Pillay, Julian DavidMurphy, BernadetteO'Connor, Laura Maié2026-08-112026-08-112025-08-25https://hdl.handle.net/10321/6453Submitted in fulfilment of the requirements of the degree of Doctor of Philosophy in Health Sciences at the Durban University of Technology, Durban, South Africa, 2026.Background Neck pain is a common disabling condition that can become chronic, leading to economic and personal burdens. A small pool of research highlights variations in motor control in neck pain, where alterations in spinal kinematic and/or muscle activation patterns, have been found during head, neck and unilateral arm tasks. Bilateral arm tasks have not been assessed, despite their common use in occupational settings and daily life. Few studies have simultaneously assessed spinal kinematics and muscle recruitment patterns in those with neck pain. The combined assessment allows a more holistic view of how neck pain may impact motor control which can lead to more effective interventions to target these processes. A common treatment for neck pain is spinal manipulative therapy (SMT). Although well utilised, the specific mechanism through which it exerts its reported clinical effects are not well understood. Mechanistic studies suggest that it activates neurophysiological processes which in addition to inducing pain relief activate reflexogenic, spinal and supraspinal structures involved in motor control. Little is known regarding the effect of SMT on spinal kinematics and muscle recruitment patterns and if changes in these motor variable correlate with pain and disability in those with neck pain. Thus, this study aimed to determine the effect of cervical SMT on pain, disability, cervicothoracic spinal kinematic and muscle activation patterns while performing a cervical range of motion (ROM), a unilateral and bilateral arm task when comparing to a control intervention. Methods Prior to the implementation of a randomised, controlled trial (RCT), a case controlled cross-sectional study was undertaken to determine the presence of cervicothoracic spinal kinematic and muscle activation patterns in adults with chronic, non-specific neck pain (n=20) compared to age and sex matched controls (n=20). Pain and disability measures and were obtained from those with neck pain; with health status being collected from all the participants. Spinal kinematics were obtained via motion capture and surface electromyography was utilised to obtain muscle recruitment patterns from six pairs of cervicothoracic muscles. Data were collected during seated cervical ROM, a seated unilateral and standing bilateral upper limb lifting task. Statistical analysis was performed with a p-value less than 0.05. A randomised, controlled trial (RCT) was then utilised to collect data from adults with chronic, non-specific neck pain (n=39), who were allocated to receive either cervical SMT or a control intervention. Four treatments were administered over 2 weeks with a one week follow up. Pain, disability and health status were measured. The same methodology as utilised in the case-controlled study was employed to obtain cervicothoracic spinal kinematics muscle recruitment patterns from six pairs of cervicothoracic muscles. Participants undertook 3 tasks: cervical range of motion, a unilateral and bilateral upper limb lifting task. Measurements were taken at baseline, the fourth visit and at a one-week follow-up. Statistical analysis was performed with a p-value less than 0.05. Results When comparing those with neck pain to those without, during cervical ROM the participants with neck pain took longer to complete right lateral flexion (p = 0,018, r = 0.35), no differences between the groups were found for peak displacement, peak and mean velocity and jerk (p ˃ 0.05). Those with neck pain had reduced left posterior cervical muscle activation time during cervical extension (p = 0.026, r = 0.34), peak activation during left and right away lateral flexion (p = 0.030, r = 0.32; p = 0.047, r = 0.29) and in the left anterior scalene during right rotation (p = 0.049, r = 0.29). Increased activation time was found in right lateral flexion in the left paraspinal muscle during the away and return phase (p = 0.042, r = 0. 0.27; p = 0.026, r = 0.34), for rotation in the left paraspinal on right rotation during the away and return phase (p = 0.042, p = 0.002), for peak and mean activation in the away and return phases (p ˂ 0.05, medium effect sizes). The unilateral arm task resulted in those with neck pain having less peak displacement in upper thoracic spine rotation (p = 0.017, r = 0.35), greater displacement in lateral flexion (p = 0.006, r = 0.40) and rotation (p = 0.038, r = 0.31) in the lower thoracic spine. During lifting and lowering those with neck pain had a greater lower thoracic spine rotation range (p = 0.036; p = 0.004). Muscle activation showed a greater percentage activation time in the right upper trapezius (UT) and left paraspinal at T4 (p = 0.039, r = 0.30; p = 0.039, r = 0.30) and lower peak activation in the left paraspinal at T4 (p = 0.032, r = 0.32) in lifting. During lowering the left paraspinal muscle at T4 had a greater activation time percentage (p = 0,005, r = 0.34), lower peak (p = 0.019, r = 0.42) and mean (p = 0.044, r = 0.24) activation in those with neck pain. The bilateral task showed greater cervical lateral flexion velocity (p = 0.024, r = 0.34) in those without neck pain, and an increased rotation range in the upper and lower thoracic spines (p = 0.024, r = 0.34; p = 0.016, r = 0.37) in those with neck pain while lifting. During lowering, those without neck pain had increased average velocity in cervical lateral flexion (p = 0.012, r = 0.34) and rotation (p = 0.022, r = 0.35). No significant differences between the groups were found for muscle activation patterns in the bilateral arm task. When comparing the effect of cervical SMT to a control group, those receiving SMT had greater pain reduction by visit four (p = 0.010; η2= 0.22), with no between group difference by the one-week follow up. The SMT had improved disability by one-week follow when compared to the control p = 0.006, η2= 0.24). At the fourth visit, the SMT group had an improvement in pain (p = 0.010, η2= 0.22), with no difference between the groups at the one week follow up (p = 0.340, η2=0.03). Greater improvements in disability occurred at the one week within the SMT group (p = 0.006, η2= 0.24). No between group differences were found for the global impression of change scale or level of satisfaction (p ˃ 0.05) between the groups, except for the SMT group reporting improved emotions (p = 0.001) and quality of life (p = 0.025). For cervical ROM assessment after one treatment, increased velocity in left rotation (p = 0.040, η2 = 0.119) and jerk in flexion (p ˂ 0.05), right (p = 0.023, η2 = 0.148) and left (p = 0.025, η2 = 0.144) lateral flexion and left rotation (p = 0.043; η2=0.115) was found in the SMT group. The control group had greater activation in the left posterior cervical (PC) (p = 0.012, η2 = 0.180) in flexion, left anterior scalene (AS) (p = 0.007, η2 = 0.215), left sternocleidomastoid (SCM) (p = 0.019, η2 = 0.164) in extension and the left paraspinal at T9 (LT9) during right lateral flexion (p = 0.049, η2 = 0.116). Less activation occurred in the control group for peak (p = 0.041, η2 = 0.128) and mean (p = 0.045, η2 = 0.123) activation in the LAS and the left LUT (p = 0.008, η2 = 0.192) in right lateral flexion; peak (p = 0.029, η2 = 0.145) and mean (p = 0.025, η2 = 0.151) activation in the right AS and in the LT9 for away time percentage (p = 0.019, η2 = 0.145) in left lateral flexion and the right PC (p = 0.024, η2 = 0.146), LT4 (p = 0.021, η2 = 0.147) and right SCM (p = 0.038, η2 = 0.127) for right rotation. After three treatments the SMT group had increased return peak jerk in flexion (p = 0.034, η2 = 0.146) and return peak (p = 0.028v) and away average jerk (p = 0.008, η2 = 0.191, η2 = 0.175) in left rotation. Muscle activation was less in the control group in the right AS during flexion (p = 0.024) and greater for right PC for peak (p = 0.034, η2 = 0.149) and mean activation (p = 0.002, η2 = 0.179) and LUT (p = 0.015, η2 =0.187) in extension, left AS in away (p = 0.048, η2 = 0.142) and return (p = 0.042, η2 = 0.042) right lateral flexion and in the LPC away (p = 0.029, η2 = 0.159) and mean (p = 0.029, η2 = 0.134) activation in left lateral flexion. The intervention group had greater activation in the left paraspinal at T4 during the return phase of flexion (p = 0.035, η2 = 0.149). When assessing the effect of the interventions on the unilateral arm task there were no differences between the groups for spinal kinematics following one treatment (p ˃ 0.05), with greater peak (p = 0.038, η2 = 0.13) and mean (p = 0.031, η2 =0.14) muscle activity in the right paraspinal muscle in the intervention group during the lowering phase. At the fourth visit the intervention group had a greater upper thoracic spine rotation range (p = 0.021, η2 = 0.17) during lifting, and increased flexion/extension range in the lower thoracic spine (p = 0.037, η2 = 0.141) during lowering. The right upper trapezius muscles had greater mean activity in the intervention group (p = 0.026, η2 = 0.141) during lifting. In the bilateral arm task, greater flexion/extension jerk occurred in the lower thoracic spine (p = 0.038, η2 = 0.13) during lowering in the control group, with the intervention group having greater jerk in lateral flexion (p = 0.014, η2 = 0.13) and rotation (p = 0.019, η2 = 0.17) and greater rotation range (p = 0.041, η2 = 0.13) and velocity (p = 0.006, η2 = 0.22) in the lower thoracic spine. The intervention group had less peak activity in the right PC muscle during lifting (p = 0.049, η2 = 0.119), with greater peak activation in the right paraspinal T4 (p = 0.026, η2 = 0.151) and activation time percentage in the right SCM during lifting (p = 0.029, η2 = 0.148) and lowering (p = 0.015, η2 = 0.186). At the fourth visit, the intervention group had greater upper thoracic spine flexion/extension jerk (p = 0.045, η2 = 0.14) during lowering, with lower left PC peak muscle activation during lifting (p = 0.029, η2 = 0.16) and mean activation during lowering (p = 0.006, η2 = 0.237). Mean activation of the left (p = 0.039, η2 = 0.154) and right (p = 0.034, η2 = 0.157) SCMs during the lowering phase was greater in the intervention group. No differences were found between those receiving SMT and the control for perceived global impression of change categories (p=0.500) or level of satisfaction (p ˃ 0.05), except that the SMT group reported improved emotions (p = 0.001) and quality of life (p = 0.025). Conclusion People with neck pain presented with an altered pattern of motor control reflected by changes in cervicothoracic spinal kinematic and muscle recruitment patterns. This presentation was influenced by the activity, which was undertaken, often involving the thoracic spine more than the cervical spine. When treated with cervical SMT there was a decrease in pain and disability with increased rotation velocity and jerk in flexion, lateral flexion and left rotation immediately after one treatment and in the short term when performing cervical ROM. This was associated with increased activation patterns of cervicothoracic muscles during lateral flexion and rotation after one treatment with lower activation levels occurring after three treatments. In the upper limb tasks, the intervention effects were few and variable. This study showed that cervical SMT had a more favourable impact on muscle recruitment patterns than spinal kinematics. Highlighting the potential role of cervical SMT to bring about changes in motor control.794 penSpinal manipulationNeck painCervical spineMotion captureSurface electromyographyMuscle recruitmentRange of motionFunctional taskThesishttps://doi.org/10.51415/10321/6453