Can a Vegetarian Diet Relieve IBS? – a Systematic Literature Review

Örebro University School of Medicine Degree project, 15 ECTS January 2018

Can a vegetarian diet relieve IBS? – A systematic literature review

Version 1

Author: Camilla Rosén Supervisor: Julia Sabet, PhD Örebro, Sweden Abstract Background IBS (Irritable Bowel Syndrome) is a functional disorder with a prevalence of approximately 10-15%. Studies show that part of the etiology is likely diet-related. Research has proven a correlation between meat consumption and intestinal diseases. Purpose The aim is to explore possible correlations between meat consumption and IBS, and the effects exclusion of meat from the diet could have on IBS symptoms. By reviewing previous study findings about meat and IBS, the objective is to strengthen or reject the hypothesis that symptoms of some IBS patients could be relieved by a transition to a vegetarian diet. Method The type of study is a systematic literature review. The electronic database PubMed was used to search for published studies that presented data on meat consumption or vegetarianism in relation to IBS. Results Six studies on the topic were found. Two were case-control studies and four were cross- sectional. The definition of ‘vegetarian’ varied extensively. All studies found an association between vegetarianism or meat consumption and IBS. The results however were contradictive and inconclusive. Two studies found a positive correlation between a consistent vegetarian diet and IBS prevalence. One found a negative correlation between a ‘lacto-vegetarian’ diet pattern and IBS prevalence. One study found that IBS patients had a higher intake of processed meat and beef than healthy controls, one that IBS patients had a higher intake of processed meat and sandwich meat and one that IBS patients had a lower meat intake but higher protein intake than the general population. Conclusion There are studies that do indicate a possibility that symptoms of some IBS patients could be eased by changing to a vegetarian diet. However, the number of studies on the topic is small and their findings are inconclusive. Also, changing to a vegetarian diet often means increasing the intake of foods that can worsen IBS symptoms. More studies are necessary to evaluate the positive or negative effect a vegetarian diet could have on IBS symptoms.

2 Abbreviations

IBS – Irritable Bowel Syndrome IBS-C – Irritable bowel syndrome with Constipation IBS-D – Irritable bowel syndrome with Diarrhea IBS-M – Irritable bowel syndrome, Mixed FGIDs – Functional gastrointestinal disorders FODMAPs – Fermentable Oligo-, Di-, Monosaccharides and Polyols OR – Odds ratio CI – Confidence Interval

3 Table of contents

1 Introduction ...... 5 1.1 IBS definition and prevalence ...... 5 1.2 Diet part of the etiology ...... 5 1.3 Correlation between meat and intestinal disorders ...... 5 1.4 Objective ...... 6

2 Method ...... 6 2.1 Design ...... 6 2.2 Search ...... 6 2.3 Inclusion and exclusion criteria ...... 6 2.4 Quality rating...... 7 2.5 Ethical considerations ...... 7

3 Results ...... 8 3.1 Selection process ...... 8 3.2 Study populations and design ...... 9 3.3 Definitions ...... 10 3.4 Meat consumption...... 11 3.5 Study results, findings and conclusions ...... 13

4 Discussion ...... 15

5 Conclusion ...... 18

6 References ...... 19

7 Attachments ...... 21 7.1 Compiled table ...... 21

4 1 Introduction

1.1 IBS definition and prevalence IBS (Irritable Bowel Syndrome) is a functional disorder defined by abdominal pain and an altered transit pattern (diarrhea/ constipation/ mixed), without proven organic changes [1]. It has a prevalence of about 10-15% [2]. Therefore, a large part of the population could be helped by better treatments for IBS. The biggest obstacle for finding new treatments is that the etiology of IBS is mostly unknown.

1.2 Diet part of the etiology Studies show that part of the etiology is likely diet-related, that diet plays an important role in managing the disease and that foods hard to digest may be a source of the problem [3,4]. One type of food that can be hard to digest is short-chain poorly absorbed carbohydrates known as FODMAPs (Fermentable Oligo-, Di-, Mono-saccharides and Polyols), which are present in for example onions, wheat, fruits and milk [3]. Research has shown that FODMAPs can worsen IBS symptoms and a low FODMAPs diet is often recommended to IBS patients [3]. Other food components that can be hard to digest are gluten and milk protein [5,6]. Non-celiac gluten sensitivity and milk protein allergy sometimes give similar symptoms as IBS [1,7,8] and recommendations for IBS patients to exclude gluten or milk can also be found [6]. However, none of these recommendations benefit all IBS patients [9].

1.3 Correlation between meat and intestinal disorders Another type of food that may be hard to digest is meat [10] and research has proven a correlation between meat consumption and intestinal diseases. For example, high consumption of red meat is a risk factor for development of colon cancer [11]. Red meat is considered to be potentially proinflammatory [12] and can therefore negatively affect the intestinal health and the barrier function of the bowel. Studies show that IBS patients often display a low-grade systemic inflammation [13]. We aimed to investigate whether exclusion of meat from the diet could benefit some IBS patients.

5 1.4 Objective The aim of this review is to explore possible correlations between IBS and meat consumption and the effects a meat free – vegetarian – diet could have on IBS symptoms. By reviewing previous study findings about meat and IBS, the objective is to strengthen or reject the hypothesis that symptoms of some IBS patients could be relieved by a transition to a vegetarian diet.

2 Method

2.1 Design The type of study is a systematic literature review.

2.2 Search The electronic database PubMed was used to search for published studies on Nov 29, 2017. A first search for studies was made using the search terms ”IBS AND Vegetarian”. Since the first search only resulted in three studies, a second, wider search was made using the search terms ("Irritable Bowel Syndrome"[Mesh] OR IBS OR irritable bowel syndrome OR irritable bowel OR colon irritable) AND ("Diet, Vegetarian"[Mesh] OR vegetarian OR vegetarianism OR herbivore OR omnivore OR meat OR meats OR beef OR animal-derived).

2.3 Inclusion and exclusion criteria The search had no geographical limitations or publishing date restrictions and all studies meeting the search criteria were included in the first round. In the second and third round, studies that did not present any data on meat consumption or vegetarianism in relation to IBS were excluded.

6 2.4 Quality rating A quality rating of the studies included was conducted, using the template ”Mall för granskning av observationsstudier”, a modified template designed for Örebro University. The rating scores can be seen in Table 1.

Table 1. Quality rating of included studies

Authors Article Quality rating

Ghoshal and Singh Frequency and risk of functional gastro-intestinal disorders Medium [14] in a rural Indian population Khayyatzadeh et al. Dietary patterns and prevalence of irritable bowel Medium [15] syndrome in Iranian adults

Buscail et al. [16] Association between self-reported vegetarian diet and the High irritable bowel syndrome in the French NutriNet cohort

Chirila et al. [17] Diet and Irritable Bowel Syndrome Medium

Tigchelaar et al. [18] Habitual diet and diet quality in Irritable Bowel Syndrome: Medium A case-control study

Böhn et al. [19] Nutrient intake in patients with irritable bowel syndrome Low compared with the general population

2.5 Ethical considerations All publications included in the scope of this literature review will be accounted for and all results presented, both those supporting and those not supporting the set hypothesis.

7 3 Results

3.1 Selection process Round 1: Search 1 provided three observational cohort studies [14–16]. Search 2 provided another 47 studies.

Round 2: The abstracts were read to assess which articles discussed meat as a diet component in relation to IBS and a total of 13 articles were identified as candidates for the review.

Round 3: After reading the full articles to determine which studies presented data about meat consumption or referred to vegetarianism in relation to IBS, a total of six studies were finally included (Fig. 1). Table 5 contains an overview of the included studies.

Figure 1. The selection process leading to six studies being included.

8 3.2 Study populations and design Two studies are case-control studies where identified IBS patients are compared to controls regarding diet habits [18,19]. The other four are cross-sectional studies measuring differences in diet habits and IBS expression within the sample group.

The number of participants span from n=193 to n=41682. In five out of six studies, the study participants are predominantly women.

Table 2. Study populations and design of included studies

Women Presenting Authors Method Study population Scope N (%) IBS, N (%) Ghoshal and Cross-sectional study. Adults from three villages Evaluation of 2774 43,3% 190 (6,8%) Singh [14] House-to house survey in northern India. prevalence and risk by interviewers using a factors of FGIDs in a translated-validated rural Indian Rome III questionnaire community. and a hospital anxiety and depression questionnaire. Khayyatzadeh Cross-sectional study Iranian adults working in Evaluation of the 3846 55,5% 828 (21,5%) et al. [15] using a 106-item self- 50 different health prevalence of IBS administered Dish- centers. among Iranian based Semi-quantitative adults depending on Food frequency diet pattern. Also questionnaire designed comparison of the for Iranian adults, and a IBS frequency modified Persian among the subjects version of the Rome III in the top quartile of questionnaire. each diet pattern to those in the lower quartiles. Buscail et al. Cross-sectional study Adult subjects Evaluation of the 41682 78,0 % 2264 (5,4%) [16] using a questionnaire participating in the association between on IBS based on Rome NutriNet-Santé cohort IBS and a vegetarian III criteria. Lifestyle data study. Participants who diet among a large including vegetarian reported organic diseases French cohort. diet was collected or alarm symptoms were previously by self- excluded. administered questionnaires at baseline and during follow-ups every six months since the start of the NutriNet-Santé

9 cohort study in 2009. Chirila et al. Cross-sectional study. A random sample of adult Assessment of IBS 193 58,5% 37 (19,1%) [17] Interview by family subjects from a general prevalence and doctors using a food urban population association of diet frequency selected from family with IBS symptoms. questionnaire and a doctors' patient lists. Rome III questionnaire.

Tigchelaar et Case-control study. IBS IBS patients participating Assessment of 380 IBS 194 (51,1%) al. [18] diagnosis and subtypes in the Maastricht IBS habitual dietary patients defined by Rome III cohort study, and healthy intake in IBS 74,7%, criteria. Self- controls from the same patients, with Controls administered food geographical region special emphasis on 62,9% frequency recruited via public the overall diet questionnaire of intake advertisements. quality. over the previous month. Böhn et al. Case-control study. Self- IBS patients participating Assessment of 561 74,3% 187 (33,3%) [19] administered 4-days in treatment trials and an nutrient intake in food registration ongoing prospective IBS patients record. study on compared to the pathophysiological general population, factors in IBS at the nutritional Sahlgrenska University differences between Hospital, and an age- and IBS subgroups, and gender-matched control whether nutrient group from a nation-wide intake met dietary survey, recommendations. Riksmaten.

3.3 Definitions A vegetarian diet/ diet pattern is considered in four studies [14–16,18]. The definition of vegetarian varies extensively between them. In the Indian study, Ghoshal and Singh [14] define vegetarian as eating no food of animal origin except milk. In the French study, Buscail et al [16] define a vegetarian diet as partial or total removal of meat, poultry and fish. Participants reporting at least once that they followed a vegetarian diet are considered vegetarian and participants who in at least three follow-up questionnaires reported a vegetarian diet are considered “stable” or “consistent” vegetarians. In the Iranian study, Khayyatzadeh et al [15] use the term ’lacto-vegetarian diet pattern’, where the subjects have a high intake of non-flatulent vegetables, tomato, citrus fruits, flatulent vegetables, fruits and low-fat dairy products. The term does not expressly involve exclusion of meat. Tigchelaar et al [18] do not define the term vegetarian in the

10 Dutch study, but state that a vegetarian diet was reported by a number of subjects. The other two studies in this review [17,19] include an association between IBS and meat consumption but do not specifically evaluate a vegetarian diet and therefore do not define the term.

In one study [18] the IBS group was collected from a cohort of IBS patients where IBS diagnosis and subtypes had already been defined, by Rome III criteria. The other five studies have used the Rome III criteria questionnaire or modified and validated versions of it to establish IBS prevalence. The criteria for different subtypes of IBS vary slightly between the studies.

3.4 Meat consumption Out of the six reviewed studies, three provided data on the prevalence of a vegetarian diet among the study participants. In the study from India, 95,7% of the subjects were vegetarian [14]. In the French study, 1,9% of the IBS patients and 2,0% of the controls reported a vegetarian diet [16]. In the Dutch study, 2,1% of the IBS patients and 6,5% of the controls reported a vegetarian diet [18].

The meat consumption in grams/day was specified in three studies. The Iranian study [15] reported the consumption of red meat to be higher among IBS patients than controls. It also stated that those in the top quartile of ‘lacto-vegetarian’ diet pattern ate less red meat than those in the bottom quartile. The French study [16] stated that the consumption of all types of meat was lower in the vegetarian group than the control group and even lower among the stable vegetarians. In the Dutch study [18] it was reported that IBS patients eat more processed meat and somewhat more poultry than controls. There were no differences regarding the consumption of non-processed meat and fish. The Romanian study [17] did not specify meat consumption in grams/day but reported the odds ratio to be 4,75 for consumption of processed meat by IBS patients compared to controls.

11 Table 3. Meat consumption of the study participants

Authors Country N Vegetarian (%) Meat consumption

Ghoshal and Singh India 2774 95,7% • Not specified

Khayyatzadeh et al. Iran 3846 Not specified • Red meat: IBS patients 72,3 g/day (Control group 71,5 g/day). Q1 'lacto-vegetarian' 69 g/day (Q4 75 g/day). Buscail et al. France 41682 IBS patients: 1,9%, • Meat, poultry: Controls: 2,0% omnivores 97,0 g/day, vegetarians 37,1 g/day, stable vegetarians 14,2 g/day. • Porc ham, poultry cuts, processed meat: omnivores 43,4 g/day, vegetarians 17,0 g/day, stable vegetarians 6,5 g/day. • Fish, shellfish, processed fish and shellfish: omnivores 62,9 g/day, vegetarians 44,6 g/day, stable vegetarians 39,5 g/day. Chirila et al. Romania 193 Not specified • Processed meat: OR 4,75 95% CI: 1,60-14,09 (IBS patients vs controls) Tigchelaar et al. The 380 IBS patients: 2,1%, • Processed meat: Netherlands Controls: 6,5% IBS patients 38,3 g/day (control group 29,4 g/day). • Meat: IBS patients 48,1 g/day (control group 48,1 g/day). • Poultry: IBS patients 11,1 g/day (control group 9,7 g/day). • Fish: IBS patients 15,7 g/day (control group 15,8 g/day). Böhn et al. Sweden 561 Not specified • Not specified

12 3.5 Study results, findings and conclusions Two studies found a significant positive correlation between IBS prevalence and a consistent vegetarian diet [14,16]. One study found a significant negative correlation between a ‘lacto-vegetarian’ diet pattern and IBS prevalence [15].

Two studies found that IBS patients have a significantly higher intake of processed meat than healthy controls [17,18]. One of them also saw a significantly higher intake of beef [17] and the other a significantly higher intake of sandwich meat [18] in IBS patients. One study reported IBS patients to have a tendency toward lower intake of meat compared to the general population, but the energy proportion from protein intake was higher in the IBS population compared to the general population [19].

Table 4. Key findings of included studies

Authors Relevant key findings Results and conclusions

Ghoshal and IBS prevalence vegetarians: 7,1% • Vegetarianism was significantly positively Singh (non-vegetarians 0,8%), correlated to the prevalence of IBS. aOR 10,77 95% CI: 1,49-77,89 Khayyatzadeh IBS prevalence lacto-vegetarian • Study subjects with a 'lacto-vegetarian' diet et al. Q1 vs Q4: pattern are less likely to have IBS than subjects aOR 0,76 95% CI: 0,59-0,98 with 'fast food', 'traditional' or 'western' dietary patterns. • A 'lacto-vegetarian' dietary pattern was strongly associated with a reduced risk for IBS. • Those in the top quartile of the 'lacto- vegetarian' diet pattern were 24% less likely to have IBS and 46% less likely to have IBS-C than those in the lowest quartile. Buscail et al. IBS prevalence stable • A stable vegetarian diet was significantly vegetarians vs controls: positively associated with IBS. aOR: 2,60 95% CI: 1,37-4,91 • Vegetarians had a significantly lower consumption of meat (including poultry, fish and shellfish) and also a lower consumption of soft sugary drinks and alcoholic beverage compared to omnivores. They had a significantly higher consumption of eggs, fruits and vegetables, wholegrain products and pulses. • Vegetarians reported lower total energy intake, with higher percent energy from carbohydrates. • Vegetarians had the highest intake of fibres and simple carbohydrates and the lowest intake of animal proteins and cholesterol. • Vegetarians had a lower intake of lactose.

13 Chirila et al. Intake of processed meat, • IBS patients ate significantly more frequently IBS patients vs controls: OR 4,75 processed meat and beef. 95% CI: 1,60-14,09. Intake of beef: not specified. Tigchelaar et Intake of processed meat, • IBS patients had a significantly higher al. IBS patients vs controls: 38 g/day consumption of processed meat and sandwich vs 29 g/day, p<0,001. meat. Intake of sandwich meat: not specified. Böhn et al. Intake of meat: not specified. • IBS patients tended to have a lower intake of %E protein IBS patients vs meat and dairy products. controls: 17% vs 16%, p=0,007. • The energy proportion from protein intake was higher in the IBS population compared to the general population.

14 4 Discussion Two of the studies in this review found a positive association and one found a negative association between IBS and a vegetarian diet. Two studies found that IBS patients ate more meat than controls and one that they ate less meat but more protein.

All studies found an association between vegetarianism or meat consumption and IBS. The results however were contradictive and inconclusive. It also seems as if the same conclusion can be drawn regardless of what the results show – if IBS patients are found to eat more meat than controls it might be that meat is what causes the IBS, and if they are found to eat less meat than controls it might be in order to relieve their symptoms. It might also be the other way around. What can be noted is that single vegetarian meals or semi-vegetarianism do not seem to have effect on IBS symptoms. However there seems to be a correlation between IBS and a stable vegetarian diet.

The definition of ”vegetarian” varied considerably between the different studies, complicating comparisons and conclusions. The ‘lacto-vegetarian’ diet pattern in the Iranian study was a diet containing lots of vegetarian food but it was not a meat exclusion diet and the subjects associated with this pattern did eat meat as well. The French study included partial removal of meat in the definition of vegetarian, and even the ‘consistent’ vegetarians reported some meat intake. In the Indian study it meant exclusion of meat and even included the exclusion of other animal products such as eggs. These findings suggest that the definitions of a vegetarian diet may differ between countries. Also, new terms such as flexitarianism and semi-vegetarianism have emerged, where subjects follow a diet that is primarily vegetarian with occasional inclusion of meat or fish [20]. This may cause subjects to either define themselves as vegetarians or not, depending on whether the term ‘vegetarian’ is poorly or well defined in a questionnaire. Different views on vegetarianism in different countries and among subjects within the studies may cause bias and consequently lead to false conclusions.

The studies in this review were all based on self-reported diet, either by self- administered questionnaires or by interviews. Self-reported diet is prone to recall bias and therefore may not be entirely accurate. Out of the six reviewed studies, the French study is the largest. It is the most recently published and received the highest quality

15 rating out of the studies in this review. It was also the only study in this review that addressed the question of whether a vegetarian diet could be considered a treatment option for IBS. In the Swedish study the control group was selected from a nation-wide dietary survey conducted about 15 years earlier and there is a chance that dietary habits of the general population have changed since then. The control group consisted of the general population and therefore included also IBS patients. The claimed results were not all accounted for and the study received the lowest quality rating.

Causality or etiology of IBS was not addressed in any of the studies included. The French study [16] found that a sporadic vegetarian diet was not associated with IBS, but a consistent vegetarian diet was. However, it could not conclude on causality, to determine whether a consistent vegetarian diet increases IBS symptoms or if IBS patients tend to adopt a vegetarian diet in order to relieve their IBS symptoms. Both the Iranian and the Romanian study [15,17] point out that cross-sectional studies may only reveal association and not causality. Since causality cannot be determined with the type of design that the studies in this review had, no conclusions on causality can be drawn. To our knowledge there is no study that addresses causality of IBS in relation to vegetarianism, and there have also been no intervention studies to test whether a vegetarian diet can ease IBS symptoms.

Lifestyle, habits and what people eat differs between countries. In the Indian study a vast majority of the subjects were vegetarian, quite contrary to all other studies. Out of the 2774 subjects in the study only 120 were non-vegetarian, and out of the 120 only one (0,8%) had IBS. This resulted in a statistically significant positive correlation between vegetarianism and IBS. However, the prevalence of IBS among the vegetarians was only 7,1%, which is very low given that the overall prevalence of IBS in most studies is estimated to 10-15% [2]. Low occurrence of IBS has been reported previously in India with a prevalence of around 4% [21,22]. Ghoshal et al [14] relates the low IBS prevalence in India to a number of factors including dietary factors such as vegetarianism.

Exclusion of meat from the diet likely leads to an increase in intake of other foods. The French study [16] showed that vegetarians had a higher intake of both simple and

16 complex carbohydrates. An increase of poorly absorbed carbohydrates (FODMAPs) can worsen IBS symptoms since these are fermented in the small intestine, leading to gas production and distension of the intestinal lumen. The Indian study [14] also notes that a vegetarian diet can be deficient in vitamin B12, which might lead to diarrhea. This means that even if exclusion of meat would have a positive effect for IBS patients, there is a risk that a diet change would lead to a higher intake of foods that would have the opposite effect, and the cumulative effect could still be negative. Future studies could address this issue using controlled diet interventions for example.

IBS is a diagnosis in which no organic changes or other explanations to the symptoms are found. All of the diagnostic criteria do not have to be fulfilled to get a diagnosis (two or more out of three criteria are sufficient) and the subtypes of IBS are considerably different from each other [1]. At the same time, the symptoms of IBS and of various types of protein sensitivities are often similar. For example, the symptoms of milk protein allergy include diarrhea, colic and/ or constipation [7], while the Rome III diagnostic criteria for IBS include a change in frequency of stool and/ or a change in form or appearance of stool [1]. With protein sensitivities, antibodies can be detected sometimes but not always [23,24]. When no antibodies can be detected, there might be a risk that food sensitivities are missed.

Considering the reported improvements of IBS symptoms in some patients when excluding for example milk protein or gluten there is a possibility that patients get misdiagnosed with IBS when they are in fact sensitive to a protein. Zar et al [25] have shown that IgG4 antibodies to beef, pork, milk, wheat, eggs and lamb are commonly elevated in IBS patients and suggest that exclusion diets based on IgG4 titers may have effect for IBS patients. Similar results have been shown by Hong Guo et al [26], where 35 patients with IBS-D had a significant symptom reduction after following a 12-week elimination diet based on IgG serology. Carroccio et al [27] have also shown that IBS patients often are hypersensitive to cow’s milk protein or gluten and can get reduction or complete elimination of symptoms by removal of offending foods. Since IBS symptoms are often similar to those of protein sensitivities, there may be a meat protein sensitivity that has not yet been discovered/ described in published studies. It may also be that a vegetarian diet is beneficial only for IBS patients who have animal protein

17 sensitivities or other sensitivities to animal products. Future studies should investigate this.

5 Conclusion The results from this review show that there are studies that do indicate a possibility that symptoms of some IBS patients could be eased by changing to a vegetarian diet. However, the number of studies on the topic is small and their findings are inconclusive. Also, changing to a vegetarian diet often means increasing the intake of FODMAPs, which can worsen the IBS symptoms. More studies are necessary to evaluate the positive or negative effect a vegetarian diet could have on IBS symptoms. Examples of adequate studies could be intervention studies, either where IBS patients exclude meat from their diet or where digestive enzymes are added. An intervention could show whether a vegetarian diet would help those who are already diagnosed with IBS. To establish causality, prospective cohort studies are needed, where diet is assessed before the onset of IBS. Questionnaires to vegetarian IBS patients on the reason why they follow this diet could also be a useful supplement, to learn if they have IBS due to their vegetarianism or if they have chosen a vegetarian diet to relieve their IBS.

18 6 References

1. Shih DQ, Kwan LY. All Roads Lead to Rome: Update on Rome III Criteria and New Treatment Options. Gastroenterol Rep. 2007;1(2):56–65. 2. Lovell RM, Ford AC. Global Prevalence of and Risk Factors for Irritable Bowel Syndrome: A Meta-analysis. Clin Gastroenterol Hepatol. 01 juli 2012;10(7):712– 721.e4. 3. Gibson PR, Barrett JS, Muir JG. Functional bowel symptoms and diet. Intern Med J. 01 oktober 2013;43(10):1067–74. 4. Simrén M, Månsson A, Langkilde AM, Svedlund J, Abrahamsson H, Bengtsson U, m.fl. Food-related gastrointestinal symptoms in the irritable bowel syndrome. Digestion. 2001;63(2):108–15. 5. Fardet A. Wheat-based foods and non celiac gluten/wheat sensitivity: Is drastic processing the main key issue? Med Hypotheses. december 2015;85(6):934–9. 6. Cozma-Petru A, Loghin F, Miere D, Dumitra cu DL. Diet in irritable bowel syndrome: What to recommend, not what to forbid to patients! World J Gastroenterol.ţ 07 juni 2017;23(21):3771–83.ş 7. Brill H. Approach to milk protein allergy in infants. Can Fam Physician. september 2008;54(9):1258–64. 8. Catassi C, Alaedini A, Bojarski C, Bonaz B, Bouma G, Carroccio A, m.fl. The Overlapping Area of Non-Celiac Gluten Sensitivity (NCGS) and Wheat-Sensitive Irritable Bowel Syndrome (IBS): An Update. Nutrients. 21 november 2017;9(11):1268. 9. Peta Hill, Jane G. Muir, Peter R. Gibson. Controversies and Recent Developments of the Low-FODMAP Diet. Gastroenterol Hepatol. januari 2017;13(1):36–45. 10. Bioactive Peptides in Animal Food Products [Internet]. [citerad 29 december 2017]. Tillgänglig vid: https://www-ncbi-nlm-nih- gov.db.ub.oru.se/pmc/articles/PMC5447911/ 11. Johnson CM, Wei C, Ensor JE, Smolenski DJ, Amos CI, Levin B, m.fl. Meta-analyses of colorectal cancer risk factors. Cancer Causes Control CCC. juni 2013;24(6):1207–22. 12. Esposito K, Maiorino MI, Bellastella G, Panagiotakos DB, Giugliano D. Mediterranean diet for type 2 diabetes: cardiometabolic benefits. Endocrine. 01 april 2017;56(1):27–32. 13. Sinagra E, Morreale GC, Mohammadian G, Fusco G, Guarnotta V, Tomasello G, m.fl. New therapeutic perspectives in irritable bowel syndrome: Targeting low-grade inflammation, immuno-neuroendocrine axis, motility, secretion and beyond. World J Gastroenterol. 28 september 2017;23(36):6593–627. 14. Ghoshal UC, Singh R. Frequency and risk factors of functional gastro-intestinal disorders in a rural Indian population. J Gastroenterol Hepatol. 01 februari 2017;32(2):378–87. 15. Khayyatzadeh SS, Esmaillzadeh A, Saneei P, Keshteli AH, Adibi P. Dietary patterns and prevalence of irritable bowel syndrome in Iranian adults. Neurogastroenterol Motil. 01 december 2016;28(12):1921–33.

19 16. Buscail C, Sabate J-M, Bouchoucha M, Torres MJ, Allès B, Hercberg S, m.fl. Association between self-reported vegetarian diet and the irritable bowel syndrome in the French NutriNet cohort. PLOS ONE. augusti 2017;12(8):e0183039. 17. Chirila I, Petrariu FD, Ciortescu I, Mihai C, Drug VL. Diet and irritable bowel syndrome. J Gastrointest Liver Dis JGLD. december 2012;21(4):357–62. 18. Tigchelaar EF, Mujagic Z, Zhernakova A, Hesselink M a. M, Meijboom S, Perenboom CWM, m.fl. Habitual diet and diet quality in Irritable Bowel Syndrome: A case- control study. Neurogastroenterol Motil. 01 december 2017;29(12):n/a-n/a. 19. Böhn L, Störsrud S, Simrén M. Nutrient intake in patients with irritable bowel syndrome compared with the general population. Neurogastroenterol Motil. 01 januari 2013;25(1):23-e1. 20. Derbyshire EJ. Flexitarian Diets and Health: A Review of the Evidence-Based Literature. Front Nutr [Internet]. 06 januari 2017 [citerad 28 december 2017];3. Tillgänglig vid: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5216044/ 21. Oshima T, Miwa H. Epidemiology of Functional Gastrointestinal Disorders in Japan and in the World. J Neurogastroenterol Motil. 31 juli 2015;21(3):320–9. 22. Ghoshal UC, Abraham P, Bhatt C, Choudhuri G, Bhatia SJ, Shenoy KT, m.fl. Epidemiological and clinical profile of irritable bowel syndrome in India: report of the Indian Society of Gastroenterology Task Force. Indian J Gastroenterol Off J Indian Soc Gastroenterol. februari 2008;27(1):22–8. 23. Anthoni S, Elg P, Haahtela T, Kolho K-L. Should milk-specific IgE antibodies be measured in adults in primary care? Scand J Prim Health Care. 2008;26(4):197–202. 24. Anthoni S, Savilahti E, Rautelin H, Kolho K-L. Milk protein IgG and IgA: The association with milk-induced gastrointestinal symptoms in adults. World J Gastroenterol WJG. 21 oktober 2009;15(39):4915–8. 25. Zar S, Mincher L, Benson MJ, Kumar D. Food-specific IgG4 antibody-guided exclusion diet improves symptoms and rectal compliance in irritable bowel syndrome. Scand J Gastroenterol. 01 januari 2005;40(7):800–7. 26. Hong Guo, Tao Jiang, Jinliang Wang, Yongchao Chang, Hai Guo, Weihong Zhang. The Value of Eliminating Foods According to Food-Specific Immunoglobulin G Antibodies in Irritable Bowel Syndrome with Diarrhoea. J Int Med Res. 01 februari 2012;40(1):204–10. 27. Carroccio A, Brusca I, Mansueto P, Soresi M, D’Alcamo A, Ambrosiano G, m.fl. Fecal Assays Detect Hypersensitivity to Cow’s Milk Protein and Gluten in Adults With Irritable Bowel Syndrome. Clin Gastroenterol Hepatol. 01 november 2011;9(11):965–971.e3.

20 7 Attachments

7.1 Compiled table

Table 5. Overview of included studies

Study Quality Women Presenting Vegetarian Meat IBS prev. Key findings, results and Authors Article Country Method Scope N population rating (%) IBS, N (%) (%) consumption vegetarian conclusions

Ghoshal and Frequency India Cross-sectional Adults from Evaluation of Medium 2774 43,3% 190 (6,8%) 95,7% Not specified 7,1% Vegetarianism, tobacco Singh and risk of study. House-to three villages in prevalence (controls chewing, aerated soft functional house survey by northern India. and risk 0,8%). aOR drinks, tea/ coffee, gastro- interviewers factors of 10,77 95% disturbed sleep, anxiety intestinal using a FGIDs in a CI: 1,49- and dyspepsia are disorders translated- rural Indian 77,89 (veg positively correlated to in a rural validated Rome community. vs non-veg) the prevalence of IBS. Indian III questionnaire population and a hospital anxiety and depression questionnaire. Khayyatzadeh Dietary Iran Cross-sectional Iranian adults Evaluation of Medium 3846 55,5% 828 Not Red meat: IBS aOR 0,76 Study subjects with a et al. patterns study using a working in 50 prevalence (21,5%) specified patients 72,3 95% CI: 'lacto-vegetarian' diet and 106-item self- different health and risk of IBS g/day (Control 0,59-0,98 pattern are less likely to prevalence administered centers. among Iranian group 71,5 (lacto-veg have IBS than subjects of irritable Dish-based Semi- adults g/day). Q1 'lacto- Q1 vs Q4) with 'fast food', bowel quantitative depending on vegetarian': 69 'traditional' or 'western' syndrome Food frequency diet pattern. g/day (Q4: 75 dietary patterns. A 'lacto- in Iranian questionnaire Also g/day). vegetarian' dietary adults designed for comparison of pattern was strongly Iranian adults, IBS frequency associated with a reduced and a modified among the risk for IBS. Those in the Persian version subjects in the top quartile of the 'lacto- of the Rome III top quartile of vegetarian' diet pattern questionnaire. each diet were 24% less likely to pattern to have IBS and 46% less those in the likely to have IBS-C than lower those in the lowest quartiles. quartile. 'Fast food' dietary pattern was associated with greater IBS risk. Buscail et al. Association France Cross-sectional Subjects Evaluation of High 41682 78,0 % 2264 IBS Meat, poultry: aOR: 2,60 A stable vegetarian diet between study using a participating in the (5,4%) patients omnivores 97,0 95% CI: was positively associated self- questionnaire on the NutriNet- association 1,9%, g/day, 1,37-4,91 with IBS. Vegetarians had reported IBS based on Santé cohort between IBS Controls vegetarians 37,1 (stable a significantly lower vegetarian Rome III criteria. study. and a 2,0% g/day, stable vegetarians consumption of meat diet and Lifestyle data Participants who vegetarian vegetarians 14,2 vs controls) (including poultry, fish the including reported organic diet among a g/day. Porc ham, and shellfish) and also a irritable vegetarian diet diseases or alarm large French poultry cuts, lower consumption of bowel was collected symptoms were cohort. processed meat: soft sugary drinks and syndrome previously by excluded. omnivores 43,4 alcoholic beverage in the self- g/day, compared to omnivores. French administered vegetarians 17,0 They had a significantly NutriNet questionnaires at g/day, stable higher consumption of cohort baseline and vegetarians 6,5 eggs, fruits and during follow- g/day. Fish, vegetables, wholegrain ups every six shellfish, products and pulses. months since the processed fish Vegetarians reported start of the and shellfish: lower total energy intake, NutriNet-Santé omnivores 62,9 with higher percent cohort study in g/day, energy from 2009. vegetarians 44,6 carbohydrates. g/day, stable Vegetarians had the vegetarians 39,5 highest intake of fibres g/day. and simple carbohydrates and the lowest intake of animal proteins and cholesterol. Vegetarians also had a lower intake of lactose. Chirila et al. Diet and Romania Cross-sectional A random Assessment of Medium 193 58,5% 37 (19,1%) Not Processed meat: Not IBS patients ate Irritable study. Interview sample of adult IBS specified OR 4,75 95% CI: specified significantly more Bowel by family doctors subjects from a prevalence 1,60-14,09 (IBS frequently processed Syndrome using a food general urban and patients vs meat, beef, canned food, frequency population association of controls) pulses, whole cereals, questionnaire selected from diet with IBS confectionary, fruit and a Rome III family doctors' symptoms. compotes and herb teas. questionnaire. patient lists.

21 Tigchelaar et Habitual The Case-control IBS patients Assessment of Medium 380 IBS 194 IBS Processed meat: Not IBS patients had a higher al. diet and Nether- study. IBS participating in habitual patients (51,1%) patients IBS patients 38,3 specified consumption of for diet quality lands diagnosis and the Maastricht dietary intake 74,7%, 2,1%, g/day (control example processed meat in Irritable subtypes defined IBS cohort study, in IBS Controls Controls group 29,4 and sandwich meat, and a Bowel by Rome III and healthy patients, with 62,9% 6,5% g/day). Meat: IBS lower consumption of Syndrome: criteria. Self- controls from the special patients 48,1 apples, pasta and A case- administered same emphasis on g/day (control alcoholic beverages. control food frequency geographical the overall group 48,1 Lower overall diet quality study questionnaire of region recruited diet quality. g/day). Poultry: in the IBS group. intake over the via public IBS patients 11,1 previous month. advertisements. g/day (control group 9,7 g/day). Fish: IBS patients 15,7 g/day (control group 15,8 g/day). Böhn et al. Nutrient Sweden Case-control IBS patients Assessment of Low 561 74,3% 187 Not Intake of meat: Not IBS patients tended to intake in study. Self- participating in nutrient (33,3%) specified not specified. specified have a higher intake of patients administered 4- treatment trials intake in IBS %E protein IBS fruits and vegetables and with days food and an ongoing patients patients vs a lower intake of meat irritable registration prospective compared to controls: 17% vs and dairy products. bowel record. study on the general 16%, p=0,007. Nutrient intake in IBS syndrome pathophysiologi- population, patients and the general compared cal factors in IBS nutritional population was similar, with the at the differences and met national nutrient general Sahlgrenska between IBS recommendations. population University subgroups, Hospital, and an and whether age- and gender- nutrient matched control intake met group from a recommenda- nation-wide tions. dietary survey, Riksmaten.