Is adiposity associated with back and lower limb pain? A systematic review

Background Back and lower limb pain have a major impact on physical function and quality of life. While obesity is a modifiable risk factor for musculoskeletal pain, the role of adiposity is less clear. This systematic review aimed to examine the relationship between both adiposity and its distribution and back and lower limb pain. Methods A systematic search of electronic databases was conducted to identify studies that examined the association between anthropometric and/or direct measures of adiposity and site specific musculoskeletal pain. Risk of bias was assessed and a best evidence synthesis was performed. Results A total of 56 studies were identified which examined 4 pain regions, including the lower back (36 studies), hip (two studies), knee (13 studies) and foot (eight studies). 31(55%) studies were assessed as having low to moderate risk of bias. 17(30%) studies were cohort in design. The best evidence synthesis provided evidence of a relationship between central adiposity and low back and knee pain, but not hip or foot pain. There was also evidence of a longitudinal relationship between adiposity and the presence of back, knee and foot pain, as well as incident and increasing foot pain. Conclusions This systematic review provides evidence of an association between both body fat and its central distribution and low back and knee pain, and a longitudinal relationship between adiposity and back, knee and foot pain. These results highlight the potential for targeting adiposity in the development of novel treatments at these sites.

Introduction Musculoskeletal conditions are a leading disease burden worldwide. They are not only the second most common cause of global disability, but disability-adjusted life years (DALYs) for musculoskeletal conditions have increased alarmingly, with a rise of up to 62% between 1990 and 2016 [1]. One in three people worldwide live with a musculoskeletal condition, which is characterised by pain and disability, leads to reduced quality of life, and results in a huge economic burden [2]. Back and lower limb pain are highly prevalent musculoskeletal conditions and make a major contribution to their increasing burden at an individual, familial and societal level. Current efforts to reduce the profound impact of these conditions have focussed on determining modifiable risk factors for management and prevention.
Obesity is an escalating, global epidemic. The 2016 Global Burden of Disease Study showed that the prevalence of obesity is not only increasing, but obese people are actually living longer, which allows for the development of co-existing conditions, such as musculoskeletal pain [3]. There is growing evidence to indicate that obesity is a modifiable risk factor for musculoskeletal pain at different sites. A meta-analysis by Shiri and colleagues reported overweight and obesity, measured by weight and body mass index (BMI), to be risk factors for low back pain [4], while a systematic review by Butterworth et al. found increased BMI to be strongly associated with non-specific foot pain in the general population [5]. While these reviews provide evidence for a relationship between obesity, measured by body weight or BMI, and musculoskeletal pain, they do not account for body composition and thus don't consider the individual contributions of fat mass and lean tissue mass (or muscle mass). This is of particular importance given there is evidence to show that fat mass or adiposity and muscle mass have different roles in the pathogenesis of musculoskeletal disease [6,7].
There is growing evidence to show that adiposity plays an important role in musculoskeletal pain. Adipose tissue acts as an endocrine organ, releasing a host of pro-inflammatory cytokines and adipokines [8], which can heighten inflammatory changes leading to destruction of tissue [9] and increasing pain and disability. A single systematic review has examined the relationship between body fat and musculoskeletal pain [10], reporting a positive cross-sectional association between higher body fat and single-site joint pain in the low back, knee and foot. However, no conclusions could be drawn from longitudinal data regarding the role of adiposity in back and lower limb pain, as there was a lack of available high quality, cohort studies. Moreover, the review focussed on studies that used direct measures of body fat, such as fat mass and percentage of body fat, and excluded those that examined anthropometric measures, such as waist circumference and waist hip ratio [11], thus limiting the opportunity to examine role of fat distribution, particularly central adiposity.
Understanding the role of adiposity in musculoskeletal pain, particularly back and lower limb pain, has huge potential to inform the development of novel prevention and treatment approaches, as well as further our understanding of mechanisms underlying the relationship between obesity and musculoskeletal pain. The aims of this systematic review were to: (i) examine the relationship between central adiposity and back and lower limb pain and (ii) investigate the longitudinal association between adiposity and both the presence, incidence and progression of pain at these sites.

Methods
This systematic review was conducted according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement (see S1 Checklist) [12].

Data sources and searches
We performed electronic searches of six databases, including MEDLINE, Embase, CINAHL, Cochane Central Register of Controlled Trials, Scopus and Web of Science from database inception to February 2, 2021. Our initial search for studies was conducted using text words and subject terms on three key databases and then based on this, we developed the search strategy, with subject classification systems investigated for each database and expanded our data sources to include all six databases for our final search. The final searches of all six databases, covering the key concepts of adiposity and musculoskeletal pain, were performed using the appropriate specifications for each database. The comprehensive search strategy for OVID Medline is provided (see S1 Medline Database search strategy in S1 Text). The searches were restricted to adult human studies but not limited based on language. To identify grey literature, we searched Google scholar, using key terms such as 'adiposity' and 'musculoskeletal pain', from 2011 to 14 February 2021, and Scopus, using our Scopus search strategy and selecting for conference proceedings, from inception to 14 February 2021. In addition, reference lists of reviews and key papers were searched to identify relevant literature.

Inclusion and exclusion criteria
Studies were included if they investigated the relationship between adiposity and low back or lower limb pain, using at least one measure of adiposity and reporting pain as an outcome measure. Studies that examined adiposity using: (i) anthropometric measures, including waist circumference, hip circumference, waist-hip ratio, waist-height ratio, and skin folds, and (ii) direct fat measures, such as fat mass and body fat percentage, using dual-energy X-ray absorptiometry (DXA) and bioelectrical impedance, were included.
For the purposes of this review, waist circumference was defined as a measurement around the trunk at the midpoint between the lower margin of the least palpable rib and the top of the iliac crest [11]. Hip circumference was considered to be a measure around the hips at the maximum posterior extension of the buttocks, while waist-hip ratio and waist-height ratio were calculated by dividing waist circumference by hip circumference measures, and waist circumference by height, respectively [11]. Skinfold measures assessed the subcutaneous fat thickness and were measured by skinfold calipers [13], while fat mass and body fat percentage were defined as the total mass of adipose tissue or percentage of total adipose tissue of the whole body mass respectively. Central adiposity, an accumulation of both subcutaneous and visceral fat in the lower torso around the abdominal area, was assessed by waist circumference or waist-hip or waist-height ratio measures, which are recommended by the World Health Organization [11].
Data on the presence, incidence and progression of pain in each region was recorded from the included studies where possible. The presence of pain, which was reported from cross-sectional, case-control and cohort studies, was defined as pain recorded at one point in time. For a cohort study, this could have been where adiposity was assessed at baseline and pain was measured at follow-up. Incident pain was defined as where pain was assessed at both baseline and follow-up in a cohort study, with pain absent at baseline and present at follow-up. Moreover, the progression of pain was described where pain was present at both baseline and follow-up in a cohort study and was assessed as increasing, decreasing or not changing over the study period There was no hierarchy given to these pain outcomes, however data from cohort studies were considered the highest level of evidence, followed by case-control studies and then cross-sectional studies.
We excluded studies that: (1) reported BMI or weight only; (2) examined only intramuscular fat; (3) reported pain in the head, neck or upper limb; (4) investigated pain other than musculoskeletal pain, i.e. abdominal pain, cardiac pain; and (5) examined multisite musculoskeletal pain where data specific to the back or lower limb were not reported separately.

Study selection
Titles and abstracts were assessed by two investigators (WP and TR) for relevance and the full texts were retrieved for relevant studies.

Data extraction
Data were extracted and tabulated by two reviewers (WP and TR) independently. Studies were categorized based on: (i) the site of pain investigated (low back, hip, knee and foot), (ii) their study design (cross-sectional, case-control or cohort) and (iii) the type of adipose measure reported (anthropometric versus direct fat measures). Data extracted from the studies included (1) author and year of publication, (2) study population characteristics (number of study participants, gender (% women), mean (SD) age, recruitment source), (3) assessment method and measure for adiposity and pain, (4) results (OR/RR, 95%CI) and (5) conclusions.

Risk of bias assessment
To assess the risk of bias of the included studies, two reviewers (TR and JF) independently assessed the included studies using the Cochrane risk of bias assessment [14]. The Cochrane risk of bias assessment examines the internal and external validity of the included studies, based on four items for cross-sectional studies and five items for cohort studies, with each item scored as low, moderate or high risk of bias. An overall assessment was then given for each study; low if every individual item scored low, moderate if all items scored low except either one high or two moderate, or high if individual items scored more than one high or more than two moderate.

Best evidence synthesis
A best evidence synthesis was used to summarise the data. It was not possible to perform a meta-analysis as there was substantial clinical and methodological heterogeneity across the studies, including differences in the clinical populations investigated, risk factors and outcomes measured, and statistical data and analyses performed. Based on the study design, the number of studies, the risk of bias rating, and consistency of the results of the studies, levels of evidence for the association between adiposity and pain was determined for each region. The studies were ranked according to their design, with cohort studies considered the highest level of evidence, followed by case-control studies and then cross-sectional studies. Studies were classified as having an association ("positive" or "negative") if the association reported was statistically significant according to the authors' predetermined alpha value (or p < .05 where this was not reported) or where the confidence interval for an odds ratio did not cross one.
The levels of evidence used were adapted from the Lievense's standardized criteria [15], which have been used previously in observational studies of musculoskeletal conditions [16]. They included: evidence of an association, conflicting evidence, limited evidence or no evidence. 'Evidence of an association' was defined as consistent findings in multiple, cohort studies, while 'conflicting evidence' was defined as inconsistent findings across the number and types of studies. 'Limited evidence' was defined as consistent findings in a small number of studies, including a single cohort study or one or two case-control or cross-sectional studies, and 'no evidence' was used when there are no studies that provided any evidence.

Identification of studies
After removal of the duplicates, 6,242 records remained (Fig 1). A total of 6,049 studies were excluded based on the screening of titles and abstracts, leaving 193 studies for full text analysis. A further 74 studies were excluded as they did not meet the review's inclusion criteria: 37 studies only included BMI as their measure of obesity , 18 studies did not examine any

Assessment of pain
A range of measures were used to assess pain (Table 1). While the Western Ontario and McMaster Universities Arthritis Index (WOMAC) and visual analogue scale (VAS) were the most commonly used validated tools, a large number of studies used structured interviews or self-administered questionnaires. Low back pain was examined using the visual analogue scale [116,121,122,132,137,138], Chronic Pain Grade scale [6,7,92,97], NHANES general wellbeing index [112], Nordic Musculoskeletal Questionnaire [133], questions regarding the history of low back pain (e.g. Have you ever had back pain? ("yes" or "no")) [88,93,95,96,100,104,105,107,108,111,113,115,117,123,130,131,[134][135][136] and structured interviews [103,114,127]. Hip pain was assessed by asking about the presence of pain (yes/no) [94] and any history of hip pain [102]. Knee pain was assessed using WOMAC Index  [99] and asking about the presence of pain (yes/no) [94,98], or the history of foot pain (Over the past month, have you had pain, aching, or stiffness in either of your feet on most days?) [90]. The follow-up periods between baseline and the assessment of pain varied between the cohort studies. Of the 8 cohort studies of back pain [87,88,92,93,[95][96][97]100], the follow-up time ranged from 2 to 20 years, with half of the studies investigating time periods less than 10 years and half of the studies examining time periods over 10 years. The single cohort study of hip pain followed up participants over 5 years [94], while the 5 cohort studies of knee pain ranged from 2 to 10.7 years [86,89,91,94,101], with 3 studies examining time periods of 5 or 6 years. Moreover, the 5 studies examining foot pain had follow-up periods ranging from 4 to 20 years [85,90,94,98,99], with 4 of the 5 studies focusing on a 3-5 year follow-up.

Relationship between adiposity and low back pain
Anthropometric fat measures. Waist circumference. Twenty one studies examined the association between waist circumference and low back pain ( Table 2). Of these studies, 13 were cross sectional studies [112,113,116,117,123,127,130,[132][133][134][135][136]138], two were case control studies [104,105] and six were cohort studies [87,88,92,93,96,100]. Eight of the 13 cross-sectional studies found significant associations between waist circumference and low back pain [112,113,117,127,132,[134][135][136], with two studies reporting an association in females only [113,127], two studies finding a relationship in males only [117,136] and the remaining 4 studies finding an association in both males and females [112,132,134,135]. Five studies did not find an association between waist circumference and radiating and non-specific low back pain [123], presence of low back pain [130,133] or low back pain intensity [116,138]. Of the two case-control studies, one study found greater waist circumference was associated with less low back pain (lasting 14 days or greater) in middle age women [104], while the other study found no association between waist circumference and chronic low back pain [105].
Of the six cohort studies, five studies found significant associations [87,88,92,96,100]. Three studies found a significant relationship between waist circumference and the presence of low Pain score of �1 over the past 7 days.
No adjustments made.

Pain intensity
Fat mass rho: -0.239 p = 0.2 % body fat rho: 0.09 p = 0.7 There was no correlation between fat mass or percentage body fat and back pain.

Pain intensity
Abdominal to lumbar fat mass ratio r = 0.32, p = 0.007 Abdominal adiposity was associated with chronic LBP.
Chou (2016) [7] Cohort split into two groups: back pain [87,88,100], with two studies reporting obese waist circumference to be associated with a larger number of days of low back pain [114,117] and one study finding waist circumference to be associated with high intensity low back pain [92]. The other two studies examined the relationship between waist circumference and incident low back pain and reported conflicting results [93], with one of the studies also examining recurrent and persistent low back pain and finding an association with waist circumference [59].
Hip circumference. Six studies examined the association between hip circumference and low back pain. Of the six studies, three cross-sectional studies found significant associations between hip circumference and low back pain intensity [116,132], but one study reported it in females only [127]. However, one cross-sectional study reported no significant association between hip circumference and low back pain intensity, a case-control study found no significant association between hip circumference and low back pain defined as pain for one or more days or 14 or more days [104] and one cohort study found no association between hip circumference and incident or recurrent/persistent low back pain [88].
Waist-hip ratio and waist-to-height ratio. Twelve studies, including eight cross sectional, two case-control and two cohort studies, examined the association between waist-hip ratio and low back pain. Of the eight cross-sectional studies [7,86,113,118,127,132,133,138], {], five found an association between waist-hip ratio and low back pain [7,113,118,127,132], while three did not find a relationship [86,133,138]. One case control study found waist-hip ratio was significantly associated with low back pain for 14 or more days [104], while another case control study found no association [105]. Both cohort studies found no association between waist-hip ratio and incident low back pain, with one study investigating this relationship in twins [93], and the other study examining females and males separately [88].
With respect to waist-to-height ratio, two studies examined the relationship between this adiposity measure and low back pain, finding an association with the presence of pain in postmenopausal women [133], and with radiating low back pain, but not chronic low back pain [123].
Direct fat measures. Body fat mass. Twelve studies, including eight cross sectional, one case-control and three cohort studies, examined the association between body fat mass and low back pain (Table 2). Of the eight cross sectional studies, one study found an association between abdominal to lumbar fat mass ratio and low back pain [122], and three studies found an association between total body fat mass and pain intensity [6,7] and the presence of low back pain [134]. The remaining four studies found no association between total body fat mass and chronic low back pain [114,121,131,137]. The case control study found an association between fat mass and presence of pain in males, but not females [107]. While one cohort study found associations between fat mass and high intensity pain in females and males, and fat mass and low intensity pain in females only [92], the remaining two cohort studies found greater fat mass was associated with a higher risk of the presence of pain [100] and high pain intensity [97].
Body fat percentage. Twelve studies, including five cross sectional, four case-control and three cohort studies, examined the association between body fat percentage and low back pain. Three cross sectional studies found associations between body fat percentage and low back pain [115,116,118], while two found no association [134,137]. Three case control studies found significant associations between body fat percentage and chronic low back pain [105,108] and presence of pain [107], while the remaining case control study found no association between body fat percentage and recurrent low back pain [103]. One cohort study found percentage fat mass to be significantly associated with high intensity pain in both females and males [92]. The second cohort study found those in the highest quartile of body fat mass were significantly more likely to develop low back pain than those in the lowest quartile [95] and the third study found no associations between percentage fat mass and incident low back pain in twins [93].

Summary of the evidence.
Overall there was evidence of an association between adiposity and low back pain from 26 of the 36 identified studies (Table 3). Specifically, there was evidence from 5 of 6 cohort studies and 12 of 15 cross-sectional studies to indicate that there is a positive relationship between central adiposity and low back pain. There was also evidence provided by six of six cohort studies for a longitudinal relationship between adiposity and presence of low back pain, but conflicting evidence for a relationship between adiposity and incident low back pain (two of three studies) and limited evidence for a relationship with increasing low back pain (one of one study) ( Table 4).

Relationship between adiposity and hip pain
Anthropometric fat measures. One case control study found no significant difference in waist circumference, hip circumference and waist-hip ratio between individuals with greater trochanteric pain and controls [102] (Table 5).
Direct fat measures. Body fat mass. A single cohort study found greater body fat mass was associated with the presence of hip pain among older individuals [94].
Overall there was limited evidence for an association between adiposity and hip pain based on two studies with conflicting results (Table 3). There was limited evidence to suggest central adiposity is not a risk factor for hip pain (one case-control study) and limited or no evidence that there is a longitudinal relationship between adiposity and the presence (one cohort study), incidence (no studies) or progression of hip pain (no studies) ( Table 4). These studies reported anthropometric measures of adiposity, such as waist circumference, waist-hip ratio, and hip circumference, to measure central adiposity.^A ll case control studies were cross-sectional in design.
# These studies used direct measures of adiposity, such as fat mass and percentage total body fat, to measure total body adiposity. � A study by Yip and colleagues found an inverse relationship between central adiposity and low back pain (included here).
�� A study by Ojoawo et al reported a relationship for hip circumference, but not waist circumference or waist-hip ratio, and low back pain (included here). https://doi.org/10.1371/journal.pone.0256720.t003

PLOS ONE
Is adiposity associated with back and lower limb pain? A systematic review

Relationship between adiposity and knee pain
Anthropometric fat measures. Waist circumference. Seven studies, including three crosssectional, one case-control and three cohort studies, examined the association between waist circumference and knee pain (Table 5). While one cross-sectional study found a significant difference in knee pain between those with a waist circumference <94cm and those with waist circumference �101.9cm [124] and a second cross-sectional study found an association between waist circumference and knee pain intensity [132], a third cross sectional study found no differences in waist circumference between those with and without knee pain [125]. Moreover, a case-control study found that a greater waist circumference was associated with increased pain intensity [110]. Of the three cohort studies, one reported a significant association between waist circumference and consistent and fluctuating knee pain [89], and the other two found a significant relationship between waist circumference and knee pain intensity [91,101].
Waist-hip ratio. One cross-sectional study and one case-control study found significant associations between waist-hip ratio and pain intensity [132] and the presence of knee pain on most days [109], while a cohort study reported an association between waist-hip ratio and fluctuating knee pain, but not consistent knee pain [89].
Direct fat measures. Body fat mass. Four studies examined the relationship between body fat mass and knee pain. Two cross sectional studies found no significant association between fat mass and knee pain [120,125], while one case-control study found no association between fat mass and presence of knee pain on most days [109] (Table 5). The single cohort study found greater body fat mass was associated with the presence of knee pain among older individuals [94].

PLOS ONE
Is adiposity associated with back and lower limb pain? A systematic review  There were no significant differences in body fat percentage or fat mass between knee pain and non-knee pain subjects.

Scott (2012) [129]
Responded yes to the question "Do you have pain at any of these sites today?" (for knee pain). Total fat mass measured by skin fold was not found to be significantly associated with the presence of knee pain.

Anthropometric fat measurement
Pan (2020) [101] Minimal pain' group: consistently low level of pain Mild pain' group: a mild level of pain consistent throughout the follow-up. Moderate pain: relatively high level of pain consistent throughout the follow-up.
Age, sex, physical activity, smoking history, unemployment, education level and radiographic knee osteoarthritis.  Body fat percentage. Three cross sectional and two cohort studies examined the association between body fat percentage and knee pain. Two cross sectional studies found an association between body fat percentage and knee pain [128], however one found this association only in females [129], while the remaining cross-sectional study found no association between body fat percentage and knee pain [120]. Of the two cohort studies, one study found an association between body fat percentage and consistent and fluctuating knee pain [89], while the other found no association between body fat percentage and incident patellofemoral pain [86].
Summary of the evidence. Overall there was evidence from nine of the 13 identified studies for an association between adiposity and knee pain (Table 3). There was evidence to indicate that central adiposity is a risk factor for knee pain (six of 8 studies) and there is a longitudinal relationship between adiposity and the presence of knee pain (three of three cohort studies) (Table 4). However, there was limited evidence for a relationship between adiposity and incident and increasing knee pain with a limited number of cohort studies identified in each case.

Relationship between adiposity and foot pain
Anthropometric fat measures. Waist circumference. One cohort study found that individuals with a larger waist circumference were at greater risk of incident and increasing foot pain [98] (Table 6).
Waist-hip ratio. One cross-sectional study, which examined the association between waisthip ratio and foot pain, found no significant association [126].
Direct fat measures. Body fat mass. Seven studies, including two cross-sectional, one case control and five cohort studies, examined the association between fat mass measures and foot pain ( Table 6). Both cross -sectional studies found significant associations between fat mass and foot pain [119,126], while the case-control study did not find any significant difference between total fat mass in individuals with foot pain compared to those without [106]. Five cohort studies found significant positive relationships between direct fat measures and foot pain, with two studies reporting an association with the presence of foot pain [90,94], three studies finding a relationship with incident foot pain [85,90,99] and two studies reporting an association with progression of foot pain [98,99]. While Laslett and colleagues reported a relationship between fat measures and increasing foot pain, no association was found for incident foot pain [98].
Summary of the evidence. Overall there was evidence from seven of the eight identified studies for an association between adiposity measures and foot pain (Table 3). While there was limited evidence for a relationship between central adiposity and foot pain (two conflicting studies), there was evidence for a longitudinal relationship between adiposity and the presence of pain (two of two studies), incident foot pain (three of four studies) and progression of foot pain from 6 months to 5 years (two of two studies) ( Table 4).

PLOS ONE
Is adiposity associated with back and lower limb pain? A systematic review

Discussion
This systematic review found that both body fat and its central distribution are associated with musculoskeletal pain. There was evidence of a relationship between central adiposity and low back and knee pain, but limited or conflicting evidence for hip and foot pain. There was also evidence of a longitudinal relationship between adiposity and the presence of low back, knee and foot pain, as well as both incident and increasing foot pain. Taken together, these findings further our understanding of the mechanisms underlying obesity-related musculoskeletal pain and highlight adiposity as a potential therapeutic target in the management of back and lower leg pain. This systematic review is the first to examine the relationship between fat distribution and musculoskeletal pain. We found evidence that central adiposity, defined as the accumulation of extra subcutaneous and visceral fat concentrated just above or around the waistline, was associated with pain in the lower back and knee. This was based on evidence of a significant, positive association in 16 of the 22 studies of low back pain, including five of six cohort studies, and 6 of 8 studies of knee pain, including 3 of 3 cohort studies. This finding is consistent with evidence that central adiposity is associated with a greater risk of major public health conditions, such as cardiovascular disease and diabetes, which are associated with huge socioeconomic burdens globally [11]. Given visceral fat associated with central adiposity is an important correlate of metabolic disturbances [139], and the cells in central, visceral fat have a much higher turnover than subcutaneous fat cells in other regions of the body [140], central adiposity may be an important way to target obesity. For instance, management strategies targeted to enhance weight loss around the abdominal region may be particularly beneficial. Overall, the evidence for an association between central adiposity and low back and knee pain indicates that it is not just extra body fat that contributes to poor health and chronic pain, but also the distribution of the fat.
This review found evidence of a longitudinal relationship between adiposity and the presence of low back, knee and foot pain, as well as incident and increasing foot pain. The findings suggest that increased adiposity can lead to back and lower limb pain in the future, and in the case of foot pain, the development or increasing intensity of pain. Our results, which take into account 17 cohort studies, build on the conclusions of a previous review of seven cohort studies [10], which suggested that such associations may exist, but was limited by a lack of high quality studies. Our results highlight the need for high quality clinical trials to examine the efficacy of approaches that target weight loss, be it through physical activity, diet and/or medical options, in the management of back and lower limb pain in overweight and obese individuals. They also suggest that investigating the efficacy of the targeted interventions, such as exercise programs that focus on reducing adipose tissue and nutrition plans that optimize health but minimize fat intake. Moreover, given current evidence collectively indicates that musculoskeletal pain has an important systemic inflammatory component, there is an exciting opportunity to examine the efficacy of pharmaceutical and complementary medicines as potential treatment targets to reduce inflammation in individuals with musculoskeletal pain with a specific overweight/obese profile.
Moreover, evidence from this review informs our knowledge of the mechanisms that underlie obesity-related musculoskeletal pain. Several mechanisms, including increased physical loading and systemic metabolic processes, have been proposed to explain the role of obesity in musculoskeletal pain. In overweight or obese individuals, excess adipose tissue may result in increased load on a region and subsequently, altered posture and abnormal movement patterns resulting in pain and disability. There is also growing evidence to support systemic metabolic processes, with evidence that adiposity is associated with pain in non-weight-bearing regions, such as the hand [94]. Adipose tissue is metabolically active, releasing a multitude of pro-inflammatory cytokines and adipokines, which may potentiate inflammatory changes in a region resulting in pain [141]. Moreover, inflammation has been shown to alter the excitation thresholds and responses to stimuli of peripheral nerves, subsequently leading to peripheral and central sensitisation [142,143]. Our findings provide evidence that both mechanical and metabolic mechanisms may be at play in lower back, knee and foot pain, with the potential for total body fat and central obesity to load these regions and increased visceral and subcutaneous fat to alter metabolic processes. However, preliminary evidence from studies that reported adiposity to be associated with pain in non-weight bearing regions, such as the neck and hand, suggest that future research examining these regions may further our understanding of the pathogenesis of obesity-related musculoskeletal pain.
This systematic review has several important strengths, including conducting a comprehensive, systematic search of the literature based on six electronic databases, performing a risk of bias assessment of studies using the Cochrane risk of bias assessment, and conducting a best evidence synthesis to summarise the strength of the available evidence. Moreover, this review is novel, as it is the first to provide evidence of the role of central adiposity in site-specific musculoskeletal pain, as well as an updated summary of the evidence examining the longitudinal association between adiposity and back and lower limb pain. While this review was not registered a priori with an international prospective register, we have provided a detailed description of our review methodology from development of our search strategy to the assigning of levels of evidence and documented any changes in our initial methodology in this publication. Furthermore, while the review was limited by the paucity of high quality cohort studies, as well as significant heterogeneity in the identified studies, which meant a meta-analysis could not be performed, we used established levels of evidence to summarise the data for each musculoskeletal pain region This systematic review found that both body fat and its distribution are associated with sitespecific musculoskeletal pain. There was evidence of a positive relationship between central adiposity and low back and knee pain and a longitudinal association between adiposity and the presence of back, knee and foot pain, as well as incident and worsening foot pain. These findings are not only important in understanding the mechanisms which underlie chronic, musculoskeletal pain, but in the development of innovative treatment approaches for these debilitating conditions.