus.28 However.* Original criteria* Hospitalization for ≥2 days during the previous 90 days Cavitary infiltrate or necrosis Residence in a nursing home or extended-care facility Rapidly increasing pleural effusion Long-term use of infusion therapy at home. recommended antibiotic regimens are effective creasing recognition that using all these risk for the majority of patients. For personal use only.6 nejm. Table 4 reviews condi- py may lead to antibiotic overtreatment of many tions in which specific testing may lead to differ- patients.33 Exotoxin production by this strain (as ative pathogens.33 with increased mortality among these patients.30. 2014 The New England Journal of Medicine Downloaded from nejm. The appropriate criteria for initial broad.28 Whereas MRSA is commonly identified in † This criterion was not included in the original criteria but patients with risk factors for health care–associ- is frequently included in many studies of health care– ated pneumonia. treatment who did not meet the definition for health care– is recommended with antibiotics that suppress associated pneumonia to have these resistant toxin production. there is in.29 MRSA that causes community-acquired pneumo- nia in previously healthy patients without health care–associated pneumonia or other risk factors including methicillin-resistant Staphylococcus aureus for MDR pathogens has increasingly been recog- (MRSA) and multidrug-resistant (MDR) gram-neg. Be- patients with culture-positive health care–asso.32. Another group of care–associated pneumonia3 or with severe com- patients at risk for pathogens resistant to the munity-acquired pneumonia requiring ICU ad- usual antibiotics for community-acquired pneu. Extensive diagnostic testing is most spectrum therapy remain controversial (see the helpful in patients with risk factors for health Areas of Uncertainty section). cause the clinical presentation of this infection ciated pneumonia were more likely than patients is disproportionately exotoxin-mediated. No other uses without permission. ent treatment. a community-acquired strain of associated pneumonia. which has been associated been associated with reduced mortality. aeruginosa infection is particularly increased in this population. 546 n engl j med 370. Table 3. 2014. . such as linezolid or clindamy- pathogens and to receive initially inappropriate cin (added to vancomycin). DIAGNOSTIC TESTING erage for Pseudomonas aeruginosa and routine MRSA The extent of testing that is warranted to iden- coverage has therefore been recommended for tify the causative microorganism in community- patients with risk factors for health care–associ. nized. All rights reserved. ‡ Pneumonia-specific criteria are from Shindo et al. diagnostic testing factors as indications for broad-spectrum thera.13 in whom the probability of the pres- monia are those with structural lung disease ence of bacteria that are resistant to usual therapy (bronchiectasis or severe chronic obstructive is greatest.org february 6.13 * Original criteria are from the American Thoracic Society and Infectious Diseases Society of America. The n e w e ng l a n d j o u r na l of m e dic i n e Table 2. will rarely affect therapy.31 Empirical broad-spectrum therapy with dual cov. in characteristic presenting features (Table 3). Criteria for Health Care–Associated Pneumonia. mission.org by amalia ro on February 9. Because the ated pneumonia (Table 2). including Gross hemoptysis (not just blood-streaked) antibiotics Concurrent influenza Hemodialysis during the previous 30 days Neutropenia Home wound care Erythematous rash Family member with multidrug-resistant pathogen Skin pustules Immunosuppressive disease or therapy† Young. Nonambulatory status Tube feedings pulmonary disease [COPD]) who have received Immunocompromised status multiple courses of outpatient antibiotics. acquired pneumonia is controversial. the Use of gastric acid suppressive agents frequency of P.

public report- ing and potential a nursing home Test for Urinary recommended SITE OF CARE Recommended if Legionella Strongly † Severe community-acquired pneumonia is defined as community-acquired pneumonia for which admission to the intensive care unit is being considered. respiratory rate ≥30 breaths per minute.34 However. unreliable home support and concern regarding adherence to therapy).. systolic blood pressure <90 mm Hg or diastolic blood pressure ≤60 mm Hg.g. calcu- Strongly recommended if the Strongly recommended if there is tracheal Strongly recommended if there is a pro- ductive cough. Scoring systems that predict short-term mor- during Influenza tality.gov/clinic/psi/psicalc.asp). 2014. Patients transferred to the ICU within 48 hours n engl j med 370. or dure they may be started only in response to a posi- tive test. and age ≥65 years. Change to specific therapy with productive cough uremia [blood urea nitrogen ≥20 mg per decili- Culture there is no cough ter].org by amalia ro on February 9.ahrq. All rights reserved.6 nejm. clinical pr actice Influenza testing in the appropriate season is therapy. both objective (e. perform the diagnostic test that is most likely to affect drainage proce- Change to specific ther. with a score ≥3 indicating the need for hospital- ization) is easy to remember and calculate but has not been as well validated as the PSI score. Health care–acquired Strategy if test result Severe community- ICU Admission Other condition or circumstance ­pneumonia† Decisions regarding initial admission to the ICU pneumonia Condition and Response to acquired of patients with community-acquired pneumonia positive Test Result and questionable cardiopulmonary stability prob- ably have the greatest potential effect on mortal- ity.. 2014 547 The New England Journal of Medicine Downloaded from nejm. Use of the PSI results in fewer admissions of patients with mild illness. point-source inves- ­patient resides in tient has traveled Recommended if pa- apy. such as the Pneumonia Severity Index Add or continue Recommended Recommended recommended Influenza Test ­oseltamivir (PSI)35 and the CURB-65 scores. Although both scores are valid for the analysis if patient has been trans- patient is hypotensive or Change to specific therapy medical unit to the ICU of groups of admissions for quality improvement Recommended if there is cirrhosis or asplenia ferred from a general Recommended or research in community-acquired pneumonia. Between 40% and 60% of patients who present to the emergency depart- Change to narrow ­antibiotic therapy recommendation Test for Urinary ment with community-acquired pneumonia are recommended recommended Pneumococcal No specific admitted.36 were devel- Strongly Season oped specifically to make admission decisions more objective. No other uses without permission. with no increase in adverse outcomes.34-36 Considerable variation in this deci- Strongly Strongly Antigen sion among patients with similar clinical charac- teristics emphasizes the opportunity for stan- dardization.* individual decisions that are inconsistent with Culture Blood the score are often made for legitimate reasons.g. Copyright © 2014 Massachusetts Medical Society. Depending on current local influenza Strongly Strongly Strongly Culture rates. Diagnostic Testing and Response.Change to specific recommended recommended recommended Pleural-Fluid treatment. Antigen Hospital Admission recently tigation A physician’s decision to hospitalize a patient with community-acquired pneumonia is the ma- jor determinant of cost.org february 6. not recommended if lating the PSI score is complex. pda. and ICU intensive care unit. low arterial saturations) and subjective (e. requiring formal Recommended if the patient has struc- ­recommended if there is productive tural lung disease or severe COPD aspirate or bronchoalveolar-lavage scoring or electronic decision support (http:// cough in a nonintubated patient ­aspirate in an intubated patient. The CURB-65 * COPD denotes chronic obstructive pulmonary disease. antiviral treatments may be started em- pirically and stopped if testing is negative. For personal use only. Table 4. Respiratory Tract score (which assigns 1 point each for confusion. .

46 32 to 15%) without substantially increasing direct The most appropriate criteria for identifying ICU admissions. broad-spectrum treatment of almost half the pa- oped and validated. with admission. 2014 The New England Journal of Medicine Downloaded from nejm. No other uses without permission. be to focus attention on patients who have high an analysis that included patients with risk fac- scores while still in the emergency department.37.26 creased mortality in association with broad- The most appropriate use of these scores may spectrum therapy in such patients. measurement tive.46 A multi- ICU admissions of patients who will never need center quality-improvement project showed in- ICU-level interventions.26 Potentially useful interventions patients who should receive initial empirical broad- include aggressive fluid resuscitation.31 subsequent ited ability to identify patients whose condition prospective studies of patients with health care– is likely to deteriorate if they are admitted to a associated pneumonia have shown markedly low- general ward.39-41 benefits of broad-spectrum antibiotic therapy in Because the PSI and CURB-65 scores have lim. trum antibiotic treatment.26. pneumonia (Table 2) were extrapolated from tial admission to a general unit. If although selection bias cannot be ruled out as an followed rigidly. 2014. pneumonia.29.29.g. multicenter study identified six risk factors of arterial blood gas in patients with borderline (Table 2) for pneumonia caused by pathogens hypoxemia or lactate in those with borderline hy. antibiotics for health care–associated pneumonia.30 tory or vasopressor therapy increases with high.26. reassessment after such inter. Copyright © 2014 Massachusetts Medical Society. A recent prospec- initiation of appropriate antibiotics.45. As compared with early observation- ies widely (ranging from 5 to 20%) depending on al studies of culture-positive cases that suggested hospital and health-system characteristics.26 and poor functional status are more important 548 n engl j med 370. The group of risk factors included in these major criteria for ICU admission would the original definition of health care–associated prevent subsequent deterioration better than ini.29 (e.44 The use of criteria should warrant consideration of ICU risk factors for health care–associated pneumonia admission (Table 5).30. studies of health care–associated bacteremia28 The percentage of hospitalized patients with and may therefore not be entirely appropriate for pneumonia who are admitted to the ICU also var. results in overuse of tial admission to the ICU of patients without antibiotics. no prospective stud. These pneumonia-specific risk factors are con- ma and COPD).. and treatment of coexisting illnesses mens recommended by IDSA–ATS guidelines.org by amalia ro on February 9. tors for health care–associated pneumonia who A quality-improvement study showed that in. administration of bronchodilators for asth. gest increased risks of adverse outcomes among These scores have many variables in common persons who are treated with broad-spectrum (Table 5) and use a similar threshold score (ap.6 nejm. persons with these risk factors.13 Other scores for predict.39-41 For each of these scores. .org february 6. The n e w e ng l a n d j o u r na l of m e dic i n e after initial admission to a general medical service A R E A S OF UNCER TA IN T Y have higher mortality than those with an obvious need for ICU care (mechanical ventilation or hy. some centers. the IDSA–ATS guidelines suggest er rates of antibiotic-resistant pathogens and high that the presence of three or more of nine minor rates of culture-negative cases.43. All rights reserved.42 prompt spectrum coverage are unclear.45 Similarly. Of particular concern are findings that sug- er numbers of criteria met or points tallied. the associated recommendation for broad-spec- ies have been performed to establish whether ini.38 However. were treated at Veterans Affairs medical centers creased attention in the emergency department showed higher mortality among those who were to patients with three or more IDSA–ATS minor given broad-spectrum therapy than among those criteria resulted in a decrease in mortality (from who received standard treatment for community- 23 to 6%) and fewer floor-to-ICU transfers (from acquired pneumonia. as the basis for antibiotic choices results in ing clinical deterioration have also been devel. tients with community-acquired pneumonia in the probability of the need for invasive ventila. sistent with those cited in other reports that in- ventions can clarify the trajectory of the patient’s dicate that recent antibiotic use or hospitalization illness. resistant to the usual inpatient antibiotic regi- potension.29. Concerns have been raised that the original defi- potension requiring vasopressors) at the time of nition of health care–associated pneumonia. all result in substantially more explanation for these findings. For personal use only. proximately 3) to consider ICU admission.13.

41 and REA-ICU40 served on radiograph Hypoxemia Ratio of partial pressure of oxygen in arterial blood SMART-COP.* Other Scoring System or Strategy Criterion Definition with Similar Criterion IDSA–ATS minor criteria Confusion None specified SMART-COP.000 platelets/mm3 — Hypotension Hypotension (systolic pressure <90 mm Hg) SMART-COP and CURXO41 39 ­requiring aggressive fluid resuscitation Hypothermia Core temperature of <36°C — Leukopenia White-cell count <4000/mm3 REA-ICU40 Other criteria Lactic acidosis Lactic acid level ≥4 mmol/liter Early goal-directed therapy42 Low pH <7. tional antibiotic regimens for community-acquired tween health care–associated pneumonia and pneumonia is safe when cultures are negative. † The criterion of a pH level of less than 7.40 depending on pH† Low albumin <3.41 and REA-ICU. cal antibiotic therapy for health care–associated modialysis. clinical pr actice Table 5. All rights reserved.41 and REA-ICU40 to fraction of inspired oxygen <250 mm Hg Thrombocytopenia <100.41 and REA-ICU40 Elevated blood urea nitrogen Blood urea nitrogen ≥20 mg/dl CURXO41 and REA-ICU40 Tachypnea Respiratory rate ≥30 breaths/min SMART-COP. For personal use only.29 However.30.5 g/dl SMART-COP39 Hyponatremia Sodium level <130 mmol/liter REA-ICU40 Leukocytosis Leukocyte count >20.org by amalia ro on February 9.35.29 The importance of distinguishing be.41 and REA-ICU40 Multilobar infiltrates ob. MRSA is an exception: the health care–associated pneumonia. Risk increases proportionally with the presence of more than three criteria. Criteria for Consideration of ICU Admission for Patients without an Obvious Need.30–7. Copyright © 2014 Massachusetts Medical Society.6 nejm.47 States. pneumonia is associated with increased mortality monia-specific risk factor may warrant MRSA among patients with culture-positive cases. .39 CURXO.39 CURXO. long-term he.43 Whereas presence of one MRSA-specific risk factor (prior studies indicate that initially inappropriate empiri- MRSA infection or colonization.29.31 coverage (but not dual antipseudomonal antibi.30 is used in the calculation of the CURXO41 score.39 CURXO. duced mortality. depending on scoring system† SMART-COP.29 Targeted diagnostic testing n engl j med 370.39 CURXO. predictors of resistant pathogens than nursing gens.43 community-acquired pneumonia depends on the and such treatment may be associated with re- local prevalence of antibiotic-resistant patho. highlighting the value of knowledge of Available data suggest that the incidence of local epidemiologic data. 2014 549 The New England Journal of Medicine Downloaded from nejm.35 is used in the calculation of the SMART-COP39 and REA-ICU40 scores. which varies markedly within the United home residence alone. No other uses without permission. None specified SMART-COP. and REA-ICU Risk of Early Admission to ICU. The criterion of a pH level of less than 7.org february 6.39 CURXO. MDR pathogens generally is not significantly Data from randomized trials are lacking to increased unless three or more risk factors are guide treatment in patients with culture-negative present.000/mm3 REA-ICU40 Tachycardia Heart rate ≥125 beats/min SMART-COP39 and REA-ICU40 Older age >80 yr CURXO41 and REA-ICU40 * A patient without an obvious need was defined as one who did not require endotracheal intubation and mechanical ventilation or as one who did not have hypotension requiring vasopressors while in the emergency department. observational data suggest that a switch to tradi- otics). or heart failure) and another pneu. 2014. IDSA–ATS denotes Infectious Diseases Society of America–American Thoracic Society.

Yende S. Interobserver reliability of the chest ra- www. Angus DC.requested if she has presented during the appro- ated pneumonia has been removed from the priate season. N Engl J Med 2013. AstraZeneca. presence of pathogens resistant to usual treat- The IDSA–ATS guidelines for community. Chest 1996. et for pneumonia after a decade of pneumo- 3.29:1303-10. Pinsky MR. No other uses without permission. Schünemann HJ. Low DE.40: 4. Infectious Diseases Society of after community-acquired pneumonia. Crit Care Respir Crit Care Med 2002. Predicting antimi- and associated costs of care. America/American Thoracic Society con- Lancet 1993. Xu J.uration. Wilson KC. ization with community-acquired pneu. Kochanek KD. All rights reserved. monia. For personal use only. acquired pneumonia. 2011. Chow JW. Hill LC. Delayed administration of antibiotics community-acquired pneumonia in adults.183:1454-62. Sanofi. An ap. Murphy M.166:717-23. U. The top 10 causes of death. tle has changed regarding antibiotic treatment of As a nursing home resident. Meehan TP.48 American Thoracic Society (ATS) community-acquired pneumonia guidelines committee and participated in the hospital-acquired. cal infections. Linde-Zwirble WT. 7. ney CG. McGeer A. 2014 The New England Journal of Medicine Downloaded from nejm. per minute.34 She has at least the full text of this article at NEJM. Gleason PP. We would measure the arterial blood gas and lactate The IDSA–ATS guidelines for community-acquired levels.org february 6. Medium-term survival after hospital. 9. Severe community-acquired pneumo. hospitalizations Vital Stat Rep 2012. and GlaxoSmithKline. Whit- Deaths: preliminary data for 2010. Lidick. Murphy SL. Angus DC.130:11-5. evaluation of American and British Tho. 1288-97. 2. et al. Vanderkooi OG. References 1. Waterer GW. Wagener MM. 2014. et al. Clin Infect Dis 2007. Copyright © 2014 Massachusetts Medical Society. Epidemiology of severe sepsis in the Unit. Ga- 550 n engl j med 370. Although ICU admission may be prudent. No other potential conflict of interest relevant to this article was reported. respiratory rate ≥30 breaths acquired pneumonia pathogens). Geneva: monia. she GUIDEL INE S would clearly benefit from further evaluation. Grijalva CG. er J.44:Suppl 2:S27-S72. Angus DC. The n e w e ng l a n d j o u r na l of m e dic i n e allows the de-escalation of therapy if cultures four minor criteria for severe community-acquired are negative (or positive for typical community.ment for community-acquired pneumonia. ed States: analysis of incidence. Kessler LA. Is pneumonia really mance measures for community-acquired 13. Natl people. Carcillo J. World Health Organization.org. 14. Influ. and health care–associated pneumonia guide- C ONCLUSIONS A ND lines committee. Criteria and antibiotic nia-specific MDR risk factors but does have risk recommendations for health care–associated pneu. Mandell LA. Am J Powis JE.13 but lit. al.342:30-3. Clermont G. Obrosky DS. 8.the current criteria for health care–associated mendations in this article are generally consis. guidelines.110:343-50. He was co- chair of the last Infectious Diseases Society of America (IDSA) and talized patients outside the ICU. Kessler LA. Dr. Influenza testing should be outdated. term mortality after pneumonia in older 11. suggesting that she would Disclosure forms provided by the authors are available with benefit from inpatient therapy. racic Society diagnostic criteria. and empirical oseltamivir started planned update of the guidelines for hospital. ence of comorbid conditions on long. Green K. diograph in community-acquired pneu- fs310/en/index. However. Vacarello SJ. Clin Infect Dis 2005.S.if the local influenza rate is high. Wunderink sensus guidelines on the management of 5. 10. We would not acquired and ventilator-acquired pneumonia and obtain blood cultures or attempt to obtain spu- will be incorporated in a future guideline by these tum cultures because of the low likelihood of the organizations.S. and uremia). Chest 2006. consulting fees from non-U. RG. multilobar infiltrates. Zhu Y.pneumonia. Griffin MR. Marrie TJ. Yu VL. the patient meets community-acquired pneumonia. Am J Respir Crit Care Med 2004.org by amalia ro on February 9.Healthcare. Waterer GW. CURB-65 score of 4. outcome. since she has no pneumo- tent with these guidelines. 2013 (http:// 169:910-4.through his institution from bioMérieux and Pfizer. praisal of the evidence underlying perfor.int/mediacentre/factsheets/ 6. acquired pneumonia differ only slightly from Dr. Moore MR. given the high respiratory rate and low sat- pneumonia were published 7 years ago. Brancati FL.pneumonia (confusion. and grant support emphasize the use of fluoroquinolones in hospi. J Am Geriatr Soc 2007.html). Fine JM.factors for severe community-acquired pneumo- monia from the older guidelines for hospital-­ nia. crobial resistance in invasive pneumococ- Med 2001. and atypical presentation in community. et al. Ali IS. Albaum MN.55:518-25. and hydrate aggressively. nia: use of intensive care services and 369:155-63.60:1-51. Wunderink RG. Accelerate Diagnostics. ventilator-acquired. Waterer has been selected as co-chair of the R EC OM MENDAT IONS next ATS–IDSA community-acquired pneumonia guidelines com- mittee and is a member of the current IDSA hospital-acquired and The woman described in the vignette has a ventilator-associated pneumonia guidelines committee.who. Am J Respir Crit Care Med A. and the recom. 12. The discussion of health care–associ. and Achaogen. European guidelines keep Crucell (Johnson & Johnson). . Anzueto the old man’s friend? Two-year prognosis pneumonia. coccal vaccination. Wunderink reports receiving fees for board membership from Pfizer.6 nejm. we would initiate treatment with ceftriaxone acquired and ventilator-acquired pneumonia28 are and azithromycin. Wunderink RG. Bayer the option of beta-lactam monotherapy and de.

and clinical outcomes in health comes in elderly patients with pneumo. Santin A. comes in healthcare-associated pneumo- pneumonia and inappropriate utilization 34. agement of possible multidrug-resistant lated to inappropriate antimicrobial ther. tors predicting mortality in necrotizing therapy. JAMA 1997. Houck PM. A Infect Dis 2004. IDSA/ATS minor criteria aided pre-ICU emergency department patients with com- M. Whitby M. Clin al. Chest 2003. Crit Care Med 2006. Late admission to the 16. Nathwani D. Arnold H. ciated. Eurich DT. in community-acquired and healthcare. J Emerg 37. Cavanaugh JE. Woodhead M. Majumdar SR.129: of severe sepsis and septic shock. 43.58:377-82. Meyssonnier V. España PP. Charles PG.47:375-84. Risk factors for drug-resistant pathogens tional prediction rule. Restrepo MI. Fac. España PP. Renaud B. et 23. process. Kumar A. El Solh AA. Quality of care. Crit Care 2009. Pietrantoni C. et al. Williams F. 46. Misdiagnosis of community-acquired 19:E142-E148. 831. Clin Infect Dis 2011. 13:R54. Kollef MH. Quintana Duration of hypotension before initiation ICU in patients with community-acquired JM.org by amalia ro on February 9. 39. 47. Bartlett JG. Cano E. Eur Respir J 2011. Reichley RM. Gross P. Eur Respir J 2013 October 31 Med 2009. No other uses without permission. ventilator-asso. Arch Intern Med 2004. lides but not fluoroquinolones. outcomes for hospitalized elderly patients Academy of Emergency Medicine. Kett DH. Crit Care clinical outcomes in community-acquired pneumonia. pneumonia: analysis of a hospital claims.137:552-7.110:451-7. Am J Respir Crit 41. Rello J. Attridge RT.28: Staphylococcus aureus necrotizing pneu. Decreased with hospital-acquired. Marrie TJ. et al. study. Fine MJ.11:181-9. Stone RA. Berbary E. Chest 2010. 33. 2007. Brar N. Fine MJ. Yealy DM. community-acquired pneumonia: a sys. domized. Labarère J. 27.164: perience. Ma A. et al.131:466-73. guideline for the management of commu. Chen YY. Am J Med 2001. For personal use only. tious Diseases Society of America. et al. Labelle AJ. Care Med 2013. Kobayashi D. 2014. An outbreak of severe Clostridium 2007. Arch Intern Med 1999. Thorax 2003. tients with community-acquired pneumo. Houck PM. monia: results from a large US database 42. Methicillin resistance is not a pre. 2007. Defining community acquired pneu. community-acquired pneumonia: controlled prediction rule to identify low-risk pa. biotic administration rule. Chest 2007. Nsa W. Clin Infect Dis 2008. nia. 48. Gorordo I.34:1589. 44. American Thoracic Society. ton-Valentine leukocidin. Guidelines for the management of adults monia.62:67-74. Meehan TP. N Engl J Med 1997. Mortensen EM. et al. Silver MP. Bateman KA. Epidemiology and nity-acquired pneumonia. Shindo Y. nity-acquired pneumonia: a controlled 96.37:2867-74.53:107-13. et al. monia: results of a prospective observa. lower respiratory tract infections: sum- 25. Clin Infect Dis 45. Timing of antibiotic adminis. Rivers E. N Engl print). Nguyen B. multicentre cohort study. Vanhems P. Crit Care Med 2006.345:1368-77. Anzueto A. community-acquired pneumonia caused care-associated pneumonia: a UK cohort nia. 32. n engl j med 370. A comparison of tration and outcomes for Medicare patients associated pneumonia and community. Implementation of guidelines for man- difficile-associated disease possibly re. Thorax 2007. Infec. and healthcare-associated pneu. septic shock. Risk stratification of early admission to ment guideline for community-acquired monia. Dean NC. 22. Frei CR. Bhat A. Roubinian N. Mukhopadhyay A. monia: improved outcomes with macro. major criteria of severe community-acquired 18. Bhora M. Khatib R. Crit of culture-positive pneumonia. Metersky ML.38:878-87. 26. Chalmers JD. Antibiotics for bacteremic pneu. Coma E. SMART-COP: a tool for predicting the before-and-after design study. 20. dose for pneumonia in the emergency de. Liu LZ. health-care-associated pneumonia. et al.org february 6.39:955-63. Singanaya- 21. before-and-after evaluation and cost-effec. Iannini P. Barlow G. (Epub ahead of print). Chest pneumonia. nursing home-acquired pneumonia. et care unit admission and outcomes for 15. with pneumonia.123:1503-11. James B. and out. Reichley RM. 38. Khanafer N. Association between timing of intensive 159:2562-72. Reducing door-to-antibiotic time in 35. Phua J. Hale D.137:1130-7. et al. Brat. Associations between partment: a white paper and position tal: an international derivation and vali- initial antimicrobial therapy and medical statement prepared for the American dation study. et al. Taylor JK. made database. Chest associated pneumonia. Guidelines for the management of adult tiveness analysis. Fine MJ. Copyright © 2014 Massachusetts Medical Society. et al. the intensive care unit of patients with no pneumonia. Johannes RS. 637-44. et al. Guideline-concordant therapy and out- MG. Macrolides V. Sligl WI. Auble TE.] J Med 2001.45:315-21. Kunkel al. et al. mary.336:243-50. Hinfey PB. Ma A. Wolfe R. Chest 2006. et al. Hadlock CJ. Bratzler DW. monia severity on presentation to hospi. nia. Impact of initial antibiotic choice on resuscitation in severe community-acquired munity-acquired pneumonia. Roberts D. pneumonia: development of an interna- zler DW. Infect Dis 2011. Fakih tional study. All rights reserved. Wood KE. Blasi F. controlled trial. Lancet Infect Control Hosp Epidemiol 2007. Lina G. Health care. mortality after implementation of a treat. Improvement of process-of-care of effective antimicrobial therapy is the pneumonia is associated with higher and outcomes after implementing a critical determinant of survival in human mortality. Auble TE. Indicators of poten- 131:1865-9. Pines JM. Polgreen PM. Quartin AA. of antibiotics: side effects of the 4-h anti. menting pneumonia guidelines: a ran. Tabak YP. gam A.174:1249-56. 212-4. Development and validation of a 19. 31. et al. sor support in community-acquired pneu- 17. Kollef MH. Capelastegui A. Kanwar M.6 nejm. [Erratum. Gupta clinical prediction rule for severe commu- L. Am J Respir Crit Care Med 2005.37:335-40.39:474-80. al.278:2080-4.51:3568-73. Micek ST. Copyright © 2014 Massachusetts Medical Society. The 36. Gillet Y. Kollef MH. Sicot N. Isserman JA. Lim WS. tematic review and meta-analysis.143:881-94. Asadi L. Ito R. Am J Respir and mortality in critically ill patients with outcomes of health-care-associated pneu.17:Suppl measurement of time to first antibiotic et al.188:985-95. 40. 29. van der Eerden MM. culture-positive and culture-negative hospitalized with community-acquired acquired pneumonia: a single-center ex. 6:1-24. Micek ST. Krumholz HM. 171:388-416. Epidemiology. Clin Microbiol Infect 2011. Antimicrob Agents Chemother 2010. 2014 551 The New England Journal of Medicine Downloaded from nejm. dictor of severity in community-acquired tional. need for intensive respiratory or vasopres- Dis 2004. Kollef KE. Brody J. Yealy DM. Early goal-directed therapy in the treatment Care Med 2013 October 23 (Epub ahead of 128:3854-62. Med 2009. . by Staphylococcus aureus containing Pan. Havstad S. Clin Microbiol Infect 2013. Coma E. et al. Renaud B. Capelastegui A. Brown RB. Shorr A. clinical pr actice lusha DH. antibiotic et al. Ann Intern tially drug-resistant bacteria in severe 24. Tjosvold 30. pneumonia in intensive care: an observa- apy for community-acquired pneumonia. Restrepo MI. Lim HF. Effect of increasing the intensity of imple. Clin Infect 28. Ewig S. et al. Chest 2005. et Med 2005. Laing R.