The association between Ehlers- Danlos syndrome hypermobility type and gastrointestinal symptoms in university students: a cross- sectional study

Transcription

1 Received: 27 May 2016 Accepted: 18 August 2016 DOI: /nmo ORIGINAL ARTICLE The association between Ehlers- Danlos syndrome hypermobility type and gastrointestinal symptoms in university students: a cross- sectional study A. Fikree 1 R. Aktar 1 J. K. Morris 2 R. Grahame 3 C. H. Knowles 1 Q. Aziz 1 1 Wingate Institute of Neurogastroenterology, Centre for Neuroscience and Trauma, Bizard Institute, Barts and the London School of Medicine and Dentistry, Queen Mary University of London, London, UK 2 Wolfson Institute of Preventive Medicine, Barts and the London School of Medicine and Dentistry, Queen Mary University of London, London, UK 3 Centre for Rheumatology Research, University College London, London, UK Correspondence Professor Qasim Aziz, Wingate Institute of Neurogastroenterology, Centre for Neuroscience and Trauma, Bizard Institute, Barts and the London School of Medicine and Dentistry, Queen Mary University of London, London, UK. q.aziz@qmul.ac.uk Abstract Background: Patients with Ehlers- Danlos syndrome hypermobility type (EDS- HT) have increased prevalence of gastrointestinal (GI) symptoms, particularly reflux and dyspepsia. EDS- HT is associated with dysautonomia, psychopathology, and chronic pain which can be associated with GI symptoms. The association between GI symptoms and EDS- HT in a non- patient population and the effect of the above- mentioned factors has never been studied. Methods: In a cross sectional study, a hypermobility questionnaire was used to screen university students; further clinical examination established the diagnosis of EDS- HT. Validated questionnaires assessed for GI, somatic, pain and autonomic symptoms, psychopathology and quality of life (QOL). These were compared in students with and without EDS- HT; logistic regression analysis examined associations between EDS- HT, GI symptoms and other variables. Key Results: Of 1998 students screened, 162 were included: 74 EDS- HT (21.0 years, 53% female) vs 88 Non- EDS- HT (21.5 years, 65% female). Compared to non- EDS- HT students, EDS- HT students were more likely to have multiple GI symptoms (41.9% vs 27.3% P=.05), particularly postprandial fullness (34.4% vs 15.9%, P=.01) and early satiety (32% vs 17%, P=.03), greater autonomic (P<.001) and somatic symptoms (P=.04) but not psychopathology (P>.8). The association between EDS- HT and postprandial symptoms was dependent on autonomic factors but independent of pain and psychopathology. Pain- related QOL scores were reduced in the EDS- HT group (80 vs 90, P=.03). Conclusions and Inferences: The previously described association between EDS- HT, dyspepsia, pain and autonomic symptoms in patients is also present in non- patient groups. Future studies are necessary to explore the etiological role of connective tissue in GI and extra intestinal symptoms. KEYWORDS autonomic, fibromyalgia, functional dyspepsia, functional gastrointestinal disorders, hypermobility Neurogastroenterol Motil 2016; 1 9 wileyonlinelibrary.com/journal/nmo 2016 John Wiley & Sons Ltd 1

2 2 Fikree et al. 1 INTRODUCTION Ehlers- Danlos syndrome hypermobility type (EDS- HT), formerly known as EDS Type III, is a common but underdiagnosed 1 4 hereditary disorder of connective tissue now considered identical to the joint hypermobility syndrome (JHS). 5 It is characterized by joint hyperflexibility, skin hyperextensibility and musculoskeletal symptoms 6 and is inherited as an autosomal dominant trait with incomplete penetrance and variable expression. 7 The exact genetic defect in EDS- HT has not been elucidated, although a small proportion of patients with EDS- HT have haploinsufficiency of Tenascin- X, an extracellular matrix glycoprotein regulating collagen deposition. 8 Due to the lack of a definite biomarker a set o f clinical diagnostic criteria are used for diagnosing EDS- HT (Villefranche criteria 9 Table S1 and JHS (1998 Brighton criteria 10 Table S2. For the purposes of this paper, the term EDS- HT will be used to refer to JHS and EDS- HT interchangeably. EDS- HT is a multisystem disorder dysautonomia, particularly postural tachycardia syndrome (PoTS), 11, 12 fibromyalgia, 13 urinary instability 14, 15 and anxiety disorders 1 can occur in affected patients. 16 Similarly there is increasing evidence that gastrointestinal (GI) symptoms such as postprandial symptoms (67%), abdominal pain (56 62%), reflux (38 69%), nausea (44 71%), constipation (36 42%), and IBStype symptoms (30 41%) are common in these patients, occur more frequently than in other EDS subtypes, 18 and are associated with reduced quality of life (QOL). 17 Case control studies in secondary care GI clinics have demonstrated that GI symptoms such as dyspepsia and gastro- esophageal reflux disease are significantly more common in patients with previously undiagnosed EDS- HT compared to those without EDS- HT, 20 and that there is an independent association between EDS- HT and functional dyspepsia (40% prevalence of EDS- HT). 21 Moreover, within functional dyspepsia, this association appears to be strongest for postprandial distress syndrome (51% prevalence of EDS- HT), which is characterized by early satiety and postprandial fullness after regular sized meals, at least several times per week. 21 It is widely recognized that autonomic factors, opiate use, psychopathology, and chronic pain conditions (e.g. back pain, pelvic pain and fibromyalgia), which are all common in EDS- HT, 11, 16 are co- associated with the presence of GI symptoms Our previous study comparing patients in secondary care GI clinics with and without EDS- HT demonstrated that all these factors were prevalent in patients with EDS- HT and that the presence of GI symptoms was interdependent on autonomic and chronic pain measures. 20 It is thus unclear whether the GI symptoms in EDS- HT are directly associated with the underlying connective tissue disorder or rather a consequence of these other factors. Key Points Ehlers-Danlos syndrome hypermobility type (EDS-HT) is a hereditary connective tissue disorder, associated with autonomic symptoms, pain, and functional dyspepsia in secondary care gastrointestinal (GI) clinics It is unknown whether individuals with undiagnosed EDS-HT not seeking medical care also have a similar presentation. University students with previously undiagnosed EDS-HT have dyspeptic, musculoskeletal, autonomic and somatic symptoms. Individuals with underlying EDS-HT may be predisposed to developing functional dyspepsia. The relation between connective tissue and GI function requires further study. A population of relatively healthy non- patients would theoretically have fewer confounders than patients who are established in secondary and tertiary care. The primary aim of the current study was to compare the prevalence of GI and extra- intestinal symptoms in non- patient university students with and without EDS- HT to determine if these symptoms are present in individuals with EDS- HT not known to be seeking medical attention. Our secondary aims were to determine whether the presence of GI symptoms was independent of extra- intestinal factors and to examine the effect of EDS- HT on QOL. 2 MATERIALS AND METHODS 2.1 Study design, subject selection and setting A cross- sectional study of students at Queen Mary University of London (QMUL) was performed. Students were selected in two parts. For the first, all undergraduate and postgraduate students were invited via to complete the validated hypermobility screening questionnaire online. 6 This consists of 5 yes/no questions (Table 1). Answering yes to 2 or more questions is considered a positive screen with 84% sensitivity and 80% specificity for diagnosing joint hypermobility. 6 For the purposes of this study, a score of 0 was considered a negative screen and scores of 1 were excluded. Those that screened positive or negative were invited to the second part of the study if eligible (age years and sufficient understanding of written English to be able to answer the questionnaires). The second part involved a structured interview and examination for EDS- HT and fibromyalgia, TABLE 1 Validated 5- point questionnaire for generalized joint hypermobility. Answering YES to two or more of these questions has 84% sensitivity and 80% specificity for diagnosing EDS- HT 1. Can you now [or could you ever] place your hands flat on the floor without bending your knees? 2. Can you now [or could you ever] bend your thumb to touch your forearm? 3. As a child, did you amuse your friends by contorting your body into strange shapes or could you do the splits? 4. As a child or teenager, did your kneecap or shoulder dislocate on more than one occasion? 5. Do you consider yourself double-jointed? EDS- HT, Ehlers- Danlos syndrome hypermobility type.

3 Fikree et al. 3 followed by completion of a set of validated questionnaires. The study was approved by Queen Mary Research Ethics Committee, Ref: QMREC2011/42, and all participants provided written informed consent to participate. 2.2 Questionnaires Subjects completed validated questionnaires to systematically assess GI symptoms (Bowel Disease Questionnaire [BDQ] 25 ), autonomic symptoms (Composite Autonomic Symptom Scale [COMPASS] 26 ), somatic and pain symptoms (Patient Health Questionnaire 15 [PHQ- 15] 27 ), anxiety and depression (Hospital Anxiety and Depression Scale 2, 28 ) and QOL (SF ). Demographic information and medication history were collected using standardized case report forms. The BDQ provides a detailed assessment of the frequency of GI symptoms experienced during the past 6 months. 25 Symptoms were considered to be present if experienced at least 2 3 times per month or often. The BDQ also contains questions which enable diagnosis of functional GI disorders by ROME III criteria. The COMPASS provides symptom scores for several defined autonomic domains (Table 4) and enables calculation of a composite autonomic score. 26 The scores are converted into a percentage of the maximum possible score (100%) for each domain, where higher scores represent greater autonomic dysfunction and 0 represents the absence of autonomic symptoms for that domain. The PHQ- 15 consists of 15 somatic symptoms, each symptom scored from 0 ( not bothered at all ) to 2 ( bothered a lot ). Scores range from 0 to 30 with higher scores indicating a higher number of somatic symptoms. 27 The Hospital Anxiety and Depression Scale (HADS) provides anxiety and depression scores, ranging from 0 to 21, with higher scores indicating a higher severity of symptoms. 28 The SF- 36 consists of 8 scales that evaluate various aspects of QOL. Scales are scored from 0 to 100, with higher scores indicating better QOL Examination and structured interview Structured interviews and examinations to diagnose EDS- HT and fibromyalgia were performed by the lead author, who had received formal training from hypermobility specialists (R.G.). EDS- HT was diagnosed if the students satisfied either the Brighton criteria for JHS 10 or the Villefranche criteria for EDS- HT. 9 The presence of fibromyalgia was determined using the 1990 Wolfe criteria by assessing the number of positive tender points and the presence of chronic widespread pain Blinding A double blind study design was used. Assessment for EDS- HT was performed before completion of questionnaires so the investigators were blinded to GI symptom presence. Students were blinded to the exact aim of the study to reduce response bias and were not informed of their hypermobility status until after they had completed the questionnaires. 2.5 Data analysis and statistics Data were described in terms of means and confidence intervals (Gaussian ordinal data), medians and IQR (non- Gaussian ordinal data) and proportions and confidence intervals (categorical data). Comparisons of GI and associated factors in students with and without EDS- HT were performed using the t test, Mann Whitney U test, chi squared test, or Fisher s exact test, as appropriate. P<.05 was considered significant. To determine whether EDS- HT was independently associated with particular GI symptoms, stepwise multivariate regression analysis was performed. The GI symptom of interest was the dependent variable, EDS- HT was the independent variable and age, gender, anxiety, depression, autonomic symptoms, somatic symptoms, and any other factors that were significantly different in the two groups were modeled as covariates. 3 RESULTS 3.1 Subjects Of the 12,000 students who were ed the screening questionnaire, 1998 students completed it and only 1576 supplied their contact details (Fig 1). Of these, 497 had negative screens (score=0/5) and 575 had positive screens (score 2/5). 125 positive screens and 98 negative screens agreed to attend the second part of the study. 74 (59%) of these had EDS- HT (screen positive and EDS- HT positive) and 88 (90%) were confirmed not to have EDS- HT (screen negative, EDS- HT negative). Thus, 162 students were finally included in the study. EDS- HT students were younger (21.6 vs 22.8, P=.048), had a lower body mass index (BMI) (22.0 vs 23.4, P=.05), were more likely to be of White ethnicity (69% vs 52%, P=.02), and to drink alcohol (86% vs 72%, P=.02) Table 2. Students with EDS- HT were more likely to have generalized joint hypermobility (82.4% vs 8.0%, P<.001), abnormal skin (64.9 vs 36.4, P<.001), recurrent dislocations (52.7 vs 7.9, P<.001), arthralgia for more than 3 months (47.3 vs 15.9, P<.001), soft tissue problems (33.8 vs 4.6, P<.001), a Marfanoid habitus (29.7 vs 11.4, P=.003) and a positive family history (25.7 vs 1.1, P<.001) Table 2. There were no significant differences in the prevalence of other medical problems or in medication use in the two groups. No students were taking opiates. 3.2 GI symptoms in students with and without EDS- HT 41.9% of the EDS- HT students experienced at least 3 GI symptoms more than once a month, and this was significantly increased compared to the non- EDS- HT students (27.3%, P=.05). There was no significant difference in the prevalence of any of the lower GI symptoms, however, EDS- HT students were more likely to have 4 or fewer bowel motions per week (33.8% vs 20.7%, P=.06) and were more likely to have harder stools (50% vs 35.2%, P=.06) though these were not significant (Table 3). The most common upper GI symptoms in EDS- HT were abdominal pain (43%), postprandial fullness (34%), early satiety (32%), and bloating (26%). However, only postprandial fullness (34%

4 4 Fikree et al. FIGURE 1 Flowchart of subject selection. Of the 1998 students who completed the online screening questionnaire, 1576 supplied their contact details and of these 497 scored 0 (screen negative), and 575 scored 2 (screen positive). 98 of the screen negative students attended the second part of the study and 88 were confirmed not to have EDS- HT (i.e. 10 false negative). 118 of the screen positive students attended the second part of the study and 74 were confirmed to have EDS- HT (i.e. 44 false positive) vs 16%, P=.01) and early satiety (EDS- HT: 32% vs 17%, P=.03) were significantly increased when compared with non- EDS- HT students (Table 3). There was no significant difference in the proportion of EDS- HT and non- EDS- HT students who had seen their GP for GI problems (25.6% vs 26.4%, P=.9), or been to hospital for GI problems (9.5% vs 4.5%, P=.2). Students with EDS- HT were significantly more likely to satisfy criteria for functional dyspepsia (39.2% vs 22.7%, P=.02) but not IBS (12.2% vs 8.0%, P=.40). 3.3 Comparison of extra- intestinal features in students with and without EDS- HT EDS- HT students did not have increased anxiety or depression but had increased somatic symptom scores on the PHQ- 15 (7 vs 4, P=.03). EDS- HT students had a higher median number of tender points: 1 vs 0, P=.004 (Table 4). Only one student (EDS- HT) was diagnosed with fibromyalgia. Autonomic scores were higher in the EDS- HT vs non- EDS- HT students (13.5 vs 9.1, P<.001) and when these were broken up into the individual domains, differences were only seen in the orthostatic and diarrhea domain (P=.005 and P=.01 respectively) (Table 4). 3.4 Factors mediating GI symptoms EDS- HT was associated with the presence of at least 3 GI symptoms, postprandial fullness and early satiety. After adjustment for demographic factors (age, BMI, white ethnicity and alcohol use), EDS- HT remained independently associated with postprandial fullness and presence of at least 3 GI symptoms (Table 5). The addition of psychological factors to the model strengthened the model for all but early satiety, and this was due to the strong association between anxiety and the GI symptoms (P<.005). When autonomic scores were added to the model, the strength of the association between EDS- HT and the GI symptom was reduced and it lost significance, suggesting that autonomic factors were mediating or confounding the association. Further addition of somatic symptom scores to the model did not adjust the odds ratios further suggesting that somatization factors were did not confound the relationship between EDS- HT and GI symptoms (Table 5). 3.5 Quality of life in students with and without EDS- HT EDS- HT students had lower QOL scores on the pain component of the SF36 compared to non- EDS- HT students (80 vs 90, P=.03) Table 4. No differences were observed in any of the emotional, social, or other physical domains. 4 DISCUSSION This is the first cross- sectional study to investigate gastrointestinal and extra- intestinal symptoms in university students with and without EDS- HT, none of which had knowledge of their EDS- HT status at the time of the study. We demonstrated: (i) dyspeptic symptoms (early satiety and postprandial fullness) were common in the EDS- HT group, and significantly increased compared to those without EDS- HT; (ii) the EDS- HT subgroup had significantly more musculoskeletal, somatic, and autonomic symptoms; (iii) the association between EDS- HT and postprandial fullness was independent of psychopathology, somatic symptoms, and pain conditions; and (iv) that despite the increased comorbidity in the EDS- HT group, only pain- related QOL scores were

5 Fikree et al. 5 TABLE 2 Comparison of demographic and hypermobility features in students with and without the Ehlers-Danlos Syndrome-Hypermobility Type (EDS-HT) Demographics No EDS- HT (n=88) EDS- HT (n=74) P Female, % (CI) 53.4 ( ) 64.9 ( ).14 Median age (range) 21.5 (18 36) 21.0 (18 35).03 BMI (CI) 23.4 ( ) ( ).05 White ethnicity, % (CI) 52.3 ( ) 68.9 ( ).02 Smoker, % (CI) 13.6 ( ) 13.5 ( ).98 Alcohol use, % (CI) 71.6 ( ) 86.5 ( ).02 Features of Villefranche classification Median Beighton score (IQR) 1 (0 2) 5 (4 5) <.001 Skin changes, % (CI) 36.4 ( ) 64.9 ( ) <.001 Arthralgia 4 joints, % (CI) 2.3 (0 5.4) 18.9 ( ) <.001 Back pain, % (CI) 7.9 ( ) 18.9 ( ).004 Dislocation, subluxation, % (CI) 7.9 ( ) 52.7 ( ) <.001 First- degree relative with hypermobility, % (CI) 1.1 ( ) 25.7 ( ) <.001 Features of Brighton classification Spondylosis, scoliosis, % (CI) 4.6 ( ) 18.9 ( ).004 Soft tissue problems, % (CI) 4.6 ( ) 33.8 ( ) <.001 Marfanoid habitus, % (CI) 11.4 ( ) 29.7 ( ).003 Eye signs, % (CI) 29.6 ( ) 37.8 ( ).3 Varicose veins, % (CI) 1.1 (0 3.4) 4.1 (0 8.6).33 Hernias, % (CI) 4.6 ( ) 4.1 ( ) 1.0 Organ prolapses, % (CI) 0 (0 0) 0 (0 0) CI, 95% confidence interval; BMI, body mass index; EDS- HT, Ehlers- Danlos syndrome hypermobility type. worse in the EDS- HT group, with no differences seen in the emotional and social domains. The prevalence of dyspepsia in non- EDS- HT (16%), was similar to that in previous studies of university students (18%), 31 and in a metaanalysis of un- investigated dyspepsia (21%). 32 The prevalence of dyspepsia in the EDS- HT group was twice as high, therefore strikingly higher than the prevalence in other community cohorts. This concurs with high reported prevalence of dyspeptic symptoms in patients with established EDS- HT attending tertiary rheumatology genetics and gastroenterology clinics 19,20,33 and in patients with previously undiagnosed EDS- HT attending gastroenterology clinics. 20 Previous patient studies have demonstrated that the presence of GI symptoms in EDS- HT is associated with multiple somatic and autonomic symptoms. 20,34 In our non- patients we demonstrated an increased incidence of autonomic, mainly orthostatic, symptoms in the EDS- HT group which supports the growing literature on the association between EDS- HT and the PoTS 11, 12, 35 but to our knowledge this is the first description of such an association in non- patients. PHQ- 15 scores, representing somatic pain or somatization, were significantly increased in the EDS- HT group as a whole, echoing findings in French undergraduate students 2 and in patient studies. 21 Interestingly, anxiety and depression scores were not increased in EDS- HT students, which contrast with findings from longitudinal studies, which showed that there is an association between EDS- HT and anxiety even in nonpatients, and that EDS- HT is a predictor of future anxiety. 1 The differences may be due to age. In the longitudinal study, the association between EDS- HT and anxiety was only seen when the EDS- HT individuals were in their mid- thirties. The students in our study were substantially younger, (mean age 22), and anxiety may not have yet developed. Despite the increased prevalence of GI and extra- intestinal symptoms in young EDS- HT individuals, their non- pain related quality of life scores were comparable, if not better, than in the non- EDS- HT students. This contrasts with findings from several studies whereby the combination of EDS- HT and multiple extra- articular symptoms is associated with severe impairments in quality of life. 19, 20, 36 This would suggest that at an early stage individuals with EDS- HT cope relatively well, despite the presence of increased symptoms. Earlier diagnosis of EDS- HT may therefore enable interventions at a stage before numerous disabilities have set in 37 potentially preserving quality of life and functional state. The etiology of GI symptoms in EDS- HT is unknown. Some hypotheses suggest that GI symptoms are a confounding effect of coexistent pain syndromes or the use of opiates which are known to be associated with the presence of GI symptoms in patient and non- patient studies. 24, 34, 38, 39 None of the students were on opiates, and only one of the EDS- HT students had a pain syndrome (fibromyalgia). Furthermore,

6 6 Fikree et al. TABLE 3 Comparison of GI symptoms in students with and without EDS- HT. Values given are percentages of students in each group who experience the symptom at least often (for lower GI symptoms) and at least 2 3 times per month (for upper GI symptoms) No EDS- HT (n=88) EDS- HT (n=74) P Lower GI symptoms Constipation (CI) 3.4 (0 7.3) 8.1 ( ).30 Diarrhea (CI) 1.1 (0 3.4) 1.4 ( ) 1.0 Alternating bowel habit (CI) 11.5 ( ) 12.2 ( ).90 4 or fewer bowel motions per week (CI) 20.7 ( ) Hard stools (Bristol type 1 3) 35.2 ( ) 50.0 ( ).06 Upper GI symptoms Abdominal pain (CI) 34.1 ( ) 42.6 ( ).28 Retrosternal chest pain (CI) 16.1 ( ) 12.3 ( ).50 Epigastric pain (CI) 5.7 ( ) 12.2 ( ).17 Gastro- esophageal reflux (CI) 20.4 ( ) 24.3 ( ).50 Dysphagia (CI) 3.5 (0 7.5) 2.7 ( ) 1.0 Postprandial fullness (CI) 15.9 ( ) 34.4 ( ).01 Early satiety (CI) 17.0 ( ) 31.5 ( ).03 Nausea (CI) 11.4 ( ) 16.2 ( ).37 Vomiting (CI) 1.1 ( ) 2.7 ( ).59 Belching (CI) 11.4 ( ) 16.4 ( ).35 Bloating (CI) 22.7 ( ) 26.4 ( ).59 Diagnoses (ROME III) Functional dyspepsia, % (CI) 22.7 ( ) 39.2 ( ).02 Irritable bowel syndrome, % (CI) 8.0 ( ) 12.2 ( ).4 EDS- HT, Ehlers- Danlos syndrome hypermobility type. PHQ- 15 scores, which can be used as a surrogate for somatic pain, did not mediate the association between EDS- HT and GI symptoms. This would suggest that GI symptoms in our students who were not seeking medical attention were not a confounding effect of opiates or pain phenotypes. Another hypothesis proposes that an underlying dysautonomia leads to visceral and somatic hypersensitivity which then results in GI and somatic symptoms. 40 In our study autonomic factors did appear to be involved in the association between EDS- HT and GI symptoms, however, the mechanism for this remains unknown, as does the underlying etiology for the autonomic symptoms. Future physiological studies investigating autonomic and GI function, and assessing for visceral and somatic hypersensitivity will be paramount in testing these hypotheses and in bringing us a step closer to understanding the etiology of EDS- HT. One possibility is that these patients have a problem with neural processing which would explain their autonomic and GI symptoms. Evidence of small fiber neuropathy and axonal polyneuropathy has been reported in patients with EDS- HT. 41, 42 In the GI tract the location of connective tissue components in very close apposition to the myenteric plexus in the GI tract makes it ideally placed for influencing gut neuromuscular function. 43 Furthermore, recent stem cell studies in animals have demonstrated that collagen is integral to the ability of enteric neural progenitor cells to differentiate into neurons and glia, supporting the notion that connective tissue is involved in the development of the enteric nervous system. 44 Thus, there is evidence that points toward a role for connective tissue in influencing GI neuromuscular function and this deserves further study. Our study had some limitations which need to be addressed in future studies. Firstly, there was a large dropout between the screening stage and the formal assessment stage. This was not unexpected as the first stage could be completed online in less than 5 minutes whereas the second part required formal attendance at the research unit for a whole hour, so required extra effort on behalf of busy students who may hence have dropped out at this stage. The information sheet sent out to the students during the first part of the study explained that the aim was to investigate GI symptoms in students and therefore another possibility was that students with GI symptoms were encouraged to take part, thus introducing a selection and response bias. However, if this were the case, it would have applied to both the EDS- HT and non- EDS- HT group and would not have accounted for the differences seen between them. Secondly, the study was questionnaire- based and so it was subject to recall bias. However, most questions related to symptoms which were being experienced currently and in the previous few months, making this less likely. Thirdly, questionnaire- based symptoms are not always equivalent to measureable pathology which is why we used questionnaires which had previously been validated to demonstrate a high incidence of concordance between symptoms and objective measurements. 26 Fourthly, we have used a cross- sectional study design rather than a case control design. However, this is acceptable

7 Fikree et al. 7 TABLE 4 Comparison of extra- intestinal features and quality of life scores in students with and without EDS- HT No EDS- HT (n=88) EDS- HT (n=74) P Psychopathology (HADS) Anxiety score (CI) 6.49 ( ) 6.50 ( ) 1.0 Depression score (CI) 2.42 ( ) 2.51 ( ).8 Fibromyalgia % with fibromyalgia (1990) (CI) 0 (0 0) 1.3 ( ).5 Median tender points (IQR) 0 (0 1) 1 (0 4).004 Somatic symptoms PHQ- 15 (IQR) 4 (3 8.5) 7 (4 9).04 Autonomic symptoms Median autonomic score (IQR) 9.1 ( ) 13.5 ( ) <.001 Orthostatic 21.9 (0 37.5) 31.2 ( ).005 Vasomotor 0 (0 0) 0 (0 0).1 Secretomotor 7.5 (7.5 15) 7.5 ( ).3 Gastrointestinal 0 (0 0) 0 (0 16.7).1 Diarrhea 0 (0 20) 0 (0 40).01 Constipation 0 (0 14.3) 0 (0 14.3).4 Urinary 0 (0 0) 0 (0 10).2 Pupillary 0 (0 10) 0 (0 20).07 Sleep 0 (0 10) 0 (0 15).1 Syncope 0 (0 0) 0 (0 0).4 Psychosomatic O (0 0) 0 (0 0).9 Quality of life scores (SF 36 + IQR) General health 75 (55 85) 75 (60 85).8 Physical function 100 (95 100) 100 (95 100).9 Role limiting physical 100 ( ) 100 (75 100).2 Emotional well- being 76 (60 84) 72 (60 84).4 Role limiting emotional 100 ( ) 100 (33 100).8 Energy and fatigue 60 (45 75) 60 (45 75) 1.0 Pain 90 ( ) 80 (67 90).03 Social functioning 100 (75 100) 87 (75 100).2 CI, confidence interval; IQR, interquartile range; EDS- HT, Ehlers- Danlos syndrome hypermobility type. TABLE 5 Stepwise logistic regression models exploring effect of various factors on association between EDS- HT and postprandial fullness, early satiety and the presence of at least 3 GI symptoms. For each model, the presence of GI symptom is the dependent variable. The odds ratios represent the strength of the association between the EDS- HT and the GI symptom after adjusting for all other factors in that model. Confidence intervals are given in brackets Postprandial fullness Early satiety GI symptoms Model 1:EDS- HT OR unadj 2.54 ( ) 2.24 ( ) 1.92 ( ) Model 2: Demographics ORadj (age, BMI, alcohol, ethnicity) Model 3: Psychopathology ORadj (demographics, anx, depn) Model 4: Autonomic ORadj (demographics, anx, depn, autonomic) Model 5: Somatization ORadj (demographics, anx, depn, autonomic, PHQ- 15) 2.27 ( ) 1.93 ( ) 2.21 ( ) 2.91 ( ) 2.03 ( ) 2.53 ( ) 2.49 ( ) 2.12 ( ) 1.76 ( ) 2.50 ( ) 2.12 ( ) 1.70 ( ) Unadj, unadjusted; BMI, body mass index; anx, anxiety; depn, depression; autonomic, autonomic score; PHQ- 15, somatization score on PHQ- 15 questionnaire; EDS- HT, Ehlers- Danlos syndrome hypermobility type.

8 8 Fikree et al. in epidemiological studies because multivariate regression analysis can be used to adjust for potential confounders such as demographic and other factors of interest 45 and this was indeed the case in our study. Finally, our study was performed in young students who are not representative of the entire population so extrapolation of our findings to older non- patients with EDS- HT must be done with caution. There remain several knowledge gaps and future work will need to address these on an epidemiological, physiological, cellular, and molecular level. The etiology and mechanistic appraisal of dyspepsia in individuals with EDS- HT will need to be determined and this will require investigation of biomechanical, autonomic, and sensorimotor function of the upper GI tract. The understanding of which EDS- HT individuals develop dyspepsia, how and why GI and extraintestinal symptoms progress over time in these young individuals require longitudinal studies. Improving our understanding of the above mentioned factors will help us manage these patients more efficiently and may help retard the progression of symptoms. Identification of the nature of the connective tissue defect in EDS- HT and its relation to pain, GI and autonomic function is undeniably an important question now and translational studies in humans and animals will be paramount in addressing these. 5 CONCLUSION We have demonstrated that over 40% of students with EDS- HT experience GI symptoms regularly and there exists an association between EDS- HT and dyspepsia in this young and healthy population, which backs up similar associations seen in patient studies. This association is partly confounded by autonomic, but not somatic, factors, which confound the association in patient studies and this requires further investigation. Although EDS- HT individuals have increased GI, autonomic and somatic symptoms, quality of life remains good at this early stage. This highlights the importance of early identification of these individuals when they do access healthcare, as they may represent the subgroup prone to developing more severe symptoms, and therefore prompt a multidisciplinary management which may potentially preserve QOL at this stage. Furthermore, identification of a phenotype which may be predisposed to developing GI, autonomic and somatic syndromes may help patho- etiological research in functional somatic syndromes which has so far struggled to identify an underlying etiology for these chronic and debilitating disorders. ACKNOWLEDGMENTS AF designed the study, performed the research, analyzed the data, and wrote the article. RA helped perform the research and critically appraised the manuscript. JKM guided the statistical analysis and critically appraised the manuscript. RG helped design the study and provided expertise and guidance with respect to the rheumatological aspects of the study. CHK provided assistance with the statistical analysis and with the development of the manuscript. QA was the principal investigator and designed the study, oversaw the research and guided the development of the manuscript. We would like to thank Georgina Wellstead for her help with data entry and database management. FUNDING This work was supported by a grant from the Pseudo- obstruction Research Trust and Bowel and Cancer Research awarded to senior author QA. DISCLOSURES None of the authors involved in this manuscript have any conflicts of interest and have nothing to disclose. CONFLICTS OF INTEREST QA has received research funding from ONO Pharmaceutical and Protexin. QA has been on advisory board for Allergan and Grunenthal. ABBREVIATIONS BDQ, Bowel Disease Questionnaire; EDS-HT, Ehlers-Danlos syndrome hypermobility type; GI, gastrointestinal; IBS, irritable bowel syndrome; JHS, joint hypermobility syndrome; PHQ-15, Patient Health Questionnaire 15; PoTS, postural tachycardia syndrome; QOL, quality of life. REFERENCES 1. Bulbena A, Gago J, Pailhez G, Sperry L, Fullana MA, Vilarroya O. Joint hypermobility syndrome is a risk factor trait for anxiety disorders: a 15- year follow- up cohort study. Gen Hosp Psychiatry. 2011;33: Baeza-Velasco C, Gely-Nargeot MC, Vilarrasa AB, Fenetrier C, Bravo JF. Association between psychopathological factors and joint hypermobility syndrome in a group of undergraduates from a French university. Int J Psychiatry Med. 2011;41: Bravo JF, Wolff C. Clinical study of hereditary disorders of connective tissues in a Chilean population: joint hypermobility syndrome and vascular Ehlers- Danlos syndrome. Arthritis Rheum. 2006;54: Grahame R. Hypermobility: an important but often neglected area within rheumatology. Nat Clin Pract Rheum. 2008;4: Tinkle BT, Bird HA, Grahame R, Lavallee M, Levy HP, Sillence D. The lack of clinical distinction between the hypermobility type of Ehlers- Danlos syndrome and the joint hypermobility syndrome (a.k.a. hypermobility syndrome). Am J Med Genet A. 2009;149A: Hakim AJ, Grahame R. A simple questionnaire to detect hypermobility: an adjunct to the assessment of patients with diffuse musculoskeletal pain. Int J Clin Pract. 2003;57: Malfait F, Hakim AJ, De Paepe A, Grahame R. The genetic basis of the joint hypermobility syndromes. Rheumatology (Oxford). 2006;45: Burch GH, Gong Y, Liu W, et al. Tenascin- X deficiency is associated with Ehlers- Danlos syndrome. Nat Genet. 1997;17: Beighton P, De Paepe A, Steinmann B, Tsipouras P, Wenstrup RJ. Ehlers- Danlos syndromes: revised nosology, Villefranche, Ehlers- Danlos National Foundation (USA) and Ehlers- Danlos Support Group (UK). Am J Med Genet. 1998;77:31 37.

9 Fikree et al Grahame R, Bird HA, Child A. The revised (Brighton 1998) criteria for the diagnosis of benign joint hypermobility syndrome (BJHS). J Rheumatol. 2000;27: Gazit Y, Nahir AM, Grahame R, Jacob G. Dysautonomia in the joint hypermobility syndrome. Am J Med. 2003;115: Mathias CJ, Low DA, Iodice V, Owens AP, Kirbis M, Grahame R. Postural tachycardia syndrome current experience and concepts. Nat Rev Neurol. 2011;8: Kozanoglu E, Coskun Benlidayi I, Eker Akilli R, Tasal A. Is there any link between joint hypermobility and mitral valve prolapse in patients with fibromyalgia syndrome? Clin Rheumatol. 2016; 35: Arunkalaivanan AS, Morrison A, Jha S, Blann A. Prevalence of urinary and faecal incontinence among female members of the Hypermobility Syndrome Association (HMSA). J Obstet Gynaecol. 2009;29: Manning J, Korda A, Benness C, Solomon M. The association of obstructive defecation, lower urinary tract dysfunction and the benign joint hypermobility syndrome: a case- control study. Int Urogynecol J Pelvic Floor Dysfunct. 2003;14: Castori M. Ehlers- danlos syndrome, hypermobility type: an underdiagnosed hereditary connective tissue disorder with mucocutaneous, articular, and systemic manifestations. ISRN Dermatol. 2012;2012: Zeitoun JD, Lefevre JH, de Parades V, et al. Functional digestive symptoms and quality of life in patients with ehlers- danlos syndromes: results of a national cohort study on 134 patients. PLoS ONE. 2013;8:e Nelson AD, Mouchli MA, Valentin N, et al. Ehlers Danlos syndrome and gastrointestinal manifestations: a 20- year experience at Mayo Clinic. Neurogastroenterol Motil. 2015;27: Castori M, Camerota F, Celletti C, et al. Natural history and manifestations of the hypermobility type Ehlers- Danlos syndrome: a pilot study on 21 patients. Am J Med Genet A. 2010;152A: Fikree A, Grahame R, Aktar R, et al. A prospective evaluation of undiagnosed joint hypermobility syndrome in patients with gastrointestinal symptoms. Clin Gastroenterol Hepatol. 2014; 12: Fikree A, Aktar R, Grahame R, et al. Functional gastrointestinal disorders are associated with the joint hypermobility syndrome in secondary care: a case- control study. Neurogastroenterol Motil. 2015;27: Farmer AD, Aziz Q. Visceral pain hypersensitivity in functional gastrointestinal disorders. Br Med Bull. 2009;91: Van Oudenhove L, Vandenberghe J, Vos R, Holvoet L, Tack J. Factors associated with co- morbid irritable bowel syndrome and chronic fatigue- like symptoms in functional dyspepsia. Neurogastroenterol Motil. 2011;23:524 e Whorwell PJ, McCallum M, Creed FH, Roberts CT. Non- colonic features of irritable bowel syndrome. Gut. 1986;27: Talley NJ, Phillips SF, Wiltgen CM, Zinsmeister AR, Melton LJ 3rd. Assessment of functional gastrointestinal disease: the bowel disease questionnaire. Mayo Clin Proc. 1990;65: Suarez GA, Opfer-Gehrking TL, Offord KP, Atkinson EJ, O Brien PC, Low PA. The Autonomic Symptom Profile: a new instrument to assess autonomic symptoms. Neurology. 1999;52: Kroenke K, Spitzer RL, Williams JB. The PHQ- 15: validity of a new measure for evaluating the severity of somatic symptoms. Psychosom Med. 2002;64: Zigmond AS, Snaith RP. The hospital anxiety and depression scale. Acta Psychiatr Scand. 1983;67: McHorney CA, Ware JE Jr, Raczek AE. The MOS 36- Item Short- Form Health Survey (SF- 36): II. Psychometric and clinical tests of validity in measuring physical and mental health constructs. Med Care. 1993;31: Wolfe F, Smythe HA, Yunus MB, et al. The American College of Rheumatology 1990 criteria for the classification of fibromyalgia. Report of the multicenter criteria committee. Arthritis Rheum. 1990;33: Basandra S, Bajaj D. Epidemiology of dyspepsia and irritable bowel syndrome (IBS) in medical students of Northern India. J Clin Diagn Res. 2014;8: JC13 JC Ford AC, Marwaha A, Sood R, Moayyedi P. Global prevalence of, and risk factors for, uninvestigated dyspepsia: a meta- analysis. Gut. 2015; 64: Hakim AJ, Grahame R. Non- musculoskeletal symptoms in joint hypermobility syndrome. Indirect evidence for autonomic dysfunction? Rheumatology (Oxford). 2004; 43: De Wandele I, Rombaut L, Malfait F, De Backer T, De Paepe A, Calders P. Clinical heterogeneity in patients with the hypermobility type of Ehlers- Danlos Syndrome. Res Dev Disabil. 2013;34: Seligman WH, Low DA, Asahina M, Mathias CJ. Abnormal gastric myoelectrical activity in postural tachycardia syndrome. Clin Auton Res. 2013;23: Rombaut L, Malfait F, Cools A, De Paepe A, Calders P. Musculoskeletal complaints, physical activity and health- related quality of life among patients with the Ehlers- Danlos syndrome hypermobility type. Disabil Rehabil. 2010;32: Adib N, Davies K, Grahame R, Woo P, Murray KJ. Joint hypermobility syndrome in childhood. A not so benign multisystem disorder? Rheumatology. 2005;44: Almansa C, Rey E, Sanchez RG, Sanchez AA, Diaz-Rubio M. Prevalence of functional gastrointestinal disorders in patients with fibromyalgia and the role of psychologic distress. Clin Gastroenterol Hepatol. 2009;7: Van Hemert S, Breedveld AC, Rovers JMP, et al. Migraine associated with gastrointestinal disorders: review of the literature and clinical implications. Front Neurol. 2014;5: Castori M, Celletti C, Camerota F. Ehlers- Danlos syndrome hypermobility type: a possible unifying concept for various functional somatic syndromes. Rheumatol Int. 2013;33: Pascarella A, Provitera V, Lullo F, et al. Evidence of small fiber neuropathy in a patient with Ehlers- Danlos syndrome, hypermobility- type. Clin Neurophysiol. 2016; 127: Voermans NC, van Alfen N, Pillen S, et al. Neuromuscular involvement in various types of Ehlers- Danlos syndrome. Ann Neurol. 2009;65: Gregersen H. Biomechanics of the Gastrointestinal Tract, New Perspectives in Motiity Research and Diagnostics. London: Springer; Raghavan S, Gilmont RR, Bitar KN. Neuroglial differentiation of adult enteric neuronal progenitor cells as a function of extracellular matrix composition. Biomaterials. 2013;34: Liang K-Y, Scott ZL, Qaqish B. Multivariate regression analyses for categorical data. J R Stat Soc. Ser B (Methodol). 1992;54:3 40. SUPPORTING INFORMATION Additional Supporting Information may be found online in the supporting information tab for this article.