Aims Assessing the potential effect of three discharge oxygen saturation (SpO2) targets ( 90%, 94% and 94%) on length of hospital admission in children with wheeze.

Methods and patients: Children (1 16 years (y)) admitted with wheeze and oxygen requirement for SpO2 (£92%), had SpO2 in air assessed 4 hourly. Time from admission for SpO2 to become stable for at least 4 hours (h) at 90%, 92% and 94% was recorded.

Results 243 children were admitted with wheeze (n=14 excluded due to inadequate demographic data or observation). Fourty eight children did not require oxygen and 39 children reached discharge criteria prior to the specified time ( 4 hour) at SpO2 targets. The median time to discharge was 23 hour (IQR 17 39) for all children admitted with wheeze (n=229), 27 hour (17 40) for children requiring supplemental oxygen (n=181) and 30 hour (IQR 17 40) for children requiring oxygen supplementation and a full data set (n=142). Data on the latter patients (median age 2.8y) were further assessed. Five children were admitted to high dependency. Seventy percent (99/142) had a viral or bacterial agent identified. Time from admission to a stable SpO2 for 4 hour was 8 hour (IQR 5 20) for SpO2 90% in air, 17 hour (IQR 7 30) for SpO2 target of 92%, and 21 hour (IQR 21 30) for SpO2 target of 94%. A time lag was observed in 61 (43%) children between SpO2 targets of 90% and 92% and in 75 (53%) children between targets of 90% and 94%. Presence of any infective agent was not associated with time to stable oxygen at any threshold, except for RSV. RSV was statistically significantly (p<0.001) associated with a more prolonged time to stable oxygen saturation at all thresholds independent of age.

Conclusions Reducing SpO2 target levels to 90% in children with acute wheeze could potentially reduce length of stay. This target level has been shown to be safe in infants with bronchiolitis, while it can be postulated that the same would be true for older children with wheeze, a randomised control trial to assess the safety and effect on clinical status is warranted.