To Sleep, Perchance to Dream

Less REM sleep was associated with higher all-cause mortality, a population-based cross-sectional study showed.

Researchers used data from two cohorts:

• The Outcomes of Sleep Disorders in Older Men (MrOS) Sleep Study of 2,675 participants (100% men, mean age 76.3, median 12.1 year follow-up).
• The Wisconsin Sleep Cohort (WSC) if 1,386 participants, (54.3% men, mean age 51.5, median 20.8 year follow-up).

The MrOS group had a 13% higher mortality rate for every 5% reduction in REM sleep (age-adjusted HR 1.12; fully adjusted HR 1.13, 95% CI 1.08-1.19).

Findings were replicated in the WSC cohort, with younger age, inclusion of women, and longer follow-up (HR 1.13, 95% CI 1.08-1.19).

The risk was independent of known risk of cardiovascular and cerebrovascular disease.

“Survival analysis of older, community-based men found an association between less REM sleep and increased mortality, which replicated in an independent data set of middle-aged men and women,” wrote Eileen Leary, PhD, RPSGT, of Stanford University and coauthors in JAMA Neurology.

“Similar effect sizes (HR ranging from 1.13- 1.19 per 5% REM sleep decrease) were observed in MrOS for all-cause, cardiovascular, cancer, and other mortality after adjusting for confounding demographic, sleep, and health related covariates,” they added. “These effect sizes are slightly larger than mortality risk resulting from aging 1 year (HR for age ranging from 1.11-1.16) based on MrOS data.”

In an accompanying editorial, Michael S. Jaffee, MD, of the University of Florida, Gainesville, and coauthors wrote, “Although association does not imply causation, the finding that a lower proportion of REM sleep was associated with a higher all-cause mortality should interest neurologists, because it raises the question of whether REM sleep in an individual patient could serve as a biomarker for general health.”

These findings “demonstrate definitively that the association between sleep and mortality extends beyond the simple measure of total sleep time,” they continued.

“Although it seems likely that REM sleep quantity is a biomarker of general health or REM sleep is decreased by disease processes that also increase mortality, rather than having a direct causal link to mortality, neurologists should remain mindful of the common clinical scenarios in which REM sleep is disturbed and follow the emerging work on sleep stage–targeted interventions closely,” they added.

While the functions of REM sleep are not fully understood, some mechanisms and roles are known. REM sleep density and melatonin both decrease with age, and melatonin regulates and stabilizes REM sleep. REM sleep has roles in memory consolidation and synaptic remodeling, and disrupted REM impairs object recognition and fear responses. Increased interest in sleep-stage-specific function and mechanisms has led to research on stimulation, behavioral, and pharmacologic interventions for multiple indications.

Self-reported sleep duration longer than 10 hours nightly has been associated with increased all-cause mortality as well as cardiovascular and cancer deaths. Decreased REM sleep has been associated with depression symptoms and depression risk.

To investigate whether REM sleep is associated with risk of mortality in two independent cohorts, researchers turned to data from MrOS (patients recruited from December 2003 to March 2005), and WSC (which began in 1988). Estimates of REM percentage of total sleep time and REM time in minutes was estimated based on a home (MrOS) or sleep lab (WSC) study. All-cause and cause-specific mortality were classified as cardiovascular, cancer, and other.

In MrOS, the percentage of REM sleep ranged from 0% to about 44%, with a normal distribution. Mean REM sleep percentage was 19.2%. Participants in the lowest quartile of REM sleep were older with higher rates of hypertension, heart attack, transient ischemic attack, antidepressant use, and lower physical activity.

Findings for the WSC cohort were similar, though mean percentage of REM sleep was lower in the WSC group (17.6%), “possibly owing to longer total sleep time during in-laboratory versus in-home studies,” the researchers noted.

Participants with less than 15% REM sleep had a higher mortality rate compared with those with 15% of more for all mortality definitions except cardiovascular.

Analysis by sleep stage “found percentage of REM sleep overwhelmingly important compared with other stages,” the authors wrote, implying that contributions from other stages were inconsequential.

“Because evaluating the quantity of REM sleep may have clinical implications, it is unfortunate that home sleep testing, the diagnostic study for obstructive sleep apnea now favored by health insurance companies, does not include information on sleep architecture,” the editorialists wrote.

They cited medication effects on sleep architecture as another important clinical dimension: “Serotonin reuptake inhibitors, selective serotonin and norepinephrine reuptake inhibitors, and tricyclic antidepressants all push REM sleep to later in the night (increased REM sleep latency) and reduce time in REM sleep,” they observed. “Some medications, such as gabapentin, prazosin, and bupropion, actually increase REM sleep.”

“Should we monitor REM sleep in patients on these medications or factor this effect into decision-making when devising a treatment plan? These remain unanswered questions,” the editorialists added. “There is no evidence to suggest that these medications provide a direct mortality benefit, and there are far too few data to suggest them as an intervention currently for patients observed to have reduced REM sleep.”

Study limitations include the possibility of unmeasured and residual confounding. Generalizability is limited because more than 90% of both cohorts were white.

1. Less REM sleep was associated with higher all-cause mortality, a population-based cross-sectional study found.

2. Be aware that this study shows an association, not causation, and confounding may have influenced results.

Paul Smyth, MD, Contributing Writer, BreakingMED™

Cat ID: 422

Topic ID: 82,422,730,422,192,50,925