Several years ago a number of parents told Andrew (Andy) Zimmerman that their autistic children did better when they had a fever. These anecdotal reports on improvement were common and significant enough for Andy to initiate a study that was finally published in the Journal Pediatrics. In a group of 30 autistic children, parents were told to complete behavioral questionnaires during and after fever episodes of 100.4 degrees or greater. More than 80% of the participants reported that fever was associated with less hyperactivity, less irritability and improved communication. Dominick Purpura suggested, based on this report and what was claimed as numerous letter he had received, that the beneficial effects of fever were due to its effect on an anatomical structure of the brainstem: the locus coeruleus. Animal studies have shown that inactivating the locus coeruleus reduces the febrile response in animals.Thus far microscopic examination of the locus coeruleus in autism spectrum disorder have not revealed any salient abnormalities. Furthermore, the original study and others that followed have been heavily criticized on methodological grounds.
The febrile response is usually a defensive mechanism meant to provide an inhospitable warm environment to invading bacteria and viruses. Fever keeps the total number of these agents at a threshold by limiting their replication. It is the result of a cascade of events involving multiple biochemical pathways of both the immune and central nervous system. In many neurological conditions, fever is related to deterioration or to a decline in function of the nervous system. Indeed, it is well known that fever, especially in the first few days after a cerebrovascular accident, confers stroke a severe prognosis. The initial fever suggests the presence of a large lesion and the possibility of increased mortality. Although controversial, some Stroke Specialty Boards recommend treating body temperature as part of stroke management.
Thus far the evidence seems to indicate that fever, if anything, is a risk factor for autism. Maternal influenza and fever during pregnancy proved to be risk factors for the later development of autism in the CHARGE study (Ousseny Zerbo et al.). In some cases autism seems to develop more often than it should after febrile seizures. Also in certain inherited conditions, e.g., mitochondrial disorders, fever may mark a regression into an autism phenotype.
The deleterious effects of fever on the nervous system have been known for a long time. Its history, effects and mechanisms have advanced our understanding of inflammatory and immune related conditions of the central nervous system. In 1890 Wilhelm Uhthoff described the temporary worsening of vision with exercise in a patient who had an inflammation of his optic nerve. Later on similar worsening of symptoms were described in patients with an autoimmune condition (i.e., multiple sclerosis) when their body temperature was elevated. In effect, infections, hot baths, the warm months of summer and stress, can all lead to the reappearance of symptoms in patients with this condition. In multiple sclerosis as in other conditions this is known as “heat intolerance”. Improvement in these conditions is brought about by lowering the body temperature, e.g., hydrotherapy, ice packs, evaporative body apparel. The underlying mechanism of the Uhtohoff phenomenon is believed to be the slowing of nerve conduction that happens with elevated body temperatures.
After the initial report on fever in autism appeared in the Journal Pediatrics I undertook a personal study asking parents of autistic individuals as to the possible effects of fever on the behavior of their children. Not wedded to any ideas I must say that by far the majority of them only related the expected findings of a high fever, that is, decreased irritability and hyperactivity from a bed-ridden patient. Otherwise the enthusiastic proclamations from some researchers like Dominick Purpura, that “a veil was lifted”, were not reported. The effects of fever in autistic individuals is an interesting phenomenon in need of further studies with adequate number of patients and controls as well as appropriate severity dependent outcome measures that bear on the core symptoms of the condition.
Ousseny Zerbo, Ana-Maria Iosif, Cheryl Walker, Sally Ozonoff, Robin L. Hansen, Irva Hertz-Picciotto. Is Maternal Influenza or Fever During Pregnancy Associated with Autism or Developmental Delays? Results from the CHARGE (CHildhood Autism Risks from Genetics and Environment) Study. 10.1007/s10803-012-1540-x