It's a Catch-22 of the highest order. People with alcohol problems often use alcohol to get to sleep -- but it actually keeps them from getting good-quality sleep all night long.

At the same time, they're highly likely to suffer from full-blown chronic insomnia that keeps them from getting enough sleep night after night – and that condition has been shown to cut their chances of getting sober again.

Meanwhile, their doctors aren't likely to prescribe them insomnia medications, because most sleeping pills can be habit-forming or have adverse effects due to an alcohol-damaged liver.

Now, a small new pilot study from a team of University of Michigan alcoholism and sleep researchers offers some sign of a possible way out of this conundrum.

The study, published in the August issue of the journal Alcoholism: Clinical and Experimental Research, suggests that the drug gabapentin might be able to reduce insomnia in recovering alcoholics, and help them stay away from alcohol more successfully. The drug, often used to treat epilepsy and chronic pain, is not habit-forming and is not processed by the liver.

Although the study involved only 21 insomniacs in recovery from alcohol dependence, and did not provide long-term gabapentin treatment or long-term follow-up on their sleep or their alcohol recovery, it was randomized, placebo-controlled, and double-blinded. In all, 30 percent of the patients who received gabapentin during alcohol recovery relapsed to drinking, compared with 80 percent of those who received a placebo.

Based on the results, the researchers have already launched additional studies of the potential role of gabapentin in alcohol recovery and sleep.

"We showed that the patients who got the real drug, rather than placebo, were less likely to relapse to drinking -- or if they relapsed it was later," says lead author Kirk Brower, M.D., FASAM, the executive director of U-M Addiction Treatment Services and a professor of psychiatry at the U-M Medical School. "In other words, gabapentin prevented and delayed relapse. Meanwhile, patients reported sleeping better in both the treatment and placebo groups, which may be due to the gabapentin in the first group and the resumption of drinking in the other."

Co-author Flavia Consens, M.D., an associate professor of neurology and member of the U-M Sleep Disorders Center, is cautiously optimistic that the new findings could open the door to better understanding of how to handle sleep problems in people who are trying to recover from their dependence on alcohol. As many as 70 percent of people with alcohol problems suffer insomnia, she says, while others cope with other sleep disturbances including breathing problems known as sleep apnea.

Nearly 14 million Americans meet the diagnostic criteria for alcohol abuse or alcoholism. Alcohol problems, alone or in combination with illicit drug problems, account for 40 percent of admissions to addiction treatment programs each year, according to the federal Substance Abuse and Mental Health Services Administration.

"There may be some underlying chemical changes in the brain that prompt alcoholics to report more insomnia as a co-existing condition than non-alcoholics," she says. "A possible explanation of these new findings is that the gabapentin might decrease the insomnia initially, and the patient may not need or crave alcohol as a treatment for the insomnia. We're also looking into other factors that may have an effect on the neurochemistry of the brain, and see how they could impact recovery and sleep."

The researchers caution that they did not observe differences in brain wave data collected during sleep studies conducted before and after patients received gabapentin. Neither did the drug appear to have a greater benefit for insomnia than placebo during the first 6 weeks of receiving study medication. Six weeks after stopping medication, however, those who had taken gabapentin reported worse insomina than those on placebo. Insomnia was measured using standardized questionnaires for a total of 12 weeks

All of the volunteers met national criteria for alcohol dependence, and were either in alcohol treatment or expressed a willingness to abstain from alcohol. They also all met criteria for insomnia that had lasted six months or more. They could not have other medical or mental health conditions, or be taking medications, that might affect their sleep, and underwent blood tests to rule out medical illnesses such as thyroid deficiency and liver disease.

Each of the study volunteers spent three nights in the U-M Sleep Disorders Center: two during the preparation for the study, and one three weeks after they began to receive gabapentin or placebo. All the volunteers received up to six brief sessions of behavioral therapy aimed not at sleep or alcohol issues, but rather at adherence to the study medication.

Fourteen of the volunteers successfully completed the entire study, including a follow-up appointment six weeks after they completed the six-week course of gabapentin or placebo, and three overnight sleep studies.

Brower notes that the medication dose and schedule used in the study may have contributed to the relatively weak effect on sleep that was seen from gabapentin. Patients took one dose each evening, rather than the three doses throughout the day that are routinely given for epilepsy or pain.

"These results raise more questions for us to explore, including the potential impact of gabapentin on people who are in recovery from alcohol dependence but do not report insomnia," he says.

The team has begun such a study, and is recruiting people who have a history of heavy drinking but have been sober for three to 12 weeks. These volunteers will spend a total of six nights in the recently opened eight-bed U-M Sleep & Chronophysiology Laboratory, and will receive study medication for one week. They'll also keep a sleep diary for 20 days, wear a wristwatch-like device to track their daily sleep/wake cycles, and come back for a follow-up visit. More information on this study is available at 734-232-0237 or by e-mailing dreamteam@umich.edu.