Undergraduate Honors Theses

Thesis Defended

Spring 2011

Document Type



Integrative Physiology

First Advisor

Kenneth P. Wright Jr., PhD


Introduction: The metabolic syndrome is reaching epidemic proportions worldwide. Sleep duration has been shown to be important in glucose homeostasis. It has been reported that the specific stages of sleep (most importantly slow wave sleep and rapid eye movement sleep) each play their part in glucose homeostasis. The aim of this thesis was to determine if partial sleep deprivation has an impact on insulin sensitivity. Specifically, we sought to determine if 3 nights of 5h sleep opportunity when compared to 3 nights of 9h sleep opportunity with food being ad libitum, would result in decreased insulin sensitivity. Methods: 13 healthy subjects (7 men, 6 women) aged 24.77 ± 4.02 (Mean ± SD), participated in the study. Subjects were admitted to the Clinical Translational Research Center (CTRC) at the University of Colorado Health Sciences Center for the duration of the study. Subjects lived in a hospital room for 14-15 days, depending on which sleep deprivation protocol they were randomized to. Total sleep time (TST) and sleep efficiency (SE) were assessed via actigraphy daily. Plasma glucose and insulin levels were obtained from oral glucose tolerance tests (OGTTs), assessed in the morning following a 9h baseline night, 3 days of 9h sleep per night, and 3 days of 5h sleep per night. Insulin sensitivity was assessed using the Matsuda index and HOMA-IR. Mixed model ANOVA with subject as a random factor and day, time and/or condition as fixed factors was used to analyze data. Results: Main effects of day (p<0.000001) and of subject (p<0.000001) were seen for TST. Planned comparisons showed significantly less TST for each day of the 5h sleep opportunity condition, as compared to baseline and each respective 9h day sleep opportunity. Main effects of day (p<0.001) and of subject (p<0.000001) were also seen for SE. Planned comparisons showed significantly higher SE for each day of the 5h sleep opportunity condition versus the 9h day sleep opportunity. Significant effects were seen for subject (p<0.00085), time (p<0.00), and significant condition × subject (0.012173) and time × subject (p<0.00) interactions for glucose levels. Significant effects were seen for condition (p<0.015392), time (p<0.00), and significant condition × subject (p<0.050201) and time × subject (p<0.000001) interactions for insulin levels. Significant main effects were also seen for condition (p<0.05) and for subject (p<0.05) for the Matsuda index. Planned comparisons revealed a significant reduction in the Matsuda index for the 5h condition versus baseline (p=0.01). Planned comparisons showed a significant increase in HOMA-IR for the 5h condition vs. baseline (p<0.05). Discussion: A sleep opportunity of 5h for 3 nights resulted in a reduction in insulin sensitivity when compared to a baseline 9h sleep opportunity. An increase in insulin levels following the sleep restriction condition was associated with normal glucose levels. These results indicate a reduction in insulin sensitivity following sleep restriction of 5h. The study’s results on insulin sensitivity are similar to those seen in obese and aging individuals and provide important implications into the effects of sleep restriction on metabolic health. Support: NIH/NHLBI R01 HL085705, with the support of UCB and UCH CTRC: physicians, nurses, dieticians, biostasticians, research advocates, informatics core staff, administrative staff, and Chronobiology Laboratory research equipment.