PHILADELPHIA — Newly diagnosed diabetes seems to confer a protective effect against fracture that disappears with increased disease duration, ultimately leading to an increased fracture risk in patients with long-term disease, according to data presented at the annual meeting of the American Society for Bone and Mineral Research.
Patients with newly diagnosed diabetes had an 11% lower risk of osteoporotic fracture and an 18% lower risk of hip fracture, compared with controls, in a retrospective study of 82,094 diabetic patients and 236,682 control subjects. Individuals who have had diabetes for less than 5 years had the same risk as controls for an osteoporotic fracture. People with diabetes of longer duration had a significantly greater risk of osteoporotic fractures in general (9% greater) and hip fractures in particular (36% greater).
The data provide “compelling evidence that long-term diabetes is associated with increased fracture risk and that newly diagnosed diabetes shows some protective effect,” said Dr. William D. Leslie, professor of endocrinology and metabolism at the University of Manitoba, Winnipeg.
A number of population-based studies have shown that bone density is increased in patients with type 2 diabetes. “Paradoxically, fracture risk [in diabetic patients] is much greater than can be explained on the basis of bone density,” said Dr. Leslie. Diabetic complications leading to falls and possibly altered bone quality are thought to mediate these effects, but the exact relationship between diabetes and fracture risk has not been determined.
For this study, the researchers used fracture data from the Population Health Information System for the Province of Manitoba from 1984 to 2004. Each diabetic adult in the database who was at least 20 years old (as of Jan. 1, 1994) was matched with three nondiabetic control subjects based on gender, birth year, area of residence, and aboriginal or nonaboriginal ethnicity.
Diabetic patients were defined as those with at least one hospitalization or two physician claims for diabetes or a diagnosis of diabetes within a 3-year period. Diabetic patients were further classified by the duration of their disease. Long-term diabetics had the disease for more than 5 years (before 1994). Short-term diabetics had the disease for less than 5 years (prior to 1994). Newly diagnosed diabetics were those who were diagnosed between 1994 and 2004. To be considered nondiabetic, controls had to be continuously nondiabetic between 1984 and 2004. Fractures were determined from ICD-9 codes. Combined osteoporotic fractures included vertebral, wrist, or hip breaks. Hip fractures were also analyzed as a predefined subgroup.
A total of 16,457 osteoporotic fractures were identified among controls and 5,591 among diabetic patients. A total of 5,224 hip fractures were identified among controls and 1,901 among diabetic patients. Overall, there was no statistical difference between diabetics and controls in terms of osteoporotic or hip fracture rates. Both types of fracture risk increased exponentially with age. In addition, longer duration of diabetes was associated with a greater fracture risk for both types of fractures.
The researchers also tried to account for comorbidity by using the number of ambulatory diagnostic groups (ADGs), developed at Johns Hopkins University. In this tool, ICD codes are grouped into 32 different ADGs, based on clinical and expected utilization criteria. Increasing number of ADGs corresponds with increasing comorbidity. The researchers used a regression model including age and comorbidity (represented by ADG number).
For diabetic patients within each age group (younger than 50, 50–59, 60–69, 70–79, and 80 years and older), fracture risk increased with increasing duration of diabetes. Interestingly, the highest fracture rates for each diabetes subgroup were among the youngest age group. In fact, there was a fourfold increase in hip fracture risk for long-term diabetics younger than 50 years, compared with nondiabetic controls. The strongest protective effect of being newly diagnosed was seen among the oldest patients.
The researchers hypothesize that the opposing effects of overweight or obesity provide short-term protection from fractures, but with time, this effect is overcome by diabetes-associated complications.