Aldridge, P. Wheeze. It’s all in the timing…, Don't Forget the Bubbles, 2019. Available at:
You’ve just treated a 5 year old with their second episode of wheeze this year. They’ve had burst therapy, and are now one hourr post treatment with no work of breathing, scattered wheeze on auscultation and oxygen saturations of 95%. You contemplate giving them steroids and decide against it.
Authors warning – this is not a full critique of the papers referenced below. I hope to nudge readers to start questioning a standard practice in paediatrics…
You think that you could probably send them home in another hour or so but your hospital guideline stretches these children to needing inhalers every four hours prior to discharge. You think that you might have to admit them as your Emergency Department mandated attention span of four hours will be breached. The paediatric team agree they need admission ‘to stretch’ but will actually send them home at three hours. The paediatric consultant suddenly comes along and discharges them (one hour post burst therapy) quoting a recent paper. You are very confused…
What is current guidance?
This is a scenario which likely plays out to varying degrees every day in the paediatric healthcare world. The recent BTS/SIGN ‘British guideline on the management of asthma’ (July 2019) (page 111) states that ‘…children can be discharged home at 3-4 hourly inhalers’. The Australian Asthma Handbook (AAH) states ‘…observe the patient for at least four hours’. This is (anecdotally) a mainstay of treatment practice in Australia, USA, UK and probably many other countries. BTS/SIGN provide reference to a paper for their recommendation whereas the AAH states its ‘…based on clinical experience and expert opinion’.
So, what titanic study is discharge at 4 hours post inhaler use based upon? Well, BTS/SIGN quote this paper. This was a randomised control study (so far so good…) from 1999 and used on 63 patients with asthma >18 months age.
They concluded that children could be safely discharged home on three hourly nebulisers based on a parental phone questionnaire two weeks post discharge. I do not mean to belittle this study as it was likely to have been ground breaking at the time and showed a novel approach. However, it was 20 years ago and is the only* study in the literature up to 2018. The study patients were discharged on nebulisers and the length of stay had 95% confidence intervals between 42 and 67 hours. As any modern 21st century paediatrician will tell you (and will be noted later) the average length of stay is rarely around these dizzy heights anymore.
But what does 3-4 hourly actually mean? Is it once a patient reaches three hours or four hours from their last inhaler for the first time? Is it reaching three hours or four hours twice between inhalers? Well, that’s where confusion reigns supreme.
At a recent UK conference #PIER19 I conducted a survey which showed the following.
Most said they send a child home on reaching four hourly inhalers for the first time, for which there is no evidence in the literature, but some said they wait twice for three or four hours (I personally call these the six or eight hour stretches…).
So why are some hospitals doing three hourly inhalers twice before discharge? Well, the most recent study is from 2018 where the Texas Children’s Hospital did a large quality improvement project, starting in 2013, including discharging children on three hourly inhalers. Multiple interventions were used including a streamlined care pathway, education, dexamethasone instead of prednisolone, a QI multi disciplinary team approach and a detailed analysis of patient care. I will not critique the study in its entirety – suffice to say they showed no adverse outcomes and a shorter average length of stay (30 hours post intervention vs 36 hours pre). This was also coupled with many other interventions over the study period but which were felt to not have had as large an impact as three hourly inhalers. How applicable all of this is to the ’real world’ is debatable, but the study is highly admirable as it’s only* the second prospective study in the literature for 20 years.
*There is a third study from 2003 but it’s retrospective and largely theoretical, so I have not dwelt upon it here.
Is one hour enough?
So, what about sending them home one hour post therapy? This comes from a very recent study which was a retrospective analysis of patients presenting to a major specialist children hospital in Australia over a two-week period in the winter of 2014 compared with the winter of 2015. In between these winters, the hospital admission criteria was changed to an admission decision on medical review one hour post initial therapy. This is based on studies (here and here) suggesting that the patient’s condition at one hour post therapy is highly predictive of the need for admission/further therapy.
They looked at a total of 105 patients (2014) vs 92 patients (2015). They showed that children who were ‘clinically well’ at one hour were more likely to not be admitted in 2015 vs 2014 (10% vs 40% p=0.001) and went home from ED faster in 2015 vs 2014. They also showed that any child with moderate symptoms (as per new guidelines) at one hour post therapy was admitted in each year. This is a small study over a two week period only and was based at a specialist children’s hospital with 75,000 ED attendances a year. How applicable this practice would be at a smaller hospital with just a few (or no) paediatricians is impossible to say at this point.
The most recent published study with discharge at one hour post initial therapy supports what I, and others, probably already do (or we wish we had the courage to) when a child is so well after initial treatment – they are sent home. Is this confirming clinical gestalt?
As for discharging a patient home at 2, 3, 4, 6 or 8 hour inhalers (and whether that’s the first or second time round) – well I will leave that to you.
This author has started a local quality improvement project around three hours (first time) and discharge home. I just won’t be able to critique it here if it ever gets published…