Generation of a Reference Dataset to Permit the Calculation of T-scores at the Distal Femur and Proximal Tibia in Persons with Spinal Cord Injury

Abstract

Persons with traumatic spinal cord injury (SCI) have severe bone loss below the level of lesion with the distal femur (DF) and proximal tibia (PT) being the skeletal regions having the highest risk of fracture. While a reference areal bone mineral density (aBMD) database is available at the total hip (TH) using the combined National Health and Nutrition Examination Survey (NHANES) III study and General Electric (GE) combined (GE/NHANES) to calculate T-score (T-scoreGE/NHANES), no such reference database exists for aBMD of the DF, and PT. The primary objectives of this study were (1) to create a reference dataset of young-healthy able-bodied (YHAB) persons to calculate T-score (T-scoreYHAB) values at the DF and PT, (2) to explore the impact of time since injury (TSI) on relative bone loss in the DF and PT regions using the two computation models to determine T-score values, and (3) to determine agreement between T-score values for a cohort of persons with SCI using the (T-scoreYHAB) and (T-scoreGE/NHANES) reference datasets. A cross-sectional prospective data collection study. A Department of Veterans Affairs Medical Center and a Private Rehabilitation Hospital. A normative reference aBMD database at the DF and PT was collected in 32 male and 32 female Caucasian YHAB participants (n=64) and then applied to calculate T-score values at the DF and PT in 105 SCI participants from a historical cohort. The SCI participants were then grouped based on TSI epochs (E-I: TSI < 1y, E-II: TSI 1–5y, E-III: TSI 6–10y, E-IV: TSI 11–20y, E-V: TSI > 20y). N/A. The knee and hip aBMD values were obtained by dual energy X-ray absorptiometry (GE Lunar iDXA) using standard clinical software for proximal femur orthopedic knee software applications. There were no significant differences in mean aBMD values across the four YHAB age subgroups (21–25, 26–30, 31–35, and 36–40 yr of age) at the TH, DF, and PT; mean aBMD values were higher in men compared to the women at all skeletal regions of interest. Using the mean YHAB aBMD values to calculate T-score values at each TSI epoch for persons with SCI, T-score values decreased as a function of TSI, and they continued to decline for 11–20 yr. Moderate kappa agreement was noted between the YHAB and the GE/NHANES reference datasets for the T-score cutoff criteria accepted to diagnose osteoporosis (i.e., SD <-2.5). A homogeneous reference dataset of YHAB aBMD values at the DF and PT was applied to calculate T-score values in persons with chronic SCI. There was a moderate level of agreement at the TH between the YHAB and GE/NHANES reference datasets when applying the conventional T-score cutoff value for the diagnosis of osteoporosis.