Candida, Fermentation, Drunkenness, GI tract, Auto-brewery, Yeast

Getting Drunk Without Drinking: The “Auto-Brewery” Syndrome

by Alan R. Gaby, MD

In a recently published case report,¹ a previously healthy 46-year-old man developed memory loss, mental changes, and episodes of depression over a period of six years. These symptoms began after he received a three-week course of the broad-spectrum antibiotic, cephalexin. On one occasion he was arrested for presumed drunken driving although he claimed that he had not consumed any alcohol. Stool culture was positive for Saccharomyces cerevisiae (brewer’s or baker’s yeast) and S. boulardii. He was also found to have Candida albicans and C. parapsilosis in the small bowel, which were presumed to be fermenting carbohydrates to alcohol. His blood alcohol level, measured eight hours after a carbohydrate meal and while under observation so that he could not drink any alcohol, was 0.057%. For comparison, behavioral and other changes may occur at blood alcohol levels as low as 0.02%; and the legal limit for “driving under the influence” (DUI) in the U.S. is 0.08%.

After treatment with a low-carbohydrate diet, antifungal medication, and a probiotic he became completely asymptomatic. Carbohydrates were gradually reintroduced into his diet without causing symptoms or increasing his blood alcohol level. At the time the case report was submitted to the journal, the patient had been symptom-free 1.5 years.

This case illustrates a condition that has been called the auto-brewery syndrome or the gut fermentation syndrome. It is characterized by symptoms of drunkenness after ingestion of carbohydrates. The first case of this syndrome was reported in 1952: that of a Japanese person who developed symptoms of drunkenness after eating. The condition was found to be due to excessive fermentation of food components by Candida species living in the stomach. As of 1976, 23 similar cases had been reported in Japan, but there had been no such case reports from other countries in the medical literature.² Symptoms in these patients included a drunken feeling, often associated with flushing of the face, headache, dizziness, sweating, fever, diarrhea, or palpitations. The symptoms typically began 30 minutes to 20 hours after eating a carbohydrate food and lasted 2 to 10 hours. Ethanol was detected in stomach contents, blood, urine, and expired breath. C. albicans or other yeasts and fungi were identified in the gastric and/or duodenal contents in these patients. Administration of antifungal drugs was dramatically effective in most cases.

In a 1990 paper, three British doctors provided evidence that fermentation of carbohydrates to alcohol in the gastrointestinal tract is actually quite common.³ They measured blood ethanol concentrations in 510 patients in whom chronic intestinal candidiasis or gut fermentation was suspected. Ethanol levels were measured in the fasting state and one hour after ingestion of 5 g of glucose. Sixty-one percent of the patients had measurable levels of ethanol one hour after glucose ingestion, with a mean concentration of 0.0025% (range, 0.001% to 0.007%). In 14 patients (2.7% of those tested), low concentrations of ethanol (less than 0.001%) were detected in the fasting state. The maximum blood alcohol concentration seen in this study was below the level at which functioning becomes impaired. However, considering that a 12-ounce soft drink contains eight times as much sugar as the challenge dose used in the study and that a typical Western diet provides daily more than 20 times as much sugar as the challenge dose, it would not be surprising if diet-induced increases in blood alcohol levels were sufficient to cause symptoms in some people.

The auto-brewery syndrome overlaps with, and may be a contributing factor to, a condition that has been given various names including Candida-related complex, Candida hypersensitivity syndrome, or the simple but less precise term, candidiasis. I will refer to this condition here as Candida-related complex (CRC). In 1978, Dr. Orian Truss introduced the concept that non-systemic yeast infections can cause a wide array of systemic symptoms and contribute to or exacerbate various disease conditions. Truss observed that CRC was a contributing factor to a wide range of symptoms and conditions, including depression, anxiety, hyperactivity, irritability, headache, difficulty with memory and concentration, chronic diarrhea, recurrent urinary tract symptoms, premenstrual syndrome, and autoimmune diseases.

In cases in which CRC was a contributing factor, oral administration of nystatin resulted in clinical improvement, sometimes dramatic.^4,5 Since that time, others have confirmed that CRC plays a role in a broad range of health conditions.⁶ Today, many nutrition-oriented practitioners recommend an “anti-Candida” program for selected patients, although many in the mainstream medical community doubt the existence of the syndrome described by Truss. In addition to antifungal medication, an anti-Candida program typically requires the avoidance of simple sugars and other refined carbohydrates. The inability of CRC patients to tolerate sugar is presumably due in part to gut fermentation, although other adverse effects of sugar (such as dysinsulinism or immune-system dysregulation) may be involved as well.

Ethanol is metabolized in the body to acetaldehyde, a compound that may be even more important than ethanol as a mediator of CRC-related illness.⁷ Some of the symptoms of CRC resemble those of alcohol hangovers, which are caused in large part by acetaldehyde. Acetaldehyde also interferes with a number of enzymes involved in intermediary metabolism, and it is thought to play a role in the pathogenesis of some autoimmune diseases.⁸

In my 19 years of clinical practice, I saw hundreds of patients (mostly women) who had a history consistent with possible CRC. I included this condition in the differential diagnosis if the patient had typical symptoms plus a history of recurrent vaginal yeast infections or oral thrush, or if they had taken medications that can promote the growth of Candida species (such as antibiotics, oral contraceptives, or glucocorticoids). Dietary modification and antifungal medication were each frequently beneficial, but combining these treatments was usually needed to achieve the best results. It was not uncommon for patients with suspected CRC to answer affirmatively when asked if eating sugar made them feel drunk. That particular symptom typically improved on the anti-Candida program. There is no generally accepted diagnostic test for CRC, so it remains a clinical diagnosis, and treatment is usually undertaken on an empirical basis.⁹