Invasive Bacterial Infections in Afebrile

Infants Diagnosed With Acute

Otitis Media
Son H. McLaren, MD, MS,a Andrea T. Cruz, MD, MPH,b Kenneth Yen, MD, MS,c Matthew J. Lipshaw, MD, MS,d
Kelly R. Bergmann, DO,e Rakesh D. Mistry, MD, MS,f Colleen K. Gutman, MD,g Fahd A. Ahmad, MD, MSCI,h
Christopher M. Pruitt, MD,i Graham C. Thompson, MD,j Matthew D. Steimle, DO,k Xian Zhao, MD,l Abigail M. Schuh, MD, MMHPE,m
Amy D. Thompson, MD,n Holly R. Hanson, MD, MS,o Stacey L. Ulrich, MD,p James A. Meltzer, MD, MS,q
Jennifer Dunnick, MD, MPH,r Suzanne M. Schmidt, MD,s Lise E. Nigrovic, MD, MPH,t Muhammad Waseem, MD, MS,u
Roberto Velasco, MD, PhD,v Samina Ali, MD,w Danielle L. Cullen, MD, MPH, MSHP,x Borja Gomez, MD, PhD,y Ron L. Kaplan, MD,z
Kajal Khanna, MD, JD,aa Jonathan Strutt, MD,ab Paul L. Aronson, MD, MHS,ac Ankita Taneja, MD,ad David C. Sheridan, MD,ae
Carol C. Chen, MD, MPH,af Amanda L. Bogie, MD,ag Aijin Wang, MS,ah Peter S. Dayan, MD, MSc,a ON BEHALF OF THE PEDIATRIC
EMERGENCY MEDICINE COLLABORATIVE RESEARCH COMMITTEE

To determine the prevalence of invasive bacterial infections (IBIs) and adverse

OBJECTIVES: abstract
events in afebrile infants with acute otitis media (AOM).
METHODS: We conducted a 33-site cross-sectional study of afebrile infants #90 days of age with
AOM seen in emergency departments from 2007 to 2017. Eligible infants were identified using
emergency department diagnosis codes and confirmed by chart review. IBIs (bacteremia and
meningitis) were determined by the growth of pathogenic bacteria in blood or cerebrospinal
fluid (CSF) culture. Adverse events were defined as substantial complications resulting from or
potentially associated with AOM. We used generalized linear mixed-effects models to identify
factors associated with IBI diagnostic testing, controlling for site-level clustering effect.
RESULTS: Of 5270 infants screened, 1637 met study criteria. None of the 278 (0%; 95%
confidence interval [CI]: 0%–1.4%) infants with blood cultures had bacteremia; 0 of 102 (0%;
95% CI: 0%–3.6%) with CSF cultures had bacterial meningitis; 2 of 645 (0.3%; 95% CI:
0.1%–1.1%) infants with 30-day follow-up had adverse events, including lymphadenitis (1)
and culture-negative sepsis (1). Diagnostic testing for IBI varied across sites and by age;
overall, 278 (17.0%) had blood cultures, and 102 (6.2%) had CSF cultures obtained.
Compared with infants 0 to 28 days old, older infants were less likely to have blood cultures
(P , .001) or CSF cultures (P , .001) obtained.
Afebrile infants with clinician-diagnosed AOM have a low prevalence of IBIs and
CONCLUSION:
adverse events; therefore, outpatient management without diagnostic testing may be
reasonable.

a WHAT’S KNOWN ON THIS SUBJECT: Guidelines do not provide recommendations

Department of Emergency Medicine, Vagelos College of Physicians and Surgeons, Columbia University, New York, about whether to evaluate for potential systemic illnesses in afebrile infants with
New York; bDepartment of Pediatrics, Baylor College of Medicine, Houston, Texas; cDepartment of Pediatrics, acute otitis media. The prevalence of invasive bacterial infection in these infants is
University of Texas Southwestern, Dallas, Texas; dDivision of Emergency Medicine, Cincinnati Children’s Hospital unclear and likely contributes to variation in care.
Medical Center and Department of Pediatrics, University of Cincinnati College of Medicine, Cincinnati, Ohio;
e WHAT THIS STUDY ADDS: Among afebrile infants aged 90 days and younger with
Department of Emergency Services, Children’s Minnesota, Minneapolis, Minnesota; fDepartment of Pediatrics,
clinician-diagnosed acute otitis media, invasive bacterial infections and adverse
School of Medicine, University of Colorado and Children’s Hospital Colorado, Aurora, Colorado; gDepartment of
events were rare. These data suggest that clinicians could minimize diagnostic
Pediatrics, Emory University, Atlanta, Georgia; hDepartment of Pediatrics, Washington University School of testing and hospitalizations for these infants.
Medicine, St. Louis, Missouri; iDepartment of Pediatrics, University of Alabama, Birmingham, Alabama;
j
Department of Pediatrics, University of Calgary and Alberta Children’s Hospital, Calgary, Alberta, Canada;
k
Department of Pediatrics, Division of Pediatric Emergency Medicine, School of Medicine, University of Utah, Salt To cite: McLaren SH, Cruz AT, Yen K, et al. Invasive
Lake City, Utah; lDepartment of Pediatrics, Division of Emergency Medicine, Children’s National Health System, Bacterial Infections in Afebrile Infants Diagnosed With
Washington, DC; mDepartment of Pediatrics, Medical College of Wisconsin, Milwaukee, Wisconsin;(Continued) Acute Otitis Media. Pediatrics. 2021;147(1):e20201571

Downloaded from www.aappublications.org/news at Indonesia:AAP Sponsored on January 6, 2021

PEDIATRICS Volume 147, number 1, January 2021:e20201571 ARTICLE
Acute otitis media (AOM) is a common departments (EDs) associated with fluid (CSF). All diagnoses of AOM
childhood infection, affecting .80% of the Pediatric Emergency Medicine were verified through the review of
children before 3 years of age.1 Collaborative Research Committee of clinician assessment in the medical
Although AOM is relatively uncommon the AAP. The participating sites records. Because the AAP diagnostic
in infants younger than 3 months of included 29 EDs in the United States, criteria for AOM are for children
age, once the diagnosis has been made, 2 EDs in Canada, and 2 EDs in Spain. 6 months and older and may be too
there is a dearth of evidence to guide All were university-affiliated teaching stringent to apply directly to this
further clinical care. National hospitals, and 25 (76%) were younger age group, we chose for our
recommendations for young infants freestanding children’s hospitals. The primary analysis to include all infants
with AOM do not exist because the study was approved by the Pediatric with documentation of $1 middle ear
current American Academy of Emergency Medicine Collaborative examination abnormality supportive
Pediatrics (AAP) guideline on the Research Committee Steering of AOM diagnosis.2 However, we
diagnosis and management of AOM Committee and the research ethics performed a subanalysis of infants
excludes infants younger than boards of all participating sites. meeting simplified AAP diagnostic
6 months of age.2 criteria, defined as the presence of
We included infants aged 90 days and
tympanic membrane erythema,
One important clinical concern in younger who presented to
bulging tympanic membrane, or
infants younger than 3 months with a participating ED between January
otorrhea not due to acute otitis
AOM is whether they have 2007 and December 2017 and were
externa.
concomitant invasive bacterial clinically diagnosed with AOM. Study
infections (IBIs). In previous studies, period varied by site depending on
researchers have suggested a low available data. We excluded infants Study Protocol
prevalence of bacteremia among with documented temperatures of
All study variables were defined
infants with AOM, but these studies $38.0°C or ,36.0°C in the ED or
a priori and described in the manual
were limited by small sample sizes within 48 hours before ED arrival,
of operations. Investigators and
and included mixed populations of antibiotic use other than topical
research coordinators received
infants with and without fever.3–8 The antibiotics within 48 hours of ED
standardized data abstraction
clinical conundrum of AOM in infants presentation, concurrent diagnosis of
training and entered their data
younger than 3 months may be most mastoiditis, evidence of focal
electronically into REDCap, a secure,
relevant to those without fever, for bacterial infection on ED examination,
web-based electronic database.
whom the appropriate diagnostic or who were transferred to the ED
Collected data included demographic
evaluation for IBI, if any, is unclear. with diagnostic testing completed
characteristics, medical history,
and/or antibiotic administration
The relative lack of data on afebrile presenting symptoms, ED physical
initiated at an outside hospital. We
infants with AOM has likely resulted examination, laboratory data, findings
chose to exclude febrile infants from
in variability among clinicians in their on assessment by otolaryngologist
our study because the presence of
diagnostic approach. If IBIs and consultants when available, and
fever in young infants, and not
adverse events are shown to be rare inpatient management for
necessarily the diagnosis of AOM, is
in a large sample of afebrile infants, hospitalized infants.
a primary driver for more expanded
clinicians may be more apt to obtain testing and/or empirical treatment
fewer diagnostic tests. Our primary To minimize bias associated with
of IBI.
aim was to determine the prevalence abstracting potentially more-
of IBIs and AOM-associated adverse Eligible infants were identified using subjective descriptions from the
events in afebrile infants 90 days and two methods. The primary method medical records, we provided
younger with clinically diagnosed was using an inclusive list of restrictive key words to guide the
AOM. Our secondary aim was to International Classification of Diseases determination of ill appearance,
describe patterns of diagnostic (ICD) 9th or 10th Revision diagnosis respiratory distress, and clinical
testing and factors associated with codes for AOM (Supplemental dehydration. These findings were
testing in these afebrile infants. Table 6). To assess for infants with categorized as present, not present,
IBI who also had AOM but for whom or unclear. To assess interrater
ICD codes for AOM were not reliability, a second abstractor at each
METHODS documented, we also retrieved and site reviewed the medical records and
reviewed the records of all infants performed an independent
Study Design and Population evaluated in the ED for whom review assessment of clinical appearance,
We conducted a cross-sectional study of microbiology databases noted respiratory status, and hydration
at 33 pediatric emergency bacteria in the blood or cerebrospinal status for a random sample of 10% of

Downloaded from www.aappublications.org/news at Indonesia:AAP Sponsored on January 6, 2021

2 MCLAREN et al
all infants, as well as for all infants (2) ED completion of CSF culture we used generalized linear mixed-
with IBI or adverse outcomes. (binary), and (3) hospitalization effects models to account for site-
(binary) from the index ED visit. Each level clustering effects. For each
Outcome Measures of these outcomes was stratified by outcome, we first examined the
The primary outcomes were IBIs and age and by site to assess for variation association between the outcome and
AOM-associated adverse events. IBIs in care. a potential explanatory variable,
included bacteremia and bacterial including those from infant
meningitis, which were defined as the Sample Size characteristics, clinical symptoms,
growth of a pathogen in the blood Because we expected IBI to be a rare and physical examination findings,
culture or CSF culture, respectively event in this sample, we determined using a univariable generalized linear
(Supplemental Table 7).9 Any our sample size on the basis of the mixed-effects model. For the
bacteria not predefined as prevalence of blood culture testing, subsequent multivariable models, we
a contaminant or pathogen, or whose a secondary outcome. Targeting first assessed for potential
categorization of pathogenicity by the evaluation of up to 20 independent collinearity between predictor
treating clinician was unclear or variables associated with this variables using generalized variance
different from the prespecified list, outcome, we aimed to study at least inflation factors. Those predictors
were reviewed by a pediatric 200 infants for whom blood cultures with univariable P value ,.1 and no
infectious disease coinvestigator were obtained. Pilot testing at the evidence of multicollinearity were
(A.C.) to determine pathogenicity. For lead site showed that 30% of eligible included in multivariable models to
those infants who did not have CSF infants had blood cultures obtained; assess for an association with the
obtained for culture, the presence or therefore, we aimed to enroll following outcomes as fixed effects:
absence of bacterial meningitis was a minimum of 667 infants. (1) obtaining a blood culture, (2)
determined through the review of the obtaining a CSF culture, and (3)
medical records. If the infant had Statistical Analysis hospitalization. Hospital sites were
a follow-up visit to the index hospital used as random effects.
To enhance clinical interpretability,
within 30 days of the ED visit that did several continuous variables were Interrater reliability for the
not identify bacterial meningitis, we categorized. Age was categorized as assessment of general appearance,
considered this outcome as negative. 0 to 28, 29 to 56, and 57 to 90 days respiratory distress, and dehydration
AOM-associated adverse events were based on cutoffs generally used when was measured by using Cohen
defined as any substantial risk-stratifying febrile infants for unweighted k, a 0 to 1 scale where
complications resulting from or IBI.10,11 The results of tests generally 0 indicates poor agreement and 1
potentially associated with AOM, available during the ED stay were indicates perfect agreement.16 All
including, but not limited to, the categorized as not performed, statistical analyses were conducted
development of mastoiditis, sepsis, or performed and within normal limits, using either the Statistical Program
death. Urinary tract infection (UTI) and performed and abnormal. For the for the Social Sciences Version 26
was not considered an AOM- complete blood count, a white blood (IBM SPSS Statistics, IBM
associated adverse event because the cell (WBC) count between 5 and 15 3 Corporation, Armonk, NY) or R (R
causal mechanism linking AOM and 103 cells/mL was considered to be version 3.6.3, 2019).
UTI is unclear. However, we collected normal; values outside of this range
information about urine testing were considered abnormal.12 RESULTS
because it is a part of the serious Urinalysis was considered abnormal
We screened 5270 infants for
bacterial infection evaluation in if there was any leukocyte esterase,
eligibility; of these, 1637 (31.1%) met
infants. For infants discharged from nitrite, or .5 WBCs per high-power
study criteria (Fig 1). Sites
the ED, we reviewed the medical field, similar to previous studies.13,14
contributed 0 (1 site) to 386 cases,
records to identify ED return visits Given the challenges of interpreting
with a median of 30 cases across
within 72 hours. Additionally, we CSF WBC count if the lumbar
sites, and ranging from 5 to 11 years
reviewed the medical records for any puncture was traumatic,15 these
of review.
outpatient clinician encounter or results were not categorized.
hospitalization within 30 days of ED Patient Characteristics
We described the prevalence of IBI,
discharge. For any visit identified, we
adverse events, and diagnostic tests The median age of included infants
evaluated the records for evidence of
obtained with proportions and 95% was 68 days (interquartile range
IBIs or adverse events.
confidence intervals [CIs]. For all [IQR]: 49–80 days) (Table 1). One
Our secondary outcomes were (1) ED tests of association regarding practice thousand four hundred fifty-nine
completion of blood culture (binary), variation (eg, testing, hospitalization), (89.1%) infants met the simplified

Downloaded from www.aappublications.org/news at Indonesia:AAP Sponsored on January 6, 2021

PEDIATRICS Volume 147, number 1, January 2021 3
AAP diagnostic criteria for AOM. The
remaining 178 infants had $1 of the
following examination findings
reported as supporting evidence for
AOM: opacification of tympanic
membrane (n = 113), dull tympanic
membrane (n = 57), decreased
visualization of middle ear structures
(n = 25), middle ear effusion (n = 9),
visualized tympanic membrane
perforation (n = 8), or decreased
tympanic membrane mobility with
insufflation (n = 5).

Prevalence of IBI and

AOM-Associated Adverse Events
None of the 278 infants with blood FIGURE 1
cultures had bacteremia (0 of 278; Patient screening and identification. aOne infant had a positive CSF culture result, which was
0%, 95% CI: 0%–1.4%). No infant clinically verified by the infectious disease expert after inclusion in the study to be a contaminant
(coagulase-negative staphylococci).
was diagnosed with bacterial
meningitis on the basis of CSF culture
results (0 of 102; 0%, 95% CI:
0%–3.6%) or combined with 30-day TABLE 1 Characteristics of Afebrile Infants With AOM (N = 1637)
follow-up (0 of 672; 0%, 95% CI: Demographic
0%–0.6%). Two of 645 infants (0.3%; Median age, d (IQR) 68 (49–80)
95% CI: 0.1%–1.1%) with 30-day Age, d, n (%)
follow-up or hospitalization history 0–28 100 (6.1)
experienced adverse events. One of 29–56 444 (27.1)
57–90 1093 (66.8)
these infants was treated for culture-
Male, n (%) 957 (58.5)
negative sepsis, a potentially AOM- Ethnicity, n (%)
related adverse event, although on Hispanic 595 (36.3)
review of medical records, it was Not Hispanic 859 (52.5)
suggested that the primary cause of Unknown or other 183 (11.2)
her illness was likely severe Term gestation, n (%) 1217 (74.3)
Chronic illnessa, n (%) 49 (3.0)
dehydration from underlying milk History of presenting illness, n (%)
protein allergy. In Table 2, we History of tactile warmth 257 (15.7)
describe the 2 infants with adverse Measured fever .2 d before ED visit 92 (5.6)
events in detail. In Supplemental Ear discharge 422 (25.8)
Difficulty feeding 429 (26.2)
Table 8, we describe the 2 infants
Vomiting 276 (16.9)
who were subsequently diagnosed Diarrhea 119 (7.3)
with UTIs. Decreased urine output 96 (5.9)
Symptoms of upper respiratory infection 1179 (72.0)
ED Management General examination, n (%); kb
Ill appearancek 30 (1.8); 0.56
Diagnostic testing and hospitalization
Respiratory distressk 75 (4.6); 0.44
rates varied by site (Supplemental Dehydrationk 28 (1.7); 0.21
Fig 2). One-fifth of all infants (355 of Ear examination $1 ear, n (%)
1637, 21.7%) had $1 diagnostic test Tympanic membrane erythema 970 (59.3)
obtained for infectious illness Bulging tympanic membrane 564 (34.5)
Otorrhea 477 (29.1)
(Table 3). One-third of infants
28 days and younger underwent All values are frequency (%) except where otherwise indicated.
a Includes congenital heart disease (19), major congenital anomaly other than congenital heart disease (10), renal or
lumbar puncture (34/100; 34.0%). Of urologic disorder (6), failure to thrive (4), disease or use of medications that would affect the immune system (3),
1179 infants with symptoms of upper neurologic disorder (3), chronic lung disease (1), presence of indwelling catheters or shunts (1), and other (7). Does not
respiratory tract infection, 58 (4.9%) total 49 because an infant may have .1 chronic illness.
b Cohen unweighted k (interrater reliability) was used to measure interrater reliability of 2 assessors’ determination of
underwent lumbar puncture, and 162 these findings. Variables with k of 0.21 to 0.40 were considered to have fair agreement, those with k of 0.41 to 0.60 were
(13.7%) had blood cultures obtained. considered to have moderate agreement, and those with k of 0.61 to 0.80 were considered to have substantial agreement.

Downloaded from www.aappublications.org/news at Indonesia:AAP Sponsored on January 6, 2021

4 MCLAREN et al
TABLE 2 Characteristics of Infants With Adverse Events
Patient Age, d Ill Appearance Culture Results Adverse Event Clinical Course
1 45 No Blood: negative, CSF: negative, urine: Lymphadenitis Initially discharged from the hospital on amoxicillin. Twenty days
negative, MEE: not obtained after index ED visit, returned with cervical lymphadenitis and
perforated AOM, confirmed by an otolaryngologist. Hospitalized
and treated with IV and otic antibiotics.
2 32 Yes Blood: negative, CSF: negative, urine: Culture-negative Presented with severe illness during the index ED visit. AOM
negative, MEE: Staphylococcus sepsis confirmed by an otolaryngologist. Source of culture-negative
aureus sepsis thought to be severe milk protein allergy. Hospitalized and
treated with IV and otic antibiotics.
IV, intravenous; MEE, middle ear effusion or ear discharge.

One hundred seventy-three infants visit with lymphadenitis, as described testing and hospitalization. Decreased
(10.6%) received either an in Table 2. urination, respiratory distress, and
intravenous or intramuscular completion of complete blood count
antibiotic in the ED (Supplemental Factors Associated With Diagnostic (regardless of whether the results
Testing and Hospitalization
Table 9). Of the 1450 infants were normal or abnormal) were also
discharged from the ED (88.6% of No potential predictor variables independently associated with
total), 1311 (90.4%) received assessed for inclusion in the hospitalization. For the subanalysis of
a prescription for an oral antibiotic. generalized linear mixed-effects infants meeting the simplified AAP
Amoxicillin was the most-frequently models were collinear (generalized diagnostic criteria, minor differences
prescribed antibiotic (n = 1228; variance inflation factors ,3) in the point estimates of the predictor
93.7%), followed by amoxicillin and (Supplemental Table 11). Adjusting variables were observed for each
clavulanate (n = 56; 4.3%). for covariates, older infants were less model (see Tables 4 and 5,
likely to have blood cultures sent Supplemental Table 12).
The 72-hour return rate was 4.3%. Of (Table 4), undergo lumbar puncture
the returning 63 infants, 15 (23.8%) (Supplemental Table 12), or be
were hospitalized for related reasons hospitalized (Table 5) as compared DISCUSSION
(median age: 49 days, IQR: 37–66) with infants 0 to 28 days old. In all 3 In this international, multicenter
(Supplemental Table 10). One patient models, history of ear discharge was study of afebrile infants aged 90 days
returned 3 weeks after the initial ED significantly associated with IBI and younger with clinically diagnosed
AOM, the prevalence of IBI and AOM-
TABLE 3 Summary of ED Management associated adverse events was low.
N (%) Despite the low probability for IBI in
Overall (N = 1637) 0–28 d Old 29–56 d Old 57–90 d Old this population, more than one-fifth
(n = 100) (n = 444) (n = 1093) underwent IBI diagnostic testing and
Diagnostic testing were hospitalized. This practice
Any testing for bacterial infectiona 355 (21.7) 58 (58.0) 177 (39.9) 120 (11.0) varied by site and was largely driven
Complete blood count 311 (19.0) 54 (54.0) 164 (36.9) 93 (8.5) by age, with younger infants more
Blood culture 278 (17.0) 53 (53.0) 147 (33.1) 78 (7.1) likely to both undergo invasive
Urine culture 207 (12.6) 46 (46.0) 102 (23.0) 59 (5.4)
testing and be hospitalized. With the
CSF culture 102 (6.2) 34 (34.0) 58 (13.1) 10 (0.9)
Respiratory pathogen testb 161 (9.8) 27 (27.0) 46 (10.4) 88 (8.1) data from our study, we suggest that
Consultations given the low rates of IBI and adverse
Otolaryngologist 64 (3.9) 8 (8.0) 40 (9.0) 16 (1.5) events, outpatient management
Treatment without IBI testing is reasonable for
IV or IM antibiotics 175 (10.7) 37 (37.0) 86 (19.4) 52 (4.8)
most afebrile infants with a clinical
Prescription for oral antibioticc 1311 (90.4) 37 (71.2) 299 (81.3) 975 (94.7)
Disposition diagnosis of AOM.
Discharged from the ED 1450 (88.5) 52 (52.0) 368 (82.9) 1030 (94.2)
Hospitalized 186 (11.4) 47 (47.0) 76 (17.1) 63 (5.8) Our data may be used to help guide
Transferred to another hospital 1 (0.1) 1 (1.0) 0 (0) 0 (0) clinical management of afebrile
IM, intramuscular; IV, intravenous. infants with clinician-diagnosed AOM,
a Defined as obtaining $1 of the following tests: complete blood count, blood culture, CSF culture, or urine culture. who are not included in the current
AAP AOM practice guideline.2 We
b Includes any respiratory testing, including point-of-care respiratory syncytial virus and influenza testing, as well viral

pathogen panels.
c Percentage reflects total out of infants discharged from the ED (1311 out of 1450 for all ages, 37 out of 52 for 0–28 days, expected variation in IBI testing and
299 out of 368 for 29–56 days, and 975 out of 1030 for 57–90 days). were not surprised that young age

Downloaded from www.aappublications.org/news at Indonesia:AAP Sponsored on January 6, 2021

PEDIATRICS Volume 147, number 1, January 2021 5
TABLE 4 Generalized Linear Mixed-Effects Model to Predict Blood Culture Testing risk of IBI, the concern for IBI in
Predictor Adjusted Odds Ratio (95% CI) P younger infants based on data from
Demographic
the febrile infant population, and the
Agea ,.001b reluctance to initiate antibiotics
29–56 d 0.37 (0.23–0.60) ,.001b without any IBI evaluation. However,
57–90 d 0.06 (0.04–0.10) ,.001b it is suggested by our data and
Gestational agec .01b previous data that the risk of IBI is
,37 weeks’ gestation 1.60 (0.94–2.72) .08
Unknown 0.59 (0.36–0.96) .03b
low in afebrile infants with AOM.4–6
History of presenting illness Additionally, IBI prevalence in
Ear discharge 2.32 (1.62–3.34) ,.001b afebrile infants with AOM appears
Tactile warmth 2.29 (1.53–3.43) ,.001b lower than the 0.8% to 2.5%
Upper respiratory infection symptomsd 0.88 (0.62–1.26) .50 prevalence described in the febrile
a Reference group: 0 to 28 d old. infant AOM population.4,5
b Statistical significance (P , .05). In this model, age, gestational age, history of ear discharge, and history of tactile fever
were independently associated with obtaining a blood culture. In the subanalysis including only the 1459 infants meeting Of note, approximately three-fourths
simplified AAP diagnostic criteria (data available on request), no meaningful changes were noted, except the predictor
variable ,37 wk gestational age was significant (P = .04).
of the infants in our study were
c Reference group: $37 wk gestation. reported to have symptoms of upper
d Include cough, new nasal congestion or sneezing, rhinorrhea, wheezing, noisy breathing, and difficulty breathing.
respiratory infections, which may
lead to viral AOM.17 Inclusion of these
infants, who may have a lower
appeared to be the major driver increased testing in the youngest
likelihood of IBI than those with
leading to evaluation for IBI with infants is likely related to the lack of
bacterial AOM, might have led to an
either blood or CSF testing. The data regarding how AOM modifies the
underestimation of IBI prevalence.
However, existing data do not provide
TABLE 5 Generalized Linear Mixed-Effects Model to Predict Hospitalization clarity regarding the ability to
distinguish viral from bacterial AOM
Predictor Adjusted Odds Ratio (95% CI) P
without performing tympanocentesis.
Demographic Because .85% of older infants and
Agea ,.001b
29–56 d 0.11 (0.06–0.22) ,.001b
children with clinically diagnosed
57–90 d 0.07 (0.03–0.14) ,.001b AOM have identifiable bacterial
Gestational agec .02b otopathogens,18 which could
,37 weeks’ gestation 1.64 (0.80–3.38) .18 potentially spread to the blood or
Unknown 0.38 (0.17–0.87) .02b CSF, it is understandable why
History of chronic illness 2.27 (0.85–6.07) .10
History of presenting illness
clinicians would manage infants with
Decreased urine output 4.13 (1.88–9.07) ,.001b AOM conservatively, regardless of the
Ear discharge 1.97 (1.17–3.31) .01b presence of concurrent viral illnesses.
New difficulty feeding 1.36 (0.81–2.30) .25
Diarrhea 0.58 (0.21–1.59) .29 One major challenge in any study of
Physical examination infants with AOM is ensuring that
Respiratory distress 40.54 (19.17–85.70) ,.001b AOM is indeed present. The diagnosis
Dehydration 2.89 (0.87–9.61) .08 of AOM is unquestionably difficult.
Ill appearance 1.64 (0.52–5.18) .40
Diagnostic test
However, in this young age group,
Complete blood countd ,.001b clinicians may be less inclined to make
Completed and within normal limits 10.95 (5.95–20.15) ,.001b this diagnosis without having clear
Completed and abnormal 9.25 (4.04–21.14) ,.001b evidence of AOM on examination,
Urinalysise .05 because once the diagnosis is made,
Completed and within normal limits 2.11 (1.15–3.87) .02
Completed and abnormal 2.14 (0.49–9.26) .31
they must make difficult decisions
a
regarding the need for invasive
Reference group: 0 to 28 d old.
b mStatistical significance (P , .05). In this model, age, gestational age, history of decreased urine output, history of ear diagnostic tests, systemic antibiotics,
discharge, respiratory distress on examination, and completion of complete blood count were independently associated and hospitalization. In previous
with hospitalization. In the subanalysis including only the 1459 infants meeting simplified AAP diagnostic criteria (data studies, researchers have used the
available on request), the following differences were observed: chronic illness was not included in the model (univariable
P . .1) and dehydrated appearance was statistically significant in the regression model (P = .02). examination by otolaryngologists, who
c Reference group: $37 wk gestation. often use surgical microscopes and/or
d Reference group: complete blood count not obtained. Test was considered abnormal if the WBC count was ,5 or .15 3
tympanocentesis results to support
103 cells/mL.
e Reference group: urinalysis not obtained. Test was considered abnormal if any of the following was present: leukocyte the clinical diagnosis of AOM.3,4,6,7
esterase, nitrite, or .5 WBCs per high-power field. However, these methods are not

Downloaded from www.aappublications.org/news at Indonesia:AAP Sponsored on January 6, 2021

6 MCLAREN et al
routinely used in current clinical except one site had the ability to AOM-associated adverse events
practice.19 Although examining the review some or all of the outpatient were uncommon. Diagnostic
ears of young infants is technically records of pediatric offices affiliated testing and hospitalization rates
challenging, previous researchers have with the hospital, augmenting our varied by site and were substantial
demonstrated successful use of ability to capture those infants in contrast to the low prevalence
otoscopy to diagnose AOM in infants presenting in the outpatient setting. of IBIs and adverse events. On
younger than 8 weeks, with 52% to We did not make similar assumptions the basis of these findings,
85% of middle ear effusion cultures regarding bacteremia, which may be outpatient management
yielding growth of true otopathogens transient or respond to standard without diagnostic testing for IBI
after tympanocentesis.4,20 therapy for AOM. Third, we may have may be reasonable for most
Our study had limitations. First, given missed infants whose discharge codes afebrile infants with AOM.
its retrospective design, we were did not include AOM, but we
unable to ensure completeness and addressed this concern by screening
accuracy of clinical data. Interrater all infants with positive blood or CSF ACKNOWLEDGMENTS
agreement for appearance of cultures for missed AOM diagnosis. We thank the faculty mentors
dehydration was only fair but was Fourth, although we identified factors and research coordinators across
higher for general appearance and associated with testing and the sites who assisted with
respiratory distress. Second, not all hospitalization, these management data review and entry. We also thank
infants had IBI testing, potentially decisions may have been driven by the PEM CRC Steering Committee for
leading to an underestimation of IBI factors that are unable to be captured their guidance.
prevalence. However, we minimized retrospectively and not primarily by
misclassification of infants with the diagnosis of AOM. Our study
meningitis by reviewing any available included only a relatively small
clinician encounter within 30 days for number of infants #28 days, which ABBREVIATIONS
discharged infants. We recognize that may reflect the challenges of AAP: American Academy of
the use of return visits and medical diagnosis or lower prevalence of AOM Pediatrics
record reviews as a proxy for in this age group. Thus, the AOM: acute otitis media
identifying those with bacterial conclusions of our study should most CI: confidence interval
meningitis may not fully capture any readily be applied to infants older CSF: cerebrospinal fluid
events occurring outside of the index than 28 days. Finally, our findings are ED: emergency department
EDs and hospital systems. However, not generalizable to febrile infants. IBI: invasive bacterial infection
the majority of the participating EDs ICD: International Classification of
are referral centers for the sickest Diseases
children in their respective regions, CONCLUSIONS UTI: urinary tract infection
making presentation to an alternative In this cohort of afebrile infants with WBC: white blood cell
hospital less likely. Additionally, all clinician-diagnosed AOM, IBIs and

n
Department of Pediatrics, Alfred I. duPont Hospital for Children, Wilmington, Delaware; oDepartment of Pediatrics, Monroe Carell Jr. Children’s Hospital at
Vanderbilt, Nashville, Tennessee; pDepartment of Pediatrics, Rady Children’s Hospital San Diego, San Diego, California; qDepartment of Pediatrics, Jacobi Medical
Center, Bronx, New York; rDepartment of Pediatrics, University of Pittsburgh Medical Center Children’s Hospital of Pittsburgh, Pittsburgh, Pennsylvania; sDepartment
of Pediatrics, Feinberg School of Medicine, Northwestern University, Chicago, Illinois; tDivision of Emergency Medicine, Boston Children’s Hospital, Boston,
Massachusetts; uDepartment of Pediatrics and Emergency Medicine, Lincoln Medical Center, Bronx, New York; vPediatric Emergency Unit, Rio Hortega University
Hospital, Valladolid, Spain; wDepartment of Pediatrics, Faculty of Medicine and Dentistry, University of Alberta, Edmonton, Alberta, Canada; xDepartment of
Pediatrics, Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania; yPediatric Emergency Department, Cruces University Hospital, Bilbao, Spain; zDepartment
of Pediatrics, School of Medicine, University of Washington and Seattle Children’s Hospital, Seattle, Washington; aaDepartment of Emergency Medicine, Stanford
University, Stanford, California; abDepartment of Pediatrics, University of Minnesota, Minneapolis, Minnesota; acDepartments of Pediatrics and Emergency Medicine,
Yale School of Medicine, Yale University, New Haven, Connecticut; adDepartment of Pediatrics, University of Florida, Jacksonville, Jacksonville, Florida; aeDepartment
of Emergency Medicine and Pediatrics, Oregon Health and Science University, Portland, Oregon; afDepartment of Emergency Medicine, University of California San
Francisco, San Francisco, California; agDepartment of Pediatrics, University of Oklahoma, Oklahoma City, Oklahoma; and ahDepartment of Biostatistics, Mailman
School of Public Health, Columbia University, New York, New York
Dr Pruitt’s current affiliation is with Department of Pediatrics, Medical University of South Carolina, Charleston, SC.
Drs McLaren and Dayan conceptualized and designed the study, designed the data collection instruments, coordinated and supervised data collection and transfer
from other sites, performed data analyses, drafted the initial manuscript, and reviewed and revised the manuscript; Dr Cruz participated in the study design,
collected local data, and reviewed and revised the manuscript; Drs Yen, Lipshaw, Bergmann, Mistry, Gutman, Ahmad, Pruitt, Thompson, Steimle, Zhao, Schuh,

Downloaded from www.aappublications.org/news at Indonesia:AAP Sponsored on January 6, 2021

PEDIATRICS Volume 147, number 1, January 2021 7
Thompson, Hanson, Ulrich, Meltzer, Dunnick, Schmidt, Nigrovic, Waseem, Velasco, Ali, Cullen, Gomez, Kaplan, Khanna, Strutt, Aronson, Taneja, Sheridan, Chen, and
Bogie collected data at their sites and critically reviewed and revised the manuscript; Ms Wang performed data analysis and reviewed and revised the manuscript;
and all authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.
DOI: https://doi.org/10.1542/peds.2020-1571
Accepted for publication Sep 24, 2020
Address correspondence to Son H. McLaren, MD, MS, Department of Emergency Medicine, Columbia University Vagelos College of Physicians and Surgeons, 622 West
168th Street, Suite VC-260, New York, NY 10032. E-mail: [email protected]
PEDIATRICS (ISSN Numbers: Print, 0031-4005; Online, 1098-4275).
Copyright © 2021 by the American Academy of Pediatrics
FINANCIAL DISCLOSURE: Dr Cruz is an Associate Editor and Dr Aronson is on the Editorial Board of Pediatrics. Dr Nigrovic is an Associate Editor of Annals of
Emergency Medicine; the other authors have indicated they have no financial relationships relevant to this article to disclose
FUNDING: No external funding.
POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.

REFERENCES
1. Teele DW, Klein JO, Rosner B. pneumoniae. Int J Pediatr younger. Pediatrics. 2018;141(2):
Epidemiology of otitis media during Otorhinolaryngol. 2013;77(6):976–980 e20173068
the first seven years of life in children 14. Schnadower D, Kuppermann N, Macias
8. Berkun Y, Nir-Paz R, Ami AB, Klar A,
in greater Boston: a prospective, CG, et al.; American Academy of
Deutsch E, Hurvitz H. Acute otitis media
cohort study. J Infect Dis. 1989;160(1):
in the first two months of life: Pediatrics Pediatric Emergency
83–94 Medicine Collaborative Research
characteristics and diagnostic
2. Lieberthal AS, Carroll AE, Chonmaitree difficulties. Arch Dis Child. 2008;93(8): Committee. Febrile infants with urinary
T, et al. The diagnosis and management 690–694 tract infections at very low risk for
of acute otitis media [published adverse events and bacteremia.
9. Thomson J, Cruz AT, Nigrovic LE, et al.;
correction appears in Pediatrics. 2014; Pediatrics. 2010;126(6):1074–1083
Pediatric Emergency Medicine
133(2):346]. Pediatrics. 2013;131(3). 15. Lyons TW, Cruz AT, Freedman SB, et al.;
Collaborative Research Committee
Available at: www.pediatrics.org/cgi/ Pediatric Emergency Medicine Clinical
(PEM CRC) HSV Study Group.
content/full/131/3/e964 Research Network (PEM CRC) Herpes
Concomitant bacterial meningitis in
3. Burton DM, Seid AB, Kearns DB, Pransky infants with urinary tract infection. Simplex Virus Study Group.
SM. Neonatal otitis media. An update. Pediatr Infect Dis J. 2017;36(9):908–910 Interpretation of cerebrospinal fluid
Arch Otolaryngol Head Neck Surg. 1993; white blood cell counts in young infants
10. Scarfone R, Murray A, Gala P, Balamuth
119(6):672–675 with a traumatic lumbar puncture. Ann
F. Lumbar puncture for all febrile
Emerg Med. 2017;69(5):622–631
4. Nozicka CA, Hanly JG, Beste DJ, Conley infants 29–56 days old: a retrospective
SF, Hennes HM. Otitis media in infants cohort reassessment study. J Pediatr. 16. Landis JR, Koch GG. The measurement
aged 0–8 weeks: frequency of 2017;187:200–205.e1 of observer agreement for categorical
associated serious bacterial disease. data. Biometrics. 1977;33(1):159–174
11. Aronson PL, Thurm C, Alpern ER, et al.;
Pediatr Emerg Care. 1999;15(4):252–254 17. Chonmaitree T, Revai K, Grady JJ, et al.
Febrile Young Infant Research
5. Turner D, Leibovitz E, Aran A, et al. Acute Collaborative. Variation in care of the Viral upper respiratory tract infection
otitis media in infants younger than febrile young infant ,90 days in US and otitis media complication in young
two months of age: microbiology, pediatric emergency departments. children. Clin Infect Dis. 2008;46(6):
clinical presentation and therapeutic Pediatrics. 2014;134(4):667–677 815–823
approach. Pediatr Infect Dis J. 2002; 18. Kaur R, Morris M, Pichichero ME.
12. Jaskiewicz JA, McCarthy CA, Richardson
21(7):669–674 Epidemiology of acute otitis media in
AC, et al.; Febrile Infant Collaborative
6. Sakran W, Makary H, Colodner R, et al. Study Group. Febrile infants at low risk the postpneumococcal conjugate
Acute otitis media in infants less than for serious bacterial infection–an vaccine era. Pediatrics. 2017;140(3):
three months of age: clinical appraisal of the Rochester criteria and e20170181
presentation, etiology and concomitant implications for management. 19. Wasserman RC, Gerber JS. Acute otitis
diseases. Int J Pediatr Pediatrics. 1994;94(3):390–396 media in the 21st century: what now?
Otorhinolaryngol. 2006;70(4):613–617 Pediatrics. 2017;140(3):e20171966
13. Tzimenatos L, Mahajan P, Dayan PS,
7. Sommerfleck P, González Macchi ME, et al.; Pediatric Emergency Care Applied 20. Bland RD. Otitis media in the first six
Pellegrini S, et al. Acute otitis media in Research Network (PECARN). Accuracy weeks of life: diagnosis, bacteriology,
infants younger than three months not of the urinalysis for urinary tract and management. Pediatrics. 1972;
vaccinated against Streptococcus infections in febrile infants 60 days and 49(2):187–197

Downloaded from www.aappublications.org/news at Indonesia:AAP Sponsored on January 6, 2021

8 MCLAREN et al
 Invasive Bacterial Infections in Afebrile Infants Diagnosed With Acute Otitis
Media
Son H. McLaren, Andrea T. Cruz, Kenneth Yen, Matthew J. Lipshaw, Kelly R.
Bergmann, Rakesh D. Mistry, Colleen K. Gutman, Fahd A. Ahmad, Christopher M.
Pruitt, Graham C. Thompson, Matthew D. Steimle, Xian Zhao, Abigail M. Schuh,
Amy D. Thompson, Holly R. Hanson, Stacey L. Ulrich, James A. Meltzer, Jennifer
Dunnick, Suzanne M. Schmidt, Lise E. Nigrovic, Muhammad Waseem, Roberto
Velasco, Samina Ali, Danielle L. Cullen, Borja Gomez, Ron L. Kaplan, Kajal
Khanna, Jonathan Strutt, Paul L. Aronson, Ankita Taneja, David C. Sheridan, Carol
C. Chen, Amanda L. Bogie, Aijin Wang, Peter S. Dayan and ON BEHALF OF THE
PEDIATRIC EMERGENCY MEDICINE COLLABORATIVE RESEARCH
COMMITTEE
Pediatrics 2021;147;
DOI: 10.1542/peds.2020-1571 originally published online December 7, 2020;

Updated Information & including high resolution figures, can be found at:
Services http://pediatrics.aappublications.org/content/147/1/e20201571
References This article cites 19 articles, 8 of which you can access for free at:
http://pediatrics.aappublications.org/content/147/1/e20201571#BIBL
Subspecialty Collections This article, along with others on similar topics, appears in the
following collection(s):
Ear, Nose & Throat Disorders
http://www.aappublications.org/cgi/collection/ear_nose_-_throat_dis
orders_sub
Otitis Media
http://www.aappublications.org/cgi/collection/otitis_media_sub
Emergency Medicine
http://www.aappublications.org/cgi/collection/emergency_medicine_
sub
Permissions & Licensing Information about reproducing this article in parts (figures, tables) or
in its entirety can be found online at:
http://www.aappublications.org/site/misc/Permissions.xhtml
Reprints Information about ordering reprints can be found online:
http://www.aappublications.org/site/misc/reprints.xhtml

Downloaded from www.aappublications.org/news at Indonesia:AAP Sponsored on January 6, 2021

 Invasive Bacterial Infections in Afebrile Infants Diagnosed With Acute Otitis
Media
Son H. McLaren, Andrea T. Cruz, Kenneth Yen, Matthew J. Lipshaw, Kelly R.
Bergmann, Rakesh D. Mistry, Colleen K. Gutman, Fahd A. Ahmad, Christopher M.
Pruitt, Graham C. Thompson, Matthew D. Steimle, Xian Zhao, Abigail M. Schuh,
Amy D. Thompson, Holly R. Hanson, Stacey L. Ulrich, James A. Meltzer, Jennifer
Dunnick, Suzanne M. Schmidt, Lise E. Nigrovic, Muhammad Waseem, Roberto
Velasco, Samina Ali, Danielle L. Cullen, Borja Gomez, Ron L. Kaplan, Kajal
Khanna, Jonathan Strutt, Paul L. Aronson, Ankita Taneja, David C. Sheridan, Carol
C. Chen, Amanda L. Bogie, Aijin Wang, Peter S. Dayan and ON BEHALF OF THE
PEDIATRIC EMERGENCY MEDICINE COLLABORATIVE RESEARCH
COMMITTEE
Pediatrics 2021;147;
DOI: 10.1542/peds.2020-1571 originally published online December 7, 2020;

The online version of this article, along with updated information and services, is
located on the World Wide Web at:
http://pediatrics.aappublications.org/content/147/1/e20201571

Data Supplement at:

http://pediatrics.aappublications.org/content/suppl/2020/12/04/peds.2020-1571.DCSupplemental

Pediatrics is the official journal of the American Academy of Pediatrics. A monthly publication, it
has been published continuously since 1948. Pediatrics is owned, published, and trademarked by
the American Academy of Pediatrics, 345 Park Avenue, Itasca, Illinois, 60143. Copyright © 2021
by the American Academy of Pediatrics. All rights reserved. Print ISSN: 1073-0397.

Downloaded from www.aappublications.org/news at Indonesia:AAP Sponsored on January 6, 2021