G
The Sleep Apnea–Autism Connection
Guest blog by Deborah Wardly, MD
I have written guest blogs (Part 1 and Part 2) in the past about the link between obstructive sleep apnea (OSA) and intracranial hypertension (IH). We know that apneas can raise intracranial pressure and that intracranial hypertension can be caused by OSA. I suspect that some of the symptoms of OSA can be explained by increases in intracranial pressure. My most recent paper discusses the idea that it may be the anatomy of the recessed jaw that (outside of respiratory factors which increase intracranial pressure) predisposes these two conditions to go hand in hand. More than likely it is the recessed jaw anatomy that also allows for temporomandibular joint dysfunction to also be present when OSA and intracranial hypertension exist in the same individual.
We have seen increasing rates of most of the chronic illnesses associated with OSA over the last 20 years, to the extent that I have wondered if the human jaw is shrinking more rapidly or has reached a critical point in its shrinkage. Along with adult human chronic disease, we have seen increases in childhood illness, not the least notable of which is an autism spectrum disorder.
Interestingly, it has been noted that there are differences in the faces of autistic children when compared to non-autistic children, and between low and high-functioning autistic children. Scientists investigating this phenomenon do not appear to be aware of how facial structure reflects underlying airway patency, or what this might mean regarding the airway etiology of autism. Autistic children are described to have very prominent sleep problems. These problems are well known to pediatric sleep specialists to reflect underlying sleep-disordered breathing (SDB). For example, 53% of autistic children have difficulty falling asleep, and 34% have frequent awakenings. These are signs indicating that autistic children have insomnia. Dr. Barry Krakow has eloquently demonstrated that chronic complex insomnia in adults is strongly associated with sleep-disordered breathing, and it doesn’t seem likely that the cause in children is very much different.
There are a great many correlations between what is found in autistic children and what is seen in OSA, and there are also many findings in autistic children that could be explained if autistic children have mild intracranial hypertension from birth. For example, leptin, IL-6, and TNFα are elevated in OSA, and in autism. Accelerated head growth in the first year of life and favorable response to substances that decrease brain edema, as is seen in autism, might be explained by intracranial hypertension. Intracranial pressure in autistic children has never been investigated. I have collected the data available prior to 2013 and presented it in my recently published paper: “Autism, sleep-disordered breathing, and intracranial hypertension: the circumstantial evidence.” If each correlation between autism and OSA, and between autism and intracranial hypertension is thought of as a “puzzle piece” in constructing the answer to the etiology of autism, then I have constructed over 90 pieces of the autism puzzle with the hypothesis presented in my paper. The ASD/OSA hypothesis is four-fold and requires that: 1) the mother has SDB during her pregnancy, 2) the infant is born with SDB, 3) both mother and infant have variations of the methylation pathway which are then triggered by the SDB, and 4) the infant is prone to intracranial hypertension.
The idea is that the combination of SDB with the tendency for intracranial pressure to increase leads to a pattern of increased intracranial pressure early on in development which contributes to autism, compounding the effect of repeated low oxygen levels in the mother’s womb due to maternal SDB. It is unlikely to present the same as the typical childhood case of intracranial hypertension because it will vary depending on waxing and waning SDB symptoms. OSA can sometimes cause optic nerve (essentially brain) swelling in the presence of normal intracranial pressures while awake, therefore this process can be very subtle. This hypothesis takes into account most of the findings seen in autism, including the multiple various gene mutations seen between individuals.
I propose that it is not so much these mutations that cause autism, but it is the underlying methylation problems as triggered by OSA/SDB that lead to random mutations of genes, producing the wide variations seen. The ASD/OSA hypothesis may also account for the association of pesticides with autism development, in multiple ways. Some pesticides have been shown to affect the growth and development of the maxilla and mandible, and it has even been noted that the risk of autism from maternal organochlorine exposure during pregnancy is greatest during the 8 weeks immediately after neural tube closure—this is the embryological period when the face is forming. Surely pesticides can be directly neurotoxic, however, if they are found to influence brain swelling then this may add to the brain edema that has already been determined to occur from OSA.
It has also been demonstrated that autistic brains are swollen. If children with autism tend to have recessed jaws that predispose them to not only compression of the airway with OSA, but also compression of their jugular veins preventing easy egress of cerebrospinal fluid (CSF), then this brain swelling becomes more clinically significant and may raise intracranial pressure.
Since the acceptance of my paper for publication, several articles were published which support my hypothesis. In 2013, Shen et al. at the MIND Institute published a study that demonstrated increased extra-axial fluid in infants who later developed autism. They concluded that this suggests an imbalance between CSF production and CSF drainage in these infants. Increased extra-axial fluid has also been seen in children with intracranial hypertension. In intracranial hypertension, the increased pressure is present in this extra-axial space surrounding the brain, pushing in on the brain, as opposed to in hydrocephalus where the increased pressure is present in the ventricles, pushing out on the brain. In December of 2012, Lemonnier et al. published a study demonstrating that bumetanide can be helpful in children with autism, improving autism rating scores and social functioning. Bumetanide is a loop diuretic that has also been used in children with intracranial hypertension, to reduce intracranial pressure. Diuretics are a mainstay of treatment in intracranial hypertension.
There is another piece of data that is more anecdotal at present. It has been reported that people with intracranial hypertension can have photophobia, and phonophobia: increased sensitivity to light and sound. (Dr. Park has also noted in his first book that he sees these characteristics in his SDB patients.) It does not seem to be generally acknowledged however, that most people with intracranial hypertension are significantly sensory defensive. I know this from knowing a great many of them. Almost 180 of them have compiled their symptoms on this spreadsheet.
If one believes this data, then 79% of patients with intracranial hypertension have auditory hypersensitivity, 33% of patients with intracranial hypertension have olfactory hypersensitivities, and 50% of patients with intracranial hypertension have sensitivity to proximity. These are very similar to the prevalence of these different types of sensory disorders among autistic children.
I believe that all of these correlations demand further investigation. It has never been determined that children with autism have normal intracranial pressures, and it has never been determined that the majority of autistic children do not have sleep-disordered breathing. Miano et al. in 2010 stated that it is not possible to conclude how significant OSA might be in causing insomnia in autism because most autistic children do not get sleep studies. Add to this the difficulties encountered in diagnosing mild sleep-disordered breathing at your average sleep lab, and it is likely to take a century before we figure out the answers to these questions. Given that it has been predicted that in ten years 50% of all children born will develop autism, we don’t have too much time.
It has been demonstrated that the degree of symptoms in SDB is inversely proportional to the AHI, therefore I believe that we need to start taking mild SDB very seriously and figure out how to diagnose it outside of the most elite university sleep centers. Given the amount of circumstantial evidence arguing for the ASD/OSA hypothesis, I think that autism researchers must rise to the challenge and rule it out formally before it is dismissed.
If you have a child with autism, does your child show the subtle signs of sleep-disordered breathing? Can you hear his breathing when he sleeps? Does he snore sometimes? Does he wake frequently? Does he sleep with his mouth open and head extended? Is he a restless sleeper? Does he fall asleep during the day? Does he have a small lower jaw (“pixie” face)?
If you are an adult with autism, do you get headaches? Do you hear whooshing sounds in your ears? Do you have visual complaints (symptoms of intracranial hypertension)?
If you are a mother of a child with autism, do you have OSA or symptoms of sleep-disordered breathing? Did you have signs of worsening SDB during your pregnancy?
