Fever Plus Mitochondrial Disease Could Be Risk Factors for Autistic Regression

24 Sep

Autism, regression, mitochondrial disease and vaccines. With a combination like that, this paper is likely going to be very important.

Fever Plus Mitochondrial Disease Could Be Risk Factors for Autistic Regression

Here is the abstract:

Autistic spectrum disorders encompass etiologically heterogeneous persons, with many genetic causes. A subgroup of these individuals has mitochondrial disease. Because a variety of metabolic disorders, including mitochondrial disease show regression with fever, a retrospective chart review was performed and identified 28 patients who met diagnostic criteria for autistic spectrum disorders and mitochondrial disease. Autistic regression occurred in 60.7% (17 of 28), a statistically significant increase over the general autistic spectrum disorder population (P < .0001). Of the 17 individuals with autistic regression, 70.6% (12 of 17) regressed with fever and 29.4% (5 of 17) regressed without identifiable linkage to fever or vaccinations. None showed regression with vaccination unless a febrile response was present. Although the study is small, a subgroup of patients with mitochondrial disease may be at risk of autistic regression with fever. Although recommended vaccinations schedules are appropriate in mitochondrial disease, fever management appears important for decreasing regression risk.

The authors note neurologic regression in general (not just autistic regression) is observed with patients who have metabolic diseases:

Patients with mitochondrial diseases, like many patients with metabolic diseases, are at increased risk of neurologic regression in conjunction with stressors such as fever, infection, and dehydration.

They studied 28 patients who met DSM-IV criteria for autism and diagnostic criteria for mitochondrial disease.

They define regression and whether it is related to fever thusly:

Autistic regression was defined as loss of developmental skills that included speech, receptive skills, eye contact, and social interests in individuals ❤ years of age. A relationship between fever and autistic regression is defined as regression beginning within 2 weeks of a febrile episode without the suggestion of infectious meningitis or encephalitis.

One comment–the definition of regression is somewhat vague to me. What is also critically important in this discussion is whether there were any signs of autism before the regression. Or, as some may put it, is this regression into autism or autistics undergoing regression? Is there a mix of pathways?

They state that 17 of the 28 patients studied underwent an autistic regression. This is higher than the roughly 25% value for autistic regression they assumed for the general autism population, and statistically significant.

In other words, they are saying that autistic regression may occur more often with kids with mitochondrial diseases.

They note that some of the fevers could be linked to vaccination:

The 17 individuals with autistic regression could be divided into 2 groups, those who regressed with fever (70.6%, 12 of 17) and those who regressed without identifiable linkage to fever or vaccinations (29.4%, 5 of 17).


No individual showed regression with vaccination unless a febrile response was present.

They discuss the concerns with vaccination in the conclusion, noting that vaccination is still recommended for children with mitochondrial diseases. My experience in discussing this issue with mitochondrial disease experts is that they find vaccination to be extremely important. If, for some reason, they decide to not vaccinate a child with mitochondrial disease, they insure that all family members are vaccinated to protect the child.

Children with identified mitochondrial diseases are routinely managed carefully by their physicians with aggressive fever control and hydration. In this context, vaccination of children with mitochondrial diseases is recommended. In our experience, the vast majority of patients with mitochondrial diseases receives a full vaccination schedule according to American Academy of Pediatric guidelines without consequences, particularly when physicians are sensitive to fever control and hydration. In our patients with mitochondrial disease and autistic spectrum disorders, the vaccines did not appear related to the neurologic regression.

I will note again that I feel autistic regression as defined is too vague. Were the patients on the spectrum before the regression? Were they typically developing before the regression?

At least two children were noted to have multiple regressions (a sibling pair). That indicates that at least in some cases, regressions occurred in people already autistic. There just isn’t other information on this.

Another area I would like to see discussed further is on siblings:

Affected siblings were identified in 35.7% (10 of 28).

Affected how? Mitochondrial disease. But, are they also autistic? It would seem not since they included one sibling pair.

This is a big question to me. While the spotlight has been shown on the possibility of mitochondrial disorders being linked to autistic regression, the more general question is more important: could fevers induced by vaccination result in any regression (autistic or otherwise) in people with mitochondrial disorders.

Another question in my mind in this study. Are there patients who underwent regression from non-autistic to autistic) after age 3? According to the Johns Hopkins group, this doesn’t happen. According to them, there is an age window where the regressions could result in autism. This is a very important question in how these patients might fit in to the broader spectrum of autism.

23 Responses to “Fever Plus Mitochondrial Disease Could Be Risk Factors for Autistic Regression”

  1. Emily September 24, 2009 at 22:30 #

    Really really interesting. MELAS, which is terminal, has widespread neurological effects.

  2. daedalus2u September 24, 2009 at 23:40 #

    This is very interesting. I don’t think it is fever per se, I think it is the high NO that usually accompanies fever.

    Normal regulation of mitochondria is via nitric oxide, by at least dozens of different mechanisms. A fever due to an immune system activation is (usually) a high NO state due to NO from iNOS. The high NO is responsible for the flushed skin, and also the hypotension of sepsis (which all by itself can be fatal).

    I think that mitochondrial shut-down in a high NO environment is a normal and absolutely necessary part of mitochondria regulation. That it happens sometimes during fever is (I think) what causes multiple organ failure sometimes observed in sepsis. I have a pretty extensive write-up on it.


    I see the fundamental “cause” of autism-like symptoms as being due to the functional connectivity of the brain. This is quite complex, and in part relates to NO levels.


    The usual technique for measuring what is called “functional connectivity” is via the fMRI BOLD technique, where the different magnetic susceptibility of oxyhemoglobin and deoxyhemoglobin is used to map out the different levels of O2 and how they change real-time in vivo. This increased O2 is caused by vasodilatation, and vasodilatation is caused by NO activating guanylyl cyclase which then generates cGMP which then relaxes the smooth muscle in the brain vasculature and acutely increases the local perfusion.

    People with autism are characterized by reduced brain blood flow, reduced functional connectivity via fMRI BOLD and a number of other symptoms characteristic of low NO (see my blog).

    Mitochondria abnormalities are going to show up (most likely) as reduced effectiveness of mitochondria which would show up as increased superoxide and oxidative stress. This results in fewer mitochondria and a greater sensitivity to high NO induced mitochondria shut-down. All the abnormalities do is increase the sensitivity. Even 100% completely normal mitochondria will exhibit this behavior, it isn’t a bug, it is a feature, a necessary feature to prevent run-away mitochondria from killing the cell.

    MELAS (mitochondrial encephalomyopathy with lactic acidosis and stroke-like episodes)

    There are likely two different types of regression, with two different time scales. The acute shutdown of mitochondria is fast, a few minutes. Some of those mitochondria might recover, but if enough don’t, then the cell they are in will die. There might be some follow on cell mortality as is often seen in strokes, where the final damage is often much worse that what is apparent right after the ischemia is corrected. The killed cells cause downstream cells to undergo apoptosis because the downstream cells have lost the upstream regulation (I think evolution configured physiology to ablate them and reduce cognitive function than risk seizure and death by predation). This type of regression is likely not recoverable from, other than how a stroke is recoverable from. This type of regression is likely very rare and is not usually observed in sepsis.

    The more common type of “regression” that is observed following sepsis is like the “regression” of neurosyphilis. This “regression” is reversible. In neurosyphilis effective clearing of the pathogen and the resolution of the state of neuroinflammation resolves the “regression”. (I am using “regression” in quotes because the cognitive decline observed in neurosyphilis isn’t recognized as “regression”). I think this is also the same as the “brain fog” that sometimes follows chemotherapy, surgery, antibiotics, trauma. I see it as being due to metabolic stress, which “resets” the NO/NOx status of the brain to a lower setpoint.

    The lactic acidosis of MELAS is due to insufficient mitochondria in the liver. It is in the liver that lactate is recycled to glucose by the Cori cycle; this requires mitochondria to consume the reducing equivalents produced by turning the lactate into pyruvate. If those reducing equivalents can’t be consumed by mitochondria, they are converted into lipid. Fatty liver essentially always accompanies lactic acidosis. It happens in Reye’s syndrome, sepsis, and MELAS.

  3. passionlessDrone September 25, 2009 at 00:57 #

    Hello friends –

    @Sullivan – I think your question regarding timeframes of development is very important. Is there something unique about brain morphology, immune system profile, or indeed, NO metabolism at earlier ages that causes regression that fades away later in childhood?

    The authors also go on to note that the nature of this study didn’t allow for precise measurements of fever (a relatively wide range), fluid intake, or other possibly pertinent factors which could have impacts. This is where things get messy, IIRC, Hannah Polling had a history of ear infections (and I believe ‘febrile events’) previous to her acute regression; teasing out the other components that might be involved is a daunting task.

    I also worry just a little bit about assigning a focus on drastic regressions in relationship to mitochondrial disorders. For example, the authors say:

    A relationship between fever and autistic regression is defined as regression beginning within 2 weeks of a febrile episode without the suggestion of infectious meningitis or encephalitis.

    Luke also had a 105 fever and trip to the ER, but this was much longer than 2 weeks previous to our noticing a loss of skills, and in any case, it was a gradual loss of skills.
    Luke has hypotonia, elevated ammonia and high lactic acid, but has never been officially diagnosed with mitochondrial disorder.

    Designing a study to analyzie timeframes like this would be difficult, or impossible, but the more I read, the more I become convinced that tools without the power to make subtle obervations aren’t very good at modeling complicated systems. The referenced study is at the least, a good start, however. Hooray for science.

    @Daedulus2u – I swear I wish I was clever enough to understand half of what is on your blog. Maybe I’ll try again soon.

    – pD

  4. dr treg September 25, 2009 at 01:07 #

    Perhaps a simpler model of the functional connectivity of the brain relates to dendrite firing rates which is proportional to the density of dendritic spines on the dendrites. Reduction in dendritic spine density due to inflammation (dendritis)results in self-obsession/ obession-compulsion and the feel-bad feelings of fear, self-pity, resentment and dishonesty.
    It is noted that dendrites have mitochondria and dendritic mitochondrial content/activity affect dendrite spine growth and the synapses upon them.

    BOLD signals do not directly measure neuronal activity.
    “BOLD signals are most strongly associated with the input to a given area rather than with the output. It is therefore possible (although unlikely) that a BOLD signal could be present in a given area even if there is no single unit activity.”

    The fever is part of the immunogenic response to an immunogen e.g. infection or immunization, which involves many other mediators as well as NO. The immunogenic response causing reduced CBF in autism involves many other mediators as well as NO.

    Fever can also improve aberrant behavior in children with autism.

  5. daedalus2u September 25, 2009 at 01:20 #

    Hannah Poling had neuropathy. The “regression” I am talking about (and what I think the authors of this study are talking about) is not associated with neuropathy. Autism is not associated with neuropathy. Neuropathy may be associated with “autism-like” symptoms, but that does not make it autism.

    As an extreme example, Terry Schaivo had neuropathy and “regressed” in that she lost language. She did not have autism. She had an acute “regression” due to neuropathy. This is not the kind of regression that is talked about in the context of autism.

    High ammonia is associated with insufficient mitochondria in the liver where ammonia is converted into urea using ATP. There is limited capacity to do that, which is why a pure protein diet is fatal, normal physiology doesn’t have the capacity to convert a day’s worth ammonia due to getting all your calories from protein into urea to get rid of it.

    To me, lactate and ammonia indicate insufficient liver capacity due to not enough NO. I think the ammonia is a feedback mechanism when it spills out on the skin to nourish the bacteria I am working with who then convert it to NO/NOx.

  6. dr treg September 25, 2009 at 08:36 #

    One could further examine your NO/cGMP theory by having a trial of Viagra in children with autism. Viagra increases cGMP and results in increased blood flow.
    Viagra is used to treat children with pulmonary hypertension and using Viagra for reasons other than erectile dysfunction is not unheard of.

  7. Natasa September 25, 2009 at 10:42 #

    No, it is not true that autistic regression does not happen after the age of 3.

    The latest documented, published case study, of regression into autism in a previously neurotypical person age 34, following herpes encephalopathy.

    There are several other documented cases of herpes encephalopathy in older individuals, ie well past 3 years of age, with autism as sequelae.

    In addition to herpes viruse, HIV infection is known to cause autistic regression, and literature is available on case studies of children as old as 6 regressing into autism following viral reactivation and dissemination through CNS.



  8. Natasa September 25, 2009 at 10:58 #


    “autism-like” symptoms do indeed make autism. Autism is defined and diagnosed SOLELY according to outism-like symptoms.

    There is nothing else to autism apart from autism like symptoms.

    Autism is never ever diagnosed or defined acorrding to underlying pathology/etimology. If a person displays autism-like symptoms, he or she has autism.

  9. Natasa September 25, 2009 at 11:06 #

    Prior/underlying mitochondrial dysfunction predisposes mice to encephalitis by a virus that does not neurotropic under normal circumstances…

    Impaired long-chain fatty acid metabolism in mitochondria causes brain vascular invasion by a non-neurotropic epidemic influenza a virus in the newborn/suckling period : Implications for influenza-associated encephalopathy
    Auteur(s) / Author(s)
    DENGFU YAO ; KUWAJIMA Masamichi ; YE CHEN ; SHIOTA Mayumi ; OKUMURA Yuushi ; YAMADA Hiroshi ; KIDO Hiroshi ;
    Résumé / Abstract
    The neuropathogenesis of influenza-associated encephalopathy in children and Reye’s syndrome remains unclear. A surveillance effort conducted during 2000-2003 in South-West Japan reveals that almost all fatal and handicapped influenza-associated encephalopathy patients exhibit a disorder of mitochondrial ?-oxidation with elevated serum acylcarnitine ratios (C16:0+C18:1)/C2. Here we show invasion by a non-neurotropic epidemic influenza A H3N2 virus in cerebral capillaries with progressive brain edema after intranasal infection of mice having impaired mitochondrial ?-oxidation congenitally or posteriorly in the newborn/ suckling periods. Mice genetically lacking of carnitine transporter OCTN2, resulting in camitine deficiency and impaired ?-oxidation, exhibited significant higher virus-genome numbers in the brain, accumulation of virus antigen exclusively in the cerebral capillaries and increased brain vascular permeability compared to in wild type mice. Mini-plasmin, which proteolytically potentiates influenza virus multiplication in vivo and destroys the blood-brain barrier, accumulated with virus antigen in the brain capillaries of OCTN2-deficient mice but only a little in wild-type mice. These results suggest that the impaired mitochondrial /3-oxidation changes the susceptibility to a non-neurotropic influenza A virus as to multiplication in the brain capillaries and to cause brain edema. These pathological findings in the brain of mice having impaired mitochondrial ?-oxidation after influenza virus infection may have implications for human influenza-associated encephalopathy.
    Revue / Journal Title
    Molecular and cellular biochemistry ISSN 0300-8177
    Source / Source
    2007, vol. 299, no1-2, pp. 85-92 [8 page(s) (article)]

    It is known that viral encepalopathy can lead to autism (with fever being the symptom of the acute phase infection?). It seems to be the case of underlying mito dysfunction predisposing one to viral neurotropism/encephalopathy….

  10. dr treg September 25, 2009 at 14:21 #

    “She had an acute “regression” due to neuropathy.”
    “Neuropathy” generally relates to inflammation of the cranial or peripheral nerves.
    Autism can be associated with neuropathy but neuropathy doesnt cause regression.

    “It is known that viral encepalopathy can lead to autism (with fever being the symptom of the acute phase infection?)”
    Autism relates to inflammation of the brain but results in different pathological changes compared to “meningitis” or “encephalitis”. (Johns Hopkins)
    The virus causes inflammation resulting in autism rather than encephalitis causing autism.

    Perhaps it is time to divide autism into sub-groups i.e.
    Primary – unknown etiology.
    Secondary – known etiology.
    Transient – self-limiting.

  11. Emily September 25, 2009 at 16:08 #

    pD–Have you considered having him tested for mitochondrial disorder? I think that they’re onto something with the mitochondrial disease as a subgroup. It makes so much sense. Can’t explain everything by any means, but the etiologies are going to turn out to be so multifactorial, “we” need to stop seeking some monolithic explanation, anyway.

    You’re right about the tools for analyzing these complex systems. The best place to start is one system at a time, I think, then incorporating them in aggregate, step by step. Takes a long time, is tedious, can be confusing, but it needs to be done.

  12. Craig Willoughby September 25, 2009 at 19:54 #

    Very interesting indeed. I agree, Sullivan, the determination of onset by age is an important thing to consider. Sadly, autism is not defined by neurological or biological diagnoses in most cases; it is defined by observing behavior. Maybe this study will help us predetermine susceptibility in children prior to vaccination.

    Thanks for this post, Sullivan; I thought it was very informative.

    • Sullivan September 25, 2009 at 20:07 #

      Craig Willoughby,

      I very much appreciate you taking the time to comment here. The whole area of mitochondrial medicine (not just autism) is amazing to me.

  13. Dawn September 25, 2009 at 20:17 #

    Very interesting, Sullivan. Autism is more and more showing itself to be a many-headed hydra, with a myriad of causes (and none of them appear to be either mercury poisoning OR the MMR.)

    Very small nit to pick: the hospital name is Johns Hopkins (no apostrophe in John’s) so the sentence should read “…Johns Hopkins’ group,…”.

    I always wondered why his parents named him Johns so looked it up; Wiki says it was a family name, one of those “mother’s last name as the child’s first name” things. The southerners of the time were big on that (and yeah, I can say that because we have a lot of those in our family tree, too, my southern is rather diluted but still there. Just go back a few generations…).

    I don’t usually go grammer nazi on people but in this case, I feel the facility name needs to be accurate. (And don’t get me started on your and you’re, its and it’s, too, to and two, lie and lay – my mother beat all these things into my head.)

    • Sullivan September 25, 2009 at 20:21 #


      you can correct my spelling or grammar anytime!

      I think that every generation of autism parents rediscovers that autism is plural.

  14. Joseph September 25, 2009 at 20:45 #

    I’d say it’s potentially interesting, but limited. In order to tell that something is a risk factor, you need a case-control study.

    More specifically, can self-selection bias be discounted in this study?

  15. Dawn September 25, 2009 at 21:55 #

    @ Sullivan…if I say I love you, will you take it the right way? 🙂

  16. Laura September 30, 2009 at 03:50 #

    Just curious if I know Dawn and Sullivan? Are you guys in Georgia? If so, I am Anslie’s mom, you should know who I am, and Jennifer was our baby sitter!

  17. mark cooper October 19, 2009 at 02:04 #

    this site is unbelievable.
    the truth is coming to expose the shills & corporate funded websites that defend what will be seen as the biggest iatrogenic harm inflicted on humanity.
    if medicine refuses to see that the emperor has no clothes, & continues to support the ‘tobacco’ science which allows this crime against humanity to carry on, then it seals it’s own fate.
    i call on scientists with integrity to come forward & remember your humanity.
    & i have no interest in comments about my spelling & grammar. i concern myself with the truth in whatever form it takes.

    • Sullivan October 19, 2009 at 02:27 #

      mark cooper,

      you are welcome to discuss your views on this site. One way to start is to discuss something in the blog post or comments. Right now you appear to be a hit-and-run commenter. I look forward to being proven wrong.


  1. Fever Plus Mitochondrial Disease Could Be Risk Factors for Autistic … | Autistic Information - September 2, 2012

    […] of low NO (see my blog). Mitochondria abnormalities are going … … Read more: Fever Plus Mitochondrial Disease Could Be Risk Factors for Autistic … ← Medical Tags – Autism – Modern Caduceus Symbol | Autism … My Autism Team […]

  2. Vaccines and Encephalitis – VAXOPEDIA - September 27, 2017

    […] Fever Plus Mitochondrial Disease Could Be Risk Factors for Autistic Regression […]

  3. What is Mitochondrial Autism? – VAXOPEDIA - October 8, 2018

    […] Fever Plus Mitochondrial Disease Could Be Risk Factors for Autistic Regression […]

Leave a Reply to Natasa Cancel reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: