Chronic Sleep Loss Takes its own Toll on Alertness


by Jim Schnabel

April 26, 2010

Is one long slumber enough to restore you to full alertness after many nights of insufficient sleep? Not according to a new study from Harvard researchers, who found that chronic sleep loss impairs alertness in ways that even a ten-hour sleep can’t fix.

“When people try to determine whether they have caught up on sleep loss, it is important that they think not only about the last night or two, but the last several weeks,” says the study’s lead author, Daniel Cohen, a neurologist and sleep researcher at Harvard Medical School and Brigham and Women’s Hospital in Boston.

The study, reported in the Jan. 13 issue of Science Translational Medicine, was designed not just to illuminate the effects of chronic sleep loss but also to give clues to the underlying brain mechanisms that normally push us to get sufficient sleep. The findings lend weight to a theory that there are both long-term and short-term mechanisms that make us sleepy.

Cohen and his colleagues recruited nine healthy young men and women, and had them spend three weeks in a controlled sleep lab. In the lab, their exposure to day and night cues followed a normal 24-hour cycle, but the volunteers had to stay awake for about 33 hours at a time, in a way that mimicked the extended shifts of doctors and military workers. They then had a 10-hour period for sleep.

The ratio of sleep time to waking time in each 43-hour cycle was about 1 to 3.3, considerably lower than the usual 1 to 2 ratio that comes from sleeping 8 hours and staying awake for 16. In this way, the subjects’ relative sleep loss accumulated throughout the study, even though their sleep times after every shift were longer than normal.

Cohen and his colleagues found that for the first several hours after waking from this long sleep, the subjects performed almost normally on a standard test of alertness. But as they got further into their 33-hour waking shift, their alertness scores fell off sharply—and this downward slope steepened as the trial wore on.

“The deterioration from staying awake 24 hours in a row was ten times greater than normal,” says Cohen.

The researchers also found that the normal “circadian” 24-hour rhythm of alertness, which corresponds to people’s usual waking and sleeping periods, became more pronounced as the trial wore on—the scores at the early-morning low point got progressively worse.

Because some of this poor performance might be due not to sleep loss but to the jet-lag-like influence of having a wake/sleep cycle constantly out of sync with the ambient day/night cycle, Cohen and his colleagues also studied a control group. Eight other volunteers went through the same three-week trial, involving the same, non-circadian 43-hour shifts, but with a normal 1:2 sleep/wake ratio. The falloff in their scores during the trial was much less than that seen in the chronic sleep-loss group.

Sleep’s enforcers

Researchers know that the normal, 24-hour cycle of alertness and drowsiness is enforced by a network of brain cells centered in the hypothalamus. For example, to help produce drowsiness during night-time hours, these cells secrete the hormone melatonin. But there also appears to be a separate mechanism—or set of mechanisms—for making us sleepy when we’ve simply stayed awake too long.

Exactly what these mechanisms are isn’t yet proved. But much of the theorizing in recent years has centered on the neurotransmitter adenosine, a byproduct of waking neuronal activity that is known to cause sleepiness. Adenosine is found in higher amounts in certain brain regions as a normal day wears on; in addition, caffeine, which boosts alertness, appears to do so by blocking adenosine receptors.

According to this idea, sleeping shuts off adenosine production, allowing the brain to clear out the excess over time and return to alertness. In principle, a long sleep should enable the brain to recover even after an extended buildup of adenosine.

Cohen’s team saw such as recovery in their subjects, but it lasted only for a few hours, and the subsequent falloff grew worse with increasing sleep loss. That finding hints at the presence of a longer-term mechanism of sleep enforcement.

Cohen suspects that the worsening falloff may be due to a long-term sensitization of the adenosine system in response to increasing sleep loss. A recent study in rats found that sleep deprivation increases the density in the brain of the most common adenosine receptor, known as the A1 receptor.

“If the receptor density increases, the brain becomes sensitized so that the same level of adenosine, from a given number of hours awake, causes a larger behavioral response,” Cohen says.

Radhika Basheer, a Harvard Medical School researcher who participated in the recent rat study, says that the research by Cohen and colleagues (with which she wasn’t involved) is an “interesting study and fits our observations.” She notes that these short-term and long-term changes in the adenosine system also seem to parallel the dynamics of drug abuse: In both cases it seems that powerful but short-term stimuli cause short-term, easily-reversible changes, while powerful chronic stimuli “force changes at the levels of gene expression,” she says, “with long-lasting effects.”

Cohen plans to continue to investigate the mechanisms of sleep-maintenance, their dynamics and their limits. For example, researchers do not yet know what it takes to reverse the effects of several weeks of chronic sleep loss: “It’s possible that it could take a couple of weeks of normal sleep to return back to a fully rested state,” he says.