Background
Bronchiolitis is the most common lower respiratory tract infection and the most common cause of admission in infants. Approximately 10% will require some airway support. The ability to identify those at risk for escalation of care would allow for appropriate disposition decisions.
Clinical Question
In infants without comorbid conditions with a first episode of clinical bronchiolitis, can demographic, history, physical examination findings and room air oxygen saturation at presentation predict the need for an escalation of care (primarily airway support) in the ED or inpatient setting?
Design
Observational: Retrospective cohort study
Primary Results
- N = 2,772
- N = 261/2,772 (9.6%) with care escalation
- Most common airway interventions: HFNC (63%), CPAP or BPAP (18%)
- 7 Independent predictors of care escalation were identified (see Table 1 below)
- Score (range 0-14 points) based on adjusted odds ratios
- 88.4% of patient with a score of 5 or less (see Table 2 below)
| TABLE 1: REGRESSION ANALYSIS | ||
| PREDICTOR | ADJUSTED OR (95% CI) | POINT SCORE |
| Age ( 2years/> 2 years) | 2.10 (1.49, 2.97) | 1 |
| Poor feeding (Y/N) | 1.85 (1.27, 2.71) | 1 |
| Oxygen saturation (<90%/90%) | 8.92 (5.08, 15.66) | 5 |
| Apnea (Y/N) | 3.01 (1.89, 4.78) | 2 |
| Nasal flaring or Grunting (Y/N) | 3.76 (2.64, 5.35) | 2 |
| Dehydration (Y/N) | 2.13 (1.37, 3.30) | 1 |
| Retractions (Y/N) | 3.02 (1.59, 5.73) | 2 |
| TABLE 2: PROPORTION WITH EACH SCORE REQUIRING ESCALATION | |||||
| Score | % Escalation | % with Score | Score | % Escalation | % with Score |
| 0 | 0.5% (1/217) | 8.0% | 8 | 40.9% (18/44) | 1.6% |
| 1 | 0.5% (1/199) | 7.3% | 9 | 69.7% (23/33) | 1.2% |
| 2 | 2.0% (11/563) | 20.7% | 10 | 52.9% (9/17) | 0.6% |
| 3 | 4.5% (33/740) | 27.2% | 11 | 81.0% (17/21) | 0.8% |
| 4 | 8.0% (34/423) | 15.5% | 12 | 58.3% (7/12) | 0.4% |
| 5 | 16.6% (44/265) | 9.7% | 13 | 100% (5/5) | 0.2% |
| 6 | 25.6% (31/121) | 4.4% | 14 | 100% (11/11) | 0.4% |
| 7 | 31.4% (16/51) | 1.9% | |||
| AUC (Area under the receiver operation curve): 84.7%, 95% CI (81.7%, 86.8%) | |||||
Strengths
- Multinational consortium of National/Regional Pediatric Emergency Medicine Research Groups
- Large sample size, 261 patients requiring care escalation
- High AUC indicating a high predictive value of the rule as a whole
Limitations
- Composite outcome. Not all interventions of equal import
- Criteria of care escalation not standardized
- Retrospective: No assessment of interrater reliability for rule predictors
- From an ED standpoint, a subgroup analysis of patients having an escalation of care in the inpatient setting would allow for appropriate distribution to inpatient ward or ICU
- Generalizability of the study’s results to non-children’s hospital settings is unclear
- Stage IV clinical decision rule (derivation with internal statistical validation only): Requires further validation before clinical use
Author's Conclusions
“We identified variables measured in the ED predictive of receipt of escalated care for bronchiolitis and derived a clinical risk score with high discriminatory ability and excellent model stability to stratify risk of this outcome during hospital stay. Prospective validation and determination of clinical use are now needed.”
Our Conclusions
Infants older than 2 months of age with an oxygen saturation greater than 90%, without hydration issues and without retraction, nasal flaring or grunting appear to be at low risk of requiring a subsequent airway intervention.
Potential Impact To Current Practice
This a stage IV clinical decision rule that requires further validation before it can be used clinically. If validated, this study has the potential to inform disposition decision making in the infant with bronchiolitis.
Read More
Core EM: Episode 46.0 – Grand Rounds (Ilene Claudius) – Pediatric SOB
