The effect of thoracic kyphosis and sagittal plane alignment on vertebral compressive loading

Abstract

To better understand the biomechanical mechanisms underlying the association between hyperkyphosis of the thoracic spine and risk of vertebral fracture and other degenerative spinal pathology, we used a previously validated musculoskeletal model of the spine to determine how thoracic kyphosis angle and spinal posture affect vertebral compressive loading. We simulated an age-related increase in thoracic kyphosis (T(1) -T(12) Cobb angle 50-75 degrees) during two different activities (relaxed standing and standing with 5-kg weights in the hands) and three different posture conditions: (1) an increase in thoracic kyphosis with no postural adjustment (uncompensated posture); (2) an increase in thoracic kyphosis with a concomitant increase in pelvic tilt that maintains a stable center of mass and horizontal eye gaze (compensated posture); and (3) an increase in thoracic kyphosis with a concomitant increase in lumbar lordosis that also maintains a stable center of mass and horizontal eye gaze (congruent posture). For all posture conditions, compressive loading increased with increasing thoracic kyphosis, with loading increasing more in the thoracolumbar and lumbar regions than in the mid-thoracic region. Loading increased the most for the uncompensated posture, followed by the compensated posture, with the congruent posture almost completely mitigating any increases in loading with increased thoracic kyphosis. These findings indicate that both thoracic kyphosis and spinal posture influence vertebral loading during daily activities, implying that thoracic kyphosis measurements alone are not sufficient to characterize the impact of spinal curvature on vertebral loading.

Figure 1

Illustration demonstrating the sagittal spinal profiles associated with an uncompensated posture, a compensated posture (in this case tilting the pelvis posteriorly), and a congruent posture.

Figure 2

a) Sagittal view of the baseline spinal curvature and pelvic orientation used in the biomechanical model, showing the degree of thoracic kyphosis (TK), lumbar lordosis (LL), pelvic incidence (PI), and pelvic tilt (PT); b) cartoon of the first activity modeled: standing upright with the arms hanging straight down at the sides; c) cartoon of the second activity modeled: standing upright with the elbows flexed 90° and 5 kg weights in each hand. Both activities were modeled in 3D and were sagittally symmetric.

Figure 3

Schematic representation of the range of spinal curvatures modeled in this study, starting with a) uncompensated posture, b) compensated posture, and c) congruent posture. The T1-T12 Cobb angle was varied from 50° to 75° in 1° increments, but for clarity the images above show only the 5° increments with the extreme spinal curves labeled.

Figure 4

Compressive force (Newtons) on T8 and T12 as a function of T1-T12 Cobb angle (degrees) for the two activities, as well as the three different postures.

Figure 5

Increase in compressive loading for every 1° increase in the T1-T12 Cobb angle. These values are the slopes of least-squares linear regressions fitted to the load versus Cobb angle data for each vertebral body.

Figure 6

Increase in T12 compressive loading for every 1° increase in the T1-T12 Cobb angle. These values are the slopes of least-squares linear regressions fitted to the load versus Cobb angle data at T12 for each height/weight combination. Note that when the subject’s height was varied, subject weight was held constant at 71 kg (the 50^th percentile value for females) and when the subject’s weight was varied, subject height was held constant at 1.62 m (the 50^th percentile value for females).