Bronchiolitis in children: diagnosis and management

Why the committee made the recommendations

The committee did not consider any of the evidence on oxygen saturation thresholds that they reviewed to be directly relevant to referral, because the studies focused on admission, management, and discharge. However, they needed to review these recommendations to ensure that they were in line with the changes to the recommendations on admission, management and discharge, and to avoid any contradictions.

For most babies and children, the oxygen saturation threshold has been lowered to 90% for admission, management and discharge, but the committee felt that the original threshold of 92% remained appropriate for referral. They agreed with the rationale in the 2015 guideline that this threshold of 92% enabled a safety margin for babies and children whose condition may be deteriorating, because their oxygen saturation level may drop further.

Although an oxygen saturation level of less than 92% indicates that a baby or child needs further clinical assessment, the committee did not feel that this criterion alone was enough to justify immediate transfer to hospital (as specified in the 2015 guideline). Oxygen saturation measurement may be less accurate in primary care than in hospital (for example, because paediatric oximeters are often not available in primary care). As a result, the committee removed the oxygen saturation threshold as a criterion that requires immediate referral to hospital and instead included it as a criterion that requires consideration of referral to hospital. They did not think that this change is likely to mean significant changes in practice, because:

Because the oxygen saturation thresholds are now lower for discharge from hospital than they are for referral to hospital, the committee discussed whether this could lead to babies and children being inappropriately referred straight back to hospital after discharge. The committee agreed that the changes to the referral recommendations would avoid this, because the recommendations now put less emphasis on using oxygen saturation level alone as the reason for referral. They also noted that many babies and children would have a post-discharge management plan, which could give guidance on when re-referral would and would not be needed.

How the recommendations might affect practice

The recommendations are unlikely to result in a significant change to the number of referrals to hospital. Oxygen saturation level alone can still be used as a reason for referral, but it is no longer a reason for immediate referral (for example by ambulance).

There may be a small increase in the number of babies and children being immediately referred to hospital if they meet one of the other criteria in recommendation 1.2.1. This is because, in the committee's experience, some babies and children who met the other criteria in recommendation 1.2.1 were not referred if they did not also have an oxygen saturation below 92%.

Other factors the committee took into account

See the rationale and impact section on when to discharge, for the committee's discussion of the variations in pulse oximeter accuracy based on skin tone.

Return to recommendations

Why the committee made the recommendation

The committee reviewed an observational study and a randomised controlled trial. They did not feel that the evidence from the observational study could be used to change the 2015 recommendations. It was assessed as being very low-quality evidence, based on the study design and a very serious risk of bias. However, the committee believed that the randomised controlled trial could be used to change the 2015 recommendations, because it involved babies and children who were assessed in an emergency department and so was directly relevant to this part of the guideline.

The 2015 recommendation specifies 'persistent' oxygen saturation levels, and the committee felt that this means babies and children would be assessed and monitored for some time in an emergency healthcare setting. This extended assessment and monitoring would allow healthcare professionals to understand the baby or child's overall health and whether it was worsening. In this context, there is very little risk of harm from reducing the oxygen saturation threshold level needed for admission from 92% to 90%. However, the committee acknowledged that some babies and children are at higher risk of severe bronchiolitis (babies under 6 weeks, and children with underlying health conditions). The randomised controlled trial did not include this group, so there was no evidence available for them. Because of this, the committee agreed it would be safer to retain the threshold of 92% for babies and children at higher risk.

Parents and carers of babies and children who are not admitted should receive key safety advice (see recommendation 1.6.1).

Reducing the oxygen saturation threshold for admission also makes these recommendations consistent with the updated management and discharge recommendations.

How the recommendation might affect practice

Babies and children assessed for bronchiolitis in hospital often meet multiple admission criteria. For these babies and children, there will be no impact from the updated recommendation, because they will still be admitted based on other criteria. There may be fewer admissions for babies and children whose only indication is persistent decreased oxygen saturation level.

There would be a reduction in NHS costs if the change in threshold led to a substantial reduction in the number of babies and children admitted to hospital, but it is unclear how large the change would be in practice.

Other factors the committee took into account

See the rationale and impact section on when to discharge, for the committee's discussion of the variations in pulse oximeter accuracy based on skin tone.

Return to recommendation

Why the committee made the recommendation

A randomised controlled trial found that using an oxygen saturation threshold of 90% (compared with a threshold of 94%) for deciding whether to provide supplementary oxygen and discharge from hospital significantly reduced the need for supplemental oxygen and the time to discharge. The trial also showed that readmission rates were not higher with a 90% threshold, compared with a 94% threshold.

The intervention, population and setting of the trial were directly relevant to management and discharge from hospital, so the committee believed it was appropriate to change the oxygen saturation thresholds used in the 2015 guideline. In addition, they noted that the 2015 thresholds were based on the 2015 committee's experience alone, because no direct evidence was available at the time. The trial included babies and children with fewer comorbidities who had less risk of poor outcomes compared with babies and children with bronchiolitis typically seen in hospital, who may have a more varied risk profile. Because there was no evidence for babies and children at higher risk (babies under 6 weeks and children of any age with underlying health conditions), the committee agreed it would be safer to retain the threshold of 92% for this group.

How the recommendation might affect practice

Reducing the oxygen saturation threshold will mean that fewer babies and children are given supplementary oxygen. All other aspects of bronchiolitis management remain unchanged and healthcare professionals should stay aware of the full clinical picture.

There is likely to be a reduction in NHS costs as a result of this recommendation, due to both a reduced duration of oxygen therapy in hospital, and a reduced length of hospital stay.

Other factors the committee took into account

See the rationale and impact section on when to discharge, for the committee's discussion of the variations in pulse oximeter accuracy based on skin tone.

Return to recommendation

Why the committee made the recommendation

A randomised controlled trial found that using an oxygen saturation threshold of 90% (compared with a threshold of 94%) for deciding whether to provide supplementary oxygen and discharge from hospital significantly reduced the need for supplemental oxygen and the time to discharge. The trial also showed that readmission rates were not higher with a 90% threshold, compared with a 94% threshold.

The intervention, population and setting of the trial were directly relevant to management and discharge from hospital, so the committee believed it was appropriate to change the oxygen saturation thresholds used in the 2015 guideline. In addition, they noted that the 2015 thresholds were based on the 2015 committee's experience alone, because no direct evidence was available at the time. The trial included babies and children with fewer comorbidities who had less risk of poor outcomes compared with babies and children with bronchiolitis typically seen in hospital, who may have a more varied risk profile. Because there was no evidence for children at higher risk (babies under 6 weeks and children of any age with underlying health conditions), the committee agreed it would be safer to retain the threshold of 92% for this group.

The committee noted that the decision to discharge should not just be based on oxygen saturation. The guideline includes other recommendations on criteria for discharge, and a recommendation on key safety criteria for parents and carers looking after the baby or child at home (recommendation 1.6.1), so the committee did not believe there was a significant risk to reducing the oxygen saturation threshold criteria.

How the recommendation might affect practice

The reduction in oxygen saturation threshold will allow more babies and children to be discharged sooner, without increasing the risk of worse outcomes. Other criteria for discharge have not changed, so the decision to discharge will not be based on oxygen saturation alone.

Other factors the committee took into account

There are emerging reports in other areas of clinical care that there may be variation in the accuracy of pulse oximetry depending on a person's skin tone. The 2021 update of the guideline did not look at the evidence in this area, so the committee did not make a recommendation to address the issue or a recommendation for further research.

The NHS Race and Health Observatory published a rapid review of the evidence in this area in March 2021. NICE will monitor for formal guidance from NHS England and NHS Improvement in this area, and update this guideline further as needed.

Return to recommendation