Autism is a neurodevelopmental disorder currently affecting as many as 1 out of 150 individuals in the United States . Autism is characterized by impairments in social interaction, difficulty with communication, and restrictive and repetitive behaviors . Autism traditionally is considered a "static" neurological disorder  and improvements in core autistic features are not common [4, 5]. Furthermore, three rigorously performed epidemiological studies demonstrate that the prevalence of autism has increased in recent years [6–8]. These facts might explain why parents of children with autism are more likely to seek alternative and off-label medical therapies than parents of children in the general population . One off-label therapy that has recently increased in popularity as a treatment for autism is hyperbaric oxygen therapy (HBOT) [10, 11]. Traditionally, HBOT involves inhaling up to 100% oxygen at a pressure greater than one atmosphere (atm) in a pressurized chamber . Most typical indications for HBOT involve the use of hyperbaric pressures above 2.0 atm. Higher atmospheric pressures are generally required to treat conditions such as carbon monoxide poisoning and to improve wound healing [12, 13].
In some studies, the use of oxygen appears to enhance neurological function. For instance, in a double-blind, placebo-controlled, cross-over study, oxygen administration in healthy young adults, when compared to room air, was demonstrated to enhance cognitive performance, including improved performance on attention, reaction times, and word recall . Additionally, in elderly patients, HBOT at 2.5 atm and 100% oxygen, when compared to a control group, improved cognitive function, including memory . Because of these outcomes, some investigators have used HBOT to treat certain neurological disorders, including chronic and traumatic brain injury [16–22], as well as fetal alcohol syndrome , and clinical improvements in these patients have been observed. Furthermore, in a recent rat model of traumatic brain injury, treatment with HBOT at 1.5 atm and 100% oxygen, when compared to a sham-treated normobaric air group, improved spatial learning and memory . Several studies, using HBOT at similar pressures, also demonstrated clinical improvements in some patients with cerebral palsy (CP) [25–28] that in some cases was dramatic ; however, some researchers have questioned the results of these studies and have called for further controlled trials and a focus on defining the mechanism of action of HBOT in individuals with CP . It is important to note that some of these studies [16, 21–24, 26] used lower hyperbaric pressures (1.5 atm or less) than the pressures typically used for most clinical indications . Given this background, some physicians have also applied similar lower hyperbaric pressures of 1.3 to 1.5 atm in autistic individuals, with oxygen concentrations ranging from 21% to 100% [10, 31].
HBOT for children is generally regarded as safe, even at pressures of 2.0 atm for 2 hours per day . However, to our knowledge, the safety of HBOT for autistic children has not been previously studied; a review of MEDLINE indicates that there are no prospective studies on the use of HBOT for autism. Yet, there are anecdotal reports of clinical improvements in autistic children with hyperbaric therapy that have been reported by some physicians. For instance, Heuser et al. treated a four year old child with autism using hyperbaric therapy at 1.3 atm and 24% oxygen and reported "striking improvement in behavior including memory and cognitive functions" after only ten sessions. This child also had marked improvement of cerebral hypoperfusion as measured by pre-hyperbaric and post-hyperbaric Single Photon Emission Computed Tomography (SPECT) scans . Another case series suggested that hyperbaric therapy at 1.3 atm led to clinical improvements in six autistic children .
Review of the pathophysiology found in some autistic individuals in conjunction with the mechanisms of action of HBOT lead to the speculation that HBOT might produce clinical improvements in autistic individuals . Several studies indicate that some autistic individuals manifest cerebral hypoperfusion [33–35], neuroinflammation [36–38], and gastrointestinal inflammation [39, 40]. HBOT might ameliorate some of these problems by improving cerebral hypoperfusion [17, 21, 31, 41], and by decreasing neuroinflammation and gastrointestinal inflammation [42–47]. However, no prospective studies have examined the role of HBOT on inflammation and cerebral hypoperfusion in autistic individuals.
Furthermore, concerns exist that HBOT might increase oxidative stress via the production of reactive oxygen species . These concerns are especially relevant because some children with autism express evidence of increased oxidative stress  including lower serum glutathione levels [50, 51], and decreased activities of antioxidant enzymes including superoxide dismutase (SOD) , glutathione peroxidase , catalase , and paraoxonase, an enzyme that prevents lipid oxidation and also inactivates organophosphate toxins in humans . Some autistic children also demonstrate evidence of increased lipid peroxidation [53, 55, 56]; this includes increased malondialdehyde which is a marker of oxidative stress and lipid peroxidation . A review of the literature indicates that oxidative stress can occur with HBOT but appears to be less of a concern at hyperbaric pressures under 2.0 atm . In fact, with long-term and repeated administration, HBOT below 2.0 atm can actually decrease oxidative stress [59–61] by reducing lipid peroxidation , and by up-regulating the activity of antioxidant enzymes including SOD [60, 63], glutathione peroxidase , catalase , and paraoxonase [62, 66]. Furthermore, at the pressures examined in this current study (1.3 to 1.5 atm), a search of the literature failed to identify any studies indicating that oxidative stress worsened with HBOT.
Alternatively, some evidence suggests that HBOT could actually alleviate oxidative stress in children with autism. For example, halving oxygen concentrations in normal healthy volunteers results in relative hypoxia and actually increases oxidative stress . There are several studies that demonstrate evidence of cerebral hypoxia, as measured by a reduction in brain Bcl-2 and an increase in brain p53, among some autistic individuals [68–71]. Elevated p53 is induced by hypoxia  and a decrease in Bcl-2 is associated with increased apoptosis provoked by hypoxia . Therefore, in theory, improving hypoxic areas in the autistic brain might decrease oxidative stress. However, the effects of HBOT on oxidative stress in autistic individuals are unknown. To our knowledge, there have been no studies performed which examine the role of HBOT on oxidative stress in autistic children.
This present study examined hyperbaric therapy at the low and the high ends of the ranges of atmospheric pressures and oxygen concentrations currently employed in individuals with autism: 1.3 atm and 24% oxygen , and 1.5 atm and 100% oxygen. This study had several objectives. First, since increased oxidative stress is found in some autistic children, the effects of HBOT on oxidative stress markers before and after 40 hyperbaric treatments were measured. Second, evidence of increased inflammation is found in many autistic individuals. HBOT is also known to have anti-inflammatory effects; therefore, the impact of HBOT on an inflammatory marker (C-reactive protein) was measured. Third, since the efficacy of HBOT in autism has not been previously evaluated, this current open-label pilot study (without a placebo-control group) examined the changes in clinical symptoms, as rated by parents or caregivers, after treatment with HBOT. Finally, the safety of HBOT, used at 1.3 and 1.5 atm, was evaluated in autistic children.