Increased chronic pain in patients with ulcerative colitis is mostly associated to increased disease activity. A cross-sectional case-control study

Abstract

Backgrounds/aims

Musculoskeletal symptoms are common in patients with ulcerative colitis (UC), but no study has compared the prevalence of chronic pain to controls from a general population.

Methods

Patients with UC (n = 1164) and controls (n = 3867) were sent questionnaires comprising demography, history of pain, pain localization and UC patients’ Patient-Simple Clinical Colitis Activity Index. Chronic regional pain (ChRP) and chronic widespread pain (ChWP) were defined as having pain for at least 3 months.

Results

The response rate for the patients with UC was 49.0% and for the control persons 61.7% (p < .001). The reported prevalence of ChRP and ChWP was higher in patients with UC versus controls (33.1% vs. 24.2%; p < .001 and 19.8% vs. 12.5%; p < .001). The patients with UC reported significantly more pain in the regions ‘lower back’, ‘hip/upper leg’ and ‘lower leg/foot’ compared to controls. The patients with P-SCCAI $\ge$5 (n = 121) reported more ChWP than patients with P-SCCAI <5 (n = 426) (46.3% vs. 12.7%; p < .001) and controls (n = 2425) (46.3 vs. 12.5%; p < .001) in all body regions. No significant difference in ChWP was found between patients with P-SCCAI <5 and controls (12.7% vs. 12.5%; p = .917).

Conclusions

Patients with UC reported more chronic pain than controls from the general population, especially from the lower back and hip region. Higher UC disease activity was associated with more pain in all body regions.

Keywords:

• Musculoskeletal pain
• spondyloarthritis ulcerative colitis
• inflammatory bowel disease
• chronic pain

Introduction

Inflammatory bowel disease (IBD) is a chronic relapsing disorder which involves an autoimmune response against the gut/microbiota [1]. The incidence of IBD is rising globally and the estimated prevalence in the western countries is approximately 0.3% [2]. There is an overlap between IBD and other chronic inflammatory conditions [3] and patients with IBD commonly exhibit extra-intestinal symptoms [4]. Peripheral arthritis is the most common extra-intestinal symptom reported with a prevalence of 21–33% [5,6]. In the Inflammatory Bowel South-Eastern Norway (IBSEN) study, 17% of the patients with IBD during a follow-up period of 20 years were at some time classified as having peripheral arthritis and 28% as having peripheral spondyloarthritis (SpA) [7]. In another IBSEN follow-up study, 7.7% of the UC patients was diagnosed with axial SpA and 11.5% with inflammatory back pain within 20 years from the IBD diagnosis [8].

Previous studies have also indicated a high prevalence of musculoskeletal (MSK) pain in patients with IBD. One study examined MSK pain prevalence in different body regions in patients with IBD, and found pain prevalence ranging from 10 to 39% depending on the body region examined, with the highest pain prevalence in the back [9]. Another study found that as many as 49% of the participants suffered from some kind of joint pain [10]. However, reported chronic MSK pain is also common in the general population with an estimated prevalence ranging from 19 to 54% [11–13]. In a large population-based survey in the Netherlands, the prevalence of reported low back pain was 27%, shoulder pain 21% and neck pain 21% [14].

One study indirectly compared patients with IBD to controls by using primary care visits for MSK symptoms and found that patients with IBD had more diagnosis for MSK symptoms than non-IBD patients [15]. When taking into account the high prevalence of chronic pain in the general population, it is surprising that there are no studies that have compared reported symptoms of chronic pain in patients with IBD to a control population. Moreover, studies on chronic MSK pain, i.e., pain persisting for more than 3 months, in patients with IBD are currently lacking.

This study focuses on patients with UC, which is the most prevalent IBD diagnosis [16].

We aimed to compare the prevalence and characteristics of chronic pain in patients with UC to a large control group representing a general population which has previously been thoroughly studied [17]. A secondary aim was to study the impact of UC disease activity on reported chronic pain.

Materials and methods

The study was designed as a cross-sectional case-control study based on questionnaires.

Patient group

Patients with UC were identified using the Swedish National Quality Registry for Inflammatory Bowel Disease (SWIBREG). All living patients aged 20–74 years, who were residents of either Västernorrland or Västerbotten, two counties in northern Sweden, and who had been diagnosed with UC by a physician in Sweden, were included in the study. Patients who did not meet all of these criteria, e.g., who were deceased or had moved to another part of the country, were excluded from the study.

Control group

The controls consist of persons from the general population included in a previous study on MSK pain by Bergman et al. [17]. In order to investigate the prevalence of chronic MSK pain in the general population, Bergman et al. sent questionnaires to 3928 randomly selected persons aged 20–74 living in two cities in the southwest of Sweden. The questionnaire contained questions about the experience and location of chronic MSK pain including a drawing of the human body with predefined body regions marked out.The postal questionnaire was sent to 3928 individuals followed by two postal reminders, the second giving the opportunity to answer only a key question about the experience of chronic pain. The complete questionnaire was answered by 2425 individuals, resulting in a response rate of 61.7%. The 2425 responders constitute the control group in this study.

Questionnaire

The patients with UC who fulfilled the inclusion criteria described above (n = 1134) were sent a questionnaire by post in the time period from October 2018 until June 2019. Of these, 591 were residents of Västernorrland and 543 were residents of Västerbotten. Non-responders were sent a second questionnaire as a reminder 2–4 months after the first questionnaire was sent. In total, 556 patients responded to the questionnaire. Of these, 312 (56.1%) patients responded to the first questionnaire and 244 (43.9%) to the second (reminder) questionnaire. The 556 responders to the questionnaire constitute the patient group in this study. The response rate was 49.0%.

The questionnaire addressed the same questions used in the study by Bergman et al. [17] and included data on age, sex, height, weight, smoking and alcohol habits. Smoking habits were dichotomized into ‘never’ or ‘ever’. Responders who reported no previous history of smoking were categorized as ‘never’, and responders who reported a history of smoking or who were currently smoking were categorized as ‘ever’. Alcohol habits were categorized as never/rare, weekly or monthly. Responders who reported one occasion of alcohol consumption per month or less were categorized as ‘never/rare’, responders who reported alcohol consumption 2–4 times per month were classified as ‘monthly’, and responders who reported alcohol consumption more than 4 times per month were categorized as ‘weekly’. Chronic MSK pain was reported using the same drawing of the human body as in the study by Bergman et al. First, the participantswere asked if he/she had experienced chronic MSK pain, i.e., pain that had lasted for more than 3 months, during the last 12 months. If this was fulfilled, the patients were then instructed to provide information about the exact location/locations of the pain using the drawing as a guide.

UC activity was measured using the patient Simple Clinical Colitis Activity Index, P-SCCAI [18]. The participants received questions regarding the severity of their UC symptoms during the last 24 h, and their answers were scored according to P-SCCAI. The questions addressed the following symptoms: bowel frequency during the day, bowel frequency during the night, the presence of urgency of defecation, the presence of diarrhea or blood in the stools, the participants’ general well-being, and the presence of extra-intestinal manifestations including arthropathy, uveitis, and skin manifestations. Patients with a P-SCCAI score of 0–4 were considered to be in remission, and patients with a P-SCCAI score of 5–19 were considered to suffer froman active colitis [18].

Definition of chronic musculoskeletal pain

Chronic MSK pain was defined as presence of pain lasting for more than 3 months during the last 12 months. Chronic widespread pain (ChWP) was defined as presence of chronic pain on both the left and the right side of the body, above and below the waist, and in the axial part (i.e., the anterior chest, the cervical, thoracic or lumbar spine). The definition followed the American College of Rheumatology criteria for ChWP from 1990 [19]. Chronic regional pain (ChRP) was defined as chronic pain that did not meet the criteria for ChWP.

Statistical analysis

Data from the questionnaires were transferred to IBM SPSS Statistics version 26.0. Mean and standard deviations (SD) were used for age. Student t-test was used for comparison of mean age. Χ² test was used for comparison of proportions.

Ethical considerations

The study was approved by the regional ethic committee in Umeå (Dnr 2018-269-31 M) and has been carried out in compliance with the Helsinki Declaration. All responders gave written informed consent.

Results

Characteristics of patients with ulcerative colitis and controls from the general population

The response rate for the patients with UC was 49.0% which was significantly lower than for the control persons (61.7%) (p < .001). The basal characteristics for the patients and control persons are shown in Table 1. The proportion of younger persons (20–34 years) was significantly higher among the controls from the general population whereas the proportion of older persons (50–74 years) was significantly higher among the patients with UC.

Table 1. Characteristics of patients with ulcerative colitis and controls from the general population.

	Patients with ulcerative colitis (n = 556)	Controls from the general population (n = 2425)	p-value
Gender
Female	274 (49.3%)	1291 (53.2%)	.092
Male	282 (50.7%)	1134 (46.8%)	.283
Age, years
Mean	52.8 (SD 13.6)	46.5 (SD 15.4)	<.001
20–34	64 (11.5%)	653 (26.9%)	<.001
35–49	160 (28.8%)	715 (29.5%)	.742
50–64	198 (35.6%)	672 (27.7%)	<.001
65–74	134 (24.1%)	385 (15.9%)	<.001
Smoking habit
Never	301 (54.5%)	1234 (51.3%)	.167
Ever	251 (45.5%)	1173 (48.7%)	.169
Alcohol habit
Never/rare	203 (36.8%)	888 (36.9%)	.962
Monthly	249 (44.8%)	961 (39.9%)	.03
Weekly	98 (17.6%)	561 (23.3%)	.005
P-SCCAI score
High (≥5)	121 (21.8%)	N.A
Low (<5)	426 (76.6%)	N.A

N.A: Not available.

Statistics: Χ² test and student t-test.

Prevalenceof chronic pain in patients with ulcerative colitis compared to controls from the general population

Patients with UC reported more ChWP and ChRP than controls from the general population (Table 2). More than half of the patients with UC reported any chronic pain and one fifth ChWP. In both sexes, ChWP was more commonly reported in patients with UC compared to controls [for women (24.6% vs. 16.1%; p < .001) and for men (15.1% vs. 8.4%; p < .001)]. Also, ChRP was more commonly reported in the patients with UC than controls [for women (32.7% vs. 24.1%; p = .005) and for men (33.3% vs. 23.8%; p = .002)]. Any reported chronic pain was more commonly reported among patients with UC in all age groups in comparison to controls and ChWP was more common in all age groups except in persons 65–75 years of age (Table 3). ChRP was more common in patients with UC in the younger age group and in the older (65–75 years of age) age group in comparison to controls. In both patients with UC and in the controls from the general population ChWP was more commonly reported among women than in men (24.6% vs. 15.1%; p = .013 and 16.1% vs. 8.4%; p < .001) whereas there was no gender difference for reported ChRP (32.7% vs. 33.3%; p = .462 and 24.1% vs. 23.8%; p = .86).

Table 2. Prevalence of ChWP, ChRP and no chronic pain in patients with ulcerative colitis and in controls from the general population.

Type of musculoskeletal pain	Patients with ulcerative colitis (n = 556)	Controls from the general population (n = 2425)	p-value
ChWP	110 (19.8%)	303 (12.5%)	<.001
ChRP	184 (33.1%)	588 (24.2%)	<.001
NCP	258 (46.4%)	1466 (60.5%)	<.001
Unknown/missing	4 (0.7%)	68 (2.8%)	.004

ChWP: chronic widespread pain; ChRP: chronic regional pain; NCP: no chronic pain.

Statistics: Χ² test.

Table 3. Prevalence of ChWP, ChRP, and no chronic pain in patients with ulcerative colitis compared to controls from the general population stratified into four different age groups.

Age group	Type of pain	Patients with ulcerative colitis (n = 556)	Controls from the general population (n = 2425)	p-value
20–34 years	ChWP	7 (10.9%)	31 (4.7%)	.035
	ChRP	20 (31.3%)	127 (19.4%)	.026
	NCP	37 (57.8%)	495 (76.0%)	.002
35–49 years	ChWP	27 (16.9%)	77 (11.0%)	<.001
	ChRP	48 (30.0%)	179 (25.0%)	.196
	NCP	83 (51.9%)	459 (64.2%)	.004
50–64 years	ChWP	52 (26.3%)	130 (19.3%)	.036
	ChRP	64 (32.3%)	179 (26.6%)	.118
	NCP	81 (40.9%)	363 (54.0%)	.002
65–74 years	ChWP	24 (17.9%)	65 (16.9%)	.786
	ChRP	52 (38.8%)	103 (26.8%)	.009
	NCP	57 (42.5%)	217 (56.4%)	.006

ChWP: chronic widespread pain; ChRP: chronic regional pain; NCP: no chronic pain.

In the ulcerative colitis patient group, four patients could not be classified regarding type of pain and have therefore been excluded in this table.

Statistics: Χ² test.

Pain based on body regions in patients with ulcerative colitis

The prevalence of chronic MSK pain based on body regions is presented in Table 4. Patients with UC significantly more often reported pain from lower back, the hip/thighand the lower foot/leg body regions. There was no statistically significant difference between the groups for the rest of the body regions.

Table 4. Prevalence of regional chronic musculoskeletal pain in different body regions in patients with ulcerative colitis compared to controls from the general population.

Body region	Patients with ulcerative colitis (n = 556)	Controls from the general population (n = 2425)	p-value
Anterior chest	34 (6.1%)	115 (4.7%)	.214
Neck	119 (21.4%)	460 (19.0%)	.266
Dorsal chest	63 (11.3%)	236 (9.7%)	.318
Lower back	168 (30.2%)	557 (23.0%)	<.001
Shoulder/upper arm	126 (22.7%)	482 (20.0%)	.205
Elbow/lower arm/hand	103 (18.5%)	405 (16.7%)	.394
Hip/upper leg	113 (20.3%)	319 (13.1%)	<.001
Knee	95 (17.1%)	335 (13.8%)	.070
Lower leg/foot	97 (17.4%)	300 (12.4%)	.003

Statistics: Χ² test.

Pain related to disease activity in patients with ulcerative colitis

In 547 of the UC patients there were complete data on P-SCCAI to estimate disease activity. Of those patients 22.1% were classified as having an active colitis (P-SCCAI $\ge$5) and 77.9% as being in remission (P-SCCAI <5). The UC patients with active colitis reported significantly more ChWP than patients in remission and controls from the general population (Table 5). The patients with UC in remission reported significantly slightly more ChRP than control persons but ChRP was reported equally in patients with active colitis and in patients in remission. Patients with active colitis reported significantly more chronic pain for all body regions compared to patients in remission and controls from the general population (Table 6). Patients in remission did not differ from control persons in reported chronic pain from different body regions except for significantly lower frequency of reported neck and arm pain than controls.

Table 5. Prevalence of ChWP, ChRP, and no chronic pain in ulcerative colitis patients with an active colitis^a, patients in remission^b, and in controls from the general population.

Type of musculoskeletal pain	Group 1. UC patients. SCCAI score $\ge 5$ (n = 121)	Group 2. UC patients. SCCAI score < 5 (n = 426)	Group 3. Controls from the general population (n = 2425)	p-value. Group 1 vs. Group 2	p-value. Group 1 vs. Group 3	p-value. Group 2 vs. Group 3
ChWP	56 (46.3%)	54 (12.7%)	303 (12.5%)	<.001	<.001	.917
ChRP	38 (31.4%)	140 (32.9%)	588 (24.2%)	.763	.075	<.001
NCP	27 (22.3%)	228 (53.5%)	1466 (60.5%)	<.001	<.001	.008

ChWP: chronic widespread pain; ChRP: chronic regional pain; NCP: no chronic pain.

^aPatient Simple Clinical Colitis Activity Index score $\ge 5.$

^bPatient Simple Clinical Colitis Activity Index score <5.

Statistics: Χ² test.

Table 6. Prevalence of chronic musculoskeletal pain in different body regions in ulcerative colitis patients with an active colitis^a, patients in remission^b, and in controls from the general population.

Body region	Group 1. UC patients. SCCAI score $\ge 5$ (n = 121)	Group 2. UC patients. SCCAI score < 5 (n = 426)	Group 3. Controls from the general population (n = 2425)	p-value. Group 1 vs. Group 2	p-value. Group 1 vs. Group 3	p-value. Group 2 vs. Group 3
Anterior chest	21 (17.4%)	13 (3.1%)	115 (4.7%)	<.001	<.001	.121
Neck	56 (46.3%)	62 (14.6%)	460 (19.0%)	<.001	<.001	.030
Dorsal chest	31 (25.6%)	32 (7.5%)	236 (9.7%)	<.001	<.001	.148
Low back	63 (52.1%)	102 (23.9%)	557 (23.0%)	<.001	<.001	.660
Shoulder/upper arm	54 (44.6%)	69 (16.2%)	482 (20.0%)	<.001	<.001	.077
Elbow/lower arm/hand	47 (38.8%)	54 (12.7%)	405 (16.7%)	<.001	<.001	.038
Hip/upper leg	45 (37.2%)	68 (16.0%)	319 (13.1%)	<.001	<.001	.119
Knee	37 (30.6%)	58 (13.6%)	335 (13.8%)	<.001	<.001	.913
Lower leg/foot	42 (34.7%)	54 (12.7%)	300 (12.4%)	<.001	<.001	.861

^aPatient Simple Clinical Colitis Activity Index score $\ge 5.$

^bPatient Simple Clinical Colitis Activity Index score <5.

Statistics: Χ² test.

Non-responders

The patients with UC who fulfilled the questionnaire (responders) were significantly older than the non-responders [mean age 52.8 years (SD 14.3) vs. 48.7 years (SD 14.6); p < .001].

A sensitivity analyses where all non-responders (both patients and controls) were included, and classified as having no pain, showed that the patients with UC still had significantly more ChWP (9.7% vs. 7.8%; p = .045) and any reported chronic pain (25.9% vs. 23.0%; p = .045) than control persons but there was no difference in ChRP (16.2% vs. 15.2%; p = .404).

Discussion

Chronic MSK pain, defined as pain persisting for more than 3 months, is associated with decreased quality of life, including decreased physical, psychological and social well-being [20,21]. On a community level, chronic pain leads to loss of productivity in the workplace [22], and has been estimated to cost millions of dollars each year [23].

Several studies have reported a high prevalence of extra-intestinal symptoms including MSK pain in patients with UC [4,6] but to our knowledge, this is the first study that compare reported chronic pain in patients with a diagnosis of UC and persons from a general population. This present study shows that the prevalence of chronic pain, both ChWP and ChRP, was significantly higher among patients with UC compared to controls. More than half of the patients with UC reported any chronic pain (ChWP or ChRP) which is slightly higher than in a previous study [11]. The patients with UC were slightly older than the controls. In order to control for this, we stratified the patients into different age groups. In all age groups, the presence of any chronic pain (ChWP and ChRP combined) was significantly more frequent in patients with UC. However, the increased frequency of reported chronic MSK pain was more explicit in the younger patient groups and less explicit in the older patient groups. One possible explanation to this finding could be that MSK pain of other causes other than those associated with IBD increases with age, which makes the difference between IBD patients and the general population less pronounced as they get older.

The increased frequency of reported ChWP in patients with UC was strongly associated to clinical disease activityin UC (P-SCCAI $\ge$ 5) whereas patients in clinical remission (P-SCCAI <5) reported ChWP similar to that in the general population. The frequency of reported chronic neck and arm pain was even lower in patients in remission than in controls. The pattern of distribution of reported chronic MSK pain in UC patients with an active colitis indicates the occurrence of axial SpA related to IBD activity. Data from the IBSEN study showed that significantly more patients with axial SpA reported chronicity in their IBD course [7].

The mechanism of chronic MSK pain in patients with IBD is not known. In a proportion of patients, it is believed to be caused by inflammatory mechanisms associated with tissue damage and related to SpA, but in other patients, the chronic pain seems to be a consequence of physiological factors or central sensitization [24]. Systemic inflammation has been proposed to alter pain perception [25,26] and inflammatory activity is also associated with worse pain experience in patients with IBD [27].

The body sites that patients with UC reported pain in was more often located in the lower half of the body, from the lower back to the foot region, in comparison to pain reported by control persons. This is consistent with the known association between SpA and UC [4]. In the present study, the overall prevalence of lower back pain in patients with UC was 30% compared to 23% in the control group. In addition to lower back pain, one study reported that pain in the ankle and the knee was common in patients with IBD which is in line with the findings in our study [28]. Moreover, the SpAs can affect the axial skeleton as well as the peripheral joints, causing MSK pain in virtually all parts of the body [5,29,30].

In both patients with UC and controls from a general population, chronic pain was more common in women according to our study. Chronic pain syndromes are known to be more common in women [31] and MSK pain has previously been reported more frequently in women than in men in the general population [14]. Also, in patients with IBD, female patients more often than men reported symptoms of peripheral arthritis and peripheral SpA [7]. Sex hormones, differences in the immune system and differences in brain response to pain has been proposed as possible gender differences in the experience of pain [32,33].

There are some limitations in the present study. First, the response rate in the patient group was lower compared to that of the control group. However, when including all non-responders (both patients with UC and controls from the general population) and classified them as having no pain, the patients with UC still had significantly more ChWP and any reported chronic pain than the controls indicating that the observed difference in pain prevalence holds true. However, the difference in response rate might have overestimated the prevalence of pain in patients with UC since those who experience pain possibly are more willing to respond. Second, it could be argued that a more objective way (i.e., fecal calprotectin) to evaluate disease activity than P-SCCAI would have been better to discriminate symptoms due to gut inflammation from symptoms due to a functional bowel disorders (i.e., irritable bowel syndrome). However, we presumed that asking the patients to perform a fecal test would likely have decreased the response rate even more, and therefore we used the validated P-SCCAI score to assess disease activity. Third, including a physical examination would have given more specific information if the reported pain were related to specific structures (i.e., joints, ligaments or muscles). Furthermore, the study lacks information from both controls and from patients with UC on other symptoms (i.e., anxiety, depressive and functional gastrointestinal disorders) that might influence the frequency of MSK pain. Finally, we cannot rule out that minor differences in the reported symptoms could be attributed to regional differences and that the controls were investigated 23 years earlier than the patients with UC.

To conclude, patients with UC commonly have chronic pain and more frequently report chronic pain than controls from a general population. However, the increased prevalence of chronic pain is mainly seen in UC patients with an active colitis, while patientsin remission report similar prevalence of chronic pain as controls. In comparison to controls, patients with UC report more pain from the lower back and lower extremities. In clinical practice, the results of this study suggest that when a patient with UC reports chronic pain, inflammatory activity must be evaluated and treated. The high prevalence of chronic low back pain in UC patients reported in this study could indicate a high prevalence of axial SpA in this patient group. However, further research is needed to conclusively ascertain the observed prevalence of axial and peripheral SpA in patients with UC.

Author contributions

N.P. and F.K. participated in designing the study, collecting the data, performing statistical analyses, interpreting the results. A.W. contributed to collecting the data and critically reviewed the final manuscript. S.B. participated in the analysis. P.K. and H.F.d.E. participated in designing the study, collecting the data, performing statistical analyses, interpreting the results and in writing the manuscript. All authors have read and approved the final manuscript.

Acknowledgements

The authors thank the health care regions of Västernorrland and Västerbotten, Sweden.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by Umeå University.