High dose compressive loads attenuate bone mineral loss in humans with spinal cord injury

Abstract

People with spinal cord injury (SCI) lose bone and muscle integrity after their injury. Early doses of stress, applied through electrically induced muscle contractions, preserved bone density at high-risk sites. Appropriately prescribed stress early after the injury may be an important consideration to prevent bone loss after SCI.

Introduction: Skeletal muscle force can deliver high compressive loads to bones of people with spinal cord injury (SCI). The effective osteogenic dose of load for the distal femur, a chief site of fracture, is unknown. The purpose of this study is to compare three doses of bone compressive loads at the distal femur in individuals with complete SCI who receive a novel stand training intervention.

Methods: Seven participants performed unilateral quadriceps stimulation in supported stance [150% body weight (BW) compressive load-"High Dose" while opposite leg received 40% BW-"Low Dose"]. Five participants stood passively without applying quadriceps electrical stimulation to either leg (40% BW load-"Low Dose"). Fifteen participants performed no standing (0% BW load-"Untrained") and 14 individuals without SCI provided normative data. Participants underwent bone mineral density (BMD) assessment between one and six times over a 3-year training protocol.

Results: BMD for the High Dose group significantly exceeded BMD for both the Low Dose and the Untrained groups (p < 0.05). No significant difference existed between the Low Dose and Untrained groups (p > 0.05), indicating that BMD for participants performing passive stance did not differ from individuals who performed no standing. High-resolution CT imaging of one High Dose participant revealed 86% higher BMD and 67% higher trabecular width in the High Dose limb.

Conclusion: Over 3 years of training, 150% BW compressive load in upright stance significantly attenuated BMD decline when compared to passive standing or to no standing. High-resolution CT indicated that trabecular architecture was preserved by the 150% BW dose of load.

Fig. 1

Schematic representation of the standing systems (left pair), and free body diagrams of the lower body during passive and active stance (right pair). In the stance models, each limb bears one half of the weight of the head, arm, and trunk segments (½W_HAT) plus the weight of the thigh (W_TH) and the weight of the shank (W_SH). During standing, thigh angle (φ)=71°, shank angle (α)=83°, and belt angle (β)=4°. The internal forces from the quadriceps and patellar tendons (F_quad and F_pat, respectively) in the active model replace the force of the supporting knee pad (F_pad) in the passive model

Fig. 2

Quadriceps adaptations to High Dose training. One-way ANOVA revealed a significant effect of training duration for quadriceps force (p=0.003), fatigue index (p=0.015), and for femur compressive load (p=0.003). *Significantly different from time bin 1 (0.25 year) (p=0.044)

Fig. 3

Longitudinal peripheral quantitative computed tomography (pQCT)-derived bone mineral density (BMD) for the distal femur (a), proximal tibia (b), and distal tibia (c). Untrained and Low Dose limbs are plotted together

Fig. 4

Longitudinal peripheral quantitative computed tomography (pQCT)-derived bone mineral density (BMD) at the year 1 and year 3 measurement points for the distal femur (a), proximal tibia (b), and distal tibia (c). NON non-spinal cord injury (SCI), AC acute SCI (<3 months), HI High Dose cohort, LO Low Dose cohort, UN Untrained cohort. *Significantly greater than LO and UN (p<0.019)

Fig. 5

Representative example of CT images from a subject in the High Dose cohort. Images in the left-hand column are from the subject’s untrained limb. Images from the right-hand column are from the subject’s trained limb. a, b Cross-sectional CT images at the 12% femur sites. c, d Cylindrical regions of interest within each medial femoral condyle. e, f Three-dimensional reconstruction of the trabecular lattice within each region of interest. The color scale corresponds to voxels assigned as plates and rods during volumetric topologic analysis (VTA). The maximum and minimum dimensions of plates and rods are listed

Fig. 6

CT-derived bone mineral density (BMD) (top row), trabecular surface width (middle row), and trabecular surface-to-curve ratio (bottom row) from a subject in the High Dose cohort