Clinical question: Does artificial elevation of pulse oximetry measurement in bronchiolitis patients during ED evaluations affect hospitalization rates?
Background: Bronchiolitis is the leading cause of hospitalization for infants younger than one year, leading to direct medical costs in the U.S. of $543 million in 2002. Compromised oxyhemoglobin saturation in bronchiolitis often leads to hospitalization and is assessed by pulse oximetry (SpO2) more commonly than by arterial blood gas (SaO2) due to ease, cost, and comfort considerations.
SpO2 can vary due to fever, acidosis, hemoglobinopathies, underperfusion, and poor probe placement. An American Academy of Pediatrics clinical practice guideline published in 2006 recommended supplemental oxygen if SpO2 drops below 90% in previously healthy infants, but the data supporting this cutoff are sparse.
Recommendations for supplemental oxygen, and thus hospitalization, are variable, with recommended minimum SpO2 ranging from 90 to 95%.
Study design: Single-center randomized, double-blind, parallel-group trial.
Setting: ED at 370-bed, urban, tertiary care children’s hospital.
Synopsis: Over a 50-month period, previously healthy infants aged four weeks to 12 months who were diagnosed with bronchiolitis in the ED and had initial triage SpO2 above 88% were enrolled by investigators. Subjects were excluded if severe respiratory distress, as measured by the Respiratory Distress Assessment Instrument (RDAI), or impending respiratory failure was present. Subjects were randomized to two groups: The control group had their true SpO2 displayed during their ED stay, and the intervention group had an SpO2 that was three points higher displayed. ED physicians were not aware of the primary hypothesis being tested. All subjects underwent concealed continuous oximetry for safety reasons, with monitors alarming if SpO2 dropped below 92%. Study nurses telephoned participants discharged home 72 hours after enrollment.
The primary outcome of hospitalization was defined as admission to an inpatient ward, hospital care provided for more than six hours in the ED if no hospital beds were available, or hospitalization after discharge if within 72 hours of enrollment. Secondary outcomes included supplemental oxygen administration, length of stay in the ED, and unscheduled return visits for bronchiolitis within 72 hours. Exploratory outcomes included delayed hospitalizations within 72 hours, active hospital treatment for more than six hours (with inhaled bronchodilators, oxygen, or intravenous fluids), and hospitalization at the index visit.
Of 1,812 patients assessed, 213 were randomized after exclusion criteria and consent. The “true” group and the “altered” group were similar in initial RDAI (8.0 vs. 8.3 respectively); 41% of children in the “true” oximetry group were hospitalized within 72 hours, compared with 25% in the “high” oximetry group (P 0.005). There were no significant differences in the secondary outcomes. The only exploratory outcome to show a significant difference was treatment for longer than six hours, with 37% of the “true group” receiving treatment for longer than six hours, compared to 20% of the “altered” group (P 0.01).
Bottom line: Perception of improved oxygenation based on falsely elevated SpO2 alone can reduce the inclination of a clinician to admit children with bronchiolitis. This brings into question the use and interpretation of SpO2 in treating children with bronchiolitis.
Citation: Schuh S, Freedman S, Coates A, et al. Effect of oximetry on hospitalization in bronchiolitis: a randomized clinical trial. JAMA. 2014;312(7):712-728.