Friday, 2 May 2014

Autism secondary to enterovirus encephalitis

"We describe a 32-month-old patient with enteroviral encephalitis confirmed by polymerase chain reaction in cerebrospinal fluid, with unfavorable clinical course with marked developmental regression, autistic features, persistent stereotypes and aphasia". So said the report by Filipa Marques and colleagues [1] describing the course and investigations of a previously healthy girl who initially presented at an emergency department with "fever (39°C) and vomiting" and who subsequently deteriorated and over the course of some some months improved although still meeting "criteria for an autistic spectrum disorder".

Unfortunately the paper is not open-access so I can't link to the complete manuscript. I would however like to pick out a few choice details with some discussion to follow. So:

  • Enterovirus encephalitis... as the authors note is "diverse and incompletely understood". Enteroviruses cover quite a bit of area with poliovirus being perhaps the most infamous. The watchwords when it comes to enterovirus encephalitis (which I'll now refer to as EE) are: immune response, cytokine production, "hyperinflammatory syndrome" and onwards the possibility of long-term "neurologic sequelae" including "neurodevelopmental delay, reduced cognitive functioning and learning and behavioral problems".
  • The child described in this report seemed to plunge into a state of developmental arrest and regression over the course of a relatively short period of time incurring a "loss of speech, eye contact and sphincter control" during the first few days of her admission to a hospital in Portugal. Even at discharge on day 31 "she maintained developmental regression with autistic features, self-mutilation and persistent stereotypes". 
  • The medical work-up for this young girl whilst comprehensively was generally unimpressive in terms of results obtained. "Enterovirus PCR was positive" although "No enterovirus serotype was identified on cerebrospinal fluid and stool specimen cultures". The authors speculate that the serotype may have been one that "usually does not circulate in Europe and therefore could not be identified in available kits". Interestingly, authors were also able to rule out anti-NMDAR (see here) and anti-ganglioside antibodies (see here) as being present.
  • Magnetic Resonance Imaging (MRI) and Single-Photon Emission Computed Tomography (SPECT) findings did detect something of a typical pattern of brain findings seen in other cases of EE [2]. I'm not a brain man in respect of being able to expertly describe what the various findings mean but will draw your attention to one sentence: "The progression and distribution of lesions on the MRI scans in our patient seemed to correlate well with clinical severity in the acute phase of the disease".
  • There is a better ending to the story of this girl, as the authors note: "she had a favorable outcome with improvement in all skills and developmental quotient enhancement" 17 months after her hospital discharge.
  • The authors conclude that their report "provides further evidence that autistic symptoms as well as loss of other developmental skills can emerge following an external event such as infection". 

There is quite a bit to discuss from this report bearing in mind that this was a single case study and as things currently stand at the time of writing, the first occasion that EE has been linked to the presentation of autism in the peer-reviewed literature. I suppose I should start with the concept of regression, that is a loss of previously acquired skills, which has appeared more than once on this blog (see here and see here) and will likely appear a few more times still. It happens. And the Marques paper is pretty good evidence of how dramatically it can appear in some cases of autism.

That also regression, including being linked to the emergence of "autistic features", can come about as a consequence of a viral illness is also important. I'm gonna be talking about congenital cytomegalovirus in relation to autism [3] in a future post so providing another example of a possible link between viruses and autism on top of other more classical correlations [4]. Another angle to this potential connection is that of a 'recovery' of sorts when it comes to behavioural presentation (something Chess also noted in the various studies on congenital rubella and autism). One wonders for example, whether some of those optimal outcomers (see here) for example, might also fall into this virus-related autism onset category?

I'll reiterate that this was a case study and hence I'm not implying that every case of autism which follows a period of regression is going to be due to EE or other viral infection. Remember autisms and the N=1. That being said, there is enough detail in papers such as the one from Marques and other peer-reviewed data to suggest that where children do undergo some sort of regression, quite a bit more clinical inspection might be indicated to rule out a viral or bacterial 'cause' to the onset of any physical or behavioural symptoms. Indeed, whether any clinical findings related to viral / bacterial infection in such cases might guide autism research towards further elucidating underlying physiology...


[1] Marques F. et al. Autism Spectrum Disorder Secondary to Enterovirus Encephalitis. J Child Neurology. 2014; 29: 708-714.

[2] Shen WC. et al. MR imaging findings of enteroviral encephaloymelitis: an outbreak in Taiwan. AJNR Am J Neuroradiol. 1999 Nov-Dec;20(10):1889-95.

[3] Sakamoto A. et al. Retrospective diagnosis of congenital cytomegalovirus infection in children with autism spectrum disorder but no other major neurologic deficit. Brain Dev. 2014 Apr 22. pii: S0387-7604(14)00094-1.

[4] Chess S. Follow-up report on autism in congenital rubella. J Autism Child Schizophr. 1977 Mar;7(1):69-81.

---------- Marques F, Brito MJ, Conde M, Pinto M, & Moreira A (2014). Autism Spectrum Disorder Secondary to Enterovirus Encephalitis. Journal of child neurology, 29 (5), 708-714 PMID: 24782421