Today, the NSRR team is taking a moment to discuss the latest sleep research with Matt Butler. Thank you Matt for providing us with insight into your research!
Matt Butler: Circadian disruptions and sleep disorders alter temporal patterns of physiology and behavior, and they increase the risk for cardiovascular disease, metabolic diseases including diabetes, and mood disorders. At the Oregon Institute of Occupational Health Science, my laboratory's goals are to determine the mechanisms by which this occurs through controlled physiology experiments (basic and clinical) as well as through prospective analyses in long term human sleep study cohorts.
Matt Butler: As a postdoctoral fellow in Steven Shea's laboratory at the Brigham and Women's Hospital, I found a circadian rhythm in the duration of apneas and hypopneas. This largely explains the common observation that respiratory events get longer through the night. This led naturally to the question of whether long or short apneas are bad for health. We approached Dr. Redline about answering this, and by chance, this coincided closely with the launch of the National Sleep Research Resource (NSRR). I became an NSRR early adopter: I provided feedback on my experiences with the NSRR and the Sleep Heart Health Study data as I worked on the preliminary analyses that led to this grant award from the American Sleep Medicine Foundation.
Matt Butler: Obstructive sleep apnea (OSA) is a common sleep disorder which increases the risks for cardiovascular disease. Currently, it is diagnosed by the mean number of events per hour of sleep (apnea-hypopnea index, AHI). This metric fails to capture the wide variability in the severity and timing of events, and importantly, it is a poor predictor of risk in women. We used the NSRR to examine the overnight sequence of events and the health outcomes for subjects in the Sleep Heart Health Study, and we found that event duration predicts mortality (Table 1). We are now testing whether event duration can outperform AHI in predicting risk, especially in women. Concurrently, we are investigating the information provided by the clustering of events in time, as the intervals between events can vary dramatically, even for a particular AHI (Figure 1).
|Event duration quartiles (sec)||Hazard Ratio||Confidence Interval||p value|
|24.0 - 57.6||1.0||Ref.|
|20.6 - 24.0||0.98||0.79 - 1.09||n.s.|
|17.7 - 20.6||1.17||1.0 - 1.36||p<0.05|
|11.1 - 17.7||1.26||1.07 - 1.48||p<0.01|
Table 1: Mortality hazard ratios by event duration quartiles, adjusted for age, sex, BMI, race, smoking, and AHI.
Figure 1. Decile plot (cross hatches show 10% intervals) of the inter-event interval (IEI) distribution for 8 subjects with AHI of 40-41. The IEI coefficient of variation (IEI-CV) varied from 0.4 to 3.6. The two left subjects have a maximum IEI of ~2 min, whereas the two right subjects have a maximum IEI of >10 min and ~25% of all intervals are >2min.
The American Sleep Medicine Foundation
2014 Focused Projects Award
Sex differences in sleep apnea and novel sleep measures to predict cardiovascular risk
Matthew P. Butler (PI)
Oregon Institute of Occupational Health Sciences and Dept. of Behavioral Neuroscience, Oregon Health & Science University, Portland, OR 97239