![A photo of purple rubber-gloved hands holding a urine sample and a toxicology test strip](http:http://assets.medpagetoday.net/media/images/108xxx/108630.jpg?width=0.8)
One patient had multiple false positive toxicology tests after taking a popular diabetes drug.
Despite not ingesting alcohol for over 10 months, a man in his early 60s tested positive for ethanol in his urine on the last four urine samples he provided a probation office. None of the urine samples contained any traces of illicit or non-prescribed substances.
But after calling up his primary care provider for fear of going to jail over the positive tests, the primary care provider's lab found no detectable ethanol in a new urine sample, Aaron L. Schwartz, MD, PhD, of the Corporal Michael J. Crescenz Department of Veterans Affairs Medical Center in Philadelphia, detailed in a correspondence.
Though they did not find ethanol on the repeat urine test, the primary care lab did find glycosuria present -- an expected finding as the patient was newly prescribed 20 mg daily of the SGLT2 inhibitor empagliflozin (Jardiance) just 5 months prior for his diabetes.
This led Schwartz to call up the city probation office to ask about exactly how they go about storing urine samples. In response, he was told that urine samples were sent once daily to an external laboratory to be tested, but samples were not refrigerated prior to transport.
To test a theory, the primary care office took their new urine sample out of refrigeration. One day later they retested the urine for ethanol after it remained at room temperature for 24 hours. Now, the sample tested positive for ethanol.
"It seems this medication caused my patient to have positive urine tests for alcohol even though he wasn't drinking alcohol," Schwartz told ľֱ. "This seems to have happened because the medication increases the amount of sugar and bacteria in the urine. Microbes can turn sugar into alcohol in the same way that grape juice turns into wine."
"When I prescribed this medication for my patient's diabetes, I never thought it might cause this harm," he said.
He explained how the SGLT2 inhibitor class of agents produces glycosuria even when blood glucose levels are within the normal range and also increases the presence of microbes in the urinary tract -- two ingredients needed for microbial fermentation.
After the positive ethanol tests at the city probation center, the urine sample collected by the primary care facility 3 days later had a glucose level of 1,000 mg/dL. It was negative for urinary ethanol, urinary ethyl glucuronide, leukocyte esterase, nitrites, ketones, blood, and bilirubin. A urine culture showed less than 50,000 CFU/mL of mixed gram-positive organisms.
"In this case, the collection and storage of urine samples by the criminal justice system appears to have been inadequate for ensuring specimen stability," Schwartz noted.
"I hope there is a greater awareness of this possible phenomenon among clinicians," he continued. "Toxicology testing for illicit drugs and alcohol can be high stakes for patients, both in the criminal justice system and the medical system."
He added that since the class of SGLT2 inhibitors -- which includes others like dapagliflozin (Farxiga), canagliflozin (Invokana), and ertugliflozin (Steglatro) -- are frequently prescribed for a wide-range of diabetes, kidney, and cardiovascular indications, this could have the potential to affect a lot of patients.
"I would encourage clinicians to be mindful of whether a patient is subject to urine toxicology testing when taking this medication," Schwartz recommended. "A clinician should be somewhat skeptical of positive tests in these situations."
Disclosures
Schwartz reported relationships with Aetna, the Lauren and John Arnold Foundation, Lown Institute, Medicare Payment Advisory Commission, Tufts Medical Center, U.S. Agency for International Development, VBID Health, and the Veterans Health Administration.
Primary Source
New England Journal of Medicine
Schwartz AL "SGLT2 inhibitors and false positive toxicology tests" N Engl J Med 2024; DOI: 10.1056/NEJMc2313463.