OBJECTIVES:

To describe inpatient management of patients with croup admitted from the emergency department (ED).

METHODS:

In a multicentered, cross-sectional observational study based on retrospective chart review, we identified children 6 months to 5 years of age with a discharge diagnosis of croup. All patients were evaluated in the ED and treated with at least 1 dose of racemic epinephrine (RE) before admission. Children with hypoxia or directly admitted to the PICU were excluded.

RESULTS:

We identified 628 admissions for croup. Significant interventions, defined as additional RE, helium-oxygen use, or PICU transfer, occurred in 142 patients (22.6%). A total of 137 children received additional RE on the inpatient ward, and 5 received RE and were transferred to the PICU. No patient was treated with helium-oxygen. A total 486 (77.4%) of patients did not receive significant interventions postadmission. Length of stay for children not requiring significant intervention was, on average, <24 hours (18.8 hours [SD 9.3]; range 1.2–111 hours). Children with tachypnea (odds ratio = 2.5; P = .002) on arrival to ED and patients who had ED radiographs (odds ratio = 1.7; P = .018) had increased odds of receiving a significant intervention after admission.

CONCLUSIONS:

Less than one-quarter of children admitted to the general wards for croup received significant interventions after admission. Tachypnea in the ED and use of radiograph were associated with an increased use of significant interventions.

Croup, or acute laryngotracheitis, is a common pediatric illness that can range in severity from rhinorrhea and a barky cough to stridor, respiratory distress, and, in extreme cases, airway obstruction. Hospital admission rates have varied widely, from 1% to 31% in older studies.1,2  Authors of a recent study including data from 26 pediatric hospitals across the United States found a median admission rate of 9.1% and significant variations in management between hospitals in inpatient care of croup.3 

Outpatient management of croup in the clinic, urgent care, or emergency department (ED) is well described and includes corticosteroids for croup of all severities and nebulized inhaled epinephrine for patients presenting with audible stridor. Racemic epinephrine (RE) has been shown to significantly improve croup scores 30 minutes after administration compared with a placebo but with no measurable difference 2 hours posttreatment, suggesting that the effect is transient.2  RE however effectively bridges the time to the observed effect of concurrently administered glucocorticoid steroids. Discharge after 1 dose of RE and ∼3 hours of observation has been shown to be safe and associated with few revisits.47 

However, there is a paucity of data or clinical practice guidelines used to describe inpatient management of croup. Many physicians choose to admit children requiring >1 RE nebulization.

The primary objective of this study was to evaluate the inpatient course and management of patients admitted with croup, focusing specifically on the subset of patients who were not critically ill. We aimed to study the inpatient course and evaluate what proportion of admitted patients received significant interventions specified as RE, helium-oxygen (heliox), or transfer to an ICU from the inpatient ward. We also aimed to evaluate any association between ED presentation and course and subsequent inpatient course. We hypothesized that most children would not require any significant interventions after admission. Second, we sought to determine if ED presentation and course could predict the presence of more interventions after admission.

We conducted a multicentered, cross-sectional observational study based on retrospective chart review involving 3 pediatric tertiary care children’s hospitals in Minnesota with a combined annual ED census of ∼115 000. The study period was March 1, 2011 to September 30, 2015. Identification of subjects was based on primary discharge diagnosis of croup (International Classification of Diseases, Ninth Revision code of 464.4) and if subjects met the following inclusion criteria: (1) evaluated by the participating ED; (2) treated with at least 1 RE dose in the ED; (3) age 6 months to 5 years of age; and (4) admitted to the hospital. Children were excluded if they had a history of congenital anomalies of the airway, previous tracheal surgery, required supplemental oxygen in the ED (saturation <90%), or were directly admitted to the PICU. At 2 children’s hospitals, a data specialist programmed the electronic health record to obtain the clinical data set. From this data set, charts were individually reviewed wherever there were missing data or uncertainty about management. The third hospital’s data were hand abstracted from International Classification of Diseases, Ninth Revision codes.

Significant interventions were defined as any number of inpatient RE, heliox treatment, or transfer to the PICU. Tachycardia and tachypnea were defined per age (Supplemental Table 5) and correspond largely with the 90th percentile for children presenting to the ED, fever was defined as a temperature ≥38°C, and an abnormal pulse oximetry was defined as <95% on room air.8,9 

If a patient had multiple hospitalizations (5.4% of analysis sample), 1 admission was randomly selected by using a random number generator (https://www.random.org). This procedure allowed data interpretation as patients rather than admissions. The study was approved by institutional review boards at each site.

Descriptive statistics include mean, SD, median, and minimum and maximum for continuous variables, and the sample size and percentage for categorical variables. χ2 tests and 2-sample t tests were used to evaluate unadjusted associations between the need for significant intervention after hospitalization for categorical and continuous variables, respectively. The nonparametric Wilcoxon rank-sum test was used to evaluate the associations between the need for significant intervention and hospital length of stay (LOS) and number of RE doses in the ED, which had skewed distributions. The nonparametric Kruskal-Wallis test was used to compare hospital LOS distribution between 3 groups: no additional RE needed, 1 additional dose of RE, 2 or more additional doses of RE after hospital admission. Post hoc Wilcoxon rank-sum tests with Bonferroni adjustment for multiple comparisons were used for pairwise comparisons of hospital LOS distribution. A multiple logistic regression model was used to estimate odds ratios (ORs) (95% confidence intervals) for significant hospital intervention adjusted for patient demographic characteristics, ED vitals, and ED treatments. Patients with missing data were not included in the logistic regression analysis. A significance level of α = .05 was used to determine statistical significance for all tests and model estimates. SAS version 9.4 (SAS Institute, Inc, Cary, NC) was used for all analyses.

We identified 702 patient visits between March 1, 2011, and September 30, 2015, who met inclusion criteria. Twenty patients were admitted directly to the PICU, and 12 had oxygen saturation <90% on room air and were not included in the analysis. Forty patients had multiple visits during the study period, and after randomly selecting 1 admission, the final cohort was 628 patients. A total of 403 (64.2%) patients were boys (Table 1). The overall mean age in months was 17.9 (SD 7.2). The number of nebulized RE given in the ED ranged between 1 and 3 (mean 1.77; SD 0.61). Thirty-three percent of patients required 1 RE in the ED, 58% required 2, and 9% required 3.

TABLE 1

Descriptive Statistics of Patients With Croup (N = 628)

Demographic CharacteristicsAll Patients
Sex, n (%)  
 Girls 225 (35.8) 
 Boys 403 (64.2) 
Age, mo  
 Mean (SD) 17.9 (7.2) 
 Median (range) 16.9 (6.0–45.7) 
Age group,an (%), mo  
 6–<12 150 (23.9) 
 12–<18 193 (30.7) 
 18–<24 160 (25.5) 
 24–<30 84 (13.4) 
 30–<36 37 (5.9) 
 36–46 4 (0.6) 
ED vitals,b treatment, imaging  
 Heart rate (n = 602), beats per minute  
  Mean (SD) 152.9 (21.9) 
  Median (range) 152 (36–212) 
  Age-defined tachycardia,cn (%) 287 (47.7) 
 Respiratory rate (n = 595)  
  Mean (SD) 38.3 (8.9) 
  Median (range) 38.0 (20–68) 
  Age-defined tachypnea,dn (%) 430 (72.3) 
 Temperature (n = 553)  
  Mean (SD) 37.6 (1.0) 
  Median (range) 37.4 (35.7–41.1) 
  Temperature ≥38°C, n (%) 169 (30.6) 
 O2 saturation (n = 572)  
  Mean (SD) 97.9 (1.7) 
  Median (range) 98 (91–100) 
  O2 saturation 91%–94%, n (%) 21 (3.7) 
 No. RE doses in EDe  
  Mean (SD) 1.77 (0.61) 
  Median (range) 2 (1–3) 
  1 RE dose in ED, n (%) 206 (32.8) 
  2 RE doses in ED, n (%) 362 (57.6) 
  3 RE doses in ED, n (%) 60 (9.6) 
 Other treatments in ED, n (%)  
  Steroids 512 (81.5) 
  Antibiotics 11 (1.8) 
  Bronchodilators 40 (6.4) 
  Any of steroids, antibiotics, and/or bronchodilators 517 (82.3) 
 Imaging, n (%)  
  Radiograph neck only 151 (24.0) 
  Radiograph chest only 29 (4.6) 
  Radiograph neck and chest 72 (11.5) 
  Any radiograph 252 (40.1) 
Hospital LOS and interventions, h  
 Mean (SD) 22.8 (17.8) 
 Median (range) 18.7 (1.2–285.8) 
 Helioxf after admission, n (%) 0 (0.0) 
 Transfer to PICU from floor, n (%) 5 (0.8) 
 Significant intervention after admission  
  Any of RE, heliox,f or transfer to PICU, n (%) 142 (22.6) 
 Other interventions after admission, n (%)  
  Steroids 273 (43.5) 
  Antibiotics 53 (8.4) 
  Bronchodilators 27 (4.3) 
  Any of steroids, antibiotics, and/or bronchodilators 309 (49.2) 
Demographic CharacteristicsAll Patients
Sex, n (%)  
 Girls 225 (35.8) 
 Boys 403 (64.2) 
Age, mo  
 Mean (SD) 17.9 (7.2) 
 Median (range) 16.9 (6.0–45.7) 
Age group,an (%), mo  
 6–<12 150 (23.9) 
 12–<18 193 (30.7) 
 18–<24 160 (25.5) 
 24–<30 84 (13.4) 
 30–<36 37 (5.9) 
 36–46 4 (0.6) 
ED vitals,b treatment, imaging  
 Heart rate (n = 602), beats per minute  
  Mean (SD) 152.9 (21.9) 
  Median (range) 152 (36–212) 
  Age-defined tachycardia,cn (%) 287 (47.7) 
 Respiratory rate (n = 595)  
  Mean (SD) 38.3 (8.9) 
  Median (range) 38.0 (20–68) 
  Age-defined tachypnea,dn (%) 430 (72.3) 
 Temperature (n = 553)  
  Mean (SD) 37.6 (1.0) 
  Median (range) 37.4 (35.7–41.1) 
  Temperature ≥38°C, n (%) 169 (30.6) 
 O2 saturation (n = 572)  
  Mean (SD) 97.9 (1.7) 
  Median (range) 98 (91–100) 
  O2 saturation 91%–94%, n (%) 21 (3.7) 
 No. RE doses in EDe  
  Mean (SD) 1.77 (0.61) 
  Median (range) 2 (1–3) 
  1 RE dose in ED, n (%) 206 (32.8) 
  2 RE doses in ED, n (%) 362 (57.6) 
  3 RE doses in ED, n (%) 60 (9.6) 
 Other treatments in ED, n (%)  
  Steroids 512 (81.5) 
  Antibiotics 11 (1.8) 
  Bronchodilators 40 (6.4) 
  Any of steroids, antibiotics, and/or bronchodilators 517 (82.3) 
 Imaging, n (%)  
  Radiograph neck only 151 (24.0) 
  Radiograph chest only 29 (4.6) 
  Radiograph neck and chest 72 (11.5) 
  Any radiograph 252 (40.1) 
Hospital LOS and interventions, h  
 Mean (SD) 22.8 (17.8) 
 Median (range) 18.7 (1.2–285.8) 
 Helioxf after admission, n (%) 0 (0.0) 
 Transfer to PICU from floor, n (%) 5 (0.8) 
 Significant intervention after admission  
  Any of RE, heliox,f or transfer to PICU, n (%) 142 (22.6) 
 Other interventions after admission, n (%)  
  Steroids 273 (43.5) 
  Antibiotics 53 (8.4) 
  Bronchodilators 27 (4.3) 
  Any of steroids, antibiotics, and/or bronchodilators 309 (49.2) 
a

99.4% of patients were <3 y.

b

The number for initial vitals is <628 because of missing data for some patients.

c

Age-defined tachycardia is 2 or more SEs above the mean heart rate for age.

d

Age-defined tachypnea is 2 or more SEs above the mean respiratory rate for age.

e

All patients had at least 1 RE treatment in ED; the maximum number of RE doses in ED was 3.

f

No patients were treated with heliox after admission.

Most children (81.5%) received dexamethasone in the ED, with 18.5% who received it before presentation from a clinic or urgent care. A total of 252 (40.1%) patients had imaging performed in the ED, most commonly radiographs of the neck (n = 223; 35.5%) (Table 1). A total of 101 children had chest radiographs performed. Twelve percent of patients were missing 1 or more elements of triage vital signs (Table 1).

Patient disposition from ED and those who had significant interventions can be seen in Fig 1. From the cohort, 486 children (77.4%) did not receive any significant interventions after admission. Significant interventions occurred in 142 patients (22.6%). All of these patients received additional RE, and 5 were also transferred to the PICU. No one received heliox. The majority of patients who received an inpatient intervention received a single dose of RE (89 patients; 62.7%). The median number of RE after admission among those with a significant intervention was 1 (mean = 1.8; range 1–17) (Table 2). All children were discharged with a primary diagnosis of croup, but many had additional diagnoses. The most common additional diagnoses were acute upper respiratory infection or viral infection not otherwise specified, which was diagnosed in 220 children (35%). One child was diagnosed with pneumonia in addition to croup. No one had an additional diagnosis of epiglottitis, bacterial tracheitis, or other cause of stridor. We also discovered that 20 children (3.2%) had a history of asthma, 51 (8.1%) had a history of perinatal problems (2 also had asthma), 7 (1.1%) had laryngotracheal anomalies that were not excluded (1 also had asthma), and 14 (2.2%) had developmental delays categorized as not otherwise specified (n = 6; 1 also had asthma), other congenital anomaly (n = 1), or Down syndrome (n = 7; 1 also had asthma).

FIGURE 1

Patient flow after ED management. a Did not require RE after admission or transfer to PICU. b Met criteria for significant intervention (n = 142; 22.6%).

FIGURE 1

Patient flow after ED management. a Did not require RE after admission or transfer to PICU. b Met criteria for significant intervention (n = 142; 22.6%).

Close modal
TABLE 2

Additional RE Doses After Admission

All PatientsPatients with Significant Interventions
No. patients 628 142 
Mean (SD) 0.41 (1.16) 1.83 (1.82) 
Median (range) 0 (0–17) 1 (1–17) 
1 RE dose, n (%) 89 (14.2) 89 (62.7) 
2 or more RE doses, n (%) 53 (8.4) 53 (37.3) 
3 or more RE doses, n (%) 29 (4.6) 29 (20.4) 
Any RE, n (%) 142 (22.6) 142 (100) 
All PatientsPatients with Significant Interventions
No. patients 628 142 
Mean (SD) 0.41 (1.16) 1.83 (1.82) 
Median (range) 0 (0–17) 1 (1–17) 
1 RE dose, n (%) 89 (14.2) 89 (62.7) 
2 or more RE doses, n (%) 53 (8.4) 53 (37.3) 
3 or more RE doses, n (%) 29 (4.6) 29 (20.4) 
Any RE, n (%) 142 (22.6) 142 (100) 

The overall mean hospital LOS, as seen in Table 1, was 22.8 hours (SD 17.8; range 1.2–285.8 hours; median 18.7 hours). The mean LOS was almost twice as long at 36.5 hours (SD 29.6; range 6.7–285.8 hours; median 27.7 hours) for patients with significant interventions as compared with 18.8 hours (SD 9.3; range 1.2–111.4 hours; median 17 hours) for patients who did not receive significant interventions (P < .0001 for Wilcoxon rank-sum test, Table 3). Median hospital LOS increased monotonically with an increasing number of RE doses after admission: 17.0 hours for those with no additional RE, 23.5 hours for those with 1 additional RE dose, and 42.7 hours for those with 2 or more doses of RE postadmission. The Kruskal-Wallis test of hospital LOS distribution differences was significant (P < .0001); post hoc Wilcoxon rank-sum tests with Bonferroni adjustment for multiple comparisons were used to identify significant differences in hospital LOS distribution between all 3 RE groups.

TABLE 3

Patient Characteristics, ED Treatment, and Hospital LOS by Significant Intervention After Hospital Admission

No Significant InterventionsSignificant InterventionaPb
No. patients 486 142  
Girls, n (%) 165 (34.0) 60 (42.3) .070 
Age, mo   .624 
 Mean (SD) 17.9 (7.3) 17.6 (6.8)  
 Median (range) 17.1 (6.0–45.7) 16.5 (7.1–5.9)  
Age-defined tachycardia (n = 602c), n (%) 211 (45.5) 76 (55.1) .048 
Age-defined tachypnea (n = 595c), n (%) 316 (68.9) 114 (83.8) <.001 
ED triage temperature ≥38°C (n = 553c), n (%) 132 (30.8) 37 (29.8) .843 
ED triage O2 saturation (n = 572c  .728 
 Mean (SD) 97.9 (1.7) 97.9 (1.7)  
 Median (range) 98.0 (91–100) 98.0 (92–100)  
No. RE in ED   .143 
 Mean (SD) 1.7 (0.6) 1.8 (0.6)  
 Median (range) 2.0 (1–3) 2.0 (1–3)  
RE in ED group, n (%)   .182 
 1 RE 164 (33.7) 42 (29.6)  
 2 RE 281 (57.8) 81 (57.0)  
 3 RE 41 (8.4) 19 (13.4)  
Imaging, n (%)   .006 
 Radiograph of neck or chest 181 (37.2) 71 (50.0)  
 No radiograph 305 (62.8) 71 (50.0)  
Steroids in ED, n (%)   .241 
 No steroid treatment 85 (17.5) 31 (21.8)  
 Any steroids in ED 401 (82.5) 111 (78.2)  
Hospital LOS   <.0001 
 Mean (SD) 18.8 (9.3) 36.5 (29.6)  
 Median (range) 17.0 (1.2–111.4) 27.7 (6.7–285.8d 
No Significant InterventionsSignificant InterventionaPb
No. patients 486 142  
Girls, n (%) 165 (34.0) 60 (42.3) .070 
Age, mo   .624 
 Mean (SD) 17.9 (7.3) 17.6 (6.8)  
 Median (range) 17.1 (6.0–45.7) 16.5 (7.1–5.9)  
Age-defined tachycardia (n = 602c), n (%) 211 (45.5) 76 (55.1) .048 
Age-defined tachypnea (n = 595c), n (%) 316 (68.9) 114 (83.8) <.001 
ED triage temperature ≥38°C (n = 553c), n (%) 132 (30.8) 37 (29.8) .843 
ED triage O2 saturation (n = 572c  .728 
 Mean (SD) 97.9 (1.7) 97.9 (1.7)  
 Median (range) 98.0 (91–100) 98.0 (92–100)  
No. RE in ED   .143 
 Mean (SD) 1.7 (0.6) 1.8 (0.6)  
 Median (range) 2.0 (1–3) 2.0 (1–3)  
RE in ED group, n (%)   .182 
 1 RE 164 (33.7) 42 (29.6)  
 2 RE 281 (57.8) 81 (57.0)  
 3 RE 41 (8.4) 19 (13.4)  
Imaging, n (%)   .006 
 Radiograph of neck or chest 181 (37.2) 71 (50.0)  
 No radiograph 305 (62.8) 71 (50.0)  
Steroids in ED, n (%)   .241 
 No steroid treatment 85 (17.5) 31 (21.8)  
 Any steroids in ED 401 (82.5) 111 (78.2)  
Hospital LOS   <.0001 
 Mean (SD) 18.8 (9.3) 36.5 (29.6)  
 Median (range) 17.0 (1.2–111.4) 27.7 (6.7–285.8d 
a

Significant intervention includes additional RE, transfer to PICU, or heliox (5 patients had additional RE and were transferred to the PICU, 137 patients had additional RE only, no patients were treated with heliox).

b

P values are from χ2 tests for categorical variables, 2-sample t tests of mean for patient age and O2 saturation, and Wilcoxon rank-sum test for RE in ED and hospital LOS.

c

The number for initial vitals is <628 because of missing data for some patients.

d

Without the outlier value of 285.8 for hospital LOS, mean (SD) and median (range) for those with significant interventions were 34.8 (20.8) and 27.4 (6.7, 119.8), respectively. The Wilcoxon rank-sum test for hospital LOS was still significant when the outlier value was excluded (P < .0001).

Comparisons of patient characteristics for children who did not require significant interventions versus those who received significant interventions are provided in Table 3. Patients were more likely to receive significant interventions if they had tachycardia, tachypnea on arrival to ED, and if they had any ED radiographs of the neck or chest. A logistic regression model was used to estimate ORs of significant intervention after admission adjusted for patient characteristics and ED interventions. Adjusted for other factors in the model, tachypnea (P = .002) and use of imaging in the ED (P = .018) were significantly associated with significant inpatient interventions (Table 4). The number of RE administered in the ED was not significantly associated with odds of significant inpatient interventions.

TABLE 4

Logistic Regression Model for Significant Intervention (Any Versus None) After Admission Adjusted for Factors in the Model

EffectAdjusted OR Estimate95% Wald Confidence LimitsP
Age, each additional mo 1.001 0.972 1.030 .955 
Sex, girls versus boys 1.187 0.772 1.826 .434 
Age-defined tachycardia, yes versus no 1.212 0.777 1.892 .397 
Age-defined tachypnea, yes versus no 2.496 1.406 4.430 .002 
Temperature, ≥38°C vs <38°C 0.815 0.500 1.328 .411 
O2 saturation, 91%–94% vs ≥95% 1.227 0.418 3.598 .709 
RE in ED, 3 RE vs 1–2 RE doses 1.812 0.943 3.482 .074 
Steroids in ED, any steroid versus none 0.743 0.438 1.260 .270 
Imaging, radiograph of neck, chest, or both versus no radiograph 1.680 1.094 2.580 .018 
EffectAdjusted OR Estimate95% Wald Confidence LimitsP
Age, each additional mo 1.001 0.972 1.030 .955 
Sex, girls versus boys 1.187 0.772 1.826 .434 
Age-defined tachycardia, yes versus no 1.212 0.777 1.892 .397 
Age-defined tachypnea, yes versus no 2.496 1.406 4.430 .002 
Temperature, ≥38°C vs <38°C 0.815 0.500 1.328 .411 
O2 saturation, 91%–94% vs ≥95% 1.227 0.418 3.598 .709 
RE in ED, 3 RE vs 1–2 RE doses 1.812 0.943 3.482 .074 
Steroids in ED, any steroid versus none 0.743 0.438 1.260 .270 
Imaging, radiograph of neck, chest, or both versus no radiograph 1.680 1.094 2.580 .018 

In our cohort of 628 children admitted to the regular inpatient floor with croup, >3 quarters (77.4%) did not receive any significant interventions after admission from the ED. Of the children receiving interventions after admission, the majority (62.7%) received a single dose of RE, but a minority (20.4%) were treated with a considerable number of RE doses or were transferred to the PICU. Our proportion of children receiving no further RE nebulization is higher than in previous studies of inpatient management of croup by Narayanan and Funkhouser10  and Rudinsky et al11  who reported rates of 51% and 34%, respectively. Our cohort before excluding patients who were hypoxic or directly admitted to the PICU consisted of 660 patients, 25(3.8%) of whom were admitted to the PICU at some point during their stay. This is lower than the average of 9% that Tyler et al3  reported. Whether this implies a more stringent use of resources after admission or a lower threshold to admit is hard to speculate because neither we nor Narayanan and Funkhouser10  or Rudinsky et al11  reported the proportion of patients presenting to the ED with croup being admitted, but it is noteworthy that almost one-third of our patients were admitted after only 1 RE nebulization in the ED. Tyler et al3  reported a significant variation in the management of croup between hospitals, and our results may also reflect the lack of published data to guide physicians caring for children with croup.

In our study, hospital LOS was significantly shorter for children not treated with RE after admission. We found that each additional administration of RE was associated with increased time spent as an inpatient, with median LOS being 6 hours longer for patients receiving 1 dose of RE compared with none and 25 hours longer for patients receiving 2 or more doses of RE after admission. Several factors may be responsible for this increase, which is longer than the typically recommended observation time after RE treatment. Time of admission and workflow on the wards likely play a role because rounds and, subsequently, discharges are often performed in the mornings; thus, a child who receives another RE dose midday may not be discharged until the next morning.

Our secondary outcome was determining if ED management or presenting vitals could predict children requiring significant intervention posthospital admission. We found that children presenting with tachypnea and those with radiographs performed in ED had increased odds of a significant intervention posthospital admission. Although tachypnea is an unspecific finding associated with respiratory distress of many kinds, it is possible or even likely that children with an elevated respiratory rate had a more severe presentation of croup, which may be more likely to require further treatment.

Neck and chest radiographs can be an adjunct in the evaluation of croup if the diagnosis is unclear12  but are not routinely indicated. In our study, 40.1% of children had some imaging performed, whereas only 1 child had an additional diagnosis of pneumonia. Causes of stridor other than croup include epiglottitis, bacterial tracheitis, or foreign body to name a few. By virtue of this study’s inclusion criteria, children with a primary diagnosis other than croup would not have been included in the cohort, and it is beyond the scope of this study to determine the rate of children with crouplike symptoms eventually being diagnosed with a different etiology. In our study, however, we did find that children diagnosed with croup who had imaging performed in the ED were more likely to receive significant interventions after admission. We speculate that this subset of children may have presented with more severe symptoms, prompting the provider to evaluate for differential etiologies. It may also again point to the variation in care provided to children with croup.

This study has several limitations. Because of its retrospective nature, data were limited to administrative data completed with chart reviews. Missed cases from misidentification errors could have occurred. Children were included from 3 different hospitals without use of croup scores, uniform admission, inpatient, or discharge guidelines or protocols. This study was not powered enough to control potentially important factors such as individual clinician practice, time of the year, ED, or hospital census at the time of admission, to name a few. In addition, the administration of intravenous fluids was not included in the analysis, and although not typically indicated for croup, it is possible that some children’s LOS was increased because of the need for intravenous hydration, which could have altered the results. As Tyler et al3  pointed out, there is significant variation in croup treatment between hospitals, and this may also be true between different providers in the same hospital. A lack of standardization between or within hospitals may increase or decrease admission rates and LOS and even lead to the overtreatment of children with RE and could make the results less generalizable.

More than three-fourths of children admitted with croup after evaluation and treatment from 3 pediatric EDs received no significant interventions after admission. Predictors for significant interventions were tachypnea on initial vital signs and imaging performed in ED. Our results reveal the largest proportion of children admitted with croup receiving no further RE or other significant interventions after admission in studies to date. In our results, a need for the standardization of admission criteria and hospital-based interventions for croup is also suggested.

Dr Asmundsson conducted the data collection, drafted and edited the manuscript, and approved the final manuscript as submitted; Ms Davey performed all data analysis and edited the final manuscript; Drs Arms, Kaila, Theiler, and Roback critically reviewed the manuscript and approved the final manuscript as submitted; Dr Louie conceptualized and designed the study and critically reviewed and approved the final manuscript as submitted; and all authors approved the final manuscript as submitted.

FUNDING: Supported by the National Institutes of Health’s National Center for Advancing Translational Sciences, grant UL1TR002494. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health’s National Center for Advancing Translational Sciences. Funded by the National Institutes of Health (NIH).

1
Marx
A
,
Török
TJ
,
Holman
RC
,
Clarke
MJ
,
Anderson
LJ
.
Pediatric hospitalizations for croup (laryngotracheobronchitis): biennial increases associated with human parainfluenza virus 1 epidemics
.
J Infect Dis
.
1997
;
176
(
6
):
1423
1427
2
Bjornson
C
,
Russell
K
,
Vandermeer
B
,
Klassen
TP
,
Johnson
DW
.
Nebulized epinephrine for croup in children
.
Cochrane Database Syst Rev
.
2013
;(
10
):
CD006619
3
Tyler
A
,
McLeod
L
,
Beaty
B
, et al
.
Variation in inpatient croup management and outcomes
.
Pediatrics
.
2017
;
139
(
4
):
e20163582
4
Chin
R
,
Browne
GJ
,
Lam
LT
,
McCaskill
ME
,
Fasher
B
,
Hort
J
.
Effectiveness of a croup clinical pathway in the management of children with croup presenting to an emergency department
.
J Paediatr Child Health
.
2002
;
38
(
4
):
382
387
5
Kelley
PB
,
Simon
JE
.
Racemic epinephrine use in croup and disposition
.
Am J Emerg Med
.
1992
;
10
(
3
):
181
183
6
Prendergast
M
,
Jones
JS
,
Hartman
D
.
Racemic epinephrine in the treatment of laryngotracheitis: can we identify children for outpatient therapy?
Am J Emerg Med
.
1994
;
12
(
6
):
613
616
7
Rizos
JD
,
DiGravio
BE
,
Sehl
MJ
,
Tallon
JM
.
The disposition of children with croup treated with racemic epinephrine and dexamethasone in the emergency department
.
J Emerg Med
.
1998
;
16
(
4
):
535
539
8
United Medical Education
.
PALS algorithms 2018 (pediatric advanced life support)
.
Available at: https://www.acls-pals-bls.com/algorithms/pals/. Accessed July 11, 2018
9
O’Leary
F
,
Hayen
A
,
Lockie
F
,
Peat
J
.
Defining normal ranges and centiles for heart and respiratory rates in infants and children: a cross-sectional study of patients attending an Australian tertiary hospital paediatric emergency department
.
Arch Dis Child
.
2015
;
100
(
8
):
733
737
10
Narayanan
S
,
Funkhouser
E
.
Inpatient hospitalizations for croup
.
Hosp Pediatr
.
2014
;
4
(
2
):
88
92
11
Rudinsky
SL
,
Sharieff
GQ
,
Law
W
,
Kanegaye
JT
.
Inpatient treatment after multi-dose racemic epinephrine for croup in the emergency department
.
J Emerg Med
.
2015
;
49
(
4
):
408
414
12
Walner
DL
,
Ouanounou
S
,
Donnelly
LF
,
Cotton
RT
.
Utility of radiographs in the evaluation of pediatric upper airway obstruction
.
Ann Otol Rhinol Laryngol
.
1999
;
108
(
4
):
378
383

Competing Interests

POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.

FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.

Supplementary data