Zolpidem Impairs Balance and Cognition on Awakening

Megan Brooks

January 20, 2011

January 20, 2011 — The nonbenzodiazepine hypnotic zolpidem produces clinically significant impairments in balance and cognition on awakening after 2 hours of sleep in both younger and older adults, new research suggests.

Dr. Kenneth Wright

"The effects on walking stability were very pronounced in older adults, and the effects on cognition were actually even more pronounced in the younger than in the older adults," Kenneth P. Wright, Jr., PhD, of the Department of Integrative Physiology, University of Colorado (UC) at Boulder, noted in a telephone interview with Medscape Medical News.

Use of these medications "may have greater consequences for health and safety than previously recognized," with nighttime falls a particular concern, Dr. Wright and colleagues say.

The study was published online January 13 in the Journal of the American Geriatrics Society.

Mechanistic Study

Whether sleep medications contribute to the risk for and occurrence of falls at night is a "concern," Lawrence Epstein, MD, a member and past president of the American Academy of Sleep Medicine, who was not involved in the study, told Medscape Medical News.

"Studies to date have shown conflicting data, some suggesting sleeping pill use increases risk, while others have not found such a risk," noted Dr. Epstein, who is chief medical officer of Sleep HealthCenters, a sleep medicine specialty group headquartered in Brighton, Massachusetts.

The current study looked at the mechanism of such a risk and "clearly shows" that sleeping medication can cause balance and cognitive impairments in the elderly "and could cause problems, such as falls, if they get out of bed and try to move about during the night," Dr. Epstein said.

The UC-Boulder study involved 12 older adults (mean age, 67.4 years) and 13 younger adults (mean age, 21.9 years). All were healthy, physically active, and mentally sharp.

Each participant spent 3 nights in a sleep laboratory, approximately 1 week apart, under 3 different test conditions. In 2 conditions, subjects received 5 mg of zolpidem (hypnotic condition) or placebo (sleep inertia/grogginess condition) 10 minutes before bedtime. In the third condition, they received a placebo and then remained sitting in bed for 2 hours past their normal bedtime (wakefulness control, limited sleep deprivation condition).

Balance and cognition were assessed right before and shortly after awakening from 2 hours of sleep. Balance was assessed with the tandem walk test, in which subjects place one foot in front of the other on a 6-inch-wide beam on the floor. Stepping off the beam constitutes a tandem walk failure. Cognition was assessed by having participants add randomly generated numbers on a computer (a test of working memory) and with the Stroop Color Word Task (a test of cognitive speed).

Impaired Balance on Zolpidem

Before the test conditions, all of the subjects had good balance; none in either group stepped off the beam on 10 practice tests. However, after taking zolpidem, 7 of 12 older adults (58%) and 3 of 11 younger adults (27%) stepped off the beam, indicating a loss of balance.

Some older adults had tandem walk failures for at least 30 minutes after they were awakened, whereas younger adults had tandem walk failures only within the first few minutes.

Number needed to harm analyses showed 1 tandem walk failure for every 1.7 (95% confidence interval [CI], 1.4 – 2.0) older and 5.5 (95% CI, 5.2 – 5.8) younger adults treated with zolpidem.

Poor performance on the tandem walk is a predictor of falls and hip fractures in older adults. The current study, the researchers say, "provides experimental evidence that may help to explain the fall risk of older adults treated with sleep medication."

We have to recognize that, with most of the sleep medications on the market, falls are a side effect, especially in high-risk populations. In our case, we studied very healthy people who are at the least risk of falls, and yet they still expressed this risk.

"We have to recognize that, with most of the sleep medications on the market, falls are a side effect, especially in high-risk populations," Dr. Wright said. "In our case, we studied very healthy people who are at the least risk of falls, and yet they still expressed this risk."

As for the other test conditions, none of the older adults and only one of the younger adults (9%) stepped off the beam during the sleep inertia condition. The researchers say further study is needed to assess possible influences of sleep inertia on fall risk.

Three older adults (25%) and none of the younger adults stepped off after going to sleep 2 hours past their bedtime. This suggests that older adults who stay up later than usual may have impaired balance, although this requires further study, the researchers say.

Cognitive Problems 'Underrecognized'

Cognition was significantly more impaired after zolpidem than after staying up late (wakefulness control) in both younger and older adults, with the effects greatest in the younger group.

Table. Treatment Effect of Hypnotic vs Control on Working Memory and Reaction Time

Measure Older Group Younger Group
Working memory (calculations) −4.3 (−7.0 to −1.7) −12.4 (−18.2 to −6.7)
Stroop Test reaction time increase, milliseconds 76 (13.5 to 138.5) 126 (34.7 to 217.5)

Effects are mean difference between conditions; lower working memory calculations and higher Stroop median reaction time indicate worse performance.

Older adults appeared to be less susceptible to sleepiness-induced changes in cognitive function caused by sleep inertia and sleep deprivation than younger adults, at least for the tasks studied.

"Our results are actually consistent with a few studies that have been done showing that older adults have less of a degree of magnitude impairment when they're sleep deprived," Dr. Wright noted. "It doesn't mean that they are performing better than younger adults, it just means that the change is not as great. That's what we find too."

Overall, Dr. Wright thinks the cognitive impairments with these medications are "probably underrecognized, especially in the young."

The recommended initial dose of zolpidem is 5 mg in adults older than 65 years and 10 mg in adults younger than 65 years. The current study used 5 mg in both groups and therefore may underestimate the influence of zolpidem on balance and cognition for younger adults who took only half the recommended dose for their age, the researchers point out.

They also note that the effects of zolpidem were analyzed for only 1 night. "We don't know the effects of taking these medications for weeks or months and whether or not any of these effects change. The long-term effects of sleep medicines require follow-up studies," Dr. Wright noted.

Andrew Krystal, MD, MS, director of the Sleep Research and Insomnia Center at Duke University Medical Center in Durham, North Carolina, said, "It is important to put these findings into context. There is increasing evidence that if you are an older individual and you have insomnia, that itself puts you at risk for falls at least for no other reason that you're up more at night and you may be a little drowsy,” noted Dr. Krystal, who was not involved in the study.

Dr. Epstein made the point that the study did not directly measure fall rates at night but used an alternative measure, namely balance and cognition.

"The counterargument," he said, "is that, with sleeping medication, people are much less likely to get up at night and fall rates may be reduced. Further evaluation of specific fall rates is needed, but the study makes a case for caution and monitoring when using sleeping medications in this population," Dr. Epstein said.

A New Option in Appropriate Patients

In March 2010, the US Food and Drug Administration approved doxepin (Silenor), in dosages of 3 and 6 milligrams, for treatment of chronic insomnia.

Doxepin is used as an antidepressant in dosages of 50 to 100 mg. According to Dr. Krystal, studies have shown that "when you drop the dose of doxepin that low, it becomes very selective for effects on sleep and the effects on other pharmacologic systems — cholinergic, adrenergic, serotonergic systems — become minimal and generally absent. Therefore, you get an excellent risk-benefit profile."

Unlike zolpidem, low-dose doxepin is not a hypnotic and does not act broadly in the brain; it is a histamine antagonist. "The mechanism of action is completely different than the benzodiazepines and nonbenzodiazepines. At very low doses, doxepin provides a very targeted surgical strike by blocking the wake-promoting effects of histamine, and it doesn't impede the normal function of the rest of the brain," Dr. Krystal said.

"Because it is not sedating, we have reason to believe that this drug would not impair balance and therefore would be particularly good in the elderly; however, there hasn't been a systematic study on this," he added.

But low-dose doxepin is not for every insomniac. "For people who have trouble falling asleep, it's not your drug," Dr. Krystal said. "It doesn't help people fall asleep; it helps people sleep better through the night without sedating you the next day. It's indicated for people who have trouble staying asleep, particularly in the early morning."

Dr. Krystal said in his experience most doctors have not been motivated to differentiate problems falling asleep with problems staying asleep. "There has not been a great reason to do it. Now there is good reason to make that differentiation," he said.

The study was supported by grants from the National Institutes of Health and UC-Boulder. Dr. Wright discloses financial relationships with Cephalon Inc, Takeda Pharmaceuticals North America, and Zeo Inc. Dr. Krystal has been a consultant to Somaxon, Sanofi-Aventis, and Sepracor. Dr. Epstein has disclosed no relevant relationships.

J Am Geriatr Soc. Published online January 13, 2011.