The Endocrine Society's 97th Annual Meeting & Expo:
Analysis of Increased Risk of Fracture in Diabetes
The higher risk of fracture in diabetes, reasons for this higher risk, and fracture prevention in this population were the topics that Ann V. Schwartz, MD, MPH, addressed in a talk on March 5, 2015, at the Endocrine Society Annual Meeting in San Diego, California. Dr. Schwartz is Professor of Epidemiology and Statistics, University of California, San Francisco.
Epidemiology and Cost of Hip Fracture
Hip fractures cost the average patient over $80,000, and the healthcare system, over $10 billion annually. One year after a hip fracture, 20% of patients have died, 30% have become permanently disabled, 40% are unable to walk independently, and 50% are no longer able to live independently.
Epidemiology of Type 2 Diabetes
The estimated percentage of people aged 45–64 years with type 2 diabetes is 13.7%, 26.9% of those aged 65 years and older.
Type 1 Diabetes and Bone Mineral Density
Bone mineral density in patients with type 1 diabetes is modestly reduced (estimated -0.22 spine Z-score, -0.37 hip Z-score). The expected relative hip fracture risk for this reduction in bone mineral density is 1.42. Though bone mineral density is only modestly reduced, hip fracture relative risk is 6.3.
Type 2 Diabetes and Bone Mineral Density
In patients with type 2 diabetes, bone mineral density is higher at the femoral neck and lumbar spine. Nevertheless, hip fracture risk is increased, at an age-adjusted relative risk of 1.4. Adjusted for body mass index and bone mineral density, relative risk of hip fracture is 1.7. In the Women’s Health Initiative (n=93,676), age-adjusted relative risk was evaluated for fractures of the hip, proximal humerus, foot, ankle, clinical spine, and forearm.
Factors Other Than Bone Mineral Density Play a Role
Type 1 and 2 diabetes carry a higher fracture risk for a given bone mineral density, which suggests that factors other than bone mineral density play a major role in bone quality in diabetes. At a given bone mineral density, diabetic bone is more fragile than nondiabetic bone. As a result, predicting fractures, which relies on bone mineral density, is compromised. The efficacy of osteoporosis therapies may also be compromised in diabetic patients. Falls are one possible explanation for this higher fracture risk at a given bone mineral density. Type 1 and 2 diabetes predisposes patients to fall more frequently. The relative risk of falling is 1.2 for patients with type 2 diabetes. Two lines of evidence, however, indicate that falls do not fully account for the higher fracture risk in patients with diabetes. After adjusting for fall frequency, type 2 diabetes is still associated with a higher fracture risk. And, bone quality has been shown to be reduced in rodent models of type 1 and 2 diabetes.
Diabetic bone fragility is characterized by:
- More rapid bone loss
- Smaller cross-sectional area of bone
- Worse trabecular bone score
- Reduced bone formation
- Cortical porosity
- Higher marrow fat
- Less resistance to microindentation
- Advanced glycation endproducts (AGEs)
- Possible microvascular damage in bone
Bone loss as determined by DXA occurs more rapidly in type 2 diabetic postmenopausal women, and osteoporotic femoral neck bone mineral density diminishes at a rate of 1.0% per year in diabetic women vs 0.6% in women without diabetes.
Predicting Fracture Risk in Patients with Type 2 Diabetes
The Fracture Risk Assessment Tool (FRAX) was found to predict but underestimate fracture risk in type 2 diabetic patients in a 2012 study of 39,603 participants in a Canadian cohort (Giangregorio et al, J Bone Miner Res). Based on further analyses in this cohort, the individual components of the FRAX algorithm tend to predict fracture risk in those with type 2 diabetes as well as they predict it in the broader population.
In 2014, age (per 10 years) was found to be a stronger predictor of hip fracture in those without diabetes (hazard ratio for subjects without diabetes, 2.59; for those with diabetes, 1.64 at P for interaction, < 0.001. (Leslie et al, Osteoporos Int).
The Canadian study did not consider possible interaction with race/ethnicity. In a US study based on NHANES data, the relationship between diabetes and fracture differed by race/ethnicity, with a hazard ratio, comparing those with and without type 2 diabetes, for non-Hispanic whites of 1.22, for non-Hispanic blacks 1.87, and for Mexican-Americans 2.37 (Looker at al, Bone, 2015).
Action to Control Cardiovascular Risk in Diabetes (ACCORD) Randomized Trial
The ACCORD trial set out to determine the effect on cardiovascular disease of intensive vs standard glycemic control in older adults with type 2 diabetes. In the ACCORD BONE ancillary study, fractures and falls were also assessed as outcomes. Participants were randomized to intensive or standard glycemic control strategies, with an achieved median hemoglobin A1C level of 6.4 and 7.5%, respectively.
In the ACCORD Bone ancillary study, fractures were assessed at 54 of the 77 ACCORD clinical sites that included 7287 of the 10,251 ACCORD participants. At annual visits, 6782 participants were asked about falls in the previous year.
During an average follow-up of 3.8 years, 198 of 3655 participants in the intensive glycemic control and 189 of 3632 participants in the standard glycemic control group experienced at least one nonspine fracture. The average rate of first nonspine fracture was 13.9 and 13.3 per 1000 person-years in the intensive and standard groups, respectively (hazard ratio 1.04).
During an average follow-up of 2.0 years, 1122 of 3364 intensive- and 1133 of 3418 standard-therapy participants reported at least one fall. The average rate of falls was 60.8 and 55.3 per 100 person-years in the intensive and standard glycemic control groups, respectively (1.10 [0.84 – 1.43]).
The conclusion drawn from the trial was that compared with standard glycemic control, intensive glycemic control did not increase or decrease fracture or fall risk. ACCORD, however, could not address the effect of poor glycemia control on fracture risk. Subsequent observational studies have addressed this question and indicate that fracture risk is increased over 8% with higher hemoglobin A1C.
Osteoporosis Therapy in Patients with Type 2 Diabetes
Theoretical concerns regard the efficacy of antiresorptive therapies in type 2 diabetes, given the reduced bone formation, rise in nonvertebral fractures, and cortical vs trabecular bone deficits associated with these agents. In addition, these therapies are most effective in the broader population in those with a T-score below -2.5. Nevertheless, diabetic patients are at increased fracture risk at higher bone mineral densities. Evidence of the agents’ efficacy, particularly for reducing fracture risk in type 2 diabetes is limited.
Trials of Osteoporosis Drugs in Reducing Fractures in Diabetic Women
A post hoc analysis of a trial of alendronate found similar efficacy in improving bone mineral density when comparing women with and without diabetes. Trial data on bisphosphonates and fracture efficacy are not available, but observational studies suggest that the drugs exert similar anti-fracture efficacy as in nondiabetic patients. A post hoc analysis of the Multiple Outcomes for Raloxifene Evaluation (MORE) trial found that raloxifene is effective in reducing vertebral fracture risk in diabetic women. In an analysis of the Direct Assessment of Nonvertebral Fractures in Community Experience (DANCE) observational study, teriparatide demonstrated similar efficacy in preventing fractures and improving bone mineral density when type 2 diabetic and nondiabetic patients were compared.
Osteoporosis Drugs and Metabolic Parameters
Whether these therapies influence glucose metabolism is in question. In rodent models, lower uncarboxylated osteocalcin (ucOC) leads to obesity, elevated glucose, and diabetes. This led to the theory that antiresorptives might increase diabetes risk by suppressing bone turnover, including lowering ucOC. In a post hoc analysis of randomized trial data, however, neither alendronate, zoledronic acid, nor denosumab have been shown to significantly affect fasting glucose or to increase incident diabetes. Alendronate and denosumab were associated with a modestly larger weight increase in type 2 diabetic women during the trials, consistent with rodent models. The degree of difference in weight gain, however, was not clinically important.
Type 1 and 2 diabetes are associated with a higher risk of fracture, and at a given bone mineral density, patients with diabetes are at higher fracture risk. As a result, DXA T-score and FRAX underestimate fracture risk in patients with diabetes.
“Reasons for the poorer bone quality in patients with diabetes are being investigated, since understanding this relationship is crucial to our efforts to reduce fracture risk in this population,” said Dr. Schwartz. “To prevent fractures in this population, good glycemic control is important, since poor control has been shown to raise fracture risk. And general guidelines should be followed when prescribing osteoporosis therapy. Hopefully, we will learn more about the anti-fracture efficacy of these therapies in diabetic patients.”
March 13, 2015