All posts by admin

Valproate – MHRA warning

General paediatrics, Neurology,

Updated December 2023 –

High risk of serious developmental disorders if exposed in womb (up to 30-40% risk, including 5x higher risk of autism) and congenital malformations (10%), including:

  • Spina bifida
  • face/skull malformations, including cleft lip/palate
  • Limb, heart, kidney, genital abnormalities
  • Deafness

In boys, pre-clinical data on transgenerational risks, and animal studies suggesting infertility.

Oral Valproate must not be started in new patients (male or female) younger than 55 years, unless two specialists independently consider and document that there is no other effective or tolerated treatment, or there are compelling reasons that the reproductive risks do not apply.”

Patient guide and checklist available.

Testing for antibiotic allergy

Allergy, General paediatrics

See also Penicillin allergy.

Systematic review – just 0.21% of unselected general paediatric outpatients exhibit positive antibiotic allergy tests, and only 6.8% of those with suspected allergy test positive.

No evidence to support using skin prick testing.  Intradermal testing has high false positive rate (64-67% for penicillin and clarithromycin). Caubet did oral provocation test (OPT) regardless of intradermal result to beta lactam, NNT=11 to avoid one OPT! [Ped All Immun 2015 ]. OPT reactions tend to be cutaneous and mild, usually more than 1hr post administration.

Where reaction is severe but non-immediate, eg Stevens Johnson syndrome, Toxic Epidermal Necrolysis (TEN), Drug Rash with Eosinophilia and Systemic Symptoms (DRESS), care needs to be taken with investigation, but studies have reported intradermal and OPT without unsafe adverse effects.  If reaction is anaphylaxis with first dose, then OPT contraindicated.

Wide variety in regimens.  Suggests 1-5 incremental doses of amoxicillin for mild reactions (timing not specified), giving cumulative dose appropriate to child, then continuing for 3 days.  Skin testing may be appropriate for severe reactions according to risk:benefit balance.  Check asthma well controlled, no antihistamines.  Warn that low risk of false negative result (absent co-factors) and low risk of re-sensitization.

Intravenous provocation only where PICU!

[Marrs, ArchDisChild 2015]

Macy article says any rash can have OPT! 5mm pos SPT for penicilloyl-polylysine has good negative predictive value for anaphylaxis with OPT. Recommends 5 days amoxicillin.

[Ann Allergy Asthma Immunol 121(2018):523−529]

Mirakian article suggests SPT for all immediate reactions! Split dose challenges, with a week between first and second doses!

For non immediate reactions (1-72hrs), OPT confirmed in 59%, ID less than 40%.

6 studies showing that benign reactions (ie witnessed macpap or urticarial, no pain/burning, <50% skin surface etc) do not need skin testing.  Geneva have done more than 800 straight to OPT.  New EAACI guidelines in press.

If delay in reaction is unclear, assume immediate.  SPT vs amoxicillin, PPL, MDM.  IgE vs BPL.  0.04ml ID volume.  Note different reference ranges!  See Brockour, Allergy 13

Recent letter claimed OPT after skin test was “unnecessary, dangerous, unethical”!  But 30-100% false negatives!

Clavulanate allergy described.

Test sensitivity falls more than 4/12 after episode, ideally do within 4-6/52????

Basophil activation test using flow cytometry looking promising for IgE mediated drug reactions.  EAACI interest group working on Drug Allergy Passport.

Atypical HUS

General paediatrics, Genetics, Haem/Oncology, Renal

Most Haemolytic Uraemic Syndrome is associated with a diarrhoeal illness (D+ HUS), esp E coli O157.

Atypical HUS is a bad name for HUS that develops in certain individuals due to a genetic, complement disorder.  Disease is typically triggered by infection, esp respiratory tract, else diarrheal illness in 80% of pediatric forms.

Consider in young infants (less than 5% of D+HUS cases occur before the age of 6 mo), severe cases, non-colitis.

The penetrance of the disease among carriers of mutations in CFH, CFI and MCP genes is approximately 50%-60%.

Low C3 levels are a clue (seen with mutations in CFH, CFI and MCP).  In almost all cases of aHUS C4 levels are normal. Normal C3 levels do not however exclude a mutation.  Check factor H and factor I too.

Diagnosis

Measure CFH, CFI and MCP levels using radio-immune-diffusion assay (RIDA) or FACS. This however fails to detect low protein levels in 25%-75% of mutations, so genetic analysis also needed.

Check ADAMTS13 activity (as seen in thrombotic thrombocytopaenic Purpura, TTP) as part of differential.

Make sure you have enough blood samples before plasma exchange!

In neonates, screen for defective cobalamin metabolism (excess homocystine and methylmalonate in urine ).  These babies have high mortality from multiorgan failure, a prompt diagnosis and B12 supplementation is their only hope.

Haemolytic Uraemic Syndrome

Gastro, General paediatrics, Infectious disease, Renal

Or HUS for short.  Mainly caused by VTEC (Verocytotoxin producing E coli), esp serotype O157.  But note also:

  • Occasionally Pneumococcal
  • Atypical HUS – not infective, but a genetic complement disorder
  • D- HUS is the shorthand for HUS which manifests in absence of diarrhoeal illness.  Still worth looking for E coli O157, but need to look at other possibilities.

Clinically

What it says on the tin!

  • Haemolysis
  • Uraemia

Mechanism is Thrombotic Microangiopathy, with microthrombi circulating that affect not just kidneys (producing uraemia) but also:

  • Thrombocytopenia,
  • MI/stroke/infarction
  • Encephalopathy

Differential

  • Acute dysentery eg campylobacter, shigella
  • Intussusception
  • Atypical HUS
  • Acute colitis (Ulcerative Colitis)

Although atypical HUS has a genetic cause, it is typically triggered by infection, esp respiratory tract, else diarrheal illness in 80% of kids, so may be difficult to exclude.  Consider in young infants, severe cases, non-colitis. Low C3 levels are a clue.

Acute colitis will usually have more extensive history or other clues.  The emphasis on managing these cases is on identifying toxic megacolon, and surgical intervention if necessary.  Antibiotics are recommended if high risk of infection (eg signs of sepsis) and/or immediately pre-surgery.

Diagnosis

Features of established disease are:

  • Microangiopathic haemolysis
    • Falling Hb, Plts – clinically pallor, petechiae and bruising
    • Fragmented rbc’s on film
    • High LDH
    • Blood/protein on urinalysis (if there’s any urine being produced to collect)
  • Rising Urea/Creatinine

But ideally you get the diagnosis early, before too much damage has been done.  Clues are:

  • Rising LDH
  • Falling Plts
  • Oliguria
  • Blood/protein on urinalysis

Stool results, to confirm E coli O157, take 24-48hrs (culture and toxin test). Discuss with microbiology if stool culture negative, consider gene PCR, serology.

[Pediatr Int. 2009 Apr;51(2):216-9 PMID 19405919]

Progression

But not everyone who gets E coli O157 gets HUS. Some people get diarrhoea without progressing to colitis (prob the majority), some get colitis without progressing to HUS (only a minority).  Predictors of progression to HUS are:

  • Fever (usually not at presentation, but in history – not very specific)
  • WCC >11
    • Normal WBC provides reliable reassurance against progression to HUS in 9/10
    • WCC>11 predicts progression to HUS in 70–90% of children
  • Raised CRP

[Ikeda, Epid infection 2000; 124:343.  Archimedes, Arch Dis Child 2007]

Ikeda proposes scoring system of 2+ features of fever, high CRP and WCC – 32% sens cf 98% spec.

Public Health

Preventing further cases is as important as managing your case.  About 20% of cases in outbreaks are secondary, but this increases to over 50% of cases in under 6yrs.  Interestingly, secondary cases are usually other children in the nursery (but rarely adults) if pre-school, but family members if primary/secondary school age!  [BMC Infect Dis. 2009 Aug 28;9:144. doi: 10.1186/1471-2334-9-144] Notify Health Protection team on suspicion of E. coli O157 syndrome.

Antibiotics?

Controversial.  Would be considered for other infective dysenteries viz Clarithro for campylobacter, Cipro/cef for salmonella, Azithro for shigella.

But do they precipitate HUS in E coli O157? “Exposure to antibiotics (aOR 3.62; 95% CI, 1.23–10.6) in 1st week independently associated with development of HUS” [Wong 2012].  Current 2011 UK guidance states “The use of antibiotics should, therefore, be governed by good paediatric practice as indicated by needs other than the suspicion of enteric VTEC infection”.

Fluids etc

Loperamide has traditionally been associated with toxic megacolon in acute dysentery so is not advised.

Pain – can be significant with colitis.  Opiates, NSAIDs contraindicated given potential for megacolon or nephropathy.

IV fluid volume and sodium during E coli O157:H7 infections, esp in first 4 days of illness, associated with oligoanuric HUS:

  • 1.6x more likely to become oligoanuric if no IV fluids were given during the first 4 pre-HUS days

[Christina Hickey (St Louis) and Jim Beattie, prospective study Arch Pediatr Adolesc Med. 2011;165(10):884-889. doi:10.1001/archpediatrics.2011.152]

Hence, seems reasonable to:

  • Give 20ml/kg at presentation
  • Rehydrate aggressively eg correct over 6-8hrs
  • Repeat boluses if urine output reduces

Monitoring

Since onset of HUS is from 5 to a maximum of 13 days into diarrhoea illness, there is a need to monitor those with suspected or proven E coli O157 without HUS viz repeat bloods at 5-7 days or earlier if symptoms worsen.  Consider admission if significant infection control issues eg young children/siblings.

Advanced Measures

  • Renal replacement therapy
  • Plasmapheresis (not done in Glasgow)
  • Eculizumab – drug of choice for aHUS, recommended for children in whom plasma exchange is technically difficult
  • Monoclonal antibodies vs STX1/2

Eculizumab = Recombinant monoclonal anti-C5 antibody. Orphan drug.  No good evidence from Germany, but they had good results from plasma exchange anyway.  Evidence from severe TTP that it is effective in cases resistant to immunosuppression and plasma exchange.

 

E. coli O157

Gastro, General paediatrics, Infectious disease, Microbiology, Renal

Notorious STEC/VTEC producing strain of E coli. STEC is shigella toxigenic E coli, VTEC is verocytotoxin toxigenic E coli (same thing).

Cause of potentially fatal Haemolytic uraemic syndrome. But other strains also recognised.

1996 Lanarkshire outbreak, traced back to meat pies from John Barr’s butcher’s in Wishaw.  21 deaths, 512 cases of HUS, esp Wishaw Parish Church Hall luncheon and a local pub birthday party. At the time it was the most deadly food related disease outbreak in world history.  As of 2023 it is still the 6th most deadly food related outbreak in history, and still the worst death toll from O157.

A sad claim to fame for the town.  John Barr was eventually fined £2500.

Emerged in late 80s, rates in Scotland have always been highest in UK until Northern Ireland outbreak with 140 cases in 2012. Odd because England has a lot more cows…

Scotland is said to have 2nd highest incidence globally, although not great data from many places, high regional variability, under-reporting…  Canada and Iran worse?

At the same time as the Lanarkshire outbreak, there was an outbreak in Sakai city, Japan – 12 000 cases of infection, mostly primary school kids.  121 developed HUS, 3 died.  Traced to white radish sprouts.

The most serious outbreak was in 2011, Northern Germany. O104 strain however, enteroaggregative plus toxin. 800 cases HUS (90% adults), 53 deaths.  Traced to organic fenugreek sprouts, although Spanish cucumbers blamed initially, exports dropped £120 million per week until consumer confidence returned.

In 2024, there was a UK wide outbreak of O145 related STEC, with 293 cases and 11 HUS cases. There were 2 deaths. Traced back to salad leaves in pre-packed sandwiches, although never proven.

Ratio of unreported human VTEC O157 infection to reports to national surveillance is estimated at 7.4 to 1.

Cows

E coli O157 is now commonly found in cattle, but causes no clinical effect, therefore no incentive for farmers to control. Supershedders are recognized, with one such cow able to contaminate a huge proportion of other animals’ hides.   Current FSA research project underway.  A vaccine has been developed.

Sheep have also been found to carry it…  Likely cause of outbreaks related to “Tough Mudder” events.

Pathology

Lots of acronyms!

  • EHEC = Enterohaemorrhagic E coli
  • O157:H7 = serotype
  • STEC = Shiga toxin producing E coli, same as VTEC (verocytotoxin)
  • STX1/2 genes (same as VTX) code for this toxin.
  • D+/- = Diarrhoeal illness associated (or not)
  • HUS – Haemolytic uraemic syndrome

More than 400 O:H serotypes – 6 account for most HUS.

Other genes are also relevant for virulence eg Intimin (an adherence factor, coded for by eae gene).

Diagnosis

  • Stool culture – should be collected and processed urgently.  State if bloody diarrhoea present and/or suspicion of STEC infection. Routinely tested for the presence of E. coli O157 at local laboratory, which takes 24-48 hours from sample receipt, but will miss non-O157 types. If positive, isolates are referred to SERL for confirmation of identity and typing.
  • PCR stool testing – if stool culture negative, and clearly bloody (or clinical info suggests likely STEC), then Scottish (SNERL) guidance is to send stool for PCR at the Scottish E. coli O157/ STEC Reference Laboratory (SERL), which detects both E. coli O157 and non-O157 STEC.
  • Serum serology – is used for suspected cases where culture/PCR negative.
  • Rectal swabs – may be submitted directly to SERL from cases of HUS who are unable to produce a stool sample.

Do not delay appropriate clinical and public health management while awaiting reference laboratory results.

Regarding laboratory processes:

  • Rapid referral of samples from diagnostic laboratories to SERL is important to improve the probability of culture confirmation.
  • Positive PCR results will be telephoned immediately to the referring diagnostic laboratory and culture results will follow.
  • The local diagnostic laboratory will inform the clinical team and the local public health team of positive PCR and culture results.
  • PCR (stool) may become available to local diagnostic laboratory, removing need for samples to be referred to SERL . However, if a patient presenting with HUS or acute bloody diarrhoea tests negative by local PCR and is causing clinical concern, please discuss referral of stools for further testing with SERL.

Insulin pumps

Clinical, Endocrinology, General paediatrics

Pumps particularly good for recurrent hypoglycaemia, suboptimal control, under 5s, better QOL.

Good for sport.

Parents report loss of control!

Continuous blood glucose monitoring – 6 days, twice daily calibration bloods. Not Prescribable. Some link to pumps, others diagnostic only. Newest pumps can auto-adjust rate but only downwards. Esp normal pre-prandial bloods but Suspected hyperglycaemic between meals.

Ketones cause insulin resistance so higher doses acceptable. Blood ketone over 3 must come to hospital, but no fixed need for IV therapy.

Tight glucose from diagnosis seems to give best results in long term – “metabolic memory”? Prolonged honeymoon phase? Better education?

ADAPT study starting, prevention.

Only 15g snack allowed if on pen eg half apple! Else low carb eg cheese, pepperoni. Pumps allowed snacks of any size, tend to calm down after first couple of weeks!

Digestion v variable. High fat, high carb? Glycemic index ie fibre. Chewing!

Pump can bolus with individual courses! Slow bolus for cinema or buffet meal eg over 30 mins.

Non-waterproof pumps disconnected for bath etc, shower cap applies to cannula site, leave pump running to avoid air space.

 

 

Encephalopathy with Status Epilepticus during Sleep (ESES)

General paediatrics, Neurology

Poorly understood complication of some epilepsies.

Clinically, global regression esp cognitive and behavioural, associated with paroxysmal epileptic activity during sleep, which may of course not be recognized.  The classic example is Benign Rolandic Epilepsy, where continuous spike waves develop druing slow wave sleep (CSWS).  Also seen in Landau Kleffner syndrome.

The problem can be unmasked by anti-epileptic medication, especially carbamazepine.  Appears to be associated with brain pathologies eg polygyria, migrational disorders plus some chromosomal problems eg 8p-.

Diagnosis should be considered when unexpected cognitive impairment (eg memory, temperospatial skills, language) or behaviour changes (eg hyperactivity, aggression, disorientation).  Motor impairments eg ataxia and dystonia have been described.

Investigations

Dramatic increase in EEG abnormalities of any kind during sleep.

Treatment

Traditional plus newer AEDs have been used, but evidence does not point to any one being superior to any other.

Steroids appear to be effective – unclear whether ACTH or hydrocortisone better.

Outcome

Poor prognosis (ie long term neuropsychiatric problems) appear to be associated with longer duration of ESES, plus frontal neuropsychological deficits and frontal EEG anomalies.

[Epilepsy Research and Treatment 2012 http://dx.doi.org/10.1155/2012/642725]

Benign Rolandic Epilepsy (BECTS, or CECTS)

General paediatrics, Neurology

Latest name “Childhood Epilepsy with centrotemporal spikes”. Not always benign…

Common form of childhood epilepsy, with some characteristic features:

  • Seizures usually in sleep
  • If awake, then start with symptoms in one side of the mouth/tongue eg tingling
  • Speech may be impaired, with gurgling noises, as throat muscles affected – child may be trying to indicate something weird is happening to them!
  • Then twitching of face, spreading to face and arm
  • Can have tonic-clonic seizures
  • Characteristic centro temporal EEG abnormalities.  May need to be done sleep-deprived to provoke.
  • Although focal, MRI not indicated if typical features.

Treatment not always required, many children only ever have one or two seizures! Benign, because almost always goes away in puberty viz before age 16.

NICE recommends treatment with carbamazepine or lamotrigine as first-line treatment.

Be aware that carbamazepine and oxcarbazepine may exacerbate or unmask continuous spike and wave during slow sleep, which may occur in some children with benign epilepsy with centrotemporal spikes.

If these are ineffective, alternatives would be levetiracetam, oxcarbazepine or sodium valproate.

Lower limb variants

Clinical, Common, General paediatrics, OPD, Orthopaedic

Beware 2 problems, with additive or compensatory effects e.g. foot and hip!

Rotation probs ie in-toeing and out-toeing.

Metatarsus adductus is common.  Outer edge of foot is curved, and vertical line through heel misses 2nd/3rd toe space.  Rigid forms need serial casting, else 90% improve without treatment by 6-9 months of age.

Internal tibial torsion can be seen by keeping patellae parallel (sitting, or kneeling), and seeing angle of foot.  No treatment required unless severe (tibial rotational osteotomy).

Femoral ante version is more common in girls and often familial.  W posture sitting, patella points in, eggbeater pattern running.  80% resolve spontaneously, else osteotomy (but high rate of complications).

Out toeing normal in first 24 months.  Usually external tibial torsion; occ femoral retroversion.  External tibial torsion associated with patellofemoral instability.  Beware Perthes and SUFE in school age children, esp  unilateral.

Pes Planus – Flexible or rigid?  Arch reforms on tiptoeing?  Rigid suggests congenital vertical talus (rocker bottom heel) or JIA, either way, usually painful.  Beware CP or muscular dystrophy or connective tissue disorder.  Asymptomatic is considered benign.  Insoles may be useful for pain and shoe deformation, do not correct flat foot!

Angular problems ie genu varum/valgum

(Bow/knock)  Gap between knees (intercondylar distance) should be <6cm, gap between ankles (intermalleolar) should be <8cm.  Beware rickets, renal osteodysplasia, tumours, skeletal dysplasias.  Note association between high impact sport and genu varum – cause or selection?  Increased risk of injury/OA in later life…  Differential includes Blounts disease (also associated with obesity).

Knock knees exacerbated by external tibial torsion, ligamentous laxity, obesity.  Less typical pattern of rickets but seen.

[Yeo, BMJ 2015;351:h3394 – videos too]