VANCOUVER – In patients with type 2 diabetes, sodium-glucose cotransporter 2 (SGLT2) inhibitors as an adjunct to metformin were not associated with an increase in fracture risk, according to a new real-world study.
There have been some reports of an increase in fracture risk associated with SGLT2 inhibitors, and it was observed in the phase 3 CANVAS trial of canagliflozin (Invokana), which led to a Food and Drug Administration warning of fracture risks associated with canagliflozin use. Some ensuing studies did not show an increased risk, but these studies were generally less than a year in duration and may have missed longer-term risk, according to Veerle van Hulten, MSc.
“Fracture risk is something that takes a long time to develop, so we wanted to have a longer follow-up. We looked into the CPRD [Clinical Practice Research Datalink], which is a beautiful database containing real-world data from primary care practices,” said Ms. van Hulten, a PhD student at Maastricht (the Netherlands) University, who presented the study at the annual meeting of the American Society for Bone and Mineral Research.
Ms. van Hulten and colleagues compared SGLT2 inhibitors with dipeptidyl peptidase–4 (DPP-4) inhibitors because the latter are used in similar populations and have been shown to have no effect on fracture risk.
“What we found is that SGLT2 inhibitors are not associated with an increased fracture risk. Even with a duration of use of over 811 days, we did not observe an increased hazard ratio for fractures when compared DPP-4 inhibitor users,” Ms. van Hulten said.
SGLT2 inhibitors reduce blood sugar by increasing elimination of sugar in the urine. They also increase phosphate, reduce calcium, and increase parathyroid hormone, which could in turn negatively affect bone turnover, according to Ms. van Hulten.
In the new study, conducted between January 2013 and June 2020, the researchers used propensity score matching to compare adult patients, including 13,807 who were prescribed SGLT2 inhibitors and 28,524 who were prescribed DPP-4 inhibitors for the first time. They matched patients based on demographics, comorbidities, comedication, and lifestyle factors.
There was no association between SGLT2 inhibitor use and overall fracture risk or major osteoporotic, hip, vertebral, humerus, radius, or ulna fractures. There was no difference in risk for any duration of use, even with the longest duration of use of 811 days (adjusted hazard ratio, 1.0). There were no differences among specific SGLT2 inhibitors, including canagliflozin (aHR, 1.12; 95% confidence interval, 0.73-1.72). Analyses by sex and age also revealed no statistically significant differences between the two drug classes.
During the Q&A session after the presentation, Sarah Berry, MD, MPH, an associate professor of medicine at Harvard Medical School and a clinical researcher at the Marcus Institute for Aging Research, both in Boston, noted the trend toward an increase in fracture risk in the first 90 days. “It looked like there was something going on in the first 90 days, and then after that the results were much closer to the null. I would put out maybe another potential mechanism whereby the SGLT2 inhibitors might cause fracture, and that’s falls. They cause polyuria, and any drug you give that causes women to rush to the bathroom may well cause fractures, particularly in the short term,” Dr. Berry said.
Ms. van Hulten agreed, and also brought up that the drugs can cause osmotic diuresis. That can lead to hypovolemia, the symptoms of which include weakness, fatigue, and dizziness. “And increased falls, of course, increases fracture risk. We do not expect anything to happen to bone metabolism in the first 90 days. I think we can agree that there would be more time needed to alter the bone enough to increase fracture risk, so we expect that this trend toward an increased risk might be attributable to that increased fall risk that might occur with SGLT2 inhibitor use,” she said.
It’s possible that such a mechanism explains increased fracture risk seen in some earlier short-term studies, she added.
Overall, Ms. van Hulten said that the results should provide some confidence in SGLT2 inhibitors, though more work needs to be done. “I think we provide reassurance that SGLT2 inhibitors are safe to use. However, we still only have a median follow-up of 1.6 years. It’s not as long as we maybe would like, but it’s the best we can do with the data available, since the SGLT2 inhibitors have only been used since 2013. So maybe it’s best to prescribe it and keep [fall risk] in mind and look into the effects later on again, but it seems to be safe to use.”
The study received funding from the Novo Nordisk Foundation. Ms. van Hulten and Dr. Berry reported no relevant financial relationships.
A version of this article appeared on Medscape.com.