Effect of subtle positioning flaws on measured bone mineral density of the hip.

Careful hip positioning is necessary for accurate and precise measurement of bone mineral density (BMD) by dualenergy X-ray absorptiometry (DXA). Large, intentional rotational deviations from best hip position reportedly increase measured BMD at the femoral neck. The aim of this study is to determine the effect of unintentional, subtle deviations from best hip position ("flaws") on expected and measured BMD of the hip and its subregions. Two hundred DXA scans (GE/Lunar Prodigy) performed in the dual-femur mode were randomly selected and scrutinized for hip-positioning flaws. Hips were sorted by side (right/left), positioning quality (flawless/flawed), and specific positioning flaw (under/over-rotation and adduction/abduction). Individual hip pairs in which at least one hip was flawlessly positioned were further sorted as flawless/flawless, flawless/vertically flawed, flawless/rotationally flaw, and flawless/dually flawed. Differences and direction of difference in BMD between sorted groups and within individual hip pairs were analyzed by univariate analysis (t-test for equal samples). Two hundred hip pairs (400 individual hip scans) were analyzed. Overall, there was no significant difference in BMD between all right and all left hips or between all flawlessly positioned and all flawed hips. In hip pairs in which one hip was flawlessly positioned and the contralateral hip position was flawed (vertically, rotationally, or both), the measured BMD of the latter hip was not predictably greater than the former. Average absolute intrapair BMD difference between hip subregions was 0.038+/-0.001 g/cm2 and was unaffected by the presence or type of positioning flaw. Net intrapair difference between sorted hip groups and within hip pairs was negligible, indicating that the direction of the variance was equally positive and negative. In the practice of clinical densitometry, subtle positioning flaws do not generate predictable changes in measured BMD at any hip region of interest. The current teaching that rotational deviations from best hip position results in a greater measured BMD needs to be reconsidered when comparing a rotationally flawed hip with a contralateral flawlessly positioned hip.