CME/CE Released: 08/07/2014 ; Valid for credit through 08/07/201
CLINICAL CONTEXT
By reduction of bone mineral density, adipocyte stimulation, and inhibition of osteoclast differentiation, thiazolidinediones (TZDs) may increase the risk for fractures. Previous research has suggested that fracture risk is increased in patients treated for diabetes with TZDs compared with those treated with metformin.
However, previous epidemiologic research has not shown an increased fracture risk with sulfonylureas. The goal of this observational study, results which were presented at the American Diabetes Association 2014 Scientific Sessions, was to compare fracture risk among multiple classes of glucose-lowering agents.
STUDY SYNOPSIS AND PERSPECTIVE
The oral diabetes drug classes of TZDs and sulfonylureas both boost the risk for fractures compared with metformin, a large database analysis has found.
The observational study confirms previous findings of increased fracture risk with TZDs but is the first to compare multiple classes of glucose-lowering agents and the first to suggest a possible increased fracture risk for sulfonylureas, Sandhya Mehta, PhD, an investigator with a technology company specializing in large data analytics, told the American Diabetes Association 2014 Scientific Sessions.
Dr. Mehta said her study findings should be taken into consideration in prescribing antidiabetic drugs, especially in those patients already at higher risk for fracture.
"Be more careful with TZDs, to be sure, and also with sulfonylureas," she told Medscape Medical News, but she added that the association with sulfonylureas needs confirmation.
Indeed, session moderator Amanda I. Adler, MD, PhD, from Addenbrooke's Hospital, Cambridge University, United Kingdom, and chair of the Technology Appraisal Committee of the United Kingdom's National Institute for Health and Care Excellence, told Medscape Medical News that despite the investigators' attempts to control for many potential confounders, "One concern is residual confounding, that patients who take sulfonylureas are at higher risk for fractures than those who don't take sulfonylureas."
More Work Needed to Drill Down Into Sulfonylurea Fracture Risk
Dr. Mehta and colleagues retrospectively analyzed 2008-2012 data from the Medical Outcomes for Effectiveness and Economics registry of more than 100 million individuals. A total of 99,892 adults were identified as new users of glucose-lowering drugs: metformin (77.8%), sulfonylureas (15.3%), dipeptidyl peptidase-4 (DPP-4) inhibitors (2.7%), TZDs (2.7%), incretins (0.81%), and meglitinides (0.6%).
Within the 5-year follow-up period, 7353 patients (7.4%) had evidence of fracture.
The incidence of fractures was 6.8% among the total 76,924 patients taking metformin, 10.9% among the 2679 taking TZDs, and 9.7% in the 15,162 receiving sulfonylureas. Rates for other glucose-lowering agents ranged from 6.1% of 799 individuals taking incretin drugs to 10.7% of the 626 receiving meglitinide.
After adjustment for age, gender, region, a variety of medical conditions, and other medications, the hazard ratios (HRs) for fracture risk compared with metformin were 1.40 for TZDs (P < .0001) and 1.09 for sulfonylureas (P = .0054). The increase in fracture risk for the other drug classes compared with metformin did not reach statistical significance.
In her presentation, Dr. Mehta noted that the findings support the hypothesis that TZDs increase fracture risk by decreasing bone mineral density, stimulating adipocytes, and inhibiting osteoclast differentiation.
She noted, however, that previous epidemiologic studies have not found an increased fracture risk with sulfonylureas. "The present study points to the need of further investigation on the association between sulfonylureas and the risk of fractures."
Dr. Adler told Medscape Medical News that what preceded the fracture is important information. "If they were having more low blood sugars and consequently fell, it would be interesting to know, but unfortunately she couldn't tell us."
She also noted that although the findings for the other glucose-lowering drugs are reassuring, "[w]e can't really know what they're doing to bones without knowing what else is going on. If you took a drug that made you more likely to topple over, your bones could be equally as strong as the group that didn't fall over, but [you could] still have more fractures."
Dr. Mehta and Dr. Adler have disclosed no relevant financial relationships.
American Diabetes Association 2014 Scientific Sessions; June 15, 2014. 165-OR
STUDY HIGHLIGHTS
- The investigators used a large administrative claims database for this longitudinal, retrospective cohort study.
- Using multivariate survival analysis, they assessed the comparative safety regarding fracture risk for antidiabetic agents including sulfonylureas, biguanides (metformin), incretin mimetic agents, meglitinide analogues, TZDs, and DPP-4 inhibitors.
- Inclusion criteria for the cohort were age 18 years or older, newly prescribed only a single class of antidiabetic agents, diagnosed with diabetes before starting treatment, continuously enrolled for 12 months before and at least 12 months after starting treatment, and not diagnosed with fracture in the 12 months preceding treatment.
- Patient data were censored at the time of fracture, when they left the cohort, or at the end of the follow-up period.
- Of 99,892 new adult users of antidiabetic drugs, 77.8% were receiving metformin, 15.3% were receiving sulfonylureas, 2.7% were receiving DPP-4 inhibitors, 2.7% were receiving TZDs, 0.81% were receiving incretins, and 0.6% were receiving meglitinides.
- During the 5-year follow-up, fracture occurred in 7353 patients (7.4%).
- Compared with metformin use, sulfonylurea use and TZD use were associated with significantly higher fracture risk.
- The HR was 1.09 for sulfonylureas (95% confidence interval [CI], 1.03 - 1.16; P = .0054) and 1.40 for TZDs (95% CI, 1.25 - 1.58; P < .0001), after adjustment for potential confounders including age, gender, region, medical comorbidities, and use of other medications.
- Fracture risk with metformin was not statistically significantly different from that seen with other classes of antidiabetic agents. Fracture rates ranged from 6.1% of 799 patients taking incretin drugs to 10.7% of 626 patients taking meglitinide.
- On the basis of the findings of their large longitudinal study, the investigators conclude that fracture risk is 9% to 40% greater in users of sulfonylureas and TZDs.
- The authors recommend that clinicians consider these findings when prescribing antidiabetic drugs, especially for those patients already at greater risk for fracture.
- Limitations of this study include its observational design with the potential for residual confounding factors, in that patients who take sulfonylureas may be at greater risk for fractures than those who do not take these drugs.
CLINICAL IMPLICATIONS
- Findings of a large, longitudinal, retrospective cohort study suggest that fracture risk is 40% greater in users of TZDs than in users of metformin. Clinicians should consider these findings when prescribing antidiabetic drugs, particularly for patients who are already at increased fracture risk.
- This study also suggests that fracture risk is 9% greater in users of sulfonylureas than in users of metformin.
No comments:
Post a Comment