My initial impression when this article popped up on email was that this is old news. We’ve been using 3% nebulised saline for Bronchiolitis in infants for many years (despite major concerns from paediatricians when we first started). A BET back in 2007 showed that there was a benefit in hospitalised children and it’s been standard practice for us since.
So a new systematic review on the subject seemed a little late in the day!
Maybe not? I vividly remember discussing the scope of the evidence back in 2007. The data at that time demonstrated a benefit for hospitalised kids, but what of admission avoidance? It’s probably safe to assume that hospitalised kids are the sicker end of the spectrum – no argument – the evidence then and now is pretty good for this group. If you are admitting the child – give them Saline nebs – it’s easy, as it often is for any patient at the more severe end of the disease spectrum – guidelines work well in those groups. However, as an emergency physician I often find the patients with moderate severity illness more challenging, and Bronchiolitis is a good example….
- 1. Runny nose, bit breathless, all obs normal, feeding, pee’ing and happy – HOME (easy)
- 2. Sick, dehydrated, not feeding, not peeing, hypoxic – ADMIT (easy)
- 3. Poorly, not drinking as much, fewer wet nappies, a bit sad – Hmmmmm (tricky)
Now, you may work in a system where there are only HOME/ADMIT options and that’s fine, you will have to refer on, but here in Virchester the ED runs a Paeds obs unit where we monitor and review many patients in the ‘moderate’ category. Just a few hours you understand, less than 12 hours usually. So, the poorly, kid with Bronchiolitis ends up there in Virchester for observation, feeding, hydration assessments and…….?Nebulised Saline?? Would you? Is there evidence to support it’s use for admission avoidance on the obs unit and in the ED?
Back in 2007 we didn’t find any, but it’s a good question, and a new review on the topic goes some way to helping me answer that question.
[learn_more caption=”What type of paper is this?”] This is a Cochrane review of the use of saline in Bronchiolitis. The authors have looked at all studies with a helpful differentiation between hospital and out patient (ED) studies.[/learn_more] [learn_more caption=”Is the quality good?”] The methodology seems sound and the authors have done a diligent search and appraisal of the literature. Trial quality is good.
11 trials are included comprising 1090 patients, and in all honestly they are all pretty small. 380 patients were included in trials based in out patients or the ED.[/learn_more] [learn_more caption=”…and the main results are?”] Saline reduces in hospital stay… we kind of knew that! The average reduction in length of stay is 1.15 days which is great, and this is reflected in a reduction in clinical severity scores at day 1 and 2.
What about the ED issues though?
There are 4 trials that address admission avoidance and these include 380 patients. For placebo (0.9% saline 25 per 189 were admitted, for 3% saline 16 per 191 were admitted. That’s a fairly low rate of admission which suggests that if this is a mild/mod group then it’s mostly mild! Having said that there is a reduction, though this does not achieve statistical significance. The relative risk of admission with 3% is0.63 (with wide confidence intervals – 0.37 to 1.07)
[/learn_more][learn_more caption=”So is my question answered?”] Yes and no. It’s great to see a trend towards admission avoidance but the numbers are tiny and the results are not statistically significant.[/learn_more] [learn_more caption=”So, a change in practice then?”] Well, I’ll admit that I’ve used saline nebs as an admission avoidance treatment for some bronchiolitics for many years. A saline neb, followed by observation on the paeds short stay ward often ‘seems’ to work, but without robust evidence I’m not sure anymore.
There is nothing here that will make me change that practice, but I am really surprised that the management of this incredibly common disease and intervention is based on a small number of patients in rather disparate trials. It would be fantastic if we could get a decent powered RCT to answer this question and to clarify a treatment strategy for moderate patients.