Obstructive sleep apnea (OSA) is a common condition associated with significant adverse health outcomes. Our overarching hypothesis is that patients with OSA and hypoxia (H-OSA) have greater degrees of insulin resistance in both liver and adipose tissue when compared to those without hypoxia (NH-OSA) thus leading to increased risk for the development of diabetes in the former group.

Obstructive sleep apnea (OSA) is a common condition associated with significant adverse health outcomes. An estimated 25% of men and 10% of women will have OSA during their lifetime. OSA is associated with an increased prevalence of insulin resistance and type 2 diabetes and, with severe degrees of OSA, non-alcoholic fatty liver disease (NAFLD) as well. The mechanisms accounting for the association between insulin resistance and OSA are not fully understood. We have previously demonstrated that experimentally induced sleep restriction in healthy volunteers led to a reduction in whole-body insulin sensitivity and increased rates of lipolysis and gluconeogenesis, accompanied by an increase in stress hormone levels. Studies by others suggest that, in animal models studied under hypoxic conditions, hepatic carbohydrate and lipid homeostasis are perturbed leading to hepatic steatosis and inflammation. Taken together, these observations form the basis of our overarching hypothesis that patients with OSA and hypoxia (H-OSA) have greater degrees of insulin resistance in both liver and adipose tissue when compared to those without hypoxia (NH-OSA), thus leading to increased risk for the development of diabetes in the former group. This hypothesis is based on the supposition that in NH-OSA insulin resistance is primarily triggered by increased levels of stress hormones due to fragmented sleep and this is manifested largely in extra-hepatic tissues (muscle and adipose), whereas in H-OSA there is additional stimulation of hepatic de novo lipogenesis, leading to liver fat accumulation and hepatic insulin resistance. The major goals of this project are to test our hypothesis and determine the impact of standard therapy for this condition, continuous positive airway pressure (CPAP), on insulin sensitivity. This will be achieved by addressing the following two specific aims.

In Aim 1 we will test the hypothesis that, although individuals with OSA have been shown to have insulin resistance in multiple target tissues (adipose, muscle, liver, beta cell), these abnormalities will be significantly greater in patients with OSA that is accompanied by hypoxia (H-OSA,) in comparison to those without hypoxia (NH-OSA). We will compare tissue-specific insulin sensitivity in 30 subjects with H-OSA and 30 with NH-OSA matched for sex, age, BMI, and apnea-hypopnea index. Hepatic and extra-hepatic insulin sensitivity will be measured using orally administered deuterated water stable isotope tracer studies of de novo lipogenesis and gluconeogenesis, both under fasting conditions and during oral glucose tolerance testing (OGTT). Lipolysis will be estimated via free fatty acid concentrations and mathematical modeling. Beta cell function and insulin kinetics will be assessed from insulin and C-peptide concentrations measured during the OGTT. Liver and pancreatic fat will be measured by magnetic resonance and total lean and fat mass by dual-energy X-ray absorptiometry.

In Aim 2 we will test the hypothesis that treatment with continuous positive airway pressure (CPAP) will improve insulin sensitivity in each of the target tissues and that these improvements will be greater in those with a greater number of OSA events per hour associated with hypoxia at baseline. Approximately 12 weeks after initiating CPAP therapy, each participant will undergo a follow-up sleep apnea test and metabolic assessments identical to those described above in Aim 1.

Sixty nondiabetic men and women

Inclusion Criteria:

• Age >19 years

• BMI >18.5 kg/m2

• Participants newly diagnosed obstructive sleep apnea (OSA) must meet the criteria for one of the two following groups:

  • OSA with hypoxia (H-OSA) defined as those with an H-Apnea Hypopnea Index (AHI) ≥15 so as to match the NH-OSA subjects in event frequency and because this is the range defined as more than mild OSA such that we would be likely to see pathology associated with OSA; or,
  • OSA without hypoxia (NH-OSA) defined as having a rate of non-hypoxic respiratory events ≥ 15 per hour (NH-AHI≥15) and having a rate of hypoxic events of less than 5 per hour (H-AHI<5,(52)).

Exclusion Criteria:

• Type 1 or 2 diabetes mellitus currently being treated with medications
• History of chronic obstructive pulmonary disease (COPD) or parenchymal lung disease
• Unstable hypertension
• Treatment for asthma (dependent on type of treatment)
• Current alcohol consumption exceeding 1 drink/day in women and 2 in men
• HIV infection
• Infectious hepatitis
• Pregnancy or lactation within the past six months
• Irregular use of any hypolipidemic agent
• History of surgery for obesity
• Hgb below the lower limit of normal
• Aspartate transaminase (AST) or alanine transaminase (ALT) greater than 3 times the upper limit of normal
• Change in body weight >5% within preceding 3 months (by self-report)

Participants will be recruited from the UCSF Clinical Sleep Medicine Laboratory. Patients seen by UCSF sleep medicine physicians (including Dr. Krystal [Lab Co-Director]) in sleep clinic and those who are referred for polysomnography (PSG) for OSA and for whom the physician has provided approval to be approached, will be contacted by the study coordinator and offered the opportunity to be screened to participate in the study. The study coordinator will review the consent form with the prospective volunteers and will ask for their approval to have their PSG data reviewed by Dr. Krystal, who will determine if the participants meet PSG entry criteria for participation in the study and if found to qualify, to be contacted by phone.