Measuring Apparent Trabecular Density and Bone Structure Using Peripheral Quantitative Computed Tomography at the Tibia: Precision in Participants With and Without Spinal Cord Injury

The objective of the study was to investigate the precision of standard outcomes obtained using peripheral quantitative computed tomography as well as apparent trabecular structure measures in adults with and without spinal cord injury (SCI). Twelve individuals with SCI, mean (standard deviation [SD]) 20 (13) yrs postinjury and mean (SD) age 44 (9) yrs, and 21 individuals without SCI (mean [SD] age: 27 [5] yrs) participated. Repeat scans of tibia epiphysis (4%) and shaft (66%) were performed using a Stratec XCT-2000 (Stratec Medizintechnik, Pforzheim, Germany). Bone mineral density and geometry variables (e.g., cortical thickness, bone area, polar moment of inertia) were derived with manufacturer's software. The following apparent trabecular structure variables were determined using custom software: average trabecular thickness (TrTh) (mm), trabecular spacing (TrSp) (mm), and trabecular number (TrN) (1/mm); average hole size (HA) and maximum hole size (H-M) (mm(2)); connectivity index (CI); cortical thickness (CTh) (mm); bone volume to total volume (BVTV) ratio. Root mean square standard deviation and root mean square coefficient of variation (RMSCV; root mean square coefficient of variation percent [RMSCV%]) were calculated. The RMSCV% for all standard bone mineral density and geometry variables was <= 2% except for total area (4% site), where precision was 3.8%. RMSCV% for bone structure variables were as follows: CTh 5.1, TrTh 1.7, TrN 1.9, TrSp 2.6, HA 9.5, HM 20.1, CI 5.1, and BVTV 1.4. Precision for bone density and geometry was excellent across a range of bone mineral densities. RMSCVs for some apparent trabecular structure variables were comparable to that of standard variables. The RMSCV for others may necessitate larger studies to detect between-group differences.