Henry Goldstein. Schrodinger’s PANDAS, Don't Forget the Bubbles, 2017. Available at:
https://doi.org/10.31440/DFTB.11670
The existence of PANDAS and other Paediatric Acute Onset Neuropsychiatric Syndromes has been controversial for the last two decades.
What is PANDAS?
Paediatric Acute-Onset Neuropsychiatric Disorder Associated with Streptococcal infection, is a concept first mooted by Swedo, an American paediatrician in 1998, as a formé fruste (or incomplete form) of Sydenhams Chorea.
Swedo SE, Leonard HL, Garvey M, et al. Pediatric autoimmune neuropsychiatric disorders associated with streptococcal infections: clinical description of the first 50 cases. Am J Psychiatry. 1998 Feb;155(2):264-71.
She postulated that Group A Beta-Haemolytic Strep infection caused sudden onset explosive OCD symptoms with choreiform movements in prepubertal children. The criteria is: (click for quotes from Swedo’s original wording)
1. Must have OCD or Tic Disorder.
2. Pre-pubertal.
3. Abrupt/Explosive Onset.
4. Association with GABHS infection.
5. Abnormal neuro exam, but NOT chorea.
Think of it as rheumatic fever for the brain; specifically, the basal ganglia.
So, a specific set of patients, with a well-stated mechanism and a clear symptom cluster. That said, this diagnosis and its criteria have been mired in controversy throughout the last twenty years, due predominantly to incongruity, poor external validity and erratic reproducibility in a number of trials.
Given the many billions of words on the topic, there have been some agreed points and areas of controversy. Hence, it is generally agreed that;
- Children with signs and symptoms compatible with Group A H Beta-Haemolytic Strep (GAS) infections should be evaluated for same.
- GAS is one of several factors that can exacerbate OCD or tic disorder in a subset of patients.
- Children with GAS infection and OCD/tic disorder require standard treatments for these problems (regardless of whether GAS and OCD/tic disorder are causally associated)
- There is no indication for routine administration of the following therapies for children who meet PANDAS criteria; prophylactic antibiotics, steroids, plasma exchange, IVIG.
Controversy exists regarding;
- It is unclear whether PANDAS is a discrete neuropsychiatric disorder sufficiently different from OCD/tic disorder to be considered a separate entity.
- The role of GAS as a precipitant of OCD/tic disorders (+/- PANDAS), and whether this is a causal or incidental relationship.
- The etiology of PANDAS as an autoimmune disorder.
- The clinical utility of seeking evidence of GAS infection in children with OCD/tic disorders.
In 2010, the PANDAS criteria was redefined and re-labelled to Paediatric Acute Onset Neuropsychiatric Syndrome (PANS). This widened the age range to <17 years, added restricted food intake as an alternative major symptom, and postulated a much broader aetiology, including a larger number of infectious causes including mycoplasma, HIV, VZV, HSV and the common cold.
By 2013, Singer proposed a move away from the term PANS to Childhood Acute onset Neuropsychiatric Syndrome (CANS), further amending the diagnostic criteria to broaden the causality and focus investigations on the exclusion of treatable causes. When none are found a diagnosis of Idiopathic CANS can be made.
So, what are the challenging parts of this diagnosis?
Firstly, it looks like OCD, and OCD is very common. 1 in 60 kids have an OCD diagnosis
CANS is extremely rare by comparison. Actually, it’s unclear the actual incience, but it is less prevalent than either choreiform disorders or Munchausen’s By Proxy, which are around 1 per 100,000.
Secondly, it’s supposed to be foudroyant, which means literally “strikes as with lightning, sudden and overwhelming in effect, stunning and dazzling.”
This in itself is a challenge to identify, as everything must start somewhere. It is the point of noticeable symptoms that really matters. To extend the metaphor, we don’t actually have random lightning strikes, as realistically the weather changes, a storm rolls in, thunder is heard in the distance and then… bang.
Thirdly, Sydenham chorea, obsessive-compulsive disorder, and tic disorders share common anatomic areas: the basal ganglia of the brain and the related cortical and thalamic sites. Some patients with Sydenham’s chorea display motor and vocal tics, obsessive-compulsive symptoms, and ADHD symptoms, adding support to the possibility that, at least in some instances, these disorders share a common etiology.
Fourthly, a review by Murphy notes a study of PANS-OCD versus Non-PANS OCD which identifies an association between Non-PANS OCD and a Family history of same.
Murphy and colleagues suggest this could be seen as an association between the infectious/autoimmune trigger for PANS-OCD. The countervailing argument is that this is instead representative of tolerability of OCD symptoms within families. That is, by the time the child is functionally impaired and the diagnosis made, it’s not controversial and there is acceptance of the diagnosis within the family.
Fifthly, it’s worth noting that the data for most studies is from the USA, where entry into the health system is very different to Australian, New Zealand or UK, particularly around self-referral to specialists. I’d speculate that it is much more socially acceptable for a child to have a neurologic than a psychiatric condition. Thus, children with OCD with or without tics are might present to a neurologist and be more likely to receive a CANS diagnosis.
Pragmatically for general paediatricians, the diagnostic dilemma here is two-fold;
- How much to investigate?
- How can we treat it?
And this is where Schroedinger comes in. Every time this box is opened, unless we are able to do so in an incisive, foudroyant manner, the child risks becoming medicalised. Indeed, we risk leading them on that journey. By simply looking into the box, we’re disproportionately more likely to find PANDAS than a child with OCD.
I don’t know how to prevent this happening. I do know that the best care for patients with suspected CANS could include general paediatrics, psychiatry, neurology and rheumatology. Ideally, these teams would collaborate to consider diagnosis and management on a case by case basis.
Medical treatments for this constellation of diagnoses have included antibiotics, steroids, immunoglobulins and plasmapheresis. Recent recommendations have highlighted the paucity of evidence for any of these therapies.
Psychiatric treatments include clomipramine, SSRI and Cognitive Behavioural therapy, as for obsessive compulsive disorder; I have not yet seen robust evidence with these interventions specifically for CANS (if you have some, please link in the comments!)
In summary, a child with explosive onset obsessions, compulsions with or without tics might have CANS. Although, on balance, they are more likely to have OCD, considering the prevalence in the population. We need to keep an open mind about these presentations.
Selected references
Swedo SE, Leonard HL, Garvey M, et al. Pediatric autoimmune neuropsychiatric disorders associated with streptococcal infections: clinical description of the first 50 cases. Am J Psychiatry. 1998 Feb;155(2):264-71.
Pichichero, ME. PANDAS: Pediatric autoimmune neuropsychiatric disorder associated with group A streptococci. UpToDate [Online database]. Accessed 22 Feb 2017.
A Paediatric Trainee based in Queensland, Australia, Henry is passionate about Adolescent Medicine & General Paediatrics, with a strong interest in Medical Education & Clinical Teaching. An admitted nerd & ironman with a penchant for Rubik's Cubes & 'Dad jokes'.
@henrygoldstein | + Henry Goldstein | Henry's DFTB posts
Could someone with Munchausen by Proxy Sundrone exploit the inaccurate diagnosis of PANDAS to which it would be hard to refute? We are concerned for a neighbor but bringing this up would be harmful to that prominent family.
It seems to be the perfect disease for MBPS victims. I’ll leave my real email this time since Insee it’s not shared.