Effects of Age and Body Mass Index on Thoracolumbar Spine X-Ray for Diagnosing Osteoporosis in Elderly Women: Tianliao Old People (TOP) Study 07.

PURPOSE: The aim of this study was to determine the effects of diagnostic discordance with or without a thoracolumbar spine lateral view X-ray in patients with osteoporosis. METHODS: We randomly enrolled 368 women over 65 years old (74.3 ± 6.0 years) from Tianliao Township in 2009 (response rate: 75.7%). A diagnosis of osteoporosis was confirmed using one of these criteria: (1) a history of non-traumatic fracture, (2) vertebral fractures based on a thoracolumbar spine lateral view X-ray, or (3) a bone mineral density T-score ≤ -2.5 for the total hip, the femoral neck, the lumbar spine, or all 3 sites. The prevalence of osteoporosis in three groups was compared based on Model I (criteria 1+2) vs. Model II (criteria 1+3) vs. Model III (criteria 1+2+3). The role of thoracolumbar X-ray reflected by the diagnostic discordance of osteoporosis between Models II and III was evaluated. RESULTS: The overall prevalence of osteoporosis was 78.3% (Model III, age-standardized 78.1%). The diagnostic discordance was 17.4% in the 368 participants. A logistic regression model showed that age was negatively associated with diagnostic discordance (odds ratio [OR] = 0.93, 95% confidence interval [CI]: 0.88-0.98, p < 0.05), but body mass index was positively associated (OR = 1.07, 95% CI: 1.00-1.15, p < 0.05). CONCLUSIONS: A thoracolumbar spine lateral view X-ray should be added for women ≥ 65 years old or with a body mass index ≥ 25 kg/m2 to minimize the diagnostic discordance in osteoporosis, especially in highly endemic regions.

Inverse relationship between central obesity and osteoporosis in osteoporotic drug naive elderly females: The Tianliao Old People (TOP) Study.

To examine the relationship between central obesity and osteoporosis in elderly females in a rural community, a total of 368 ambulatory elderly women were recruited by random sampling during July 2009. Structured questionnaires were completed to survey possible osteoporosis-related risk factors. Subjects were dichotomized into either noncentral obese (waist circumference [WC]<80cm) or central obese subgroups (WC≥80cm) for further analysis. Bone mineral densities were scanned by dual-energy X-ray absorptiometry installed in a mobile bus. Thoracolumbar spine X-ray examination was interpreted by the same radiologist. Of the 365 subjects with completed data, 275 (75.3%) aged women were classified as having osteoporosis based on diagnostic Model III. Compared with the nonosteoporosis subjects, the subjects with osteoporosis had relatively higher mean age, lower body mass index, and a lower percentage of central obesity. Using the binary logistic regression method, central obesity was negatively associated with osteoporosis in all 3 models (odds ratios in the 3 models were 0.348, 95% confidence interval [CI]: 0.130-0.927; 0.444, 95% CI: 0.218-0.905; and 0.415, 95% CI: 0.184-0.936, respectively; p<0.05). Our study suggests that the paradox between central obesity and osteoporosis in elderly women should be of concern and warrants further study.