The purpose of this study is to examine the influence of sleep effectiveness on glucose and insulin metabolism in health and disease (prediabetes and type two diabetes).

We will monitor sleep effectiveness using the sleep spectrogram, obtain serial nocturnal blood glucose and insulin measurements, and assess the impact of pharmacologic enhancement [using eszopiclone (Lunesta), a medication that promotes stable sleep)] on glucose and insulin homeostasis.

We hypothesize that 1: Effective sleep is associated with enhanced insulin sensitivity, relative to ineffective sleep states, and 2: Enhancing sleep effectiveness using eszopiclone (Lunesta) improves 24-hour glucose metabolism in prediabetics and diabetics compared to baseline.

Evidence from experimental studies supports the hypothesis that fragmented or insufficient sleep contributes to impaired glucose and insulin homeostasis. The sleep spectrogram, an EEG-independent measure of sleep effectiveness, maps coupled oscillations of heart rate variability and ECG-derived respiration. In a sample of non-diabetic subjects with and without sleep apnea, we previously explored the association between ECG-spectrogram derived biomarkers and glucose metabolism and found that the marker of effective sleep, High Frequency Coupling (HFC), is associated with reduced diabetes risk (increased Disposition Index). HFC is also enhanced by sedative medications (unpublished data). In this study we will 1.) explore the relationship between sleep effectiveness and insulin sensitivity across the sleep period, by frequently sampling glucose and insulin during nocturnal polysomnography in healthy and prediabetic subjects; and 2.) evaluate the impact of pharmacologic enhancement of effective sleep with nightly eszopiclone (1 week, home environment) on glycemic profiles (continuous glucose monitoring, 72 hrs) in prediabetics and diabetics compared to pretreatment baseline. We expect that desirable glycemic profiles will correlate with the spectrographic marker of effective sleep while undesirable glucose profiles will correlate with the marker of ineffective sleep. Using pharmacologic enhancement of effective sleep, we expect to demonstrate improvement in glycemic profiles in prediabetic and diabetic subjects compared to pre-treatment baseline.

Inclusion Criteria:

• Healthy volunteers, men and women 18-64 years of age.
• Fluent English speakers.
• Health status as per criteria listed for prediabetes and diabetes (based on 2003 American Diabetes Association criteria and 2009 International Expert Committee Report: Prediabetics will have impaired glucose tolerance with fasting plasma glucose (FPG) 100-125 mg/dL, Hemoglobin A1C 5.7-6.4%, or 2-hour plasma glucose (PG) 140-199 mg/dL after 75-g oral glucose tolerance test (OGTT). Diabetics will have FPG ≥ 126 mg/dL, Hemoglobin A1C ≥ 6.5%, or 2-hour PG ≥ 200 mg/dL on OGTT.

Exclusion Criteria:

• Primary psychiatric disease or conditions which may independently contribute to sleep fragmentation or may hinder the subject's ability to complete the proposed testing:
• Respiratory, liver, or clotting disorders
• History of sleep disordered breathing, Restless legs syndrome or Periodic limb movement disorder or high clinical suspicion of sleep disordered breathing or other sleep disorder (e.g., snoring, excessive daytime sleepiness, frequent napping, excessive motor activity)
• Shift worker or circadian phase disorder
• Abnormal resting ECG, pacemaker, atrial fibrillation or other arrhythmia
• Seizure disorder
• History of depression, bipolar disorder, anxiety disorder, schizophrenia or use of psychiatric medication
• Narcolepsy
• Tobacco or recreational drug use
• Pregnancy or lactation
• Regular use of stimulants or hypnotic medication
• Evidence of sleep apnea (Apnea-Hypopnea Index > 10 on screening sleep study)