The highly popular ketogenic or "keto" diet, which restricts carbohydrates in favor of high-fat foods, could help ease the severity of alcohol withdrawal and reduce dependence on drug therapies for that condition, suggests a small study from the National Institutes of Health. 

Published Friday in Science Advances, the paper highlights a potential clinical strategy for people undergoing treatment for alcohol use disorders, said Nora D. Volkow,  a researcher with the National Institute on Alcohol Abuse and Alcoholism, director of the National Institute on Drug Abuse and a co-author of the study. "It's very exciting to see how we can use nutrition, instead of just medication, to enable and help with recovery," she explained. "I think this study illustrates how powerful this tool may be." 

The study adds to a growing body of research on the potential medical uses for ketogenic diets, including evidence that the keto diet could benefit people living with Alzheimer's disease and epilepsy.

The trial involved 33 people seeking treatment in a clinical setting — 19 were assigned a keto diet and 14 ate a standard American diet — over a period of three weeks. In the American diet, 50% of calories came from carbohydrates, 35% from fats and 15% from protein. The participants on the keto diet got 80% of their calories from fats, 15% from protein and 5% from carbs. 

"You dramatically reduce the consumption of carbohydrates," Volkow said of the keto diet. "Carbohydrates are glucose. When you don't have glucose, you start to favor a completely different production of energy." 

The researchers used magnetic resonance spectroscopy scans to monitor the participants' levels of brain ketones — a backup fuel source produced by the liver — amino acids and markers of neuroinflammation, in addition to weekly measurements of ketones in the blood.

The patients on the keto diet reported lower cravings for alcohol than those on the typical American diet. Significantly, the benefits were observed in some participants within 48 hours of starting the diet, suggesting that it could serve as a strategy for people anticipating an imminent withdrawal period.

Another promising result: 57% of patients on the standard American diet needed benzodiazepine drugs to alleviate their withdrawal symptoms, versus 26% of patients on the keto diet.

Previous research has shown that when people are intoxicated with alcohol, their capacity to metabolize acetate — a byproduct of alcohol being broken down by the liver — goes up, while their ability to metabolize glucose goes down, Volkow said. 

"Your body is adapting to optimize the use of acetate that is so widely available because you're always drinking," she said. "If you're an alcoholic and all of the sudden you go cold turkey — no alcohol — that means the acetate in your brain is going to go down. How are you going to make the transition to glucose for the production of energy?" 

Detoxing off alcohol would therefore leave patients in a state of deprivation in which the brain would not have access to its now-primary energy source, contributing to the symptoms of withdrawal.  

The researchers hypothesized that ketosis, the metabolic state in which the body produces ketones, could ease the transition from acetate as the primary source of energy in the brains of alcoholics. The new findings suggest that theory could hold true, though with a small-sample-size caveat, Volkow said. Some of the results were also subjective: participants self-reported their alcohol cravings. However, the researchers also used a rat model to achieve a more objective understanding of the keto diet's effects on alcoholism. 

"It was important to do the study in the rats," Volkow said. "It would have been very interesting to see when you're on the ketogenic diet and you get access to alcohol, does your desire to keep on drinking change? We hypothesized that it will curtail that thirst alcoholics have — once they get the first sip, they cannot stop. But we cannot do that in human [studies] because it's not ethical to give alcohol to somebody who's on treatment."  

Rats on the keto diet significantly reduced their alcohol intake, suggesting a potentially long-lasting effect on drinking patterns that goes beyond the withdrawal period. "I think that's extraordinary," Volkow said. "We did not expect that. It was a major surprise." 

In the future, Volkow wants to assess whether the ketogenic diet interferes with the body's metabolism of alcohol, and whether it could help prevent heavy drinking patterns and relapses before they begin. 

"We're very interested, in our testing, whether keeping with the ketogenic diet long-term will help people prevent relapse and also help them recover brain function, which, in turn, will help them achieve sobriety," she explained. "This is the next important question."  

The study, "Ketogenic diet reduces alcohol withdrawal symptoms and alcohol intake in rodents," published April 9 in Science Advances, was authored by Peter Manza, Ehsan Shokri-Kojori, Danielle Kroll, Dana Feldman, Katherine McPherson, Erin Biesecker, Rui Zhang, Kimberly Herman, Melanie Schwandt, Dardo Tomasi, Nancy Diazgranados, and Gene-Jack Wang, National Institute on Alcohol Abuse and Alcoholism; Corinde E. Wiers, National Institute on Alcohol Abuse and Alcoholism and Perelman School of Medicine, University of Pennsylvania; Leandro F. Vendruscolo, Sophie K. Elvig, and Janaina C.M. Vendruscolo, National Institute on Drug Abuse; George F. Koob and Nora D. Volkow, National Institute on Alcohol Abuse and Alcoholism and National Institute on Drug Abuse; Jan-Willem van der Veen, National Institute of Mental Health, Maryland; Sara A. Turner and Shanna Yang, Clinical Center Nutrition Department, National Institutes of Health; Mackenzie C. Cervenka, Department of Neurology, Johns Hopkins University; Anders Fink-Jensen, University Hospital of Copenhagen; and Helene Benveniste, Department of Anesthesiology, Yale University.