The Relationship Of ABMD And VBMD With Bone Area And Their Effects On Diagnosis Of Osteoporosis

ObjectiveTo probe into the relationship of areal bone mineral density (aBMD) and volumetric bone mineral density (vBMD) with bone area (BA) and their effects on diagnosis of osteoporosis(OP) at various skeletal regions in Chinese women.MethodsBA, bone mineral content (BMC), aBMD and vBMD were measured by QDR-4500A fan beam Dual-energy X-ray absorptiometry bone densitometer in 5585 females aged 5 to 96 years at the posteroanterior spine, lateral spine, and femur neck. The relationships between BA and BMC, aBMD, and vBMD were assessed by linear regression analysis. The subjects were divided into four groups according to BA quartile at different skeletal regions. The differences of prevalence rates of OP among groups were investigated.Results①At the posteroanterior spine, there was a significant positive correlation between BMC and BA (r = 0.706, P = 0.000). A intermediate correlation was found between aBMD and BA (r =0.394, P = 0.000). The correlation between vBMD and BA was markedly decreased (r =0.141, P = 0.000). The subjects aged more than 40 years (average age 52.7±9.58 years) (n=4012) were divided into Q1, Q2, Q3 and Q4 group according to BA quartile at the posteroanterior spine. BMC, aBMD, and vBMD presented an gradient increase among the groups. When we use aBMD to diagnose OP, detection rates of OP among these four groups were Q1 = 27.39%, Q2 = 18.30%, Q3 = 15.89%, and Q4 = 13.99%, respectively, presenting an obvious gradient decrease in the detection rates and risks of OP. When we use vBMD to diagnose OP, detection rates of OP among these four groups were Q1 = 27.39%, Q2 = 18.30%, Q3 = 15.89%, and Q4 = 13.99%, respectively. The differences among the groups in the detection rates and risks of OP were lower than aBMD. Among all subjects aged more than 40 years, the concordance rates for diagnosing OP by aBMD and vBMD were 87.9%.②At the lateral spine, there were significant positive correlations between BA and both BMC (r = 0.669, P = 0.000) and aBMD (r = 0.266, P = 0.000). BA was negatively correlated with vBMD1 (r =—0.045, P = 0.011), but not with vBMD2 (r =0.032, P = 0.080). The subjects aged more than 40 years (average age 52.7±9.58 years) (n=4012) were divided into Q1, Q2, Q3 and Q4 group according to BA quartile at the lateral spine. BMC and aBMD demonstrated an obvious gradient increase among the groups. No significant differences were found in the vBMD among the groups. Detection rates of OP by aBMD were Q1 = 30.04%, Q2 = 20.65%, Q3 = 14.70%, and Q4 = 15.14%, respectively. Detection rates of OP by vBMD1 were Q1 = 19.76%, Q2 = 18.49%, Q3 = 16.31%, and Q4 = 20.27%, respectively. Detection rates of OP by vBMD2 were Q1 = 28.15%, Q2 = 27.76%, Q3 = 24.19%, and Q4 = 28.57%, respectively. No significant differences were found among the groups. Prevalence rates of OP by vBMD2 were markedly higher than that by vBMD1. The differences of prevalence risks of OP by vBMD were obviously lower than that by aBMD. The concordance rates for diagnosing OP by aBMD and vBMD were 83.9% among all subjects.③At the femur neck, there was a significant positive correlation between BMC and BA (r = 0.366, P = 0.000). No significant correlations were found between aBMD and BA (r =—0.012, P = 0.104). vBMD showed significantly negative correlations with BA (r =—0.385, P = 0.000). The subjects aged more than 40 years (average age 52.7±9.58 years) (n=4012) were grouped into Q1, Q2, Q3 and Q4 group according to BA quartile at the femoral neck. BMC presented an obvious gradient increase among the groups. vBMD showed an obvious gradient decrease. No significant differences were found in the aBMD among the groups. When we use aBMD to diagnose OP, detection rates of OP among these four groups were Q1 = 8.59%, Q2 = 7.67%, Q3 = 8.86% and Q4 = 9.07%, respectively. No significant differences were found among the groups. When we use vBMD to diagnose OP, detection rates of OP among these four groups were Q1 = 1.90%, Q2 = 4.88%, Q3 = 6.87% and Q4 = 13.4%, respectively, showing an obvious gradient increment in the detection rates and risks of OP. Detection rates of OP by aBMD (8.55%) were significantly higher than that by vBMD (6.80%) among all subjects aged more than 40 years. The concordance rates for diagnosing OP by aBMD and vBMD were 67.1%.Conclusions①At the posteroanterior spine, aBMD and vBMD are positively correlated with bone size, while correlation of the latter was significantly lower than the former. The superiority of vBMD over aBMD in diagnosing OP is only adapted to subjects with larger or smaller skeletal size.②At the lateral spine, aBMD is significantly correlated with skeletal size. vBMD measured by DXA is not correlated with skeletal size. vBMD may be a sensitive marker for diagnosing OP. vBMD measured by DXA can be more accurately used to diagnose OP.③At the femur neck, aBMD is not correlated with bone size. While vBMD is negatively correlated with bone size. Detection rates of OP by aBMD is significantly higher than that by vBMD.