Anal incontinence and vesico-sphincter events in systemic sclerosis: An epidemiologic bicentric cohort study

Seminars in Arthritis and Rheumatism ] (2016) ]]]–]]]

Contents lists available at ScienceDirect

Seminars in Arthritis and Rheumatism journal homepage: www.elsevier.com/locate/semarthrit

Anal incontinence and vesico-sphincter events in systemic sclerosis: An epidemiologic bicentric cohort study Mickaël Martina,n, Nadine Meaux-Ruaulta, Nadine Magy-Bertranda, Guillaume Beraudb, Bernard Parrattec, Pascal Roblotb a

Department of Internal Medicine, Besancon University Hospital, BESANCON Cédex, France Department of Internal Medicine, Infectious and Tropical Diseases, Poitiers University Hospital, Poitiers, France c Department of Physiotherapy and Rehabilitation, Besancon University Hospital, BESANCON Cédex, France b

a r t i c l e in fo

Keywords: Systemic sclerosis Fecal incontinence, Urinary bladder Quality of life

a b s t r a c t Objective: To estimate the frequency and severity of anal incontinence and vesico-sphincter events, associated factors, and impact on the quality of life of patients with systemic sclerosis. Methods: Questionnaires assessing anal incontinence (Miller score), vesico-sphincter events (Urogenital Distress Inventory) and quality of life [Short Form Health Survey 36v2 (SF-36), and Hospital Anxiety and Depression Scale] were mailed to 139 patients with systemic sclerosis at the university hospitals of Besançon and Poitiers, France. Clinical data were collected from the medical records to identify risk factors. Results: Among the 121 (87%) responders, severe vesico-sphincter events or severe anal incontinence occurred in 3.4% and 12.4% of cases, respectively. Frequent urination (66.3%) and anal incontinence to gas (50.4%) were the most frequent symptoms. Anal incontinence was associated positively with vesicosphincter events, unrelated to obstetrical factors. No correlations were seen with age, sex, or systemic sclerosis characteristics. In multivariate analysis, moderate or severe vesico-sphincter events was associated with higher anxiety and depression scores and lower SF-36 scores; the same results were observed for anal incontinence, but did not reach significance. Conclusion: Vesico-sphincter events and anal incontinence are common in systemic sclerosis, and sometimes severe, with a potential negative impact in quality of life. These results will be confirmed by a casecontrol study with dynamic and manometric assessment, and could legitimate a systematic screening to ensure early therapy and multidisciplinary individual management. & 2016 Elsevier Inc. All rights reserved.

Introduction Systemic sclerosis (SSc) is a systemic connective tissue disease and a source of morbidity and mortality. It preferentially affects women than men, between 45 and 64 years old [1]. Among the complications, anal incontinence (AI) has been reported in 22–38% of cases, with varying severity and symptomatology [2,3]. The prevalence of AI is often underestimated due to patients’ reluctance to report symptoms or to seek care [4]. Vesico-sphincter events (VSE) in SSc have been rarely and poorly described in the literature. They can manifest as irritative

Abbreviations: AI, anal incontinence; HADS, hospital anxiety and depression scale; QoL, quality of life; SF-36, short form health survey 36v2; SSc, systemic sclerosis; UDI-6, urogenital distress inventory; UI, urinary incontinence; VAS, visual analogic scales; VSE, vesico-sphincter events. n Corresponding author at: Mickaël Martin, Department of Internal Medicine, Besancon University Hospital, BESANCON Cédex, France. E-mail address: [email protected] (M. Martin). http://dx.doi.org/10.1016/j.semarthrit.2016.03.013 0049-0172/& 2016 Elsevier Inc. All rights reserved.

symptoms [5–9], obstructions [5,6,8–10], or more rarely, as suprapubic pain [5,6], or isolated hematuria [5]. In the general population, risk factors for urinary incontinence (UI) are obesity and vaginal delivery, whereas diabetes mellitus type II and chronic diarrhea are risk factor for AI. Age above 80 years old and depression reassociated with dual incontinence (UI and AI) [11]. The impact of VSE and AI on quality of life (QoL) has rarely been studied in SSc. Only a few studies have reported that AI negatively impacts patient QoL; however, the symptoms related to AI were not reported [2,12]. One cohort study explored the impact of VSE on QoL in SSc and found that overactive bladder was associated positively with overall disability and anxiety [8]. The main objective of the current study was to estimate the frequency and the severity of VSE and AI in SSc. The secondary objectives were to assess associated symptoms, risk factors, and to estimate the impact of VSE and AI on patient’s QoL.

2

M. Martin et al. / Seminars in Arthritis and Rheumatism ] (2016) ]]]–]]]

Methods Clinical records We report here a descriptive and analytic transversal and retrospective bicentric cohort study. Patients with a SSc diagnosis according to the International Statistical Classification of Diseases and related Health Problems (ICD) from the registries of the university hospital of Besançon or Poitiers, (France), and followed in these centers between January 2010 and December 2012 were first selected. After this screening the patients only of 18 years or older, with a definite SSc diagnosis according to the criteria of LeRoy et al. [13] or LeRoy and Medsger [14] were retained. Exclusion criteria were as follows: non– French-speaking patients; localized scleroderma; and overlap syndromes (systemic lupus erythematosus (SLE) and/or rheumatoid arthritis). In fact, interstitial cystitis is reported in SLE and rheumatoid arthritis [15,16], and SLE may cause enteritis [17] and rectal ischemia [18]. Other exclusion criteria were chronic inflammatory bowel disease during treatment and/or symptomatic Crohn’s disease or ulcerative colitis, that can also affect the digestive tract. Five self-questionnaires were sent to each patient: (1) the Urogenital Distress Inventory (UDI-6) [19], testing by 6 items, each heighted from 0 to 3, (final ranging 0–18) the intensity of stress and emergency UI, as well as dysuria and pelvic pain. This scale was completed by questioning about the presence or absence of four or more urinary tract infections/year and/or macroscopic hematuria not associated with the menstrual cycle or cystitis; (2) the Miller score [20], testing AI by giving variable weights to the same frequencies according to the different type of AI: to gas (1–3), to liquid stools (4–6), or to solid stools (7–9). For each type, a total absence of AI was scored 0 (final ranging 0–18). This scale was completed by questioning about the presence or absence of clinical gastroesophageal reflux, clinical gastroparesis, frequent constipation, daily bloating, chronic diarrhea ( Z3 months almost daily), anal seepage, urge AI, inability to distinguish gas or stool before emptying, and regular use of protective underwear; (3) the hospital anxiety and depression scale (HADS) [21] allowing distinct anxiety and depression scores, ranging from 0 (no sign) to 21 (major anxiety or depression); (4) the short form health survey 36v2 standard Canada (French) version 2.0 (SF-36) [22,23]. Each of the eight dimensions of this scale were individually scored, as well as the physical summary score (PSs) and the mental summary score (MSs), each varying from 0 (maximal QoL alteration) to 100 (no impact in QoL); and (5) visual analogue scales (VAS) exploring the impact of AI, VSE, and SSc in general, in the QoL, varying from 0 (no impact) to 10 (maximal negative impact). In order to identify potential QoL confounders, patients were also invited to indicate “yes” or “no” at the following question “Did you have any personal factors unrelated to SSc during the past 12 months that could have altered your quality of life?”. The patients returned the questionnaires using an enclosed pre-paid envelope. Non-respondents were contacted by phone 1 month after the questionnaires were sent. Patients with AI and/or or VSE not previously known have been then addressed for a specific medical consultation. On the basis of medical records, SSc patients with anticentromere antibodies were classified as limited form and those with anti-Scl70 antibodies, regardless of their skin fibrosis extension, were classified as diffuse form because of the better association of anti-Scl70 with disease severity than skin sclerosis extension [24] and the high risk of progression to diffuse cutaneous involvement and pulmonary fibrosis [25]. SSc patients without SSc-related specific auto-antibody were classified according to their skin extension and criteria of LeRoy et al. [13] or LeRoy and Medsger [14]. Patient’s general data (age and sex), and SSc characteristics were collected retrospectively from medical records: SSc duration

(time since the first non-Raynaud’s symptom), SSc-specific autoantibodies (anti-centromere, anti-Scl70, anti-RNApolymeraseIII, and anti-PmScl), and others (rheumatoid factor and anti-SSA/ SSB); current immunomodulatory/immunosuppressive drugs (corticosteroids, hydroxychloroquine, methotrexate, or others), current modified Rodnan skin score [26]. The current or past following clinical manifestations were also recorded: arthralgia and/or clinical arthritis; history of clinical digital ulcers; pulmonary arterial hypertension confirmed by right heart catheterization (mean pulmonary arterial pressure 4 25 mmHg and pulmonary capillary wedge pressure r15 mmHg); pulmonary fibrosis (carbon monoxide diffusing capacity o 70% and typical interstitial pneumopathy on tomodensitometry); renal crisis (acute arterial hypertension with schizocytosis and proteinuria or microscopic hematuria, and acute renal insufficiency with or without renal histologic confirmation); and peripheral neuropathy supposed to be related to SSc by the practician and notified to it in the medical record. Potential confounders for VSE and AI were also collected from medical records: parity, history of UI related to postpartum, hysterectomy or bladder prolapse (operated or not), history of benign prostatic hyperplasia (treated or not), and current treatment with antidepressants and/or benzodiazepines. This study was approved by the Besançon and Poitiers university hospital local ethical committees.

Statistical analysis The data were entered into a computer database (Microsoft ACCESS software, PEDYN, Inc., Dallas, Fort Worth). Qualitative and ordinal variables were described using the number and frequency of each modality. Quantitative variables were expressed as the mean 7 standard deviation. The UDI-6 and Miller scores were subdivided into four categories, with 0 indicating no impairment, 1–6 slight impairment, 7–12 moderate impairment, and 13–18 severe impairment, based on the study of White et al. [27]. Groups with no or slight impairment (UDI-6 or Miller score r 6) were subsequently compared to those with modest or severe impairment (UDI-6 or Miller score Z 6) about risk factors and quality of life. Univariate analysis was performed to study the correlations between the various factors by comparing percentages using Fisher’s exact test for qualitative variables. Quantitative variables being expressed as the mean 7 standard deviation have a Gaussian distribution, so comparison were made using the parametric Student’s t-test due to their wide distribution. Regarding Fisher’s exact test in bilateral formulation, the 95% CI is only an approximation and may sometimes contain an odds ratio (OR) of 1 despite p o 0.05 [28]. The SF-36 scores were weighted by age and sex based on the scores of the U.S. population. Statistical analysis was performed with the statistical programming environment R/2.15. Univariate analysis of the quality of life was performed and the Student’s t-test was used to analyze the associations between different factors and QoL and anxiety/depression scores. Multiple linear regression analysis was then performed for each QoL and anxiety/depression score (dependent variables), adjusted for independent variables that might affect QoL, regardless of the VSE or AI, such as the presence of personal life events unrelated to SSc and antidepressants or benzodiazepines use. All significance levels are set to p o 0.05. Covariate interactions with the main explanatory variable were sought, and the analysis was stratified by the interaction variable if necessary. The alpha risk was set to 5% and statistical tests were performed bilaterally.

M. Martin et al. / Seminars in Arthritis and Rheumatism ] (2016) ]]]–]]]

3

excluded from the final analysis because of the lack of return of the self-administered questionnaires, and one related to a very incomplete questionnaire (Fig. 1). In the end, 121 patients have been analyzed, corresponding to a response rate of 87%. The mean age of the responders was 61.1 7 14.1 years, and 83.5% of which were women. SSc was mostly limited cutaneous type (72.7%), with a mean disease duration since diagnosis of 8 7 7.8 years. No significant differences were found between responders and non-responders about age, sex ratio, SSc type, modified Rodnan score, SSc duration since the first non-Raynaud’s symptom, overall SSc treatments, antidepressant or benzodiazepine use, neither about the rate of digital ulcers, PAH, pulmonary fibrosis, or arthralgia. There was also no difference according to the rate of the anti-centromere/anti-Scl70 positivity or other non-SSc related autoantibodies. Comparisons for neuropathic and kidney involvement were not performed because of the small numbers of patients (Table 1).

Vesico-sphincter events

Fig. 1. Flow-chart diagram.

Results Clinical and biologic characteristics Of the 186 patients examined for eligibility from the SSc registries, 139 met the inclusion criteria. Among them, 17 were

Among the 121 self-administered questionnaires received, 119 UDI-6 scores were interpretable (2 were insufficiently completed), and data on macroscopic hematuria and urinary tract infections were missing for five responders. Frequent urination was reported by 79 (66.3%) patients. Emergency UI (urge incontinence leaks) occurred in 47 (39.4%) patients. Urine leakage related to coughing, sneezing, or laughing was reported by 50 (42%) patients, and small amount of urine leakage (drops) by 28 (23.6%) patients. These two last forms of UI were regrouped into the term stress UI. Dysuria and pelvic pain were reported, by 41 (34.5%) and 22 (18.4%) patients, respectively. Macroscopic hematuria and urinary tract infections were reported by four (3.4%) of 116 patients and three (2.6%) of 116 patients (Fig. 2). Mixed UI (emergency and stress) occurred in 37 (31.1%) patients. Symptomatic intensity was reported as mild to moderate in most cases (Fig. 2). The average total UDI-6 score was 3.7 7 3.5.

Table 1 Patient's characteristics

Age (years)a Male/female ratio Diffuse/limited SSc ratio Time since first non-Raynaud's symptom (years)a Maintenance SSc treatment Corticosteroids Hydroxychloroquine and/or methotrexate Others immunosuppressive drugs Antidepressants Benzodiazepines Digital ulcers Modified Rodnan scorea PAH Pulmonary fibrosis Scleroderma renal crisis Arthralgia/arthritis Neuropathy ANA Anti-centromere Anti-Scl70 Anti-RNA polymerase III Anti-PMScl Other

Responders (n ¼ 121)

Non-responders (n ¼ 18)

p

61.1 7 14.1 0.2 0.3 8 7 7.8 37 (30.6%) 20 (16.5%) 17 (14%) 18 (14.9%) 24/116 (20.7%) 14/116 (12.1%) 38/114 (33.3%) 9.4 7 6.7 9 (7.4 %) 23 (19%) 1 (0.8%) 39 (32.2%) 4 (3.3%) 118 (97.5%) 65 (53.7%) 25 (20.1%) 0% 2 (1.6%) 15 (12.4%)

59 7 16.2 0.06 0.2 11.9 7 9.5 3 (16.7%) 1 (5.5%) 2 (11.1%) 1 (5.5%) 4 (26.7%) 3 (20%) 10 (55.5%) 11.2 7 9.8 1 (5.5%) 1 (5.5%) 0% 3 (16.7%) 0% 18 (100%) 9 (50%) 2 (11.1%) 0% 0% 3 (16.7%)

0.6b 0.31c 0.4c 0.11b 0.28c ND ND ND 0.74c 0.41c 0.11c 0.8c 1c 0.31c ND 0.27c ND 1c 0.8c 0.52c ND ND 0.7c

SSc, systemic sclerosis; DMARDs, disease-modifying antirheumatic drugs; PAH, pulmonary arterial hypertension; ANA, antinuclear antibodies; ND, not done. a b c

Mean 7 standard deviation. Student's test. Fischer's exact test.

M. Martin et al. / Seminars in Arthritis and Rheumatism ] (2016) ]]]–]]]

4

70

60

50

%

40

Not explored

30

Severe Moderate

20

Slight

10

0

Fig. 2. Frequency and intensity of vesico-sphincter events, hematuria and urinary infections in SSc.

The mean subscores were 1.9 7 1.7 for irritative symptoms (frequent urination and / or urge incontinence leak), 1.1 7 1.6 for stress UI, 0.5 7 0.7 for dysuria, and 0.3 7 0.6 for discomfort/pelvic pain. An UDI-6 score r 6 (no or slight disease)

was found in 82.3% of patients and a moderate and severe disease in 14.3% and 3.4%, respectively. We compared patients with a UDI-6 score r 6 (n ¼ 98) and those with a UDI-6 score Z 7 (n ¼ 21). A history of hysterectomy (OR ¼ 5.5, 95% CI: 1.5–20.6) or bladder prolapse (OR ¼ 10.2, 95% CI: 1.9–73) was associated with a UDI-6 score Z 7 (p o 0.01). The number of delivery and postpartum incontinence were more prevalent in the moderate or severe VSE groups, but the differences were not significant (2.4 vs. 1.8, p ¼ 0.1 and 11.8% vs. 9.1%, p ¼ 0.7, respectively). No man with a UDI-6 score Z 7 had a history of prostatic adenoma. A higher Miller score was observed in the group with a UDI-6 score Z 7 (p o 0.01), as well as inability to distinguish gas from stools before emptying, anal urge incontinence, intestinal bloating, and chronic constipation. These patients most often reported using protective underwear and anal seepage (Table 2). Patients were treated with benzodiazepines significantly more frequently in the group with a UDI-6 score Z 7 (p o 0.01). The difference was not significant for antidepressants (Table 2). Concerning the SSc characteristics, no significant difference was seen regarding the type, disease duration, visceral involvement, treatment, or auto-antibody status (Table 2). Anal incontinence and lower gastro-intestinal symptoms All the 121 questionnaires returned were interpretable. AI to gas was reported by 61 (50.4%) patients, followed by AI to liquid stool by 40 (33%) patients, and to solid stool by 20 (16.5%) patients.

Table 2 Associated factors to vesico-sphincter events

a

Age (years) Male/female ratio Diffuse/limited SSc ratio Time since first non-Raynaud's symptom (years)a Modified Rodnan scorea Maintenance SSc treatment Corticosteroids Hydroxychloroquine and/or methotrexate Others immunosuppressive drugs Antidepressants Benzodiazepines Anti-centromere Anti-Scl70 Digital ulcers Arthralgia/arthritis PAH Pulmonary fibrosis Miller score Clinical gastroesophageal reflux Clinical gastroparesis Chronic constipation Bloating Chronic diarrhea Underwear soiling Use of underwear protections Urge to defecate Non distinction gas/stools Rectal prolapse

UDI-6

UDI-6

0–6

7–18

(n ¼ 98)

(n ¼ 21)

60.2 7 15.5 0.2 0.3 7.8 7 8.2 8.9 7 6.8 28 (28.6%) 14 (14.3%) 14 (14.3%) 13 (13.4%) 18/93 (19.3%) 7/93 (7.5%) 50 (51%) 24 (24.5%) 32/92 (34.8%) 33 (33.7%) 7 (7.1%) 18 (18.4%) 3,1 7 1.7a 69 (70.4%) 40 (40.8%) 23 (23.5%) 25 (25.5%) 8 (8.2%) 16 (16.3%) 18 (18.3%) 22/97 (22.7%) 14 (14.3%) 3 (3.1%)

65.2 7 11.9 0 0.2 9 7 6.6 11.5 7 5.8 9 (42.8%) 6 (28.6%) 3 (14.3%) 5 (23.8%) 5 (23.8%) 7 (33.3%) 13 (61.9%) 1 (4.8%) 6 (28.6%) 6 (28.6%) 1 (4.8%) 4 (19%) 8.1 7 4.4a 15 (71.4%) 13 (61.9%) 12 (57.1%) 14 (66.7%) 3 (14.3%) 10 (47.6%) 11/20 (55%) 14 (66.7%) 12 (57.1%) 2 (9.5%)

p

0.1b 0.02c 0.18c 0.49b 0.21b 0.21c 0.12c 1c 0.31c 0.76c 0.004c 0.47c 0.07c 0.79c 0.79c 1c 1c 0.003b 1c 0.09c 0.003c o 0.001c 0.41c 0.003c 0.001c o 0.001c o 0.001c 0.21c

OR [95% CI]

NA 0.4 [0.1–1.4]

1.9 2.4 1 2 1.3 6 1.5 0.2 0.8 0.8 0.6 1

[0.6–5.4] [0.6–8] [0.2–4.1] [0.5–7.1] [0.3–4.4] [1.5–23.7] [0.5–4.7] [0.03–1.1] [0.2–2.3] [0.2–2.4] [0.01–5.6] [0.2–3.8]

1 2.3 4.3 5.7 1.9 4.6 5.3 6.7 7.8 3.3

[0.3–3.6] [0.8–7.1] [1.4–13.1] [1.9–189] [0.3–8.8] [1.5–14.3] [1.7–17.1] [2.2–22.2] [2.5–25.5] [0.3–30.8]

Bold values corresponds to Student's test, which is non parametric. OR, odds ratio; SSc, systemic sclerosis; DMARDs, disease-modifying antirheumatic drugs; NA, not applicable (CI contains 0); PAH, pulmonary arterial hypertension. a b c

Mean 7 standard deviation Student's test. Fischer's exact test.

M. Martin et al. / Seminars in Arthritis and Rheumatism ] (2016) ]]]–]]]

5

Table 3 Associated factors to anal incontinence

a

Age (years) Male/female ratio Diffuse/limited SSc ratio Time since fisrt non-Raynaud's syymptom (years)a Modified Rodnan scorea Maintenance SSc treatment Corticosteroids Hydroxychloroquine and/or methotrexate Others immunosuppressive drugs Antidepressants Benzodiazepines Anti-centromere Anti-Scl70 Digital ulcers Arthralgia/arthritis PAH Pulmonary fibrosis UDI-6 score Clinical gastroesophageal reflux Clinical gastroparesis Chronic constipation Bloating Chronic diarrhea Underwear soiling Use of underwear protections Urge to defecate Non distinction gas/stools Rectal prolapse

Miller

Miller

0–6

7–18

(n ¼ 91)

(n ¼ 30)

59.9 7 143 0.17 0.32 7.3 7 7.7 9 7 7.1 27 (29.7%) 12 (13.2%) 12 (13.2%) 14 (15.5%) 16/86 (18.6%) 10/86 (11.6%) 46 (50.5%) 21 (23.1%) 33/86 (38.4%) 28 (30.8%) 6 (6.6%) 15 (16.4%) 2.9 7 1.2a 63 (69.2%) 37 (41.1%) 24 (26.4%) 24/90 (26.7%) 5 (5.5%) 11 (12.1%) 11/90 (12.2%) 19 (20.9%) 12 (13.2%) 3 (3.3%)

64.8 7 16.4 0.13 0.13 10.2 7 8.2 11.2 7 4.9 10 (33.3%) 8 (26.7%) 5 (16.7%) 4 (13.3%) 8 (26.7%) 4 (13.3%) 19 (63.3%) 4 (13.3%) 5/28 (17.8%) 11 (36.7%) 3 (10%) 8 (26.7%) 5.8 7 3.1a 22 (73.3%) 16 (53.3%) 12 (40%) 15 (50%) 6 (20%) 15 (50%) 18 (60%) 17 (58.6%) 14 (46.7%) 2 (6.7%)

p

0.14b 0.78c 0.06c 0.09b 0.15b 0.82c 0.09c 0.76c 1c 0.43c 0.75c 0.29c 0.31c 0.06c 0.65c 0.68c 0.28c 0.001b 0.82c 0.29c 0.17c 0.02c 0.03c o0.001c o0.001c o0.001c o0.001c 0.59c

OR [95% CI]

0.7 [0.2–2.5] 0.3 [0.07–1.1]

1.2 2.4 1.3 0.8 1.6 1.2 1.7 0.5 0.3 1.3 1.6 1.8 1.2 2.6 1.8 2.7 4.2 7.1 10.5 5.3 5.7 2.1

[0.4–3.1] [0.7–7.3] [0.3–4.5] [0.2–3] [0.5–4.6] [0.2–4.5] [0.7–4.4] [0.1–1.7] [0.1–1.1] [0.5–3.3] [0.2–7.9] [0.6–5.4] [0.5–3.6] [0.7–4.1] [0.7–4.8] [1.1–7] [0.98–19.2] [2.5–21] [3.7–31.6] [2–14.5] [2–16.3] [0.2–19.1]

OR, odds ratio; SSc, systemic sclerosis; DMARDs, disease-modifying antirheumatic drugs; PAH, pulmonary arterial hypertension. a b c

Mean 7 standard deviation. Student's test. Fischer's exact test.

The average Miller score was 3.96 7 5.2. A score of 6 or lower (no AI or slight) was observed in 75.2% of patients, and 12.4% had moderate or severe AI. Anal seepage and regular use of protective underwear were reported, by 26 (21.5%) and 29 (24.2%) patients, respectively. Rectal urge incontinence was observed in 36 (30%) patients, and 26 (21.5%) patients reported an inability to distinguish gas from stool before emptying. We compared patients with Miller score r 6 (n ¼ 91), and those with a score Z 7 (n ¼ 30). The frequency of obstetrical factors (number of deliveries, hysterectomy, history of postpartum UI, or bladder prolapse) did not differ significantly between the two groups (data not shown). A significantly higher average UDI-6 total score was observed in patients with a Miller Z 7 score (p o 0.01), as well as inability to distinguish gas from stool before emptying, urge AI, and, at the limit of significance, intestinal bloating or diarrhea for longer than 3 months (Table 3). The use of protective underwear and anal seepage also occurred significantly more frequently in these patients (p o 0.001) (Table 3). Conversely, there was no significant difference in the presence of a clinical gastroesophageal reflux disease, clinical gastroparesis, chronic constipation, or rectal prolapse.

Specific VAS scores of VSE and AI, but also the VAS reflecting impaired QoL related to SSc, were significantly higher in patients with a UDI-6 score Z 7 (p o 0.01). The same results were observed for the HADS results. Regarding the SF-36, mental and physical summary scores, and the eight dimensions studied were significantly lower in patients with a UDI-6 score Z 7 (p r 0.01). We conducted multiple linear regression analyses for anxiety and depression scores, as well as for each SF-36 subscale, which were adjusted for independent variables that might have affected QoL, regardless of the VSE or AI, such as the presence of personal life events unrelated to SSc (i.e., family or professional issues) and antidepressants or benzodiazepines use. A life event unrelated to SSc in the past 12 months was reported by 62% of patients with a UDI-6 score Z 7, compared with 53.1% of those with a UDI-6 score r 6 (p ¼ 0.2) (Table 4). Moderate or severe VSE remained independently associated with a decrease in all scores of the SF-36, and with higher scores for anxiety and depression (Table 4). However, physical summary score and the physical activity score were no longer significant when taking antidepressants. Conversely, the decreased mental health score was significant only in the presence of personal life events unrelated to SSc (Table 4).

Quality of life Vesico-sphincter events

Anal incontinence

We compared the patients with a UDI-6 score r 6 (n ¼ 98) vs. those with a UDI-6 score Z 7 (n ¼ 21) (Table 4).

We compared the patients with a Miller score r 6 (n ¼ 91) vs. those with a Miller score Z 7 (n ¼ 30) (Table 5).

M. Martin et al. / Seminars in Arthritis and Rheumatism ] (2016) ]]]–]]]

6

Table 4 Vesico-sphincter events and quality of life UDI-6

UDI-6

Univariate

Multivariate

0–6

7–18

p

p

o0.001b 0.002b o0.001b 0.001b 0.007b 0.01b 0.001b 0.002b 0.01b 0.002b 0.007b 0.002b 0.001b 0.002b o0.001b

ND ND ND o0.001c 0.003c 0.002c,d o0.001c o0.001c,d 0.03c 0.003c 0.02c 0.005c o 0.001c 0.002c o0.001c,e

(n ¼ 98) VAS VSE VAS AI VAS overall SSc HADa HADd PSs MSs Physical functioning Role limitations due to physical problems Bodily pain General health perception Vitality Social functioning Role limitations due to emotional problems Mental health

0.7 0.7 2.9 7.5 5.7 41.15 42.5 42 39.3 40.9 38.7 46 42.8 38 42.2

7 7 7 7 7 7 7 7 7 7 7 7 7 7 7

(n ¼ 21) a

1.3 2a 2.6a 4.3a 4.3a 9.8a 11.2a 11.4a 12a 10.1a 10.1a 10.9a 11.3a 13.8a 9.9a

4.7 7 2.9a 3.5 7 3.5a 5.7 7 3.2a 11.6 7 4.9a 9.3 75.2a 35.2 7 9.4a 32.2 7 14.2a 32.2 7 11.9a 32.1 7 11a 33 7 9.7a 32.5 7 8.5a 37.7 7 9.4a 32.4 7 12.5a 26.8 7 13.3a 29.2 7 13.9a

Bold values corresponds to Student's test, which is non parametric. VAS, visual analogique scale; VSE, vesico-sphincter events; AI, anal incontinence; HADa, hospital anxiety and depression scale anxiety score; HADd, hospital anxiety and depression scale depression score; PSs, SF-36 physical summary score; MSs, SF-36 mental summary score; ND, not done. a

Mean 7 standard deviation. Student's test. c Multiple linear regression (after adjustment on life events, benzodiazepines and antidepressants use). d Without antidepressants use (p ¼ 0.41 if antidepressants use). e If life events associated (p ¼ 0.17 without life events associated). b

Specific VAS scores of the VSE, AI, and VAS reflecting impaired quality of life related to SSc were significantly higher in patients with a Miller score Z 7 (p o 0.01). Anxiety and depression scores were generally comparable between patients with no or slight AI and those with moderate or severe AI. The overall SF-36 scores were slightly lower in the group with a Miller score Z 7, however, no differences reached significance.

A life event unrelated to SSc in the past 12 months was reported by 46.7% of patients with a Miller score Z 7 compared with 58.2% of those with a Miller score r 6 (p ¼ 0.09). Multivariate analysis showed that, unlike VSE, decreased SF-36 scores or increased anxiety or depression scores in the group with a Miller score Z 7 did not differ significantly from those with a Miller score r 6 (Table 5). Moreover, taking antidepressants or benzodiazepines or the presence of life events were not associated with AI.

Table 5 Anal incontinence and quality of life Miller

Miller

Univariate

Univariate

0–6

7–18

p

p

0.008b o0.001b 0.004b 0.39b 0.31b 0.23b 0.51b 0.12b 0.23b 0.21b 0.73b 0.19b 0.48b 0.31b 0.29b

ND ND ND 0.81c 0.82c 0.41c 0.84c 0.56c 0.59c 0.43c 0.7c 0.61c 0.92c 0.88c 0.81c

(n ¼ 91) VAS VSE VAS AI VAS overall SSc HADa HADd PSs MSs Physical functioning Role limitations due to physical problems Bodily pain General health perception Vitality Social functioning Role limitations due to emotional problems Mental health

0.4 1.1 2.9 8 6.1 40.8 40.8 41.3 39 40.4 37.7 45.3 41.4 36.9 40.6

7 7 7 7 7 7 7 7 7 7 7 7 7 7 7

(n ¼ 30) a

1.4 1.9a 2.8a 4.5a 4.6a 9.5a 12.3a 11.5a 11.6a 10.4a 10.3a 10.9a 12.1a 14.1a 11a

3.7 2.6 4.8 8.8 7.1 38 39 37 35.7 37.5 37 42.2 39.6 33.7 37.7

7 7 7 7 7 7 7 7 7 7 7 7 7 7 7

3.6a 2.8a 2.9a 4.9a 4.7a 11.1a 13.4a 13.3a 13a 10.8a 9.5a 11.2a 12.1a 14.8a 13.5a

Bold values corresponds to Student's test, which is non parametric. VAS, visual analogique scale; VSE, vesico-sphincter events; AI, anal incontinence; HADa, hospital anxiety and depression scale anxiety score; HADd, hospital anxiety and depression scale depression score; PSs, SF-36 physical summary score; MSs, SF-36 mental summary score; ND, not done. a b c

Mean 7 standard deviation. Student's test. Multiple linear regression (after adjustment on life events, benzodiazepines, and antidepressants use).

M. Martin et al. / Seminars in Arthritis and Rheumatism ] (2016) ]]]–]]]

Discussion In this study, we found that AI and VSE are associated, common, and sometimes severe in SSc patients, with a potential negative impact on QoL, especially for VSE. Although several publications have demonstrated a negative effect of gastro-intestinal manifestations in QoL of patients with SSc, our work is the first to date to specifically explore in details potential associated VSE and their impact in SSc QoL The high response rate to the questionnaires in our study (87%) is comparable to that published in studies about lower gastrointestinal involvement in SSc, but which included fewer patients [3,12], and substantially higher than that in the two series that evaluated VSE based on self-reporting questionnaires [8,9]. In addition, responders and non-responders are comparable in our study, which limits selection bias, and the patient’s characteristics seems not to be different from the SSc general population (Table 1) what could confer a good external validity to our study. As described in the Method section, SSc patients anti-Scl70 positive were classified as diffuse SSc, regardless of their skin fibrosis extension, based on the fact that this antibody status seems more closely associated to disease severity than skin sclerosis extension [24] and a good predictive marker of progression to diffuse cutaneous involvement and pulmonary fibrosis [25]. This might reduce the statistical weight of SSc with diffuse skin sclerosis, however, anti-Scl70 antibodies are more frequent in the UDI-6 and Miller r 6 groups, supporting the hypothesis that a diffuse form would not be associated with more severe VSE or AI than limited SSc. Likewise, Umar et al. [29] did not find any difference in lower gastro-intestinal and pelvic floor symptoms rates between limited and diffuse SSc. In addition, urinary symptoms and urodynamic changes were most often reported in association with the limited cutaneous form [7], and there were no significant differences in voiding volumes based on the SSc types [30]. The UDI-6 and Miller scales are validated, and have been selected over other scales related to their simplicity, and shortness. Moreover, these scales are weighted allowing a slight discrimination for intensity of the symptoms. Therefore, the main objective of our study being to determine the frequency and the severity of VSE and AI in SSc, we excluded symptoms scales that included a quality-of-life questionnaire, that would have added a subjectivity factor leading to an over- or underestimate total score for a given symptom intensity from one patient to another. The UDI-6 was validated in women [31] and also explores irritative signs and dysuria, making its use interesting and validated in men [32]. Our study shows that VSE in SSc are dominated by stress symptoms. However, our work did not exclude secondary Sjögren’s syndrome (SS) associated with SSc, which could lead to some analysis bias. In fact, as in SLE, interstitial cystitis seems more important in SS than in general population and may overestimate the irritative rate in our SSc cohort [33]. Nevertheless, this is consistent with unpublished epidemiologic data in which overactive bladder was reported by nearly 90% of patients of both sexes [8,9]. Moreover, predominant irritative bladder seems consistent with the few available pathophysiologic data to date. Some authors reported a fibrotic [6,10] or microthrombotic [5] microvascular injury of the bladder, resulting in predominantly hypocompliant and small bladder capacity compared control subjects [30]. Others highlighted an autonomic nervous system dysfunction without correlation between bladder fibrosis level and bladder dysfunction [7]. Our study confirms a high prevalence of AI in SSc, which is moderate or severe in 12.4% of cases. Franck-Larsson et al. [12] compared 79 SSc patients to 158 age-matched healthy control and found that AI was 3–4 times more prevalent in patients than controls, with significantly higher mean Miller score (2.79 vs. 1.27) . Our current patients’ mean score was higher (3.96), which can be explained by an AI to solid stool twice more frequent. The

7

frequencies of anal seepage, inability to distinguish stool from gas, and regular use of protective underwear are comparable. The lack of a significant association between AI and rectal prolapse in our study was likely due to the small number of affected patients. Chronic diarrhea is positively associated with AI in our study. However, anti-diarrheic treatments have not been recorded. It would be interesting to further analysis their impact on a potential AI improvement, as well as for the role of stool softeners or laxatives use. The physiopathology of AI in SSc seems related to fibrosis and/or smooth muscle cells atrophy of the internal anal sphincter, by microvascular injury and by ischemic or autoimmune neuropathy [34–36]. However, the too small number of patients with anti-PmScl in our study (n ¼ 2) did not allow to assess a potential role of this auto-antibody in AI in SSc. Only few studies have explored a potential relationship between AI and UI in SSc. Umar et al. [29] recently highlighted that 23% (n ¼ 37) of SSc women suffered from dual incontinence (AI and UI). Our results are consistent with this finding with a positive association between VSE and AI. Moreover, we explored these symptoms both in men and women, with no difference between gender. All symptoms intensities combined, 58.8% of patients had AI and Z 1 symptom related to VSE. This frequency was 8.2% when considering only moderate or severe intensity. In the general population, the association between AI and UI stress [37] or UI overall [38,39] has been reported. However, the questionnaires and average patient age were not comparable. Interestingly, in our current study and obstetric factors, including parity are not positively associated with AI, unlike the general population [37,40]. In the other part, autoantibodies against type 3 muscarinic receptors were recently highlighted in the SSc’s internal anal sphincter [36]. These receptors are also located in the detrusor muscle physiologically. Thus, specific autoantibodies against these receptors may be involved in bladder irritative symptoms and AI in SSc. Unexpectedly, the AI had a negative, but not significant impact on the HAD or SF-36 scores, contrary to the results about VSE. Recently, despite different evaluation scales, Bodukam et al., in a recent study of 152 SSc patients, showed a negative, but not significant impact of fecal soilage on mood, contrary to the other gastro-intestinal symptoms. The authors did not explain this difference [41]. Umar et al. also reported a negative but not significant impact of AI on Scleroderma Health Assessment Questionnaire (SHAQ) QoL total score, contrary to UI and dual incontinence [29]. In our current study, we speculate that the SF-36 could lack sensitivity, mainly reflecting the impact of painful involvements and impaired mobility [42,43]. Irritative bladder symptoms are felt as more painful than AI by the patients, leading to a significant impact on the SF-36 score, unlike AI. Another explanation could be related to digital ulcers that are known to decrease the quality of life in SSc [43], and which are predominant in the Miller r 6 group, although the difference was not significant. Another explanation could be a lack of sensitivity related to the subdivision of UDI-6 and Miller scores into four categories. Therefore, a complementary perspective would be to analyze the scores continuously to define a threshold score at which its impact is significant on the QoL. Moreover, due to potential other confounders as medical comorbidities that could be linked to AI or VSE (e.g.,: diabetes mellitus type II, obesity), not explored in our study, the relationship between VSE and QoL remain to be interpreted with caution.

Our study has some limitations The first is the lack of a non-SSc control group. The main objective of our study being to assess the frequency and the

8

M. Martin et al. / Seminars in Arthritis and Rheumatism ] (2016) ]]]–]]]

severity of VSE and VSE in SSc, a control group group was not essential in first intention. Moreover, the SSc patients are divided into two groups according to their UDI-6 and Miller score, making it possible intra-cohort comparisons. Our work is a preliminary study, especially about VSE, providing data about VSE rate and type in SSc, which will be essential to define precisely the number of control to include in a further case-control study. It would be interesting to compare SSc with other rheumatic diseases and with patients from general population matched by age, gender, and comorbidities. Schmeiser et al. in a comparison of 34 SSc patients with 34 patients with rheumatoid arthritis, SLE, or psoriasic arthritis did not found difference in fecal incontinence, neither in meteorism, diarrhea, or constipation rates. SSc seemed related to a more prevalent upper gastro-intestinal tract involvement rather than lower GI tract involvement [44]. Another limitation is the lack of data about other potential cofounders for AI and/or VSE, which might lead to under/overprevalence of AI, and VSE and analysis bias. For example, diabetes type II mellitus, as risk factor for AI, would require to be recorded, and the potential role of anti-cholinergic or diuretics medications, history of gastro-intestinal tract or bladder surgery or nephrolithiasis would need to be further explored. A third limitation is the lack of objective confirmation of the reported AI and VSE symptoms. However, there were no retrospective data available in the medical records about anorectal/ pelvic manometry or specific digital rectal examination, even for patients turned out to have AI or VSE. This suggest that, although prevalent in SSc, AI and VSE are underestimated or overlooked and hence, undertreated. Moreover, an objective approach could be difficult, especially in AI, because of important discrepancies between self-reported and objective assessment. In fact, normal resting and squeeze anal pressures by digital rectal examination could be seen in 30% and, 12%, respectively of self-reported AI patients [45] and normal high-resolution anorectal manometry in 10–40% of such patients [40].

Conclusion AI and VSE are common and associated in SSc, and sometimes severe with a potential negative impact on QoL. According to the limitations of our study, a prospective case-control study with dynamic and manometric tests for AI and VSE is needed to confirm these findings, identify potential risk factors, and clarify the potential pathophysiologic relationship between AI and VSE in SSc. The aim is to provide early, individualized, and multidisciplinary management, including psychological support, to SSc patients with VSE and/or AI.

References [1] Chifflot H, Fautrel B, Sordet C, Chatelus E, Sibilia J. Incidence and prevalence of systemic sclerosis: a systematic literature review. Semin Arthritis Rheum 2008;37:223–5. [2] Omair MA, Lee P. Effect of gastrointestinal events on quality of life in 87 consecutive patients with systemic sclerosis. J Rheumatol 2012;39:992–6. [3] Trezza M, Krogh K, Egekvist H, Bjerring P, Laurberg S. Bowel problems in patients with systemic sclerosis. Scand J Gastroenterol 1999;34:409–13. [4] Cotterill N, Norton C, Avery KN, Abrams P, Donovan. JL. A patient-centered approach to developing a comprehensive symptom and quality-of-life assessment of anal incontinence. Dis Colon Rectum 2008;51:82–7. [5] Oumaya C, Ouali M, Ciofu C, Haab F, Thibault P. Bladder involvement in scleroderma: report of 3 cases. Ann Urol (Paris) 2003;37:267–71. [6] Lazzeri M, Beneforti P, Benaim G, Corsi C, Ciambrone V, Marrapodi E, et al. Vesical dysfunction in systemic sclerosis (scleroderma). J Urol 1995;153: 1184–7. [7] Minervini R, Morelli G, Minervini A, Pompaloni S, Tognetti A, Fiorentini L, et al. Bladder involvement in systemic sclerosis: urodynamic and histological assessment in 23 patients. Eur Urol 1998;34:47–52.

[8] K. Sanchez, P. Denys, A. Berezne, F. Giuliano, H. Abid, S. Poiraudeau, et al. Genital and sexual and urinary disorders in 83 patients with systemic sclerosis. Oral communication in the 62e congress of the French National Society of Internal Medicine, Paris, France; 2010 December. [9] H. Abid, P. Denys, A. Berezne, F. Giuliano, K. Sanchez, S. Poiraudeau, et al. Evaluation of urinary disorders in scleroderma: about 69 patients. Oral communication in the 26e congress of the Rehabilitation and Physical Medicine, Nantes, France; 2011 October. [10] La Civita L, Fiorentini L, Tognetti A, Pasero G, Ferri C. Severe urinary bladder involvement in systemic sclerosis. Case report and review of the literature. Clin Exp Rheumatol 1998;16:591–3. [11] CA. Matthews, WE. Whitehead, MK. Townsend, F. Grodstein. Risk factors for urinary, fecal, or dual incontinence in the Nurses’ Health Study. Obstet Gynecol 2013;122:539–45. [12] Franck-Larsson K, Graf W, Rönnblom. A. Lower gastrointestinal symptoms and quality of life in patients with systemic sclerosis: a population-based study. Eur J Gastroenterol Hepatol 2009;2:176–82. [13] LeRoy EC, Black C, Fleischmajer R, Jablonska S, Krieg T, Medsger TA Jr, et al. Scleroderma (systemic sclerosis): classification, subsets and pathogenesis. J Rheumatol 1988;15:202–5. [14] LeRoy EC, Medsger TA Jr. Criteria for the classification of early systemic sclerosis. J Rheumatol 2001;28:1573–6. [15] Nishizaki Y, Tamaki H, Yukawa S, Matsui Y, Okada M. Comparison between Japanese and non-Japanese features of lupus cystitis based on case reports including novel therapy and a literature review. Intern Med 2011;50:961–8. [16] Keller JJ, Liu SP, Lin. HC. A case–control study on the association between rheumatoid arthritis and bladder pain syndrome/interstitial cystitis. Neurourol Urodyn 2013;32:980–5. [17] Janssens P, Arnaud L, Galicier L, Mathian A, Hie M, Sene D, et al. Lupus enteritis: from clinical findings to therapeutic management. Orphanet J Rare Dis 2013;8:67. [18] Chattopadhyay P, Abby Philips C, Dhua D, Saha S. Systemic lupus erythematosus presening as ischaemic proctitis. Lupus 2011;20:653–5. [19] Uebersax JS, Wyman JF, Shumaker SA, McClish DK, Fantl JA. Short forms for Assessment to life quality and symptom distress for urinary incontinence in women: the Incontinence Impact Questionnaire and the Urogenital Distress Inventory. Continence Program for Women Research Group. Neurourol Urodyn 1995;14:131–9. [20] Miller R, Bartolo DC, Locke-Edmunds JC, Mortensen NJ. Prospective study of conservative and operative treatment for faecal incontinence. Br J Surg 1988;75:101–5. [21] Friedman S, Samuelian JC, Lancrenon S, Even C, Chiarelli P. Three-dimensional structure of the Hospital Anxiety and Depression Scale in a French primary care population wide suffering from major depression. Psychiatry Res 2001;104:247–57. [22] Ware JE, Kosinski M, Dewey JE. How to score Version 2 of the SF-36 health survey. Lincolnshire, RI: QualityMetric Incorporated; 2000. [23] Bullinger M, Alonso J, Apolone G, Leplege A, Sullivan M, Wood-Dauphinee S, et al. Translating health status questionnaires and evaluating their quality: IQOLA the project approach. International quality of life assessment. J Clin Epidemiol 1998;51:913–23. [24] Reveille JD, Solomon DH. American College of Rheumatology Ad Hoc Committee of Immunologic Testing Guidelines. Evidence-based guidelines for the use of immunologic tests: anticentromere, Scl-70, and nucleolar antibodies. Arthritis Rheum 2003;15:399–412. [25] Walker UA, Tyndall A, Czirják L, et al. Clinical risk assessment of organ manifestations in systemic sclerosis: a report from the EULAR Scleroderma Trials And Research group database. Ann Rheum Dis 2007;66:754–63. [26] Clements PJ, Lachenbruch PA, Seibold JR, Zee B, Steen VD, Brennan P, et al. Skin thickness score in systemic sclerosis: an assessment of interobserver variability in 3 independent studies. J Rheumatol 1993;20:1892–6. [27] White E, Hermieu JF, Ravery V, Moulinier F, Delmas V, Boccon-Gibod. L. Value of the use of a questionnaire, in the assessment of incontinence surgery. Prog Urol 1999;9:88–94. [28] Fay MP. Confidence intervals that match Fisher’s exact or Blaker’s exact tests. Biostatistics 2010;11:373–4. [29] Umar SB, Griffing L, Garcia H, Foxx-Oreinstein AE, DiBaise JK, Crowell. MD. The Impact of pelvic floor and lower gastrointestinal symptoms on quality of life in women with systemic sclerosis. J Clin Gastroenterol 2015 [Epub ahead of print]. [30] Kucharz EJ, Jonderko G, Rubisz-Brzezinska J, Konca A, Jarczyk R. Premictional volume and contractility of the urinary bladder in patients with systemic sclerosis. Clin Rheumatol 1996;15:118–20. [31] Donovan J, Bosch R, Gotoh M, Jackson S, Naughton M, Radley S, et al. Symptom and quality of life assessment. In: Abrams P, Cardozo L, Khoury S, Wein A, editors. Incontinence: Proceedings of the Third International Consultation on Incontinence, June 26–29, 2004. 3rd ed. Plymouth, United Kingdom: Health Publication Ltd., 2005. p. 519–84. [32] Utomo E, Korfage IJ, Wildhagen MF, Steensma AB, Bangma CH, Blok BF. Validation of the urogenital distress inventory (UDI-6) and Incontinence Impact Questionnaire (IIQ-7) in a Dutch population. Neurourol Urodyn 2015;34:24–31. [33] Darrieutort-Laffite C, André V, Hayem G, Saraux A, Le Guern V, Le Jeune C, et al. Sjögren’s syndrome complicated by interstitial cystitis: a case series and literature review. Joint Bone Spine 2015;82:245–50.

M. Martin et al. / Seminars in Arthritis and Rheumatism ] (2016) ]]]–]]] [34] Fynne L, Worsøe J, Laurberg S, Krogh K. Faecal incontinence in patients with systemic sclerosis: is an impaired internal anal sphincter the only cause? Scand J Rheumatol 2011;40(6):462–6. [35] Thoua NM, Abdel-Halim M, Forbes A, Denton CP, Emmanue AV. Fecal incontinence in systemic sclerosis is secondary to neuropathy. Am J Gastroenterol 2012;107(4):597–603. [36] Singh J, Mehendiratta V, Del Galdo F, Jimenez SA, Cohen S, DiMarino AJ, et al. Immunoglobulins from scleroderma patients inhibit the muscarinic receptor activation in internal anal sphincter smooth muscle cells. Am J Physiol Gastrointest Liver Physiol 2009;297:1206–13. [37] Abramov Y, Sand PK, Botros SM, Gandhi S, Miller JJ, Nickolov A, et al. Prevalence and correlates of fecal incontinence impact among older women. Dis Colon Rectum 2013;56:1080–6. [38] Halland M, Koloski NA, Jones M, Byles J, Chiarelli P, Forder P, et al. Risk factors for female anal incontinence: new insight through the EvanstonNorthwestern twin sisters study. Obstet Gynecol 2005;106:726–32. [39] Ditah I, Devaki P, Luma HN, Ditah C, Njei B, Jaiyeoba C, et al. Prevalence, Trends, and Risk Factors for Fecal Incontinence in Adults United States, 2005– 2010. Clin Gastroenterol Hepatol 2014;12:636–43.

9

[40] Lam TJ, Kuik DJ, Felt-Bersma RJ. Anorectal function assessment and predictive factoring for faecal incontinence in 600 patients. Colorectal Dis 2012; 14:214–23. [41] Bodukam V, Hays RD, Maranian P, Furst DE, Seibold JR, Impens A, et al. Association of gastrointestinal involvement and depressive symptoms in patients with systemic sclerosis. Rheumatology (Oxford) 2011;50: 330–4. [42] Georges C, Chassany O, Toledano C, Mouthon L, Tiev K, Meyer O, et al. Impact of pain in health related quality of life of patients with systemic sclerosis. Rheumatology (Oxford) 2006;45:1298–302. [43] Mouthon L, Mestre-Stanislas C, Berezne A, Rannou F, Guilpain P, Revel M, et al. Impact of digital ulcers on disability and health-related quality of life in systemic sclerosis. Ann Rheum Dis 2010;69:214–7. [44] Schmeiser T, Saar P, Jin D, Noethe M, Müller A, Soydan N, et al. Profile of gastrointestinal involvement in patients with systemic sclerosis. Rheumatol Int 2012;32:2471–8. [45] Dobben AC, Terra MP, Deutekom M, Gerhards MF, Bijnen AB, Felt-Bersma RJF, et al. Anal inspection and digital rectal examination compared to anorectal physiology tests and endoanal ultrasonography in evaluating fecal incontinence. Int J Colorectal Dis 2007;22:783–90.