By Dennis Thompson
People on sodium-glucose cotransporter2 (SGLT2) inhibitors were twice as likely to require a lower limb amputation as people taking another type of diabetes medication, Scandinavian researchers found.
Patients also had a doubled risk of diabetic ketoacidosis, a life-threatening complication in which acids called ketones build up in the bloodstream.
"Patients at high risk of amputation, for example those with peripheral artery disease or foot ulcers, might be monitored more closely if SGLT2 inhibitors are used, and the risk of this adverse event may be considered when deciding on which drugs to use," said lead researcher Dr. Peter Ueda, a postdoctoral researcher with Karolinska University Hospital in Stockholm, Sweden.
SGLT2 inhibitors include dapagliflozin (Farxiga), empagliflozin (Jardiance) and canagliflozin (Invokana and Invokamet).
"The way this class of medications works is if you have higher blood sugars in you, it actually causes an increase in urination because that's how your body will dispose of the extra sugar," explained Dr. David Lam. He's an assistant professor of medicine, endocrinology, diabetes and bone disease at the Icahn School of Medicine at Mount Sinai in New York City.
The U.S. Food and Drug Administration issued a warning in 2017 that two large clinical trials had linked canagliflozin to an increased risk of leg and foot amputations.
However, other clinical trials have revealed no such amputation risk in either dapagliflozin or empagliflozin, said Dr. Kevin Pantalone, an endocrinologist with the Cleveland Clinic.
In this new observational study, 61 percent of patients were using dapagliflozin, 38 percent were on empagliflozin and just 1 percent on canagliflozin.
"They're reporting an increased risk that hasn't been observed in prospective, randomized, placebo-controlled trials, and that's the gold standard," Pantalone said. "Yes, it's interesting they find this observation in patients who are on SGLT2 inhibitors, but only 1 percent of the patients were on the drug that is really concerning for harm."
Ueda agreed that the clinical trial data on record for dapagliflozin or empagliflozin does not jibe with the results.
For this study, Ueda and his colleagues analyzed national health data from Sweden and Denmark for 17,213 patients taking SGLT2 inhibitors and 17,213 patients taking GLP1 receptor agonists between July 2003 and December 2016.
Use of SGLT2 inhibitors was associated with a twofold increased risk of lower limb amputation compared to people on GLP1 receptor agonists. Risk of diabetic ketoacidosis also was doubled.
Researchers tried to control for a large number of other factors that could otherwise explain this association, such as disease history, other medications, and social and economic conditions for the patients. But the study did not prove that these drugs caused amputation risk to rise.
"Although we used a strict study design and accounted for a large number of patient-related variables in our analyses, the results could be affected by unmeasured differences in the characteristics of the patients receiving SGLT2 inhibitors vs. the comparator drug," Ueda said. "This is always the case with observational studies and the reason why findings from such studies should be considered with caution."
Pantalone and Lam said one potential way that SGLT2 inhibitors might increase amputation risk is due to the way they work in the body.
Many people who are diabetic have poor circulation in their legs and feet, and these drugs cause people to excrete more urine to lower their blood sugar, the doctors said.
"You could potentially get more dehydrated if your blood sugars are very elevated," Lam said. "Because of the decrease in blood volume, it's decreasing overall blood flow and that might be compromising someone who's already at risk for having poor blood circulation to their lower extremities. It could be making an existing problem worse."
The conflicting results between this observational study and previous clinical trials means doctors will have to take a patient-by-patient approach, Pantalone and Lam said.
Not every patient taking the drugs needs to stop. "When I have patients come in and they've been on it for three years and they're doing great, they have no history of peripheral vascular disease and no problems, I don't just take everybody off it," Pantalone said.
On the other hand, there clearly are patients who might want to avoid SGLT2 inhibitors.
"You just need to think twice," Lam said. "If this patient has circulatory problems or an active foot ulcer, maybe we should think about a different agent for them."
"If I have somebody sitting in front of me who already has a history of amputation, this is probably a drug I'm going to avoid," Pantalone said. "Or if there's somebody who does have established peripheral vascular disease, maybe this is someone I'm going to avoid prescribing this medication."
The findings were published Nov. 14 in the journal BMJ.