Abstract
Little is known about the relationship between specific composition of the airway microbiota and severity of bronchiolitis. We aimed to identify nasopharyngeal microbiota profiles and to link these profiles to acute severity in infants hospitalized for bronchiolitis.
We conducted a multicenter prospective cohort study of 1005 infants (age <1 year) hospitalized for bronchiolitis over three winters, 2011-2014. By applying a 16S rRNA gene sequence and clustering approach to the nasopharyngeal aspirates collected within 24 hours of hospitalization, we determined nasopharyngeal microbiota profiles and their association with bronchiolitis severity. The primary outcome was intensive care use – i.e., admission to an intensive care unit or use of mechanical ventilation.
We identified four distinct nasopharyngeal microbiota profiles – three profiles were dominated by either Haemophilus, Moraxella, or Streptococcus, while the fourth profile had the highest bacterial richness. The rate of intensive care use was highest in infants with a Haemophilus-dominant profile and lowest in those with a Moraxella-dominant profile (20.2% vs 12.3%; unadjusted OR, 1.81; 95%CI, 1.07-3.11; P=0.03). After adjusting for 11 patient-level confounders, the rate remained significantly higher in infants with a Haemophilus-dominant profile (OR, 1.98; 95%CI, 1.08-3.62; P=0.03). These findings were externally validated in a separate cohort of 307 children hospitalized for bronchiolitis.
Introduction
Bronchiolitis is a significant public health problem worldwide (1-3). Indeed, bronchiolitis is the leading cause of hospitalizations for U.S. infants, accounting for 18% of all infant hospitalizations (3). The severity of bronchiolitis can range from a minor nuisance to fatal infection; however, the reasons for this severity difference are not fully explained by traditional risk factors, such as age, prematurity, and viral etiology (4, 5).
While common viruses and bacteria are identifiable using conventional techniques (e.g., cultures), these pathogens represent only a small fraction of the microbes living within any one individual (6, 7). The recent advances of genome sequencing have revealed the presence of highly functional bacterial communities inhabiting humans – the microbiota. Emerging evidence shows that airway microbiota may influence immune responses (8-13), suggesting a role of airway microbiota in the development and morbidity of acute respiratory infections (ARIs), including bronchiolitis (14). However, the limited literature about the association of airway microbiota with ARI incidence during infancy is conflicting. Specifically, ARI incidence has been found to be lower (15) and higher (16) among infants with more abundant Moraxella genus in their nasopharynx. None of the aforementioned studies of airway microbiota have externally validated their findings. Moreover, to the best of our knowledge, no studies have investigated the airway microbiota and its relation with severity of illness in infants hospitalized for bronchiolitis – a vulnerable population with high morbidity.
In this context, we conducted a multicenter prospective cohort study of infants hospitalized for bronchiolitis to identify airway microbiota profiles and to link these profiles to acute severity (i.e., intensive care use and hospital length-of-stay). We also externally validated the findings in a separate multicenter study of children hospitalized for bronchiolitis.
Materials and Methods
Study Design, Setting, and Participants
We conducted a multicenter prospective cohort study of infants (age <1 year) hospitalized for bronchiolitis (severe bronchiolitis). This study, called the 35th Multicenter Airway Research Collaboration (MARC-35) (17), was coordinated by the Emergency Medicine Network (EMNet)(18), a collaboration of 235 participating hospitals.
Using a standardized protocol, site investigators at 17 sites across 14 U.S. states (Table E1 in the Online Supplement) enrolled infants hospitalized with an attending physician diagnosis of bronchiolitis during three consecutive bronchiolitis seasons from November 1 to April 30 (2011-2014). Bronchiolitis was defined by the American Academy of Pediatrics guidelines – acute respiratory illness with some combination of rhinitis, cough, tachypnea, wheezing, crackles, and retractions (19). We excluded infants with previous enrollment, those who were transferred to a participating hospital >24 hours after the original hospitalization, those who were consented >24 hours after hospitalization, or those with known heart-lung disease, immunodeficiency, immunosuppression, or gestational age <32 weeks. All patients were treated at the discretion of the treating physician. The institutional review board at each of the participating hospitals approved the study. Written informed consent was obtained from the parent or guardian.
Data Collection
Investigators conducted a structured interview that assessed patients' demographic characteristics, medical and family history, and details of the acute illness. Emergency department and hospital chart reviews provided further clinical data, such as vital signs, physical examination, medical management, and disposition. All data were reviewed at the EMNet Coordinating Center and site investigators were queried about missing data and discrepancies identified by manual data checks.
Nasopharyngeal aspirates were collected by trained site investigators using the same standardized protocol utilized in a previous cohort study of children with bronchiolitis (4, 20). All sites used the same collection equipment (Medline Industries, Mundelein, IL) and collected the samples within 24 hours of hospitalization. The nasopharyngeal sample was added to transport medium, immediately placed on ice and then stored at −80°C. Frozen samples were shipped in batches on dry ice to Baylor College of Medicine, where they were tested for: 1) 17 respiratory viruses (e.g., respiratory syncytial virus [RSV], rhinovirus) by using real time polymerase chain reaction (PCR) assays (4, 20, 21), and 2) microbiota by using 16S rRNA gene sequencing.
16S rRNA Gene Sequencing and Compositional Analysis
16S rRNA gene sequencing methods were adapted from the methods developed for the NIH-Human Microbiome Project (6, 7). The details of the methods are described in the Online Supplement (Supplemental Methods). Briefly, bacterial genomic DNA was extracted using MO BIO PowerSoil DNA Isolation Kit (Mo Bio Laboratories; Carlsbad, CA). The 16S rDNA V4 region was amplified by PCR and sequenced in the MiSeq platform (Illumina; SanDiego, CA) using the 2×250 bp paired-end protocol yielding pair-end reads that overlap almost completely. The primers used for amplification contain adapters for MiSeq sequencing and single-end barcodes allowing pooling and direct sequencing of PCR products (22, 23).
Sequencing read pairs were demultiplexed based on the unique molecular barcodes, and reads were merged using USEARCH v7.0.1090 (24). Rarefaction curves of bacterial operational taxonomic units (OTUs) were constructed using sequence data for each sample to ensure coverage of the bacterial diversity present. Samples with suboptimal amounts of sequencing reads were re-sequenced to ensure that the majority of bacterial taxa were encompassed in our analyses.
16S rRNA gene sequences were clustered into OTUs at a similarity cutoff value of 97% using the UPARSE algorithm (25). OTUs were determined by mapping the centroids to the SILVA database (26) containing only the 16S V4 region to determine taxonomies. A custom script constructed a rarefied OTU table (rarefaction was performed at only one sequence depth) from the output files generated in the previous two steps for downstream analyses of alpha-diversity (e.g., Shannon index) and beta-diversity (e.g., weighted UniFrac)(27, 28). We utilized multiple quality control measures, including the use of non-template controls, at the microbial DNA extraction, 16S rRNA gene amplification, and amplicon sequencing processes. The details of the quality control measures are described in the Online Supplement (Supplemental Methods).
Outcome Measures
The primary outcome was intensive care use, defined as admission to intensive care unit and/or use of mechanical ventilation (continuous positive airway pressure and/or intubation during inpatient stay, regardless of location) at any time during the index hospitalization (20, 29). The secondary outcome was hospital length-of-stay of ≥3 days, defined using the median length-of-stay of 2 days and similar to the approach used in prior studies (4, 20).
Statistical Analyses
For each nasopharyngeal sample, the relative abundance of each OTU was calculated. Analyses were conducted at the genus-level; as each genus was dominated by one OTU (e.g., the detected Haemophilus genus consisted solely of one OTU), all OTUs assigned to the same genus were collapsed into a single group for reporting (16). To identify nasopharyngeal microbiota profiles, we performed unbiased clustering by partitioning around medoids (PAM) (30) using weighted UniFrac distance. Each PAM cluster is defined by a point designated as the center (the “medoid”) and minimizes the distance between samples in a cluster. The number of clusters to choose for the data was determined using the average silhouette score (Figure E1) (31).
Additionally, as the presence and/or abundance of an individual genus likely interacts and influences other genera in the microbial community, we displayed the microbiota association network on the basis of the approach of Faust et al. (32). The details of the method are described in the Online Supplement (Supplemental Methods).
Next, across the identified nasopharyngeal microbiota profiles, we compared patient characteristics and hospital course using chi-square test or Kruskal-Wallis test as appropriate. To examine the association of microbiota profiles with the severity outcomes (i.e., intensive care use and hospital length-of-stay of ≥3 days), we constructed two logistic regression models for each outcome with the Moraxella-dominant profile as the reference (15). First, we fitted an unadjusted model that included only microbiota profiles as the independent variable. Second, we constructed a 2-level mixed-effects model to account for patient clustering at the hospital level. We also adjusted for 11 patient-level variables (i.e., age, sex, race/ethnicity, gestational age, history of breathing problems, daycare attendance, siblings at home, lifetime history of antibiotic use, history of corticosteroid use, use of antibiotics during the pre-hospitalization visit, and respiratory viruses detected by PCR). We chose these potential confounders on the basis of clinical plausibility and a priori knowledge (4, 5, 20, 29, 33). We did not adjust for markers of severity (e.g., vital signs) because these were considered intermediate factors in the association of interest.
We performed sensitivity analyses to assess the robustness of our findings. First, we repeated the analysis stratifying by viral pathogen, similar to the approach used in prior studies (4, 34, 35). Second, we modeled length-of-stay as a count variable by fitting a mixed-effects Poisson regression model. Analyses used R version 3.2 with the lme4 package for the mixed effects models and the phyloseq package (36). All P-values were two-tailed, with P<0.05 considered statistically significant.
Validation of Microbiota Profiles and Outcome Models
Next, to validate our findings in MARC-35, we analyzed data from another 16-center prospective cohort of children aged <2 years with severe bronchiolitis (MARC-30) (4, 20), in which nasopharyngeal samples were collected using the same methods as in MARC-35. The technical methods (i.e., microbial DNA extraction, 16S rRNA gene amplification, and amplicon sequencing) used for MARC-30 were also identical to those for MARC-35. Using nasopharyngeal samples from 156 children with intensive care use and 156 children with no intensive care use and a length-of-stay ≤1 day (i.e., nested case-control sample with a greater severity contrast), we tested for the nasopharyngeal microbiota, identified microbiota profiles, and determined the associations of these profiles with the severity outcomes.
Results
Population and Sequence
At 17 participating hospitals, 1,016 infants with severe bronchiolitis were enrolled in MARC-35. We analyzed nasopharyngeal samples from all of the enrolled infants by 16S rRNA gene sequencing and obtained 17,399,260 high-quality merged sequences, of which 16,685,637 (95.9%) were mapped to the database. Of 1,016 infant samples, 1,005 (98.9%) had sufficient sequence depth (rarefaction cutoff: 2,128 reads per sample) and were eligible for the current analysis. Among the analytic cohort, the median age at hospitalization was 3 months (IQR, 2-6 months), 60.0% were male, and 42.6% were non-Hispanic white. The sequencing identified 24 phyla and 379 genera. The nasopharyngeal microbiota was dominated by 3 genera – Streptococcus (30.9%), Moraxella (29.8%), and Haemophilus (20.2%) – followed by Prevotella (2.3%) and Staphylococcus (2.2%).
Nasopharyngeal Microbiota Profiles and Network
PAM clustering of nasopharyngeal microbiota identified 4 distinct microbiota profiles: 1) Haemophilus-dominant profile (19.2%), 2) Moraxella-dominant profile (21.9%), 3) Streptococcus-dominant profile (28.2%), and 4) a mixed profile (30.7%) (Figure 1). The first 3 profiles were dominated by either Haemophilus, Moraxella, or Streptococcus genera. By contrast, the mixed profile had the highest bacterial richness (P<0.001) and alpha-diversity index (Shannon index, P<0.001) with highest relative abundance of Prevotella, Alloprevotella, and Veillonella (Table E2). Subclustering of the mixed profile identified three subclusters: a subcluster co-dominated by Streptococcus and Moraxella, a subcluster co-dominated by Streptococcus and Haemophilus, and a subcluster with a relatively high abundance of Streptococcus, Prevotella and Alloprevotella (Table E3).
Figure 1. Clustering and Composition in Nasopharyngeal Microbiota of 1005 Infants Hospitalized for Bronchiolitis, MARC-35.
All nasopharyngeal microbiota profiles of infants were clustered using partitioning around medoids clustering method with weighted UniFrac distance. Colored bars indicate 4 microbiota profiles: Haemophilus-dominant profile (green), Moraxella-dominant profile (red), Streptococcus-dominant profile (yellow), and mixed profile (blue). The optimal number of clusters was identified by the Silhouette index. To obtain further information about the bacterial composition of samples within microbiota profiles, the 10 most abundant genera present in an adjacent heatmap were displayed. The taxonomy depicted is on the genus level because each genus was dominated by one operational taxonomic unit.
HDP = Haemophilus-dominant profile; MDP = Moraxella-dominant profile; SDP = Streptococcus-dominant profile; MP = mixed profile.
The presence or abundance of individual bacteria likely affects those of others due to ecologic interactions. To examine the community structure of the microbiota, we created a microbiota association network (Figure E2). Haemophilus, Moraxella, and Streptococcus genera were negatively associated with each other, which is consistent with the observed Haemophilus-dominant, Moraxella-dominant, and Streptococcus-dominant profiles. The Streptococcus genus was positively associated with Veillonella, which was also positively associated with Prevotella and Alloprevotella. This community structure was consistent with the mixed microbiota profile.
Microbiota Profiles and Patient Characteristics
Patient characteristics differed across the four microbiota profiles (Table 1). For example, infants with the Haemophilus-dominant profile, compared to the others, were older, less likely to be non-Hispanic white, and more likely to have used antibiotics before the index hospitalization (all P<0.05). At presentation to the hospital, infants with the Haemophilus-dominant profile also had higher weight, and were more likely to receive antibiotics during prehospitalization visit (e.g., emergency department before the enrollment) and have sole rhinovirus infection (all P<0.001), compared to the others. By contrast, there was no significant difference in the vital signs across the profiles.
Table 1. Characteristics and Clinical Presentation of Infants Hospitalized for Bronchiolitis by Nasopharyngeal Microbiota Profile in the MARC-35 Cohort.
| Haemophilus-dominant profile | Moraxella-dominant profile | Streptococcus-dominant profile | Mixed profile | P-value | |
|---|---|---|---|---|---|
| Variables | n=193 (19.2%) | n=220 (21.9%) | n=283 (28.2%) | n=309 (30.7%) | |
| Characteristics | |||||
| Age (mo), median, (IQR) | 4 (3-8) | 3 (2-6) | 3 (1-5) | 3 (2-6) | <0.001 |
| <2 | 39 (20.2) | 69 (31.4) | 108 (38.2) | 93 (30.1) | <0.001 |
| 2-5.9 | 82 (42.5) | 92 (41.8) | 127 (44.9) | 149 (48.2) | |
| 6-12 | 72 (37.3) | 59 (26.8) | 48 (17.0) | 67 (21.7) | |
| Male sex | 112 (58.0) | 127 (57.7) | 172 (60.8) | 192 (62.1) | 0.70 |
| Race/ethnicity | 0.02 | ||||
| Non-Hispanic white | 71 (36.8) | 89 (40.5) | 145 (51.2) | 123 (39.8) | |
| Non-Hispanic black | 39 (20.2) | 58 (26.4) | 57 (20.1) | 79 (25.6) | |
| Hispanic | 76 (39.4) | 65 (29.5) | 72 (25.4) | 93 (30.1) | |
| Other | 7 (3.6) | 8 (3.6) | 9 (3.2) | 14 (4.5) | |
| Parental history of asthma | 58 (30.1) | 68 (30.9) | 109 (38.8) | 106 (34.4) | 0.16 |
| Maternal smoking during pregnancy | 21 (11.3) | 25 (11.7) | 40 (14.3) | 58 (19.0) | 0.051 |
| Mode of birth | 0.95 | ||||
| Vaginal birth | 122 (63.2) | 143 (65) | 179 (63.3) | 199 (64.4) | |
| C-section | 64 (34.4) | 72 (33.5) | 101 (36.1) | 106 (34.8) | |
| Prematurity (32-37 weeks) | 33 (17.1) | 35 (15.9) | 51 (18.0) | 64 (20.7) | 0.52 |
| Previous breathing problems before the index hospitalization* | 49 (25.4) | 38 (17.3) | 55 (19.4) | 61 (19.8) | 0.42 |
| History of eczema | 27 (14.0) | 30 (13.6) | 37 (13.1) | 52 (16.8) | 0.58 |
| Ever attended daycare | 52 (26.9) | 60 (27.3) | 50 (17.7) | 68 (22.0) | 0.03 |
| Aeroallergen sensitization† | 1 (0.7) | 3 (1.8) | 4 (1.7) | 5 (2.1) | 0.75 |
| Food sensitization† | 37 (19.2) | 42 (19.1) | 45 (15.9) | 55 (17.8) | 0.75 |
| Sibling at home | 144 (74.6) | 182 (82.7) | 223 (78.8) | 250 (80.9) | 0.20 |
| Mostly breastfed for the first 3 months of age | 87 (49.7) | 94 (49.2) | 123 (50.6) | 115 (42.9) | 0.29 |
| Smoke exposure at home | 30 (15.5) | 33 (15.0) | 39 (13.8) | 51 (16.5) | 0.83 |
| Antibiotic use before index hospitalization | 92 (47.7) | 46 (20.9) | 84 (29.7) | 90 (29.1) | <0.001 |
| Corticosteroid use before index hospitalization | 32 (16.6) | 34 (15.5) | 35 (12.4) | 45 (14.6) | 0.60 |
| Clinical presentation | |||||
| Month of hospitalization | 0.005 | ||||
| November | 13 (6.7) | 29 (13.2) | 18 (6.4) | 22 (7.1) | |
| December | 26 (13.5) | 56 (25.5) | 46 (16.3) | 58 (18.8) | |
| January | 59 (30.6) | 56 (25.5) | 91 (32.2) | 98 (31.7) | |
| February | 42 (21.8) | 45 (20.5) | 76 (26.9) | 78 (25.2) | |
| March | 37 (19.2) | 23 (10.5) | 34 (12.0) | 36 (11.7) | |
| April | 16 (8.3) | 11 (5.0) | 18 (6.4) | 17 (5.5) | |
| Duration of breathing problem before the index hospitalization (day), median (IQR) | 4 (2-5) | 3 (2-5) | 3 (2-5) | 3 (2-4) | 0.14 |
| Weight at presentation (kg), median (IQR) | 7.0 (5.2-8.2) | 6.1 (4.80-7.93) | 5.5 (4.4-7.2) | 5.9 (4.8-7.4) | <0.001 |
| Respiratory rate at presentation (per minute), median (IQR) | 48 (40-60) | 50 (40-60) | 48 (40-60) | 48 (40-60) | 0.48 |
| Oxygen saturation at presentation | 0.20 | ||||
| <90% | 26 (13.8) | 12 (5.6) | 24 (8.7) | 27 (8.9) | |
| 90%-93% | 33 (17.6) | 28 (13.1) | 47 (17.0) | 45 (14.8) | |
| ≥94% | 129 (68.6) | 174 (81.3) | 205 (74.3) | 233 (76.4) | |
| Retractions on examination | 0.003 | ||||
| None | 28 (14.5) | 49 (22.5) | 66 (23.4) | 47 (15.3) | |
| Mild | 74 (38.3) | 105 (48.2) | 116 (41.1) | 131 (42.5) | |
| Moderate/severe | 83 (43.0) | 55 (25.2) | 95 (33.7) | 121 (39.3) | |
| Wheezing on examination | 106 (54.9) | 144 (65.5) | 154 (54.4) | 192 (62.1) | 0.10 |
| Received antibiotics during prehospitalization visit | 45 (23.3) | 13 (6.0) | 58 (20.7) | 59 (19.4) | <0.001 |
| Received corticosteroids during pre-hospitalization visit | 21 (10.9) | 16 (7.3) | 26 (9.3) | 26 (8.6) | 0.64 |
| Virology | <0.001 | ||||
| Sole RSV infection | 83 (43.0) | 119 (54.1) | 202 (71.4) | 176 (57.0) | |
| Sole rhinovirus infection | 24 (12.4) | 15 (6.8) | 7 (2.5) | 14 (4.5) | |
| RSV + rhinovirus coinfection | 20 (10.4) | 30 (13.6) | 24 (8.5) | 46 (14.9) | |
| RSV + non-rhinovirus pathogens | 32 (16.6) | 21 (9.5) | 27 (9.5) | 33 (10.7) | |
| Rhinovirus + non-RSV pathogens | 9 (4.7) | 13 (5.9) | 4 (1.4) | 5 (1.6) | |
| Neither RSV nor rhinovirus‡ | 19 (9.8) | 18 (8.2) | 12 (4.2) | 25 (8.1) | |
| No viral pathogens | 6 (3.1) | 4 (1.8) | 7 (2.5) | 10 (3.2) | |
| Viral genomic load (CT-value), median IQR | |||||
| RSV | 23 (21-26) | 22 (20-24) | 23 (21-25) | 23 (21-26) | <0.001 |
| RV | 29 (26-34) | 28 (26-35) | 28 (27-36) | 30 (27-35) | 0.89 |
| Outcomes | —‖ | ||||
| Intensive care use‡ | 39 (20.2) | 27 (12.3) | 48 (17.0) | 47 (15.2) | |
| Hospital length-of-stay ≥3 days | 99 (51.3) | 73 (33.2) | 111 (39.2) | 112 (36.2) | |
| Hospital length-of-stay (day), median (IQR) | 3 (2-4) | 2 (1-3) | 2 (1-3) | 2 (1-3) | |
Data are no. (%) of infants unless otherwise indicated. Percentages may not equal 100, because of missingness
Patient characteristics and hospital course were compared using chi-square test or Kruskal-Wallis test across the identified nasopharyngeal microbiota profiles.
Abbreviations: CT, cycle threshold, IQR, interquartile range; RSV, respiratory syncytial virus
Defined as an infant having cough that wakes him/her at night and/or causes emesis, or when the child has wheezing or shortness of breath without cough
Defined by having one or more positive values for allergen-specific IgE
Consisted of adenovirus (n=6), coronavirus (n=15), human bocavirus (n=9), human metapneumovirus (n=38), influenza virus (n=11), and parainfluenza virus (n=13). Note 18 patients had coinfection.
Defined as admission to intensive care unit and/or use of mechanical ventilation (continuous positive airway pressure and/or intubation during inpatient stay, regardless of location) at any time during the index hospitalization
Unadjusted P-values presented in Table 2
Microbiota Profiles and Bronchiolitis Severity
Overall, 161 (16.0%) infants hospitalized for bronchiolitis required intensive care use. The rate of intensive care use was highest in infants with a Haemophilus-dominant profile and lowest in those with a Moraxella-dominant profile (20.2% vs. 12.3%; unadjusted OR, 1.81; 95%CI, 1.07-3.11; P=0.03; Table 2). In the multivariable model adjusting for 11 patient characteristics (e.g., demographics, use of antibiotics during the pre-hospitalization visit, virus) and clustering at the hospital-level, the rate remained significantly higher in infants with a Haemophilus-dominant profile (OR for comparison with Moraxella-dominant profile, 1.98; 95%CI, 1.08-3.62; P=0.03; Tables 2 and E4). There was also a positive linear relationship of relative abundance of Haemophilus with rate of intensive care use after adjusting for 11 patient characteristics (P=0.03; Figure 2A). By contrast, compared to infants with Moraxella-dominant profile, the rate of intensive care use in those with a Streptococcus-dominant or mixed profile was not significantly different. Stratified analyses across virology strata are shown in Table E5. Although there is limited statistical power in most viral strata, infants with RSV infection and a Haemophilus-dominant profile had higher odds of intensive care use. Additionally, in another sensitivity analysis (n=978) that excluded all infants with no viral pathogen detected, the results did not change materially – i.e., the rate of intensive care use remained significantly higher in infants with a Haemophilus-dominant profile (OR for comparison with Moraxella-dominant profile, 2.15; 95%CI, 1.16-3.99; P=0.01).
Table 2. Unadjusted and Multivariable Associations of Nasopharyngeal Microbiota Profiles with Bronchiolitis Outcomes, MARC-35 Cohort.
| Unadjusted model | Adjusted model* | |||
|---|---|---|---|---|
| Outcome by microbiota profile | OR (95% CI) | P-value | OR (95% CI) | P-value |
| Intensive care use† | ||||
| Haemophilus-dominant profile | 1.81 (1.07-3.11) | 0.03 | 1.98 (1.08-3.62) | 0.03 |
| Moraxella-dominant profile | Reference | Reference | ||
| Streptococcus-dominant profile | 1.46 (0.88-2.45) | 0.14 | 1.32 (0.74-2.34) | 0.34 |
| Mixed profile | 1.28 (0.78-2.16) | 0.34 | 1.19 (0.68-2.09) | 0.54 |
| Hospital length-of-stay ≥3 days | ||||
| Haemophilus-dominant profile | 2.12 (1.43-3.17) | <0.001 | 2.47 (1.60-3.83) | <0.001 |
| Moraxella-dominant profile | Reference | Reference | ||
| Streptococcus-dominant profile | 1.30 (0.90-1.88) | 0.16 | 1.06 (0.71-1.57) | 0.78 |
| Mixed profile | 1.14 (0.80-1.65) | 0.47 | 1.01 (0.68-1.48) | 0.97 |
Abbreviations: CI, confidence interval; OR, odds ratio
Bold results are statistically significant
Full adjusted models are included in Table E3.
Mixed-effects logistic regression model adjusting for 11 patient-level variables (age, sex, race/ethnicity, gestational age, history of breathing problems, daycare attendance, siblings at home, lifetime history of antibiotic use, history of corticosteroid use, use of antibiotics during the pre-hospitalization visit, and respiratory viruses detected by PCR) and sites as random effect
Defined as admission to intensive care unit and/or use of mechanical ventilation (continuous positive airway pressure and/or intubation during inpatient stay, regardless of location) at any time during the index hospitalization
Figure 2. Independent Association of Relative Abundance of Haemophilus Genus with the Rate of Severity Outcomes in Infants Hospitalized for Bronchiolitis, MARC-35.
Two-level mixed-effects models were constructed to account for patient clustering at the hospital level. The models adjusted for 11 patient-level variables (i.e., age, sex, race/ethnicity, gestational age, history of breathing problems, daycare attendance, siblings at home, lifetime history of antibiotic use, history of corticosteroid use, use of antibiotics during the pre-hospitalization visit, and respiratory viruses detected by PCR).
A) There was a positive linear association between relative abundance of Haemophilus genus and the rate of intensive care use (adjusted OR, 1.07 [per 0.1 increase in the relative abundance of Haemophilus]; 95%CI, 1.01-1.13; P=0.03).
B) There was a positive linear association between relative abundance of Haemophilus genus and the rate of a hospital length-of-stay of ≥3days (adjusted OR, 1.11 [per 0.1 increase in the relative abundance of Haemophilus]; 95%CI, 1.06-1.17; P<0.001).
Similar to results of the primary outcome, infants with a Haemophilus-dominant profile, compared to those with a Moraxella-dominant profile, had a higher rate of length-of-stay ≥3 days (51.3% vs. 33.2%; unadjusted OR, 2.12; 95%CI, 1.43-3.17;P<0.001; Tables 1 and 2). In the multivariable model, the rate remained significantly higher in infants with a Haemophilus-dominant profile (OR, 2.47; 95%CI, 1.60-3.83; P<0.001; Tables 2 and E4). There was also a positive linear relationship of relative abundance of Haemophilus with rate of length-of-stay ≥3 days (P<0.001; Figure 2B). Likewise, in the sensitivity analysis modeling length-of-stay as count variable (i.e., not as binary variable), the significant association between microbiota profiles and the outcome persisted (Table E6).
Validation of Profiles and Association with Severity
In 312 nasopharyngeal samples selected from children in the MARC-30 study, 307 (98.4%) had sufficient sequencing depth (rarefaction cutoff: 1,064 reads per sample) and were used for validation. The use of the PAM clustering method of microbiota, with the use of the average silhouette score, revealed 4 microbiota profiles (Figure E3) similar to those in MARC-35 – i.e. 3 profiles were dominated by either Haemophilus, Moraxella, or Streptococcus genera, while the fourth profile (i.e., mixed profile) had the highest bacterial richness and alpha-diversity levels (Table E7). In addition, the subclustering of the mixed profile identified a subcluster with a relatively high abundance of Streptococcus, Prevotella and Alloprevotella (Table E8). Likewise, the significant associations between the Haemophilus-dominant profile and the higher rate of severity outcomes were also observed, with larger effect sizes, in this nested case-control sample with a greater severity contrast – e.g., adjusted OR for intensive care use of Haemophilus-dominant profile in comparison with Moraxella-dominant profile, was 5.34 (95%CI, 1.96-14.5; P=0.001; Table E9).
Discussion
In this prospective multicenter cohort of 1005 infants with severe bronchiolitis, we identified 4 distinct microbiota profiles in their nasopharynx. Infants with Haemophilus-dominant profile had significantly higher rates of intensive care use and prolonged hospital length-of-stay than those with Moraxella-dominant profile. In contrast, the rate of outcomes in those with Streptococcus-dominant or mixed profile was not significantly different. These findings were externally validated in a separate multicenter study of severe bronchiolitis. To our knowledge, this is the first study to have examined the association of airway microbiota with severity in infants with severe bronchiolitis. Our data corroborate and build on previous reports linking bacteria composition in the airway to ARI outcomes (15, 16, 37-40), findings of both research and clinical importance.
Previous reports on ARI in young children have reported inconsistent relationships of airway bacteria with ARI incidence and severity. For example, Kloepfer et al., by applying quantitative PCR to nasal samples (n=380), found no associations between H. influenzae detection alongside rhinovirus and ARI severity in school-age children (37). Similarly, Carlson et al., by using a culture-dependent technique in the COPSAC2000 cohort (n=283), also found no association between H. influenzae detection in the hypopharynx and duration of wheezing episodes in children aged <3 years (38). By contrast, Teo et al., by using 16S rRNA gene sequence in the Childhood Asthma Study cohort (n=234) in Western Australia, reported that there were 6 nasopharyngeal microbiota profiles (Haemophilus, Moraxella, Streptococcus, Corynebacterium, Alloiococcus, and Staphylococcus) in infants with high risk of atopy and that Haemophilus-dominant nasopharyngeal microbiota was associated with higher incidence of ARI and higher severity (16). In addition to the apparent inconsistency in the Haemophilus-ARI link, we also observed a lower abundance of Corynebacterium, Dolosigranulum, and Staphylococcus genera in our population (i.e., young infants hospitalized with bronchiolitis) compared to that in the previous studies of healthy children (15) and children with mild upper respiratory infection (41, 42). Potential reasons for these discrepancies across studies include differences in patient populations, clinical settings (e.g., community vs. inpatient setting), sampling (e.g., anterior nasal swab vs. nasopharyngeal aspirate), and laboratory techniques (e.g., culture, PCR) for bacteria identification. By contrast, the validity of our findings is buttressed by the use of 16S rRNA gene sequencing of the nasopharyngeal microbiota, a sample size that is many times larger than any other prior study on this topic, and validation in an additional multicenter cohort of severe bronchiolitis.
The observed microbiota-severity association challenges the conventional virus-centric view of bronchiolitis. However, the nature of this microbial association warrants clarification. It is possible that there is a causal relationship – i.e., Haemophilus-dominant microbiota in the infant's airway alters immune responses to increase the severity of bronchiolitis. Indeed, Følsgaard et al. reported that H. influenzae colonization of the airways in asymptomatic neonates is associated with upregulated Th1/Th2/Th17-type inflammatory response of the upper airway mucosa (11). The presence of both Th2 and Th17 mediators suggests that colonization with H. influenzae may counteract the Th1 response needed to eradicate viral pathogens of bronchiolitis. Similarly, studies of primary bronchial epithelial cells demonstrated that exposure to H. influenzae upregulates intercellular adhesion molecule-1 expression and enhances chemokine release induced by subsequent infection by RSV (43) and rhinovirus (43, 44). Alternatively, Haemophilus-dominant microbiota may be simply a marker of an infant who is prone to develop more-severe bronchiolitis. Additionally, reverse causation – i.e., more-severe illness results in a rapid overgrowth of Haemophilus in the airway – is also possible (45). These possibilities are not mutually exclusive. Notwithstanding the complexity, the identification and validation of Haemophilus-dominant profile as the primary culprit in the association between airway microbiota and bronchiolitis severity is an important finding. Our data should facilitate further investigations (e.g., animal models, metagenomics, proteomics, metabolomics) to disentangle the complex web of the airway microbiome, viral pathogens, host immune responses, and bronchiolitis pathogenesis (14).
Potential Limitations
The study has several potential limitations. First, bronchiolitis is a disease of the lower airways, and our study was based on samples obtained from the infant's nasopharynx. However, lower airway sampling in infants is technically and ethically challenging. Nevertheless, prior studies have shown strong correlations between upper and lower airway microbiology (46) and virology (47) in children. Therefore, we believe that the nasopharyngeal microbiota in the infant is likely indicative of that in the lower respiratory tract. Second, with the use of 16S rRNA gene sequencing, we did not analyze the microbiota at the species-level, and therefore, were limited in the confirmation of their phenotypic features. Third, the study design precluded us from evaluating the relationship between succession of the airway microbial ecosystem and respiratory health in children. To address this important question, the study population is currently being followed longitudinally up to age 6 years with airway specimen sampling at multiple time-points. Fourth, as with any observational studies, the associations of microbiota profiles with bronchiolitis severity does not necessarily prove causality and might be confounded by unmeasured factors – e.g., virus genotypes, institutional variation in resource use. One may surmise that our inferences are biased by co-circulating viruses within hospitals or regions. However, MARC-35 enrolled patients at 17 states across 14 US states across three consecutive bronchiolitis seasons (2011-2014). Their findings were further validated in a separate 16-center cohort of severe bronchiolitis during 2007-2010. Furthermore, the observed association between the microbiota and bronchiolitis severity remained significant after accounting for patient-level clustering within hospital by the use of a mixed-effects model. Fifth, we did not have the information of a “control” group (e.g., infants hospitalized for non-respiratory events. However, the study objective is not to assess the role of microbiota on the development of bronchiolitis (yes/no) but to investigate its relationship with disease severity. Finally, although the study cohorts consisted of racially, ethnically, and geographically diverse U.S. sample of severe bronchiolitis, our inferences might not be generalizable to those in the ambulatory setting. Nevertheless, our data remain highly relevant for 130,000 children hospitalized yearly in the U.S.– a vulnerable population with high morbidity (3).
Conclusions
In this prospective multicenter cohort of 1005 infants hospitalized for bronchiolitis, we identified 4 distinct microbiota profiles in their nasopharynx. Infants with Haemophilus-dominant profile had a higher bronchiolitis severity – i.e., significantly higher rates of intensive care use and prolonged length-of-stay – than those with Moraxella-dominant profile. These findings were validated in a separate multicenter study of severe bronchiolitis. The findings should serve as an important starting point for further mechanistic and interventional investigations of the interplay between the airway microbiome, viral pathogens, host immune responses, and bronchiolitis pathogenesis.
Supplementary Material
Acknowledgments
This study was supported by the grants U01 AI-087881, R01 AI-114552, R01 AI-108588, and R21 HL-129909 from the National Institutes of Health (Bethesda, MD) and a grant from the William F. Milton Fund (Boston, MA). The content of this manuscript is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. We thank Ms. Pamela Luna at Rice University (Houston, TX) for her assistance with statistical analysis.
Footnotes
Study Collaborators: The following individuals were MARC-35 site investigators and are Collaborators on this paper. Anne K. Beasley, MD, Phoenix Children's Hospital, Phoenix, AZ; Juan C. Celedon, MD, DrPH, Children's Hospital of Pittsburgh, Pittsburgh, PA; Ari R. Cohen, MD, Massachusetts General Hospital, Boston, MA; Michelle B. Dunn, MD, Children's Hospital of Philadelphia, Philadelphia, PA; Michael Gomez, MD, MS-HCA, Children's Hospital at St. Francis, Tulsa, OK; Nancy Inhofe, MD, Children's Hospital at St. Francis, Tulsa, OK; Sujit Iyer, MD, Dell Children's Medical Center of Central Texas, Austin, TX; Federico R. Laham, MD, MS, Arnold Palmer Hospital for Children, Orlando, FL; Charles G. Macias, MD, MPH, Texas Children's Hospital, Houston, TX; Thida Ong, MD, Seattle Children's Hospital, Seattle, WA; Brian M. Pate, MD, The Children's Mercy Hospital & Clinics, Kansas City, MO; Wayne G. Schreffler, MD, PhD, Massachusetts General Hospital, Boston, MA; Michelle D. Stevenson, MD, MS, Kosair Children's Hospital, Louisville, KY; Richard T. Strait, MD, Cincinnati Children's Hospital and Medical Center, Cincinnati, OH; Stephen J. Teach, MD, MPH, Children's National Medical Center, Washington, D.C.; Henry T. Puls, MD, The Children's Mercy Hospital & Clinics, Kansas City, MO; Amy D. Thompson, MD, Alfred I. duPont Hospital for Children, Wilmington, DE; Vincent J. Wang, MD, MHA, Children's Hospital of Los Angeles, Los Angeles, CA; and Ilana Waynik, MD, Connecticut Children's Medical Center, Hartford, CT.
Declaration of Interests: Dr. Mansbach has provided bronchiolitis-related consultation for Regeneron. Drs. Ajami and Petrosino own shares at Diversigen Inc., a microbiome research company. Dr. Piedra provided bronchiolitis-related consultation for Gilead, Novavax, and Regeneron. The other authors have no financial relationships relevant to this article to disclose.
Take Home Message: Haemophilus-dominant airway microbiota was associated with higher risk of intensive care in infants with bronchiolitis.
References
- 1.Nair H, Nokes DJ, Gessner BD, Dherani M, Madhi SA, Singleton RJ, O'Brien KL, Roca A, Wright PF, Bruce N, Chandran A, Theodoratou E, Sutanto A, Sedyaningsih ER, Ngama M, Munywoki PK, Kartasasmita C, Simoes EA, Rudan I, Weber MW, Campbell H. Global burden of acute lower respiratory infections due to respiratory syncytial virus in young children: A systematic review and meta-analysis. Lancet. 2010;375:1545–1555. doi: 10.1016/S0140-6736(10)60206-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Hasegawa K, Tsugawa Y, Brown DF, Mansbach JM, Camargo CA., Jr Temporal trends in emergency department visits for bronchiolitis in the united states, 2006-2010. Pediatr Infect Dis J. 2014;33:11–18. doi: 10.1097/INF.0b013e3182a5f324. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Hasegawa K, Tsugawa Y, Brown DF, Mansbach JM, Camargo CA., Jr Trends in bronchiolitis hospitalizations in the united states, 2000-2009. Pediatrics. 2013;132:28–36. doi: 10.1542/peds.2012-3877. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Mansbach JM, Piedra PA, Teach SJ, Sullivan AF, Forgey T, Clark S, Espinola JA, Camargo CA, Jr Investigators M. Prospective multicenter study of viral etiology and hospital length of stay in children with severe bronchiolitis. Arch Pediatr Adolesc Med. 2012;166:700–706. doi: 10.1001/archpediatrics.2011.1669. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Hasegawa K, Mansbach JM, Camargo CA., Jr Infectious pathogens and bronchiolitis outcomes. Exp Rev Anti Infect Ther. 2014;12:817–828. doi: 10.1586/14787210.2014.906901. [DOI] [PubMed] [Google Scholar]
- 6.Human Microbiome Project C. A framework for human microbiome research. Nature. 2012;486:215–221. doi: 10.1038/nature11209. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Human Microbiome Project C. Structure, function and diversity of the healthy human microbiome. Nature. 2012;486:207–214. doi: 10.1038/nature11234. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Herbst T, Sichelstiel A, Schar C, Yadava K, Burki K, Cahenzli J, McCoy K, Marsland BJ, Harris NL. Dysregulation of allergic airway inflammation in the absence of microbial colonization. Am J Respir Crit Care Med. 2011;184:198–205. doi: 10.1164/rccm.201010-1574OC. [DOI] [PubMed] [Google Scholar]
- 9.Olszak T, An D, Zeissig S, Vera MP, Richter J, Franke A, Glickman JN, Siebert R, Baron RM, Kasper DL, Blumberg RS. Microbial exposure during early life has persistent effects on natural killer T cell function. Science. 2012;336:489–493. doi: 10.1126/science.1219328. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Hollams EM, Hales BJ, Bachert C, Huvenne W, Parsons F, de Klerk NH, Serralha M, Holt BJ, Ahlstedt S, Thomas WR, Sly PD, Holt PG. Th2-associated immunity to bacteria in teenagers and susceptibility to asthma. Eur Respir J. 2010;36:509–516. doi: 10.1183/09031936.00184109. [DOI] [PubMed] [Google Scholar]
- 11.Folsgaard NV, Schjorring S, Chawes BL, Rasmussen MA, Krogfelt KA, Brix S, Bisgaard H. Pathogenic bacteria colonizing the airways in asymptomatic neonates stimulates topical inflammatory mediator release. Am J Respir Crit Care Med. 2013;187:589–595. doi: 10.1164/rccm.201207-1297OC. [DOI] [PubMed] [Google Scholar]
- 12.Lynch SV. Viruses and microbiome alterations. Ann Am Thorac Soc. 2014;11(1):S57–60. doi: 10.1513/AnnalsATS.201306-158MG. [DOI] [PubMed] [Google Scholar]
- 13.Huang YJ, Nariya S, Harris JM, Lynch SV, Choy DF, Arron JR, Boushey H. The airway microbiome in patients with severe asthma: Associations with disease features and severity. J Allergy Clin Immunol. 2015;136:874–884. doi: 10.1016/j.jaci.2015.05.044. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Hasegawa K, Camargo CA., Jr Airway microbiota and acute respiratory infection in children. Exp Rev Clin Immunol. 2015;11:789–792. doi: 10.1586/1744666X.2015.1045417. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Biesbroek G, Tsivtsivadze E, Sanders EA, Montijn R, Veenhoven RH, Keijser BJ, Bogaert D. Early respiratory microbiota composition determines bacterial succession patterns and respiratory health in children. Am J Respir Crit Care Med. 2014;190:1283–1292. doi: 10.1164/rccm.201407-1240OC. [DOI] [PubMed] [Google Scholar]
- 16.Teo Shu M, Mok D, Pham K, Kusel M, Serralha M, Troy N, Holt Barbara J, Hales Belinda J, Walker Michael L, Hollams E, Bochkov Yury A, Grindle K, Johnston Sebastian L, Gern James E, Sly Peter D, Holt Patrick G, Holt Kathryn E, Inouye M. The infant nasopharyngeal microbiome impacts severity of lower respiratory infection and risk of asthma development. Cell Host Microbe. 2015;17(5):704–15. doi: 10.1016/j.chom.2015.03.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Mansbach JM, Hasegawa K, Henke DM, Ajami NJ, Petrosino JF, Shaw CA, Piedra PA, Sullivan AF, Espindola PA, Camargo CA., Jr Respiratory syncytial virus and rhinovirus severe bronchiolitis are associated with distinct nasopharyngeal microbiota. J Allergy Clin Immunol. 2016 doi: 10.1016/j.jaci.2016.01.036. Epub ahead of print. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Emergency medicine network. [cited May 13 2016]; Available from: http://www.emnet-usa.org/
- 19.Ralston SL, Lieberthal AS, Meissner HC, Alverson BK, Baley JE, Gadomski AM, Johnson DW, Light MJ, Maraqa NF, Mendonca EA, Phelan KJ, Zorc JJ, Stanko-Lopp D, Brown MA, Nathanson I, Rosenblum E, Sayles S, 3rd, Hernandez-Cancio S American Academy of P. Clinical practice guideline: The diagnosis, management, and prevention of bronchiolitis. Pediatrics. 2014;134:e1474–1502. doi: 10.1542/peds.2014-2742. [DOI] [PubMed] [Google Scholar]
- 20.Hasegawa K, Jartti T, Mansbach JM, Laham FR, Jewell AM, Espinola JA, Piedra PA, Camargo CA. Respiratory syncytial virus genomic load and disease severity among children hospitalized with bronchiolitis: Multicenter cohort studies in the us and finland. J Infect Dis. 2015;211:1550–1559. doi: 10.1093/infdis/jiu658. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Beckham JD, Cadena A, Lin J, Piedra PA, Glezen WP, Greenberg SB, Atmar RL. Respiratory viral infections in patients with chronic, obstructive pulmonary disease. J Infect. 2005;50:322–330. doi: 10.1016/j.jinf.2004.07.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Caporaso JG, Kuczynski J, Stombaugh J, Bittinger K, Bushman FD, Costello EK, Fierer N, Pena AG, Goodrich JK, Gordon JI, Huttley GA, Kelley ST, Knights D, Koenig JE, Ley RE, Lozupone CA, McDonald D, Muegge BD, Pirrung M, Reeder J, Sevinsky JR, Turnbaugh PJ, Walters WA, Widmann J, Yatsunenko T, Zaneveld J, Knight R. Qiime allows analysis of high-throughput community sequencing data. Nature Methods. 2010;7:335–336. doi: 10.1038/nmeth.f.303. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Caporaso JG, Lauber CL, Walters WA, Berg-Lyons D, Huntley J, Fierer N, Owens SM, Betley J, Fraser L, Bauer M, Gormley N, Gilbert JA, Smith G, Knight R. Ultra-high-throughput microbial community analysis on the illumina hiseq and miseq platforms. ISME J. 2012;6:1621–1624. doi: 10.1038/ismej.2012.8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Edgar RC. Search and clustering orders of magnitude faster than Blast. Bioinformatics. 2010;26:2460–2461. doi: 10.1093/bioinformatics/btq461. [DOI] [PubMed] [Google Scholar]
- 25.Edgar RC. Uparse: Highly accurate otu sequences from microbial amplicon reads. Nature Methods. 2013;10:996–998. doi: 10.1038/nmeth.2604. [DOI] [PubMed] [Google Scholar]
- 26.Quast C, Pruesse E, Yilmaz P, Gerken J, Schweer T, Yarza P, Peplies J, Glockner FO. The SILVA ribosomal rna gene database project: Improved data processing and web-based tools. Nucleic Acids Res. 2013;41:D590–596. doi: 10.1093/nar/gks1219. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Lozupone C, Lladser ME, Knights D, Stombaugh J, Knight R. Unifrac: An effective distance metric for microbial community comparison. ISME J. 2011;5:169–172. doi: 10.1038/ismej.2010.133. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Lozupone C, Knight R. Unifrac: A new phylogenetic method for comparing microbial communities. Appl Environ Microbiol. 2005;71:8228–8235. doi: 10.1128/AEM.71.12.8228-8235.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Jartti T, Hasegawa K, Mansbach JM, Piedra PA, Camargo CA., Jr Rhinovirus-induced bronchiolitis: Lack of association between virus genomic load and short-term outcomes. J Allergy Clin Immunol. 2015;136:509–512. doi: 10.1016/j.jaci.2015.02.021. e511. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Wu GD, Chen J, Hoffmann C, Bittinger K, Chen YY, Keilbaugh SA, Bewtra M, Knights D, Walters WA, Knight R, Sinha R, Gilroy E, Gupta K, Baldassano R, Nessel L, Li H, Bushman FD, Lewis JD. Linking long-term dietary patterns with gut microbial enterotypes. Science. 2011;334:105–108. doi: 10.1126/science.1208344. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Rousseeuw PJ. Silhouettes: A graphical aid to the interpretation and validation of cluster analysis. J Computational Appl Math. 1987;20:53–65. [Google Scholar]
- 32.Faust K, Sathirapongsasuti JF, Izard J, Segata N, Gevers D, Raes J, Huttenhower C. Microbial co-occurrence relationships in the human microbiome. PLoS Computational Biol. 2012;8:e1002606. doi: 10.1371/journal.pcbi.1002606. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Mansbach JM, Piedra PA, Stevenson MD, Sullivan AF, Forgey TF, Clark S, Espinola JA, Camargo CA, Jr, Investigators M. Prospective multicenter study of children with bronchiolitis requiring mechanical ventilation. Pediatrics. 2012;130:e492–500. doi: 10.1542/peds.2012-0444. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Hasegawa K, Mansbach JM, Teach SJ, Fisher ES, Hershey D, Koh JY, Clark S, Piedra PA, Sullivan AF, Camargo CA., Jr ulticenter study of viral etiology and relapse in hospitalized children with bronchiolitis. M Pediatr Infect Dis J. 2014;33:809–813. doi: 10.1097/INF.0000000000000293. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Hasegawa K, Stevenson MD, Mansbach JM, Schroeder AR, Sullivan AF, Espinola JA, Piedra PA, Camargo CA. Association between hyponatremia and higher bronchiolitis severity among children in the icu with bronchiolitis. Hosp Pediatr. 2015;5:385–389. doi: 10.1542/hpeds.2015-0022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.McMurdie PJ, Holmes S. Phyloseq: An r package for reproducible interactive analysis and graphics of microbiome census data. PLoS One. 2013;8:e61217. doi: 10.1371/journal.pone.0061217. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Kloepfer KM, Lee WM, Pappas TE, Kang TJ, Vrtis RF, Evans MD, Gangnon RE, Bochkov YA, Jackson DJ, Lemanske RF, Jr, Gern JE. Detection of pathogenic bacteria during rhinovirus infection is associated with increased respiratory symptoms and asthma exacerbations. J Allergy Clin Immunol. 2014 doi: 10.1016/j.jaci.2014.02.030. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Carlsson CJ, Vissing NH, Sevelsted A, Johnston SL, Bonnelykke K, Bisgaard H. Duration of wheezy episodes in early childhood is independent of the microbial trigger. J Allergy Clin Immunol. 2015;136:1208–1214. doi: 10.1016/j.jaci.2015.05.003. e1205. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Vissing NH, Chawes BL, Bisgaard H. Increased risk of pneumonia and bronchiolitis after bacterial colonization of the airways as neonates. Am J Respir Crit Care Med. 2013;188:1246–1252. doi: 10.1164/rccm.201302-0215OC. [DOI] [PubMed] [Google Scholar]
- 40.von Linstow ML, Schonning K, Hoegh AM, Sevelsted A, Vissing NH, Bisgaard H. Neonatal airway colonization is associated with troublesome lung symptoms in infants. Am J Respir Crit Care Med. 2013;188:1041–1042. doi: 10.1164/rccm.201302-0395LE. [DOI] [PubMed] [Google Scholar]
- 41.Pettigrew MM, Laufer AS, Gent JF, Kong Y, Fennie KP, Metlay JP. Upper respiratory tract microbial communities, acute otitis media pathogens, and antibiotic use in healthy and sick children. Appl Environ Microbiol. 2012;78:6262–6270. doi: 10.1128/AEM.01051-12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Laufer AS, Metlay JP, Gent JF, Fennie KP, Kong Y, Pettigrew MM. Microbial communities of the upper respiratory tract and otitis media in children. mBio. 2011;2:e00245–00210. doi: 10.1128/mBio.00245-10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Gulraiz F, Bellinghausen C, Bruggeman CA, Stassen FR. Haemophilus influenzae increases the susceptibility and inflammatory response of airway epithelial cells to viral infections. FASEB journal : official publication of the Federation of American Societies for Experimental Biology. 2015;29:849–858. doi: 10.1096/fj.14-254359. [DOI] [PubMed] [Google Scholar]
- 44.Sajjan US, Jia Y, Newcomb DC, Bentley JK, Lukacs NW, LiPuma JJ, Hershenson MB. H. Influenzae potentiates airway epithelial cell responses to rhinovirus by increasing icam-1 and tlr3 expression. FASEB J. 2006;20:2121–2123. doi: 10.1096/fj.06-5806fje. [DOI] [PubMed] [Google Scholar]
- 45.McGillivary G, Mason KM, Jurcisek JA, Peeples ME, Bakaletz LO. Respiratory syncytial virus-induced dysregulation of expression of a mucosal beta-defensin augments colonization of the upper airway by non-typeable haemophilus influenzae. Cellular Microbiol. 2009;11:1399–1408. doi: 10.1111/j.1462-5822.2009.01339.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Charlson ES, Bittinger K, Haas AR, Fitzgerald AS, Frank I, Yadav A, Bushman FD, Collman RG. Topographical continuity of bacterial populations in the healthy human respiratory tract. Am J Respir Crit Care Med. 2011;184:957–963. doi: 10.1164/rccm.201104-0655OC. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Mosser AG, Vrtis R, Burchell L, Lee WM, Dick CR, Weisshaar E, Bock D, Swenson CA, Cornwell RD, Meyer KC, Jarjour NN, Busse WW, Gern JE. Quantitative and qualitative analysis of rhinovirus infection in bronchial tissues. Am J Respir Crit Care Med. 2005;171:645–651. doi: 10.1164/rccm.200407-970OC. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.