MRI-Based Assessment of Trunk and Hip Muscle Morphology and Strength in Chronic Low Back Pain

Introduction

Low back pain (LBP) syndrome is a significant health condition that compromises overall health status and diminishes quality of life. Debate continues as to whether LBP should be classified as a disease entity or merely a symptom. Several well-established facts regarding LBP warrant attention. It is known that nonspecific LBP, or lumbago, a musculoskeletal disorder, develops in nearly 80% of individuals during their lifetime, often resulting in reduced capacity to perform physical tasks [1].

Epidemiological studies show that LBP is more common in women and individuals aged 40 to 69 years, with higher prevalence in affluent countries and a positive correlation with the Human Development Index [2,3]. LBP is the leading cause of disability worldwide. Only 58% of affected individuals seek medical care, with women, those with recurrent episodes, and individuals with multiple health conditions more likely to seek help. Most episodes resolve within a few months, but up to 10% progress to chronic pain. Since LBP affects mainly the working-age population, it significantly increases economic and public health burdens [4–6].

The muscles included in the hip joint, examined in this study, play an important postural role. These muscles maintain an upright posture; for example, the iliopsosas muscle increases lumbar lordosis and indirectly influences thoracic kyphosis. Additionally, it stabilizes the lumbar spine, thereafter the hip joint, and the sacroiliac joints [7,8].

Previous investigations have demonstrated an association between LBP and atrophic changes in the paraspinal musculature [9–11]. Moreover, degenerative changes of the lumbar spine have been linked to the development of LBP. The influence of muscle strength on the occurrence of LBP has been studied as well [12,13]. However, findings to date remain inconsistent, and it is still not fully clear which observed factors are causal and which are consequential to LBP. We hypothesized that chronic patients with LBP would show reduced cross-sectional areas (CSA) and strength in trunk and hip flexors, compared with controls. The objective of this study was to further clarify the mechanisms underlying the pathogenesis of LBP.

Material and Methods

STUDY POPULATION:

The participants were divided into 2 groups: a patient group (experimental group) and a healthy control group. The experimental group consisted of 50 participants who met the diagnostic criteria for chronic LBP syndrome. Exclusion criteria included a positive Lasègue test, documented history of spinal or hip surgery, neoplastic disease at any site, and acute disabling conditions. The control group consisted of 30 healthy participants. Inclusion criteria were the absence of symptoms of chronic LBP syndrome, along with the same exclusion criteria as above. Both male and female participants were included. The study protocol was approved by the Ethics Committee of the University of Rijeka School of Medicine (approval number: 01-000-00-431-2/2019), and all participants provided written informed consent in accordance with the Declaration of Helsinki. The experimental group included 22 men and 28 women, with a mean age of 53 years and a mean body mass index (BMI) of 27.2. The healthy group included 17 men and 13 women, with a mean age of 42.3 years and a mean BMI of 26.4.

All participants underwent magnetic resonance imaging (MRI) of the lumbosacral spine, followed by isokinetic dynamometric testing of trunk and hip flexor and extensor muscles.

MAGNETIC RESONANCE IMAGING:

MRI of the lumbar spine was performed using a 1.5 T Siemens Magnetom Avanto scanner (Erlangen, Germany). Standard sagittal and axial T1- and T2-weighted sequences were obtained (repetition time 169–5870 ms; echo time 4.76–133 ms; slice thickness 4–5 mm; field of view 180×240 to 380×380 mm²).

Muscles were manually segmented using integrated measurement tools within the Picture Archiving and Communication System (PACS – Jivex, VISUS Health IT GmbH, Bochum, Germany) on a diagnostic-grade monitor. Measurements were independently performed by 2 examiners. Comparison of results showed no significant differences. For normally distributed data, the difference method was applied; for non-normally distributed data, the Wilcoxon matched-pairs test was used. As no significant inter-examiner differences were found, subsequent measurements were performed by a single examiner.

CSAs were measured at the mid-disc level from L1/L2 to L5/S1 for the following muscles: psoas major, quadratus lumborum, erector spinae, and multifidus (Figure 1). At the S2/S3 level, where the intervertebral disc is developmentally replaced by bony synostosis, the following muscles were segmented: iliacus, rectus abdominis, and gluteus maximus (Figure 2).

ISOKINETIC DYNAMOMETRIC TESTING:

Isokinetic dynamometric testing was performed on a Cybex isokinetic system, model HUMAC NORM (Stoughton, MA, USA). Testing was performed by a licensed physiotherapist.

Hip flexors and extensors were tested bilaterally in the supine position, with the non-tested leg supported and the tested leg positioned against the dynamometer lever above the knee (distal third of the thigh). The rotational axis of the dynamometer was aligned with the greater trochanter, and thoracic stabilization was provided using a fixation strap. After proper positioning and programming of the range of motion, testing was performed in active concentric mode over 2 cycles. The first cycle was conducted at 60°/s for 5 repetitions, following 4 familiarization trials. The second cycle was performed at 240°/s for 15 repetitions, also after 4 familiarization trials.

Trunk flexor and extensor muscles were tested in the standing position using the Trunk Modular Component module connected to the dynamometer. The thigh and lower leg were fixed with the knees in semiflexion, and the pelvic girdle was stabilized with the module’s rotational axis aligned at the level of the anterior superior iliac spine. Support pads were positioned anteriorly and posteriorly on the thorax. Testing was performed in active concentric mode at 60°/s and 90°/s for 4 repetitions, following 3 familiarization trials at each speed [14].

STATISTICAL ANALYSIS:

Data were processed using Statistica 14.1.0.8 (Cloud Software Group, Inc) and initially recorded in Microsoft Excel. Sample size was calculated in a preliminary study using power analysis, with a test power of 80% and significance level of 0.05 considered sufficient. Using guidelines from Cohen et al [15], a sample size was conducted based on the previous research [16–19], and it appeared that a sample size (n) ranging from 29 to 42 would be needed for a statistical power of 0.80 and an alpha level 0.05.

Descriptive and inferential statistical methods were applied. Normality was assessed using the Kolmogorov-Smirnov test. Normally distributed variables are presented as mean±standard deviation (SD), and parametric tests were used. Comparisons between the patient and healthy groups were performed using the t test for independent samples. The Pearson correlation coefficient (r) was used to determine relationships between variables within groups. Statistical significance was set at P≤0.05, corresponding to 95% confidence intervals.

Results

DEMOGRAPHIC AND ANTHROPOMETRIC CHARACTERISTICS:

No significant differences were observed between the groups in terms of sex distribution or BMI. Participants in the healthy group were significantly younger (mean±SD), which is acknowledged as a limitation of the study. However, given that over 80% of the population experienced at least 1 episode of LBP syndrome during their lifetime, it was not possible to form a fully uniform control group across all parameters (Table 1).

MUSCLE CROSS-SECTIONAL AREAS ANALYSIS:

Analysis of muscle CSAs revealed significant differences between patients with chronic LBP and healthy controls in selected trunk muscles. The psoas major exhibited reduced CSA at multiple lumbar levels in patients, most prominently at L2/L3, L3/L4, and L4/L5. Similarly, the rectus abdominis at S2/S3 showed smaller CSAs in patients compared with controls. No significant differences were observed in the quadratus lumborum, erector spinae, multifidus, or hip muscles (iliacus and gluteus maximus). These results indicate selective atrophy of specific trunk muscles in patients with chronic LBP (Table 2).

ISOKINETIC DYNAMOMETRIC TESTING ANALYSIS:

To ensure comparability during isokinetic dynamometric testing, age-adjusted corrections were applied when entering participants into the dynamometer. The resulting values were interpreted using reference tables that account for both the age and physical conditioning of the participants. The isokinetic dynamometric testing revealed significant differences in trunk and hip muscle strength between the experimental and control groups. Notably, trunk flexors and trunk extensors exhibited markedly lower absolute and relative (%) strength values in the experimental group compared with controls (P<0.001 for all comparisons; Table 3).

The trunk flexor/trunk extensor ratio was also significantly lower in the experimental group (0.7666±0.20 vs 0.9583±0.16, P<0.001), indicating an imbalance in trunk muscle strength. Similarly, hip flexors and hip extensors demonstrated significantly lower absolute and corrected strength values bilaterally in the experimental group (P < 0.001 for all comparisons). The hip flexor/hip extensor ratio showed a trend toward imbalance, particularly on the left side (0.6026±0.15 vs 0.668±0.12, P=0.047), although this difference was not statistically significant on the right side (P=0.241). Age-adjusted corrections ensured comparability across groups, further strengthening the validity of these results (Table 4).

CORRELATION ANALYSIS:

Table 5 shows the correlation coefficients (r(X,Y)) and P values for trunk flexors and extensors across various spinal levels in the experimental and control groups. The analysis included the following muscles: rectus abdominis, quadratus lumborum, erector spinae, multifidus, psoas major, iliacus, and gluteus maximus, stratified by spinal levels. Key findings include significant positive correlations in the experimental group for trunk extensors, particularly in the lumbar region. Trunk flexors showed weaker or non-significant correlations in both groups, with the experimental group exhibiting slightly stronger correlations. In the control group, trunk extensors demonstrated consistently strong correlations. The erector spinae and multifidus showed minimal correlations in the experimental group, while the control group displayed moderate correlations at specific levels (erector spinae at L3/L4: r=0.366, P=0.046; data not shown). The iliacus exhibited moderate correlations in both groups, with stronger associations in the control group. These findings suggest differential activation patterns in trunk muscles, with the experimental group potentially exhibiting altered neuromuscular control or compensatory mechanisms, particularly in the lumbar region.

Table 6 shows the correlation coefficients (r(X,Y)) and P values for muscle activity across various spinal levels in the experimental and control groups. The analysis focuses on hip flexors and hip extensors, comparing bilateral (right and left) muscle activity patterns. Results are stratified by specific spinal levels and muscle groups. Key findings include strong positive correlations in the experimental group for hip flexors and extensors, particularly in the lumbar region (L1/L5), with the rectus abdominis showing significant correlations at L5/S1. Bilateral differences were observed, with some spinal levels exhibiting stronger correlations on one side, such as the quadratus lumborum at L3/L4. Weak or non-significant correlations were noted in the control group, especially in the erector spinae and multifidus at higher spinal levels (eg, L3/L4, L4/L5; data not shown). The psoas major demonstrated the strongest correlations in the experimental group, particularly at lower spinal levels L4/L5 and L5/S1.

Discussion

SUBSEQUENT ANALYSIS OF CSA AND RELATIVE PEAK TORQUE RELATIONSHIPS:

In the latter part of the study, we examined the association between the CSAs of the investigated muscles and relative muscle group strength. We aimed to compare our findings with prior research on trunk flexors and extensors, as reported by Verbrugge et al [27]. As previously noted, a positive association exists between LBP and reduced muscle strength of trunk flexors and extensors. Our results demonstrated a statistically significant positive correlation between the CSAs of the m. psoas major at nearly all levels and the normalized strength of all tested muscle groups in patients with LBP. In contrast, healthy participants exhibited significant positive associations only at specific levels and exclusively with relative trunk extensor and hip flexor strength.

A similar pattern was observed for the m. rectus abdominis, with the exception that healthy participants showed significant correlations only with relative trunk extensor strength. Unlike the m. rectus abdominis, the m. iliacus in LBP patients displayed no significant association with normalized trunk flexor strength, whereas healthy participants demonstrated a correlation solely with normalized trunk extensor strength. Furthermore, no significant association was found between the CSA of the m. quadratus lumborum and normalized trunk extensor strength in patients with LBP. Notably, LBP patients exhibited a significant correlation between the CSA of the m. quadratus lumborum and normalized strength of both trunk flexors and extensors, a relationship absent in healthy participants.

REPETITIVE PATTERNS IN CSA-STRENGTH ASSOCIATIONS:

Across muscles, a consistent pattern emerged between CSA and normalized strength. In healthy participants, the CSAs of the hip and trunk muscles did not correlate with trunk flexor strength, whereas in patients with LBP, significant associations were observed for hip and trunk flexors, particularly hip flexors (significant at almost all levels). A similar tendency appeared for hip extensors, although neither trunk nor hip extensors showed statistically significant CSA-strength associations overall. The m. quadratus lumborum, functionally a trunk lateroflexor [8], was an exception to this pattern.

For the m. erector spinae, m. multifidus, and m. gluteus maximus, no significant associations between CSA and relative strength were found in either group. Thus, in healthy individuals, the CSA is not related to relative strength of trunk or hip extensors, whereas in patients with LBP, recurrent weakness of trunk and hip flexors may contribute to functional impairment.

In summary, patients LBP showed reduced trunk flexor strength and CSA compared with healthy participants, as well as reduced trunk extensor strength without differences in extensor CSA. A similar pattern was observed in the hip muscles. These findings support a link between LBP and both the morphological and functional characteristics of trunk muscles, and indicate that hip musculature is also involved in the LBP “chain”. This study re-examines previous findings and lays the foundation for further efforts in elucidating the complex mechanism of the occurrence of lumbar pain syndrome.

Conclusions

This study demonstrated significant morphological and functional differences in trunk and hip musculature between individuals with chronic LBP and healthy controls. The CSA of the m. psoas major and m. rectus abdominis was significantly smaller in patients with LBP, while no differences were observed for the m. quadratus lumborum, m. erector spinae, m. multifidus, m. iliacus, and m. gluteus maximus. Functionally, patients with LBP exhibited lower absolute and relative trunk and hip flexor and extensor strength, with a marked imbalance favoring extensors.

The analysis of CSA-strength relationships further revealed that, in patients with LBP, muscle size, particularly of the m. psoas major and m. rectus abdominis, was positively associated with relative trunk and hip flexor strength, a pattern largely absent in healthy participants. These findings indicate that LBP is characterized by both the morphological atrophy and functional weakness of key stabilizing and flexor muscle groups, supporting the hypothesis that muscular imbalance and altered load distribution contribute to spinal instability and pain.

Importantly, this study underscores that LBP extends beyond local spinal musculature, implicating hip muscles as integral components of the lumbopelvic kinetic chain. This integrative perspective emphasizes the need for rehabilitation strategies that target not only trunk extensors but also trunk and hip flexors, such as the m. psoas major and rectus abdominis, aiming to restore balanced muscle function and improve spinal stability.