Guidelines to Manage

C o m m u n i t y - A c q u i red
Pneumonia
Richard G. Wunderink, MD

KEYWORDS
 Pneumonia  Community-acquired pneumonia  Macrolide  Viral pneumonia  Corticosteroids
 Procalcitonin

KEY POINTS
 Community-acquired pneumonia (CAP) accounts for 78% of deaths from infectious diseases in the
United States, but outcomes continue to be highly variable even on a county level.
 The preponderance of evidence favors use of b-lactam/macrolide combination therapy over b-lac-
tam monotherapy for hospitalized, non–intensive care unit patients.
 The decreasing incidence of pneumococcal pneumonia and greater viral detections in CAP due to
public health efforts, especially pneumococcal conjugate vaccine, will impact future guideline man-
agement recommendations.
 The health care–associated pneumonia definition overestimates the small subgroup of CAP pa-
tients with pathogens resistant to the usual antibiotic regimens; new risk stratification is needed
for individual pathogens, such as methicillin-resistant Staphylococcus aureus.

In the areas of pulmonary, infectious diseases, and challenges continue to emerge. Outcomes for
critical care medicine, no guideline has greater val- lower respiratory tract infections also remain high-
idity and acceptance than that for management of ly variable even on a county-by-county level, much
community-acquired pneumonia (CAP).1 These higher than for other major infectious diseases.3
guidelines have been incorporated into quality This review addresses the current status and dis-
metrics, pay-for-performance, and public report- cusses areas of current debate.
ing of physician and hospital care. Because pneu-
monia is the leading cause of adult admissions in RATIONALE FOR COMMUNITY-ACQUIRED
the United States,2 the leading cause of infectious PNEUMONIA GUIDELINES
deaths,3 and in the differential of the most frequent
symptom complexes in outpatient primary care,4 The rationale for CAP guidelines has evolved over
this attention to CAP guidelines is neither surpris- the last few decades. Significant variability in anti-
ing nor inappropriate. Of all infectious deaths in biotic prescription for CAP without differences in
the United States, lower respiratory tract infection outcome was a primary driver of the initial efforts
accounted for 78.8%, a log greater than human to derive guidelines. Almost every new antibiotic
immunodeficiency virus/AIDS (7.0%) and 2 logs received a US Food and Drug Administration
greater than tuberculosis (0.75%).3 However, (FDA) indication for CAP, resulting in a plethora
despite this high level of acceptance, areas in of individual antibiotics among which to choose.
the guidelines remain controversial, and new The cost/benefit relationship of these new more
chestmed.theclinics.com

Disclosure: Dr R.G. Wunderink has personally received consulting fees from Accelerate Diagnostics, Arsanis,
Biotest, Curetis, GenMark, Glaxo/Smith/Klein, Inflarex, KBP Biosciences, Merck, Nabriva, Pfizer, Roche.
Department of Medicine, Pulmonary and Critical Care, Northwestern University Feinberg School of Medicine,
240 East Huron Street, McGaw M-336, Chicago, IL 60611, USA
E-mail address: [email protected]

Clin Chest Med 39 (2018) 723–731

https://doi.org/10.1016/j.ccm.2018.07.006
0272-5231/18/Ó 2018 Elsevier Inc. All rights reserved.
724 Wunderink

expensive agents compared with each other and Many other countries and regions have also
to generic antibiotics was unclear. Initial CAP developed CAP guidelines, including published
guidelines attempted to address these issues guidelines from Great Britain,12 Spain,13 the
through the opinion of experts in the field and Netherlands,14 Sweden,15 Japan,16 and China.17
were sponsored by professional societies. In the Most guidelines roughly parallel those of the
United States, the American Thoracic Society IDSA/ATS or northern Europe. Although differ-
(ATS) and the Infectious Diseases Society of ences are relatively minimal, these local and
America (IDSA) led these efforts.1,5,6 regional differences are actual internally consistent
Despite the basis of expert opinion only and a with a major emphasis in most guidelines for local
lack of randomized controlled trials specifically adaptation of recommendations.
comparing guideline-recommended therapy to The major difference between the IDSA/ATS
usual care, subsequent studies consistently CAP guidelines1 and those of northern European
demonstrated improved outcomes as a greater countries12,14,15 is the need for macrolide
proportion of CAP patients were managed accord- combination with beta-lactams for hospitalized,
ing to ATS/IDSA CAP guidelines.7–9 The Centers non-ICU patients. Several justifications for macro-
for Medicare and Medicaid Services recognized lide combination exist. The primary is coverage of
the benefit of compliance with these guidelines atypical bacterial pathogens (Mycoplasma,
by incorporating antibiotic choices consistent Chlamydophila, and Legionella). Because detec-
with ATS/IDSA guidelines as a major quality metric tion of these pathogens was difficult, empirical
for public reporting. coverage was recommended. However, rates of
The guidelines have expanded to address other Mycoplasma and Chlamydophila vary by seasons
issues, including appropriate diagnostic testing and can occur in occult epidemics.18,19 Other po-
and the use of adjunctive therapies. New chal- tential benefits include an anti-inflammatory effect
lenges for antibiotic management have also arisen, on host immune response,20 as well as suppres-
including the role of Legionella and community- sion of the pore-forming exotoxin pneumolysin,21
acquired methicillin-resistant Staphylococcus a major virulence factor for Streptococcus
aureus (MRSA),10 and the emergence of CAP pneumoniae.22
cases with other resistant pathogens typically Despite these theoretic benefits of macrolides,
associated with nosocomial infections and tempo- the major driving force for the IDSA/ATS guideline
rarily designated as health care–associated pneu- recommendation for beta-lactam/macrolide com-
monia (HCAP).11 bination therapy was a consistent survival benefit
for combination therapy in large public and admin-
CURRENT GUIDELINES istrative databases.18,23–28 Theoretic benefits of
macrolides were only explored to explain the
The current IDSA/ATS CAP guidelines were observed effect.
published in 2007,1 and a new revision is antici- In contrast, proponents of beta-lactam mono-
pated soon. The current recommended antibiotic therapy base their recommendation on the
choices are listed in Table 1 and do not differ sub- overwhelming dominance of pneumococcus (in
stantially from previous versions. addition of other Streptococci) as the cause of

Table 1
Current recommended antibiotic therapy for community-acquired pneumonia

Patient Category Recommended Alternative

Outpatient: previously healthy Macrolide Doxycycline
Outpatient: underlying disease/ Fluoroquinolone Beta-lactam combined with
previous treatment macrolide or doxycycline
Non-ICU inpatient Beta-lactam combined with Beta-lactam combined with
macrolide or fluoroquinolone doxycycline
monotherapy
ICU patient Beta-lactam combined with Add linezolid or vancomycin for
macrolide or beta-lactam suspected MRSA
combined with fluoroquinolone Change to anti-pseudomonal
beta-lactam and quinolone if
suspect pseudomonas
 Guidelines to Manage Community-Acquired Pneumonia 725

CAP.12,14,15 Atypical pathogen coverage is then published.37 Designed similar to antibiotic FDA-
reserved for patients with specific risk factors or registration trials,38 the primary endpoint was the
failure of prior beta-lactam therapy. Routine mac- more clinically relevant time to clinical stability, a
rolide combination therapy is not warranted, measurement associated with safe discharge
outside of intensive care unit (ICU) admissions, and low risk of readmission.39 The difference in
based on drug toxicity and the potential for devel- proportions achieving clinical stability by day 7
opment of resistance. was 7.6% (95% confidence interval of the differ-
Three informative major studies have been pub- ence 0.8%–16%). In addition, serious adverse
lished since the last guidelines. First, the large events of death and ICU transfer only occurred in
multicenter Centers for Disease Control and monotherapy patients, and readmission rates
Prevention-sponsored Epidemiology of Pneu- were higher in the monotherapy group. Based on
monia In the Community (EPIC) study found that this study, if beta-lactam monotherapy was a
the proportion of adults with detection of an atyp- new antibiotic, the probability of FDA approval
ical bacterial pathogen was 3.7%.29 In contrast, would be very low.
pneumococcal detection occurred in only 5.1%, A synthesis of these data is that a large propor-
despite extensive diagnostic testing. Additional tion of CAP patients can be treated successfully
cases of pneumococcal pneumonia were subse- with beta-lactam monotherapy but are more likely
quently found with research-only diagnostic to fail without careful attention and potentially
tests30 but still remain less than 15% of all detec- longer hospitalization. If the role of guidelines is
tions. These dramatically lower detection rates to indicate the best treatment of most CAP pa-
for the pneumococcus compared with previous tients, especially for primary care physicians and
epidemiologic studies31 likely represent the effect hospitalists, combination therapy clearly gives
of near universal pediatric conjugate pneumo- more consistent and lower risk. An analysis of
coccal vaccination32,33 and increased smoking risk benefit confirms that macrolides are not
cessation efforts. benign but that the mortality cost of not including
A large multicenter cluster-randomized trial in macrolides with beta-lactams outweighs this risk.
the Netherlands appeared to support beta-
lactam monotherapy as equivalent to both IDSA/ CHALLENGES FOR CURRENT AND FUTURE
ATS-recommended treatment regimens.34 How- COMMUNITY-ACQUIRED PNEUMONIA
ever, many aspects of the trial design call into GUIDELINES
question this conclusion. Because of a public
health orientation, the chosen primary endpoint Evidence-based guidelines for an infectious dis-
was 90-day mortality. This endpoint is compro- ease are difficult, with continued evolution in anti-
mised by a greater effect of underlying diseases, biotic resistance, diagnostic tests, and new
such as metastatic cancer, and intercurrent antibiotics invalidating older studies and their
illnesses, especially cardiovascular events. The resultant recommendations. Many important clin-
latter may be related to CAP, but the relationship ical questions do not have sufficient high-quality
to type of antibiotic treatment is very unclear. A studies to include in meta-analyses, and therefore,
more pertinent endpoint is need for alteration of recommendations are often weak or there is no
the antibiotic regimen, which was 8.8% for beta- recommendation. Issues that current and future
lactam monotherapy compared with 6.1% for guidelines committees may face include the
combination and 3.7% for fluoroquinolone mono- following.
therapy. The study was also compromised by a
Minimizing Fluoroquinolone Use
large proportion of patients who crossed over
treatment from that assigned. Differences in health Lost in the debate regarding the need for macro-
care systems are relevant: the hospital length of lides are the excellent results with respiratory
stay (LOS) for the Netherlands is greater than fluoroquinolones, which have CAP outcomes
6 days in contrast to approximately 3 days in the equivalent or better than beta-lactam/macrolide
United States.34–36 This longer observation time combination therapy. However, a major emphasis
allows recognition of failing therapy before of current antibiotic stewardship efforts is to
discharge, thus avoiding readmission, which is minimize use of quinolones when other legitimate
now a major issue for public reporting of the quality alternatives exist.40 This pressure to minimize
of CAP management in the United States. quinolones is based on emerging data on toxicities
A head-to-head randomized controlled trial of and the fact that quinolones remain one of the few
macrolide/beta-lactam combination with mono- orally active agents for serious gram-negative in-
therapy with the identical beta-lactam for non- fections.41 These factors may alter the IDSA/ATS
ICU hospitalized CAP patients has finally been guideline recommendation of fluoroquinolones as
726 Wunderink

equivalent to beta-lactam/macrolide combination the equation for respiratory viral testing. Treatment
therapy for non-ICU hospitalized CAP. of respiratory syncytial virus is in late stage trials,
Only one antibiotic with an entirely new mecha- and others are under development.
nism of action is on the FDA fast track for approval
for CAP. Lefamulin, the first of the pleuromutilin New Diagnostic Platforms
class of antibiotics,42 appears to have activity
One of the other major findings of the EPIC study
equivalent to moxifloxacin, arguably the best fluo-
is that routine diagnostic testing for CAP cause
roquinolone for CAP, while having a spectrum that
is restricted and incomplete: 64% of adults had
includes MRSA. Drawbacks include limited gram-
no pathogen detected.29 Bacteremia rates have
negative coverage, even for the occasional CAP
fallen dramatically, partially because of greater
pathogen. The entirely different mechanism of ac-
emphasis on rapid delivery of antibiotics. Even a
tion makes salvage therapy for prior failures for
single dose of antibiotic may be enough to make
either fluoroquinolones or beta-lactam/macrolide
blood and respiratory tract cultures negative.
attractive. Other new antibiotics are next genera-
Molecular diagnostics offer an alternative to
tions of currently available antibiotic classes (mac-
traditional culture-based methods. They are
rolides, quinolones). Although covering holes in
already the standard for respiratory viral diagnosis.
the current regimens, including MRSA, beta-
Several platforms under development use various
lactam resistance, and macrolide resistance, the
technologies. The greatest limitation is the inability
cost/benefit equation for these agents is less
to detect pathogens directly from blood, and
favorable.
therefore pathogens are susceptible to some of
the same limitations as sputum cultures. Their
Increasing Evidence of Viral Cause in Adult highest yield and most likely benefit will be with
Community-Acquired Pneumonia lower respiratory specimens from intubated pa-
tients, especially those with suspected infection
The EPIC study found that viral detections were
with more resistant pathogens, such as Pseudo-
significantly more common in adult CAP than
monas, Enterobacteriaceae, and MRSA.
bacterial.29 A large proportion (65%) of the EPIC
Most molecular tests will only detect the pres-
patients without etiologic detections had procalci-
ence of a pathogen and offer no data on antibiotic
tonin (PCT) levels in the range (0.25 ng/dL) seen
susceptibility, similar to what is currently available
with viral infection and safely managed without an-
with urinary antigen detection of S pneumoniae or
tibiotics.43–46 The combination of more extensive
Legionella pneumophila. The number of mutations
use of respiratory viral panels and increased avail-
giving rise to beta-lactam resistance in S pneumo-
ability of PCT offers the opportunity to not only di-
niae is too extensive for any polymerase chain
agnose viral CAP but also, more importantly, avoid
reaction–based technology to use for antibiotic
prolonged courses of antibiotics. Several meta-
susceptibility. Conversely, detection of the mecA
analyses demonstrate the safety of avoiding antibi-
gene in S aureus is highly predictive of beta-
otics in patients with persistently low PCT levels,
lactam resistance, as is a specific mutation for
and a guideline committee will have to address
macrolide resistance in Mycoplasma.
this approach.46,47 Although demonstrated safe in
No study yet has compared management based
a variety of CAP settings in European studies,
on these rapid diagnostic tests to standard limited
concern remains regarding the need to use antibi-
testing and empirical antibiotic therapy. The ability
otics for prevention of secondary bacterial pneu-
to detect S pneumoniae or Haemophilus influen-
monia following viral infection. A critical aspect of
zae is unlikely to dramatically change antibiotic
these PCT protocols was a repeat assay if initially
therapy if standard regimens are used, although
negative, because a small proportion of patients
a positive detection in a patient being treated
will have increased levels with time or in response
with broad spectrum agents may have clinical
to a dose of antibiotics. A compromise compatible
value.
with the US health care system emphasis on time
to first antibiotic dose for suspected infections
Management of Severe Community-Acquired
would be to give a single dose of antibiotics while
Pneumonia
awaiting a respiratory viral panel and repeat PCT
level. Persistently low PCT, especially in the setting No prospective randomized controlled trial has
of a positive respiratory panel, can be used to stop been performed specifically on antibiotic treat-
antibiotics and shift to symptomatic care. ment of severe CAP patients. The definition of se-
The only viral pneumonia with a legitimate treat- vere CAP is variable but, specifically, mechanically
ment option currently is influenza. The develop- ventilated and vasopressor-dependent CAP pa-
ment of other antivirals will significantly change tients are routinely excluded from antibiotic trials.
 Guidelines to Manage Community-Acquired Pneumonia 727

Therefore, treatment regimens are based on case data demonstrate worse outcome with broad-
series or cohort studies, often retrospective.48,49 spectrum antibiotics compared with usual CAP
Most guidelines recommend use of combination therapy.60–64 New guidelines will have to address
therapy with a beta-lactam and either a macrolide this small subgroup with alternative risk
or a fluoroquinolone. Support for the former in- stratification.
cludes retrospective studies of bacteremic cases, An important strategy is to avoid combining
although not all studies find a survival advantage pathogens with different risk factors and antibiotic
for combination therapy.50–53 treatment, such as MRSA, Pseudomonas, and
Consistently, retrospective studies demonstrate Acinetobacter, into the single entity of HCAP.
prolonged time to appropriate therapy is associ- MRSA is difficult because of separate risks for
ated with an increased risk for adverse clinical con- true community-acquired strains (USA300 and
sequences in CAP, including mechanical 400)10,65 and the more traditional hospital-
ventilation, acute respiratory distress syndrome, acquired strains that extend into the commu-
septic shock, and acute kidney injury.54–58 Many nity.60,66 Treatment of these 2 MRSA categories
studies also demonstrate a slight but clearly higher may also vary, with a greater need for toxin-
incidence of less common CAP pathogens in se- suppression therapy in the true community-
vere CAP,29,59 mainly S aureus but also including acquired strain.67 Table 2 lists some risk factors
those bacteria often associated with hospital- for these drug-resistant pathogens60,68; applica-
acquired infections. The tendency to use broad- tion of these risk factors to define antibiotic treat-
spectrum treatment is somewhat understandable. ment has not been prospectively validated.
However, the current level of evidence, although
retrospective, suggests worse outcome with broad
spectrum therapy. These contrasting findings result Corticosteroids and Other
in an additional 2 different challenges to CAP guide- Immunomodulatory Agents
line development: definition of patients who do Given the frequency of hospital and ICU admission
benefit from broad-spectrum therapy and, alterna- and the associated mortality despite appropriate
tively, need for adjunctive therapy in severe CAP antibiotics, severe CAP is a likely target for immu-
patients on appropriate antibiotics. nomodulatory therapy. In fact, CAP predominates
in most studies of immunomodulatory treatment of
septic shock, and CAP subgroups often demon-
Alternative to Health Care–Associated
strate the greatest benefit.69–72 However, no agent
Pneumonia Risk Factors for Resistant
is currently on the market, now that drotrecogin
Community-Acquired Pneumonia Pathogens
alfa activated has been withdrawn.73
The original HCAP definition was developed for Although routinely available and inexpensive,
cases of primary bacteremia and then was applied corticosteroid use in CAP remains very controver-
to pneumonia cases.11 A small subgroup of pa- sial. Meta-analysis suggests that steroid treatment
tients with community-onset pneumonia will have can decrease hospitalization by a day,74 but these
pathogens resistant to the usual CAP antibiotic studies were performed in settings wherein the
regimens. Unfortunately, the original HCAP defini- usual LOS was 6 to 7 days and combination ther-
tions grossly overestimated that population, and apy with macrolides was rarely used.35,36 The

Table 2
Risk factors for pathogens resistant to usual community-acquired pneumonia treatment

All drug-resistant 1.Prior broad-spectrum antibiotic therapy

pathogens 2.Recent hospitalization
3.Immunocompromising drugs or conditions
4.Prior colonization/infection with drug-resistant pathogen
5.Nursing home patients unable to perform activities of daily living,
gastrostomy, tracheostomy
CA-MRSA Hospital Strains MRSA Pseudomonas Enterobacteriaceae
Hemoptysis Hemodialysis Structural lung disease Alcoholism
Influenza season Congestive heart failure Severe chronic Cigarette smoking
Neutropenia from obstructive Gastric acid suppression
infection pulmonary disease
Bronchiectasis
728 Wunderink

greatest theoretic benefit was demonstrated in a 9. Gattarello S, Borgatta B, Sole-Violan J, et al.

randomized controlled trial of CAP patients with Decrease in mortality in severe community-
very high inflammatory biomarkers, but this popu- acquired pneumococcal pneumonia: impact of
lation is very infrequent.75 improving antibiotic strategies (2000-2013). Chest
2014;146(1):22–31.
10. Lobo LJ, Reed KD, Wunderink RG. Expanded clin-
SUMMARY
ical presentation of community-acquired methicillin-
Current guidelines for CAP have had a demon- resistant Staphylococcus aureus pneumonia. Chest
strated benefit on outcomes and are widely 2010;138(1):130–6.
accepted. Differences among guidelines likely 11. Guidelines for the management of adults with
reflect the local health care systems and, as hospital-acquired, ventilator-associated, and
such, are appropriately different. Despite this, healthcare-associated pneumonia. Am J Respir
new data, diagnostic testing, and evolving resis- Crit Care Med 2005;171(4):388–416.
tance will require regular updates to guidelines 12. Eccles S, Pincus C, Higgins B, et al, Guideline
but will challenge consensus. Development Group. Diagnosis and management
of community and hospital acquired pneumonia in
REFERENCES adults: summary of NICE guidance. BMJ 2014;
349:g6722.
1. Mandell LA, Wunderink RG, Anzueto A, et al. Infec- 13. Alfageme I, Aspa J, Bello S, et al. Guidelines for the
tious Diseases Society of America/American diagnosis and management of community-acquired
Thoracic Society consensus guidelines on the man- pneumonia. Spanish Society of Pulmonology and
agement of community-acquired pneumonia in Thoracic Surgery (SEPAR). Arch Bronconeumol
adults. Clin Infect Dis 2007;44(Suppl 2):S27–72. 2005;41(5):272–89 [in Spanish].
2. Pfuntner A, Wier LM, Stocks C. Most frequent con- 14. Wiersinga WJ, Bonten MJ, Boersma WG, et al.
ditions in U.S. Hospitals, 2011: statistical brief #162. SWAB/NVALT (Dutch Working Party on Antibiotic
Healthcare cost and utilization project (HCUP) sta- Policy and Dutch Association of Chest Physicians)
tistical briefs. Rockville (MD): Agency for Healthcare guidelines on the management of community-
Research and Quality; 2013. Available at: http:// acquired pneumonia in adults. Neth J Med 2012;
www.ncbi.nlm.nih.gov/pubmed/24228292. 70(2):90–101.
3. El Bcheraoui C, Mokdad AH, Dwyer-Lindgren L, 15. Hedlund J, Stralin K, Ortqvist A, et al. Swedish
et al. Trends and patterns of differences in infectious guidelines for the management of community-
disease mortality among US Counties, 1980-2014. acquired pneumonia in immunocompetent adults.
JAMA 2018;319(12):1248–60. Scand J Infect Dis 2005;37(11–12):791–805.
4. Schappert SM, Burt CW. Ambulatory care visits to 16. Miyashita N, Matsushima T, Oka M, et al, Japanese
physician offices, hospital outpatient departments, Respiratory Society. The JRS guidelines for the man-
and emergency departments: United States, 2001- agement of community-acquired pneumonia in
02. Vital Health Stat 13 2006;(159):1–66. adults: an update and new recommendations. Intern
5. Niederman MS, Bass JB Jr, Campbell GD, et al. Med 2006;45(7):419–28.
Guidelines for the initial management of adults with 17. Cao B, Huang Y, She DY, et al. Diagnosis and treat-
community-acquired pneumonia: diagnosis, ment of community-acquired pneumonia in adults:
assessment of severity, and initial antimicrobial ther- 2016 clinical practice guidelines by the Chinese
apy. American Thoracic Society. Medical Section of Thoracic Society, Chinese Medical Association.
the American Lung Association. Am Rev Respir Clin Respir J 2018;12(4):1320–60.
Dis 1993;148(5):1418–26. 18. Houck PM, MacLehose RF, Niederman MS, et al.
6. Niederman MS, Mandell LA, Anzueto A, et al. Guide- Empiric antibiotic therapy and mortality among
lines for the management of adults with community- medicare pneumonia inpatients in 10 western states
acquired pneumonia. Diagnosis, assessment of : 1993, 1995, and 1997. Chest 2001;119(5):1420–6.
severity, antimicrobial therapy, and prevention. Am 19. Diaz MH, Benitez AJ, Cross KE, et al. Molecular
J Respir Crit Care Med 2001;163(7):1730–54. detection and characterization of mycoplasma
7. Dean NC, Bateman KA, Donnelly SM, et al. pneumoniae among patients hospitalized with
Improved clinical outcomes with utilization of a community-acquired pneumonia in the United
community-acquired pneumonia guideline. Chest States. Open Forum Infect Dis 2015;2(3):ofv106.
2006;130(3):794–9. 20. Amsden GW. Anti-inflammatory effects of macrolides–
8. Dean NC, Silver MP, Bateman KA, et al. Decreased an underappreciated benefit in the treatment of
mortality after implementation of a treatment guide- community-acquired respiratory tract infections
line for community-acquired pneumonia. Am J Med and chronic inflammatory pulmonary conditions?
2001;110(6):451–7. J Antimicrob Chemother 2005;55(1):10–21.
 Guidelines to Manage Community-Acquired Pneumonia 729

21. Anderson R, Steel HC, Cockeran R, et al. Clarithro- 34. Postma DF, van Werkhoven CH, van Elden LJ, et al.
mycin alone and in combination with ceftriaxone in- Antibiotic treatment strategies for community-
hibits the production of pneumolysin by both acquired pneumonia in adults. N Engl J Med 2015;
macrolide-susceptible and macrolide-resistant 372(14):1312–23.
strains of Streptococcus pneumoniae. J Antimicrob 35. Meijvis SC, Hardeman H, Remmelts HH, et al. Dexa-
Chemother 2007;59(2):224–9. methasone and length of hospital stay in patients
22. Dessing MC, Hirst RA, de Vos AF, et al. Role of Toll- with community-acquired pneumonia: a rando-
like receptors 2 and 4 in pulmonary inflammation mised, double-blind, placebo-controlled trial. Lan-
and injury induced by pneumolysin in mice. PLoS cet 2011;377(9782):2023–30.
One 2009;4(11):e7993. 36. Snijders D, Daniels JM, de Graaff CS, et al. Effi-
23. Meehan TP, Fine MJ, Krumholz HM, et al. Quality of cacy of corticosteroids in community-acquired
care, process, and outcomes in elderly patients with pneumonia: a randomized double-blinded clinical
pneumonia. JAMA 1997;278(23):2080–4. trial. Am J Respir Crit Care Med 2010;181(9):
24. Brown RB, Iannini P, Gross P, et al. Impact of initial 975–82.
antibiotic choice on clinical outcomes in 37. Garin N, Genne D, Carballo S, et al. beta-Lactam
community-acquired pneumonia: analysis of a hos- monotherapy vs beta-lactam-macrolide combination
pital claims-made database. Chest 2003;123(5): treatment in moderately severe community-acquired
1503–11. pneumonia: a randomized noninferiority trial. JAMA
25. Garcia Vazquez E, Mensa J, Martinez JA, et al. Lower Intern Med 2014;174(12):1894–901.
mortality among patients with community-acquired 38. Spellberg B, Fleming TR, Gilbert DN. Executive
pneumonia treated with a macrolide plus a beta- summary: workshop on issues in the design and
lactam agent versus a beta-lactam agent alone. Eur conduct of clinical trials of antibacterial drugs in
J Clin Microbiol Infect Dis 2005;24(3):190–5. the treatment of community-acquired pneumonia.
26. Gleason PP, Meehan TP, Fine JM, et al. Associations Clin Infect Dis 2008;47(Suppl 3):S105–7.
between initial antimicrobial therapy and medical 39. Halm EA, Fine MJ, Marrie TJ, et al. Time to clinical
outcomes for hospitalized elderly patients with stability in patients hospitalized with community-
pneumonia. Arch Intern Med 1999;159(21):2562–72. acquired pneumonia: implications for practice
27. Rodrigo C, McKeever TM, Woodhead M, et al. Sin- guidelines. JAMA 1998;279:1452–7.
gle versus combination antibiotic therapy in adults 40. Weiss K, Tillotson GS. Fluoroquinolones for respira-
hospitalised with community acquired pneumonia. tory infection: too valuable to overuse (and too valu-
Thorax 2013;68(5):493–5. able to misuse!). Chest 2002;122(3):1102–3 [author
28. Tessmer A, Welte T, Martus P, et al. Impact of intra- reply: 1103].
venous b-lactam/macrolide versus b-lactam mono- 41. Spellberg B. Community-acquired pneumonia.
therapy on mortality in hospitalized patients with N Engl J Med 2014;370(19):1861–2.
community-acquired pneumonia. J Antimicrob Che- 42. Mendes RE, Farrell DJ, Flamm RK, et al. In vitro ac-
mother 2009;63(5):1025–33. tivity of lefamulin tested against streptococcus
29. Jain S, Self WH, Wunderink RG, et al. Community-ac- pneumoniae with defined serotypes, including
quired pneumonia requiring hospitalization among multidrug-resistant isolates causing lower respira-
U.S. adults. N Engl J Med 2015;373(5):415–27. tory tract infections in the United States. Antimicrob
30. Wunderink RG, Self WH, Anderson EJ, et al. Agents Chemother 2016;60(7):4407–11.
Pneumococcal community-acquired pneumonia de- 43. Self WH, Wunderink RG, Jain S, et al, Etiology of
tected by serotype-specific urinary antigen detec- Pneumonia in the Community (EPIC) Study Investi-
tion assays. Clin Infect Dis 2018;66(10):1504–10. gators. Procalcitonin as a marker of etiology in
31. Marston BJ, Plouffe JF, File TM Jr, et al. Incidence of adults hospitalized with community-acquired pneu-
community-acquired pneumonia requiring hospitali- monia. Clin Infect Dis 2018;66(10):1640–1.
zation. Results of a population-based active surveil- 44. Christ-Crain M, Jaccard-Stolz D, Bingisser R, et al.
lance Study in Ohio. The Community-Based Effect of procalcitonin-guided treatment on antibiotic
Pneumonia Incidence Study Group. Arch Intern use and outcome in lower respiratory tract infec-
Med 1997;157(15):1709–18. tions: cluster-randomised, single-blinded interven-
32. Griffin MR, Zhu Y, Moore MR, et al. U.S. hospitaliza- tion trial. Lancet 2004;363(9409):600–7.
tions for pneumonia after a decade of pneumo- 45. Christ-Crain M, Stolz D, Bingisser R, et al. Procalci-
coccal vaccination. N Engl J Med 2013;369(2): tonin guidance of antibiotic therapy in community-
155–63. acquired pneumonia: a randomized trial. Am J
33. Whitney CG, Farley MM, Hadler J, et al. Decline in Respir Crit Care Med 2006;174(1):84–93.
invasive pneumococcal disease after the introduc- 46. Schuetz P, Briel M, Christ-Crain M, et al. Procalcito-
tion of protein-polysaccharide conjugate vaccine. nin to guide initiation and duration of antibiotic treat-
N Engl J Med 2003;348(18):1737–46. ment in acute respiratory infections: an individual
730 Wunderink

patient data meta-analysis. Clin Infect Dis 2012; antibiotic treatment and empirical broad spectrum
55(5):651–62. antibiotic treatment in patients with community ac-
47. Soni NJ, Samson DJ, Galaydick JL, et al. Procalci- quired pneumonia: a prospective randomised study.
tonin-guided antibiotic therapy: a systematic re- Thorax 2005;60(8):672–8.
view and meta-analysis. J Hosp Med 2013;8(9): 60. Shindo Y, Ito R, Kobayashi D, et al. Risk factors for
530–40. drug-resistant pathogens in community-acquired
48. Frei CR, Attridge RT, Mortensen EM, et al. Guideline- and healthcare-associated pneumonia. Am J Respir
concordant antibiotic use and survival among Crit Care Med 2013;188(8):985–95.
patients with community-acquired pneumonia 61. Attridge RT, Frei CR, Restrepo MI, et al. Guideline-
admitted to the intensive care unit. Clin Ther 2010; concordant therapy and outcomes in healthcare-
32(2):293–9. associated pneumonia. Eur Respir J 2011;38(4):
49. Rodriguez A, Mendia A, Sirvent JM, et al. Combina- 878–87.
tion antibiotic therapy improves survival in patients 62. Chalmers JD, Rother C, Salih W, et al. Healthcare-
with community-acquired pneumonia and shock. associated pneumonia does not accurately identify
Crit Care Med 2007;35(6):1493–8. potentially resistant pathogens: a systematic re-
50. Baddour LM, Yu VL, Klugman KP, et al. Combination view and meta-analysis. Clin Infect Dis 2014;
antibiotic therapy lowers mortality among severely ill 58(3):330–9.
patients with pneumococcal bacteremia. Am J Re- 63. Garcia-Vidal C, Viasus D, Roset A, et al. Low inci-
spir Crit Care Med 2004;170(4):440–4. dence of multidrug-resistant organisms in patients
51. Harbarth S, Garbino J, Pugin J, et al. Lack of effect with healthcare-associated pneumonia requiring
of combination antibiotic therapy on mortality in pa- hospitalization. Clin Microbiol Infect 2011;17(11):
tients with pneumococcal sepsis. Eur J Clin Micro- 1659–65.
biol Infect Dis 2005;24(10):688–90. 64. Webb BJ, Dangerfield BS, Pasha JS, et al. Guide-
52. Martinez JA, Horcajada JP, Almela M, et al. Addition line-concordant antibiotic therapy and clinical out-
of a macrolide to a beta-lactam-based empirical comes in healthcare-associated pneumonia. Respir
antibiotic regimen is associated with lower in- Med 2012;106(11):1606–12.
hospital mortality for patients with bacteremic pneu- 65. Gillet Y, Vanhems P, Lina G, et al. Factors predicting
mococcal pneumonia. Clin Infect Dis 2003;36(4): mortality in necrotizing community-acquired pneu-
389–95. monia caused by Staphylococcus aureus containing
53. Waterer GW, Somes GW, Wunderink RG. Monother- Panton-Valentine leukocidin. Clin Infect Dis 2007;
apy may be suboptimal for severe bacteremic pneu- 45(3):315–21.
mococcal pneumonia. Arch Intern Med 2001; 66. Kollef MH, Shorr A, Tabak YP, et al. Epidemiology
161(15):1837–42. and outcomes of health-care-associated pneu-
54. Rodriguez A, Lisboa T, Blot S, et al. Mortality in ICU monia: results from a large US database of
patients with bacterial community-acquired pneu- culture-positive pneumonia. Chest 2005;128(6):
monia: when antibiotics are not enough. Intensive 3854–62.
Care Med 2009;35(3):430–8. 67. Sicot N, Khanafer N, Meyssonnier V, et al. Methicillin
55. Kojicic M, Li G, Hanson AC, et al. Risk factors resistance is not a predictor of severity in
for the development of acute lung injury in pa- community-acquired Staphylococcus aureus necro-
tients with infectious pneumonia. Crit Care 2012; tizing pneumonia–results of a prospective observa-
16(2):R46. tional study. Clin Microbiol Infect 2013;19(3):E142–8.
56. Garcia-Vidal C, Fernandez-Sabe N, Carratala J, 68. Wunderink RG, Waterer GW. Clinical practice.
et al. Early mortality in patients with community- Community-acquired pneumonia. N Engl J Med
acquired pneumonia: causes and risk factors. Eur 2014;370(6):543–51.
Respir J 2008;32(3):733–9. 69. Abraham E, Anzueto A, Gutierrez G, et al. Double-
57. Leroy O, Santre C, Beuscart C, et al. A five-year blind randomised controlled trial of monoclonal anti-
study of severe community-acquired pneumonia body to human tumour necrosis factor in treatment
with emphasis on prognosis in patients admitted to of septic shock. NORASEPT II Study Group. Lancet
an intensive care unit. Intensive Care Med 1995; 1998;351(9107):929–33.
21(1):24–31. 70. Laterre PF, Garber G, Levy H, et al. Severe
58. Gattarello S, Lagunes L, Vidaur L, et al. Improve- community-acquired pneumonia as a cause of se-
ment of antibiotic therapy and ICU survival in severe vere sepsis: data from the PROWESS study. Crit
non-pneumococcal community-acquired pneu- Care Med 2005;33(5):952–61.
monia: a matched case-control study. Crit Care 71. Wunderink RG, Laterre PF, Francois B, et al. Recom-
2015;19:335. binant tissue factor pathway inhibitor in severe
59. van der Eerden MM, Vlaspolder F, de Graaff CS, community-acquired pneumonia: a randomized trial.
et al. Comparison between pathogen directed Am J Respir Crit Care Med 2011;183(11):1561–8.
 Guidelines to Manage Community-Acquired Pneumonia 731

72. Wunderink RG, Leeper KV Jr, Schein R, et al. Fil- community-acquired pneumonia: a systematic re-
grastim in patients with pneumonia and severe view and meta-analysis. Ann Intern Med 2015;
sepsis or septic shock. Chest 2001;119:523–9. 163(7):519–28.
73. Ranieri VM, Thompson BT, Barie PS, et al. Drotreco- 75. Torres A, Sibila O, Ferrer M, et al. Effect of cortico-
gin alfa (activated) in adults with septic shock. steroids on treatment failure among hospitalized pa-
N Engl J Med 2012;366(22):2055–64. tients with severe community-acquired pneumonia
74. Siemieniuk RA, Meade MO, Alonso-Coello P, et al. and high inflammatory response: a randomized clin-
Corticosteroid therapy for patients hospitalized with ical trial. JAMA 2015;313(7):677–86.