0570 Apnea-Hypopnea Index is Positively Correlated with Mood Disturbance

Introduction

The prevalence of mood disorders such as depression is higher in individuals with obstructive sleep apnea (OSA). Previous studies have found no significant correlation between the apnea-hypopnea index (AHI) and measures of mood and have only included participants who met diagnostic criteria for OSA. The current analysis sought to determine whether mood correlated with AHI in individuals with any AHI values including those that did not meet diagnostic criteria for OSA.

Methods

31 volunteers were studied (BMI=29.2±1.0 kg/m2, mean±SE), free from medication and without psychiatric illness or chronic medical conditions with the exception of untreated OSA, uncomplicated hypertension (BP<160/100), or obesity. Following 1-3 weeks of an 8h habitual at home sleep schedule, participants completed the POMS-Brief questionnaire (POMS-B) to assess mood after undergoing overnight polysomnography to determine AHI. Total mood disturbance (TMD) scores were calculated by adding the scores on the POMS-B for each mood state subscale and subtracting the score for vigor-activity.

Results

The average AHI was 15.3±3.1 (range of 1.1-74.1) events per hour. The average POMS-B TMD score was 21±1.5 (range of 4-46). There was a significant correlation between the POMS-B TMD score and AHI (p=0.037, r2=0.14). This result was also seen in only those individuals with AHI scores >5 (p=0.002, r2=0.4).

Conclusion

In this sample, individuals with higher AHI values displayed higher TMD scores. These results differ from previous data that showed no significant correlation between AHI and TMD. This is the first analysis to demonstrate a correlation between TMD and AHI while including individuals who didn’t meet diagnostic criteria for OSA. However, the relationship between AHI and TMD was also significant in those with AHI>5. More data on these measures with larger sample sizes and a more equal representation of AHI values should be gathered to provide additional evidence for this relationship.

Support: Support

NIH R01-HL125893; CTSA UL1TR000128, R21HL140377

0035 Resting Metabolism and the Metabolic Response to Exercise Follow Circadian Patterns with Day/Night Differences in Substrate Utilization Between Lean and Obese Adults

Introduction

Resting energy expenditure (EE) follows a circadian rhythm in healthy lean participants, with a nadir in the early morning hours. We determined: (1) whether this pattern persists (or how substrate utilization may change), when challenged with exercise, and; (2) whether obesity affects these responses.

Methods

Fourteen participants (aged 48.5±12.8y; 6-female; 5-obese, BMI 31.9±1.4kg/m2 [avg±SD]) underwent a 5-day inpatient forced desynchrony protocol, comprised of ten 5h 20min ‘days’ in dim-lighting and free of time cues. Resting EE was measured immediately prior to a 15-minute cycle ergometer exercise bout at 50% of estimated heart rate maximum. Substrate utilization was determined from the respiratory quotient (RQ). Circadian phase was calculated using the salivary dim-light melatonin onset (>3pg/mL threshold). EE data were analyzed using a mixed-effect model with group (lean vs. obese) and circadian phase as fixed factors; subject was a random factor. RQ was analyzed using t-tests to determine day/night differences in groups at rest and in response to exercise.

Results

Resting and exercising EE both displayed endogenous circadian rhythms (p<0.05) with nadirs in the early morning (~5:30am), without any differences between groups (p>0.22). Resting RQ was similar between the day and night in the lean group (p=0.66), but decreased (suggesting lower carbohydrate utilization) at night within the obese group (-2.5±1.6%, p=0.02). The lean group increased RQ in response to exercise both during the day (+8.9±2.8%) and night (+8.0±2.8%) (both p<0.001), but there was no increase in RQ in the obese group during either day or night exercise (p>0.16).

Conclusion

These data demonstrate that EE during rest and exercise follows a circadian pattern, with limited influence of obesity. Circadian differences in substrate utilization between lean and obese in the resting state and in response to exercise may play a role in expression and maintenance of unwanted weight gain and impaired metabolic health.

Support

R01HL125893, R01HL140577, KL2TR002370, K01HL146992, F32HL131308, Medical Research Foundation of Oregon, Ford Foundation, and CTSA grant (UL1TR000128)

0847 Association Between Cannabis Use Frequency and Sleep Duration Among a Representative Sample of US Adults

Introduction

Cannabis use is on the rise in the US, and while the cannabis plant and related compounds are considered to have low toxicity, the impact on physiology including sleep remains unclear. Further, the bulk of cannabis research has focused on adolescents and young adults despite growing use among the elderly. Thus, this analysis sought to determine the impact of the frequency of cannabis use on sleep duration and determine if this relationship varies by age.

Methods

This cross-sectional study used data from the 2016-2018 Behavioral Risk Factor Surveillance System surveys. Multinomial logistic regression was used to evaluate the association between the frequency of cannabis use, 0-30 times over the prior 30 days; and average sleep duration, short (<7 hours); recommended (7-9 hours); and long (>9 hours) sleep.

Results

The study sample included 235,667 participants (48% male, 43% 18-44 years old and 35% 44-64 years old) of which 14,122 consumed cannabis. The weighted proportion who reported using cannabis 1-4x, 5-15x, or more than 16x/month were 3.1% (95% CI, 3.0%-3.3%), 2.1% (2.0%-2.2%), and 4.4% (4.2%-4.6%) respectively. In an age stratified model adjusted for sociodemographic and clinical characteristics, there were no associations between sleep and cannabis use up to 15x/month. Among those adults who consumed cannabis more than 16x/month: (1) 18-44 year olds were more likely to report either short sleep (risk ratio (RR), 1.21; 95% CI, 1.05-1.39) or long sleep (RR, 1.57; 95% CI, 1.11-2.22) as opposed to the recommended amount of sleep; whereas (2) 44-64 year olds demonstrated an increased risk for long sleep (RR, 1.77; 95% CI, 1.09-2.89); and (3) participants 65 years and older demonstrated an increased risk for short sleep (RR, 1.61; 95% CI, 1.04-2.49).

Conclusion

We demonstrate that modest cannabis use was not associated with sleep, whereas daily cannabis use is associated with age-dependent sleep duration above and below the recommended amount.

Support

KL2TR002370, 2R25HL105444-09.

0223 Interactions Between Home, Work, and Sleep Among Full-Time Firefighters

Introduction

Firefighters endure large occupational burdens (e.g., heat, exposure to toxic fumes, witnessed trauma) and generally operate under conditions of chronic sleep deficiency due to long shifts plus disrupted sleep and circadian disruption due to emergency calls during the night. A typical shift for firefighters is 24-hours on/48-hours off, and firefighters are expected to use time-off to recover from any sleep debt. However, firefighters need to balance that recovery with social/family needs and home maintenance. We conducted focus groups and qualitative analysis of responses to understand how firefighters’ sleep recoverability is affected by occupational burdens and home/family dynamics.

Methods

Focus groups were conducted via convenience sampling in Portland, Oregon, with full-time firefighters, battalion chiefs, and spouses of firefighters to assess current strategies and coping mechanisms used to manage occupational burdens, home/family obligations, and sleep loss based on their 24-hours on/48-hours off shift schedule. Grounded theory, using NVivo 12 plus, was used to code focus group transcripts to reveal reoccurring concepts that were further grouped into themes.

Results

Major themes that emerged among firefighters and spouses (n=48) centered on spousal resentment of firefighters, driven by understanding a firefighters’ heroic occupation and need to recover from accumulated sleep loss and shift schedule, but also wanting a partner physically/emotionally present to share home/social responsibilities. While married firefighters discussed choosing family/home obligations over reducing sleep debt to maintain social/family relationships, single and divorced firefighters spoke of fewer conflicts impeding their ability to prioritize sleep at home.

Conclusion

This study improves our understanding of how work impacts home life in firefighters and can inform future strategies to address work-family conflict and sleep loss concerns, and highlights the importance of managing expectations of time-off to promote a healthier work-life balance.

Support

Oregon Healthy Workforce Center

Chronic, noninvasive glucocorticoid administration suppresses limbic endocannabinoid signaling in mice

Limbic endocannabinoid signaling is known to be sensitive to chronic stress; however, studies investigating the impact of prolonged exposure to glucocorticoid hormones have been limited by the concurrent exposure to the stress of daily injections. The present study was designed to examine the effects of a noninvasive approach to alter plasma corticosterone (CORT) on the endocannabinoid system. More precisely, we explored the effects of a 4-week exposure to CORT dissolved in the drinking water of mice (100 μg/ml) and measured cannabinoid CB(1) receptor binding, endocannabinoid content, activity of the endocannabinoid degrading enzyme fatty acid amide hydrolase (FAAH), and mRNA expression of both the CB(1) receptor and FAAH in both the hippocampus and amygdala. Our data demonstrate that CORT decreases CB(1) receptor binding site density in both the hippocampus and amygdala and also reduced anandamide (AEA) content and increased FAAH activity within both structures. These changes in both CB(1) receptor binding and FAAH activity were not accompanied by changes in mRNA expression of either the CB(1) receptor or FAAH in either brain region. Interestingly, our CORT delivery regimen significantly increased 2-AG concentrations within the hippocampus, but not the amygdala. Collectively, these data demonstrate that the confounder of injection stress is sufficient to conceal the ability of protracted exposure to glucocorticoids to reduce CB(1) receptor density and augment AEA metabolism within limbic structures.