Dual X-ray absorptiometry of the lumbar spine: the precision of paired anteroposterior/lateral studies

Abstract

Supine lateral dual X-ray absorptiometry (DXA) has the potential to provide a sensitive and precise method of measuring changes in trabecular bone mass in the lumbar vertebral bodies. An important reason for the improvement in precision compared with the earlier decubitus lateral technique is the use of baseline compensation, an algorithm which computes the changes in lateral bone mineral density (BMD) using data from both anteroposterior (AP) and lateral scans. This report examines the precision of supine lateral DXA based on scans of 12 female volunteers (age range 21-56 years). Five AP/lateral scan pairs were performed on each subject and precision expressed by the coefficient of variation (CV). For the AP scan the CV for both L1-L4 and L2-L4 BMD was 0.8%. Precision of the L2-L4 lateral vertebral body, mid-vertebral and volumetric BMD parameters with (without) the baseline compensation correction were 1.2% (2.1%), 2.4% (3.0%) and 1.5% (2.5%), respectively. Two methods of performing baseline compensation, referred to as the multiplicative and additive algorithms, were compared and the former shown to give significantly better precision. The precisions of different combinations of lumbar vertebrae for the measurement of lateral vertebral body BMD were compared and the precision for L2-L4 shown to be significantly better than for L3-L4. When individual vertebrae were examined L3 had the best precision although the differences with L2 and L4 were not significant. The study confirmed that supine lateral measurements of vertebral body BMD with baseline compensation give better precision than the decubitus method. The improvement makes it feasible to follow patients longitudinally and may allow the sensitivity of the lateral scan to supersede that of AP measurements.