Aims The utilisation of blood gas analysis (BGA) in acute bronchiolitis is common with wide variation between hospitals. Guidelines recommend its use only in those with severe respiratory distress and who are tiring but evidence for such practice is sparse. This study investigated clinical indicators that demarcate clinically important rise in carbon dioxide (CO2).

Methods We undertook a prospective observational study of children admitted from the emergency department (ED) to a tertiary care university hospital with a diagnosis of bronchiolitis (October 2014–January 2015). Data was obtained from hospital charts and electronic patient information and analysed using STATA/IC 12.1 for Mac. Univariate analysis examined the correlation between blood gas CO2 and oxygen requirement, SpO2 in air, vital signs, feeding support requirement, gender, age, co-morbidities and history of prematurity. Statistically significant variables were entered into a logistic regression model.

Results 220 patients with bronchiolitis were admitted (mean age of 0.57 years (95% CI:0.05,0.64)). 113 (51%) had at least one BGA done. Those with more than three BGAs (32/113 (28%)) were significantly younger and more likely to be premature. 14% (30/220) were admitted to intensive/high dependency care (ITU/HDU). In ED a CO2 >7kPA was associated with an admission to HDU/ITU (OR 3.75(95% CI:1.13,12.47), p = 0.031); prematurity and young age also independently predicted ITU/HDU (IRR1.53(95% CI:1.21,1.93), p < 0.0001 and IRR0.68(95% CI:0.53,0.86), p = 0.001 respectively). All but one patient with CO2 >7kPa in ED were < 3m and/or premature. Length of stay (LOS) was also significantly longer in this patient group. For BGA taken during the admission, only oxygen requirement and age (particularly <3months (m)) were significantly associated with CO2 >7kPa in the regression model (OR 2.46(95% CI:1.42,4.26), p = 0.001 and OR 0.08(95% CI:0.03,0.17), p < 0.001; respectively). There was no association between amount of oxygen supplied and level of CO2 measured. Overall, LOS was significantly higher in those who had BGA during admission (4.0 days (d) (95% CI 3.5,4.5) versus 2.3d (95% CI:2.0,2.6)).

Conclusion Age under 3 months, history of prematurity and CO2 >7kPa done in ED identify those with prolonged LOS and/or HDU/ITU admission. Significantly raised CO2 is not seen in bronchiolitis without oxygen requirement. Further work will look to elucidate when BGA may be helpful in the management of bronchiolitis.