Original Article National Health and Nutrition Examination Survey Whole-Body Dual-Energy X-Ray Absorptiometry Reference Data for GE Lunar Systems Bo Fan, * ,1 John A. Shepherd, 1 Michael A. Levine, 2 Dee Steinberg, 1 Wynn Wacker, 3 Howard S. Barden, 3 David Ergun, 3 and Xin P. Wu 4 1 Department of Radiology and Bioimaging, University of California San Francisco, San Francisco, CA, USA; 2 Division of Endocrinology and Diabetes, The Children’s Hospital of Philadelphia, Philadelphia, PA, USA and Department of Pediatrics, University of Pennsylvania Perelman School of Medicine, Philadelphia, PA, USA; 3 GE-Lunar, Madison, WI, USA; and 4 Institute of Metabolism and Endocrinology, Second Xiang-Ya Hospital, Central South University, Changsha, China Abstract The National Health and Nutrition Examination Survey (NHANES 1999e2004) includes adult and pediatric comparisons for total body bone and body composition results. Because dual-energy x-ray absorptiometry (DXA) measurements from different manufacturers are not standardized, NHANES reference values currently are applica- ble only to a single make and model of Hologic DXA system. The purpose of this study was to derive body com- position reference curves for GE Healthcare Lunar DXA systems. Published values from the NHANES 1999e2004 survey were acquired from the Centers for Disease Control and Prevention website. Using previously reported cross- calibration equations between Hologic and GE-Lunar, we converted the total body and regional bone and soft-tissue measurements from NHANES 1999e2004 to GE-Lunar values. The LMS (LmsChartMaker Pro Version 3.5) curve fitting method was used to generate GE-Lunar reference curves. Separate curves were generated for each sex and ethnicity. The reference curves were also divided into pediatric (20 years old) and adult (O20 years old) groups. Adult reference curves were derived as a function of age. Additional relationships of pediatric DXA values were derived as a function of height, lean mass, and bone area. Robustness was tested between Hologic and GE-Lunar Z-score values. The NHANES 1999e2004 survey included a sample of 20,672 participants’ (9630 female) DXA scans. A total of 8056 participants were younger than 20 yr and were included in the pediatric reference data set. Participants enrolled in the study who weighed more than 136 kg (over scanner table limit) were excluded. The average Z-scores comparing the new GE-Lunar reference curves are close to zero, and the standard deviation of the Z-scores are close to one for all variables. As expected, all measurements on the GE-Lunar reference curves for participants younger than 20 yr increase monotonically with age. In the adult population, most of the curves are constant at younger age and drop moderately as age increases. We have presented NHANES reference curves ap- plicable to DXA whole-body scans acquired on GE Healthcare Lunar systems by age, sex and ethnicity. Users of GE Healthcare GE-Lunar DXA systems can now benefit from the large body composition reference data set collected in the NHANES 1999e2004 study. Key Words: Body composition; bone density; reference curve; whole-body DXA; Z-score. Introduction Dual-energy X-ray absorptiometry (DXA) is currently a criterion method to measure bone mineral density (BMD) and whole-body composition (% Fat, fat mass, muscle Received 07/31/13; Accepted 08/14/13. *Address correspondence to: Bo Fan, MD, Department of Radi- ology and Bioimaging, University of California San Francisco, 1 Irv- ing Street, MRSC Room 109, San Francisco, CA 94143. E-mail: bo.fan@ucsf.edu 344 Journal of Clinical Densitometry: Assessment & Management of Musculoskeletal Health, vol. 17, no. 3, 344e377, 2014 Ó Copyright 2014 by The International Society for Clinical Densitometry 1094-6950/17:344e377/$36.00 http://dx.doi.org/10.1016/j.jocd.2013.08.019