Paediatric Emergencies Update – Noosa Day 1

Estimated reading time: 14 minutes

You might remember I wrote up my learning and reflections from the PEM virtual colloquium back in 2020 – well, this year I was there in person!

The conference is a round-up of paediatric emergency medicine topics, often delivered by related specialists rather than by emergency medicine clinicians, but provides a valuable update on a number of PEM topics and is often very practical. Although day 1 speakers were all quite aesthetically similar, there was more diversity present among faculty on days 2 & 3!

Paediatric Stroke – Ian Andrews

Stroke is rare; mimics are common; MRI & MRA are the best modality but complicated. Considering stroke as a diagnosis in children is often the biggest hurdle, cognitively and time-wise for meaningful management.

Reduced tissue blood supply initially produces reversible brain functional failure – in the ischaemic penumbra, the subacute phase (4-6 hours) leads to the accumulation of toxic metabolites and this area we are trying to preserve. Children’s brains usually have healthy blood vessels, unlike adult brains. For adults, with acute ischaemic stroke, perfusion imaging helps us to understand regions of hypoperfusion around the core of the infarct and we can measure the area with reduced flow. This has informed earlier treatment decisions around reperfusion therapies.

Australia has published guidance on diagnosis and management of children’s stroke, which you can find here^​1​.

In 2010-2019, NSW saw 16.5 strokes per year in children aged 1m-18yrs. Neonates have a much higher incidence. Stroke is generally rare and mimics are common in children with acute focal neurological dysfunction; data from Boston^​2​ (where MRI is available 24/7, can you imagine?!) found only 9% of those having urgent MRI for ?stroke were having an ischaemic stroke. The most common diagnosis/mimic is hemiplegic migraine and often the first hemiplegic migraine (though, of course, migraine is not usually diagnosed after a single episode).

There are risk factors, commonly cardiac disorders – but most children have no risk factors before presenting with ischaemic stroke.

Headache occurs in about half of paediatric strokes and can be localising; 25% have symptoms on waking but onset is generally sudden.

Imaging is ideally MRI and MRA head and neck vessels. It is possible to do a limited MR in 15 minutes with enough detail to diagnose and inform treatment decisions. CTA is the next best choice – non-contrast CTB plus CT angiogram of the head and neck is what we are after. Non-contrast CT can sometimes show an acute stroke; it can identify or exclude haemorrhage and may show clot in large vessels but is only minimally helpful in stroke mimics. In contrast to MR, CT is usually readily and rapidly available, does not require GA, and is better than an MRA at showing blood vessels, so it may provide the courage needed to trigger protocols for thrombectomy and/or TPA. The radiation dose of CT perfusion is considerable and the algorithm for imaging is based on adult cerebral blood flow; there is limited understanding of how this applies, for example, for a 4 year old.

So, what should we do when they walk through the door?

Good quality critical care is the answer: normoxia, treat fever (it increases metabolic rate and we want to reduce the metabolic demand of the penumbra), maintain circulation, aim for normal glucose, identify and treat seizures. If there is a confirmed ischaemic stroke: mechanical thrombectomy is available for virtually any large vessel occlusion, likely applicable to M2 branches, it requires a CT and angio and we should aim for <6hrs though this is frequently longer in children, up to 24h (and sometimes even longer). Heparin is used for embolic stroke, and extracranial or intracranial dissection, with aspirin for other diagnoses (although this is an evidence-light area.)

In NSW data, about 26% had a large vessel occlusion – and all of these patients are potentially eligible for mechanical thrombectomy if we can consider and make the diagnosis in a timely manner. Data tells us that thrombectomy provides better functional outcome in paediatric stroke patients^​3​ Basilar artery stroke is less common, but the data may support thrombectomy for these too^​4​.

tPA is tricky, with a tight set of inclusions and many exclusions, including safety-related, stroke-related, imaging-related and drug-related exclusions. tPA is relatively safe in children but the data is limited. In the NSW data, only around 1 in 5 children was eligible for tPA, so it’s understandable that it’s difficult to research it.

We are particularly bad at suspecting stroke in those presenting with seizure, those with posterior fossa symptoms, those with pain (headache) or injury (for example, if they aren’t using a limb or have fallen), and those with waxing and waning symptoms.

Malignant middle cerebral artery infarction is a syndrome worth knowing about, accounting for 1.3-9% of paediatric strokes. Patients present with prolonged seizures (>5 minutess), a big stroke clinically and radiologically, and often hyperglycaemia at admission in those over 2 years.

In summary – this is a rare diagnosis but one we should probably think about more because we are missing opportunities – and, at a system level, we should be working on our diagnostic pathways for these rare times we might want to use them.

Bronchiolitis – Andrew Numa

Respiratory Syncitial Virus (RSV) was first identified in 1957 and remains highly transmissible (with an R0 similar to COVID19) and therefore sequelae make this a common PED presentation. RSV causes 70% of bronchiolitis – 98% of children get it in their first two years of life and there are two main subtypes, A&B, with A more virulent. There is alternating annual dominance, which explains why some bronchiolitis seasons seem worse than others. Worldwide, there are 30 million cases, 3.2 million hospitalisations and 200,000 deaths each year with deaths due to RSV usually happening in low and middle-income countries.

RSV accounts for 15% of all paediatric ED visits, 15% of PICU admissions in Australasia (the commonest non-elective PICU admission diagnosis) and the commonest cause of admission to hospital in the first year of life.

Andrew notes that in the PREDICT guideline, most recommendations are around NOT doing things – and there are only three actual recommendations (they need oxygen if indicated, hydration if indicated, and mechanical support if indicated – and time!). Notably, despite the lack of evidence and recommendations against giving medications, many patients do receive medications (most commonly, salbutamol). The adage that there are no beta receptors in small kids is nonsense; the reason that bronchodilators don’t tend to work in small, wheezy kids, is that the pathophysiology is not bronchoconstriction but mechanical obstruction from epithelial sloughing and oedema. Remember, salbutamol is not a harmless drug and can cause VQ mismatching including increasing oxygen consumption and requirement for a few hours after administration (we recorded a Resuscitology podcast discussing this here).

Interestingly, patients who are admitted to PICU with RSV bronchiolitis are more likely to receive medications (including adrenaline^​5​, steroids and adrenaline^​6​, dexamethasone with beta agonist in the 1-2 age group^​7​, dexamethasone^​8​, adrenaline with dexamethasone in a four-way trial^​9​, with some evidence suggesting benefits from the combination of steroid and adrenaline in returning to normal feeding and quiet breathing, though this has not been statistically significantly demonstrated.

This study, almost 30 years old (!)^​10​, selecting children with purely obstructive symptoms in bronchiolitis showed around half of the patients demonstrated beta responsiveness at 4m of age – so if anyone tells you children don’t have the receptors under the age of 2yrs you can wave this paper at them. That said, in bronchiolitis (as mentioned above) we wouldn’t expect a response to bronchodilator therapy as the medications don’t address the underlying pathophysiology. Other studies of salbutamol have shown no difference in oxygen saturations, clinical score, admission rates or length of stay – so Andrew’s big takehome is that for bronchiolitis we shouldn’t be giving salbutamol (and 25% of kids in ED with bronchiolitis do get it).

So what about the crossover between RSV bronchiolitis and asthma in those older infants (around 1yr of age, where the crossover becomes more blurry) – well, we can consider trialling salbutamol and if there is a response we should probably consider an asthma diagnosis and for patients highly responsive to salbutamol, adding a steroid would be appropriate. Most of the children who have repeated wheezy episodes in early childhood grow out of it and don’t become asthmatic – and there’s definitely an association between RSV exposure and the development of childhood asthma, though it’s not clearly delineated. There is a strong association between rhinovirus exposure and IgE to peanut developing wheezy episodes, but this wasn’t replicated for RSV A or B.

Antibiotics are not indicated^​11​ for the majority of bronchiolitis presentations; for those admitted to PICU, those with no fever, no infiltrates on CXR and PCR positive do not get antibiotics but most others do (based on this paper^​12​, showing high rates of bacterial growth from BAL samples of kids in PICU with bronchiolitis).

For mechanical support – up to 2L/kg/min of high flow does seem to decrease work of breathing and patients look better but this doesn’t translate into truly meaningful clinical outcomes; the evidence^​13​ for higher rates does not support them, and if they do really need respiratory support CPAP is better than high flow^​14​.

Donovan Dwyer – Tips and Tricks for Painful Procedures

Donovan’s structure for thinking about painful procedures is PLAN – Parents/Psychological/Play Therapy/Pain Score/Place, Local Anaesthetic, Analgesic, Nitrous/Narcoleptics and other sedation

A lot of his thoughts around the P part of his structure of managing kids with procedures is covered in my SMACC talk, which you can find here. I’ve recently updated the talk and presented it at the Sydney HEMS education day – that version will be available online soon and I’ll add the link here when it is. He also made a good case for procedure rooms for paediatric procedures; providing a separate space for procedures to take place that is away from other children in the department (so they don’t hear the scary sounds from behind the curtain and think you are coming for them next) and so that the child undergoing the procedure has a safe place to retreat to with their family afterwards. Makes sense to me.

We are pretty good at thinking about topical local anaesthetic for venepuncture and cannulation, but there are other procedures (LP, suprapubic aspiration etc) where topical local might help too. I covered this in the context of femoral nerve blocks a long long time ago… Do you routinely give co-phenylcaine for NGT insertion? We probably should. What about local anaesthetic topically, perianally prior to enema usage?

A nice tip is the technique of trimming the needle cap down to prevent the full length of the needle being able to be plunged (eg for suprapubic aspiration or for quinsy aspiration):

Donovan also has a well-practiced technique for nasal FB:

For gluing wounds, the choice of tissue adhesive matters for paediatric patients – liquiband is less painful than dermabond^​15​and works better if we combine with steristrips (and cover with a non-adherent dressing that doesn’t stick to the steristrips). Dehiscence risk should be part of the way we consent for and explain this wound care technique.

Local anaesthetic injection is painful and there are several things we can do to make it less painful.

Digital nerve blocks can be undertaken as a single shot from the volar aspect^​15​ and this might be more effective for fingertip injuries – posterior tibial blocks are very effective for sole of foot lesions too. Vein lies posterior to posterior tibial artery, landmark works well especially if local anaesthetic is used first.

Regarding analgesia, Donovan echoed a lot of the points made in Greta Palmer’s talk as part of the 2021 PEM Colloquium (my notes are here).

Be aware of the role of dead space in the circuit for free-flowing nitrous in the under 5s (see image) as this can mean they aren’t actually getting any nitrous.

Regarding ketamine sedation – remember that the risk of vomiting is not directly related to fasting time; however, ondansetron administration pre-procedure can significantly reduce vomiting. Midazolam pre-med doesn’t work brilliantly for children undergoing ketamine sedation and generally just prolongs recovery time. Propofol has the fastest recovery profile but the trade-off is greater haemodynamic instability and apnoea.

He wrapped his talk (and the day 1 sessions) up with a summary of his approach to a 6-year-old with an abscess of the foot requiring debridement.

Don’t forget to talk about ongoing analgesia – children leave the Emergency Department with no pain due to our wonderful interventions but when the local wears off the pain will return – so give advice about getting on top of and staying on top of analgesia.

That’s all for day 1 – day 2 & 3 notes will be coming together soon!