Dual-energy X-ray absorptiometry underestimates in vivo lumbar spine bone mineral density in overweight rats.

Dual-energy X-ray absorptiometry (DXA) is currently the most widely used technique for measuring areal bone mineral density (BMD). However, several studies have shown inaccuracy, with either overestimation or underestimation of DXA BMD measurements in the case of overweight or obese individuals. We have designed an overweight rat model based on junk food to compare the effect of obesity on in vivo and ex vivo BMD and bone mineral content measurements. Thirty-eight 6-month old male rats were given a chow diet (n = 13) or a high fat and sucrose diet (n = 25), with the calorie amount being kept the same in the two groups, for 19 weeks. L1 BMD, L1 bone mineral content, amount of abdominal fat, and amount of abdominal lean were obtained from in vivo DXA scan. Ex vivo L1 BMD was also measured. A difference between in vivo and ex vivo DXA BMD measurements (P < 0.0001) is evidenced with an underestimation of in vivo BMD by (8.47 ± 10.54)%. This difference was found for the chow and high fat, high sucrose diets (P = 0.008), and a significant interaction between in vivo measurements, ex vivo measurements, and diet was observed (P = 0.030). Also, the data show a positive significant correlation of ex vivo BMD with body weight, perirenal fat, abdominal fat, and abdominal lean. Multiple linear regression analysis shows that body weight, abdominal fat, and abdominal lean were independently related to ex vivo BMD. DXA underestimated lumbar in vivo BMD in overweight rats, and this measurement error is related to body weight and abdominal fat. Therefore, caution must be used when one is interpreting BMD among overweight and obese individuals.