Journal of the Pediatric Infectious Diseases Society

INVITED REVIEW

Macrolides in Children With Community-Acquired

Pneumonia: Panacea or Placebo?
Christopher C. Blyth1,2 and Jeffrey S. Gerber3
1
School of Medicine, University of Western Australia, Crawley; 2Wesfarmers Centre for Vaccines and Infectious Diseases, Telethon Kids Institute, University of Western Australia;
and 3Division of Infectious Diseases, Children’s Hospital of Philadelphia and the Perelman School of Medicine at the University of Pennsylvania

Pneumonia, most often caused by a respiratory virus, is common in childhood. Mycoplasma pneumoniae also is detected frequently,
particularly in older children in the era of pneumococcal conjugate vaccination. Despite recommendations for β-lactam antibiot-
ics, macrolide antibiotics, including erythromycin, clarithromycin, and azithromycin, are prescribed frequently to children with
acute lower respiratory infection. However, the significance of detecting “atypical” pathogens, including M pneumoniae, in children
remains contentious. Considering the potential for antibacterial and anti-inflammatory activities of macrolides, our understanding
of the role of these drugs in acute and chronic infections and in inflammatory conditions is changing. Some observational data have
revealed improved outcomes in adults and children with pneumonia who are prescribed macrolides, although its widespread use
has led to increases in macrolide resistance in Streptococcus pneumoniae and M pneumoniae. Clinical trials to define the role of mac-
rolides in pediatric acute respiratory infection must be prioritized.
Keywords.  community-acquired; macrolides; Mycoplasma pneumonia; pneumonia; Streptococcus pneumonia.

Pneumonia is one of the leading causes of hospitalization for
children in the United States and results in a significant burden
to children, families, and the healthcare system [1]. Current
Infectious Diseases Society of America guidelines for children
and adolescents with community-acquired pneumonia (CAP)
recommend amoxicillin for mild-to-moderate CAP and ampi-
cillin or benzyl-penicillin for children with more severe disease
[2]. In addition, consideration for withholding antimicrobial
therapy for young children with mild disease is recommended
given the burden of respiratory viruses in this age group.
Macrolide antibiotics are recommended for children with find-
ings compatible with CAP secondary to atypical pathogens (eg,
Mycoplasma pneumoniae) or in the setting of severe β-lactam
allergy. British guidelines state that macrolide antibiotics can be
added to amoxicillin if there is no response to first-line therapy,
if atypical pathogens are considered likely, or if disease is very
severe [3]. Despite this relatively narrow recommendation for
macrolide antibiotics, they continue to be prescribed to 35% to
42% of children with CAP [4, 5].
Recent studies in which the etiology of CAP in the setting
of high pneumococcal conjugate vaccine (PCV) coverage was
explored found frequent detection of respiratory viruses and M
pneumoniae in children [6]. A lack of rapid and reliable tests for
Received 4 May 2017; editorial decision 22 August 2017; accepted 9 October 2017.
Correspondence: C. C. Blyth, MBBS, PhD, University of Western Australia, M561, 35 Stirling
Hwy, Crawley, WA 6009, Australia (christopher.blyth@uwa.edu.au).
Journal of the Pediatric Infectious Diseases Society   2017;00(00):1–7
© The Author 2017. Published by Oxford University Press on behalf of The Journal of the
Pediatric Infectious Diseases Society. All rights reserved. For permissions, please e-mail:
journals.permissions@oup.com.
DOI: 10.1093/jpids/pix083
bacterial pneumonia in children and the nonspecific clinical fea-
tures of atypical pneumonia leave clinicians uncertain about the
likely etiology at the point of antibiotic prescription. Conflicting
data about the true role of M pneumoniae in pediatric CAP [7],
increasing macrolide resistance in both Streptococcus pneumo-
niae and M pneumoniae, and observational data on improved
outcomes in those with CAP who are prescribed non-β-lactam
antibiotics make empiric antibiotic choices for children with
CAP an area of ongoing controversy. Given this background, it
is time to reconsider the 6-year-old recommended options for
children with CAP [2, 3]. In this review, we focus particularly
on the role of macrolide antibiotics.
ETIOLOGY OF CAP IN CHILDHOOD
The recent Centers for Disease Control and Prevention–funded
Etiology of Pneumonia in the Community (EPIC) study sought
to determine the viral and bacterial etiologies of radiologically
confirmed CAP in hospitalized children in the United States
between January 2010 and June 2012 [6]. It should be noted
that this study occurred nearly 1 decade after the introduction
of PCVs and concurrent with the shift from the 7-valent to the
13-valent PCV; therefore, generalizing these results to other
settings, particularly those with higher rates of pneumococcal
colonization or lower PCV coverage, should be done cautiously.
In the EPIC study, bacterial pathogens, including S pneumo-
niae, Streptococcus pyogenes, Haemophilus influenzae, and
Staphylococcus aureus, were identified by blood culture, whole-
blood polymerase chain reaction (PCR) assays (for S pneumo-
niae and S pyogenes only), and, when available, pleural fluid
culture and PCR assays. The same pathogens were identified

Macrolides in Children With CAP: Panacea or Placebo? • JPIDS 2017:XX (XX XXXX) • 1

Downloaded from https://academic.oup.com/jpids/advance-article-abstract/doi/10.1093/jpids/pix083/4583480

by University of Tasmania Library user
on 04 January 2018
 from endotracheal aspirates and bronchoalveolar lavage fluid
samples (culture only), when such samples were available. M
pneumoniae and Chlamydophila pneumoniae were detected by
using PCR assays on nasopharyngeal (NP) and oropharyngeal
(OP) swabs. Viral pathogens were detected by PCR on NP and
OP swabs and by acute- and convalescent-phase serology test-
ing [6]. Healthy asymptomatic child controls were recruited
between February 2011 and June 2012. A pathogen was
detected in 81% of these children, which is significantly more
than the 38% observed in a concurrently conducted study in
adults [8]. Respiratory viruses, including respiratory syncytial
virus (RSV), human rhinovirus, and human metapneumovirus,
were the most frequently identified pathogen in all pediatric
age groups. Of the bacterial pathogens identified, M pneumo-
niae was the most frequently detected (8% of patients, ranging
from 2% of those aged <2 years to 23% of those aged 10 years or
older); this pathogen was detected significantly more often than
was pneumococcus (4%) and S aureus (1%) [6]. Other intracel-
lular pathogens, including C pneumoniae, were detected infre-
quently in all age groups (<1%). Apart from human rhinovirus,
other viral and bacterial pathogens (including M pneumoniae)
were identified infrequently from NP/OP swabs of asymptom-
atic control children.
The more frequent detection of M pneumoniae in patients
than in controls in the EPIC study lies in contrast to that in a
recently published Dutch case-control study undertaken during
a period of increased M pneumoniae activity in Europe. Using
PCR and mycoplasma culture, Spuesens et al [7] identified sim-
ilar rates of M pneumoniae detection and load in both popula-
tions (16% in children with acute respiratory infection and 21%
in asymptomatic children). These results confirm the possibility
of detection (or carriage) of M pneumoniae in asymptomatic
children, particularly during periods of increased M pneumo-
niae activity, and they raise questions about the significance of
M pneumoniae detection in children with and in those without
pneumonia.
Despite the advancements in rapid diagnostics, many of
which were used in the aforementioned studies, there remains
an ongoing lack of sensitive and specific diagnostics for pedi-
atric CAP. Diagnostic testing is performed frequently on
upper respiratory tract samples; because pneumonia is a dis-
ease of the lower airways, this approach, although most practi-
cal, might not reflect the true bacterial etiology of pneumonia
(which, for example, could account for the relatively low rate
of identification of pneumococcus in the EPIC study). The
lack of sensitive and specific diagnostics is compounded fur-
ther by our inability to reliably and clinically distinguish CAP
caused by different pathogens; in a meta-analysis of prospec-
tive cohort studies in which children with CAP were exam-
ined, clinical features that distinguished those with from those
without M pneumoniae–associated CAP included chest pain
and crepitations (more frequent in those with Mycoplasma
2  • JPIDS 2017:XX (XX XXXX) •  Blyth and Gerber
Downloaded from https://academic.oup.com/jpids/advance-article-abstract/doi/10.1093/jpids/pix083/4583480
by University of Tasmania Library user
on 04 January 2018
infection) and wheeze (less frequent in those with Mycoplasma
infection), yet no features were sufficiently reliable to distin-
guish Mycoplasma-associated CAP on the basis of clinical
signs [9]. This inherent limitation leads to empiric antibiotic
use for pediatric CAP.
MACROLIDES: ANTIBACTERIAL ACTION AND
RESISTANCE
The macrolide class of antibiotics, derived from natural com-
pounds produced by Gram-positive Streptomyces species, are
characterized by a 14-membered (erythromycin, roxithro-
mycin, clarithromycin), 15-membered (azithromycin), or
16-membered (josamycin, spiramycin) lactone ring to which
amino and neutral side sugars are attached. As a class, mac-
rolides are characterized by moderately broad-spectrum
activity that includes most Gram-positive bacteria, some
Gram-negative bacteria, and several bacteria responsible for
intracellular infection (eg, Mycoplasma spp, Chlamydia spp,
Legionella spp, and numerous Mycobacterium spp) [10]. The
bacteriostatic properties of macrolides are attributable to
reversible binding to the 23S ribosomal RNA (rRNA) of the
50S bacterial ribosomal subunit, which thereby inhibits RNA-
dependent protein synthesis.
A rise in macrolide resistance in respiratory pathogens,
including in both S pneumoniae and M pneumoniae, has been
reported globally. Resistance to macrolide antibiotics most fre-
quently occurs as a result modification of the ribosomal target
by methylation or mutation, decreased uptake of the mole-
cules, and/or active efflux of the drug [11]. In pneumococci,
methylation of the 23S rRNA (most frequently attributable to
a family of rRNA methyltransferases designated Erm enzymes)
and drug efflux (most frequently attributable to msrA gene-en-
coded ABC transporter and mefA gene-encoded MFS [major
facilitator superfamily] pumps) are the most frequent causes
of resistance [11]. In M pneumoniae, a number of specific point
mutations in the 23s rRNA gene that affect attachment of the
macrolide molecule and various degrees of drug resistance
have been described [12, 13]. Rates of macrolide resistance in
pneumococcus vary greatly among those in different countries
[14]. Contemporary data from US children (2012–2014) sug-
gest that half of pneumococci are macrolide resistant, but <10%
are resistant to penicillin [15]. Macrolide-resistant M pneumo-
niae has also emerged more recently with resistance rates that
range from <10% in the United States [16] and United Kingdom
[17] to >90% in some Asian countries [12]. The rates of resis-
tance in both S pneumoniae and M pneumoniae closely parallel
community antimicrobial use, particularly in children [16–18].
Recently, solithromycin, a next-generation fluoroketolide mac-
rolide that retains activity against most macrolide-resistant
strains, was developed. Safety and efficacy trials in pediatric
patients with CAP are underway (ClinicalTrials.gov identifier
NCT02605122).
 MACROLIDES: IMMUNOMODULATORY AND
ADDITIONAL ANTIBACTERIAL AND ANTIVIRAL
EFFECTS
In ecological studies, patients prescribed erythromycin for dif-
fuse pan-bronchiolitis, an inflammatory disorder that is char-
acterized by progressive neutrophil-associated suppurative and
obstructive airway disease with progression to bronchiectasis
and was first noted in east-Asian adults [19–21], had a signifi-
cantly improved survival rate and experienced improvement in
their symptoms and radiological features. Similar effects have
been found with other 14- and 15-member macrolides [22–24].
Findings of the immunomodulatory effect of macrolides, in
addition to any direct antibacterial effect, have led to their use
for a number of other inflammatory and infective conditions.
Several randomized placebo-controlled clinical trials of
macrolides in children and adults with chronic inflammatory
or infectious conditions have been conducted. A meta-anal-
ysis of clinical trials found improvement in forced expiratory
volume over 1 second (FEV1) (mean difference in FEV1, 4.0%
[95% confidence interval (CI), 1.7; 6.1%]) and an increase in the
odds of being free of pulmonary exacerbations after 6 months
in selected subjects with cystic fibrosis (CF) who were given
azithromycin compared with those who were given placebo
[25]. In a recently published report from a placebo-controlled
clinical trial in non-CF-associated bronchiectasis, a reduction
in pulmonary exacerbations over a 6-month period was found
in subjects prescribed azithromycin, despite the failure to detect
a difference in symptom scores or FEV1s [26]. In adults after
lung transplantation, bronchiolitis obliterans–free survival was
found in those given prophylactic azithromycin compared with
those given placebo (hazard ratio, 0.27 [95% CI, 0.09–0.82 ])
[27]. In a clinical trial in adults with chronic obstructive pulmo-
nary disease, the median time to first exacerbation was 92 days
longer in subjects prescribed long-term azithromycin than in
those given placebo (P < .001) [28]. Variations in the macrolide
used, doses, and frequency of administration have made gener-
alizability of the findings from clinical trials difficult to imple-
ment in everyday clinical practice.
There has been significant interest in the role of macrolides
in preventing acute exacerbations of chronic asthma. A recent
meta-analysis of 23 clinical trials that compared children and
adults prescribed prolonged macrolide therapy or placebo failed
to find any difference in exacerbations, other treatments, symp-
toms, or quality of life [29]. Investigators have also examined
the role of macrolides in the treatment of acute exacerbations
of asthma. As reported by Johnston et al [30], a 10-day course
of telithromycin, a structurally similar ketolide antibiotic, when
prescribed within 24 hours of a medically attended acute asthma
exacerbation in adults aged 18 to 55 years with a history of
asthma, resulted in a small but significant reduction in symptom
scores compared with those who were given placebo (no differ-
ence in FEV1 values was detected). More recently, children aged
Macrolides in Children With CAP: Panacea or Placebo? • JPIDS 2017:XX (XX XXXX) • 3
Downloaded from https://academic.oup.com/jpids/advance-article-abstract/doi/10.1093/jpids/pix083/4583480
by University of Tasmania Library user
on 04 January 2018
1 to 3 years with recurrent asthma-like symptoms were allocated
randomly to receive either 3 days of azithromycin or placebo
in a blinded fashion during an acute episode (at least 3 days of
cough, wheeze, or dyspnea) [31]. Of 158 randomized episodes,
therapy for 154 episodes was completed. In the 148 episodes
with adequate follow-up, a significant reduction in the number
of symptomatic days was noted (3.4 vs 7.7 days [azithromycin vs
placebo, respectively]; 63.3% reduction with azithromycin [95%
CI, 56.0; 69.3%]). No effect on the time to next exacerbation was
identified. The authors of both trials acknowledged that undiag-
nosed bacterial infection or macrolide-associated immunomod-
ulatory effects could have resulted in the outcomes observed.
The mechanism by which macrolides exert their immuno-
modulatory action are complex, and an in depth description is
beyond the scope of this review [32, 33]. In brief, macrolides
exert their effect through (1) reducing neutrophil accumulation
in the airway epithelium, which results in a reduction in local
production of interleukin 8 (IL-8), (2) reducing the production
of proinflammatory IL-1β and tumor necrosis factor α (TNF-α)
in macrophages, which results in a further reduction in proin-
flammatory cytokines (IL-6 and IL-12) and increased anti-in-
flammatory cytokine production (IL-10), (3) downregulating
production of proinflammatory mediators (prostaglandin E2,
nitric oxide, TNF-α, IL-8, and IL-1a), (4) modulating key tran-
scription factors, such as activator protein 1 (AP-1) and nuclear
factor κB (NFκB), and (5) decreasing airway mucus production.
As found in cell-culture studies, certain macrolides also
can influence acute viral infection by downregulating antiviral
receptors and inhibiting virus production; the production of
intercellular admission molecule 1 (ICAM-1), the receptor for
entry by human rhinoviruses, and activated RhoA, a receptor
for respiratory syncytial virus F protein, is downregulated by
macrolides [34, 35], and late-stage influenza A  virus produc-
tion also seems to be blocked by macrolides [36]. Additional
antibacterial effects have been found at subtherapeutic mac-
rolide concentrations, particularly in in vitro models with
Pseudomonas aeruginosa; downregulation of virulence factor
production, motility, and biofilm formation have all been found
[37, 38]. It is highly possible that these additional actions influ-
ence outcomes in acute and chronic respiratory conditions.
ADULT PNEUMONIA: EVIDENCE SUPPORTING
MACROLIDE USE
Current Infectious Diseases Society of America guidelines [39]
(published in 2007) recommend combination β-lactam–mac-
rolide therapy or respiratory fluoroquinolone monotherapy for
adults with CAP who have an underlying comorbidity, are at
increased risk of drug-resistant S pneumoniae (DRSP) infection,
or require hospitalization (regardless of comorbidities or DRSP
risk). In previously healthy adults with outpatient CAP (CAP
not requiring hospitalization), macrolide or doxycycline mono-
therapy is recommended.
 Numerous analyses have explored the role of macrolides in
adults with pneumonia, and conflicting results have been found.
After pooling outcomes from clinical trials that enrolled adults
hospitalized with CAP, Eliakim-Raz et al [40] failed to find any
advantage of giving empiric coverage (including macrolides)
for atypical organisms. In their study of adults with acute lower
respiratory infection (including acute bronchitis, acute exac-
erbations of chronic bronchitis, and pneumonia) enrolled in
clinical trials, Laopaiboon et al [41] also did not identify any
difference in outcomes when azithromycin was compared with
β-lactam therapy. These results are in contrast to those from
observational studies; Asadi et al [42] found a significant reduc-
tion in the mortality rate in adults hospitalized with CAP who
were prescribed macrolides compared with those who were not
prescribed macrolides (relative risk [RR], 0.78 [95% CI, 0.64–
0.95]; P = .01; n = 137 574). This effect disappeared, however,
when the analysis was restricted to data from clinical trials (RR,
1.13 [95% CI, 0.65–1.98]; P = .66; n = 1092). Superior outcomes
for combined β-lactam and macrolide therapy were also found
in a pooled analysis of cohort studies (odds ratio [OR], 0.67
[95% CI, 0.61–0.73]; P < .001; n = 42 942) [43].
CHILDHOOD PNEUMONIA: EVIDENCE SUPPORTING
MACROLIDE USE
Despite their frequent prescription for children, there remains a
paucity of pediatric data from specific examinations of the role
of empiric and targeted macrolide therapy for CAP. Published
data from clinical trials and observational studies in which the
efficacy of macrolide therapy was assessed in comparison with
that of the standard of care are listed in Table 1. Apart from a
small single-center trial that found more rapid improvement in
radiological findings in physician-defined “classic” pneumonia
and more rapid improvement in symptoms in physician-de-
fined atypical pneumonia [44], clinical trials have yet to show
any advantage of prescribing macrolides, either alone or in
combination, over β-lactams in younger children. None of the
studies recruited sufficient numbers of older children or com-
pared β-lactams and macrolides in the groups. Observational
studies, with the attendant difficulties in adjusting for differ-
ences in prescribing between patient groups, have explored the
impact of macrolide use. Ambroggio et al [56] compared 1164
children with CAP who were managed in an outpatient setting
and prescribed β-lactam monotherapy with the same number
of children who were prescribed macrolide monotherapy. No
difference in treatment failure (hospitalization or re-presenta-
tion requiring a change of drug within 14 days) was observed in
children aged ≤5 years (adjusted OR [aOR], 0.90 [95% CI, 0.37–
2.22]), but a trend toward better outcomes was seen in children
aged >5 years (aOR, 0.48 [95% CI, 0.22–1.01]). The report from
another study performed by the same authors, a large multi-
center retrospective cohort study of children hospitalized with
4  • JPIDS 2017:XX (XX XXXX) •  Blyth and Gerber
Downloaded from https://academic.oup.com/jpids/advance-article-abstract/doi/10.1093/jpids/pix083/4583480
by University of Tasmania Library user
on 04 January 2018
CAP, described outcomes for children given combination β-lac-
tam–macrolide therapy compared with those for children pre-
scribed β-lactam monotherapy. A 20% reduction in length of
stay was observed in those prescribed combination therapy, and
the effect size was greatest in older children (Table 1).
A lack of trial data supporting treatment recommenda-
tions that specifically consider the role of macrolide therapy
in children with proven atypical pneumonia also remains
[45]. Evidence that supports treatment recommendations are
derived from observational data; Principi et al [46] described
210 and 87 children aged 2 to 14  years with laboratory-con-
firmed (according to results from serology and/or PCR test-
ing of respiratory specimens) M pneumoniae or Chlamydia
pneumoniae infection, respectively, the majority of whom had
radiologically confirmed pneumonia. Of the 191 children with
M pneumoniae and/or C pneumoniae evaluated, 97.2% of those
treated with macrolides and 81.7% treated with other antibi-
otics were considered cured or improved after 4 to 6 weeks
(P < .05). In addition to highlighting improved outcomes with
macrolides in children diagnosed with M pneumoniae and/or
C pneumoniae infection, these data also reveal a high rate of
clinical improvement in the absence of targeted therapy. Other
studies have found moderate reductions in length of fever and
hospital length of stay in children with M pneumoniae CAP
who were prescribed macrolides [47, 48].
CONCLUSIONS
Given recent evidence of the burden of respiratory viruses in
children with CAP, increasing rates of macrolide resistance in
both S pneumoniae and M pneumoniae, and the lack of clini-
cal trial data that show a clear benefit over alternative agents,
the role of macrolide antibiotics in children with CAP remains
unclear. Yet, with M pneumoniae frequently detected (particu-
larly in older children), observational data indicating modest
improvements in outcomes with therapy, and data indicating
the additional activity of macrolide antibiotics in patients with
acute and chronic infective and inflammatory conditions, it is
likely that specific patient groups can benefit from treatment
with macrolide antibiotics. Defining this group and the role
that macrolide antibiotics play is a research priority, particularly
given the frequent prescription of macrolides [5, 49–51] and the
development of newer macrolide antibiotics. Despite decades of
use for acute respiratory tract infection, an adequately powered
clinical trial is the only way to determine the benefits, risks, and
costs (both economic and resistance) according to age group and
in children with CAP attributable to specific pathogens. A num-
ber of challenges need to be considered when designing such
trials in children, including defining optimal end points and
standardizing diagnostic procedures, particularly for pathogens
such as M pneumoniae. Given the self-limiting nature of viral
pneumonia and M pneumoniae disease, it is likely that the time
on 04 January 2018
Table 1.  Clinical Trials and Observational Studies of Pediatric CAP Treated With Macrolide Therapy, Either Alone or With β-Lactam Antibiotics

by University of Tasmania Library user

Reference Study Design Population Primary Outcome Participants Drug Results
Harris et al (1998) [53] Clinical trial US children aged 6 mo to Cure: complete resolution of 456 children, 36 excluded from primary Azithromycin (5 days) vs comparator Overall treatment success: 71.8% vs 72.3% (NS); children
15 y with clinically and symptoms and signs with analysis (n = 420; 285 given (10 days) (amoxicillin-clavulanic aged ≤5 y, 67.2% vs 66.7% (NS); children aged >5 y, 75.7%
radiologically defined CAP radiologically indicated azithromycin; 135 given comparator) acid [≤5-y-olds] or erythromycin vs 77.6% (NS)
improvement (day 15–19) [>5-y-olds])
Wubbel et al (1999) [54] Clinical trial US children aged 6 mo to Cure: complete resolution of 174 children enrolled, 27 excluded from Azithromycin (5 days) vs comparator Overall treatment success: 98.6% vs 96.2% (NS); children
16 y with clinically and symptoms and signs with primary analysis (n = 147; 69 given (10 days) (amoxicillin-clavulanic aged ≤5 y, 97.4% vs 95.9% (NS); children aged >5 y, 100%
radiologically defined CAP radiologically indicated azithromycin, 78 given comparator) acid [≤5-y-olds] or erythromycin vs 96.6% (NS)
improvement (day 14–35) [>5-y-olds])
Kogan et al (2003) [44] Clinical trial Chilean children aged 1 mo Classic CAP: fever on day 3 and/ 110 children enrolled, 4 excluded from Azithromycin (3 days) vs comparator Classic CAP: absence of fever on day 3, 91.3% vs 87.5%
to 14 y with clinically and or radiologically indicated primary analysis (n = 106; 56 given (7 days) (amoxicillin for clinically (NS); improved radiologically by day 7, 81.0% vs 60.9%
radiologically defined CAP improvement by day 7; atypical azithromycin, 50 given comparator) and radiologically defined classic (P < .01); atypical CAP, mean duration of cough, 3.6 vs
CAP: no. of days of cough CAP and erythromycin for clinically 5.5 days (P = .02); improved radiologically by day 14, 100%
and radiologically indicated and radiologically defined atypical vs 81% (NS)
improved by day 14 CAP)
Aurangzeb and Hameed Clinical trial Pakistani children aged 3–72 Clinical improvement after 48 h 126 patients enrolled, 2 excluded from Clarithromycin vs amoxicillin vs Overall: 97.6% vs 97.6% vs 95.2%
(2003) [55] mo hospitalized with primary analysis (n = 124; 42 given cefuroxime (IV followed by oral)
clinically defined CAP clarithromycin, 43 given amoxicillin, 41
given cefuroxime)
Ambroggio et al (2015) Matched retrospective US children aged 1–18 y Treatment failure, defined 1164 children included in the analysis and Macrolide therapy or β-lactam Overall treatment failure: children aged ≤5 y, 2.7% vs 2.9%
[56] cohort study with ICD-coded CAP by hospitalization or matched 1:1, using propensity score monotherapy (aOR, 0.90 [95% CI, 0.37–2.22]); children aged >5 y, 4.9%
from outpatient pediatric re-presentation with a change adjusted probability of assignment vs 9.4% (aOR, 0.48 [95% CI, 0.22–1.01])
practicesa of drug within 14 days
Ambroggio et al (2012) Retrospective cohort US children aged 1–18 y with Hospital LOS and readmission 20 743 children included in the analysis Combination β-lactam + macrolide LOS: aRR, 0.80 (0.75–0.86) (ref = monoRx); LOS (≤5-y-olds),
[57] study ICD-coded CAP admitted within 14 days therapy vs β-lactam monotherapy aRR, 0.96 (0.86–1.06); LOS (6- to 11-y-olds); aRR, 0.85

Downloaded from https://academic.oup.com/jpids/advance-article-abstract/doi/10.1093/jpids/pix083/4583480

to a freestanding children’s (0.79–0.91); LOS (12- to 18-y-olds), aRR, 0.69 (0.49–
hospital 0.98); readmission, aRR, 0.69 (0.41–1.12) (ref = monoRx)
Ambroggio et al (2016) Retrospective cohort US children aged 1–18 y Treatment failure, defined 717 children included in the analysis; Combination β-lactam + macrolide Overall treatment failure: children aged ≤5 y, 8.3% vs 4.9%
[58] study with ICD-coded CAP by hospitalization or analysis was adjusted by propensity of therapy vs β-lactam monotherapy (aOR, 1.34 [95% CI, 0.83–2.18]); children aged >5 y, 4.0%
from outpatient pediatric re-presentation with a change being prescribed combination therapy vs 12.9% (aOR, 0.51 [95% CI, 0.28–0.95])
practices of drug within 14 days
Leyenaar et al (2014) Retrospective cohort US children aged 1–17 y with Hospital LOS and readmission 13 593 children included in the analysis; Combination ceftriaxone + macrolide LOS (<5-y-olds), 2.43 vs 2.37 days (aRR, 0.99 [0.96–
[50] study ICD-coded CAP admitted to within 28 days analysis was adjusted by propensity of therapy vs ceftriaxone monotherapy 1.02]); LOS (5- to 17-y-olds), 2.48 vs 2.60 days (aRR, 0.95
an acute care hospital being prescribed combination therapy [0.92–0.98]); readmission (<5-y-olds), 0.8% vs 0.7% (NS);
readmission (5- to 17-y-olds), 0.6% vs 0.5% (NS)

Abbreviations: aOR, adjusted odds ratio; aRR, adjusted relative risk; CAP, community-acquired pneumonia; ICD, International Classification of Diseases; IV, intravenous; LOS, length of stay; monoRx, monotherapy; NS, not significant; ref, reference.
a
This is the population that was recruited.

Macrolides in Children With CAP: Panacea or Placebo? • JPIDS 2017:XX (XX XXXX) • 5
 to symptom resolution or time to pathogen clearance is a more
feasible end point than the more traditional pneumonia trial
end points, including hospitalization and clinical failure [52].
Such trials need to be prioritized.
Note
Potential conflicts of interest.  All authors: No reported conflicts of
interest. All authors have submitted the ICMJE Form for Disclosure of
Potential Conflicts of Interest. Conflicts that the editors consider relevant to
the content of the manuscript have been disclosed.
References
1. Keren R, Luan X, Localio R, et al; Pediatric Research in Inpatient Settings (PRIS)
Network. Prioritization of comparative effectiveness research topics in hospital
pediatrics. Arch Pediatr Adolesc Med 2012; 166:1155–64.
2. Bradley JS, Byington CL, Shah SS, et al.; Pediatric Infectious Diseases Society and
the Infectious Diseases Society of America. The management of community-ac-
quired pneumonia in infants and children older than 3  months of age: clinical
practice guidelines by the Pediatric Infectious Diseases Society and the Infectious
Diseases Society of America. Clin Infect Dis 2011; 53:e25–76.
3. Harris M, Clark J, Coote N, et al. British Thoracic Society guidelines for the man-
agement of community acquired pneumonia in children: update 2011. Thorax.
2011; 66(Suppl 2):ii1–23.
4. Bowen SJ, Thomson AH. British Thoracic Society Paediatric Pneumonia Audit: a
review of 3 years of data. Thorax 2013; 68:682–3.
5. Handy LK, Bryan M, Gerber JS, Zaoutis T, Feemster KA. Variability in antibiotic
prescribing for community-acquired pneumonia. Pediatrics 2017; 139:e20162331.
6. Jain S, Williams DJ, Arnold SR, et  al.; CDC EPIC Study Team. Community-
acquired pneumonia requiring hospitalization among U.S. children. N Engl J Med
2015; 372:835–45.
7. Spuesens EB, Fraaij PL, Visser EG, et al. Carriage of Mycoplasma pneumoniae in
the upper respiratory tract of symptomatic and asymptomatic children: an obser-
vational study. PLoS Med 2013; 10:e1001444.
8. Jain S, Self WH, Wunderink RG, et  al.; CDC EPIC Study Team. Community-
acquired pneumonia requiring hospitalization among U.S. adults. N Engl J Med
2015; 373:415–27.
9. Wang K, Gill P, Perera R, et al. Clinical symptoms and signs for the diagnosis of
Mycoplasma pneumoniae in children and adolescents with community-acquired
pneumonia. Cochrane Database Syst Rev 2012; 10:CD009175.
10. Grayson ML, Crowe SM, McCarthy JS, et  al. Kucers’ The Use of Antibiotics.
London: ASM Press; 2010.
11. Jelic D, Antolovic R. From erythromycin to azithromycin and new potential ribo-
some-binding antimicrobials. Antibiotics (Basel) 2016; 5.
12. Principi N, Esposito S. Macrolide-resistant Mycoplasma pneumoniae: its role in
respiratory infection. J Antimicrob Chemother 2013; 68:506–11.
13. Spuesens EB, Meyer Sauteur PM, Vink C, van Rossum AM. Mycoplasma pneumo-
niae infections—does treatment help? J Infect 2014; 69(Suppl 1):S42–6.
14. Lynch JP III, Zhanel GG. Streptococcus pneumoniae: does antimicrobial resistance
matter? Semin Respir Crit Care Med 2009; 30:210–38.
15. Pfaller MA, Mendes RE, Castanheira M, Flamm RK, Jones RN, Sader HS.
Ceftaroline activity tested against bacterial isolates causing community-acquired re-
spiratory tract infections and skin and skin structure infections in pediatric patients
from United States hospitals: 2012–2014. Pediatr Infect Dis J 2016; 36:486–91.
16. Yamada M, Buller R, Bledsoe S, Storch GA. Rising rates of macrolide-resistant
Mycoplasma pneumoniae in the central United States. Pediatr Infect Dis J 2012;
31:409–11.
17. Brown RJ, Macfarlane-Smith L, Phillips S, Chalker VJ. Detection of macrolide re-
sistant Mycoplasma pneumoniae in England, September 2014 to September 2015.
Euro Surveill 2015; 20:30078.
18. Lynch III JP, Martinez FJ. Clinical relevance of macrolide-resistant Streptococcus
pneumoniae for community-acquired pneumonia. Clin Infect Dis 2002; 34(Suppl
1):S27–46.
19. Akira M, Higashihara T, Sakatani M, Hara H. Diffuse panbronchiolitis: follow-up
CT examination. Radiology 1993; 189:559–62.
20. Azuma A, Kudoh S. Diffuse panbronchiolitis in east Asia. Respirology 2006;
11:249–61.
21. Kudoh S, Azuma A, Yamamoto M, et al. Improvement of survival in patients with
diffuse panbronchiolitis treated with low-dose erythromycin. Am J Respir Crit
Care Med 1998; 157:1829–32.
6  • JPIDS 2017:XX (XX XXXX) •  Blyth and Gerber
Downloaded from https://academic.oup.com/jpids/advance-article-abstract/doi/10.1093/jpids/pix083/4583480
by University of Tasmania Library user
on 04 January 2018
22. Kadota J, Mukae H, Ishii H, et al. Long-term efficacy and safety of clarithromycin
treatment in patients with diffuse panbronchiolitis. Respir Med 2003; 97:844–50.
23. Li H, Zhou Y, Fan F, et al. Effect of azithromycin on patients with diffuse panbro-
nchiolitis: retrospective study of 51 cases. Intern Med 2011; 50:1663–9.
24. Nakamura H, Fujishima S, Inoue T, et al. Clinical and immunoregulatory effects
of roxithromycin therapy for chronic respiratory tract infection. Eur Respir J
1999; 13:1371–9.
25. Southern KW, Barker PM, Solis-Moya A, Patel L. Macrolide antibiotics for cystic
fibrosis. Cochrane Database Syst Rev 2012; 11:CD002203.
26. Wong C, Jayaram L, Karalus N, et  al. Azithromycin for prevention of exac-
erbations in non-cystic fibrosis bronchiectasis (EMBRACE): a randomised,
double-blind, placebo-controlled trial. Lancet 2012; 380:660–7.
27. Vos R, Vanaudenaerde BM, Verleden SE, et al. A randomised controlled trial of
azithromycin to prevent chronic rejection after lung transplantation. Eur Respir J
2011; 37:164–72.
28. Albert RK, Connett J, Bailey WC, et  al.; COPD Clinical Research Network.
Azithromycin for prevention of exacerbations of COPD. N Engl J Med 2011;
365:689–98.
29. Kew KM, Undela K, Kotortsi I, Ferrara G. Macrolides for chronic asthma.
Cochrane Database Syst Rev 2015; 9:CD002997.
30. Johnston SL, Blasi F, Black PN, et  al.; TELICAST Investigators. The effect of
telithromycin in acute exacerbations of asthma. N Engl J Med 2006; 354:1589–600.
31. Stokholm J, Chawes BL, Vissing NH, et al. Azithromycin for episodes with asth-
ma-like symptoms in young children aged 1–3 years: a randomised, double-blind,
placebo-controlled trial. Lancet Respir Med 2016; 4:19–26.
32. Cameron EJ, McSharry C, Chaudhuri R, et al. Long-term macrolide treatment of
chronic inflammatory airway diseases: risks, benefits and future developments.
Clin Exp Allergy 2012; 42:1302–12.
33. Kanoh S, Rubin BK. Mechanisms of action and clinical application of macrolides
as immunomodulatory medications. Clin Microbiol Rev 2010; 23:590–615.
34. Asada M, Yoshida M, Suzuki T, et  al. Macrolide antibiotics inhibit respiratory
syncytial virus infection in human airway epithelial cells. Antiviral Res 2009;
83:191–200.
35. Suzuki T, Yamaya M, Sekizawa K, et al. Erythromycin inhibits rhinovirus infec-
tion in cultured human tracheal epithelial cells. Am J Respir Crit Care Med 2002;
165:1113–8.
36. Miyamoto D, Hasegawa S, Sriwilaijaroen N, et al. Clarithromycin inhibits progeny
virus production from human influenza virus-infected host cells. Biol Pharm Bull
2008; 31:217–22.
37. Ichimiya T, Takeoka K, Hiramatsu K, et  al. The influence of azithromycin on
the biofilm formation of Pseudomonas aeruginosa in vitro. Chemotherapy 1996;
42:186–91.
38. Molinari G, Guzmán CA, Pesce A, Schito GC. Inhibition of Pseudomonas aerugi-
nosa virulence factors by subinhibitory concentrations of azithromycin and other
macrolide antibiotics. J Antimicrob Chemother 1993; 31:681–8.
39. Mandell LA, Wunderink RG, Anzueto A, et al.; Infectious Diseases Society of
America; American Thoracic Society. Infectious Diseases Society of America/
American Thoracic Society consensus guidelines on the management of
community-acquired pneumonia in adults. Clin Infect Dis 2007; 44(Suppl
2):S27–72.
40. Eliakim-Raz N, Robenshtok E, Shefet D, et  al. Empiric antibiotic coverage of
atypical pathogens for community-acquired pneumonia in hospitalized adults.
Cochrane Database Syst Rev 2012; 9:CD004418.
41. Laopaiboon M, Panpanich R, Swa Mya K. Azithromycin for acute lower respira-
tory tract infections. Cochrane Database Syst Rev 2015; 3:CD001954.
42. Asadi L, Sligl WI, Eurich DT, et al. Macrolide-based regimens and mortality in
hospitalized patients with community-acquired pneumonia: a systematic review
and meta-analysis. Clin Infect Dis 2012; 55:371–80.
43. Nie W, Li B, Xiu Q. β-Lactam/macrolide dual therapy versus β-lactam mono-
therapy for the treatment of community-acquired pneumonia in adults: a system-
atic review and meta-analysis. J Antimicrob Chemother 2014; 69:1441–6.
44. Kogan R, Martínez MA, Rubilar L, et al. Comparative randomized trial of azith-
romycin versus erythromycin and amoxicillin for treatment of community-ac-
quired pneumonia in children. Pediatr Pulmonol 2003; 35:91–8.
45. Gardiner SJ, Gavranich JB, Chang AB. Antibiotics for community-acquired lower
respiratory tract infections secondary to Mycoplasma pneumoniae in children.
Cochrane Database Syst Rev 2015; 1:CD004875.
46. Principi N, Esposito S, Blasi F, Allegra L; Mowgli Study Group. Role of Mycoplasma
pneumoniae and Chlamydia pneumoniae in children with community-acquired
lower respiratory tract infections. Clin Infect Dis 2001; 32:1281–9.
47. Shah SS, Test M, Sheffler-Collins S, et al. Macrolide therapy and outcomes in a
multicenter cohort of children hospitalized with Mycoplasma pneumoniae pneu-
monia. J Hosp Med 2012; 7:311–7.
 48. Lu YJ, Chen TH, Lin LH, et  al. Macrolide use shortens fever duration in
Mycoplasma pneumoniae infection in children: a 2-year experience. J Microbiol
Immunol Infect 2008; 41:307–10.
49. Donnelly JP, Baddley JW, Wang HE. Antibiotic utilization for acute respiratory tract infec-
tions in U.S. emergency departments. Antimicrob Agents Chemother 2014; 58:1451–7.
50. Leyenaar JK, Lagu T, Shieh MS, et al. Variation in resource utilization for the man-
agement of uncomplicated community-acquired pneumonia across community
and children’s hospitals. J Pediatr 2014; 165:585–91.
51. Williams DJ, Edwards KM, Self WH, et al. Antibiotic choice for children hospitalized
with pneumonia and adherence to national guidelines. Pediatrics 2015; 136:44–52.
52. Bradley JS, McCracken GH. Unique considerations in the evaluation of antibac-
terials in clinical trials for pediatric community-acquired pneumonia. Clin Infect
Dis 2008; 47(Suppl 3):S241–8.
53. Harris JA, Kolokathis A, Campbell M, et al. Safety and efficacy of azithromycin in
the treatment of community-acquired pneumonia in children. Pediatr Infect Dis
J 1998; 17:865–71.
Macrolides in Children With CAP: Panacea or Placebo? • JPIDS 2017:XX (XX XXXX) • 7
Downloaded from https://academic.oup.com/jpids/advance-article-abstract/doi/10.1093/jpids/pix083/4583480
by University of Tasmania Library user
on 04 January 2018
54. Wubbel L, Muniz L, Ahmed A, et  al. Etiology and treatment of communi-
ty-acquired pneumonia in ambulatory children. Pediatr Infect Dis J 1999;
18:98–104.
55. Aurangzeb B, Hameed A. Comparative efficacy of amoxicillin, cefuroxime and
clarithromycin in the treatment of community-acquired pneumonia in children.
J Coll Physicians Surg Pak 2003; 13:704–7.
56. Ambroggio L, Test M, Metlay JP, et al. Comparative effectiveness of beta-lactam
versus macrolide monotherapy in children with pneumonia diagnosed in the out-
patient setting. Pediatr Infect Dis J 2015; 34:839–42.
57. Ambroggio L, Taylor JA, Tabb LP, et  al. Comparative effectiveness of empiric
β-lactam monotherapy and β-lactam-macrolide combination therapy in chil-
dren hospitalized with community-acquired pneumonia. J Pediatr 2012;
161:1097–103.
58. Ambroggio L, Test M, Metlay JP, et  al. Beta-lactam versus beta- lactam/mac-
rolide therapy in pediatric outpatient pneumonia. Pediatr Pulmonol 2016;
51:541–8.