Research Review By Dr. Brynne Stainsby©


Download MP3

Date Posted:

June 2019

Study Title:

Changes in spinal stiffness with chronic thoracic pain: Correlation with pain and muscle activity


Pagé I, Nougarou F, Lardon A & Descarreaux M

Author's Affiliations:

Departments of Anatomy, Electrical and Computer Engineering, Human Kinetics – all at the Université du Québec à Trois-Rivières, Quebec, Canada.

Publication Information:

PLoS ONE 2018; 13(12): e0208790.

Background Information:

Spinal pain is a very common musculoskeletal condition, with a one-year prevalence of up to 43% in the general population (1, 2). Spinal manipulative therapy (SMT) is an evidence-based treatment for spinal pain (3) and is often used based on a clinician’s ability to identify a ‘dysfunctional’ spinal level using static or in-motion palpation to assess movement (3-5). Typically, clinicians assess for the quality of motion and the patient’s pain response, while also potentially comparing to past experience assessing the same spinal level in other patients (or, the same patient) (4). Several factors have been shown to influence the clinician’s sensations or tissue behaviour during this assessment (such as patient anthropometrics, position, breathing, contact area, applied load, velocity and angulation), which may explain the conflicting evidence on the reliability and validity of palpation (5, 6). Mechanical devices allow for the standardization of assessment of spinal stiffness by gradually applying a specific load over a targeted spinous process with a controlled velocity (5).

Two studies have demonstrated an association between an immediate decrease in lumbar spine stiffness following lumbopelvic SMT and a clinically significant improvement in disability (7, 8); however, the effect on thoracic spine pain (TSP) remains unknown. As such, the main objective of this study was to compare thoracic spinal stiffness between healthy subjects and those with chronic TSP, and to evaluate the association between spinal stiffness and both muscle activity and pain intensity, in order to attempt to better understand the mechanisms by which spinal stiffness could be modulated. Finally, the within- and between-day reliability was evaluated in a subgroup of participants.

Pertinent Results:

  • Healthy subjects and those with chronic thoracic spine pain (TSP) were similar regarding age, weight, height and BMI at baseline. Those with chronic TSP reported significantly higher scores on the self-reported clinical outcomes (p < 0.01). For those subjects who completed two sessions, there was a significant decrease (p = 0.01) between the two sessions, however this was not clinically meaningful.
  • Reliability analyses suggest “good” to “excellent” within- and between-day reliability in all subjects, with the exception of between-day reliability in healthy participants at T5.
  • A significant main effect of groups for both global and terminal spinal stiffness coefficients was observed; specifically, lower spinal stiffness was observed in those subjects with chronic TSP (this is opposite to what may have expected!). Significant main effect of spinal levels for the global spinal stiffness was also observed with Bonferroni post hoc test revealing that T5 was less stiff than T7, though this difference was not observed for the terminal stiffness.
  • Muscle activity level was not significantly correlated to the terminal and global spinal stiffness.
  • All subjects could tolerate the 45N load, though a statistically higher proportion of those with chronic TSP reported pain during the assessment of spinal stiffness: At T5, T6, T7 and T8 respectively – 13%, 24%, 14% and 10% of healthy participants reported pain, compared to 89%, 90%, 82% and 73% of those with chronic TSP.

Clinical Application & Conclusions:

This study is the first to measure spinal stiffness at multiple levels of the thoracic spine and the results demonstrate a decrease in stiffness in participants with chronic thoracic spine pain (TSP). This contrasts the commonly-held assumption that chronic TSP is associated with increased stiffness. It is also interesting to note, the findings of this study do not support the hypothesis of an association between spinal stiffness and muscle contraction due to pain during the assessment of stiffness. Although pain intensity was greater in subjects with TSP, it was only significantly associated with stiffness at T6.

The authors caution that clinicians should not ‘look for’ a stiff or hypomobile spine, and should also consider the possibility of increased mobility as a mechanism for TSP. More research is required to determine the value and clinical utility of instrumented spinal stiffness assessment in the evaluation and management of patients with chronic spinal pain.

Study Methods:

  • Participants between the ages of 18-60 years were recruited for the study. Twenty-five healthy subjects (with no significant thoracic pain in the past year) and 50 with chronic thoracic spine pain (TSP) for at least three months within the T5-T8 region were recruited. Those who were pregnant or those diagnosed non-spine-related conditions that can refer to the chest wall, inflammatory or neuromuscular disease, myelopathy, malignancy, infection, scoliosis, advanced osteoporosis or osteoarthritis, neurological symptoms or uncontrolled hypertension were excluded.
  • To assess within- and between-day reliability of the spinal stiffness assessment, all healthy and the first 25 subjects with chronic TSP underwent two identical experimental sessions within 24-48 hours. The remaining 25 subjects with TSP participated in the first session only. Subjects were asked to complete the Visual Analog Scale (VAS), the Quebec Back Pain Disability Questionnaire (QBPDQ), the Tampa Scale of Kinesiophobia (TSK) and the STarT Back Screening Tool (SBST). Following an interview and physical examination with an experienced clinician, subjects lay on a treatment table in the prone position for 30 minutes.
  • Surface electromyography (sEMG) electrodes were applied bilaterally at approximately 2 cm from the spine in line with the muscle fibers of the thoracic erector spinae muscles at the level of the T5 and T8 spinous processes (SP) and between the T6 and T7 SP (after preparing the skin) in order to assess muscle activity during the assessment of spinal stiffness. All sEMG were recorded at 2000 Hz using two differential EMG inputs with two stabilizing references. Participants relaxed quietly while sEMG activity was recorded for 4s, and the signal recorded by each electrode during this time was subsequently used to normalize the respective electrode signal during the spinal stiffness assessments.
  • An a priori randomization was used to determine the order of spinal level testing for each participant (T5, T6, T7, T8). Spinal stiffness was assessed four times at each level and sEMG was concurrently recorded during each assessment. Spinal stiffness was measured by an apparatus with a servo-controlled linear actuator motor and an indenter (with a round, padded rod) positioned directly over a spinous process, with a constant rate of force application of 18 N/s from 5N to a peak force of 45N (which was maintained for 1s). The subjects were instructed to hold his/her breath at the end of exhalation while the total of 45N was gradually applied. Each assessment was separated by at least 45s. Immediately following each assessment, subjects were asked to rate mean pain intensity perceived during the trial (0-100 VAS).
  • The bipolar sEMG signals were digitally band-pass filtered in the 40-400 Hz frequency bandwidth. After computing and normalizing root mean square (nRMS) values for each electrode for a 1s window during spinal stiffness assessment, the mean value of the closest right and left electrodes (to the level of assessed spine stiffness) was used to calculate muscle activity.
  • Global and terminal spinal stiffness coefficients were calculated using the force-displacement data of each spinal stiffness assessment.
Differences in baseline characteristics and clinical status were assessed. To assess spinal stiffness measurement reliability using data from all healthy subjects and the first 25 with chronic TSP, within- and between-day reliability were determined using an intraclass correlation coefficient (ICC). To study the differences in spinal stiffness between groups and between spinal levels, mixed model analyses of variance (ANOVA) were independently computed for global and terminal stiffness coefficients. Between-group differences regarding muscle activity during assessments were determined using a 2(groups) x 4(levels) mixed-model ANOVA. Post-hoc tests using Bonferroni correction were computed for significant effects. Between-group comparisons for pain intensity were performed by Fisher’s exact test and descriptive analyses. Association between spinal stiffness and both muscle activity and pain intensity were studied with Pearson’s correlation or its estimated value from the Kendall tau rank correlation coefficient.

Study Strengths / Weaknesses:

  • Though preliminary in nature, this study quantified spinal stiffness at T5-T8 spinal segments in a healthy population and those with chronic TSP.
  • This study examined muscle activation and pain intensity during the assessment of spinal stiffness to inform clinical hypotheses regarding these potential relationships.
  • Forces were applied in a calibrated and reproducible manner using a standardized testing protocol.
  • The methodology in the study appears to be appropriate for between- and within-day comparisons, and this should inform future research.
  • The greatest limitation of this study is its preliminary nature, as clinical relevance cannot be determined. It is important to note however that this work provides a critical foundation for future research.
  • The minimal difference representing a clinically meaningful change in spinal stiffness has not been established, and again, this limits the clinical relevance of this study.
  • This study specifically measured stiffness from T5-8, however, future studies may assess the entire thoracic spine for a more complete picture of thoracic spinal stiffness.
  • Although sEMG was used to capture the activity of superficial muscles, the myoelectric activity of deep muscles could not be assessed in this way. Future research may use needle EMG to study the activation of deeper muscles.
  • Given the set up of the apparatus, the angulation could not be modified, and thus, some measurements may not have been perpendicular to the vertebra, which has been reported to decrease spinal stiffness (such as T5 being less stiff than T7).
  • Although done for patient safety, generalizability is limited by the exclusion of patients with pathology, more specifically, osteoarthritis.

Additional References:

  1. Leboeuf-Yde C, Nielsen J, Kyvik KO et al. Pain in the lumbar, thoracic or cervical regions: do age and gender matter? A population-based study of 34,902 Danish twins 20–71 years of age. BMC Musculoskelet Disord 2009; 10: 39.
  2. Briggs A, Smith A, Straker L et al. Thoracic spine pain in the general population: Prevalence, incidence and associated factors in children, adolescents and adults. A systematic review. BMC Musculoskeletal Disorders 2009; 10(1): 77.
  3. Hurwitz EL. Epidemiology: spinal manipulation utilization. J Electromyogr Kinesiol 2012; 22(5): 648–654.
  4. Abbott JH, Flynn TW, Fritz JM et al. Manual physical assessment of spinalsegmental motion: Intent and validity. Manual Therapy 2009; 14(1): 36–44.
  5. Wong AYL, Kawchuk GN. The clinical value of assessing lumbar posteroanterior segmental stiffness: A narrative review of manual and instrumented methods. PM&R 2017; 9(8): 816-830.
  6. Snodgrass SJ, Haskins R, Rivett DA. A structured review of spinal stiffness as a kinesiological outcome of manipulation: its measurement and utility in diagnosis, prognosis and treatment decision-making. J Electromyog Kinesiol 2012; 22(5):708–23.
  7. Fritz JM, Koppenhaver SL, Kawchuk GN et al. Preliminary investigation of the mechanisms underlying the effects of manipulation: exploration of a multivariate model including spinal stiffness, multifidus recruitment, and clinical findings. Spine 2011; 36(21): 1772–1781.
  8. Wong AY, Parent EC, Dhillon SS et al. Do participants with low back pain who respond to spinal manipulative therapy differ biomechanically from nonresponders, untreated controls or asymptomatic controls? Spine 2015; 40(17): 1329–1337.