Background

Bronchiolitis is the leading cause of hospitalizations among infants in the United States. Hospitalizations for this disease have increased and it has been hypothesized that reliance on oximetry values may be at least partly responsible for the increase. There is no absolute consensus on whether supplemental oxygen is required in patients with saturations between 90-95%. Additionally, there is little evidence examining the effect of presenting oximetry readings on Emergency Physician’s decision to admit or discharge.

These authors sought to help clarify the influence of oximetry on admission decisions in a novel way. They created a randomized study in which half of the patients had “true” oximetry readings and half of the patients had “false” readings (the actual reading was falsely increased by 3% to the eyes of the clinicians).

Clinical Question

What is the effect of artificially increasing oximetry readings on Emergency Physician decision to admit or discharge infants presenting with bronchiolitis?

Population

Previously healthy infants aged 4 weeks to 12 months who presented to the ED with a diagnosis of bronchiolitis.

Intervention

Pulse oximetry displaying altered saturation values of 3% higher than that read.

Control

True saturation pulse oximetry display.

Outcomes

Primary Outcome: Hospitalization within 72 hours (defined as hospitalization a at the index visit, hospitalization within 72 hours after discharge or active hospital care for greater than 6 hours)
Secondary Outcome: Supplemental oxygen use, ED LOS, unscheduled bronchiolitis visits within 72 hours.

Design

Randomized, double-blind, parallel group trial conducted in a tertiary-care ED in Toronto.

Excluded

Children with significant prior medical history (cardiopulmonary, neuromuscular, hematologic or congenital airway anomalies), triage saturation below 88%, patients transferred from other institutions and the presence of severe respiratory distress.

Primary Results

213 patients randomized (1812 patients assessed for eligibility), 33% admitted
Primary Outcome (hospitalization within 72 hours)

  • 41% (44/108) of patients in the true oximetry group
  • 25% (26/105) of patients in the altered oximetry group

Difference: 16% (95% CI 3.6 – 28.4%) p = 0.005 (NNT = 6.25)

Adjusted Odds Ratio: 4.0

In a subgroup analysis only the active hospital care for greater than 6 hours demonstrated a statistically significant difference

Secondary Outcome (unscheduled medical visits for bronchiolitis)

  • 21.3% (23/108) of patients in true oximetry group
  • 14.3% (15/105) of patients in the altered oximetry group

This difference was not statistically different.
None of the other secondary endpoints had statistically significant differences.

Strengths

  • Successful randomization – baseline characteristics very similar
  • Primary outcome is important and patient oriented
  • Randomization well done
  • Authors took unique approach to address a significant clinical question

Limitations

  • Single center
  • High number of enrollment refusals from parents introduces selection bias.
  • Most infants had near-normal oxygen saturation or mild hypoxia. Only 16% of the altered group and 10% of the true group had oxygen saturation < 94% potentially limiting the impact of the intervention
  • Although physicians were blinded to allocation, they were aware the study was being performed. This may have altered behavior.

Author's Conclusions

“Among infants presenting to an emergency department with mild to moderate bronchiolitis, those with an artificially elevated pulse oximetry reading were less likely to be hospitalized within 72 hours or to receive active hospital care for more than 6 hours than those with unaltered oximetry readings. This suggests that oxygen saturation should not be the only factor in the decision to admit, and its use may need to be reevaluated.”

Our Conclusions

Mild hypoxia appears to disproportionately influence Emergency Physician disposition in infants with bronchiolitis. This likely results from poor understanding of the oxygen dissociation curve. This curve is relatively flat between 90-95% saturation. In fact, 3% oxygen saturation differences in this range correspond with an arterial pressure difference of only 9 mm Hg. Instead of concentrating on small differences in oxygen saturation, we should focus our attention on the patient and how they look clinically.

Potential Impact To Current Practice

Mild hypoxia should not be a major determinant in disposition decision making for infants with bronchiolitis.

Bottom Line

Emergency Physicians rely too heavily on oximetry readings in deciding whether to admit or discharge patients with bronchiolitis. Over-reliance on this value leads to excess, possibly unnecessary, hospitalizations.