Objective: To analyze bone mineral density (BMD) of elderly patients in Beijing suffering from osteoporotic fracture, and discuss Chinese population BMD threshold of osteoporosis complicated by fracture.
Methods: Randomly selected three hundred male and female patients over the age of fifty years old with hip, shoulder or wrist fracture forming a research group. And randomly selected one hundred and fifty male and female patients over fifty years old with lumbar diseases forming a comparison group. Hip and spine BMD values of both groups were measured by dual energy X-ray absorptiometry and statistically analyzed by t-test; The BMD values were analyzed using a receiver operating characteristic (ROC) curve to determine the optimal cutoff point as the fracture thresholds.
Results: BMD values relate negatively to age (research group, L2-L4 r value: -0.267, femur neck r value: -0.284, Ward triangle r value: -0.368, great trochanter r value: -0.314, all P < 0.01; comparison group, femur neck r value: -0.383, Ward triangle r value: -0.476, great trochanter r value: -0.305, all P < 0.01); There is evident difference between statistical comparisons of experimental and control group (P < 0.01); The cutoff point of the Ward triangle BMD at 0.732 g/cm(2) (91.0% sensitivity, 56.1% specificity, 9% missed diagnosis, 43.9% misdiagnosis) yields the likelihood of osteoporotic fracture for male; and it is at 0.577 g/cm(2) (72.6% sensitivity, 64.5% specificity, 27.4% missed diagnosis, 35.5% misdiagnosis) for female.
Conclusion: It is reasonable to take the cutoff point values as thresholds of fracture risk of osteoporosis.