01.08.14
Repeat bone mineral density testing in older men and women may not improve the ability to predict fracture beyond the insight gained from the original bone scan, researchers claim.
In a population-based cohort of older men and women not yet treated for bone loss, the result of a second bone density test within four years of the first did not "meaningfully improve" fracture prediction, according to Sarah Berry, M.D., and her colleagues at the Institute for Aging Research at Hebrew SeniorLife in Boston, Mass.
The finding suggests that repeated bone scans might not be needed, at least among people at relatively low risk for fractures, Berry and colleagues argue. Experts, however, warn that doctors should careful analyze the study population to decide if the findings apply to any particular patient. "This is not a generalized recommendation that is going to fit all people," one doctor noted.
An initial bone mineral density test can help manage osteoporosis, the authors noted. For instance, guidelines for starting drug therapy for osteoporosis are based on bone density combined with risk classification scores. But repeated tests have not proven their value.
Nevertheless, Medicare will pay for a bone density screening every two years without limitation on the number of tests and regardless of the baseline bone density. If those tests cannot help manage patients, Berry and her cohorts argued, there's no need to conduct them.
To assess the issue, they turned to the Framingham Osteoporosis Study, whose participants were enrolled starting in 1948. Between 1987 and 1999, living participants were invited for three bone mineral density tests, about 4 years apart.
For their analysis, Berry and her colleagues examined 310 men and 492 women with an average age of 74.8 who had two measurements of femoral neck bone density. The study subjects were followed from the second measure until death or through 2009, or 12 years of follow-up.
The primary outcome of the study was the risk of hip or other major osteoporotic fracture.
After a median follow-up of 9.6 years, Berry and colleagues found, 113 participants --- or 14.1 percent --- met the primary endpoint, with 88 hip, 24 spine, five shoulder, and 33 forearm fractures.
The average change in bone mineral density was a drop of 0.6 percent annually, ranging from a gain of 5.6 percent to a loss of 9 percent per year.
In a model adjusted for clinical characteristics and baseline bone density, every standard deviation of decrease in annual percentage bone density change was associated with a hazard ratio for incidence hip fracture of 1.43, with a 95 percent confidence interval from 1.16 to 1.78.
At 10 years' follow-up, the absolute risk of hip fracture was 10.2 percent among participants whose bone density change was average, compared with 14.1 percent among those whose change was a standard deviation below the mean.
Similarly, every standard deviation decrease in annual percentage annual percentage bone density change was associated with a hazard ratio of 1.21 for major osteoporotic fractures (with a 95 percent confidence interval from 1.01 to 1.45).
Again, at 10 years' follow-up, participants with an average change had an absolute risk of major osteoporotic fracture of 15.6 percent, compared with 18.3 percent among those whose change was a standard deviation lower.
Despite those significant associations, Berry and colleagues found, there was little difference in the area under the receiver operating curve --- a measure of predictive value --- for models based on baseline bone density or percentage change over time.
Combining baseline bone density and change over time also did not change the area under the curve, they reported.
The researchers cautioned that the findings are based on an untreated, but screened cohort, and might not apply to people using osteoporosis medications. They also noted that the study population included only Caucasians and might not apply to other ethnic or racial groups.
In a population-based cohort of older men and women not yet treated for bone loss, the result of a second bone density test within four years of the first did not "meaningfully improve" fracture prediction, according to Sarah Berry, M.D., and her colleagues at the Institute for Aging Research at Hebrew SeniorLife in Boston, Mass.
The finding suggests that repeated bone scans might not be needed, at least among people at relatively low risk for fractures, Berry and colleagues argue. Experts, however, warn that doctors should careful analyze the study population to decide if the findings apply to any particular patient. "This is not a generalized recommendation that is going to fit all people," one doctor noted.
An initial bone mineral density test can help manage osteoporosis, the authors noted. For instance, guidelines for starting drug therapy for osteoporosis are based on bone density combined with risk classification scores. But repeated tests have not proven their value.
Nevertheless, Medicare will pay for a bone density screening every two years without limitation on the number of tests and regardless of the baseline bone density. If those tests cannot help manage patients, Berry and her cohorts argued, there's no need to conduct them.
To assess the issue, they turned to the Framingham Osteoporosis Study, whose participants were enrolled starting in 1948. Between 1987 and 1999, living participants were invited for three bone mineral density tests, about 4 years apart.
For their analysis, Berry and her colleagues examined 310 men and 492 women with an average age of 74.8 who had two measurements of femoral neck bone density. The study subjects were followed from the second measure until death or through 2009, or 12 years of follow-up.
The primary outcome of the study was the risk of hip or other major osteoporotic fracture.
After a median follow-up of 9.6 years, Berry and colleagues found, 113 participants --- or 14.1 percent --- met the primary endpoint, with 88 hip, 24 spine, five shoulder, and 33 forearm fractures.
The average change in bone mineral density was a drop of 0.6 percent annually, ranging from a gain of 5.6 percent to a loss of 9 percent per year.
In a model adjusted for clinical characteristics and baseline bone density, every standard deviation of decrease in annual percentage bone density change was associated with a hazard ratio for incidence hip fracture of 1.43, with a 95 percent confidence interval from 1.16 to 1.78.
At 10 years' follow-up, the absolute risk of hip fracture was 10.2 percent among participants whose bone density change was average, compared with 14.1 percent among those whose change was a standard deviation below the mean.
Similarly, every standard deviation decrease in annual percentage annual percentage bone density change was associated with a hazard ratio of 1.21 for major osteoporotic fractures (with a 95 percent confidence interval from 1.01 to 1.45).
Again, at 10 years' follow-up, participants with an average change had an absolute risk of major osteoporotic fracture of 15.6 percent, compared with 18.3 percent among those whose change was a standard deviation lower.
Despite those significant associations, Berry and colleagues found, there was little difference in the area under the receiver operating curve --- a measure of predictive value --- for models based on baseline bone density or percentage change over time.
Combining baseline bone density and change over time also did not change the area under the curve, they reported.
The researchers cautioned that the findings are based on an untreated, but screened cohort, and might not apply to people using osteoporosis medications. They also noted that the study population included only Caucasians and might not apply to other ethnic or racial groups.