Effect of Parathyroid Hormone Combined With Gait Training on Bone Density and Bone Architecture in People With Chronic Spinal Cord Injury

Objective: To evaluate the response of bone to 2 anabolic stimuli, teriparatide and mechanical loading, in subjects with spinal cord injury. Design: A pilot study, 1 group, pretest-posttest. Setting: A rehabilitation hospital. Participants: A convenience sample of 12 nonambulatory chronic spinal cord injury subjects. Methods: The subjects were administered open-label teriparatide 20 mu g/d while undergoing robotic-assisted stepping 3 times a week for 6 months, followed by 6 months of teriparatide alone. Main Outcome Measurements: Bone status was evaluated at 3, 6, and 12 months by using dual-energy x-ray absorptiometry to calculate bone mineral density (BMD) at the spine and hip, magnetic resonance imaging to assess bone microarchitecture of the distal tibia, and serum bone markers. Results: Mean (SD) baseline BMD measurements at the spine and the left and right total hip were 1.05 +/- 0.162 g/cm(2), 0.638 +/- 0.090 g/cm(2) and 0.626 +/- 0.088 g/cm(2), respectively. After 6 months of treatment, BMD changed 2.19% +/- 3.61%, 0.02% +/- 2.21%, and 0.74% +/- 2.80% at the spine, and left and right total hip, respectively. These changes were not statistically significant (P > .05 for all). Magnetic resonance imaging supported an anabolic effect after 3 months of treatment with significant (P < .05) changes in trabecular thickness, 4.4% +/- 4.06%; surface-to-curve ratio, 23.6% +/- 22.3%; and erosion index, -17.04% +/- 12.9%. Although the trend remained after 6 months, statistical significance was not retained. At 6 months, bone markers indicated an increase in mean levels of bone-specific alkaline phosphatase, 53.8% +/- 62.9%; C-terminal telopeptides of type I collagen, 137.6% +/- 194.6%; and intact aminoterminal propeptide of type I procollagen, 61.4% +/- 99.3%. Conclusion: In this limited pilot study, teriparatide and mechanical loading resulted in a numerical but not statistically significant increase in lumbar spine BMD and no significant BMD changes at the hip. Magnetic resonance imaging at the distal tibia suggested an anabolic effect, but the high sensitivity offered by this technique was challenged by the limited ability to obtain analyzable data from all the subjects. Further studies that involve longer treatment periods and greater mechanical loading are warranted.