Abstract
Introduction
The role of pelvic stabilizing muscles, particularly the gluteus muscles, remains largely unexplored in the context of sagittal alignment. Understanding the relative vulnerability of each muscle group during the sequential breakdown of sagittal alignment is thus important. We hypothesize that gluteal muscle degeneration precedes or accompanies the failure of pelvic compensation in sagittal malalignment, making it a key indicator of early decompensation. Here, we examine which trunk or pelvic muscles demonstrate the greatest deterioration in morphology and quality across progressive stages of sagittal alignment decompensation, and aim to identify muscle-related predictors of spinopelvic malalignment.
Methods
In this cross-sectional study, we analyzed 204 outpatients stratified into groups based on standing whole-spine radiographic parameters: normal (sagittal vertical axis [SVA] ≤50 mm and pelvic tilt [PT] ≤20°), compensated (SVA ≤50 mm and PT >20°), and decompensated (SVA >50 mm). We measured computed tomography-derived cross-sectional area (CSA) and Hounsfield unit (HU) values for the paravertebral (erector spinae and multifidus), psoas, gluteus medius, and gluteus maximus muscles. Group comparisons were performed using non-parametric tests, and logistic regression models were used to identify independent predictors of alignment deterioration.
Results
Both gluteal and paraspinal muscles exhibited significant declines in CSA and HU values across worsening alignment stages, whereas the psoas muscle did not. In multivariate analyses, reduced HU values of the gluteus medius and maximus and decreased paraspinal CSA values were independently associated with abnormal and decompensated alignment. Notably, the gluteus medius HU value was positively associated with compensation maintenance, whereas the gluteus maximus HU value had a negative association.
Conclusions
Degeneration of the gluteal muscles, especially in muscle quality, is strongly associated with the breakdown of sagittal alignment compensation. Our results suggest the gluteal musculature is a useful early biomarker and therapeutic target in preventing age-related spinal imbalance.
Keywords: gluteal muscles, spinopelvic alignment, sagittal balance, muscle degeneration, Hounsfield unit, cross-sectional area, compensation failure, age-related kyphosis
Introduction
Age-related kyphosis and posterior pelvic tilt (PT) are common sagittal alignment disorders in the elderly, arising from progressive musculoskeletal degeneration with significant functional impact1,2). Disruption of the sagittal vertical axis (SVA) and PT are critical indicators of global imbalance and deterioration3,4). These alignment changes reflect the failure of compensatory mechanisms that normally maintain upright posture with minimal energy expenditure5).
Prior work on the relationship between alignment and trunk muscle degeneration has mainly examined the paraspinal muscles and psoas major6,7). Reduced CSA and increased fatty infiltration in these muscles are associated with sagittal imbalance, chronic low back pain, and poorer postoperative outcomes8,9). Importantly, patients categorized as ‘compensated’ (normal SVA but elevated PT) already exhibit reduced quality of life and back pain, highlighting the relevance of early compensatory stages10).
In contrast, the contribution of pelvic stabilizing muscles―particularly the gluteus medius and maximus―remains underinvestigated11). Anatomically, the gluteus medius is a primary hip abductor and stabilizes the pelvis in the frontal plane during single-leg stance12). The gluteus maximus supports posterior pelvic stability, counters anterior trunk flexion, and assists the lumbopelvic complex during sagittal movements13,14). These muscles act synergistically with paraspinal and psoas muscles to preserve upright posture, particularly in older adults with weakened trunk extensors.
We aim to clarify the relative vulnerability of these muscle groups during sequential sagittal alignment breakdown. We hypothesize that gluteal muscle degeneration precedes or accompanies pelvic compensation failure, serving as an early indicator of decompensation. Accordingly, we compare muscle morphology and quality across the paraspinal, psoas, and gluteal groups in normal, compensated, and decompensated alignment categories, and identify candidate muscle groups as early therapeutic targets for preventive and rehabilitative strategies15).
Materials and Methods
Study design and population
This cross-sectional observational study was conducted at a single tertiary orthopedic center and included consecutive outpatients at a spine clinic between January 2023 and March 2025. Participants were eligible if they had no prior treatment for spinal disorders (including conservative management) and had both standing lateral whole-spine radiographs and lumbar computed tomography (CT) scans obtained within a three-month interval. Exclusion criteria were known neurological disorders or incomplete imaging data.
Spinopelvic alignment classification
Sagittal spinal alignment was assessed using standardized standing lateral whole-spine radiographs. Radiographic parameters included SVA, pelvic incidence, lumbar lordosis (LL), PT, and thoracic kyphosis. Two independent board-certified spine surgeons, blinded to clinical and imaging group assignments, performed digital measurements. Intra- and inter-rater reliability were determined using intraclass correlation coefficients (ICCs) in a randomly selected subsample of 30 patients. ICCs for all parameters exceeded 0.90 (0.91-0.97), indicating excellent agreement.
Participants were classified into three sagittal alignment groups based on SVA and PT: normal (SVA ≤50 mm, PT ≤20°), compensated (SVA ≤50 mm, PT >20°), and decompensated (SVA >50 mm). This classification was adapted and simplified from a previous study to clarify progressive muscular changes associated with alignment deterioration10).
Age-related sagittal malalignment is generally understood as a sequential transition from normal to compensated and ultimately decompensated stages, consistent with the cone of economy theory5) and supported by studies describing biomechanical adaptations and eventual compensatory failure in the aging spine11,12).
Muscle evaluation using computed tomography
Axial CT images of the lumbar spine and pelvis were obtained using a 64-slice multidetector CT scanner (Revolution EVO, GE Healthcare, Chicago, IL, USA) with patients in the supine position. Acquisition parameters included a tube voltage of 120 kVp, a tube current of 200-250 mA with automated exposure control, and a slice thickness of 3 mm.
Bilateral evaluation was performed at standardized anatomical levels: paravertebral muscles (erector spinae and multifidus) at T12/L1 and L4/5, psoas major at L4/5, gluteus medius at the superior femoral head, and gluteus maximus at the greater trochanter. These landmarks have demonstrated clinical relevance and reproducibility for assessing muscle morphology and quality6,9,11,16). The thoracolumbar junction (T12/L1) and lower lumbar segment (L4/5) are widely used for evaluating paravertebral degeneration due to their susceptibility to age-related and mechanical changes6,9). The L4/5 level represents the thickest psoas belly6. The superior femoral head and greater trochanter provide consistent peak CSA levels for the gluteus medius and maximus, respectively11,16).
Muscles were manually traced using DICOM software (OsiriX, Pixmeo, Switzerland) by trained analysts blinded to group classification. CSA (mm2) was used to assess muscle quantity, and mean attenuation (Hounsfield unit [HU]) was used to assess muscle quality. Bilateral values were averaged. Care was taken to avoid including bone, fascia, or intermuscular fat during muscle delineation17,18).
To ensure reliability, intra- and inter-rater ICCs were calculated using a randomly selected subsample of 30 cases. ICCs for CSA and HU exceeded 0.90 across all muscle groups (0.91-0.96), confirming excellent reproducibility.
Statistical analysis
All statistical analyses were performed using JMP version 17.0 (SAS Institute Inc., Cary, NC, USA). Continuous variables were first assessed for normality using the Shapiro-Wilk test. Normally distributed variables were summarized as means with standard deviations and compared using one-way analysis of variance, whereas non-normally distributed variables were presented as medians with interquartile ranges and analyzed using the Kruskal-Wallis test. When overall group differences were significant, post hoc pairwise comparisons were performed using Dunn's test with Bonferroni correction. Categorical variables, including sex distribution, were analyzed using the chi-square test.
To identify independent predictors of sagittal alignment deterioration, two multivariate logistic regression models were constructed. Model 1 evaluated predictors of abnormal alignment by comparing the normal group with a combined group of compensated and decompensated patients, both classified as having spinopelvic malalignment. Model 2 assessed predictors of further deterioration by comparing the compensated and decompensated groups. This analytic structure reflects the hypothesized clinical trajectory of alignment breakdown, with Model 1 examining early compensation failure and Model 2 capturing progression to advanced decompensation, enabling identification of stage-specific muscle predictors.
Independent variables included demographic factors and CT-derived muscle parameters, specifically CSA and HU values of the paravertebral (erector spinae and multifidus), psoas, gluteus medius, and gluteus maximus muscles. Multicollinearity was assessed using the variance inflation factor (VIF), with all variables demonstrating acceptable levels (VIF <2.0). No significant interaction terms were identified in either model.
A priori power analysis based on pilot data assumed an odds ratio (OR) of 2.0 for the association between gluteal muscle degeneration and alignment deterioration. To detect this effect with 80% power at a two-sided α of 0.05, a minimum of 186 participants was required; thus, the final sample size of 204 was deemed adequate. All statistical tests were two-tailed, with p-values <0.05 considered statistically significant.
Results
Patient characteristics
A total of 218 patients initially provided informed consent, of whom 204 met the inclusion criteria and were analyzed. These patients were categorized into the normal (n=58), compensated (n=54), and decompensated (n=92) groups (Figure 1).
Figure 1.
Study Flow Diagram. Flowchart depicting the inclusion and exclusion process for patient selection. A total of 204 participants were included in the final analysis and classified into three groups based on sagittal alignment: normal (sagittal vertical axis ≤50 mm and pelvic tilt ≤20°), compensated (sagittal vertical axis ≤50 mm and pelvic tilt >20°), and decompensated (sagittal vertical axis >50 mm). These classifications reflect progressive stages of sagittal malalignment, with increasing pelvic retroversion and anterior trunk shift.
As shown in Table 1, mean age increased progressively with worsening alignment: 69.6±10.4 years in the normal group, 73.7±8.7 in the compensated group, and 75.9±8.2 in the decompensated group (p<0.001). Overall, 84 men and 120 women were included. Among men, 51.7% were in the normal group, 18.0% in the compensated group, and 30.3% in the decompensated group; among women, the corresponding proportions were 21.3%, 35.1%, and 43.6%, respectively (p=0.0076).
Table 1.
Patient Characteristics by Sagittal Alignment Group.
| Variable | Normal (n=58) | Compensated (n=54) | Decompensated (n=92) | p-Value |
|---|---|---|---|---|
| Sex (female % in group) | 48.3% | 74.1% | 67.4% | 0.0076† |
| Sex (group % by sex) | ||||
| – Male (n=84) | 51.7% | 18.0% | 30.3% | |
| – Female (n=120) | 21.3% | 35.1% | 43.6% | |
| Age (years) | 69.6±10.4 | 73.7±8.7 | 75.9±8.2 | <0.001‡ |
| BMI (kg/m2) | 24.5±3.5 | 23.5±3.3 | 23.9±3.5 | 0.3425‡ |
| SVA (mm) | 14.5 [7.2-26.3] | 11.0 [3.6-21.5] | 96.0 [75.5-124.1] | <0.0001† |
| PI (°) | 45.5±10.9 | 48.0±7.6 | 47.9±14.1 | 0.3971‡ |
| LL (°) | 43.6±11.2 | 34.6±13.0 | 28.7±15.4 | <0.0001‡ |
| TK (°) | 25.1±10.3 | 30.1±14.2 | 32.1±15.5 | 0.0564‡ |
| PT (°) | 13.0 [10.2-16.1] | 27.3 [24.1-30.6] | 25.5 [19.2-31.2] | <0.0001† |
Data are presented as mean±standard deviation or median [interquartile range], as appropriate based on the Shapiro–Wilk test for normality.
BMI: body mass index; LL: lumbar lordosis; PI: pelvic incidence; PT: pelvic tilt; SVA: sagittal vertical axis; TK: thoracic kyphosis
†p-value calculated using chi-square test (for categorical variable) or Kruskal–Wallis test (for non-normally distributed continuous variables)
‡p-value calculated using one-way analysis of variance (ANOVA)
A p-value of <0.05 was considered statistically significant.
SVA showed marked elevation in the decompensated group (median 96.0 mm [75.5-124.1]) compared with the compensated (11.0 mm [3.6-21.5]) and normal groups (14.5 mm [7.2-26.3]; p<0.0001). PT was significantly higher in the compensated (27.3° [24.1-30.6]) and decompensated groups (25.5° [19.2-31.2]) than in the normal group (13.0° [10.2-16.1]; p<0.0001). LL declined with alignment deterioration: 43.6±11.2° in the normal group, 34.6±13.0° in the compensated group, and 28.7±15.4° in the decompensated group (p<0.0001).
Muscle quantity and quality across alignment groups
CSA and HU values for paravertebral, gluteal, and psoas muscles were compared across alignment categories (Table 2). At T12/L1, paravertebral CSA declined significantly from 1,325 mm2 (1,136-1,452; normal) to 1,135 mm2 (932-1,291; compensated) and 1,070 mm2 (921-1,245; decompensated) (p<0.0001), a 19.2% reduction. At L4/5, CSA declined from 1,842 mm2 (1,650-2,060) to 1,543 mm2 (1,325-1,754) (p<0.0001), a 16.2% reduction. Muscle density at L4/5 (HU) also declined significantly (median, 19.0-9.6; p=0.0383), suggesting progressive fatty infiltration.
Table 2.
Comparison of Muscle Quantity and Quality Among the Three Groups.
| Variable | Normal | Compensated | Decompensated | p_KW | Post-hoc |
|---|---|---|---|---|---|
| T12/L1 Paraspinal CSA (mm2) | 1,325 [1,136-1,452] | 1,135 [932-1,291] | 1,070 [921-1,245] | <0.0001 | Normal vs Comp: p=0.0042 Normal vs Decomp: p<0.0001 Comp vs Decomp: p=0.5874 |
| T12/L1 Paraspinal HU | 35.2 [31.0-41.3] | 29.1 [18.0-35.0] | 24.2 [15.5-31.4] | <0.0001 | Normal vs Comp: p=0.0050 Normal vs Decomp: p<0.0001 Comp vs Decomp: p=0.2971 |
| L4/5 Paraspinal CSA (mm2) | 1,842 [1,650-2,060] | 1,744 [1,495-2,023] | 1,543 [1,325-1,754] | <0.0001 | Normal vs Comp: p=0.1887 Normal vs Decomp: p<0.0001 Comp vs Decomp: p=0.1571 |
| L4/5 Paraspinal HU | 19.0 [11.2-28.5] | 14.1 [5.5-21.2] | 9.6 [2.0-17.9] | 0.0383 | Normal vs Comp: p=0.1999 Normal vs Decomp: p=0.0380 Comp vs Decomp: p=0.6848 |
| Psoas CSA (mm2) | 785 [565-980] | 748 [520-940] | 705 [510-911] | 0.1516 | Not tested |
| Psoas HU | 33.0 [26.5-41.3] | 31.1 [23.4-39.5] | 27.4 [18.0-36.1] | 0.1240 | Not tested |
| Gluteus Medius CSA (mm2) | 2,180 [1,972-2,390] | 1,832 [1,575-2,000] | 1,610 [1,380-1,835] | <0.0001 | Normal vs Comp: p=0.0016 Normal vs Decomp: p<0.0001 Comp vs Decomp: p=0.0049 |
| Gluteus Medius HU | 46.3 [42.5-50.0] | 45.5 [42.0-48.1] | 39.3 [29.0-45.5] | 0.0596 | Not tested |
| Gluteus Maximus CSA (mm2) | 2,460 [2,200–2,720] | 2,090 [1,885–2,330] | 1,830 [1,510–2,100] | <0.0001 | Normal vs Comp: p=0.0003 Normal vs Decomp: p<0.0001 Comp vs Decomp: p=0.0679 |
| Gluteus Maximus HU | 27.2 [20.2-33.6] | 30.9 [25.1-35.7] | 20.2 [12.5-27.4] | 0.0052 | Normal vs Comp: p=0.3066 Normal vs Decomp: p=0.0064 Comp vs Decomp: p=0.1381 |
Values are presented as median (interquartile range) based on the results of the Shapiro–Wilk test for normality.
Group comparisons were performed using the Kruskal–Wallis test due to non-normal data distribution.
If the overall p-value was statistically significant (p<0.05), pairwise post hoc comparisons were conducted using Dunn’s test with Bonferroni correction.
Post hoc comparisons are shown only when the Kruskal–Wallis test revealed significant differences among the three groups.
Compensated: Pelvic compensated group (SVA ≤50 mm and PT >20°, n=54); CSA: cross-sectional area; Decompensated: Pelvic decompensated group (SVA >50 mm, n=92); Gluteus Maximus: gluteus maximus muscle; Gluteus Medius: gluteus medius muscle; HU: Hounsfield unit; Normal: Normal alignment group (SVA ≤50 mm and PT ≤20°, n=58); Psoas: psoas major muscle; SVA: sagittal vertical axis; T12/L1 and L4/5 Paravertebral: paravertebral muscles measured at T12/L1 and L4/5 levels, respectively.
Degeneration of the gluteal muscles was more pronounced. Gluteus medius CSA decreased from 2,180 mm2 (1,972-2,390) to 1,610 mm2 (1,380-1,835) (p<0.0001), a 26.1% reduction; gluteus maximus CSA decreased by 25.6%, from 2,460 mm2 (2,200-2,720) to 1,830 mm2 (1,510-2,100) (p<0.0001). Gluteus maximus HU declined significantly (27.2-20.2, p=0.0052), with the largest decrement between the normal and decompensated groups (p=0.0064). These findings indicate a strong association between gluteal muscle degeneration and more severe spinopelvic malalignment.
In contrast, the psoas major showed no significant differences in CSA or HU across groups (p=0.1516 and p=0.1240, respectively), indicating relative preservation.
Predictors of abnormal alignment (normal vs. compensated/decompensated)
Multivariate logistic regression identified independent predictors of abnormal sagittal alignment (Table 3). Female sex showed a significant association (OR, 5.60; 95% confidence interval [CI], 1.39-22.66; p=0.0156).
Table 3.
Predictors of Abnormal Postural Alignment (Normal Vs. Compensated/Decompensated).
| Variable | Odds Ratio | 95% CI | p-Value |
|---|---|---|---|
| Sex (female) | 5.60 | 1.39-22.66 | 0.0156* |
| Age (per one-year increase) | 1.05 | 1.00-1.11 | 0.0692 |
| BMI (kg/m2) | 1.07 | 0.90-1.27 | 0.4226 |
| Paraspinal CSA at T12/L1 (mm2) | 1.00 | 0.998-1.002 | 0.8660 |
| Paraspinal HU at T12/L1 | 0.92 | 0.87-0.97 | 0.0006* |
| Paraspinal CSA at L4/5 (mm2) | 1.00 | 0.999-1.000 | 0.1165 |
| Paraspinal HU at L4/5 | 0.98 | 0.95-1.01 | 0.1808 |
| Psoas CSA (mm2) | 1.00 | 0.999-1.003 | 0.2433 |
| Psoas HU | 1.01 | 0.98-1.05 | 0.4772 |
| Gluteus medius CSA (mm2) | 1.00 | 0.996-0.999 | <0.0001* |
| Gluteus medius HU | 0.88 | 0.80-0.97 | 0.0102* |
| Gluteus maximus CSA (mm2) | 1.00 | 0.997- 0.999 | 0.0004* |
| Gluteus maximus HU | 1.10 | 1.02-1.19 | 0.0172* |
Multivariate logistic regression was performed using abnormal postural alignment (combined compensated and decompensated groups) as the dependent variable. Values are presented as odds ratios with 95% confidence intervals.
BMI: body mass index; CSA: cross-sectional area; HU: Hounsfield unit
Normal=Normal alignment group (n=58); Compensated=Pelvic compensated group (n=54); Decompensated=Pelvic decompensated group (n=92); T12/L1 and L4/5 Paraspinal=cross-sectional measurements of the paraspinal muscles at the T12/L1 and L4/5 levels, respectively; Psoas=psoas major muscle; Gluteus Medius=gluteus medius muscle; Gluteus Maximus=gluteus maximus muscle.
A p-value <0.05 was considered statistically significant. Statistically significant predictors are indicated by an asterisk.
Among muscle parameters, lower HU values of the paravertebral muscles at T12/L1 (OR, 0.92; 95% CI, 0.87-0.96; p=0.0006) and the gluteus medius (OR, 0.88; 95% CI, 0.80-0.97; p=0.0102) were significant predictors of abnormal alignment. In contrast, higher HU values of the gluteus maximus were associated with abnormal alignment (OR, 1.10; 95% CI, 1.02-1.20; p=0.0172). Reduced CSA of the gluteus medius and maximus also remained significant after adjustment (both p<0.001). Overall, variations in both gluteal and paraspinal muscle quality and quantity were significantly associated with the presence of sagittal malalignment.
Predictors of decompensation (compensated vs. decompensated)
In comparisons between compensated and decompensated groups (Table 4), female sex was an exceptionally strong predictor of decompensation (OR, 66.74; 95% CI, 11.33-393.30; p<0.0001).
Table 4.
Predictive Factors for Transition from Pelvic Compensation to Decompensation.
| Variable | Odds ratio | 95% CI | p-Value |
|---|---|---|---|
| Sex (female) | 66.74 | 11.33-393.30 | <0.0001 * |
| Age (per one-year increase) | 1.01 | 0.96-1.06 | 0.6764 |
| BMI (kg/m2) | 1.19 | 0.99-1.42 | 0.0611 |
| Paraspinal CSA at T12/L1 (mm2) | 0.998 | 0.996-1.000 | 0.0411 * |
| Paraspinal HU at T12/L1 | 1.0 | 0.98-1.02 | 0.9968 |
| Paraspinal CSA at L4/5 (mm2) | 0.998 | 0.997-1.000 | 0.0036 * |
| Paraspinal HU at L4/5 | 1.0 | 0.98-1.02 | 0.9274 |
| Psoas CSA (mm2) | 1.0 | 0.99-1.01 | 0.5118 |
| Psoas HU | 1.0 | 0.97-1.03 | 0.8487 |
| Gluteus medius CSA (mm2) | 1.0 | 0.97-1.03 | 0.8681 |
| Gluteus medius HU | 1.046 | 1.00-1.13 | 0.0002 * |
| Gluteus maximus CSA (mm2) | 1.0 | 0.98-1.02 | 0.9576 |
| Gluteus maximus HU | 0.893 | 0.83-0.96 | 0.0014 * |
Values are expressed as odds ratios with 95% confidence intervals. A multivariate logistic regression analysis was conducted to identify independent predictors of transition from pelvic compensation to decompensation. Variables included demographics and muscle quality, and quantity indices derived from imaging data.
Statistically significant values are indicated with an asterisk (p<0.05).
CSA: cross-sectional area; HU: Hounsfield unit; Paraspinal: paraspinal muscle;
Psoas: psoas major muscle; Gluteus medius: gluteus medius muscle; Gluteus maximus: gluteus maximus muscle; T12/L1 and L4/5: vertebral levels at which the paraspinal muscles were evaluated; BMI: body mass index; CI: confidence interval
Lower paravertebral CSA at T12/L1 (OR, 0.86; 95% CI, 0.75-0.99; p=0.0411) and L4/5 (OR, 0.85; 95% CI, 0.76-0.95; p=0.0036) was also associated with decompensation. The gluteus medius HU value showed a positive association (OR, 1.046; 95% CI 1.02-1.07; p=0.0002), whereas the gluteus maximus HU value showed a negative association (OR, 0.893; 95% CI, 0.84-0.96; p=0.0014), indicating distinct associations between gluteal muscle quality and the severity of sagittal compensation status.
Discussion
Summary of the results
We identified progressive declines in trunk and pelvic muscle morphology and quality across the three alignment categories. Gluteal and paravertebral muscles showed significant reductions in CSA and HU values with worsening alignment. Multivariate logistic regression revealed that degeneration in the gluteus medius and maximus, along with paravertebral muscle quality, was a significant predictor of abnormal sagittal alignment. Moreover, the transition from compensated to decompensated states was strongly associated with reductions in paravertebral CSA and distinct alterations in gluteal HU values, underscoring their differential roles in maintaining compensatory mechanisms.
Novel findings and strengths of this study
This is the first study to comprehensively investigate the relationship between gluteal muscle morphology and quality (CT-based CSA and HU) and the progression of sagittal malalignment in a clinical outpatient cohort. Despite its critical role in pelvic stability, the gluteal musculature has been largely overlooked. Our findings demonstrate that deterioration of the gluteus medius and maximus is evident and is strongly associated with early alignment changes, positioning these muscles as potential early biomarkers and therapeutic targets. Another strength is the radiographically verified classification of alignment compensation and the use of a two-stage logistic regression model designed to distinguish predictors of initial compensation failure (Model 1: normal vs. compensated/decompensated) from those of further deterioration (Model 2: compensated vs. decompensated), reflecting the progressive biomechanical breakdown of sagittal alignment.
Comparison with previous studies
Associations between paravertebral muscle atrophy and sagittal imbalance have been documented, particularly in older adults19,20). Reduced paravertebral muscle CSA correlates with impaired global sagittal alignment and increased disability19). Additionally, sagittal malalignment―especially increased SVA―has been linked to impaired locomotive function and decreased physical performance in community-dwelling older women20). In contrast, the psoas major shows inconsistent associations with alignment parameters21), and our findings align with these reports.
Few studies have evaluated the gluteal muscles in this context. One investigation identified a relationship between gluteus medius atrophy and PT in hip osteoarthritis16), yet their broader contribution to spinopelvic alignment remains insufficiently characterized. Our results fill this gap by demonstrating that both CSA and HU of the gluteal muscles decline in parallel with sagittal malalignment, supporting their dynamic role in age-related kyphosis.
Interpretation of the findings
The progressive deterioration in gluteal muscle morphology and quality likely reflects age-related disuse, myosteatosis, and altered recruitment associated with compensatory spinal loading. The inverse relationship between HU values and alignment deterioration suggests increasing fatty infiltration, a known contributor to impaired muscle contractility and endurance22).
Notably, gluteus maximus HU was negatively associated with decompensation, whereas gluteus medius HU showed a positive association. This divergence may reflect their distinct biomechanical roles. As a primary coronal-plane stabilizer, the gluteus medius may be over-recruited during compensation to sustain balance under heightened sagittal load. Chronic overactivation could induce metabolic stress and fatty degeneration, explaining the elevated HU in the compensated state and its eventual vulnerability to failure. Conversely, reduced ambulatory demand in older adults may promote disuse atrophy of the gluteus maximus, which may be associated with more severe decompensated posture.
These findings suggest that gluteal muscle decline is closely associated with compensation breakdown, although causal relationships cannot be determined from the present cross-sectional design.
Additionally, the strong association between female sex and decompensation mirrors previous work19,23), indicating heightened susceptibility in women. Potential mechanisms include lower baseline muscle mass, postmenopausal estrogen deficiency, and a predominance of type I fibers, which are vulnerable to functional decline under chronic low-load conditions. Estrogen loss accelerates intramuscular fat accumulation and mitochondrial dysfunction, promoting sarcopenia24,25). These factors may compound gluteal decline in postmenopausal women and highlight the need for early, sex-specific preventive strategies.
Clinical implications and future directions
These results suggest a window for early preventive intervention, particularly in the compensated stage before irreversible deterioration. Given the known responsiveness of gluteal and paravertebral muscles to resistance training, future studies should evaluate targeted strengthening to halt or reverse alignment decline. Potential strategies include hip extensor and abductor resistance programs, neuromuscular re-education to improve motor control, and posture-correction interventions to restore sagittal balance. This is especially pertinent in postmenopausal women, in whom sarcopenic and hormonal changes accelerate alignment failure. Integrating muscle preservation into spine care may reduce fall risk, enhance quality of life, and delay or prevent surgical intervention.
Limitations
Among several limitations, the cross-sectional nature of this study precludes causal inference. Gluteal muscle degeneration is suggested to be strongly associated with the failure of sagittal alignment compensation, but longitudinal research is required to determine whether such muscular changes precede or follow alignment deterioration. Furthermore, compensation was defined solely based on static radiographic parameters (SVA and PT), although compensatory mechanisms are inherently functional phenomena involving dynamic posture control and coordinated muscle activation. Given that no objective functional assessments―such as gait analysis, electromyography, or balance testing―were included, the functional aspects of compensation could not be evaluated. Also, although CT provides a reliable and reproducible assessment of muscle morphology, it cannot assess neuromuscular activation or dynamic biomechanical function, further underscoring the absence of functional evaluation in the present study. Future studies might incorporate magnetic resonance imaging, gait analysis, or electromyography for a better understanding of muscle involvement in spinal alignment. Moreover, our cohort was drawn from a clinical outpatient population with spinal symptoms, which may limit generalizability. In addition, although acute vertebral fractures were excluded to avoid transient posture-related effects, we did not collect data on the number of pre-existing vertebral fractures, other osteoporosis-related fractures, or bone mineral density. These unmeasured factors may influence both muscle status and sagittal alignment and therefore represent an important limitation of the present study. Finally, various potentially confounding factors were not assessed here, and they should be accounted for in future research because they may influence the interpretation of the associations between muscle degeneration and sagittal alignment.
Conclusions
Gluteus medius and maximus degeneration appears to be closely associated with more severe spinopelvic malalignment, although its temporal relationship to compensation failure cannot be determined from this cross-sectional study. Although causality remains unclear, the strong associations observed suggest that the gluteal muscles may represent potential modifiable targets for future preventive strategies. Targeted resistance training, neuromuscular re-education, and postural correction programs should be investigated in longitudinal or interventional studies to determine whether they can help preserve pelvic muscle function and support sagittal balance in aging populations.
Author Contributions: S.M. and H.H. conceptualized and designed the study. S.M., H.Iw., S.T., M.Ta., K.N., Y.I., and M.Te. contributed to data collection and management. S.M. and K.M. performed the data analysis and interpreted the results. R.T., T.Ko., R.N., Y.Ya., and K.Mu. contributed to the interpretation of data and provided clinical insights. H.Ya. supervised the overall project. S.M. drafted the initial manuscript. H.H. and H.Ya. critically revised the manuscript for important intellectual content. All authors reviewed and approved the final manuscript.
Conflicts of Interest: The authors declare that there are no relevant conflicts of interest.
Hiroshi Hashizume is one of the Editors of Spine Surgery and Related Research and on the journal's Editorial Committee. He was not involved in the editorial evaluation or decision to accept this article for publication at all.
Ethics Approval: All patient procedures performed in this study were conducted in accordance with the ethical standards of the Research Ethics Committee of Shingu Municipal Medical Center (approval number: 104) and with the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards. Informed consent was obtained from all participants prior to their inclusion in the study.
Acknowledgments
We acknowledge proofreading and editing by Benjamin Phillis, a Board-Certified Editor in the Life Sciences (BELS), at Wakayama Medical University.
References
- 1.Katzman WB, Wanek L, Shepherd JA, et al. Age-related hyperkyphosis: its causes, consequences, and management. J Orthop Sports Phys Ther. 2010;40(6):352-60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Reid DBC, Daniels AH, Ailon T, et al. Frailty and health-related quality of life improvement following adult spinal deformity surgery. World Neurosurg. 2018;112:e548-54. [DOI] [PubMed] [Google Scholar]
- 3.Glassman SD, Bridwell K, Dimar JR, et al. The impact of positive sagittal balance in adult spinal deformity. Spine. 2005;30(18):2024-9. [DOI] [PubMed] [Google Scholar]
- 4.Barrey C, Roussouly P, Perrin G, et al. Sagittal balance disorders in severe degenerative spine. Can we identify the compensatory mechanisms? Eur Spine J. 2011;20(suppl 5):626-33. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Le Huec JC, Thompson W, Mohsinaly Y, et al. Sagittal balance of the spine [published correction in Eur Spine J. 2019;28(11):2631]. Eur Spine J. 2019;28(9):1889-905. [Google Scholar]
- 6.Yagi M, Hosogane N, Watanabe K, et al. The paravertebral muscle and psoas for the maintenance of global spinal alignment in patient with degenerative lumbar scoliosis. Spine J. 2016;16(4):451-8. [DOI] [PubMed] [Google Scholar]
- 7.Xia W, Fu H, Zhu Z, et al. Association between back muscle degeneration and spinal-pelvic parameters in patients with degenerative spinal kyphosis. BMC Musculoskelet Disord. 2019;20(1):454. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.He K, Head J, Mouchtouris N, et al. The implications of paraspinal muscle atrophy in low back pain, thoracolumbar pathology, and clinical outcomes after spine surgery: a review of the literature. Glob Spine J. 2020;10(5):657-66. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Muellner M, Haffer H, Moser M, et al. Paraspinal musculature impairment is associated with spinopelvic and spinal malalignment in patients undergoing lumbar fusion surgery. Spine J. 2022;22(12):2006-16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Murata S, Hashizume H, Tsutsui S, et al. Pelvic compensation accompanying spinal malalignment and back pain-related factors in a general population: the Wakayama spine study [published correction in Sci Rep. 2023;13(1):12791]. Sci Rep. 2023;13(1):11862. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Lee KY, Lee JH, Im SK. Effect of gluteal muscle strengthening exercise on sagittal balance and muscle volume in adult spinal deformity following long-segment fixation surgery. Sci Rep. 2022;12(1):9063. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Semciw AI, Pizzari T, Murley GS, et al. Gluteus medius: an intramuscular EMG investigation of anterior, middle and posterior segments during gait. J Electromyogr Kinesiol. 2013;23(4):858-64. [DOI] [PubMed] [Google Scholar]
- 13.Nascimento MB, Vilarinho LG, Lobato DFM, et al. Role of gluteus maximus and medius activation in the lower limb biomechanical control during functional single-leg tasks: a systematic review. Knee. 2023;43:163-75. [DOI] [PubMed] [Google Scholar]
- 14.Chang M, Slater LV, Corbett RO, et al. Muscle activation patterns of the lumbo-pelvic-hip complex during walking gait before and after exercise. Gait Posture. 2017;52:15-21. [DOI] [PubMed] [Google Scholar]
- 15.Miura T, Ito T, Sato K, et al. Is dynamic spino-pelvic alignment during gait associated with lumbar function in patients with degenerative lumbar spinal stenosis? Gait Posture. 2023;103:210-4. [DOI] [PubMed] [Google Scholar]
- 16.Momose T, Inaba Y, Choe H, et al. CT-based analysis of muscle volume and degeneration of gluteus medius in patients with unilateral hip osteoarthritis. BMC Musculoskelet Disord. 2017;18(1):457. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Kim HK, Kim KW, Kim EH, et al. Age-related changes in muscle quality and development of diagnostic cutoff points for myosteatosis in lumbar skeletal muscles measured by CT scan. Clin Nutr. 2021;40(6):4022-8. [DOI] [PubMed] [Google Scholar]
- 18.Wesselink EO, Pool JJM, Mollema J, et al. Is fatty infiltration in paraspinal muscles reversible with exercise in people with low back pain? A systematic review. Eur Spine J. 2023;32(3):787-96. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Takahashi S, Hoshino M, Ohyama S, et al. Relationship of back muscle and knee extensors with the compensatory mechanism of sagittal alignment in a community-dwelling elderly population. Sci Rep. 2021;11(1):2179. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Yahata M, Watanabe K, Tashi H, et al. Impact of spinal sagittal malalignment on locomotive syndrome and physical function in community-dwelling middle aged and older women. BMC Musculoskelet Disord. 2023;24(1):620. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Jang HD, Won SH, Kim DW, et al. Magnetic resonance imaging characteristics and age-related changes in the psoas muscle: analysis of 164 patients with back pain and balanced lumbar sagittal alignment. World Neurosurg. 2019;131:e88-95. [DOI] [PubMed] [Google Scholar]
- 22.Goodpaster BH, He J, Watkins S, et al. Skeletal muscle lipid content and insulin resistance: evidence for a paradox in endurance-trained athletes. J Clin Endocrinol Metab. 2001;86(12):5755-61. [DOI] [PubMed] [Google Scholar]
- 23.Lorbergs AL, Allaire BT, Yang L, et al. A longitudinal study of trunk muscle properties and severity of thoracic kyphosis in women and men: the Framingham study. J Gerontol A Biol Sci Med Sci. 2019;74(3):420-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Geraci A, Calvani R, Ferri E, et al. Sarcopenia and menopause: the role of estradiol. Front Endocrinol (Lausanne). 2021;12:682012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Lu L, Tian L. Postmenopausal osteoporosis coexisting with sarcopenia: the role and mechanisms of estrogen. J Endocrinol. 2023;259(1):e230116. [DOI] [PubMed] [Google Scholar]