Secondary Logo

Journal Logo

Musculoskeletal Pain in Individuals With Inflammatory Bowel Disease Reflects Three Distinct Profiles

Falling, Carrie BPhty(Hons)*; Stebbings, Simon FRACP-FRCP; Baxter, George D. PhD*; Gearry, Richard B. FRACP; Mani, Ramakrishnan PhD*

The Clinical Journal of Pain: July 2019 - Volume 35 - Issue 7 - p 559–568
doi: 10.1097/AJP.0000000000000698
Original Articles
Free

Objectives: Pain affects over 70% of individuals with inflammatory bowel disease (IBD), with abdominal and musculoskeletal pain representing the most common symptoms. Musculoskeletal pain in IBD is reported to be associated with multiple clinical features, however the scope and nature of pain is not well understood. Primary aims were to identify subgroups of musculoskeletal pain in individuals with IBD based on clinical features of pain and assess how these subgroups differ in aspects of demographics, comorbidity, and IBD characteristics.

Methods: Cross-sectional online survey was performed on adults with IBD. Domains included: demographics, comorbidity, and clinical IBD and pain features. Latent class analysis was used to identify subgroups with similar attributes of: pain (severity, location, interference, and quality), IBD (activity, quality of life, and abdominal pain), and symptoms related to central sensitization. Correlation and regression analyses were used to profile identified subgroups.

Results: Of 305 included participants, 208 indicated the presence of musculoskeletal pain. Three identified subgroups were characterized as “mixed mechanism,” “central mechanism,” and “regional and remission.” Between subgroup differences included: total comorbidity score (P=0.005), osteoarthritis (P=0.027), osteoporosis (P=0.045), depression (P=0.001), anxiety (P=0.025), and chronic fatigue syndrome (P=0.020). Sex and age were identified as confounders for depression and anxiety.

Conclusions: Study results suggest multiple mechanisms contributing to pain experiences in IBD, to include central mechanisms. Features related to demographics, extraintestinal manifestations, IBD subtype, and clinical IBD features were not predictors of subgroup membership. However, total comorbidity demonstrated association with pain subgroups in this population.

*School of Physiotherapy

School of Medicine, University of Otago, Dunedin

Department of Medicine, University of Otago, Christchurch, New Zealand

The authors declare no conflict of interest.

Reprints: Carrie Falling, BPhty(Hons), 325 Great Kings Street, Dunedin 9010, New Zealand (e-mail: carrie.falling@otago.ac.nz).

Received August 12, 2018

Received in revised form February 1, 2019

Accepted February 20, 2019

Inflammatory bowel disease (IBD), comprising the Crohn disease (CD) and ulcerative colitis (UC), are chronic inflammatory conditions characterized by relapsing-remitting gastrointestinal tract inflammation resulting from dysregulated immune responses.1–4 In addition to symptoms associated with intestinal inflammation, pain is recognized as a prominent symptom of IBD affecting over 70% of patients, with abdominal and musculoskeletal (MSK) pain representing the most common symptoms in this population.5–9 Previous studies have tended to investigate MSK pain in relation to inflammatory arthropathies, which are common extraintestinal manifestations (EIMs) of IBD.10,11 Inflammatory arthritis affecting peripheral and axial joints are reported in 30% to 60% of IBD patients, and often present in a pattern typical of spondyloarthritides.4–6,10 However, a recent study reported only 17.5% of peripheral joint symptoms fulfilled standard spondyloarthritis criteria, indicating that the majority of joint pain in IBD is potentially noninflammatory in nature.6 Unfortunately, studies on joint pain in IBD commonly exclude noninflammatory joint pain, or mention it without further investigation, resulting in a lack of clarity between types of pain in IBD.9

Population-based surveys have described significant diversity in IBD-related pain, with reports of fluctuating and varying patterns of clinical presentation.5,6,8 MSK pain in IBD is reported to be associated with a multiple features, such as quality of life, disease activity, and concomitant comorbidities.6,12 However, to our knowledge the scope and contribution of these features to pain experiences, clinical pain presentations, and processes mediating MSK pain have not been described in this population.5,6,11,13 The use of mechanism-based assessments to characterize MSK pain has been promoted in chronic conditions, as a means of improving patient management by suggesting underlying pathologic processes.14–16 Such assessments aim to classify clinical presentations of pain based on mechanisms typically related to nociception, peripheral neuropathic pain, and central pain mechanisms.14–16

In an effort to contextualize multifactorial conditions subgrouping analyses, such as latent class analysis (LCA), has been used to explore known associations by describing distinct patient subgroups with similar attributes.17–20 The aim of subgrouping patients in this manner is to identify clear and distinct clinical profiles and to provide greater insight into the complexity of multifactorial conditions.20,21 This exploratory analysis approach works well in populations such as IBD, where there is not a strong a priori hypothesis with regard to the number or characteristics of pain profiles.22,23

Identifying patient profiles characterized by pain and IBD features through subgrouping, may provide empirical evidence for mechanisms participating in pain experiences, as well as facilitate targeted treatment pathways. Likewise, the opportunity to describe associations of additional patient features unique to each subgroup provides context to identified pain profiles.20,21,24,25 Therefore, the aims of this study are: to explore MSK-related pain in individuals with IBD through self-reported measures, in order to: identify subgroups based on clinical features of pain (location, severity, quality, and interference); and assess how these subgroups differ in aspects of demographics, comorbidity, and IBD characteristics (subtype, disease activity, and aspects of disease course).

Back to Top | Article Outline

METHODS

Research Design

The present cross-sectional survey was granted ethical approval by the University of Otago Human Ethics Committee (Health) (approval number: H17/095). Details of the protocol for this study have been published.26 Individuals with an IBD diagnosis aged 18 years and older were invited to participate in an online survey through Crohn’s and Colitis New Zealand Charitable Trust (CCNZ) email database and additional social media outlets associated with: IBD research groups, New Zealand health forums, patient support groups, and practitioner resource groups. CCNZ is a national organization that provides support, advice, and information to individuals with IBD in New Zealand. Participants were excluded if they reported any of the following: pregnancy, nerve injuries, neurological conditions (eg, stroke, multiple sclerosis, peripheral neuropathy, and Parkinson’s disease), and surgery within the last 3 months.

The present online survey included validated questionnaires identified in current literature used in the assessment of IBD, pain, and similar chronic inflammatory conditions. The survey included 3 sections: (1) demographics and comorbidity, (2) IBD status, and (3) pain status. Measures used in each of the survey sections are listed in Table 1 and described below.

TABLE 1

TABLE 1

Back to Top | Article Outline

Survey Section 1: Demographics and Comorbidity

Participant demographics, included: age, sex, and ethnicity. Comorbidities assessed in the present study included health conditions identified on the Self-Administered Comorbidity Questionnaire,27 EIM checklist,28 and conditions identified on part B of the Central Sensitization Inventory (CSI). Central sensitization (CS) in the present study refers to the International Association for the Study of Pain (2017) definition of CS as: “increased responsiveness of nociceptive neurons in the central nervous system to their normal or subthreshold afferent input.” Exploration of CS in the present study was through part A of the CSI. At present, there are no direct measures of CS in humans. Therefore, a variety of diagnostic surrogate markers are commonly used to identify the possibility of CS in patient populations, to include indirect screening measures such as CSI.29,30 CSI has been validated to identify symptomology related to syndromes where the underlying etiology is believed to be related to CS.30,31 CSI indicates the presence of CS symptomology as scores ≥40.30,31

Back to Top | Article Outline

Survey Section 2: IBD Status

IBD status was evaluated through dimensions of: IBD subtype, IBD activity, health-related quality of life (HRQOL), and clinical features (number of hospitalizations and surgeries, current medications, and EIM status). IBD activity was evaluated using the patient Harvey-Bradshaw Index (P-HBI) for CD, and patient Simple Clinical Colitis Activity Index (P-SCCAI) for UC/IC. Both P-SCCAI and P-HBI identify disease remission as scores of ≤4.32,33 HRQOL was assessed through the Short IBD Questionnaire (SIBDQ), with scoring interpreted as poor (10 to 29), moderate (30 to 49), and optimal (50 to 70).34,35 Medications were recorded under the following categories: immunosuppressants, biologics, gut specific anti-inflammatories, and steroids. The presence of EIMs was recorded for each participant using a 20-item checklist developed from current IBD literature, including the European Crohn’s and Colitis Organization (ECCO) Guidelines developed from expert consensus.10,36–38

Back to Top | Article Outline

Survey Section 3: Pain Status

MSK pain dimensions evaluated in the present study included: location, intensity, interference, and quality (nociceptive and neuropathic). A body diagram to record regional pain location and distribution (n=47) was used in the present study. This diagram was developed by recommendations from current literature with regard to MSK conditions in IBD5,6 as well as investigations for the reliability of pain drawing instruments.39 Generalized pain was distinguished from regional pain through the modified widespread pain criterion which requires having pain in 4 of 5 pain regions (4 quadrants plus axial pain) described by Wolfe et al.40 Pain severity was recorded relative to regions identified as the individual’s “main area of pain” using a Numerical Rating Scale (NRS), with positive findings as mild (1 to 4), moderate (5 to 6), or severe (7 to 10).41 Evaluation of pain interference and nociceptive pain in the present survey was evaluated through Patient-Reported Outcomes Measurement Information System (PROMIS) Pain Interference 4a and Nociceptive Pain Quality 5a short forms.42 Scoring of PROMIS short forms identify positive findings as: mild (50 to 59), moderate (60 to 69), or severe (≥70). The presence of neuropathic pain quality was evaluated through PainDETECT, with possible scores ranging from 0 to 38.43 Interpretation of PainDETECT scores identify neuropathic components as likely (≥19), unlikely (≤12), and unclear (13 to 18).43

Secondary evaluation of abdominal pain was also included in the present study. Abdominal pain was characterized through dimensions of severity and interference, measured by NRS for both items. Positive findings for abdominal pain intensity and interference are identified as mild (1 to 4), moderate (5 to 6), or severe (≥7).41

Back to Top | Article Outline

Sample Size Estimation

Current literature with regard to the subgrouping using LCA offers no straightforward guidelines about the minimum nor maximum sample size necessary for LCA.44 However, supporting evidence exists for minimum sample size of n=200 with respect to the number of present indicator variables and use of a medium to strong covariate.45,46 Therefore, minimum sample size estimate for the present study is 200 participants.

Back to Top | Article Outline

Data Reduction

Data collected during the present study was used to quantify latent class indicators, potential latent class covariates, and external variables directly from raw data, scores from individual measures, or criteria identified from composite data. Categorical indicator variables were utilized in the present study including: strongest pain severity, average pain severity, pain interference, generalized pain, neuropathic quality, nociceptive quality, CSI, IBD activity, and presence of abdominal pain. The use of categorical indicators reflect the clinical utility of variables such as IBD activity whereby assessment scores are typically interpreted as active versus not active, as opposed to continuous values. Variables considered for covariate analysis included: IBD subtype and SIBDQ. The variable(s) not utilized as a covariate was included in the list of external variables used for subgroup profiling. External variables, included: sex, age, number of EIMs, total comorbidity score, hospitalizations, surgeries, and current medications.

Back to Top | Article Outline

Statistical Analyses

Descriptive statistics (frequencies, means, and SD) were used to characterize demographic, comorbidity, IBD, and pain characteristics of study participants. LCA was performed in 2 stages using R statistical software. Initial LCA was used to identify models with 1 to 6 subgroups (or classes) across the sample using indicator variables quantified during data reduction. Model fit was assessed using model fit statistics (Akaike information criterion and Bayesian information criterion), goodness of fit G2 which follows a χ2 distribution, and model entropy.47 Lower values for information criteria and higher values for G2 and entropy between models suggests models with optimal balance between fit and parsimony.47 Interpretability was also considered along with fit statistics when selecting the final model. Further assessment of the final model included average posterior probabilities and classification error. Average posterior probabilities ≥0.7047 and classification error ≤0.10 were considered acceptable. Participants were assigned class membership based on their highest posterior probability. χ2 tests were used to identify a covariate with significant association (P≤0.05) with the final model. A subsequent LCA was then performed using the final model and the identified covariate to predict indicator class membership. Conditional item probabilities of indicator variables reaching ≥0.500 were used to characterize each latent class.47

Subgroup profiling of identified latent classes was performed for internal latent class variables (indicators and covariate) and external variables. χ2 test of association for dichotomous variables and Kruskal-Wallis 1-way analysis of variance for variables with skewed data were used to identify latent class and demographic variables that differed significantly between classes. Univariate logistic regression analysis was used to explore association of individual comorbidities (independent variable) with latent class membership (dependent variable). Where Peduzzi’s criteria for sample size allowed, variables demonstrating significance (P≤0.05) were adjusted for age, sex, and IBD subtype.48

Back to Top | Article Outline

RESULTS

Results of this study are presented in 2 sections. The first section describes participant characteristics (demographics, IBD, and pain). The second section presents subgrouping results of the LCA and comparative analysis of the subgroup profiles.

Back to Top | Article Outline

Participant Characteristics

Demographics

A total of 370 individuals with IBD volunteered to participate in the online survey. Eleven respondents were excluded due to minimum age requirements. An additional 54 respondents were excluded due to incomplete survey completion. Essential data sets for the present study were identified as completion of all components related to internal latent class variables identified above. The remaining 305 respondents were included as study participants. Demographics (age and sex), IBD subtype, and comorbidity status of study participants are presented in Table 2. Of the 305 participants, 201 (66%) individuals reported IBD subtype as CD, 94 (31%) reported UC, and 10 (3%) reported indeterminate colitis. Participants in the study represented the following ethnic groups: New Zealand European (n=274), Maori (n=18), Indian (n=4), English (n=5), Australian (n=5), North American (n=4), Fijian (n=3), Dutch (n=2), Scottish (n=1), German (n=1), South African (n=2), Croatian (n=1), and other European (n=1).

TABLE 2

TABLE 2

Back to Top | Article Outline

IBD Status

Results from questionnaires assessing aspects of IBD status in this study are presented in Table 3. Of the participants who reported CD (n=201), 4 indicated the presence of an ileostomy and therefore, were unable to utilize the P-HBI to indicate disease activity. Consequently, results of the P-HBI are reported for the remaining 197 participants with CD.

TABLE 3

TABLE 3

Back to Top | Article Outline

Pain Status

Results from questionnaires assessing aspects of abdominal and MSK pain in this study are presented in Table 4. Of the included participants, 162 (53%) individuals reported the presence of abdominal pain, 208 (68%) individuals reported the presence of MSK pain, and 126 (41%) individuals reported both abdominal and MSK pain. Of the MSK regions identified as painful by study participants, the low back was overall the most frequently reported region (n=124, 60%), while also identified most frequently as the “main area of pain” (n=41, 20%).

TABLE 4

TABLE 4

Back to Top | Article Outline

Subgrouping

Latent Class Analysis

Fit statistics for initial latent class models (1 to 6) are reported in Table 5. As indicated by the lowest Bayesian information criterion results, a 2-class model was most parsimonious, where Akaike information criterion supported a 4-class model. Consideration of fit statistics along with interpretability of the models suggested that a 3-class model was optimal. Classification error of the 3-class model was acceptable at 0.087. Average posterior probabilities (SD) of the 3-class model were 0.920 (0.13), 0.851 (0.14), and 0.892 (0.15), respectively.

TABLE 5

TABLE 5

Class 1 included 30.8% of study participants and was characterized as “mixed mechanism.” Class 1 represents a high probability for presenting with positive CSI scores, active IBD, abdominal pain, severe MSK pain, and moderate MSK pain interference. In addition, class 1 demonstrated increased probability of presenting with nociceptive and/or neuropathic pain qualities when compared with classes 2 and 3. Class 2 was characterized as “central mechanism” and represented the largest group (42.1%). Class 2 represented a high probability for presenting with positive CSI scores, active IBD, mild MSK pain interference, and no additional pain qualities (nociceptive or neuropathic). Class 2 also presented with moderate probability for presenting with abdominal pain and moderate MSK pain severity. The third, and smallest (26.9%), latent class was characterized as “regional and remission.” Class 3 represented low probability for demonstrating positive CSI scores, active IBD, abdominal pain, or additional pain qualities (nociceptive or neuropathic). In addition, class 3 demonstrated a high probability of presenting with mild to no MSK pain interference, and a moderate probability of presenting with regional MSK pain and moderate MSK pain severity. Conditional item responses of indicator variables for each class are reported in Table 6.

TABLE 6

TABLE 6

Back to Top | Article Outline

Subgroup Profiles

Descriptive statistics with between class differences for external variables (demographics, total comorbidity, EIMs, and clinical IBD features) and internal latent class variables (indicator and covariate) are shown in Tables 79. No significant differences between latent classes were found for sex, age, EIMs, IBD subtype, hospitalizations, surgeries, or medications. All internal latent class variables (indicator and covariate) demonstrated significance between class differences (P≤0.05) (Table 8), while total comorbidity score was the sole external variable demonstrating significance (P=0.005) (Table 7). Univariate logistic regression analysis for individual comorbidities, to include EIMs, identified statistically significant differences (P≤0.05) between latent classes for: osteoarthritis (P=0.027), osteoporosis (P=0.045), depression (P=0.001), anxiety (P=0.025), and chronic fatigue syndrome (P=0.020). Peduzzi’s criteria for sample size were solely met by depression and anxiety variables. As such, subsequent logistic regression analysis adjusting for age, sex, and IBD subtype indicated sex as a confounder for both depression and anxiety, and age as a confounder for anxiety Table 9.

TABLE 7

TABLE 7

TABLE 8

TABLE 8

TABLE 9

TABLE 9

Back to Top | Article Outline

DISCUSSION

Primary aims of this cross-sectional study were to explore MSK-related pain in individuals with IBD, in order to describe distinct patient subgroups and identify between subgroup differences based on profiles of external variables. Current study results describe 3 clinically relevant subgroups characterized as “mixed mechanism” for class 1, “central mechanism” for class 2, and “regional and remission” for class 3. Both classes 1 and 2 demonstrated high probabilities of presenting with positive CSI scores (≥40) and active IBD states. However, class 1 was the only class to demonstrate increased probability of presenting with neuropathic and/or nociceptive pain qualities. Class 1 also presented with the highest MSK pain severity and pain interference scores, as well as the highest probability for presenting with abdominal pain. Class 3 was the only class to demonstrate increased probability for IBD remission and regional MSK pain, while also demonstrating low probability for positive CSI scores and abdominal pain.

In the current study, positive CSI scores were a dominant feature in classes 1 and 2, thereby identifying symptomology related to the presence of CS in these classes. In addition, classes 1 and 2 also presented with increased MSK pain severity and interference profiles compared with the class with low probability of positive CSI scores (class 3). As described by Woolf,25,49 the increased responsiveness of neurons within the central nervous system as a result of CS leads to pain hypersensitivity. Previous studies of persistent pain in other populations, such as osteoarthritis, have shown an increase in pain severity in individuals demonstrating CS.29,50,51 CS is a broad concept that includes numerous and complex pathophysiological mechanisms, including changes to pain facilitation, inhibition, and sensory processing.25,49

Of the 2 classes presenting with CS symptomology, class 1 demonstrated a 91% probability of presenting with severe MSK pain, whereas class 2 demonstrated a 69% probability of presenting with moderate pain. The presence of CS has been described as different degrees over a continuum,50 as opposed to simply present or not. The question then becomes, to what degree is CS contributing to the clinical picture?50 Differences in pain profiles stated here may be the result of greater contribution from CS to the overall clinical picture of class 1, leading to worse pain experiences. Previous investigation of chronic spinal pain found that higher CSI scores, described as increased CSI severity, correlated with increased pain severity.52 Although the use of CSI as a measure of CS “severity” has not been validated, current results demonstrate similar findings whereby higher CSI scores corresponded with higher pain severity across subgroups.

Alternatively, increased MSK pain severity seen in class 1 may relate to a higher probability of presenting with neuropathic and/or nociceptive pain qualities compared with the other classes. Individuals in class 1 were more likely to present with abdominal pain and demonstrated the highest total comorbidity score. In addition, class 1 demonstrated a higher prevalence of osteoarthritis and osteoporosis suggesting these individuals may present with additional sources of nociception compared to the other classes. Incidentally, the presence of CS is thought to be an important contributor to increased painful comorbidities in chronic pain populations, as hypersensitivity due CS may result in pain from minimal nociceptive input of other structures (eg, arthritic joints).29 As such, the presence of multiple comorbidities and/or overlapping peripheral mechanisms potentially in the presence of CS, may have a cumulative effect leading to worse pain experiences in class 1.

In addition to positive CSI scores, individuals in classes 1 and 2 also demonstrated an increased probability for presenting with active IBD states. Results indicated a notable pattern between CSI and IBD activity measures across the study, whereby a high probability of demonstrating positive CSI also represented a high probability of demonstrating active IBD (class 1 and 2), with the opposite being true as well (class 3). IBD literature has previously described animal models of CS and IBD activity, specifically with regard to the development and maintenance of chronic abdominal pain.53,54 These models describe modulation of neural activity as a consequence of proinflammatory mediators present in active IBD states leading to sensitized nervous systems.53,54 Results from the present study suggests that contributions of CS to pain states in IBD may extend beyond models of abdominal pain to include MSK-related pain as well.

Subgroup profiling in the present study indicated no between class differences for observed demographics, EIMs, IBD subtype, and clinical IBD features. Of the variables assessing IBD in this study, disease activity and HRQOL were the only ones to demonstrate significant association. Collective evaluation of historic and present clinical features have been used to describe severity of an individual’s IBD course.35,55 Although a Delphi consensus described the presence of mucosal lesions, identified by colonoscopies, as the most important feature associated with overall disease severity,55 complicated IBD courses have been reported to include accumulation of: IBD-related hospitalizations and surgeries, multiple EIMs, use of disease modifying medications, and multiple steroid courses following IBD diagnosis.35,55 Results from the present study indicate that these features in isolation do not predict subgroup membership, suggesting that either these features are not related to MSK pain in IBD, or that the absence of clinical tests decreased the association of these combined features. Therefore, future research should explore IBD severity, inclusive of clinical tests, as an independent risk factor for MSK pain experiences in this population.

Total comorbidity scores were the sole external variable to demonstrate association with class membership, with increased prevalence in class 1. Although comorbidities have shown to be independent predictors of several outcomes, such as mortality and disability, relatively little is known about the effect of disease combinations on outcomes.56 Investigation of individual disease combinations on disability indicated that the effects of some combinations were additive, whereas other effects were synergistic, leading to increased disability.56 Results from the present study indicate that an increase in total comorbidity (eg, disease count) is predictive of membership to the subgroup demonstrating worse MSK pain profiles (class 1). However, it is unknown whether specific disease combinations would demonstrate differing effects on subgroup membership.

Analysis of individual comorbidities identified significant between class differences for osteoarthritis, osteoporosis, chronic fatigue syndrome, depression, and anxiety. Distribution of these comorbidities followed the total comorbidity distribution, with increased prevalence in class 1. Similar to previous irritable bowel syndrome and IBD investigations,57 the present study identified sex as a confounder for both depression and anxiety. Rates of mood disorders have shown to be higher in IBD compared with other diseases and the general population.58 Depression and anxiety, along with poor HRQOL, has shown association with increased number of relapses, suggesting mood disorders may be a risk factor in IBD.58,59 However, it is unclear whether the temporal presentation and fluctuation of mood disorders in relation to the relapsing-remitting nature of IBD is significant beyond chance.60

To our knowledge, this is the first study to characterize self-reported MSK pain through subgrouping analysis of multiple IBD and pain features. The present study included numerous standalone questionnaires, validated to assess respective pain and disease features. Face and content validity of survey questionnaires was provided by experts in the field of gastroenterology and chronic pain. Subgroups in the present study represent categorical interpretation of included questionnaires to reflect clinical utility of these measures. Authors acknowledge that LCA performed on continuous indicator variables may influence subgrouping results.

A common limitation of online surveys relates to assessment of IBD activity through self-reported measures. Standard clinical practice for estimating IBD activity typically includes clinical investigations, such as colonoscopies and serum biomarkers, alongside measures used in the present study.33,35,55 Therefore, estimation of IBD activity in the present study may not fully reflect findings from more invasive clinical assessments. Although the present study recorded all MSK pain regions reported by each participant, exploration of pain features were solely recorded for regions identified as the participant’s “main area of pain.” Therefore, results may not characterize features specific to all painful regions reported in the present study. The primary source for recruitment of study participants was through CCNZ. Study results may, therefore, overly represent individuals who have previously or are currently seeking support in managing their IBD.61 Further, the present study represents individuals who have digital and online access, and may not be generalizable to the broader IBD population.

Future research should consider not only the presence and implications of CS in IBD patients with and without persistent MSK pain, but exploration of multiple and potentially overlapping mechanisms (neuropathic/nociceptive) as well. Investigations exploring, for instance, somatosensory functioning, psychosocial features, and clinical IBD assessments would provide further insight to the nature of associations between pain mechanisms and IBD, in the broader scope of MSK pain experienced in this population. Future research of this nature would allow for the development of theoretical pain models beyond traditional inflammatory mechanisms, leading to targeted assessment and treatment pathways for MSK pain in IBD.

Back to Top | Article Outline

CONCLUSIONS

Exploration of MSK-related pain through subgrouping, in the current study, provides empirical evidence for the possibility of multiple mechanisms contributing pain experiences in IBD, including nociceptive, neuropathic, and central mechanisms. Study results demonstrate significant associations between CSI scores, IBD activity, HRQOL, and worse MSK pain, implicating the possibility of central mechanisms in the maintenance of pain states. In addition, results indicate variables related to demographics, EIMs, IBD subtype, and clinical IBD features are not predictors of subgroup membership. Conversely, increased number of comorbidities demonstrated association with MSK pain subgroups in this population.

Back to Top | Article Outline

ACKNOWLEDGMENTS

Professor Jo Nijs, PhD: Vrije Universiteit Brussel (Brussels, Belgium); Professor Andrew Day, FRACP: Head of Department of Paediatrics/Cure Kids Chair of Paediatric Research, University of Otago Christchurch (Christchurch, New Zealand); and Anneleen Malfliet, PhD: Postdoctoral Researcher, Vrije Universiteit Brussel (Brussels, Belgium) are acknowledged for their contributions in providing expert consensus during survey development. Biostatistical support was provided by Andrew Grey, BA BCom(Hons): Senior Research Fellow, Dunedin School of Medicine, University of Otago (Dunedin, New Zealand). CCNZ is acknowledged for support with participant recruitment. Author C.F. is a full-time University of Otago PhD candidate and acknowledges receipt of a University of Otago doctoral scholarship.

Back to Top | Article Outline

REFERENCES

1. Bliddal H, Danneskiold-Samsøe B. Chronic widespread pain in the spectrum of rheumatological diseases. Best Pract Res Clin Rheumatol. 2007;21:391–402.
2. Burisch J, Munkholm P. The epidemiology of inflammatory bowel disease. Scand J Gastroenterol. 2015;50:942–951.
3. M’Koma AE. Inflammatory bowel disease: an expanding global health problem. Clin Med Insights Gastroenterol. 2013;6: CGast-S12731.
4. Sheth T, Pitchumoni CS, Das KM. Musculoskeletal manifestations in inflammatory bowel disease: a revisit in search of immunopathophysiological mechanisms. J Clin Gastroenterol. 2014;48:308–317.
5. Zeitz J, Ak M, Müller-Mottet S, et al. Pain in IBD patients: very frequent and frequently insufficiently taken into account. PloS One. 2016;11:e0156666.
6. van Erp SJ, Brakenhoff LK, van Gaalen FA, et al. Classifying back pain and peripheral joint complaints in inflammatory bowel disease patients: a prospective longitudinal follow-up study. J Crohns Colitis. 2016;10:166–175.
7. Palm Ø, Bernklev T, Moum B, et al. Non-inflammatory joint pain in patients with inflammatory bowel disease is prevalent and has a significant impact on health related quality of life. J Rheumatol. 2005;32:1755–1759.
8. van der Have M, Brakenhoff LK, Erp SJ, et al. Back/joint pain, illness perceptions and coping are important predictors of quality of life and work productivity in patients with inflammatory bowel disease: a 12-month longitudinal study. J Crohns Colitis. 2015;9:276–283.
9. Brakenhoff LK, van der Heijde DM, Hommes DW. IBD and arthropathies: a practical approach to its diagnosis and management. Gut. 2011;60:1426–1435.
10. Levine JS, Burakoff R. Extraintestinal manifestations of inflammatory bowel disease. Gastroenterol Hepatol (NY). 2011;7:235–241.
11. Sheth T, Pitchumoni CS, Das KM. Management of musculoskeletal manifestations in inflammatory bowel disease. Gastroenterol Res Pract. 2015;2015:387891.
12. Vavricka SR, Schoepfer A, Scharl M, et al. Extraintestinal manifestations of inflammatory bowel disease. Inflamm Bowel Dis. 2015;21:1982–1992.
13. Bourikas LA, Papadakis KA. Musculoskeletal manifestations of inflammatory bowel disease. Inflamm Bowel Dis. 2009;15:1915–1924.
14. Woolf CJ, Bennett GJ, Doherty M, et al. Towards a mechanism-based classification of pain? Pain. 1998;77–227–229–.
15. Smart KM, Blake C, Staines A, et al. Mechanisms-based classifications of musculoskeletal pain: part 1 of 3: symptoms and signs of central sensitisation in patients with low back (±leg) pain. Man Ther. 2012;17:336–344.
16. Nijs J, Torres-Cueco R, van Wilgen P, et al. Applying modern pain neuroscience in clinical practice: criteria for the classification of central sensitization pain. Pain Physician. 2014;17:447–457.
17. Lee YC, Frits ML, Iannaccone CK, et al. Subgrouping of patients with rheumatoid arthritis based on pain, fatigue, inflammation, and psychosocial factors. Arthrit Rheumatol. 2014;66:2006–2014.
18. Loevinger BL, Shirtcliff EA, Muller D, et al. Delineating psychological and biomedical profiles in a heterogeneous fibromyalgia population using cluster analysis. Clin Rheumatol. 2012;31:677–685.
19. Rehm SE, Koroschetz J, Gockel U, et al. A cross-sectional survey of 3035 patients with fibromyalgia: subgroups of patients with typical comorbidities and sensory symptom profiles. Rheumatology (UK). 2010;49:1146–1152.
20. de Luca K, Parkinson L, Downie A, et al. Three subgroups of pain profiles identified in 227 women with arthritis: a latent class analysis. Clin Rheumatol. 2016;36:625–634.
21. Edwards RR, Dworkin RH, Turk DC, et al. Patient phenotyping in clinical trials of chronic pain treatments: IMMPACT recommendations. Pain. 2016;157:1851–1871.
22. Laudy O, Boom J, Hoijtink H. Bayesian computational methods for inequality constrained latent class analysis. In: Andries van der Ark L, Croon MA, Sijtsma K, eds. New developments in categorical data analysis for the social and behavioral sciences. Psychology Press. 2005:63–82.
23. Hoijtink H. Confirmatory latent class analysis: model selection using Bayes factors and (pseudo) likelihood ratio statistics. Multivariate Behav Res. 2001;36:563–588.
24. Jensen MP. Review of measures of neuropathic pain. Curr Pain Headache Rep. 2006;10:159–166.
25. Woolf CJ. Central sensitization: implications for the diagnosis and treatment of pain. Pain. 2011;152:S2–S15.
26. Falling C, Stebbings S, Baxter GD, et al. Profile of musculoskeletal pain in patients with inflammatory bowel disease: a study protocol for assessing the multidimensional experience of self-reported pain. Phys Ther Rev. 2018;23:1–9.
27. Sangha O, Stucki G, Liang MH, et al. The self‐administered comorbidity questionnaire: a new method to assess comorbidity for clinical and health services research. Arthritis Care Res (Hoboken). 2003;49:156–163.
28. Levine JB, Lukawski-Trubish D. Extraintestinal considerations in inflammatory bowel disease. Gastroenterol Clin N Am. 1995;24:633–646.
29. Arendt‐Nielsen L, Morlion B, Perrot S, et al. Assessment and manifestation of central sensitisation across different chronic pain conditions. Eur J Pain. 2018;22:216–241.
30. Mayer TG, Neblett R, Cohen H, et al. The development and psychometric validation of the Central Sensitization Inventory. Pain Pract. 2012;12:276–285.
31. Neblett R, Cohen H, Choi Y, et al. The Central Sensitization Inventory (CSI): establishing clinically significant values for identifying central sensitivity syndromes in an outpatient chronic pain sample. J Pain. 2013;14:438–445.
32. Evertsz FB, Nieuwkerk PT, Stokkers PC, et al. The patient simple clinical colitis activity index (P-SCCAI) can detect ulcerative colitis (UC) disease activity in remission: a comparison of the P-SCCAI with clinician-based SCCAI and biological markers. J Crohns Colitis. 2013;7:890–900.
33. Walsh AJ, Bryant RV, Travis SP. Current best practice for disease activity assessment in IBD. Nat Rev Gastroenterol Hepatol. 2016;13:567–579.
34. Irvine E, Zhou Q, Thompson A. The short inflammatory bowel disease questionnaire: a quality of life instrument for community physicians managing inflammatory bowel disease. Am J Gastroenterol. 1996;91:1571–1578.
35. Peyrin-Biroulet L, Panés J, Sandborn WJ, et al. Defining disease severity in inflammatory bowel diseases: current and future directions. Clin Gastroenterol Hepatol. 2016;14:348–354; e17.
36. Trikudanathan G, Venkatesh PG, Navaneethan U. Diagnosis and therapeutic management of extra-intestinal manifestations of inflammatory bowel disease. Drugs. 2012;72:2333–2349.
37. Huang V, Mishra R, Thanabalan R, et al. Patient awareness of extraintestinal manifestations of inflammatory bowel disease. J Crohns Colitis. 2013;7:e318–e324.
38. Harbord M, Annese V, Vavricka SR, et al. The first European evidence-based consensus on extra-intestinal manifestations in inflammatory bowel disease. J Crohns Colitis. 2016;10:239–254.
39. Margolis RB, Chibnall JT, Tait RC. Test-retest reliability of the pain drawing instrument. Pain. 1988;33:49–51.
40. Wolfe F, Clauw DJ, Fitzcharles M-A, et al. 2016 Revisions to the 2010/2011 fibromyalgia diagnostic criteria. Semin Arthrit Rheumatism. 2016;46:319–329.
41. Williamson A, Hoggart B. Pain: a review of three commonly used pain rating scales. J Clin Nurs. 2005;14:798–804.
42. Gershon RC, Rothrock N, Hanrahan R, et al. The use of PROMIS and assessment center to deliver patient-reported outcome measures in clinical research. J Appl Measure. 2010;11:304–314.
43. Freynhagen R, Baron R, Gockel U, et al. Pain DETECT: a new screening questionnaire to identify neuropathic components in patients with back pain. Curr Med Res Opin. 2006;22:1911–1920.
44. Wurpts IC, Geiser C. Is adding more indicators to a latent class analysis beneficial or detrimental? Results of a Monte-Carlo study. Front Psychol. 2014;5:920.
45. Finch WH, Bronk KC. Conducting confirmatory latent class analysis using M plus. Struct Equation Model. 2011;18:132–151.
46. Gudicha DW, Tekle FB, Vermunt JK. Power and sample size computation for Wald tests in latent class models. J Classification. 2016;33:30–51.
47. Collins LM, Lanza ST. Latent class and Latent Transition Analysis: With Applications in the Social, Behavioral, and Health Sciences. Hoboken, New Jersey: John Wiley & Sons, Inc.; 2013.
48. Peduzzi P, Concato J, Kemper E, et al. A simulation study of the number of events per variable in logistic regression analysis. J Clin Epidemiol. 1996;49:1373–1379.
49. Woolf CJ. What to call the amplification of nociceptive signals in the central nervous system that contribute to widespread pain? Pain. 2014;155:1911–1912.
50. Lluch E, Torres R, Nijs J, et al. Evidence for central sensitization in patients with osteoarthritis pain: a systematic literature review. Eur J Pain. 2014;18:1367–1375.
51. Akinci A, Al Shaker M, Chang M, et al. Predictive factors and clinical biomarkers for treatment in patients with chronic pain caused by osteoarthritis with a central sensitisation component. Int J Clin Pract. 2016;70:31–44.
52. Neblett R, Hartzell MM, Williams M, et al. Use of the Central Sensitization Inventory (CSI) as a treatment outcome measure for patients with chronic spinal pain disorder in a functional restoration program. Spine J. 2017;17:1819–1829.
53. Bielefeldt K, Davis B, Binion DG. Pain and inflammatory bowel disease. Inflamm Bowel Dis. 2009;15:778–788.
54. Farrell KE, Callister RJ, Keely S. Understanding and targeting centrally mediated visceral pain in inflammatory bowel disease. Front Pharmacol. 2014;5:27.
55. Siegel CA, Whitman CB, Spiegel BM, et al. Development of an index to define overall disease severity in IBD. Gut. 2016;67:244–254.
56. de Groot V, Beckerman H, Lankhorst GJ, et al. How to measure comorbidity: a critical review of available methods. J Clin Epidemiol. 2003;56:221–229.
57. Geng Q, Zhang Q-E, Wang F, et al. Comparison of comorbid depression between irritable bowel syndrome and inflammatory bowel disease: a meta-analysis of comparative studies. J Affective Disord. 2018;237:37–46.
58. Regueiro M, Greer JB, Szigethy E. Etiology and treatment of pain and psychosocial issues in patients with inflammatory bowel diseases. Gastroenterology. 2017;152:430–439; e4.
59. Mittermaier C, Dejaco C, Waldhoer T, et al. Impact of depressive mood on relapse in patients with inflammatory bowel disease: a prospective 18-month follow-up study. Psychosomatic Med. 2004;66:79–84.
60. Mikocka‐Walus AA, Turnbull DA, Moulding NT, et al. Controversies surrounding the comorbidity of depression and anxiety in inflammatory bowel disease patients: a literature review. Inflamm Bowel Dis. 2007;13:225–234.
61. Šimundić A-M. Lessons in biostatistics. Biochemia Medica. 2013;23:12–15.
Keywords:

musculoskeletal pain; central sensitization; inflammatory bowel disease

Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved.