Low bone mineral density (BMD) has been considered as a risk factor for mechanical complications after spine and hip surgeries. Currently, the Dual-energy X-ray absorptiometry (DXA) is the gold standard for the evaluation of bone mineral density. However, DXA scan is not always readily available for spine and hip surgery patients. In this study, we focused on clavicle cortical bone thickness that can be easily assessed on routine chest radiograph to predict low bone mineral density among spine and hip surgery patients.
METHODS
The bilateral clavicles shown on the chest radiograph was divided equally into four equal portions (A: distal, B: middle-distal, C: middle-proximal, D: proximal) and the thickness of each cortical and trabecular bone was measured at each point. The clavicle index was defined as the following formula: (cortical + trabecular) / trabecular bone. Pearson’s correlation coefficients were calculated to compare the clavicle cortical index at each point, unilateral and bilateral average values, and BMDs in the proximal femur, lumbar spine, and distal radius. Multivariate linear regression analyses were conducted with setting BMDs at each anatomical region as the outcome variable. The receiver operating characteristic (ROC) curve analysis was also performed to assess the predictive value of the index for diagnosing osteoporosis defined as BMD <70% of young adult mean value.
A total of 219 patients (61.2% female) with mean age of 68.03 years were included in the final analysis. Mild but significant correlations were found between the bilateral average middle-distal cortex index (MDCI) and BMDs in the radius (r = 0.409) but weak correlations in the femur (r = 0.313) and spine (r = 0.209). Female was a factor for stronger correlations between MCDI and BMD in the radius (r = 0.63). In spine surgery patient, mild correlations were observed between the MDCI and BMDs in the femur (r = 0.42) and radius (r = 0.432) but weak correlations in lumbar BMD (r = 0.223). In hip surgery patient, mild correlations were observed between the MDCI and BMDs in lumbar BMD (r = 0.223) and radius (r = 0.446) but weak correlations in femur BMD (r = 0.186). Multivariate analysis adjusted with sex and age demonstrated that MDCI was an independent factor for lumbar (β (95% CI (confidential interval)) = 0.31 (0.1-0.53) p<0.001) and radius (β (95% CI) = 0.16 (0.04-0.25) p<0.001). The ROC analyses demonstrated that the areas under the curve (95%CI) of the bilateral average MDCI for osteoporosis diagnosis were moderately accurate: AUC:0.69 (0.55-0.83) with femur BMD, AUC:0.716 (0.57-0.861) with lumbar BMD, and AUC:0.834 (0.737-0.931) with radius BMD.The ICCs (95%CI) at each measurement points varied by measured points ranging 0.4-0.8. The ICC of the averaged values at points B and G were 0.74(0.40-0.89)
Our results indicated that the MDCI was significantly associated with BMDs especially in the radius among spine and hip surgery patients. The clavicle cortex index is calculated easily using a routine chest radiograph, hence, the index can be utilized as an osteoporosis screening tool for spine and hip surgery patients.