RESEARCH ARTICLE

Variation in Care and Clinical Outcomes inChildren Hospitalized With Orbital CellulitisJessica L. Markham, MD, MSc,a Matthew Hall, PhD,a,b Jessica L. Bettenhausen, MD,a Angela L. Myers, MD, MPH,a,c Henry T. Puls, MD,a Russell J. McCulloh, MDa,c

A B S T R A C TOBJECTIVES: To describe variation in the care of children hospitalized with orbital cellulitis and todetermine associations with length of stay (LOS), emergency department (ED) revisits, and hospitalreadmissions.

METHODS: By using the Pediatric Health Information System, we performed a multicenter,retrospective study of children aged 2 months to 18 years with a primary International Classificationof Diseases, Ninth Revision, Clinical Modification discharge diagnosis code for orbital cellulitis from2007 to 2014. We assessed hospital-level variation in the use of diagnostic tests, corticosteroids, andantibiotics individually and in aggregate for association with outcomes (LOS, ED revisits,readmissions) after risk-adjusting for important clinical and demographic factors.

RESULTS: A total of 1828 children met inclusion criteria. Complete blood cell counts (median[interquartile range]: 81.8% [66.7–89.6]), C-reactive protein levels (57.1% [22.2–84.0]), bloodcultures (57.9% [48.9–63.6]), and computed tomography imaging (74.7% [66.7–81.0]) were themost frequently performed diagnostic tests, with significant variation observed across hospitals(all P , .001). Corticosteroids were used in 29.2% of children (interquartile range: 18.4–37.5).There was significant variation in antibiotic exposure across hospitals (P , .001). Increased totaldiagnostic test usage was associated with increased LOS (P 5 .044), but not with 30-day ED revisits(P 5 .176) or readmissions (P 5 .403).

CONCLUSIONS: Children hospitalized with orbital cellulitis experience wide variation in clinicalmanagement. Increased hospital-level usage is associated with increased LOS. Our findingshighlight a critical need to identify treatment strategies that optimize resource use and outcomesfor children hospitalized with orbital cellulitis.

aDivisions of PediatricHospital Medicine andcInfectious Diseases,

Children’s Mercy KansasCity, Kansas City,

Missouri; and bChildren’sHospital Association,

Lenexa, Kansas

www.hospitalpediatrics.orgDOI:https://doi.org/10.1542/hpeds.2017-0040Copyright © 2018 by the American Academy of Pediatrics

Address correspondence to Jessica L. Markham, MD, MS, Division of Pediatric Hospital Medicine, Children’s Mercy Kansas City,2401 Gillham Rd, Kansas City, MO 64108. E-mail: jlmarkham@cmh.edu

HOSPITAL PEDIATRICS (ISSN Numbers: Print, 2154-1663; Online, 2154-1671).

FINANCIAL DISCLOSURE: The 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.

Dr Markham conceptualized and designed the study, analyzed and interpreted the data, drafted the initial manuscript, and coordinated all editsof the manuscript. Dr Hall was involved in the study design, supervised the data analysis and interpretation, and reviewed and revised themanuscript as submitted. Drs Bettenhausen, Myers, and Puls assisted with study design, participated in the interpretation of data, andreviewed and revised the manuscript. Dr McCulloh supervised the conceptualization and design of the study, participated in the interpretationof data, and reviewed and revised the manuscript; and all authors approved the final version of the manuscript as submitted.

28 MARKHAM et al

Orbital cellulitis infections are serious andcan lead to life-altering conditions, includingvision loss, meningitis, brain abscess, andcavernous sinus thrombosis.1–3 Theseinfections account for an estimated2525 hospital admissions in children peryear within the United States, withestimated aggregate annual charges of$49.3 million.4 Although several hospitalshave implemented local clinical practiceguidelines and diagnostic algorithms, nonational guidelines or clinical trials haveestablished optimal diagnostic or treatmentregimens in the United States.5–7 Withoutconsensus guidelines, there is likely greatvariation in clinical practice within andacross hospitals. Investigations ofdiagnostic testing and management of otherdisease processes, including pneumoniaand diabetic ketoacidosis, havedemonstrated that increased variation andresource use is associated with increasedrates of hospitalization, higher costs, andincreased length of stay (LOS).8–14 However,similar studies focused on variation inmanagement and clinical outcomes oforbital infections are lacking.

Identifying variation in care and itsrelationship to patient outcomes is animportant step in the national effort toimprove health care delivery and patientoutcomes.15 The aims of this study are todescribe hospital-level variation indiagnostic testing, corticosteroid use, andantibiotic selection in a large, nationalcohort of children with orbital cellulitis andto describe the association of variation inresource use with LOS, emergencydepartment (ED) revisits, and hospitalreadmissions.

METHODSStudy Design and Data Source

This is a multicenter, retrospective studyof children hospitalized with a principalInternational Classification of Diseases,Ninth Revision, Clinical Modification (ICD-9-CM)discharge diagnosis code of 376.01(orbital cellulitis). Data were derived fromthe Pediatric Health Information System(PHIS), an administrative and billingdatabase of 46 tertiary care pediatrichospitals affiliated with the Children’sHospital Association (Lenexa, KS). Patient

data have been deidentified within PHIS;however, encryption of patient identifiersallows for tracking of individual patientsacross visits. The current study includeddata from 42 hospitals, with 4 hospitalsexcluded for billing data quality concerns.This study was approved by the localinstitutional review board.

Study Population

Inclusion Criteria

Children aged 2 months to 18 years whowere admitted to a PHIS-participatinghospital from January 1, 2007, throughDecember 31, 2014, with a principalICD-9-CM discharge diagnosis code of376.01 were eligible for inclusion. If patientshad multiple hospitalizations within a 30-dayperiod, only the index hospitalization wasincluded.

Exclusion Criteria

To identify children with orbital cellulitiswho were otherwise healthy, we excludedpatients with congenital malformations,prematurity, low birth weight, underlyingmalnutrition, or complex chronic conditions(Supplemental Table 2).16 Children withcompeting ophthalmologic diagnoses wereexcluded. Children with underlyingdiagnoses that would increase thelikelihood of corticosteroid administration(eg, asthma, adrenal insufficiency) wereexcluded to allow for examination ofcorticosteroid use. Children with secondarydiagnoses of intracranial abscess andtrauma were excluded because thesechildren would be unlikely to undergomanagement for orbital cellulitis alone.Children who received antifungal orantiviral therapy, and those who did notreceive systemic antibiotics within the first2 days of hospitalization, were excluded forthe possibility of a nonbacterial infection.Because of a high proportion of preseptalinfections identified during chart reviewwith our initial inclusion and exclusioncriteria, we subsequently excluded childrendischarged within the first 2 days becauseof their increased likelihood for preseptalinfection.

Validation

An internal validation study was performedthrough manual chart review at 2 PHIS

hospitals to assess the accuracy of our caseidentification strategy. Medical records witha principal ICD-9-CM code of 376.01 werereviewed by board-certified pediatriciansand classified as periorbital cellulitis (ie,preseptal infection) versus orbital cellulitiswith or without abscess (ie, post-septalinfection). Diagnostic category assessmentwas confirmed by a second reviewer, withdiscrepancies adjudicated by a board-certified pediatric infectious diseasesphysician. Of the 468 records reviewed,301 met inclusion criteria. Of includedrecords, our identification strategy wasassociated with a positive predictive valuefor identifying orbital infection of any type(pre- or postseptal infection) of 100.0% anda positive predictive value for orbitalcellulitis with or without abscess of 89.7%.Among children who underwent surgicalintervention, the positive predictive valuefor orbital cellulitis with or without abscesswas 98.4%.

Diagnostic Testing

All records were assessed for the presenceof a billing code for selected diagnostictests performed within the first 2 days ofhospitalization. Laboratory measuresincluded complete blood cell (CBC) count,C-reactive protein (CRP), erythrocytesedimentation rate (ESR), chemistries,blood cultures, cerebrospinal fluid (CSF)cultures, fungal wound cultures, andbacterial wound cultures. Imaging studiesincluded computed tomography (CT), MRI,and other imaging. Other imaging wasdefined as radiographs or ultrasoundimages of the orbits, sinuses, or head.

Corticosteroid and AntibioticExposure

Corticosteroid use was defined as oralor parenteral administration of at least1 of the following during the first 2 daysof hospitalization: dexamethasone,methylprednisolone, prednisolone, orprednisone. Empirical antibiotic selectionwas defined as antibiotics administeredduring the first 2 days of hospitalization andwas divided into 2 broad categories: onlyoral or any parenteral. Use within the first2 days of hospitalization was chosen tomost closely correlate with empirical

HOSPITAL PEDIATRICS Volume 8, Issue 1, January 2018 29

antibiotic choice and to minimize theinfluence of microbiological test results onantibiotic choice, a factor that could not becaptured in the PHIS database. Antibioticselection was defined by using ahierarchical algorithm (Supplemental Fig 5,Supplemental Table 3). Within our study,children who received oral preparationsalone were categorized as only oral.Children who received parenteral antibioticswere then categorized on the basis of theantibiotic’s spectrum of activity. Childrenwho received vancomycin, daptomycin, orlinezolid were defined as vancomycin aloneor in combination. Children who receivedclindamycin but did not receive a b-lactam/b-lactamase inhibitor antibiotic weredefined as clindamycin alone or incombination. Children who receivedclindamycin and a b-lactam/b-lactamaseinhibitor were defined as clindamycin/b-lactam combination. Children who did notreceive clindamycin but received ab-lactam/b-lactamase inhibitor antibioticwere defined as b-lactam alone or incombination. Finally, children who did notreceive clindamycin or a b-lactam/b-lactamase inhibitor antibiotic typicallyprescribed for the management of orbitalor sinus infection were defined as otherantibiotic. Categorization was reviewedand confirmed by the 2 board-certifiedpediatric infectious diseases physicians inthe study group.

Resource Use and Outcome Measures

Resource use included hospital-level ratesof diagnostic testing, corticosteroid use,and antibiotic selection. Outcome measuresincluded LOS, 30-day ED revisit rates, and 30-day readmission rates. Only ED revisits orreadmissions secondary to complications oforbital cellulitis or its management wereconsidered (Supplemental Table 4).

Demographics and ClinicalCharacteristics

Collected demographics and clinicalcharacteristics included age, sex, raceand/or ethnicity, primary payer admissionseason, region of the United States, surgicalintervention, peripherally inserted centralcatheter (PICC) placement, ICU transfer, andcase mix index (CMI). PICC placement was

defined by the presence of a correspondingprocedural code for catheter placement.Surgical intervention was defined by thepresence of procedural codes forophthalmic or sinonasal surgicalprocedures. ICU transfer and CMI werechosen to serve as surrogate markers forthe severity of illness (SOI). ICU transfer wasdefined as any child requiring transfer tothe ICU after admission. CMI in PHIS is arelative weight assigned to each dischargeon the basis of the All-Patient RefinedDiagnostic Group (APR-DRG) (3M, St Paul,MN) assignment and APR-DRG SOI, whichranges from 1 (minor) to 4 (extreme). Theweights are derived by Truven HealthAnalytics (Ann Arbor, MI) from its nationally-representative pediatric database as theratio of the average charge for dischargeswithin a specific APR-DRG and SOIcombination to the average charge for alldischarges in the database. For simplicity ofreporting and interpretation, we split theweights at the median into 2 groups: minorand major.

Statistical Analysis

Continuous variables were summarized withmedians and interquartile ranges (IQRs),whereas categorical variables weresummarized with frequencies andpercentages. We calculated hospital-levelsummary statistics to assess variation indiagnostic test use, corticosteroid use, andantibiotic selection across hospitals, andmade comparisons across hospitals byusing the x2 test or Kruskal-Wallis test.Hospital-level diagnostic test usage ratesand outcomes were risk-adjusted for ageand continuous CMI by using appropriatedistributions for generalized linear mixedeffects models with a random intercept foreach hospital. After adjustment, hospitalswere assigned risk-adjusted diagnostic testusage scores for individual diagnostic tests(ie, a hospital-level usage score for anindividual diagnostic test) on the basis oftheir absolute percentage above and belowthe median rate. For each diagnostic test,hospitals received 2 points if their risk-adjusted rate was .10% over the median,1 point if it was 5% to 10% over themedian, 0 points if it was within 5% of themedian, 21 point if it was 5% to 10% under

the median, and 22 points if it was .10%under the median. We then assigned a totaldiagnostic test usage score to each hospitalby summing the risk-adjusted diagnostictest scores for individual tests (ie, wegenerated an aggregate score for use of alldiagnostic tests at an individual hospital).We correlated the risk-adjusted totaldiagnostic test usage score with the risk-adjusted outcomes by using Pearson’scorrelation coefficient. We similarlycorrelated risk-adjusted total diagnostictest usage scores with risk-adjustedoutcomes in a substudy analysis of childrenwith orbital cellulitis who underwentsurgical intervention. All statistical analyseswere performed by using SAS version 9.4(SAS Institute, Cary, NC), and P values , .05were considered statistically significant.

RESULTSGeneral Characteristics

A total of 1828 children across 42 hospitalsmet the inclusion criteria (Fig 1). Of these,70.8% were ,10 years of age, 66.0% wereboys, and 2.1% required transfer to the ICU(Table 1). The median CMI for the studypopulation was 0.95 (IQR: 0.77–2.49). Amongchildren included in our cohort, 23.9%underwent surgical intervention. Themedian percentage of children acrosshospitals that underwent surgicalintervention was 50.0% (IQR: 23.4%–76.6%).

Variation in Resource Use andAssociation With LOS, ED Revisits,and Readmission

We observed substantial variation indiagnostic test use across hospitals (Figs 2and 3). CBC counts, CRP tests, bloodcultures, and CT scans were the mostfrequently performed diagnostic tests. CSFcultures, MRIs, and other imaging wereobtained infrequently.

Variation in the treatment of orbitalcellulitis with systemic corticosteroids andantibiotics was also observed acrosshospitals. Use of adjunctive corticosteroidsoccurred in a subset of pediatric patients(29.2% [IQR: 18.4–37.5], P , .001), with useobserved within 41 of the 42 hospitalsincluded in the study. Antibiotic selectionvaried widely across hospitals with morethan 200 unique combinations of antibiotics

30 MARKHAM et al

prescribed in the first 2 days ofhospitalization before categorization. Themost frequently prescribed antibioticregimens included vancomycin alone or incombination (43.1% [IQR: 29.2–62.9], P ,.001) and clindamycin/b-lactamcombinations (30.9% [IQR: 14.3–54.2], P ,.001). Twelve of the 1828 (0.7%) childrenwithin the cohort had exposure toantibiotics included within the parenteralother antibiotic category, whereas 1 (0.05%)child received oral antibiotics alone.Variation in antibiotic selection was alsoobserved within hospitals, as demonstratedby the distribution of antibiotic groupswithin a majority of hospitals (Fig 3). Weobserved similar variation in diagnostictest use, corticosteroid use, andantibiotic selection in a subanalysis ofchildren with orbital cellulitis whounderwent surgical intervention(Supplemental Figs 6 and 7).

After risk adjustment for age andcontinuous CMI, there was significantvariation across hospitals in LOS (4.1 days

[IQR: 3.8–4.2], P , .001), but not for 30-dayreadmission rates (1.6% [IQR: 0.0–3.5], P 5.082), or 30-day ED revisit rates (2.3% [IQR:0.0–4.7], P 5 .053). An increased totaldiagnostic test usage score at the hospital-level was significantly associated with anincreased LOS (P 5 .044) (Fig 4). For every5-point increase in the score, the risk-adjusted LOS increased by 0.19 days(95% confidence interval: 0.01–0.37). Nostatistically significant association wasfound between the total diagnostic testusage score and the risk of 30-day EDrevisits (P 5 .176) or 30-day readmissions(P 5 .403). We observed similar variation inLOS across hospitals in a subanalysis ofchildren with orbital cellulitis whounderwent surgical intervention(Supplemental Fig 8).

DISCUSSION

In this multicenter, retrospective study, weobserved substantial variation acrosshospitals in diagnostic test use,corticosteroid use, and empirical antibiotic

TABLE 1 Demographic and ClinicalCharacteristics for ChildrenHospitalized With Orbital Cellulitis,N 5 1828

Characteristic n (%)

Age

2 mo–1 y 301 (16.5)

2–4 y 386 (21.1)

5–9 y 608 (33.3)

10–14 y 443 (24.2)

15–18 y 90 (4.9)

Sex

Boys 1207 (66.0)

Race and/or ethnicity

Non-Hispanic white 933 (51.0)

Non-Hispanic African American 398 (21.8)

Hispanic 248 (13.6)

Asian American 38 (2.1)

Other 211 (11.5)

Payer

Government 807 (44.1)

Private 873 (47.8)

Other 148 (8.1)

Season

Spring 528 (28.9)

Summer 343 (18.8)

Fall 386 (21.1)

Winter 571 (31.2)

Region

Midwest 422 (23.1)

Northeast 266 (14.6)

South 714 (39.1)

West 426 (23.3)

Surgical interventiona

Yes 436 (23.9)

PICCb

Yes 305 (16.7)

ICUc

Yes 38 (2.1)

CMId

Minor 888 (48.6)

Major 940 (51.4)

a Surgical intervention was defined on the basis ofthe presence or absence of procedural codes forophthalmic or sinonasal surgical procedures.

b PICC placement was defined on the basis of thepresence or absence of a correspondingprocedural code for catheter placement.

c ICU transfer was defined as any child requiringtransfer to the ICU after admission.

d CMI is a relative weight assigned to eachdischarge on the basis of the ARP-DRGassignment and ARP-DRG SOI. For simplicity ofreporting and interpretation, we split theweights at the median into 2 groups.

FIGURE 1 Cohort flow diagram.

HOSPITAL PEDIATRICS Volume 8, Issue 1, January 2018 31

selection in a large, national cohort ofchildren with orbital cellulitis. Extensive

variation in empirical antibiotic selection

was observed not only across hospitals but

also within individual hospitals. We found

that increased diagnostic test use was

associated with increased LOS but was notassociated with ED revisits or hospitalreadmissions.

Significant variation was observed for theuse of nearly every diagnostic test in ourcohort. The variation we observed may

reflect clinical uncertainty of the utility ofsome diagnostic tests in predictingoutcomes for patients with orbital cellulitis.For example, although the authors of somestudies have suggested that elevated whiteblood cell counts are associated with thepresence of a drainable abscess, authors ofother studies have shown that the degree ofelevation does not reliably predict thedevelopment of a complicated courseor the need for surgical intervention.17,18

Additionally, expert opinion varies regardingthe appropriate use of blood cultures in theevaluation of orbital cellulitis, challengingthe formation of a consensus guideline.3,5,7,19

The authors of previous studies havereported that the rates of blood culturepositivity among children with orbitalcellulitis range from 2.27% to 7%.5,19–22

However, studies have revealed that rates ofblood culture contamination amongpediatric patients range from 1% to 4%,23–25

making it difficult to develop a universalrecommendation for obtaining a bloodculture in the setting of low likelihood ofpositivity in patients with orbital cellulitis.

Within our study, we observed significantvariation in the use of corticosteroids andin the choice of empirical antibiotic forchildren hospitalized with orbital cellulitis.

FIGURE 2 Box plot distributions for individual diagnostic tests in children hospitalized withorbital cellulitis. Bacterial Culture, bacterial wound culture; Fungal Culture, fungalwound culture.

FIGURE 3 Heat map of risk-adjusted diagnostic test, corticosteroid, and empirical antibiotic selection by hospital presented as variance from themedian across hospitals. b-Lactam includes b-lactam/b-lactamase inhibitor antibiotics.

32 MARKHAM et al

The authors of previous studies havesuggested that using adjunctivecorticosteroids in the management oforbital cellulitis may be beneficial; however,the generalizability of these findings to allpediatric patients is limited by small samplesizes derived from single instiutions.26–28

Within our study, corticosteroids were usedin 29.2% (IQR: 18.4–37.5) of children, withuse observed in 41 hospitals, suggestingthat corticosteroids were used for aminority of patients at many of thehospitals. The variation in corticosteroiduse may reflect the lack of strong evidencefor clinical benefit as reported in theliterature, as well as reflecting physicianconcern for the adverse effect profile ofcorticosteroids, particularly for hostimmune suppression and the increased riskfor infectious complications. The variation inempirical antibiotic selection that weobserved is consistent with observationsfrom previous studies of orbitalcellulitis.1,18,20,29,30 The wide variation inempirical antibiotic selection that weobserved both within and across hospitalsmay similarly reflect a lack of a clearconsensus recommendation as to optimalempirical antibiotic choice. Regardless, theuse of such a diverse range of antibioticcombinations highlights the uncertaintyclinicians face when choosing optimaltreatment regimens for orbital cellulitis.

Higher diagnostic test use at the hospitallevel was associated with increased LOS, afinding that occurs in other seriousinfections in children.8–14 Not only areprolonged hospitalizations more costly, butthey also increase the risk of acquiring

nosocomial infections, place an increasedpsychosocial burden upon families, andreduce the quality of life for the hospitalizedchild.31–34 Although one might infer thatincreased test use might provide morecomprehensive management, and thereforedecrease rates of revisits, we did notobserve any significant reduction in revisitsamong hospitals with high diagnostictesting usage versus hospitals with lowdiagnostic testing usage, even aftercontrolling for age and SOI. Understandingthe significant variation in pediatric orbitalcellulitis hospitalizations may proveinfluential in the development of clinicalguidelines aimed at improving the quality ofcare delivered in this subset of infections,because guideline development has provento be beneficial within other diseaseprocesses.35

This study has several important limitations.First, the ICD-9-CM discharge diagnosis code376.01 encompasses a range of orbitalinfections, including periorbital cellulitis,orbital cellulitis, orbital abscess, andsubperiosteal orbital abscess. Consequently,some of the variation in care we observedmay reflect differences in admissionpractices and in how clinicians manage pre-versus postseptal infections. Although thereare limitations in using the ICD-9-CMdiagnosis code to identify orbital cellulitis,this is a previously published strategy.36

Additionally, we attempted to mitigatemisclassification bias by identifying thosemost likely to have postseptal infectionthrough our application of exclusion criteriaand through examination of a cohort ofchildren who underwent surgical

intervention. Although our case definitionincludes both pre- and postseptal orbitalinfections, our internal test of validityrevealed that our identification strategy wasassociated with a positive predictive valueof 89.7% for postseptal disease in ourgeneral population and with a positivepredictive value of 98.4% among those whounderwent surgical intervention. As thevalidation occurred at 2 hospitals, theresults may have differed (higher or lowerpositive predictive value) with the inclusionof additional hospitals. Our substudyanalysis of children who underwent surgicalintervention (ie, children with highlikelihood for postseptal disease), revealedsimilar results on Pearson’s correlationcompared with the entire study population,further supporting our case definition’svalidity. Within our study, LOS was used asan exclusion criterion as well as anoutcome measure. Although thismethodology has the potential to biastoward detecting a difference, the results ofthe analyses were unchanged regardless ofwhether children hospitalized for ,2 dayswere included or excluded. The PHISdatabase contains administrative data only,thus limiting our ability to evaluate theassociation of patient presentation withclinical decision-making. Particularly, wehad limited ability to assess orbital cellulitisseverity or illness severity, microbiologictest results, or local patterns ofantimicrobial resistance, factors that mighthave influenced LOS and decision-making,including diagnostic test use andantimicrobial prescription practices. Weattempted to mitigate these effects by

FIGURE 4 Relationship between hospital-level risk-adjusted total diagnostic test usage score and risk-adjusted outcomes. A, LOS. B, Thirty-day EDrevisits. C, Thirty-day hospital readmissions.

HOSPITAL PEDIATRICS Volume 8, Issue 1, January 2018 33

assessing surrogate markers for illnessseverity, such as CMI, and by focusing onempirical antibiotic selection. Finally, thedifferences in LOS observed betweenhospitals were modest, and additionalfactors, including confounding fromunmeasured variables, may havecontributed to the association betweendiagnostic testing and LOS that weobserved.

CONCLUSIONS

Our study provides a comprehensiveoverview of the wide variation in clinicalmanagement that children hospitalized withorbital cellulitis experience. With ourfindings, we highlight an opportunity forantimicrobial stewardship for orbitalcellulitis and a definitive need for futureprospective investigations aimed atidentifying optimal diagnosis andmanagement strategies. Future studiesexamining patient-level factors inassociation with outcomes may moreclearly identify opportunities fortargeted management strategies andhelp guide the development of diagnosticand treatment guidelines for orbitalcellulitis.

REFERENCES

1. Ferguson MP, McNab AA. Currenttreatment and outcome in orbitalcellulitis. Aust N Z J Ophthalmol. 1999;27(6):375–379

2. Chaudhry IA, Shamsi FA, Elzaridi E, et al.Outcome of treated orbital cellulitis in atertiary eye care center in the middleEast. Ophthalmology. 2007;114(2):345–354

3. Hauser A, Fogarasi S. Periorbital andorbital cellulitis. Pediatr Rev. 2010;31(6):242–249

4. Agency for Healthcare Research andQuality. HCUPnet: Healthcare Cost andUtilization Project. Available at: https://hcupnet.ahrq.gov/. Accessed December1, 2015

5. Uzcátegui N, Warman R, Smith A, HowardCW. Clinical practice guidelines for themanagement of orbital cellulitis.J Pediatr Ophthalmol Strabismus. 1998;35(2):73–79; quiz 110–111

6. Bedwell J, Bauman NM. Management ofpediatric orbital cellulitis and abscess.Curr Opin Otolaryngol Head Neck Surg.2011;19(6):467–473

7. Howe L, Jones NS. Guidelines for themanagement of periorbital cellulitis/abscess. Clin Otolaryngol Allied Sci.2004;29(6):725–728

8. Florin TA, French B, Zorc JJ, Alpern ER,Shah SS. Variation in emergencydepartment diagnostic testing anddisposition outcomes in pneumonia.Pediatrics. 2013;132(2):237–244

9. Leyenaar JK, Lagu T, Shieh M-S, PekowPS, Lindenauer PK. Variation in resourceutilization for the management ofuncomplicated community-acquiredpneumonia across community andchildren’s hospitals. J Pediatr. 2014;165(3):585–591

10. Knapp JF, Simon SD, Sharma V. Variationand trends in ED use of radiographs forasthma, bronchiolitis, and croup inchildren. Pediatrics. 2013;132(2):245–252

11. Tieder JS, McLeod L, Keren R, et al;Pediatric Research in Inpatient SettingsNetwork. Variation in resource use andreadmission for diabetic ketoacidosis inchildren’s hospitals. Pediatrics. 2013;132(2):229–236

12. Aronson PL, Thurm C, Alpern ER, et al;Febrile Young Infant ResearchCollaborative. Variation in care of thefebrile young infant ,90 days in USpediatric emergency departments.Pediatrics. 2014;134(4):667–677

13. Rice-Townsend S, Chen C, Barnes JN,Rangel SJ. Variation in practice patternsand resource utilization surroundingmanagement of intussusception atfreestanding children’s hospitals.J Pediatr Surg. 2013;48(1):104–110

14. Rice-Townsend S, Barnes JN, Hall M,Baxter JL, Rangel SJ. Variation inpractice and resource utilizationassociated with the diagnosis andmanagement of appendicitis atfreestanding children’s hospitals:implications for value-basedcomparative analysis. Ann Surg. 2014;259(6):1228–1234

15. Karnon J, Partington A, Horsfall M, ChewD. Variation in clinical practice: apriority setting approach to the stagedfunding of quality improvement. ApplHealth Econ Health Policy. 2016;14(1):21–27

16. Feudtner C, Feinstein JA, Zhong W, HallM, Dai D. Pediatric complex chronicconditions classification system version2: updated for ICD-10 and complexmedical technology dependence andtransplantation. BMC Pediatr. 2014;14:199

17. Rudloe TF, Harper MB, Prabhu SP, RahbarR, Vanderveen D, Kimia AA. Acuteperiorbital infections: who needsemergent imaging? Pediatrics. 2010;125(4). Available at: www.pediatrics.org/cgi/content/full/125/4/e719

18. Robinson A, Beech T, McDermott AL,Sinha A. Investigation and managementof adult periorbital and orbital cellulitis.J Laryngol Otol. 2007;121(6):545–547

19. McKinley SH, Yen MT, Miller AM, Yen KG.Microbiology of pediatric orbitalcellulitis. Am J Ophthalmol. 2007;144(4):497–501

20. Baring DEC, Hilmi OJ. An evidence basedreview of periorbital cellulitis. ClinOtolaryngol. 2011;36(1):57–64

21. Huang SF, Lee TJ, Lee YS, Chen CC, ChinSC, Wang NC. Acute rhinosinusitis-related orbital infection in pediatricpatients: a retrospective analysis. AnnOtol Rhinol Laryngol. 2011;120(3):185–190

22. Sharma A, Liu ES, Le TD, et al. Pediatricorbital cellulitis in the Haemophilusinfluenzae vaccine era. J AAPOS. 2015;19(3):206–210

23. Stoll ML, Rubin LG. Incidence of occultbacteremia among highly febrile youngchildren in the era of the pneumococcalconjugate vaccine: a study from achildren’s hospital emergencydepartment and Urgent Care Center.Arch Pediatr Adolesc Med. 2004;158(7):671–675

24. Myers AL, Hall M, Williams DJ, et al.Prevalence of bacteremia in hospitalizedpediatric patients with community-

34 MARKHAM et al

acquired pneumonia. Pediatr Infect DisJ. 2013;32(7):736–740

25. Herz AM, Greenhow TL, Alcantara J, et al.Changing epidemiology of outpatientbacteremia in 3- to 36-month-oldchildren after the introduction of theheptavalent-conjugated pneumococcalvaccine. Pediatr Infect Dis J. 2006;25(4):293–300

26. Pushker N, Tejwani LK, Bajaj MS,Khurana S, Velpandian T, Chandra M.Role of oral corticosteroids in orbitalcellulitis. Am J Ophthalmol. 2013;156(1):178–183.e1

27. Yen MT, Yen KG. Effect of corticosteroidsin the acute management of pediatricorbital cellulitis with subperiostealabscess. Ophthal Plast Reconstr Surg.2005;21(5):363–366; discussion 366–367

28. Davies BW, Smith JM, Hink EM, DurairajVD. C-reactive protein as a marker for

initiating steroid treatment in childrenwith orbital cellulitis. Ophthal PlastReconstr Surg. 2015;31(5):364–368

29. Nageswaran S, Woods CR, Benjamin DKJr, Givner LB, Shetty AK. Orbital cellulitisin children. Pediatr Infect Dis J. 2006;25(8):695–699

30. Seltz LB, Smith J, Durairaj VD, EnzenauerR, Todd J. Microbiology and antibioticmanagement of orbital cellulitis.Pediatrics. 2011;127(3). Available at:www.pediatrics.org/cgi/content/full/127/3/e566

31. Commodari E. Children staying inhospital: a research on psychologicalstress of caregivers. Ital J Pediatr. 2010;36:40

32. Rennick JE, Dougherty G, Chambers C,et al. Children’s psychological andbehavioral responses following pediatricintensive care unit hospitalization: the

caring intensively study. BMC Pediatr.2014;14:276

33. El-Jawahri AR, Traeger LN, Kuzmuk K,et al. Quality of life and mood of patientsand family caregivers duringhospitalization for hematopoietic stemcell transplantation. Cancer. 2015;121(6):951–959

34. Coyne I. Children’s experiences ofhospitalization. J Child Health Care.2006;10(4):326–336

35. McCulloh RJ, Alverson BK. Thechallenge–and promise–of local clinicalpractice guidelines. Pediatrics. 2012;130(5):941–942

36. Mahalingam-Dhingra A, Lander L,Preciado DA, Taylormoore J, Shah RK.Orbital and periorbital infections: anational perspective. Arch OtolaryngolHead Neck Surg. 2011;137(8):769–773

HOSPITAL PEDIATRICS Volume 8, Issue 1, January 2018 35