Monthly Archives: October 2014

Optimal oxygen saturation level uncertain for guiding hospitalization in children with bronchiolitis

Bottom Line:

Artificially inflating pulse oximetry measurements by 3% resulted in a significant decrease in hospitalization rates in infants presenting to the emergency department (ED) with acute bronchiolitis. The authors suggest that oxygen saturation rates should not be the only factor in the decision to admit or discharge infants from the ED. (LOE = 1b)


Schuh S, Freedman S, Coates A, et al. Effect of oximetry on hospitalization in bronchiolitis. A randomized clinical trial. JAMA 2014;312(7):712-18.

Study Design: Randomized controlled trial (double-blinded)

Setting: Emergency department


The optimal use of pulse oximetry in the management of acute bronchiolitis in infants is uncertain. These investigators identified previously healthy infants, aged 4 weeks to 12 months, who were diagnosed with acute bronchiolitis using standard clinical criteria. Exclusion criteria included a true oxygenation saturation below 88% (as measured by the triage nurse) and those meeting clinical parameters for severe respiratory distress based on a prevalidated scoring tool. Eligible infants (N = 213) randomly received assignment (concealed allocation) to either a true saturation group (ie, true oximetry saturation displayed throughout the ED visit) or the altered saturation group (ie, the saturation measurements displayed were 3 percentage points higher, to a maximum of 100%). Half the study oximeters used in the ED were altered by the manufacturers so that the saturation display was increased by 3 percentage points. All individuals involved in the care of the infants, including clinical personnel and the infants’ parents, remained masked to treatment group assignment. Outcomes assessed included hospitalization for bronchiolitis within 72 hours of enrollment, active hospital care for more than 6 hours in the ED, or unscheduled visits for recurrent bronchiolitis. Complete follow-up occurred for all patients at 72 hours. The study protocol did not specify an oxygen saturation cutoff dictating a need for hospitalization, but instead left that decision to the clinical judgment of the attending ED physicians. Only 28 of the 213 infants at triage had saturation levels below 94%. Using intention-to-treat analysis, significantly more infants in the true oximetry group were hospitalized within 72 hours compared with those in the altered oximetry group (41% vs 25%, respectively). No significant group differences occurred in the length of ED stay, use of supplemental oxygen, or rates of unscheduled return visits. No participants were admitted to the intensive care unit. These results suggest that medicolegal issues should not be specifically decided by measurements such as oxygen saturation rates, but instead by sound clinical judgment and the understanding that sometimes adverse outcomes are not always predictable.

David Slawson, MD
Director of Information Sciences
University of Virginia Health System
Charlottesville, VA