You are here

  • Home
  • Online First
  • MRI-derived abdominal adipose tissue is associated with multisite and widespread chronic pain

Article Text

Article menu
Chronic and interventional pain
MRI-derived abdominal adipose tissue is associated with multisite and widespread chronic pain
Free
  1. Zemene Demelash Kifle1,
  2. Jing Tian1,
  3. Dawn Aitken1,
  4. Phillip E Melton1,2,
  5. Flavia Cicuttini3,
  6. Graeme Jones1 and
  7. Feng Pan1
    1. 1University of Tasmania Menzies Institute for Medical Research, Hobart, Tasmania, Australia
    2. 2School of Global and Population Health, The University of Western Australia, Perth, Western Australia, Australia
    3. 3Monash University School of Public Health and Preventive Medicine, Melbourne, Victoria, Australia
    1. Correspondence to Dr Feng Pan, University of Tasmania Menzies Institute for Medical Research, Hobart, Tasmania, Australia; feng.pan{at}utas.edu.au

    Abstract

    Introduction Musculoskeletal pain typically occurs in multiple sites; however, no study has examined whether excessive visceral and subcutaneous adipose tissue are associated with musculoskeletal pain. This study therefore aimed to describe the associations between MRI-derived abdominal adipose tissue and multisite and widespread chronic musculoskeletal pain.

    Methods Data from the UK Biobank, a large prospective, population-based cohort study, were used. Abdominal MRI scans were performed at two imaging visits to quantify visceral adipose tissue and subcutaneous adipose tissue. Pain in the neck/shoulder, back, hip, knee or ‘all over the body’ was assessed at the corresponding visits. Mixed-effects ordinal/multinomial/logistic regression models were used for the analyses.

    Results A total of 32 409 participants were included (50.8% women, mean age 55.0±7.4 years). In multivariable analyses, there was a dose–response association of visceral adipose tissue, subcutaneous adipose tissue and their ratio with the number of chronic pain sites in both women (visceral adipose tissue: OR 2.04 per SD (95% CI 1.85 to 2.26); subcutaneous adipose tissue: OR 1.60 (95% CI 1.50 to 1.70); and their ratio: OR 1.60 (95% CI 1.37 to 1.87)) and men (visceral adipose tissue: OR 1.34 (95% CI 1.26 to 1.42); subcutaneous adipose tissue: OR 1.39 (95% CI 1.29 to 1.49); and their ratio: OR 1.13 (95% CI 1.07 to 1.20)). Higher levels of adipose tissue were also associated with greater odds of reporting chronic pain in both sexes. The effect estimates of these adipose measures were relatively larger in women than in men.

    Conclusion Abdominal adipose tissue was associated with chronic musculoskeletal pain, suggesting that excessive and ectopic fat depositions may be involved in the pathogenesis of multisite and widespread chronic musculoskeletal pain. The identified stronger effects in women than men may reflect sex differences in fat distribution and hormones.

    • Pain Management
    • CHRONIC PAIN
    • Back Pain

    Data availability statement

    Data from the UK Biobank is available to researchers by application via the UK Biobank online Access Management System.

    https://doi.org/10.1136/rapm-2024-105535

    Statistics from Altmetric.com

      Request Permissions

      If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

      What is already known on this topic

      • Obesity has been linked to musculoskeletal pain; however, the relationship between excessive visceral and subcutaneous adipose tissue and chronic pain in multiple sites and widespread chronic pain has not been investigated.

      What this study adds

      • Abdominal adipose tissue is associated with chronic pain in multiple sites and widespread chronic pain. There is a sex difference in the relationship between adipose tissue and chronic musculoskeletal pain.

      How this study might affect research, practice or policy

      • Excessive abdominal adiposity is implicated in the pathogenesis of chronic pain, including visceral and subcutaneous adipose tissue. Therefore, reducing abdominal adiposity may be considered a target for chronic pain management, particularly in those with pain in multiple sites and widespread pain.

      Introduction

      Obesity is a complex public health issue and has nearly tripled since 1975.1 Overweight and obesity are linked to multiple chronic diseases, contributing to a decrease in life expectancy.2 Musculoskeletal pain is another common health concern, causing poor quality of life, decreased physical function and disability.3 According to the Global Burden of Disease 2019 data, 1.71 billion people worldwide experienced chronic musculoskeletal pain.4 Multisite musculoskeletal pain, which is often characterised as having ≥2 pain sites, is more prevalent than single-site pain. Previous studies, including our own, have shown that a substantial proportion of middle-aged and older adults ranging from 41% to 75% experience multisite musculoskeletal pain.5 6

      Although obesity and musculoskeletal pain are linked,7 the directionality of their relationship and mechanisms are not well understood. Most previous studies exploring the association between obesity and musculoskeletal pain were cross-sectional and used body mass index (BMI) to classify obesity,7 but BMI cannot distinguish distribution of body fat. We have previously reported that fat mass measured by dual X-ray absorptiometry (DXA) was associated with multisite pain in a cohort of 1099 community-dwelling older adults.5 In addition, while substantial sex differences in body fat distribution have been documented,8 most previous studies did not report gender-specific associations between obesity and musculoskeletal pain.

      There is evidence that two primary adipose depots, visceral adipose tissue (VAT) and subcutaneous adipose tissue (SAT), have a distinct metabolic profile.9 The VAT plays a more important role in the development of obesity-related chronic diseases than fat stored in other areas of the body.10 Evidence suggests that SAT might improve metabolism,11 whereas other data suggest that excess SAT causes metabolic syndrome.12 So far, only one study that investigated the associations of DXA-derived VAT and SAT with number of painful joints found that VAT but not SAT was associated with the number of painful joints.13

      Total body adipose tissue and different depots of adipose tissue can be visualised and quantified using MRI, making it an attractive imaging option for clinical and research purposes.14 Currently, no study has investigated the relationship between MRI-derived adipose tissue and multisite chronic musculoskeletal pain. Therefore, this study aimed to investigate the associations between MRI-derived abdominal VAT, SAT, and their ratio and multisite and widespread chronic musculoskeletal pain. We hypothesised that greater MRI-derived adipose tissue was dose-responsively associated with an increased number of chronic pain sites and presence of chronic pain.

      Methods

      Data source and study population

      This study used data from the UK Biobank study, a large population-based prospective cohort study with the aim to improve the prevention, diagnosis and treatment of a wide range of serious and life-threatening illnesses. A detailed description of this study has been previously published.15 In brief, all participants were registered with the UK National Health Service and lived within 25 miles of one of 22 assessment centres across England, Wales and Scotland. Overall, half a million participants aged 40–69 years were recruited between 2006 and 2010, with a participation rate of 5.5%. At baseline, participants provided information on a range of sociodemographic, physical, lifestyle and health-related factors via questionnaires, cognitive measures, physical measures and blood samples. Participants were also invited to attend the initial repeat assessment at the UK Biobank assessment centers between 2012 and 2013.

      After baseline and the initial repeat assessment, the UK Biobank imaging study was launched in 2014 and a repeat imaging visit was conducted in 2019. This study included participants with available abdominal MRI-measured body composition data during the first imaging visit and/or the first repeat imaging visit (figure 1).

      Figure 1
      Figure 1

      Flowchart of participants selected for analyses. After baseline recruitment (2006–2010), the initial repeated measures were obtained (2012–2013). Subsequently, participants underwent periodic reconnection for additional data assessments, including the first imaging visit (2014 – ongoing), and the first repeat imaging visit (2019 – ongoing).

      Abdominal MRI measures (exposures)

      Abdominal MRI scans at the first imaging visit and the first repeat imaging visit with 2.7 years apart were performed to quantify the composition of VAT and SAT. All scans were performed on a Siemens Aera 1.5 T scanner (Syngo MR D13; Siemens, Erlangen, Germany) following the 6 min dual-echo Dixon Vibe protocol from neck to knees.16 The VAT and SAT volumes were extracted from images of fat and water using validated and standardised methods and their quantities were then determined. Analyses were conducted in Advanced MR Analytics Profiler (AMRA AB, Linköping, Sweden) using the automated segmentation software and were visually verified for accuracy by an analysis engineer.17 The VAT/SAT ratio was calculated as a metric assessing relative body fat composition and distinguishing the tendency to store fat viscerally versus subcutaneously. The intraclass correlation coefficients for VAT and SAT volume were 0.997 and 0.996, respectively.18

      Pain assessment (outcomes)

      Participants were asked at the assessment centers, through a touchscreen pain questionnaire, if they had experienced any pain in the last month that interfered with their regular activities. If participants reported pain, they were then asked about musculoskeletal pain in different body parts including neck/shoulder, back, hip, knee or ‘all over the body’. Participants who reported pain ‘all over the body’ were not provided with further options to select any additional specific pain sites. If participants reported pain at any of the sites or ‘all over the body’, they were further asked if the pain had persisted for 3 months or longer. Chronic pain was defined as having pain that lasted more than 3 months. In this study we used musculoskeletal pain questions data during the first imaging visit and the first repeat imaging visit, and categorised participants into six groups according to their number of chronic pain sites: no chronic pain, chronic pain in one, two, three, four sites, and those with chronic pain ‘all over the body’. Additionally, participants were categorised into two groups: those experiencing chronic pain and those without.

      Covariates

      Based on existing literature within the field,5 13 19–26 covariates measured at the first imaging visit that are related to both pain and obesity were considered. These included:

      1. Adhering to common epidemiological practices, we considered a range of sociodemographic factors, such as age (in continuous years), household income, highest education qualification, and ethnicity.5 13

      2. Standing height was measured using the Seca 202 device; this adjustment is crucial as adiposity can vary significantly depending on an individual’s height, ensuring a fair comparison between participants.24

      3. Lifestyle factors have been reported to be associated with obesity and pain,7 22 23 including meeting recommended moderate/vigorous physical activity (yes/no), frequency of alcohol consumption (daily or almost daily, 3–4 times a week, 1–2 times a week, 1–3 times/month, occasional drinker and never), and smoking status (current, past, and never).

      4. Health factors were considered as several health-related factors have been reported to affect the association between obesity and pain,21 22 including the presence of any comorbidities such as diabetes, hypertension, cardiovascular disease, stroke, cancer and lung disease.

      5. Psychological problems were considered since there is increasing interest in the role of psychological well-being in obesity7 25 and their links to pain.7 22 26 Participants were classified as having a psychological problem if they answered ‘yes’ to either of the following two questions: “Have you ever seen a general practitioner for nerves, depression, tension or anxiety?” or “Have you ever seen a psychiatrist for nerves, depression, tension or anxiety?”.

      6. Sleep duration was included due to its documented associations with both obesity and pain.22 26 Sleep duration responses, measured in hourly increments, were categorised into: less than recommendation (<7 hours/day), meeting the recommendation (≥7 and ≤8 hours/day), and more than recommendation (>8 hours).

      Statistical analyses

      The categorical and continuous variables were described using percentages (number) and means±SD, respectively. T-tests for continuous variables and χ2 tests for categorical variables were used to compare the characteristics of participants with abdominal MRI measures to the rest of the UK Biobank cohort, as well as participants included at the first imaging visit compared with those at the first repeat imaging visit. Analysis of variance was used to determine p values for trends in VAT, SAT, VAT/SAT ratio and BMI across pain groups. A mixed-effects ordinal logistic regression model was used to explore the associations of VAT, SAT, VAT/SAT ratio and BMI with the number of chronic pain sites, providing estimates of cumulative ORs to capture general trends. Additionally, a mixed-effects multinomial logistic regression model was used to examine the relative risk ratios (RRRs) for each pain category relative to a reference category. In further analyses we investigated the associations between VAT, SAT, VAT/SAT ratio and BMI with the presence of chronic pain using a mixed-effects logistic regression model. This method can use all available data from participants and take into account repeated participant observations.5 The following models were performed: Model 1: VAT, SAT and VAT/SAT ratio adjusted for age and height, BMI adjusted for age only; Model 2: model 1 plus ethnicity, house income, highest education level, alcohol frequency, smoking status, physical activity, comorbid conditions, sleep duration, psychological problems and follow-up time, ensuring that the potential confounding factors were considered during the investigation of associations between adiposity and pain.

      The interactions between each adipose measure and BMI with sex were tested and significant interactions were found. Therefore, stratified analyses by sex were performed. Additionally, we tested the interaction between each adipose measure and BMI with follow-up time to assess whether the effect of VAT, SAT, VAT/SAT ratio and BMI on pain changed over time. We performed inverse probability weighting analyses to assess the impact of missing data on our results. To directly compare RRR and OR, VAT, SAT, VAT/SAT ratio and BMI were standardised. All statistical analyses were performed with STATA software, version V.17.0 (Stata Corp, College Station, Texas, USA). A two-tailed p value <0.05 was considered statistically significant.

      Results

      Participant characteristics

      This study included participants who underwent MRI scans and had complete data on pain and covariates, resulting in 32 409 participants at the first imaging visit and 638 participants at the repeat imaging visit, with a mean follow-up of 2.7±1.1 years (figure 1). The characteristics of participants at the first imaging visit by sex are shown in table 1. The mean age of the participants was 55.0±7.4 years, 50.8% were women, and mean BMI was 26.4±4.2 kg/m2. On average, women were younger, had lower BMI, VAT and VAT/SAT ratio but higher SAT and a greater number of chronic pain sites than men. The characteristics of women and men at the first imaging visit across the number of chronic pain sites are also shown in online supplemental tables 1 and 2, respectively.

      Supplemental material

      View this table:
      Table 1

      Characteristics of participants at the first imaging visit

      Figure 2 shows the relationship between adipose measures and BMI with the number of chronic pain sites by sex at the first imaging visit. The VAT, SAT, VAT/SAT ratio and BMI exhibited an increasing trend with increasing number of chronic pain sites in both women and men.

      Figure 2
      Figure 2

      Association between (A) visceral adipose tissue (VAT), (B) subcutaneous adipose tissue (SAT), (C) VAT/SAT ratio,(D) body mass index (BMI) and number of chronic pain sites in women and men at the first imaging visit. The bar graph represents the mean values of VAT, SAT and VAT/SAT ratio, and error bars indicate SDs. P for trend determined by analysis of variance.

      MRI-derived adipose measures and number of chronic musculoskeletal pain sites

      Tables 2 and 3 show the associations between MRI-derived adipose measures and BMI with number of chronic pain sites by sex. There was a dose–response relationship between adipose measures and BMI and number of chronic pain sites (table 2). In the age and height-adjusted model (Model 1), VAT, SAT, VAT/SAT ratio and BMI were associated with chronic pain in one, two, three, four sites and ‘all over the body’ compared with those without chronic pain in both women and men. In men the VAT/SAT ratio was not associated with chronic pain in four sites and ‘all over the body’ (table 3). After further adjustment for ethnicity, house income, highest education level, alcohol frequency, smoking status, physical activity, comorbid conditions, sleep duration, psychological problems and follow-up time (Model 2), the associations remained significant while the effect sizes were slightly attenuated in both women and men.

      View this table:
      Table 2

      Mixed-effects ordinal logistic regression for the association between MRI-derived adipose measures and body mass index and number of chronic pain sites

      View this table:
      Table 3

      Mixed-effects multinomial logistic regression for the association between MRI-derived adipose measures and body mass index and number of chronic pain sites

      View this table:
      Table 4

      Association between MRI-derived adipose measures and body mass index and chronic pain

      MRI-derived adipose measures and chronic musculoskeletal pain

      The associations between MRI-derived adipose measures and BMI and chronic pain by sex are shown in table 4. In the age and height-adjusted model, greater VAT, SAT, VAT/SAT ratio and BMI were associated with greater odds of reporting chronic pain compared with those without chronic pain in both sexes. After further adjustment for ethnicity, house income, highest education level, alcohol frequency, smoking status, physical activity, comorbid conditions, sleep duration, psychological problems and follow-up time, the associations remained statistically significant although the effect sizes were slightly reduced in both sexes.

      Further analyses

      Statistically significant interactions were found between each adipose measure and BMI with sex (tables 2 and 4). The estimates of the associations between adipose measures with the number of chronic pain sites and chronic pain were relatively larger in women than those in men (tables 2–4). No statistically significant interactions between VAT, SAT, VAT/SAT ratio and BMI with follow-up time were found, suggesting that the effects of adipose measures and BMI on the number of chronic pain sites and chronic pain did not change over time (data not shown). There were significant differences between the participants included in this analysis and the rest of the cohort in terms of sex, age, height, weight, BMI, ethnicity, meeting recommended physical activity, household income, highest education qualification, frequency of alcohol consumption, smoking status, sleep duration, psychological problems and presence of any comorbidities (see online supplemental table 3). Differences were also observed between the participants included at the first imaging visit and those at the first repeat imaging visit in terms of age, household income, highest education qualification, alcohol consumption and presence of any comorbidities (see online supplemental table 4). Sensitivity analyses using an inverse probability weighting method did not show substantial changes in our results (see online supplemental tables 5–7).

      Discussion

      This study found a dose–response association between abdominal VAT, SAT, VAT/SAT ratio and BMI with the number of chronic pain sites in both sexes. Additionally, greater VAT, SAT, VAT/SAT ratio and BMI were associated with greater odds of reporting chronic pain in both sexes. These associations persisted even after adjusting for age, height, ethnicity, house income, highest education level, alcohol frequency, smoking status, physical activity, comorbid conditions, sleep duration, psychological problems and follow-up time. These results suggest that excessive abdominal adipose tissue may be implicated in the pathogenesis of chronic musculoskeletal pain. Notably, the effect estimates of the associations of the adipose measures and BMI with the number of chronic pain sites and chronic pain were relatively larger in women than in men. To our knowledge, this study is the first to examine the associations between MRI-derived adipose measures and both the number of chronic pain sites and chronic pain.

      The study found a dose–response relationship between VAT and SAT with the number of chronic pain sites in both sexes. Additionally, greater VAT and SAT were associated with greater odds of reporting chronic pain in both sexes. These findings align in part with only one recent study with 5 years of follow-up among 2961 community-dwelling older adults with or at high risk for knee osteoarthritis. In that study, VAT but not SAT, measured by DXA, was associated with the number of painful joints.13 Our finding that SAT was associated with the number of chronic pain sites in both sexes could be attributed to our larger sample size, which provided greater power to detect differences. Additionally, differences in characteristics of the study populations, methods used to measure SAT, potential covariates considered and number of chronic pain sites assessed may also contribute to these discrepancies. Our findings appear to be partially supported by previous studies in which fat mass was used to examine the association between adiposity and multisite pain.5 27 28 Our group has reported that fat mass, measured by DXA, was associated with multisite pain in a cohort of 1099 community-dwelling older adults aged 50–79, with an average 5.1 years of follow-up.5 Similarly, two cross-sectional studies conducted by Yoo et al27 found associations between fat mass and fat–muscle mass ratio with widespread pain, while Brady et al28 reported that fat mass and fat mass index were associated with a greater number of lower body pain sites. Taken together, combining the findings from this study with previous research may suggest that excess adiposity may be involved in the pathogenesis of chronic musculoskeletal pain, and that VAT, SAT and their ratio could potentially serve as markers in managing chronic musculoskeletal pain.

      Moreover, the current study found that the VAT/SAT ratio was associated with the number of chronic pain sites and chronic pain in both sexes. While no direct research has investigated this specific relationship, several studies have found an association between the VAT/SAT ratio and cardiometabolic risks.29 30 This ratio may reflect a difference in the distribution of adipose tissue and the propensity for VAT relative to SAT. Therefore, our findings highlight a connection of fat deposition with chronic musculoskeletal pain. Several mechanisms have been proposed to explain adiposity-related musculoskeletal pain including biomechanical, inflammatory and metabolic factors. The accumulation of VAT and SAT has been linked to increased levels of cytokines (eg, interleukin-6, C-reactive protein, tumor necrosis factor) as well as adipokines (eg, leptin, visfatin).31 These inflammatory markers have been suggested to be implicated in the pathogenesis of musculoskeletal pain.7 Furthermore, inflammation may modify excitation thresholds and reactions of peripheral nerves, leading to both peripheral and central sensitisation,32 which could be an additional mechanism for adiposity-related meta-inflammatory processes contributing to musculoskeletal pain. Additionally, the link between ectopic fat and proinflammatory molecules involves complex interactions impacting inflammation.33 Dietary choices, including the balance of pro- and anti-inflammatory foods, influence inflammation, obesity and pain by affecting the microbiome, oxidative stress and diet-related pain sensitivity.34 Moreover, non-steroidal anti-inflammatory drugs, adiposity and pain have a bidirectional relationship where their effects on inflammation and pain can counteract each other.7 35 The sympathetic nervous system also modulates inflammation by balancing pro- and anti-inflammatory responses.36 These intricate interactions between dietary patterns, inflammation and sympathetic activation may predispose individuals to increased abdominal adiposity, potentially playing critical roles in the pathogenesis of chronic musculoskeletal pain.

      In this study women had more SAT and less VAT than men, and the magnitude of effect estimates for the association between VAT, SAT and their ratio and the number of chronic pain sites and chronic pain were relatively larger in women than in men. These findings appear to be supported by studies in cardiovascular and diabetes research which show that fat deposition, including VAT, SAT and their ratio, has a more pronounced impact on adverse cardiometabolic risk in women than in men.37 38 Despite having less VAT than men, women appear to have a high susceptibility to musculoskeletal pain. This may suggest sex differences in the relationship between adiposity and musculoskeletal pain. The observed sex differences in the effect estimates of adipose measures with number of chronic pain sites may be attributed to several mechanisms including the influence of sex hormones on adipose tissue distribution, function and storage.39 In women, SAT tends to expand primarily through an increase in adipocyte size rather than by proliferating new cells.40 This could indicate that SAT may not adequately expand to accommodate excess fat storage, resulting in the accumulation of fat in VAT or other ectopic depots. Additionally, adipocyte size plays an important role in regulating adipokine secretion, with larger adipocytes leading to an increased secretion of proinflammatory adipokines.7 Furthermore, women have been reported to exhibit a stronger inflammatory response than men, which could contribute to the sex differences in pain experience.41

      The strengths of this study include the fact that it is the first of its kind to examine the associations between MRI-derived adipose tissue and number of chronic pain sites in a large population-based sample. We used a highly specific and reproducible MRI-derived assessment of fat depots rather than anthropometric surrogate measures of body fat composition. However, this study has limitations that should be acknowledged. First, the pain questionnaire used in this study did not include an assessment of pain severity, so we were unable to examine the relationships between fat measures and pain severity. Second, MRI was conducted on two occasions; more visits may allow more information on patterns and fluctuations in the number of chronic pain sites. Third, although the application of the propensity score in our sensitivity analysis has improved the reliability of our findings, it is important to acknowledge that this approach does not mitigate the influence of unmeasured confounders or address the issue of potential bidirectional causality. Moreover, despite considering a range of covariates, identifying certain variables as either confounders, mediators or colliders is challenging due to the bidirectional relationships between some variables (ie, psychological issues and sleep disturbances and both adiposity and pain).7 26 Fourth, the use of only two brief screening questions to assess psychological issues may not fully capture the range of psychological status, potentially leading to under-reporting of subtle or complex psychological issues. This limitation affects our ability to draw conclusions about the role of psychological factors in the relationship between adiposity and chronic pain. Fifth, the relatively small size of the imaging sample compared with the original baseline sample may limit the generalisability of the findings to a broader population. However, our sensitivity analyses using the inverse probability weighting method showed that any potential biases from this limitation did not alter our results. Furthermore, the predominant white ethnicity of participants may also limit the generalisability of findings to diverse populations, as prior studies have shown that ethnic, cultural and socioeconomic factors can influence pain perception and reporting.42

      Conclusions

      Abdominal adipose tissue was associated with chronic musculoskeletal pain, suggesting that excessive and ectopic fat depositions may be involved in the pathogenesis of multisite and widespread chronic musculoskeletal pain. The identified stronger effects in women than men may reflect sex differences in fat distribution and hormones.

      Data availability statement

      Data from the UK Biobank is available to researchers by application via the UK Biobank online Access Management System.

      Ethics statements

      Patient consent for publication

      Ethics approval

      The North-West Multi-centre Research Ethics Committee provided approval for the study (reference 11/NW/0382). Participants in the UK Biobank study gave informed consent to participate in the study before taking part.

      Acknowledgments

      Acknowledgements of participants and financial support: The authors extend their sincerest thanks to the UK Biobank participants and staff for their contribution. The study was conducted utilising the UK Biobank resource, authorised under application number 68213. We acknowledge support of this study by the Australian National Health and Medical Research Council (grants 302204). The funding organisation played no part in designing, conducting, or managing the study, nor in collecting, analysing, or interpreting the data. They were also not involved in preparing, reviewing, or approving the manuscript, or in deciding to submit it for publication.

      References

      1. ↵
        1. World Health Organization
        . Noncommunicable diseases country profiles 2018. Licence: CC BY-NC-SA 3.0 IGO. Geneva, 2018.
      2. ↵
        1. Sarma S,
        2. Sockalingam S,
        3. Dash S
        . Obesity as a multisystem disease: trends in obesity rates and obesity-related complications. Diabetes Obes Metab 2021;23 (Suppl 1):3–16. doi:10.1111/dom.14290
        OpenUrl
      3. ↵
        1. Blyth FM,
        2. Noguchi N
        . Chronic musculoskeletal pain and its impact on older people. Best Pract Res Clin Rheumatol 2017;31:160–8. doi:10.1016/j.berh.2017.10.004
        OpenUrl
      4. ↵
        1. Cieza A,
        2. Causey K,
        3. Kamenov K, et al
        . Global estimates of the need for rehabilitation based on the Global Burden of Disease Study 2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet 2020;396:2006–17. doi:10.1016/S0140-6736(20)32340-0
        OpenUrl
      5. ↵
        1. Pan F,
        2. Laslett L,
        3. Blizzard L, et al
        . Associations between fat mass and multisite pain: a five‐year longitudinal study. Arthritis Care Res 2017;69:509–16. doi:10.1002/acr.22963
        OpenUrl
      6. ↵
        1. Tian J,
        2. Jones G,
        3. Lin X, et al
        . Association between chronic pain and risk of incident dementia: findings from a prospective cohort. BMC Med 2023;21:169. doi:10.1186/s12916-023-02875-x
      7. ↵
        1. Walsh TP,
        2. Arnold JB,
        3. Evans AM, et al
        . The association between body fat and musculoskeletal pain: a systematic review and meta-analysis. BMC Musculoskelet Disord 2018;19:233. doi:10.1186/s12891-018-2137-0
      8. ↵
        1. Barry VW,
        2. Baruth M,
        3. Beets MW, et al
        . Fitness vs. fatness on all-cause mortality: a meta-analysis. Prog Cardiovasc Dis 2014;56:382–90. doi:10.1016/j.pcad.2013.09.002
        OpenUrlCrossRefPubMed
      9. ↵
        1. Shuster A,
        2. Patlas M,
        3. Pinthus JH, et al
        . The clinical importance of visceral adiposity: a critical review of methods for visceral adipose tissue analysis. Br J Radiol 2012;85:1–10. doi:10.1259/bjr/38447238
        OpenUrlAbstract/FREE Full Text
      10. ↵
        1. Brown JC,
        2. Harhay MO,
        3. Harhay MN
        . Anthropometrically-predicted visceral adipose tissue and mortality among men and women in the third National Health and Nutrition Examination Survey (NHANES III). Am J Hum Biol 2017;29:e22898. doi:10.1002/ajhb.22898
      11. ↵
        1. Stefan N,
        2. Schick F,
        3. Häring H-U
        . Causes, characteristics, and consequences of metabolically unhealthy normal weight in humans. Cell Metab 2017;26:292–300. doi:10.1016/j.cmet.2017.07.008
        OpenUrlCrossRefPubMed
      12. ↵
        1. Bertoli S,
        2. Leone A,
        3. Vignati L, et al
        . Metabolic correlates of subcutaneous and visceral abdominal fat measured by ultrasonography: a comparison with waist circumference. Nutr J 2015;15:1–8. doi:10.1186/s12937-015-0120-2
        OpenUrl
      13. ↵
        1. Li S,
        2. Schwartz AV,
        3. LaValley MP, et al
        . Association of visceral adiposity with pain but not structural osteoarthritis. Arthritis Rheumatol 2020;72:1103–10. doi:10.1002/art.41222
        OpenUrl
      14. ↵
        1. Blake T,
        2. Gullick NJ,
        3. Hutchinson CE, et al
        . More than skin-deep: visceral fat is strongly associated with disease activity, function and metabolic indices in psoriatic disease. Arthritis Res Ther 2023;25:108. doi:10.1186/s13075-023-03085-9
        OpenUrl
      15. ↵
        1. Sudlow C,
        2. Gallacher J,
        3. Allen N, et al
        . UK Biobank: an open access resource for identifying the causes of a wide range of complex diseases of middle and old age. PLoS Med 2015;12:e1001779. doi:10.1371/journal.pmed.1001779
      16. ↵
        1. West J,
        2. Dahlqvist Leinhard O,
        3. Romu T, et al
        . Feasibility of MR-based body composition analysis in large scale population studies. PLoS One 2016;11:e0163332. doi:10.1371/journal.pone.0163332
      17. ↵
        1. Bland VL,
        2. Klimentidis YC,
        3. Bea JW, et al
        . Cross-sectional associations between adipose tissue depots and areal bone mineral density in the UK Biobank imaging study. Osteoporos Int 2022;33:391–402. doi:10.1007/s00198-021-06140-w
        OpenUrl
      18. ↵
        1. Langner T,
        2. Gustafsson FK,
        3. Avelin B, et al
        . Uncertainty-aware body composition analysis with deep regression ensembles on UK Biobank MRI. Comput Med Imaging Graph 2021;93:101994. doi:10.1016/j.compmedimag.2021.101994
        OpenUrl
      19. ↵
        1. Toussirot E,
        2. Michel F,
        3. Béreau M, et al
        . Serum adipokines, adipose tissue measurements and metabolic parameters in patients with advanced radiographic knee osteoarthritis. Clin Rheumatol 2017;36:2531–9. doi:10.1007/s10067-017-3789-0
        OpenUrl
      20. ↵
        1. Hussain SM,
        2. Urquhart DM,
        3. Wang Y, et al
        . Fat mass and fat distribution are associated with low back pain intensity and disability: results from a cohort study. Arthritis Res Ther 2017;19:1–10. doi:10.1186/s13075-017-1242-z
        OpenUrlCrossRef
      21. ↵
        1. Heuch I,
        2. Heuch I,
        3. Hagen K, et al
        . A comparison of anthropometric measures for assessing the association between body size and risk of chronic low back pain: the HUNT study. PLoS One 2015;10:e0141268. doi:10.1371/journal.pone.0141268
      22. ↵
        1. Machado LAC,
        2. Viana JU,
        3. da Silva SLA, et al
        . Correlates of a recent history of disabling low back pain in community-dwelling older persons: the Pain in the Elderly (PAINEL) study. Clin J Pain 2018;34:515–24. doi:10.1097/AJP.0000000000000564
        OpenUrl
      23. ↵
        1. Yoshimoto T,
        2. Ochiai H,
        3. Shirasawa T, et al
        . Sex differences in the association of metabolic syndrome with low back pain among middle-aged Japanese adults: a large-scale cross-sectional study. Biol Sex Differ 2019;10:33. doi:10.1186/s13293-019-0249-3
      24. ↵
        1. Bergman RN,
        2. Stefanovski D,
        3. Buchanan TA, et al
        . A better index of body adiposity. Obesity (Silver Spring) 2011;19:1083–9. doi:10.1038/oby.2011.38
        OpenUrlCrossRefPubMed
      25. ↵
        1. Tanamas SK,
        2. Wluka AE,
        3. Berry P, et al
        . Relationship between obesity and foot pain and its association with fat mass, fat distribution, and muscle mass. Arthritis Care Res 2012;64:262–8. doi:10.1002/acr.20663
        OpenUrlCrossRef
      26. ↵
        1. Harrison L,
        2. Wilson S,
        3. Munafò MR
        . Pain-related and psychological symptoms in adolescents with musculoskeletal and sleep problems. Clin J Pain 2016;32:246–53. doi:10.1097/AJP.0000000000000252
        OpenUrl
      27. ↵
        1. Yoo JJ,
        2. Cho NH,
        3. Lim SH, et al
        . Relationships between body mass index, fat mass, muscle mass, and musculoskeletal pain in community residents. Arthritis Rheumatol 2014;66:3511–20. doi:10.1002/art.38861
        OpenUrl
      28. ↵
        1. Brady SRE,
        2. Mamuaya BB,
        3. Cicuttini F, et al
        . Body composition is associated with multisite lower body musculoskeletal pain in a community-based study. J Pain 2015;16:700–6. doi:10.1016/j.jpain.2015.04.006
        OpenUrl
      29. ↵
        1. Ladeiras-Lopes R,
        2. Sampaio F,
        3. Bettencourt N, et al
        . The ratio between visceral and subcutaneous abdominal fat assessed by computed tomography is an independent predictor of mortality and cardiac events. Rev Esp Cardiol (Engl Ed) 2017;70:331–7. doi:10.1016/j.rec.2016.09.010
        OpenUrl
      30. ↵
        1. Storz C,
        2. Heber SD,
        3. Rospleszcz S, et al
        . The role of visceral and subcutaneous adipose tissue measurements and their ratio by magnetic resonance imaging in subjects with prediabetes, diabetes and healthy controls from a general population without cardiovascular disease. Br J Radiol 2018;91:20170808. doi:10.1259/bjr.20170808
      31. ↵
        1. Pou KM,
        2. Massaro JM,
        3. Hoffmann U, et al
        . Visceral and subcutaneous adipose tissue volumes are cross-sectionally related to markers of inflammation and oxidative stress: the Framingham Heart Study. Circulation 2007;116:1234–41. doi:10.1161/CIRCULATIONAHA.107.710509
        OpenUrlAbstract/FREE Full Text
      32. ↵
        1. Neogi T
        . The epidemiology and impact of pain in osteoarthritis. Osteoarthr Cartil 2013;21:1145–53. doi:10.1016/j.joca.2013.03.018
        OpenUrlCrossRefPubMedWeb of Science
      33. ↵
        1. Suganami T,
        2. Tanaka M,
        3. Ogawa Y
        . Adipose tissue inflammation and ectopic lipid accumulation. Endocr J 2012;59:849–57. doi:10.1507/endocrj.ej12-0271
        OpenUrlCrossRefPubMedWeb of Science
      34. ↵
        1. Ma C,
        2. Searle D,
        3. Tian J, et al
        . Dietary inflammatory index and magnetic resonance imaging-detected knee structural change and pain: a 10.7‐year follow‐up study. Arthritis Care Res 2024;76:813–20. doi:10.1002/acr.25307
        OpenUrl
      35. ↵
        1. Pattanittum P,
        2. Turner T,
        3. Green S, et al
        . Non-steroidal anti-inflammatory drugs (NSAIDs) for treating lateral elbow pain in adults. Cochrane Database Syst Rev 2013;2013:CD003686. doi:10.1002/14651858.CD003686.pub2
      36. ↵
        1. Pongratz G,
        2. Straub RH
        . The sympathetic nervous response in inflammation. Arthritis Res Ther 2014;16:1–12. doi:10.1186/s13075-014-0504-2
        OpenUrlCrossRef
      37. ↵
        1. Kammerlander AA,
        2. Lyass A,
        3. Mahoney TF, et al
        . Sex differences in the associations of visceral adipose tissue and cardiometabolic and cardiovascular disease risk: the Framingham Heart Study. J Am Heart Assoc 2021;10:e019968. doi:10.1161/JAHA.120.019968
      38. ↵
        1. Karlsson T,
        2. Rask-Andersen M,
        3. Pan G, et al
        . Contribution of genetics to visceral adiposity and its relation to cardiovascular and metabolic disease. Nat Med 2019;25:1390–5. doi:10.1038/s41591-019-0563-7
        OpenUrlPubMed
      39. ↵
        1. Gavin KM,
        2. Bessesen DH
        . Sex differences in adipose tissue function. Endocrinol Metab Clin North Am 2020;49:215–28. doi:10.1016/j.ecl.2020.02.008
        OpenUrl
      40. ↵
        1. McLaughlin T,
        2. Lamendola C,
        3. Coghlan N, et al
        . Subcutaneous adipose cell size and distribution: relationship to insulin resistance and body fat. Obesity (Silver Spring) 2014;22:673–80. doi:10.1002/oby.20209
        OpenUrlCrossRefPubMed
      41. ↵
        1. Straub RH
        . The complex role of estrogens in inflammation. Endocr Rev 2007;28:521–74. doi:10.1210/er.2007-0001
        OpenUrlCrossRefPubMedWeb of Science
      42. ↵
        1. Krupić F,
        2. Čustović S,
        3. Jašarević M, et al
        . Ethnic differences in the perception of pain: a systematic review of qualitative and quantitative research. Med Glas (Zenica) 2019;16:108–14. doi:10.17392/966-19
        OpenUrl

      Supplementary materials

      • Supplementary Data

        This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

      Footnotes

      • Contributors Study concept and design: FP. Statistical analysis: ZDK, JT and FP. Acquisition, analysis or interpretation: ZDK, JT, DA, PEM, FC, GJ and FP. Drafting of the manuscript: ZDK and FP. Critical revision of the manuscript for important intellectual content: ZDK, JT, DA, PEM, FC, GJ and FP. Study supervision: FP, DA and PEM. All the authors read and approved the final manuscript. Accepts full responsibility for the work and/or the conduct of the study, had access to the data, and controlled the decision to publish: FP.

      • Funding We acknowledge support of this study by the Australian National Health and Medical Research Council (grants 302204).

      • Competing interests None declared.

      • Provenance and peer review Not commissioned; externally peer reviewed.

      • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.