Abstract
Objective
A recent study demonstrated that subcutaneous fat tissue thickness (SFTT) at the L1–L2 level is predicting paraspinal muscle fatty infiltration. Given that fat distribution patterns differ between males and females, we calibrated the SFTT index to obtain a standardized subcutaneous fat index (SSFI).
Methods
A total of 175 patients admitted between January 2023 and July 2024 were included in this study. Based on lumbar CT attenuation values, patients were classified into an osteoporosis group (HU ≤ 110) and a normal bone density group (HU > 110). Demographic characteristics and clinical variables, including age, BMI, sex, medical history, and the standardized subcutaneous fat index (SSFI), were compared between the two groups. Pearson correlation analysis was subsequently performed to evaluate the associations between SSFI1 and other variables. Multiple linear regression analyses were conducted to further assess the relationships between SSFI1 and multifidus fatty infiltration, erector spinae fatty infiltration, frailty index, and vertebral bone quality (VBQ). Finally, receiver operating characteristic (ROC) curve analysis was performed to evaluate the ability of SSFI1 to discriminate osteoporosis.
Results
Patients with osteoporosis were older than the control group, while no significant differences were observed in smoking, alcohol consumption, diabetes, or hypertension between the two groups. Compared with controls, patients with osteoporosis showed lower SSFI1 values, greater paraspinal muscle fatty infiltration, and higher frailty index and VBQ scores. Moreover, SSFI was lower in both male and female patients with osteoporosis. Pearson correlation analysis further demonstrated that SSFI1 was significantly associated with paraspinal muscle fatty infiltration, lumbar bone mineral density, femoral bone mineral density, frailty index, and VBQ. Multiple linear regression analysis revealed that SSFI1 was associated with VBQ. ROC curve analysis indicated that SSFI1 had potential value in distinguishing osteoporosis from non-osteoporotic individuals. When the cutoff value of SSFI1 was < 0.29, the area under the curve (AUC) was 0.642, with a sensitivity of 0.696 and a specificity of 0.921, suggesting that SSFI1 may serve as a potential imaging marker for osteoporosis.
Conclusion
SSFI1 was significantly reduced in patients with osteoporosis and was closely associated with paraspinal muscle fatty infiltration, bone mineral density, frailty index, and VBQ. In addition, SSFI1 demonstrated potential value in distinguishing osteoporosis from non-osteoporotic individuals. These findings suggest that SSFI1 may serve as a simple and accessible imaging biomarker for evaluating bone quality and may provide a novel approach for the early identification and risk stratification of osteoporosis.
Keywords: SSFI, Osteoporosis, Lumbar, HU
Introduction
As the population ages, osteoporosis, a condition characterised by progressive bone loss, fragility, and an increased risk of fractures, is becoming more prevalent and occurring at a younger age. This represents a growing public health concern [1–3]. Epidemiologic statistics indicate that in China, 20.73% of middle-aged and elderly men and 38.05% of middle-aged and elderly women suffer from different degrees of osteoporosis. Studies have demonstrated that if the progression and occurrence of osteoporosis are effectively prevented, vertebral and hip fractures of varying degrees can be prevented [4].
Osteoporosis is a multifactorial disorder influenced by both modifiable and non-modifiable determinants. Non-modifiable factors include genetics, aging, sex, race, and hormonal status, whereas modifiable factors encompass dietary habits, physical activity, obesity, and prolonged corticosteroid exposure [5]. As people age, fat tissue redistributes: subcutaneous fat decreases while intramuscular and intramedullary fat increases, increasing bone fragility [6, 7]. Body Mass Index (BMI) has been widely adopted as a standard index for assessing adipose-related metabolic disorders. However, because BMI fails to reflect the regional distribution and biological characteristics of adipose tissue, it is no longer considered a reliable indicator in adipose-related research [8–16]. Localised fat tissue (visceral fat and subcutaneous fat) is a better indicator of skeletal health than overall fat distribution [17–19]. Recent studies have further revealed that excess visceral fat combined with reduced subcutaneous fat exerts deleterious effects on bone microarchitecture [20].
Berikol et al. demonstrated that subcutaneous fat tissue thickness (SFTT) in the upper lumbar region provides a more accurate assessment of intervertebral disc degeneration [17]. SFTT is a better predictor of lumbar spine degeneration in males, with disc degeneration serving as a risk factor for bone changes [21]. The reason for this gender difference is thought to be due to differences in how fat is stored between the sexes. This study proposes a new metric: the standardised subcutaneous fat index (SSFI), which is calculated as the ratio of SFTT squared to lumbar vertebral body area. The SSFI was associated with vertebral fat, paraspinal muscle fat infiltration, and frailty index.
The vertebral bone quality (VBQ) score, based on magnetic resonance imaging (MRI), is a new technique for the evaluation of bone quality. The principle behind it is the measurement of the fat content of the vertebra, with indirect reflection of bone quality [22, 23]. The main aim of this study is to analyse the predictive ability of SSFI and SFTT for osteoporosis. Further analysis will examine the correlation between SSFI and fat content within tissues, as well as overall frailty levels. These findings may provide new insights into the relationship between spinal fat, muscle, and bone health.
Materials and methods
Patients
This single-center retrospective study included patients who underwent preoperative dual-energy X-ray absorptiometry (DXA) and 3.0-T lumbar spine magnetic resonance imaging (MRI) at the Third Hospital of Hebei Medical University between January 2023 and July 2024. The study was approved by the Ethics Committee of the Third Hospital of Hebei Medical University (No. Ke2025-402-1). All procedures were conducted in accordance with the Declaration of Helsinki (1964) and its subsequent amendments or comparable ethical standards. Written informed consent was obtained from all participants. Participants were excluded if they met any of the following criteria: Prior lumbar spine surgery; exposure to pharmacological agents known to alter fat distribution or bone metabolism (particularly systemic corticosteroids such as dexamethasone within the preceding year); comorbid metabolic or endocrine conditions with potential effects on adipose tissue distribution or bone metabolism; incomplete preoperative DXA or MRI data; diagnosis of pathological osteoporosis (e.g., tumor-associated osteoporosis); or radiological evidence of scoliosis.
Baseline demographic and radiological data were collected for all enrolled participants. Demographic variables included age, sex, body mass index (BMI), smoking history, alcohol consumption, diabetes, and hypertension. Radiological assessments included preoperative lumbar spine and bilateral hip bone mineral density (BMD) measured by DXA, as well as lumbar spine MRI for evaluating subcutaneous fat thickness and paraspinal muscle fat infiltration on T2-weighted images. Based on a lumbar CT value of 110 HU as the cutoff [22], participants were subsequently classified into two groups: the osteoporosis group and the normal bone density group.
SSFI measurement
SSFI was further quantified based on the subcutaneous fat tissue thickness (SFTT) previously described by Berikol et al. [17]. The measurements were performed on T2-weighted lumbar spine MRI images. All MRI scans were acquired using identical parameters: repetition time (TR) = 560 ms, echo time (TE) = 11.45 ms, slice thickness = 4 mm, field of view (FOV) = 300 mm, and number of signal averages = 3.
The SFTT was defined as the perpendicular distance from the tip of the spinous process to the overlying skin surface, measured on axial MRI images at the L1–L4 intervertebral disc levels (Fig. 1A). A region of interest (ROI) encompassing the entire vertebral body was selected to measure the vertebral body area at the L1–L4 levels (Fig. 1B). The SSFI was subsequently calculated as the square of the subcutaneous fat thickness divided by the lumbar vertebral body area.
Fig. 1.
Schematic diagram of SSFI measurement. A: Measure the distance from the spinous process to the skin at the mid-lumbar level (SFTT). B: Measure the area of the corresponding vertebral body. 
Calculation of VBQ score
In our analysis, the VBQ scores were derived from T1-weighted magnetic resonance imaging. For the VBQ score, a region of interest (ROI) was centrally positioned on the L1-L4 vertebral bodies within the median sagittal plane of the lumbar spine. For patients with abnormalities in the midsagittal slice (e.g., hemangioma, venous plexus, scoliosis changes), measurements were performed using the parasagittal slices. In the case of abnormal invasion of the entire vertebral body, this level was excluded from the calculation and only the remaining vertebrae were used to calculate the VBQ score. A meticulous approach was employed to maintain a 3 mm margin from the superior and inferior endplates of the vertebral bodies, and the cerebrospinal fluid level was identified at the L3 vertebral level [22, 23]. The VBQ score could be determined using the following formula.
 |
Paraspinal muscle fatty infiltration
This study quantified the degree of fatty infiltration (FI) in the paraspinal muscles using lumbar magnetic resonance imaging (MRI). Axial T2-weighted images at the mid-vertebral level of L3–L4 were analyzed to evaluate the cross-sectional area and fat replacement of the psoas major, multifidus, and erector spinae muscles [23]. All image analyses were performed using ImageJ software (National Institutes of Health, Bethesda, MD, USA).
Frailty index measurement
The frailty index in this study was assessed using Conlon’s 5i-mFI. The 5i-mFI is based on five variables (nonindependent functional status, diabetes mellitus treated with oral agents or insulin, COPD, hypertension requiring medication, and CHF), with each variable assigned a score of 1, resulting in a total score ranging from 0 to 5. Frailty status was stratified as follows: nonfrail = 0, prefrail = 1, frail = 2, and severely frail ≥ 3.
Statistical analysis
To evaluate the intra-observer and inter-observer reliability of the measurements, the intraclass correlation coefficient (ICC) was calculated for the preoperative and postoperative measurements of SSFI, psoas major FI, multifidus FI, and erector spinae FI. Independent sample t-tests were used to compare continuous variables, including age, BMI, SSFI, frailty index, psoas major FI, multifidus FI, and erector spinae FI. The chi-square test was applied to categorical variables, including sex, smoking status, alcohol consumption, diabetes, and hypertension. Subsequently, Pearson linear correlation analysis was performed to assess the relationships between SSFI1 and other variables. Multiple linear regression analyses were then conducted to further evaluate the independent associations of SSFI1 with multifidus FI, erector spinae FI, frailty index, and vertebral bone quality (VBQ). In addition, receiver operating characteristic (ROC) curve analysis was performed to evaluate the predictive performance of the model. All statistical analyses were performed using SPSS 27.0 (IBM, Armonk, New York, USA). A two-tailed P value < 0.05 was considered statistically significant.
Results
Baseline characteristics comparison
Compared to the control group, osteoporosis patients were older. No significant differences were observed between the two groups regarding smoking history, alcohol consumption, diabetes, or hypertension. Compared with the control group, patients with osteoporosis showed significantly lower SSFI1 values, increased fatty infiltration of the psoas major, multifidus, and erector spinae muscles, and a higher frailty index. No significant differences were observed between groups for SSFI2, SSFI3, or SSFI4 (P > 0.05). Compared with the control group, patients with osteoporosis had a lower SFTT, but the difference was not statistically significant (P > 0.05) (Table 1).
Table 1.
Comparison of baseline demographics between the osteoporosis group and the normal bone mineral density group
| N | Osteoporosis（63） | Normal（112） | P | |
|---|---|---|---|---|
| Age ( year ) | 62.53 ± 7.55 | 64.27 ± 7.50 | 59.99 ± 6.93 | ＜0.001 |
| BMI ( kg/㎡ ) | 26.31 ± 4.02 | 26.11 ± 4.11 | 26.59 ± 3.90 | 0.437 |
| Gender | 175 | 0.103 | ||
| man | 55 | 15 ( 23.8% ) | 40 ( 35.7% ) | |
| female | 120 | 48 ( 76.2% ) | 72 ( 64.3% ) | |
| Smoking | 175 | 0.684 | ||
| Yes | 10 | 3 ( 4.8% ) | 7( 6.3% ) | |
| No | 165 | 60 ( 95.2% ) | 105( 93.8% ) | |
| Drinking | 175 | 0.890 | ||
| Yes | 6 | 2 ( 3.4% ) | 4( 3.6% ) | |
| No | 169 | 61 ( 96.8% ) | 108( 96.4% ) | |
| Diabetes | 175 | 0.156 | ||
| Yes | 37 | 17 ( 27.0% ) | 20( 17.9% ) | |
| No | 138 | 46 ( 73.0% ) | 92( 82.1% ) | |
| Hypertension | 175 | 0.221 | ||
| Yes | 92 | 37 ( 58.7% ) | 55 ( 49.1% ) | |
| No | 83 | 26 ( 41.3% ) | 57 ( 50.9% ) | |
| SFTT ( mm ) | 1.45 ± 0.74 | 1.38 ± 0.66 | 1.57 ± 0.84 | 0.096 |
| SSFI | ||||
| SSFI1 | 0.19 ± 0.21 | 0.13 ± 0.14 | 0..21 ± 0.23 | 0.002 |
| SSFI2 | 0.16 ± 0.17 | 0.14 ± 0.15 | 0.19 ± 0.20 | 0.075 |
| SSFI3 | 0.21 ± 0.21 | 0.19 ± 0.19 | 0.22 ± 0.23 | 0.259 |
| SSFI4 | 0.31 ± 0.27 | 0.29 ± 0.23 | 0.33 ± 0.31 | 0.284 |
| FI ( % ) | ||||
| psoas | 12.96 ± 7.16 | 13.85 ± 7.64 | 11.65 ± 6.21 | 0.046 |
| multifidus | 22.82 ± 8.80 | 24.22 ± 7.55 | 22.02 ± 9.37 | 0.113 |
| erector spinae | 18.42 ± 6.88 | 24.22 ± 7.54 | 17.26 ± 6.52 | 0.003 |
| Frailty Index | 1.75 ± 1.27 | 2.24 ± 1.32 | 1.61 ± 1.30 | 0.002 |
| VBQ | 3.55 ± 0.53 | 3.77 ± 0.47 | 3.21 ± 0.42 | ＜0.001 |
Gender-specific analysis revealed that SFTT exhibited sex-related differences in the assessment of bone mineral density among patients with osteoporosis, likely reflecting the distinct patterns of fat accumulation between males and females during aging. In contrast, the adjusted SSFI showed statistically significant differences in both sexes, suggesting that SSFI1 may provide a more stable indicator for evaluating bone health across genders (Table 2).
Table 2.
Comparison of SFTT and SSFI between osteoporotic and non-osteoporotic patients stratified by sex
| man | P | feman | P | |||
|---|---|---|---|---|---|---|
| Osteoporosis | Normal | Osteoporosis | Normal | |||
| SFTT | 1.17 ± 0.37 | 1.32 ± 0.63 | 0.397 | 1.30 ± 0.65 | 1.62 ± 0.84 | 0.032 |
| SSFI 1 | 0.06 ± 0.03 | 0.14 ± 0.16 | 0.005 | 0.15 ± 0.15 | 0.26 ± 0.26 | 0.009 |
| SSFI 2 | 0.16 ± 0.19 | 0.13 ± 0.15 | 0.537 | 0.12 ± 0.13 | 0.19 ± 0.20 | 0.026 |
| SSFI 3 | 0.23 ± 0.25 | 0.17 ± 0.16 | 0.344 | 0.29 ± 0.21 | 0.35 ± 0.33 | 0.221 |
| SSFI 4 | 0.33 ± 0.30 | 0.27 ± 0.17 | 0.326 | 0.16 ± 0.15 | 0.24 ± 0.24 | 0.041 |
Pearson correlation analysis of SSFI1 with other variables
Pearson correlation analysis showed that SSFI1 was significantly associated with several clinical variables. SSFI1 was negatively correlated with age (r = − 0.187, P = 0.013), multifidus FI (r = − 0.174, P = 0.022), erector spinae FI (r = − 0.193, P = 0.011), frailty index (r = − 0.164, P = 0.030), and VBQ score (r = − 0.191, P = 0.011). In contrast, SSFI1 showed positive correlations with femur BMD (r = 0.260, P = 0.006) and lumbar BMD (r = 0.154, P = 0.042). No significant correlation was observed between SSFI1 and psoas FI (r = 0.051, P = 0.503) (Table 3).
Table 3.
Pearson correlations between SSFI1 and various variables
| Variable | r | P |
|---|---|---|
| Age | -.0.187 | 0.013 |
| Femur BMD | 0.26 | 0.006 |
| Lumbar BMD | 0.154 | 0.042 |
| Psoas FI | 0.051 | 0.503 |
| Multifidus FI | -0.174 | 0.022 |
| Erector spinae FI | -0.193 | 0.011 |
| Frailty Index | -0.164 | 0.03 |
| VBQ | -0.191 | 0.011 |
Multivariate linear regression analysis of the associations between SSFI1 and paraspinal muscle fatty infiltration, VBQ, and the frailty index.
Multivariate linear regression analysis demonstrated that age was independently associated with multifidus FI (Table 4), erector spinae FI (Table 5), frailty index (Table 6), and VBQ (Table 7) (all P < 0.05). SSFI1 was negatively associated with VBQ (β = −0.183, P = 0.038), but showed no significant association with paraspinal muscle FI or frailty index. BMI was not significantly associated with any of the outcomes.
Table 4.
Multivariate linear regression analysis of multifidus FI
| Variable | β | P | VIF |
|---|---|---|---|
| Age | 0.230 | 0.002 | 1.039 |
| BMI | -0.115 | 0.185 | 1.407 |
| SSFI 1 | -0.069 | 0.430 | 1.428 |
Table 5.
Multivariate linear regression analysis of erector spinae FI
| Variable | β | P | VIF |
|---|---|---|---|
| Age | 0.241 | 0.001 | 1.039 |
| BMI | -0.119 | 0.168 | 1.407 |
| SSFI 1 | -0.084 | 0.333 | 1.428 |
Table 6.
Multivariate linear regression analysis of frailty index
| variable | β | P | VIF |
|---|---|---|---|
| Age | 0.368 | ＜0.001 | 1.039 |
| BMI | 0.088 | 0.291 | 1.407 |
| SSFI 1 | -0.143 | 0.090 | 1.428 |
Table 7.
Multivariate linear regression analysis of VBQ
| variable | β | P | VIF |
|---|---|---|---|
| Age | 0.209 | 0.006 | 1.039 |
| BMI | 0.059 | 0.495 | 1.407 |
| SSFI 1 | -0.183 | 0.038 | 1.428 |
ROC analysis of SSFI1 for predicting osteoporosis
SSFI1 showed a moderate ability to discriminate osteoporosis, with an AUC of 0.642. At the optimal cut-off value of 0.29, the sensitivity and specificity were 0.696 and 0.921, respectively, indicating that SSFI1 may have potential utility in distinguishing osteoporosis from non-osteoporotic individuals (Fig. 2).
Fig. 2.
ROC analysis of SSFI1 for discriminating osteoporosis
Discussion
Our findings suggest that higher SSFI1 levels are associated with a lower prevalence of osteoporosis. While recent studies have highlighted the utility of upper lumbar SFTT in predicting lumbar degeneration and low back pain [9], our results suggest that the SSFI1 demonstrates superior performance compared with SFTT in distinguishing osteoporosis from non-osteoporotic individuals. We hypothesize that this enhanced predictive capability may be attributed to gender-specific variations in fat deposition patterns during aging, which may limit the reliability of SFTT in evaluating skeletal health. In addition, SSFI1 was inversely associated with VBQ, suggesting its potential as an imaging marker of vertebral bone quality. Furthermore, our findings suggest that when SSFI1 is lower than 0.29, individuals may be at an increased risk of osteoporosis, indicating that greater attention should be paid to early monitoring and preventive strategies in this population.
Interestingly, SSFI measured at the L1 level, but not at lower lumbar sites (SSFI2–SSFI4), was predictive of osteoporosis. This finding may be explained by several factors. First, the L1 vertebra is located at the thoracolumbar junction, a biomechanically vulnerable region where alterations in bone quality may be more readily detected [22]. Second, adipose tissue distribution varies along the lumbar spine, and subcutaneous fat at the upper lumbar levels may better reflect trunk adiposity than lower segments [22]. Finally, the anatomical composition of paraspinal muscles is relatively stable at the upper lumbar spine, which may provide more reliable measurements of muscle fat infiltration [22].
Current research has confirmed the importance of fat distribution rather than total body fat in maintaining bone health [24, 25]. Subcutaneous fat is considered protective adipose tissue because it produces leptin and adiponectin, which enhance bone mass by stimulating osteogenic activity [26–28]. One of the causes of osteoporosis is the uncoupling of osteogenesis and adipogenesis [29, 30]. Osteoblasts and adipocytes originate from a common progenitor cell [30]. This study demonstrated that SSFI1 was positively correlated with bone mineral density and T-score. The VBQ score, a novel indicator for evaluating bone quality, is derived from T1-weighted MRI images and reflects the degree of vertebral fatty infiltration based on T1 signal intensity. SSFI1 showed a significant negative correlation with the VBQ score [23]. These findings suggest a potential link between SSFI1 and the uncoupling of adipogenic and osteogenic differentiation in mesenchymal stem cells.
Recent studies have reported that the VBQ score may serve as a comprehensive imaging biomarker for assessing both bone quality and paraspinal muscle status [23]. The degree of fat infiltration in the erector spinae and multifidus muscles is significantly correlated with osteoporosis [21, 31]. This study found that SSFI1 showed a significant positive correlation with the degree of fat infiltration in both muscles. This suggests that SSFI1 is an effective method of assessing fat infiltration in the paraspinal muscles of the lower lumbar spine.
Additionally, muscles and bones are closely coupled as a musculoskeletal unit. Some studies have found that in osteoporosis patients, fat is primarily deposited in muscles and bone marrow. This study observed a higher degree of fat infiltration in the paraspinal muscles of osteoporosis patients. Other related studies have suggested that dysfunction of subcutaneous fat leads to the accumulation of visceral fat [27]. Our study further revealed a negative correlation between SSFI1 and paraspinal muscle fat infiltration. Low subcutaneous fat lack sufficient adipocytes and have limited fat storage capacity, leading to fat deposition within and around organs. Moreover, related studies have found that the degree of paraspinal muscle fat infiltration is associated with preoperative frailty in patients [28]. Consistent with these findings, SSFI1 was negatively correlated with the frailty index, suggesting a potential association between lower subcutaneous fat levels and greater frailty.
Taken together, our study suggests that SSFI1 serves as a practical, non-invasive tool for quantifying the pathogenic interplay between adipose tissue, muscle, and bone. Implementing SSFI1 assessment could aid in the pre-symptomatic identification of patients at risk for osteoporosis and frailty, going beyond conventional SFTT. Future longitudinal studies should investigate whether therapeutic strategies designed to preserve subcutaneous fat are effective in curbing the progression of bone loss and functional decline.
Limitations
This study has several limitations. First, it is a cross-sectional analysis; therefore, a causal relationship between subcutaneous fat thickness and osteoporosis or frailty cannot be established. Second, longitudinal data on fat distribution and changes in musculoskeletal health were not available, preventing assessment of temporal relationships. Third, lumbar spine imaging was performed in the supine position, which may slightly compress subcutaneous fat and affect measurement accuracy.
Conclusion
SSFI1 was significantly reduced in patients with osteoporosis and was closely associated with paraspinal muscle fatty infiltration, bone mineral density, frailty index, and VBQ. In addition, SSFI1 demonstrated potential value in distinguishing osteoporosis from non-osteoporotic individuals. These findings suggest that SSFI1 may serve as a simple and accessible imaging biomarker for evaluating bone quality and may provide a novel approach for the early identification and risk stratification of osteoporosis.
Acknowledgements
Not applicable.
Abbreviations
- SSFI
Standardized subcutaneous fat index
- VBQ
Vertebral bone quality
- FI
Fatty infiltration
Authors’ contributions
Xiaowei Lian: Methodology, Investigation, Data curation, Conceptualization, Visualization, Writing-original draft, Writing-review & editing. Yilai Li: Writing-original draft, Writing-review & editing. Zhizhou Yang: Writing-review & editing. Ranxu Yang: Data curation, Writing-original draft. Wenshuai Li: Investigation, Data curation. Yunsheng Wang: Writing-review & editing, Visualization. Feng Wang: Supervision, Writing-review & editing. Linfeng Wang: Writing-review & editing, Conceptualization, Funding acquisition, Project administration, Resources, Supervision.
Funding
No funding has been received by any agency in relation to this research.
Data availability
Data cannot be provided due to identifying information of participants but are available from the corresponding author on reasonable request.
Declarations
Ethics approval and consent to participate
The study was approved by the Ethics Committee of the Third Hospital of Hebei Medical University (No. Ke2025-402-1). All procedures were conducted in accordance with the Declaration of Helsinki (1964) and its subsequent amendments or comparable ethical standards. Written informed consent was obtained from all participants.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Narayanan R, Tarawneh OH, Trenchfield D, Meade MH, Lee Y, Opara O et al. Preoperative hounsfield units predict pedicle screw loosening in osteoporotic patients following short segment lumbar fusion. SPINE. 2024. [DOI] [PubMed]
- 2.Morin SN, Leslie WD, Schousboe JT, Osteoporosis. A Review. JAMA-J AM MED ASSOC. 2025 2025–06. [DOI] [PubMed]
- 3.LeBoff MS, Greenspan SL, Insogna KL, Lewiecki EM, Saag KG, Singer AJ et al. The clinician’s guide to prevention and treatment of osteoporosis. Osteoporosis Int. 2022;33(10):2049–102. [DOI] [PMC free article] [PubMed]
- 4.Yuan C, Yu XT, Wang J, Shu B, Wang XY, Huang C, et al. Multi-modal molecular determinants of clinically relevant osteoporosis subtypes. CELL DISCOV. 2024;10(1):28. 2024-03-12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Samartzis D, Karppinen J, Mok F, Fong DY, Luk KD, Cheung KM. A population-based study of juvenile disc degeneration and its association with overweight and obesity, low back pain, and diminished functional status. J Bone Joint Surg Am. 2011;93(7):662–70. [DOI] [PubMed]
- 6.Nguyen TT, Corvera S. Adipose tissue as a linchpin of organismal ageing. Nat Metab. 2024;6(5):793–807. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Ou M, Zhang H, Tan P, Zhou S, Li Q. Adipose tissue aging: mechanisms and therapeutic implications. Cell Death Dis. 2022;13(4):300. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Liuke M, Solovieva S, Lamminen A, Luoma K, Leino-Arjas P, Luukkonen R, et al. Disc degeneration of the lumbar spine in relation to overweight. INT J Obes. 2005;29(8):903–8. 2005-08-01. [DOI] [PubMed] [Google Scholar]
- 9.Ozcan-Eksi EE, Kara M, Berikol G, Orhun O, Turgut VU, Eksi MS. A new radiological index for the assessment of higher body fat status and lumbar spine degeneration. SKELETAL RADIOL. 2022;51(6):1261–71. 2022-06-01. [DOI] [PubMed] [Google Scholar]
- 10.Samartzis D, Karppinen J, Chan D, Luk KD, Cheung KM. The association of lumbar intervertebral disc degeneration on magnetic resonance imaging with body mass index in overweight and obese adults: a population-based study. Arthritis Rheum. 2012;64(5):1488–96. 2012-05-01. [DOI] [PubMed] [Google Scholar]
- 11.Samartzis D, Karppinen J, Mok F, Fong DY, Luk KD, Cheung KM. A population-based study of juvenile disc degeneration and its association with overweight and obesity, low back pain, and diminished functional status. J Bone Joint Surg Am. 2011;93(7):662–70. [DOI] [PubMed]
- 12.Cornier MA, Despres JP, Davis N, Grossniklaus DA, Klein S, Lamarche B et al. Assessing adiposity: a scientific statement from the American Heart Association. Circulation. 2011;124(18):1996–2019. [DOI] [PubMed]
- 13.Coutinho T, Goel K, Correa DSD, Kragelund C, Kanaya AM, Zeller M, et al. Central obesity and survival in subjects with coronary artery disease: a systematic review of the literature and collaborative analysis with individual subject data. J AM COLL CARDIOL. 2011;57(19):1877–86. 2011-05-10. [DOI] [PubMed] [Google Scholar]
- 14.Dutton DJ, McLaren L. The usefulness of corrected body mass index vs. self-reported body mass index: comparing the population distributions, sensitivity, specificity, and predictive utility of three correction equations using Canadian population-based data. BMC Public Health. 2014;14:430. [DOI] [PMC free article] [PubMed]
- 15.Gurunathan U, Myles PS. Limitations of body mass index as an obesity measure of perioperative risk. Brit J Anaesth. 2016;116(3):319–21. [DOI] [PubMed]
- 16.Roubenoff R. Applications of bioelectrical impedance analysis for body composition to epidemiologic studies. Am J Clin Nutr. 1996;64(3 Suppl):S459–62. [DOI] [PubMed]
- 17.Berikol G, Eksi MS, Aydin L, Borekci A, Ozcan-Eksi EE. Subcutaneous fat index: a reliable tool for lumbar spine studies. Eur Radiol. 2022;32(9):6504–13. [DOI] [PubMed]
- 18.Lin Y, Zhong X, Lu D, Yao W, Zhou J, Wu R, et al. Association of visceral and subcutaneous fat with bone mineral density in US adults: a cross-sectional study. Sci Rep-UK. 2023;13(1):10682. [DOI] [PMC free article] [PubMed]
- 19.Fox CS, Massaro JM, Hoffmann U, Pou KM, Maurovich-Horvat P, Liu CY et al. Abdominal visceral and subcutaneous adipose tissue compartments: association with metabolic risk factors in the Framingham Heart Study. Circulation. 2007;116(1):39–48. [DOI] [PubMed]
- 20.Kim JH, Choi HJ, Ku EJ, Hong AR, Kim KM, Kim SW et al. Regional body fat depots differently affect bone microarchitecture in postmenopausal Korean women. Osteoporosis Int. 2016;27(3):1161–8. [DOI] [PubMed]
- 21.Han G, Zou D, Liu Z, Zhou S, Li W, Gong C, Sun Z, Li W. Paraspinal muscle characteristics on MRI in degenerative lumbar spine with normal bone density, osteopenia and osteoporosis: a case-control study. BMC Musculoskel Dis. 2022;23(1):73. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Pickhardt PJ, Pooler BD, Lauder T, Del Rio AM, Bruce RJ, Binkley N. Opportunistic screening for osteoporosis using abdominal computed tomography scans obtained for other indications. Ann Intern Med. 2013;158(8):588–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Li W, Wang F, Chen J, Zhu H, Tian H, Wang L. MRI–based vertebral bone quality score is a comprehensive index reflecting the quality of bone and paravertebral muscle. Spine Journal: Official J North Am Spine Soc. 2024;24(3):472–8. [DOI] [PubMed] [Google Scholar]
- 24.Sun Q, Cui L, Liu Y, Pang Q, Chi Y, Jiajue R, Liu W, Jiang Y, Wang O, Li M et al. Association of body fat distribution with bone mineral density: evidence from observational and mendelian randomization analyses. Bone. 2025:117742. [DOI] [PubMed]
- 25.Zhu W, Guan W, Chen B, Lu Y, Li J, Yuan X, Li W, Yu F, Liu J, Yin H, et al. Association between bone marrow adipose tissue, abdominal adipose tissue distribution, and volumetric bone mineral density in the Chinese adult population: a retrospective cohort study. PeerJ. 2025;13:e20446. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Crivelli M, Chain A, Da SI, Waked AM, Bezerra FF. Association of visceral and subcutaneous fat mass with bone density and vertebral fractures in women with severe obesity. J Clin Densitom. 2021;24(3):397–405. [DOI] [PubMed]
- 27.Zhang W, Ma X, Xue P, Gao Y, Wu X, Zhao J et al. Associations between fat distribution and volumetric bone mineral density in Chinese adults. Endocrine. 2014;47(3):862–8. [DOI] [PubMed]
- 28.Wang W, Huang W, Liu P, Fu S, Zhang M, Zhang X et al. Lower subcutaneous fat index predicts bone metastasis in breast cancer. Cancer Biomark. 2023;38(1):121–30. [DOI] [PMC free article] [PubMed]
- 29.Ru J, Wang Y. Osteocyte apoptosis: the roles and key molecular mechanisms in resorption-related bone diseases. Cell Death Dis. 2020;11(10):846. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Andreev D, Liu M, Weidner D, Kachler K, Faas M, Grüneboom A, Schlötzer-Schrehardt U, Muñoz LE, Steffen U, Grötsch B, et al. Osteocyte necrosis triggers osteoclast-mediated bone loss through macrophage-inducible C-type lectin. J Clin Investig. 2020;130(9):4811–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.zcan-Ekşi ö EE, Osama M, ^ Ekşi M. Crosstalk between osteoporosis and fat-infiltrated psoas at the upper lumbar levels. Eur Spine Journal: Official Publication Eur Spine Soc Eur Spinal Deformity Soc Eur Sect Cerv Spine Res Soc. 2025;34(11):5023–33. [DOI] [PubMed]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Data cannot be provided due to identifying information of participants but are available from the corresponding author on reasonable request.