Mild Intermittent Hypoxia and Its Multipronged Effect on Sleep Apnea

The dogma over the past 3 decades, particularly in the field of sleep medicine, has been that intermittent hypoxia (IH) is a detrimental stimulus that leads to a number of co‐morbidities including autonomic (e.g. increased sympathetic nervous system activity), cardiovascular (e.g. hypertension, atherosclerosis, arterial fibrillation), cognitive (e.g. loss of gray matter, neural injury and impaired neural function coupled to sleepiness) and metabolic dysfunction (dyslipidemia, hyperglycemia, insulin resistance). This belief was based principally on animal studies that employed protocols that were for the most part severe in nature in regards to length and/or intensity of the hypoxic stimulus. However, the elimination of IH in humans with sleep apnea using continuous positive airway pressure (CPAP) has often been ineffective in mitigating the above mentioned co‐morbidities. The lack of compliance with CPAP, length of treatment with CPAP (i.e. short durations), and the possibility that IH or other hallmarks of sleep apnea are not the primary mechanism response for the listed co‐morbidities, are possible reasons for the absence of improvement in humans.

In contrast to the findings outlined in the previous paragraph, work completed over a similar time frame indicates that some forms of IH may be beneficial in nature. Many studies using a variety of protocols and species, including humans, established that exposure to mild IH initiates sustained increases in the activity of motoneurons, nerves and muscles that contribute to the enhancement of ventilation and the maintenance of upper airway patency. This sustained increase has been termed long‐term facilitation (LTF) and this phenomenon has been the focus of PI's research program for two decades. Long‐term facilitation is the principle form of respiratory plasticity that we documented in healthy humans, and in humans with obstructive sleep apnea (OSA) and spinal cord injury (SCI). The initiation of this phenomenon is mediated by a number of neuromodulators (e.g. serotonin, adenosine, noradrenaline) that trigger components of at least two cellular pathways, deemed the Q and S pathways, which mediate the phenomenon. Besides the initiation of LTF, studies in rats and humans have provided compelling evidence that mild IH might be cardiovascular (e.g. angiogenesis, reductions in blood pressure, reductions in infarct size), neurocognitive (e.g. brain neurogenesis, reduced oxidative stress and inflammation) and metabolically (e.g. decreased cholesterol, decreased low density and very low lipoprotein, increased high density lipoproteins and reduced hyperglycemia) protective. Many reviews over the past decade, including reviews from PI's laboratory have addressed the underlying physiological cellular mechanisms and the translation to whole animals and humans. Briefly, mild IH may lead to moderate increases in reactive oxygen species. These moderate levels of reactive oxygen species activate transcription factors (e.g. hypoxia‐inducible factor 1α, nuclear factor erythroid‐derived 2‐like 2, GATA binding protein 4) that lead to the induction of many cytoprotective proteins. These proteins serve, for example, to reduce oxidative stress (e.g. superoxide dismutase, glutathione, thioredoxin), inflammation (e.g. inducible nitric oxide synthase), apoptosis (e.g. B‐cell lymphoma 2), and promote vasodilation (e.g. heme oxygenase 1) and the formation of blood vessels (e.g. vascular endothelial growth factor). These modifications that ultimately manifest in improved cardiovascular, autonomic and neurocognitive outcomes indicate that beneficial responses can be initiated by IH in a dose dependent fashion without accompanying maladaptive responses. Despite this recognition, the beneficial responses to IH in humans with sleep apnea have not been fully delineated.

Based on the findings outlined in the previous paragraphs the working hypothesis for the present proposal is that exposure to mild IH leads to LTF of upper airway muscle activity that manifests in increased stability of the upper airway, which could ultimately reduce the CPAP required to treat OSA. As previously reported, a reduction in the therapeutic pressure necessary for the maintenance of airway patency leads to improved comfort and ultimately treatment compliance, which is approximately 40 % amongst users. Indeed, the preliminary data shows that following acute exposure to IH during sleep, and repeated daily exposure to IH during wakefulness, the therapeutic pressure required for the maintenance of upper airway patency was significantly reduced during sleep. The reduced therapeutic pressure was also coupled to a reduction in upper airway resistance and the critical closing pressure. These modifications ultimately led to increased CPAP compliance. The numerous co‐morbidities listed in the initial paragraph, which have been linked to hallmarks of sleep apnea (e.g. sleep fragmentation and severe IH), could be significantly improved by increased compliance to CPAP. In addition, as outlined above, IH may directly impact on a variety of co‐morbidities associated with sleep apnea independent of CPAP compliance. Collectively, exposure to IH could impact on comorbidities linked to sleep apnea both directly and via improved therapeutic compliance to CPAP.

Thus, our proposal will determine if mild IH can serve as an adjunct therapy coupled to CPAP to mitigate associated co‐morbidities via its direct effects on a variety of autonomic, cardiovascular, neurocognitive and metabolic measures and indirectly by improving CPAP compliance. Autonomic and cardiovascular modifications will be the primary outcome measures coupled to secondary neurocognitive measures. In addition, metabolic, inflammatory and angiogenic/vasculogenic biomarkers will be measured to indicate the safety and efficacy of exposure to IH.