Research Review By Dr. Josh Plener©


Download MP3

Date Posted:

July 2020

Study Title:

Lumbar muscle stiffness is different in individuals with low back pain than asymptomatic controls and is associated with pain and disability, but not common physical examination findings


Koppenhaver S, Gaffney E, Oates A et al.

Author's Affiliations:

Baylor University Doctoral Program in Physical Therapy; U.S. Army-Baylor University Doctoral Program in Physical Therapy, Texas, USA; University of New Brunswick, Canada

Publication Information:

Musculoskeletal Science and Practice 2020; 45: 102078.

Background Information:

There have been various studies examining the association between lumbar multifidus muscle impairments in relation to low back pain (LBP). Generally speaking, these studies have identified that LBP patients have multifidus muscles that are structurally and functionally deficient. For example, structurally there is atrophy and fatty infiltrate (1) and functionally there is delayed and attenuated contraction (2, 3).

A previous study by Masaki et al. found that for LBP patients, greater stiffness of the lumbar multifidus at rest was the strongest independent predictor of low back pain, with an odds ratio of 4.13. However, this study had a small sample size as their LBP group contained 9 subjects (4).

Masaki and colleagues used shear-wave elastography (SWE), which quantifies shear modulus. Shear modulus estimates the elasticity or stiffness of the soft tissue by measuring the propagation speed of ultrasound-induced shear-waves within the tissue. This method is superior to electromyography (EMG) and B-mode ultrasound imaging, which are the most commonly used ways to measure spinal muscle function. SWE is superior to EMG, which is more invasive – requiring needle electrodes (5), while B-mode ultrasound requires a measurement with the muscle at rest in order to determine the muscle thickness (during contraction). However, postural tone and muscle spasm do not allow for a valid ‘rest’ position with this modality (6). SWE, in contrast, is non-invasive and measures the biomechanical properties of muscle tissue without the need for comparisons to a resting value (7).

The purpose of this current study was to compare the stiffness of resting and contracted lumbar spine musculature in individuals with and without LBP. The secondary aim of this study was to explore the relationships between self-reported and physical examination measures with resting and contracted muscle stiffness in LBP individuals.

Pertinent Results:

120 subjects participated in this study – 60 with LBP and 60 without. Baseline characteristics between both groups were statistically equivalent except for BMI, which was higher in the LBP group (p = 0.01).

At rest, the shear modulus (stiffness) was greater for the erector spinae, which was 20% stiffer, and the lumbar multifidus, which was 10% stiffer, in LBP participants compared to asymptomatics. During submaximal lumbar multifidi contraction, there was no statistical difference between the groups.

Five self-reported measures were correlated to muscle stiffness:
  • Lumbar multifidi stiffness during contraction was correlated with higher activity level, lower LBP disability and lower BMI.
  • Lumbar multifidi stiffness at rest and during contraction was correlated with lower current pain intensity.
  • Erector spinae stiffness at rest was correlated with a subjective report of worsening LBP with bending forward.
Overall, no physical examination findings were correlated to muscle shear modulus. One possible reason for this is that the physical exam tests are dynamic, where SWE measures are static and therefore may assess motor function differently.

Clinical Application & Conclusions:

The results of this study demonstrate that there is greater muscle stiffness for resting erector spinae and lumbar multifidi muscles in LBP patients compared to asymptomatic subjects. These results were associated with pain and disability, but no physical exam findings were correlated. These findings support the clinical picture of muscle spasm or protective hypertonicity in LBP patients.

This study adds to the body of research regarding the association of multifidus muscle dysfunction and low back pain. However, further work needs to occur as studies thus far examining muscle stiffness only provide an associative, but no causative, link. Until this occurs, creating specific assessment and treatment recommendations based on this information remains speculative.

Study Methods:


60 LBP and 60 asymptomatic participants were included in this cross-sectional study.

Inclusion criteria:
  • Age between 18 and 65 years
  • Department of Defense beneficiaries (active duty military and civilian dependents)
  • Able to read and speak English
  • Able to lie prone and elevate their arms in order to be in the contracted muscle position for imaging
  • LBP patients had pain between the 12th rib and buttocks, with a minimum 10% score on the Oswestry Disability Index (ODI), signifying a minimal level of physical disability
Exclusion criteria:
  • Pregnancy
  • Reported history of lumbosacral surgery
  • Signs of lumbar radiculopathy and non-musculoskeletal pathology (i.e. cancer, cauda equina syndrome, infection)
Baseline Procedures

Baseline history and physical examination:
Baseline data was collected such as demographic data, activity levels and BMI. Low back pain participants additionally went through a standardized clinical examination including a brief history, self-reported pain on an 11-point NRS, and self-reported disability measured through an ODI. A physical examination was also carried out including:
  • Lumbosacral range of motion
  • Segmental mobility assessment through posterior to anterior (PA) palpation of L3-5 spinous processes.
  • Prone instability test to determine if less pain was reported with lumbar segmental posterior-anterior pressure when their legs were lifted in the air while prone, compared to when they were resting.
  • Multifidus lift test to assess lumbar multifidus muscle function at L3-5 by abducting their arm while prone to assess if pain was present in the contralateral lumbar multifidus muscles.
Experimental Procedure:

Participants laid prone with pillows under their pelvis and ankles to increase comfort and reduce their lumbar lordosis. Lumbar erector spinae muscles (illiocostalis and longissimus) for LBP participants were imaged at rest on a single side at a single level. This level and side were determined after the participant reported maximal tenderness following palpation of L3-L5. All asymptomatic participants were imaged on their right side at the L4 spinal level. The lumbar erector spinae muscles were imaged at the most prominent portion of the musculature immediately above the iliac crest at the selected spinal level. The transducer was oriented in the sagittal plane and placed parallel to the muscle fibres (8). The lumbar multifidus muscles were imaged at rest and during submaximal contraction. The submaximal contraction consisted of the participant abducting their left arm 5 cm high, for 5-10 seconds, while holding a weight proportional to their body weight. This contraction method results in approximately 30% maximal voluntary isometric contraction (9). The multifidi were imaged with the transducer moved just lateral to the L4/5 facet joint, parallel to the muscle.

The following sequence was performed three times with the average values recorded:
  1. Erector spinae imaged at rest
  2. Lumbar multifidus imaged at rest
  3. Lumbar multifidus imaged during submaximal contraction

Study Strengths / Weaknesses:

  • This study used SWE which is thought to be superior to other methods used to assess muscle stiffness (like ultrasound).
  • This study helps build on previous research with improved study methods (i.e. larger sample size than Masaki et al. 2017).
  • The SWE measurement technique is relatively new with regards to assessing muscle tissue and therefore errors or misinterpretations may result.
  • No direct causal conclusions can be made from this study, since this was cross-sectional in design.

Additional References:

  1. Goubert D, Oosterwijck JV, Meeus M et al. Structural changes of lumbar muscles in non-specific low back pain: a systematic review. Pain Physician 2016; 19: E985-E1000.
  2. Hungerford B, Gilleard W, Hodges P. Evidence of altered lumbopelvic muscle recruitment in the presence of sacroiliac joint pain. Spine 2003; 28: 1593-1600.
  3. Wallwork TL, Stanton WR, Freke M et al. The effect of chronic low back pain on size and contraction of the lumbar multifidus muscle. Man Ther 2009; 14: 496-500.
  4. Masaki M, Aoyama T, Murakami T et al. Association of low back pain with muscle stiffness and muscle mass of the lumbar back muscles, and sagittal spinal alignment in young and middle-aged medical workers. Clin Biomech 2017; 49: 128-133.
  5. Hug F, Hodges PW, Tucker K. Muscle force cannot be directly inferred from muscle activation: illustrated by the proposed imbalance of force between the vastus medialis and vastus lateralis in people with patellofemoral pain. J Orthop Sport Phys Ther 2015 45: 360-365.
  6. Koppenhaver S, Hebert J, Parent E et al. Rehabilitative ultrasound imaging is a valid measure of trunk muscle size and activation during most isometric sub-maximal contractions: a systematic review. Aust J Physiother 2009; 55: 153-169.
  7. Hug F, Tucker K, Gennisson JL et al. Elastography for muscle biomechanics: toward the estimation of individual muscle force. Exerc Sport Sci Rev 2015; 43: 125-133.
  8. Creze M, Timoh K, Gagey O et al. Feasibility assessment of shear wave elastography to lumbar back muscles: a Radioanatomic Study. Clin Anat N Y N 2017; 30: 774-780.
  9. Kiesel K, Uhl T, Underwood F et al. Measurement of lumbar multifidus muscle contraction with rehabilitative ultrasound imaging. Man Ther 2007; 12: 161-166.