DXA-generated Z-scores and T-scores may differ substantially and significantly in young adults

Abstract

Central dual-energy X-ray absorptiometry (DXA) is the gold standard for non-invasive measurement of bone mineral density (BMD). Using this value and subject demographics, DXA software calculates T-scores and Z-scores. Professional society guidelines for the management of osteoporosis are based on T-scores and Z-scores, rather than on the actual BMD value. Although one expects T-scores and Z-scores to be very similar in young men and women for any given BMD measurement, little literature exists on this issue. Our clinical experience shows that some younger adult individuals (premenopausal women and men younger than 50 yr) have larger than expected difference between their DXA T-score and Z-score. This cross-sectional study evaluates the extent of this discordance between Z-scores and T-scores in a sample of 4275 men and women aged 20-49 yr. All subjects were scanned by central DXA using equipment manufactured by GE Lunar, GE, Madison, WI, or Hologic, Inc., Bedford, MA. Significant differences between Z-scores and T-scores were seen within individuals at the lumbar spine, total hip, femoral neck, and trochanter (p value<0.001) for both DXA systems. Although these differences were less than half a standard deviation (SD) in most instances, the magnitude of difference was substantial at times, being 1 or more SD in up to 11% of cases (range: -1.95 to +1.54 SD). The smallest differences were seen at the total hip and the largest differences were seen at the femoral neck for both technologies. This is in part because there is no single standard Z-score definition, resulting in different methods of calculation across, and even within, DXA manufacturers. Standardization of Z-score definition and method of calculation is indicated. DXA Z-scores should be interpreted with caution in men and women aged 20-50 yr.