Surface electromyography study on asymmetry in paravertebral muscle degeneration in patients with degenerative lumbar scoliosis

Abstract

The asymmetry of paravertebral muscle (PVM) degeneration in degenerative lumbar scoliosis (DLS) patients has been extensively studied by imaging and histological examination and has not yet been verified by surface electromyography (sEMG) techniques. To study the relationship between the surface electromyography (sEMG) and degenerative characteristics of paravertebral muscles (PVMs) in patients with degenerative lumbar scoliosis (DLS). In twenty DLS patients and fifteen healthy subjects, sEMG activity of the PVMs at the level of the upper end vertebra (UEV), apical vertebra (AV) and lower end vertebra (LEV) was measured during static standing and dynamic standing forward flexion and backward extension tasks. Action segmentation was achieved according to inertial measurement unit (IMU) data. The sEMG characteristics of the PVMs on the convex and concave sides were compared, and the relationship of these data with the Cobb angle and lumbar lordotic angle (LL) was analyzed. In the DLS group, there was no difference in sEMG activity between the convex and concave sides at the UEV or AV level, but in the motion and return phases of the standing forward flexion task (P = 0.000, P = 0.015) and the maintenance and return phases of the standing backward extension task (P = 0.001, P = 0.01), there was a significant difference in sEMG activity between the convex and concave sides at the LEV level. Asymmetrical sEMG activity at the LEV level was negatively correlated with the Cobb angle (F = 93.791, P < 0.001) and LL angle (F = 65.564, P < 0.001). In the DLS group, asymmetrical sEMG activity of the PVMs appeared at the LEV level, with the concave side being more active than the convex side. This sEMG characteristics were consistent with their imaging and histological degenerative features and correlated with bone structural parameters.

Figure 1

Male, 75 years, DLS with left-sided convexity. (A) The UEV was the L1 vertebra, the AV was the L3 vertebra, the LEV was the L5 vertebra, and the Cobb angle was 14.3°. (B) The LL angle was 44.9°.

Figure 2

Six bipolar electrodes were placed at three levels of the bilateral PVMs (UEV, AV and LEV). After confirmation by radiographs, the sEMG sensors and IMUs were placed in the appropriate areas. (a) Three sEMG sensors on the right. (b) Three sEMG sensors on the left, each integrated with an IMU, placed on the corresponding position of the spinous process (L1, L3 and L5). (c) Three axes of the IMU.

Figure 3

A subject performs movement tasks. (A) Standing forward flexion task. (B) Standing backward extension task. (a) Static standing phase. (b) Motion phase of the standing flexion/extension tasks. (c) Maintenance phase of the standing flexion/extension tasks. (d) Return phase of the standing flexion/extension tasks.

Figure 4

Typical images of the three sets of signals collected with the IMU after processing: acceleration along the vertical axis (x-axis), horizontal axis (y-axis), and anterior–posterior axis (z-axis). (A) Standing forward flexion task. (B) Standing backward extension task. (a) Static standing phase. (b) Motion phase of the standing flexion/extension tasks. (c) Maintenance phase of the standing flexion/extension tasks. (d) Return phase of the standing flexion/extension tasks. The x-axis is the data collection frequency (Hz); the y-axis is the acceleration of the movement (mg).

Figure 5

Typical sEMG signal of the subject's PVMs after pretreatment on the convex/concave sides at the three measurement levels. (A) Standing forward flexion task in DLS group. (B) Standing backward extension task in DLS group. The x axis is the sampling time (seconds), and the y axis is the EMG amplitude (uV).

Figure 6

sEMG activity of PVMs at different levels in the control and DLS groups during the static standing phase.

Figure 7

sEMG activity of PVMs at different levels in the control and DLS groups in the standing forward flexion task, *Significant difference in the sEMG activity of PVMs between the convex and concave sides.

Figure 8

sEMG activity of PVMs at different levels in the control and DLS groups in the standing backward extension task. *Significant difference in sEMG activity of PVMs between the convex and concave sides.

Figure 9

sEMG activity of PVMs at the LEV level in the DLS group in all movement tasks. *Significant difference in sEMG activity of PVMs between the convex and concave sides.

Figure 10

Correlation of the convex/concave sEMG ratio (sEMG activity on the convex side divided by sEMG activity on the concave side) at the LEV level and the Cobb angle in the DLS group during the differential movement phase.

Figure 11

Correlation of the convex/concave sEMG ratio at the LEV level and the LL angle in the DLS group during the differential movement phase.