[Correlation between prevalent and incident osteoporotic vertebral fractures in Beijing male elderly]

Wei Shang,Wei Yu,Ling Xu,Qiang Lin,Jun-Ping Tian
2005-01-12
Abstract:Objective: To investigate the correlation between prevalent and incident osteoporotic vertebral fractures in Beijing male elderly. Methods: Roentgenography of the thoracic and lumber vertebrae (T4-L4) and dual X-ray absorptiometry at the lumber spine (L2-4) and proximal femur (neck, greater trochanter, and Ward's triangle) were conducted so as to observe the vertebrae and measure the bone density upon 271 Beijing men, aged 72 +/- 6 (63.1 - 92.7). The subjects were followed up for 3.7 +/- 0.4 years. Results: Prevalent fractures involving 98 vertebrae, mostly T12-L1 and mostly I degrees fracture, occurred in 71 men with an incidence rate of 28.3% (95% CI 22.8% - 34.3%). Incident fracture, mostly T12-L1 and mostly I degrees fracture too, occurred in 8 patients (95% CI 1.4% - 6.2%). The average baseline age of those with incident fracture was 77 +/- 8, significant higher than those of the subjects without fracture (71 +/- 6) and the subjects with only prevalent fractures (72 +/- 5) (both P < 0.05). The body height and weight of the subjects with only prevalent fractures were significantly lower than those of the other 2 groups (P = 0.025 and P = 0.05). However, the body mass index (BMI) was not significantly different among these 3 groups. The bone density values of femur neck and greater trochanter of the group with prevalent fracture but without incident fracture were 0.81 g/cm(2) +/- 0.11 and 0.4 g/cm(2) +/- 0.11 g/cm(2) respectively, both significantly lower than those of the group without fracture (0.85 g/cm(2) +/- 0.11 g/cm(2) and 0.78 g/cm(2) +/- 0.11 g/cm(2) respectively, both P < 0.05). The BMI value decreased in the order of the group without fracture, the group with only prevalent fracture, and the group with prevalent and incident fracture. The incidence rate of incident fracture was 7% in the group with prevalent fracture, significantly lower than that in the group without fracture (1.7%, P < 0.05). Among those with prevalent fracture the higher the degree of the prevalent fracture the more cases of incident fracture (P = 0.02). The incidence rate of incident fracture in those with prevalent fracture of 4 vertebrae was 33.3%, higher than those in other 2 groups. Logistic analysis showed that the baseline age and severity of vertebral fracture were closely correlated with incident fracture (OR = 1.229, 95% CI 1.037 - 1.407, and OR = 4.270, 95% CI 1.691 - 19.785). Conclusion: As independent of BMI, baseline age and severity of prevalent fracture are closely correlated with incident fracture in Beijing male elderly. Baseline BMI is a strong predictor of fracture risk.
What problem does this paper attempt to address?