Angharad Griffiths. VP shunts, Don't Forget the Bubbles, 2020. Available at:
https://doi.org/10.31440/DFTB.25996
Rhiannon is a 2 year old with spina bifida who had a VP shunt inserted during the first month. It was revised because of a “blockage” at 18 months. She has not been herself for the last 24hours; more lethargic than usual- especially this morning when she woke up where she also felt hot. She vomited in the car on the way to the Emergency Department. She has had previous urinary tract infections.
Rhiannon was afebrile on presentation in ED. In triage, she was observed as being awake and drinking from a bottle but wasn’t perturbed by having her observations taken or finger-prick glucose. Overall she was a little quiet in the ED and her mum was concerned that this was absolutely different from her baseline. Her urine microscopy was not concerning. She had a mild neutrophilia and a normal CRP. Subcutaneous examination of her shunt, from her skull to the right clavicle was normal. She vomited once in the ED.
This is a common presentation and one that can be challenging to manage. Missing a shunt problem can have catastrophic consequences. This post will take you through some pearls and pitfalls of managing children with VP shunts presenting to the ED.
What is a VP shunt?
A ventriculoperitoneal (VP) shunt is a medical device used to drain fluid via a pressure gradient, away from the brain for conditions of excessive cerebrospinal fluid (CSF). The intention is to shunt fluid away and avoid excessive pressure on the brain. It is one of the commonest performed neurosurgical procedures and is the treatment of choice for the vast majority of patients with hydrocephalus.
Shunts drain according to the differential pressure gradient between the ventricle and the tip of the distal catheter. The ventricular end of the catheter is inserted through a burr hole in the right parietooccipital region and the valve often sits behind the right ear. The distal portion is subcutaneously tunneled down into the abdomen where it’s positioned inside the peritoneal cavity.
- Case courtesy of Assoc Prof Frank Gaillard, Radiopaedia.org. From the case rID: 6079
- Case courtesy of Assoc Prof Frank Gaillard, Radiopaedia.org. From the case rID: 6079
The diagnosis of raised intracranial pressure in children with VP shunts is challenging. The symptoms are non-specific and the commonest causes often benign. Rhiannon could easily have a simple viral illness or her symptoms could be associated with rising intracranial pressure. Missing a shunt malfunction in these patients can be catastrophic.
Physiology of CSF circulation and drainage
- CSF is produced at a rate of approximately 20ml/h in children and adults. In normal circumstances, in a normal adult, CSF is recycled three times a day. The normal adult volumes of CSF (approximately 150mls) are reached by age 5. The volume in a neonate can be estimated at 2ml/kg.
- Around 50% of CSF is created by the choroid plexus of the lateral, third and fourth ventricles. The rest of the CSF arises from the extracellular fluid of the brain. CSF travels out of the foramen of Lushka and foramen of Magendie (at the level of the fourth ventricle), and heads through to the subarachnoid spaces, along the spinal cord to the cerebral hemispheres. Over the cerebral hemispheres, the CSF is reabsorbed by arachnoid villi and then into the venous sinuses which drain into the jugular (internal) veins.
- Intracranial pressure (ICP) rises when CSF production exceeds absorption.
- Hydrocephalus is the consequence of the excessive accumulation of CSF. This could be from disruptions in formation, flow or absorption.
- As hydrocephalus worsens and intracranial pressure increases, the temporal and frontal horns dilate, sometimes asymmetrically; and there’s elevation of the corpus callosum and stretching of white matter tracts.
Shunting CSF is an effective way to avoid the neurological damage that ensues if the build-up of CSF is left untreated.
Three shunts types are mainly used to shunt CSF: Ventriculoperitoneal (VP), ventriculopleural (VPL), and ventriculoatrial (VA). By far, the commonest are VP shunts.
Reasons for shunt placement
There is a myriad of reasons as to why a VP shunt should be placed. They can be categorized into congenital or acquired causes.
Having a VP shunt in itself is a cause for concern for patients and caregivers. Common areas for concern in paediatric patients include:
- Flying and travel: there is no evidence that flying is dangerous but patients have concern over access to neurosurgical help should they need it. Shunt alert cards displaying the type of shunt are available. The need for augmented travel insurance is also an area of concern.
- Sports: Contact sports such as boxing are contra-indicated, and the US paediatric neurosurgeons named wrestling and football (soccer) as the commonest sports with adverse events. Neurosurgeons in the UK are advising that football and rugby wearing a skullcap are acceptable. You can still go scuba diving.
- Future employment: in the UK, the Royal Air Force, Royal Navy and Police Force are the only services that preclude entry.
- Programmable shunts and magnets: Background magnetic fields of household objects such as microwaves etc are safe as are walk-through metal detectors. Post MRI check is advised with some programmable shunts but all are MRI safe. The iPad2 has a strong magnetic field and can re-program some shunt valves, thus important to keep them at safe distances.
- Shunt length: reassurance that placing sufficient length inside the abdomen will suffice and allow for growth.
- Weight: Obesity is a risk factor for failure of VP shunts and dislodgement.
Shunt related complications
Failure rates are quoted as 30-40% at 1 year and 50% at 2 years in the paediatric cohort. A patient can expect to have 2-3 shunt revisions over the course of 20 years and the median time to shunt failure is just 1 and a half years. Paediatric revisions are more commonplace than adult revisions.
Risk factors for shunt failure include:
- Younger patients (<6months), particularly neonates
- Complex comorbidities, for example, cardiac, myelomeningocele, IVH’s, tumour and post-meningitic hydrocephalus, spinal dysraphism, and congenital hydrocephalus
- Prior shunt failure and short time intervals between revisions
- Male sex
- Low socioeconomic status
Causes of shunt failure
Obstruction, blockage or occlusion
The commonest cause of shunt malfunction is proximal occlusion. There’s no clear data on whether programmable or non-programmable shunts are less likely to occlude.
It’s hypothesized that on insertion, the lumen can be obstructed with brain parenchyma from the cerebral cortex before reaching the ventricle, or choroid plexus when inserted near to the foramen of Monroe. It can also be occluded with blood following choroid plexus haemorrhage along with other cellular matter such as macrophages, giant cells, connective tissue, fibrin networks, debris, neoplastic cells.
Distal catheter blockage tends to occur later and when it occurs, it should raise the suspicion for an intra-abdominal pseudocyst or adhesions which may affect future peritoneal catheter placements.
Shunt infection
Shunt infection is the second most common reason for malfunction. The data is mixed, particularly as some older papers use data from before the time of antibiotic-impregnated catheters, skewing the data. Risk factors include young age (including neonates), post-op CSF leak, previous shunt infections, and the presence of a gastrostomy tube.
The vast majority of shunt infections are acute. Far fewer late-onset infections have been reported. They can be attributed, mostly, to peritonitis, abdominal pseudocyst, bowel perforation, and haematogenous inoculation.
Shunt infection is associated with an increased risk of seizure disorder, decreased intellectual performance, and a two-fold increase in long term mortality. Re-infection occurs in one-quarter of children.
The proportion of shunt infections falls off rapidly after the first several months following implantation. The vast majority occur during the first 6 months. Children will present with signs of shunt failure, as well as systemic infection. Fever is common and if we were to sample CSF (which is not often done in the ED), the presence of >10% neutrophils in the ventricular fluid is highly specific and sensitive of infection.
The commonest organisms are skin flora, including Staphylococcus epidermidis, Staph. aureus and gram-negative rods. Infections with Staph. aureus and epidermidis are associated with an earlier onset as they are skin commensals. Infections with Staph. aureus are associated with a significantly increased likelihood of subsequent shunt infection.
Shunt fracture
This is often a late complication and almost always occurs along the distal portion between the valve and peritoneum. With age, fibrous tissue becomes calcified and does not slide freely within the subcutaneous tissue then the tubing can crack. This is more likely to happen in the neck where most movement occurs.
Tension can also form along areas of calcification causing tethering and stretching as the child grows. It is important to note that early shunt fracture can occur and this could be a consequence of trauma to the tubing during surgery.
CSF can still drain resulting in an insidious duration of symptoms, clouding and confusing the diagnostic process.
Shunt series radiographs should always be sought in this cohort, though frequently fractures and disconnections are incidental findings during surveillance exams.
VP shunt fracture. Reused with permission from Education and Practice, Archives of Disease of Childhood
Shunt disconnections
Catheters are generally multi-component and hubbed together so disconnections can occur soon after surgery. The disconnection impedes the flow of CSF and it may still leak. The onset of these symptoms may be slow. Disconnections can happen at either the proximal or distal aspect of the valve.
Case courtesy of Dr Adam Eid Ramsey, Radiopaedia.org. From the case rID: 71794
Abdominal pseudocyst
This is a rare complication of VP shunts and is usually a late complication occurring years after initial placement. A pseudocysts is a fluid-filled sac that collects at the distal tip of the catheter. It is thought that they form because of inflammation or due to abdominal adhesions. It can present with abdominal pain or distention with, or without, a palpable abdominal mass. Neurological symptoms occur when there is elevated ICP.
Shunt migration
The proximal or distal catheter tip may migrate. With growth, the proximal catheter can withdraw from the ventricle (extremely rare), or the distal catheter can shift away from the peritoneum. The distal tubing can become tethered and cause traction on some of the components causing a disconnection. Distal migration occurs as the child grows.
Over-drainage
It is possible that a VP shunt can over-drain as well and ‘under-drain’. With rapid over-drainage, the dura can be stressed and subdural haematomas and/or extra-axial fluid collections can form.
A slit ventricular syndrome occurs when gravitational forces exert a siphoning effect on the ventricles. This effect is generally amplified by pressure.
Clinical presentation of shunt malfunction
Children with a blocked shunt can present with a myriad of symptoms including:
- headache
- nausea
- vomiting
- fever
- irritability
- abnormal level of consciousness
Infants and older children may present differently.
Infants
- difficulty feeding
- bulging fontanelle
Older children may present more specifically with
- Nausea, somnolence, lethargy, cognitive difficulties, or eye pain.
Predictably, fever is commoner in children with shunt infections. Those with shunts because of myelomeningoceles may present with symptoms such as:-
- weakness,
- difficulty walking
- bowel/bladder dysfunction
- lower cranial nerve palsies.
Children present with these symptoms all the time to the ED. They are clearly not specific to a shunt problem. As a consequence, diagnosing shunt malfunction on clinical grounds alone is incredibly difficult. Patients with shunt fracture or disconnection can present with a slow onset of symptoms. They may have pain/tenderness localized to the area of fracture/disconnection or an area of calcification of an area of fluctuant swelling.
Diagnosis, evaluation, and imaging
The diagnosis of a shunt malfunction requires a combination of CT, shunt series radiographs, and occasionally (though seldom in the ED), CSF sampling.
A CT is likely to show an increase in ventricular size and occasionally, periventricular lucency representing oedema. There may be increasing ventricular size on cross-sectional imaging but up to 15% will have “such profound alterations on brain compliance that their ventricles will not enlarge in the face of shunt failure and increased ICP”. Ventricular size doesn’t appear to reach a plateau until approximately 14months after placement of the shunt (regardless of type implanted).
A lumbar puncture (LP) may demonstrate increased opening pressures, but not always. It is also used for evidence of infection. This not performed commonly in the ED in the context of possible shunt malfunction.
Shunt series (SS) radiographs are used to check the overall course of the catheter, looking for disconnection or disruption. The series will not show obstructions, only damage to the catheter. It can rarely demonstrate complications such as a CSF pseudocyst (abnormal separation of bowel loops near the catheter tip) but shouldn’t be relied upon for this.
The number of radiographs needed varies according to the size of the child. It is usually 3-4 radiographs, including two views of the skull and the continuous trajectory of the shunt tubing down the neck, chest, and then looping into the abdomen.
If a series is performed after the scan, theoretically a 2 view skull radiographs can be eliminated, provided that the chest x-ray includes the base of the neck. Unnecessary radiation may then be avoided.
The use of ultrasound is an area of ongoing research and has been largely unvalidated in children with VP shunts.
No clear cause for Rhiannon’s symptoms was found following a thorough examination. A CT was performed because of concern over shunt failure. Her ventricles were noted to be slightly larger than a CT performed previously. Shunt series radiographs showed continuous, non-kinked tubing. She was admitted under the care of the Neurosurgeons and her shunt was revised. No physical reason for shunt obstruction was found.
Selected references
Mansson PK, Johansson S, Ziebell M, Juhler M. Forty years of shunt surgery at Rigshospitalet, Denmark: A retrospective study comparing past and present rates and causes of revision and infection. BMJ Open. 2017;7(1).
Berry JG, Hall MA, Ph D. A multi-institutional 5 year analysis of Initial and multiple ventricular shunt revisions in children. Neurosurgery. 2008;62(2):445–54.
Pople IK. Hydrocephalus and shunts: What the neurologist should know. Neurol Pract. 2002;73(1).
Paff M, Alexandru-Abrams D, Muhonen M, Loudon W. Ventriculoperitoneal shunt complications: A review. Interdiscip Neurosurg Adv Tech Case Manag. 2018;13(June 2017):66–70.
Gonzalez DO, Mahida JB, Asti L, Ambeba EJ, Kenney B, Governale L, et al. Predictors of Ventriculoperitoneal Shunt Failure in Children Undergoing Initial Placement or Revision. Pediatr Neurosurg. 2016;52(1):6–12.
Wu Y. V Entriculoperitoneal S Hunt C Omplications in C Alifornia : 1990 To 2000. 2007;61(3):557–63.
Brinker T, Stopa E, Morrison J, Klinge P. A new look at cerebrospinal fluid circulation. Fluids Barriers CNS. 2014;11(1):1–16.
Rochette A, Malenfant Rancourt MP, Sola C, Prodhomme O, Saguintaah M, Schaub R, et al. Cerebrospinal fluid volume in neonates undergoing spinal anaesthesia: A descriptive magnetic resonance imaging study. Br J Anaesth [Internet]. 2016;117(2):214–9. Available from: https://dx.doi.org/10.1093/bja/aew185
Desai KR, Babb JS, Amodio JB. The utility of the plain radiograph “shunt series” in the evaluation of suspected ventriculoperitoneal shunt failure in pediatric patients. Pediatr Radiol. 2007;37(5):452–6.
Stone JJ, Walker CT, Jacobson M, Phillips V, Silberstein HJ. Revision rate of pediatric ventriculoperitoneal shunts after 15 years: Clinical article. J Neurosurg Pediatr. 2013;11(1):15–9.
Brian W. Hanak et al. Cerebrospinal fluid shunting compliations in children. Pediatr Neur. 2017;52(6):381–400.
Hanak BW, Ross EF, Harris CA, Browd SR, Shain W. Toward a better understanding of the cellular basis for cerebrospinal fluid shunt obstruction: Report on the construction of a bank of explanted hydrocephalus devices. J Neurosurg Pediatr. 2016;18(2):213–23.
Khan F, Shamim MS, Rehman A, Bari ME. Analysis of factors affecting ventriculoperitoneal shunt survival in pediatric patients. Child’s Nerv Syst. 2013;29(5):791–802.
Shastin D, Zaben M, Leach P. Life with a cerebrospinal fluid (CSF) shunt. BMJ [Internet]. 2016;355(October):1–5. Available from: https://dx.doi.org/doi:10.1136/bmj.i5209
Simon TD, Butler J, Whitlock KB, Browd SR, Holubkov R, Kestle JRW, et al. Risk factors for first cerebrospinal fluid shunt infection: Findings from a multi-center prospective cohort study. J Pediatr [Internet]. 2014;164(6):1462-1468.e2. Available from: https://dx.doi.org/10.1016/j.jpeds.2014.02.013
Buster BE, Bonney PA, Cheema AA, Glenn CA, Conner AK, Safavi-Abbasi S, et al. Proximal ventricular shunt malfunctions in children: Factors associated with failure. J Clin Neurosci [Internet]. 2016;24:94–8. Available from: https://dx.doi.org/10.1016/j.jocn.2015.08.024
Mcgirt MJ, Zaas A, Fuchs HE, George TM, Kaye K, Sexton DJ. Factors Infecton for Pediatrc and Venticulopertoneal of Shunlt Predictors Infectous Patiogens. 2014;36(7):858–62.
Mcclinton D, Carraccio C, Englander R. Predictors of ventriculoperitoneal shunt pathology. Pediatr Infect Dis J. 2001;20(6):593–7.
Erol FS, Ozturk S, Akgun B, Kaplan M. Ventriculoperitoneal shunt malfunction caused by fractures and disconnections over 10 years of follow-up. Child’s Nerv Syst. 2017;33(3):475–81.
Dabdoub CB, Dabdoub CF, Chavez M, Villarroel J, Ferrufino JL, Coimbra A, et al. Abdominal cerebrospinal fluid pseudocyst: A comparative analysis between children and adults. Child’s Nerv Syst. 2014;30(4):579–89.
Boyle TP, Kimia AA, Nigrovic LE. Validating a clinical prediction rule for ventricular shunt malfunction. Pediatr Emerg Care. 2018;34(11):751–6.
Tuli S, O’Hayon B, Drake J, Clarke M, Kestle J. Change in ventricular size and effect of ventricular catheter placement in pediatric patients with shunted hydrocephalus. Neurosurgery. 1999;45(6):1329–35.
DeFlorio RM, Shah CC. Techniques that decrease or eliminate ionizing radiation for evaluation of ventricular shunts in children with hydrocephalus. Semin Ultrasound, CT MRI [Internet]. 2014;35(4):365–73. Available from: https://dx.doi.org/10.1053/j.sult.2014.05.002
Smyth MD, Narayan P, Tubbs RS, Leonard JR, Park TS, Loukas M, et al. Cumulative diagnostic radiation exposure in children with ventriculoperitoneal shunts: A review. Child’s Nerv Syst. 2008;24(4):493–7.
PEM Adventures Chapter 1
Dani Hall, Rachael Mitchell and Sarah Davies. PEM Adventures Chapter 1, Don't Forget the Bubbles, 2020. Available at:
https://doi.org/10.31440/DFTB.29589
Stories are a powerful vehicle for education. Combine a story with some active participation and you have the recipe for some great learning. And so, it’s with great delight, that we bring you Chapter 1 of PEM Adventures. First presented at EuSEM 2018 and then again with some spectacular twists by Dan Lumsden, Paediatric Neurologist extraordinaire, Dani has a particular soft spot for Tomas, a little boy who dreams of being a footballer. Join us on a journey (with an inbuilt time travel machine) in managing Tomas, a little boy with a dream…
Meet Tomas, an 8-year-old boy who dreams of playing professional football. He’s been completely well until an ill-fated shopping trip for some new football boots. At 2 o’clock hours, while trying to persuade his mother that he definitely did need the new Premier League football to add to the collection, he developed sudden onset right-sided facial drooping. His mum bundled him into the car and drove him directly to your ED. You look at your watch: it’s now 3.30 pm.
Your assessment is as follows: Tomas is alert and he seems orientated. He has right sided facial weakness and weakness of both his right arm and leg. He has no obvious sensory changes but is struggling to communicate with you as he has global aphasia.
Suspecting the worse, you have a critical decision to make. But what are you going to do?
Call neurology
You bleep neurology.
And wait…
They don’t answer so you bleep again…
But they still don’t answer…
You bleep a third time…
But they still don’t answer. So you give up and call radiology instead.
Call radiology
You call radiology. And they ask… What imaging do you want?
CT? CT with contrast? CTA?
The radiologist says, “Sure, imaging sounds like a good idea. Let’s do a combination of both a CT brain with CTA to look for blood and clots.”
The CT scanner is available at 16:00.
“But,” she adds, “if you’d like MR imaging, we could do that at 18:00.”
Do you…
Proceed with CT and CTA
Tomas’ CT and CTA shows evidence of an arterial ischaemic stroke with thrombus occluding the middle cerebral artery. There is no intracranial haemorrhage.
It is now 16:15, 2 ¼ hours after the onset of Tomas’ symptoms.
You’re doing great. Close the toggles and move on to the next part of Tomas’ case.
Say, “Actually, I’ll have MR imaging please.”
Tomas has an MRI and MRA.
It shows an arterial ischaemic stroke with thrombus occluding the middle cerebral artery.
It is now 18:45, almost 5 hours after symptom onset – something tells you this is a bit too late.
Luckily for you, the inbuilt time machine whizzes you back to make that last decision again.
This time when you’re told you can have an MR and MRA at 18:00 or CT and CTA now you say… “I’ll have a CT and CTA now please.”
MRI? MRV? MRA?
The radiologist says “Sure, imaging like a good idea. Let’s do a combination of both an MRI plus MRA to check the brain and look for clots.”
You phone MR. They say they can do the MR at 18:00. The CT scanner, however, is free now.
Do you…
Proceed with MR at 18:00?
Tomas has an MRI and MRA. It shows an arterial ischaemic stroke with thrombus occluding the middle cerebral artery. It is now 18:45, almost 5 hours after symptom onset – something tells you this is a bit too late.
Let’s travel back in time…
This time when you’re told you can have an MR and MRA at 18:00 or CT and CTA now you say… “I’ll have a CT and CTA now please.”
Say, no thanks, I’ll have a CT and CTA please.
Tomas has a CT and CTA. It shows evidence of an arterial ischaemic stroke with thrombus occluding the middle cerebral artery. There is no intracranial haemorrhage.
It is now 16:15, 2 ¼ hours after the onset of Tomas’ symptoms.
You’re doing great. Close the toggles and move on to the next part of Tomas’ case.
With a little luck, Tomas has now had neuroimaging and you know he’s had an arterial ischaemic stroke with thrombus occluding the middle cerebral artery without intracranial haemorrhage.
So, what now? You haven’t managed to get hold of a neurologist for love nor money. So do you…
Admit for supportive care
Tomas has supportive care.
Despite physio, OT and lots of assistance at the best neuro-rehabilitation centre, Tomas has a persistent hemiparesis.
You spend your life wishing you’d treated his stroke differently.
So let’s try that choice again.
Give Tomas aspirin
You prescribe 5mg/kg aspirin.
Tomas has repeat imaging with an MRI and MRA 24 hours later. His clot has not increased in size but the original clot remains in the middle cerebral artery.
Tomas has a persistent hemiparesis.
He becomes a demon-swimmer and wins Gold in the 2028 Paralympics
However, you spend your life wishing you’d treated his stroke differently.
Why don’t you try that choice again.
Anticoagulate with heparin
You prescribe heparin.
24 hours later Tomas deteriorates, dropping his GCS to 6
Repeat neuroimaging shows a large haemorrhage in the infarcted territory with significant pressure effect.
Let’s go back in time and try that choice again.
Thrombolyse
You decide to thrombolyse. Tomas will need to go to PICU after thrombolysis but there isn’t a PICU at your hospital.
Do you…
Transfer to the regional centre and thrombolyse there
You opt for thrombolysis at the regional centre but will you…
Transfer yourself
You work fast to mobilise your anaesthetist, ED nurse and emergency kit as quickly as possible.
Tomas arrives at the regional centre at 19:15, 5 ¼ hours after the onset of his symptoms.
It is too late to thrombolyse.
Let’s hop in the time travel machine and go back in time to decide whether to transfer for thrombolysis or thrombolyse in your ED (Hint: you may want to thrombolyse in your own ED as the clock is ticking…)
Ask the retrieval team to transfer
The retrieval team are mobilised. They collect Tomas from your ED and deliver him safely to the regional centre at 19:15, 5 ¼ hours after the onset of Tomas’ symptoms.
But it’s now too late to thrombolyse.
Let’s hop in the time travel machine and go back in time to decide whether to transfer for thrombolysis or thrombolyse in your ED (Hint: you may want to thrombolyse in your own ED as the clock is ticking…)
Thrombolyse in your ED and then transfer
“Time is critical,” you think to yourself, and tell the team you’re going to thrombolyse in resus.
Tomas is thrombolysed with tissue plasminogen activator (tPA for short) at 18:00, 4 hours after onset of symptoms. His symptoms start to improve.
After intensive neuro-rehab he has no residual neurological deficit.
He grows up to become a professional football player for Bayern Munich, scoring a hat-trick to win the 2028 UEFA champion’s league.
Congratulations! You successfully treated a stroke in childhood. Now close the toggles and read on…
Organise an angiographic thrombectomy
Although you opt for angiographic thrombectomy, the interventional neuroradiologist is on study leave and no-one is able to cover.
You suspect they are actually scared of children.
Either way Tomas can’t have the clot removed. You’re going to have to choose again.
Refer to neurosurgeons for a hermicraniotomy
You phone the neurosurgeons and ask them to do a hemicraniectomy.
They ask you to go through all his neurology and review his imaging.
They say “Sorry, but his PedNIHSS isn’t high enough for us to take to theatre.”
You think, “PedNIHSS?” and make a mental note to look it up later.
Let’s try that choice again.
After your shift you do a quick google search to look at the evidence around using tPA in children and you stumble across this paper:
Rivkin, M.J., deVeber, G., Ichord, R.N., Kirton, A., Chan, A.K., Hovinga, C.A., Gill, J.C., Szabo, A., Hill, M.D., Scholz, K. and Amlie-Lefond, C., 2015. Thrombolysis in pediatric stroke study. Stroke. 2015: 46(3); 880-885.
Rivkin’s team were part of a huge multi-state stroke research team in North America. They designed the incredibly well thought out and well put together TIPS (Thrombolysis in Paediatric Stroke) study, to look at (A) safety of and (B) dose of tissue plasminogen activator (tPA) in children presenting with and arterial ischaemic stroke (AIS). They set out to recruit children aged 2 – 17 with acute AIS and PedNIHSS score between 4 – 24 to receive tPA if initiated within 4.5 hours of symptom onset. Centres were given protocols to manage complications such as intracranial haemorrhage, systemic bleeding, hypotension or angioedema.
Sounds good, right?
Well, in principle, yes. The study opened in April 2012 but closed only 20 months later in December 2013 because only 1 child had been enrolled and they hadn’t actually been treated due to complications following extubation prior to tPA administration.
93 children had been screened with 43 having confirmed AIS and the other 50 having a stroke mimic such as migraine, seizure or tumour etc.
Of the 43 children with AIS about half had medical contraindications to tPA (including moyamoya disease & anticoagulation treatment); 10 were outside the treatment window (including 1 who missed the treatment window by 15 minutes due to delay at scanner); some had a PedNIHSS that was too low ; 1 had a PedNIHSS that was too high; and a couple didn’t have arterial occlusion on imaging.
But it wasn’t a total disaster. Preparing for TIPS also led to the development of Paediatric Stroke Networks in North America. And designing the TIPS study led to consensus guidelines on the management of stroke in children.
These consensus guidelines derived from the TIPS study design have been extrapolated to the 2017 RCPCH Stroke in Childhood guideline, based on expert opinion and the best available evidence. As well as the full guideline, there’s a simple, easy to follow pathway poster that can be grabbed for quick reference whenever a child presents with potential stroke symptoms.
The poster gives a list of potential stroke presentations, from an unexplained persistent drop in GCS, through acute focal neurology (even if resolved), focal seizures, headaches, ataxia, dizziness, speech disturbance and a prompt to consider stroke in children with sickle cell disease.
It includes a simple, easy to follow, Paediatric National Institute of Health Stroke Scale (that PedNIHSS we’ve talked about) a bit like a Glasgow Coma Scale but specific for paediatric stroke. The PedNIHSS makes up a really important part of the neurological assessment, a way of scoring the severity of the stroke. It is vitally important that the PedNIHSS is calculated because if the score is very low, with a very minimal deficit at the outset, the risk of thrombolysis outweighs the potential benefit. And if the PedNIHSS score is very high, it’s likely that the child has a very large area of brain damage, with a high risk of haemorrhage into that infarcted territory, again making the risk : benefit ratio too risky. The child’s PedNIHSS score guides your subsequent management.
The pathway lists investigations (which must include coagulation profile and group and save, because of that risk of bleeding), monitoring and neuroimaging. Timing of imaging is key. The guideline states that children should be scanned within 1 hour of presentation to the ED. Pragmatically, this is usually CT with CTA (the angiography component to look at the arteries), because organising an MRI with MRA takes longer. But, if you’re in an institution with great access to MR and you can get your imaging within an hour of presentation then it’s definitely worth a discussion with the radiologist.
If a child has a confirmed AIS, what do we do? The guideline offers two either / or treatments: EITHER aspirin 5mg/kg within an hour, as long as there is no parenchymal haemorrhage OR thrombolysis. The guideline suggests that thrombolysis may be considered in children aged 2-8 and could be considered in children over 8 (some careful wording there because extrapolating evidence from adult studies to an 8 year old is easier than to a 2 year old) provided the PedNIHSS is between 4 and 24 and tPA can be administered within 4.5 hours of symptom onset. There must be either MRA evidence of thrombus or normal or only minimally ischaemic changes on CTA (no huge areas of ischaemia, because the risk of bleeding into it is just too high), with or without evidence of thrombus. And as the biggest risk of giving tPA is haemorrhage, there must be no contraindications, such as abnormal clotting, an underlying bleeding disorder, malignancy, hypertension or moyamoya disease.
It’s really important to note that the treatment for AIS is not the same as for a child with a haemorrhagic stroke (these children need urgent discussion with a neurosurgeon for consideration of evacuation) or a child with an ischaemic stroke secondary to sickle cell disease (pick up the phone, call a haematologist and organise an exchange transfusion). Although not included on the poster, the guideline summary and full guideline give indications for surgical and endovascular interventions in stroke, as well as those nuggets for managing stroke in a child with sickle cell disease or haemorrhagic stroke.
And what about thrombectomy? This is a very active area of interest. In the world of adult AIS there has been a big move towards primary clot removal by thrombectomy rather than clot busting with thrombolysis. In the world of paediatric stroke, although there are some published case series and case reports, we don’t have a clear evidence base or national guidance. Yet.
So, what is the take-home from Tomas’ case? Although stroke is rare in children, it does occur. Thrombolysis is a potential management option given the right conditions, as long as it’s given within the 4.5-hour window. So, next time you see a child with stroke-like symptoms, send bloods early, get early neuroimaging with angio, and pull out the RCPCH Stroke in Childhood poster.
With special thanks to Dr Dan Lumsden, Paediatric Neurologist at the Evelina London Children’s Hospital, who inspired the creation of Tomas’ case and presented him so fabulously at the Royal Society of Medicine. Thank you, Dan.