Irritable bowel syndrome (IBS) is the most commonly diagnosed

Transcription

1 CLINICAL GASTROENTEROLOGY AND HEPATOLOGY 2012;10: Increased Risk of Miscarriage and Ectopic Pregnancy Among Women With Irritable Bowel Syndrome ALI S. KHASHAN,*, EAMONN M. M. QUIGLEY, ROSEANNE McNAMEE, FERGUS P. McCARTHY,* FERGUS SHANAHAN, and LOUISE C. KENNY* *Anu Research Centre, Department of Obstetrics and Gynecology, and Department of Medicine, Alimentary Pharmabiotic Centre, University College Cork, National University of Ireland, Cork, Ireland; and Maternal and Fetal Health Research Group, and Biostatistics Group, University of Manchester, Manchester, United Kingdom BACKGROUND & AIMS: Irritable bowel syndrome (IBS) is the most commonly diagnosed gastrointestinal condition and is most prevalent in women of reproductive age. We investigated the effects of IBS on risk for adverse outcomes from pregnancy. METHODS: We conducted a cohort study by using the United Kingdom General Practice Research Database. The study cohort consisted of 100,000 women selected by stratified random sampling from all women with a diagnosis of pregnancy from January 1, 1990, to December 31, Those with a recorded diagnosis of IBS before pregnancy were identified (n 26,543). Outcome measures were spontaneous miscarriage, ectopic pregnancy, preeclampsia, and stillbirth. Odds ratios (ORs) and 95% confidence intervals (CIs) for the association between IBS and pregnancy outcomes were estimated by using logistic regression adjusted for several potential confounders. RESULTS: Of women diagnosed with IBS before pregnancy, 6578 (7%) had a spontaneous miscarriage, 741 (0.74%) had an ectopic pregnancy, 425 (0.43%) developed preeclampsia, and 217 (0.22%) had a stillbirth. Maternal IBS was associated with a moderately increased risk of miscarriage (OR, 1.21; 95% CI, ) and ectopic pregnancy (OR, 1.28%; 95% CI, ). There did not appear to be an association between IBS and preeclampsia (OR, 1.09; 95% CI, ) or stillbirth (OR, 1.00; 95% CI, ). CONCLUSIONS: IBS, a common disorder in women of reproductive age, appears to increase the risk of miscarriage and ectopic pregnancy. These findings indicate the importance of prenatal care for women with IBS. Keywords: Abdominal Pain; Bloating; Intestinal; Pregnant; Complication. Irritable bowel syndrome (IBS) is the most commonly diagnosed gastrointestinal condition. It is characterized by the presence of abdominal pain or discomfort that occurs in association with altered bowel habit and bloating or distention during a period of at least 3 months in the absence of organic gastrointestinal disease. 1 IBS is estimated to have a prevalence varying between 7% and 20% of the population, 2,3 affects women more commonly than men, 4 8 and results in significantly impaired quality of life. 9,10 At present, there is no verified laboratory test to aid in the diagnosis of IBS, which is based on the identification of positive symptoms such as those proposed by Manning et al 11 and, where appropriate, the exclusion of other diagnoses. Typical symptoms include pain relieved with defecation, looser stools at the onset of pain, more frequent stools at the onset of pain, passage of mucus, sensation of incomplete evacuation, and visible abdominal distention. The likelihood of IBS is proportional to the number of positive Manning criteria. Although more recent approaches and the various iterations of the Rome consensus process, in particular, have produced criteria that have been widely adopted in clinical research, their use in clinical practice has been less widespread. 12,13 IBS might be associated with depression and anxiety, with reported comorbidity rates as high as 12% and 50%, respectively. 14,15 The pathophysiology of IBS is uncertain. Because of the heterogeneity of the patient population, it seems likely that IBS might involve multiple underlying causes. Although initial enthusiasm for dysmotility as the basic abnormality has waned, various studies have advocated roles for visceral hypersensitivity or hyperalgesia, altered brain-gut signaling, psychological dysfunction, microbial dysbiosis, and immunoinflammatory disturbances. 16 Although it is unrealistic to expect that any one of these phenomena will explain all of IBS, it is possible that each might define a particular and as yet undefined subset Disturbances of circulating cytokines have been implicated in both the pathophysiology of IBS 11 and adverse pregnancy outcomes such as spontaneous miscarriage, ectopic pregnancy, preeclampsia, and stillbirth We addressed the possibility that women with IBS might have increased risk of adverse pregnancy outcomes by using data from the General Practice Research Database (GPRD). Considering the high depression/anxiety prevalence in IBS patients, patients were stratified according to their depression/ anxiety status to ensure that any observed association between IBS and pregnancy outcome is independent from depression/ anxiety. Materials and Methods We performed a cohort study by using data from the GPRD. 26 The database contains longitudinal records for more than 10 million people entered routinely by general practitioners across the United Kingdom. The database contains demographic information, major diagnoses and medical events from Abbreviations used in this paper: BMI, body mass index; CI, confidence interval; GPRD, General Practice Research Database; IBD, inflammatory bowel disease; IBS, irritable bowel syndrome; IBS-C, constipationpredominant irritable bowel syndrome; IBS-D, diarrhea-predominant irritable bowel syndrome; OR, odds ratio by the AGA Institute /$

2 August 2012 IBS AND MISCARRIAGE RISK 903 general practitioners, referrals to specialists, and hospital admissions. Several IBS studies and several pregnancy studies have used GPRD data. 6,27 29 Study Cohort During the study period there were 3.66 million women with up-to-standard data aged years. Up-to-standard is a term used in the GPRD to indicate prospective data recording by the General Practice. Any data before the up-to-standard period were recorded retrospectively after the General Practice joined the GPRD. The study cohort consisted of 100,000 women selected by stratified random sampling from all women with a diagnosis of pregnancy in their clinical or referral records during a 19-year period between January 1, 1990, and December 31, Any record of pregnancy constituted evidence of a pregnancy; the first pregnancy during this period was considered the index pregnancy and used in the analysis. The date of index pregnancy was defined as the time when pregnancy was first recorded in the GPRD. Women were years of age on the date of the index pregnancy and had at least 2 years of up-to-standard data records before the date of their index pregnancy (cohort 0). Two cohorts were selected by stratified random sampling from cohort 0: (1) a random sample of women who had no records of IBS or depression/anxiety (unexposed cohort) and (2) a random sample of women who had records of IBS (exposed cohort). The exposed cohort was divided into 2 groups on the basis of whether the woman had a record of depression/anxiety, because IBS is frequently associated with depression/anxiety. 15 The unexposed women had no diagnosis of IBS or depression/anxiety (before or after pregnancy) in their clinical or referral record between January 1, 1987, and December 31, The exposed cohort consisted of women with a record of IBS in their clinical or referral records between January 1, 1987, and December 31, The date of IBS diagnosis must have preceded the date of the index pregnancy for a woman to be included in the exposed cohort. Women who had their first diagnosis of IBS after the date of the index pregnancy were not eligible for selection. Initially we requested data on 50,000 women with no IBS or depression/ anxiety, 10,000 women with IBS but not depression/anxiety, and 10,000 women with IBS and depression/anxiety. However, the data we received were on 100,000 women including 18,159 with IBS but no depression/anxiety and 8384 with IBS and depression/anxiety. Table 1. Maternal Characteristics in Relation to IBS Variable No IBS or depression/anxiety (n 73,457) IBS only (n 18,159) IBS and depression/anxiety (n 8384) Maternal age at index pregnancy date (y), n (%) (5.3) 268 (1.5) 61 (0.7) (6.6) 595 (3.3) 159 (1.90) ,417 (16.9) 2804 (15.4) 1102 (13.1) ,732 (24.1) 5002 (27.5) 2139 (25.5) ,881 (25.7) 5238 (28.8) 2284 (27.2) ,615 (14.4) 2816 (15.5) 1587 (18.9) (6.9) 1436 (7.9) 1052 (12.5) BMI (kg/m 2 ), n (%) (3.6) 723 (4.0) 371 (4.4) ,157 (42.4) 9007 (49.6) 4085 (48.7) ,346 (15.4) 3070 (16.9) 1540 (18.4) (5.5) 1002 (5.5) 646 (7.7) (3.0) 493 (2.7) 394 (4.7) Data missing 22,053 (30.0) 3864 (21.3) 1348 (16.1) Social deprivation score, n (%) 1 (least deprived) 14,049 (19.1) 3815 (21.0) 1575 (18.8) 2 12,684 (17.3) 3352 (18.5) 1505 (17.9) 3 13,584 (18.5) 3195 (17.6) 1626 (19.4) 4 15,016 (20.4) 3666 (20.2) 1668 (19.9) 5 (most deprived) 18,124 (24.7) 4131 (22.7) 2010 (24.0) Smoking status, n (%) Never smoked 35,672 (48.6) 9348 (51.5) 3679 (43.9) Current smoker 18,759 (25.5) 4813 (26.5) 3027 (36.1) Ex-smoker 6759 (9.2) 2107 (11.6) 1205 (14.4) Data missing 12,267 (16.7) 1891 (10.4) 473 (5.6) Alcohol status, n (%) Nondrinker 10,000 (13.6) 2283 (12.6) 1181 (14.1) Drinker 40,177 (54.7) 11,598 (63.9) 5621 (67.0) Ex-drinker 495 (0.7) 139 (0.8) 104 (1.2) Data missing 22,785 (31.2) 4139 (22.8) 1478 (17.6) No diabetes, n (%) 71,751 (97.7) 17,641 (97.1) 8097 (96.6) No hypertension, n (%) 66,635 (90.7) 16,245 (89.5) 7479 (89.2) No asthma, n (%) 66,795 (90.9) 15,906 (87.6) 7188 (85.7) No IBD, n (%) 72,923 (99.3) 17,702 (97.5) 8187 (97.6) No celiac disease, n (%) 73,320 (99.8) 18,072 (99.5) 8351 (99.6)

3 904 KHASHAN ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 10, No. 8 Outcome measures. Women who had a recorded diagnosis of the following pregnancy complications were identified: spontaneous miscarriage, ectopic pregnancy, preeclampsia, and stillbirth. Because the index pregnancy was the first pregnancy between 1990 and 2008, some of the women in the cohort might have had pregnancies before and/or after the index pregnancy. To ensure that the majority of the outcome data are related to the index pregnancy, the following rules were constructed: (1) spontaneous miscarriages and ectopic pregnancies were considered as those that occurred during the 90 days after the date of index pregnancy, and (2) preeclampsia and stillbirth cases were considered as those that occurred during the year after the date of index pregnancy. Potential confounders. Data were extracted on the following variables: maternal age on the date of the index pregnancy, body mass index (BMI), smoking status, alcohol intake, and social deprivation score. We also identified women with recorded diagnoses of diabetes, high blood pressure ( 140/90 mm Hg), asthma, inflammatory bowel disease (IBD), and celiac disease (maternal comorbidity). Social deprivation scores were based on the practice post code by using the Index of Multiple Deprivation. 30 The rest of the variables were based on the latest measurements on or before date of index pregnancy. Statistical Analysis Logistic regression in Stata Software 10.0 (Stata Corporation, College Station, TX) was used to estimate the odds ratios (ORs) of the association between IBS and miscarriage, ectopic pregnancy, preeclampsia, and stillbirth. The models were adjusted for the following maternal characteristics: age, BMI, alcohol, smoking, social deprivation, diabetes, hypertension, IBD, celiac disease, and asthma, all of which were represented as categorical variables (Table 1). The models accounted for the clustering effect of general practice by using the cluster option in Stata. Women with missing data on BMI, smoking, or alcohol were grouped separately, as reported in Table 1. Adjustment for year of pregnancy in 1-year categories did not change the results and thus was not included in the analyses. Subgroup Analyses It was decided a priori to examine interactions between maternal age and IBS and smoking and IBS. Women were grouped in 2 categories: and years old. The logistic model was repeated with an added interaction term between the binary age variable and the IBS exposure variable. The effect of IBS on pregnancy outcome in smokers and nonsmokers was explored by following the same strategy as for maternal age. Also it was decided a priori to repeat the models with restriction to women with no asthma, diabetes, hypertension, celiac disease, or IBD (women with no additional comorbidity). Sensitivity Analyses The miscarriage and ectopic pregnancy analyses were restricted to women who had at least 2 years of up-to-standard data before the date of IBS diagnosis. It is worth noting that there are 2 relevant dates: (1) date of index pregnancy and (2) Table 2. Association Between IBS and Spontaneous Miscarriage Variable Miscarriage, n Crude OR (95% CI) Adjusted a OR (95% CI) Model I No IBS or depression/anxiety 4500 Reference Reference IBS only ( ) 1.21 ( ) IBS and depression/anxiety ( ) 1.28 ( ) Subgroup analyses b Young mothers ( 24 y) No IBS or depression/anxiety 1038 Reference Reference IBS only ( ) 1.15 ( ) IBS and depression/anxiety ( ) 1.30 ( ) Older mothers ( 24 y) No IBS or depression/anxiety 3462 Reference Reference IBS only ( ) 1.22 ( ) IBS and depression/anxiety ( ) 1.28 ( ) Smokers c No IBS or depression/anxiety 1184 Reference Reference IBS only ( ) 1.16 ( ) IBS and depression/anxiety ( ) 1.23 ( ) Nonsmokers c No IBS or depression/anxiety 2723 Reference Reference IBS only ( ) 1.22 ( ) IBS and depression/anxiety ( ) 1.28 ( ) No additional comorbidity d No IBS or depression/anxiety 3540 Reference Reference IBS only ( ) 1.22 ( ) IBS and depression/anxiety ( ) 1.27 ( ) a Adjusted for maternal age, BMI, smoking, alcohol, social deprivation, asthma, diabetes, hypertension, celiac disease, and IBD. b In subgroup analyses, variables adjusted for were all those referred to in footnote a except for the subgroup variable. c Nonsmokers include never and ex-smokers. Subgroup analyses by smoking status excluded women with missing smoking data. d This analysis excluded women with additional comorbidities.

4 August 2012 IBS AND MISCARRIAGE RISK 905 Table 3. Association Between IBS and Ectopic Pregnancy Variable Ectopic pregnancy, n Crude OR (95% CI) Adjusted a OR (95% CI) Model I No IBS or depression/anxiety 486 Reference Reference IBS only ( ) 1.28 ( ) IBS and depression/anxiety ( ) 1.35 ( ) Subgroup analyses b Young mothers ( 24 y) No IBS or depression/anxiety 86 Reference Reference IBS only ( ) 1.88 ( ) IBS and depression/anxiety ( ) 1.55 ( ) Older mothers ( 24 y) No IBS or depression/anxiety 400 Reference Reference IBS only ( ) 1.19 ( ) IBS and depression/anxiety ( ) 1.32 ( ) Smokers c No IBS or depression/anxiety 168 Reference Reference IBS only ( ) 1.41 ( ) IBS and depression/anxiety ( ) 1.33 ( ) Nonsmokers c No IBS or depression/anxiety 256 Reference Reference IBS only ( ) 1.18 ( ) IBS and depression/anxiety ( ) 1.43 ( ) No additional comorbidity d No IBS or depression/anxiety 367 Reference Reference IBS only ( ) 1.35 ( ) IBS and depression/anxiety ( ) 1.30 ( ) a Adjusted for maternal age, BMI, smoking, alcohol, social deprivation, asthma, diabetes, hypertension, celiac disease, and IBD. b In subgroup analyses, variables adjusted for were all those referred to in footnote a except for the subgroup variable. c Nonsmokers include never and ex-smokers. Subgroup analyses by smoking status excluded women with missing smoking data. d This analysis excluded women with additional comorbidities. date of IBS diagnosis. The up-to-standard condition applies to date of pregnancy only, ie, it was an inclusion criterion. The sensitivity analysis adds another inclusion criterion that is upto-standard before IBS date. A variable was generated to indicate whether the time between diagnosis of IBS and index pregnancy is 5 years or 5 years. This variable was used to examine whether the time between IBS diagnosis and pregnancy has an effect on the association. Results The study cohort consisted of 100,000 women who had a diagnosis of pregnancy in their clinical or referral records between 1990 and Of these, 6578 (6.6%) had a miscarriage, 741 (0.74%) had an ectopic pregnancy, 425 (0.43%) had preeclampsia, and 217 (0.2%) had a stillbirth. There were 18,159 women with a diagnosis of IBS and 8384 with a diagnosis of IBS associated with depression/anxiety. Additional details are summarized in Table 1. Irritable Bowel Syndrome Is Associated With Spontaneous Miscarriage There was an increased risk of miscarriage in women with IBS only (OR, 1.21; 95% confidence interval [CI], ) and women with IBS and depression/anxiety (OR, 1.28; 95% CI, ) (Table 2). Subgroup analyses showed that the risk of miscarriage in women with IBS was consistent regardless of the subgroup of women examined (all tests of interaction were nonsignificant). Young women, women who smoked, and women with no additional comorbidities who had IBS and depression/anxiety had an increased risk of miscarriage by approximately 25% 30% (Table 2). Irritable Bowel Syndrome Is Associated With Ectopic Pregnancy There was an increased risk of ectopic pregnancy in women with IBS only (OR, 1.28%; 95% CI, ) and women with IBS and depression/anxiety (OR, 1.35; 95% CI, ) (Table 3). The subgroup analyses showed that the highest risk of ectopic pregnancy was related to IBS in young women (OR, 1.88; 95% CI, ; interaction P value.07). Apart from age, there was little variation in the association in women with IBS and depression/anxiety on the basis of the subgroup analyses, with all the tests of interaction being nonsignificant (Table 3). Little Evidence of Association With Preeclampsia Overall, there was little evidence of an association between preeclampsia and IBS (OR, 1.09; 95% CI, ) or IBS and depression/anxiety (OR, 1.18; 95% CI, ). The subgroup analyses showed a trend of increased risk of preeclampsia in IBS women who were young (OR, 1.40; 95% CI, ), smokers (OR, 1.66; 95% CI, ), and in those with no other comorbidities (OR, 1.39; 95% CI, ) (Table 4). However, none of the tests of interaction were significant.

5 906 KHASHAN ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 10, No. 8 Table 4. Association Between IBS and Preeclampsia Variable Preeclampsia, n Crude OR (95% CI) Adjusted a OR (95% CI) Model I No IBS or depression/anxiety 300 Reference Reference IBS only ( ) 1.09 ( ) IBS and depression/anxiety ( ) 1.18 ( ) Subgroup analyses b Young mothers ( 24 y) No IBS or depression/anxiety 68 Reference Reference IBS only ( ) 1.40 ( ) IBS and depression/anxiety 3 Older mothers ( 24 y) No IBS or depression/anxiety 232 Reference Reference IBS only ( ) 1.02 ( ) IBS and depression/anxiety ( ) 1.17 ( ) Smokers c No IBS or depression/anxiety 45 Reference Reference IBS only ( ) 1.66 ( ) IBS and depression/anxiety ( ) 1.09 ( ) Nonsmokers c No IBS or depression/anxiety 213 Reference Reference IBS only ( ) 0.98 ( ) IBS and depression/anxiety ( ) 1.17 ( ) No additional comorbidity d No IBS or depression/anxiety 142 Reference Reference IBS only ( ) 1.39 ( ) IBS and depression/anxiety ( ) 1.42 ( ) a Adjusted for maternal age, BMI, smoking, alcohol, social deprivation, asthma, diabetes, hypertension, and IBD. Models were not adjusted for celiac disease because numbers were too small. b In subgroup analyses, the variables adjusted for were all those referred to in footnote a except for the subgroup variable. c Nonsmokers include never and ex-smokers. Subgroup analyses by smoking status excluded women with missing smoking data. d This analysis excluded women with maternal comorbidities. Irritable Bowel Syndrome Is Not Associated With Stillbirth There was no evidence of an association between maternal IBS and risk of stillbirth (adjusted OR, 1.00; 95% CI, ) (Table 5). Women with IBS and depression/anxiety had a consistently increased risk of stillbirth in all analyses, although numbers were too small for robust conclusions. Sensitivity Analyses When the 2-year up-to-standard data before the IBS diagnoses were applied, the conclusions did not change, although the ORs of miscarriage and ectopic pregnancy were 1.31 ( ) and 1.47 ( ), respectively. The time interval between IBS diagnosis and date of index pregnancy did not change the conclusions. Discussion These findings suggest that IBS might play a role in increasing the risk of miscarriage and ectopic pregnancy but not stillbirth. There was some evidence that the association between IBS and preeclampsia might be dependent on maternal age, smoking, and the presence of additional comorbidities. The risk of miscarriage was consistent regardless of maternal age, smoking, or the presence of maternal comorbidities. The study has several strengths. This study investigates the association between clinical diagnosis of IBS in women who sought health care and the risk of pregnancy complications. The study had the advantage of a very large population-based cohort of women with well-documented IBS diagnosis and pregnancy outcomes. Although IBS lacks a reproducible diagnostic biomarker and its diagnosis rests on clinical grounds alone, there is a considerable body of evidence to support the accuracy and integrity of a diagnosis of IBS made in clinical practice. Thompson et al 31 studied the accuracy of IBS diagnosis made by 36 general practitioners among more than 3000 patients and confirmed the diagnosis in 82%; all but 9% were considered to be either IBS or another related, functional gastrointestinal disorder. A recent meta-analysis of the accuracy of symptom-based criteria for IBS diagnosis in primary care concluded that there was a low pretest probability of organic disease among patients who met symptom-based criteria. 32 Furthermore, an IBS diagnosis made clinically is reliable, and a change in diagnosis is unlikely over time, especially in a younger age group such as that studied here. 33 Accordingly, 2 previous studies have used GPRD records to study IBS. 27,34 They identified IBS cases in the GPRD and contacted the general practitioners to assess the accuracy of the records. Both studies reported that 88% of the IBS cases were confirmed by the general practitioner. Because both studies were interested in first diagnosis of IBS, they reported that 77% of the identified IBS cases were confirmed as first-time recorded diagnosis of IBS. Moreover, Lewis et al 35 reported a 92% confirmation rate of IBD diagnosis in the GPRD. This further supports the validity of the diagnostic codes of gastrointestinal diseases in the GPRD. Because of the known association between IBS and

6 August 2012 IBS AND MISCARRIAGE RISK 907 Table 5. Association Between IBS and Stillbirth Variable Stillbirth, n Crude OR (95% CI) Adjusted a OR (95% CI) Model I No IBS or depression/anxiety 153 Reference Reference IBS only ( ) 1.00 ( ) IBS and depression/anxiety ( ) 1.35 ( ) Subgroup analyses b Young mothers ( 24 y) No IBS or depression/anxiety 37 Reference Reference IBS only ( ) 1.08 ( ) IBS and depression/anxiety 3 Older mothers ( 24 y) No IBS or depression/anxiety 116 Reference Reference IBS only ( ) 0.99 ( ) IBS and depression/anxiety ( ) 1.34 ( ) Smokers c No IBS or depression/anxiety 36 Reference Reference IBS only ( ) 1.63 ( ) IBS and depression/anxiety ( ) 1.24 ( ) Nonsmokers c No IBS or depression/anxiety 88 Reference Reference IBS only ( ) 0.78 ( ) IBS and depression/anxiety ( ) 1.38 ( ) No additional comorbidity d No IBS or depression/anxiety 112 Reference Reference IBS only ( ) 1.04 ( ) IBS and depression/anxiety ( ) 1.60 ( ) a Adjusted for maternal age, BMI, smoking, alcohol, social deprivation, asthma, diabetes, hypertension, and IBD. Models were not adjusted for celiac disease because numbers were too small. b In subgroup analyses, the variables adjusted for were all those referred to in footnote a except for the subgroup variable. c Nonsmokers include never and ex-smokers. Subgroup analyses by smoking status excluded women with missing smoking data. d This analysis excluded women with maternal comorbidities. depression/anxiety, the present analyses were careful to minimize the confounding influence of these disorders by separately analyzing women with IBS alone and women with records of IBS depression/anxiety. Furthermore, the patients medical history came from medical records and thus was not subject to recall bias. Confounding by maternal characteristics such as age, smoking, and social deprivation is unlikely; statistical models were adjusted for several potential confounders. The study has some limitations. Pregnancy outcome diagnoses in the GPRD have been validated in several studies that suggested that these outcomes are accurately recorded and complete. 28 However, in the present study and another recent GPRD study, the rates of the outcome measures, mainly preeclampsia and stillbirth, appeared to be lower than those normally reported in the general population. 28 Despite this limitation, the rate of underreporting of these outcomes is unlikely to have differed between women with and without IBS. Therefore, this is unlikely to have biased the present results. It could be argued that women with IBS are more likely to detect their pregnancy earlier and hence visit their general practitioner earlier, and consequently, their miscarriage is more likely to be recorded. However, we are not aware of any evidence to support this. The study cohort lacks data on the subtype and severity of IBS. Such data would have facilitated a further exploration of the potential biological mechanisms of the observed associations. Although neither the Rome nor Manning criteria require constipation or diarrhea to be present for IBS diagnosis, Rome II classified IBS into diarrhea-predominant (IBS-D) and constipation-predominant (IBS-C). 8,9,36 However, Quigley and Shanahan 9 have argued that these are not fixed entities, because the current evidence suggests that IBS-D and IBS-C overlap and criss-cross over time. For example, it has been reported that the prevalence of IBS in North America is equally distributed among IBS-D, IBS-C, and IBS alternating between both. 37 Two previous GPRD studies reported that approximately 52% of IBS episodes were mild, and only 3% were severe. 27,34 Data on the severity of IBS would have allowed an examination of whether there is increasing risk of adverse pregnancy outcome with increasing severity. However, it is very difficult to assess the severity of a condition that is subjective and there are no objective signs. IBS severity is fluctuant, and patients with active IBS symptoms experience frequent flares characterized by days with severe symptoms, followed by days with minimal or no symptoms. 7,8,38 Although data on prescribed medications were available, it was difficult to explore the effect of these medications on the reported associations because of the large and varied number of medications used for IBS. A previous GPRD study reported that 13% of female IBS patients did not receive any drug. 34 Among female patients who did receive drug treatment, 69% received antispasmodics, 13% received laxatives, and 9% received both. However, no data were available on over-the-counter drugs received by IBS patients including laxatives, antidiarrheal, and health food store products.

7 908 KHASHAN ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 10, No. 8 Although the mechanism(s) underpinning a link between IBS and risk of adverse outcome in pregnancy is unclear, several issues arise. First, the accuracy of the diagnosis of IBS and whether it might be confused with or overlap with pelvic inflammatory disease will need further study in this population. The 2 conditions often have similar symptoms, and pelvic inflammatory disease is an established risk factor for ectopic pregnancy. Furthermore, Whorwell et al 39 reported a high prevalence of urologic and gynecologic symptoms among women with IBS and went on to describe abnormal bladder smooth muscle function in a small group of IBS patients, 38 suggesting the involvement of smooth muscle outside the gastrointestinal tract in IBS. Second, because IBS has been associated with a diversity of comorbidities in different organ systems, a systemic defect with disturbed cytokine balance requires prospective assessment. It is noteworthy that IBS is a poor descriptor for a condition that is neither irritable nor confined to the bowel. It is a syndrome with a spectrum of hypersensitivity in many viscera and structures containing innervated smooth muscle, including the genitourinary tract. 39,40 Indeed, Scully et al 41 noted that although interleukin-6 and interleukin-8 levels were elevated in IBS subjects without comorbidity, tumor necrosis factor- and interleukin-1 levels were also elevated in IBS subjects with such comorbidities as fibromyalgia, chronic fatigue syndrome, and premenstrual syndrome. Determining whether it is these comorbidities that confer the pregnancy risk in IBS should be a topic for future study. Of interest, an association between elevated levels of tumor necrosis factor- during pregnancy and a history of prior trauma has been reported recently. 42 Although this study did not comment on the prevalence of IBS among the women they studied, an association between IBS and various life traumas has also been well described 43 Third, the possibility that prescribed and over-the-counter medication might have an effect on either fallopian tube peristalsis or myometrial contractility needs further examination. Conclusions In view of the high prevalence of IBS in women of reproductive age, evidence for an increased risk of miscarriage and ectopic pregnancy in such patients merits further study and potentially has significant implications for antenatal management. References 1. Brandt LJ, Chey WD, Foxx-Orenstein AE, et al. An evidence-based position statement on the management of irritable bowel syndrome. Am J Gastroenterol 2009;104(suppl 1):S Hungin AP, Whorwell PJ, Tack J, et al. The prevalence, patterns and impact of irritable bowel syndrome: an international survey of 40,000 subjects. Aliment Pharmacol Ther 2003;17: Hungin AP, Chang L, Locke GR, et al. Irritable bowel syndrome in the United States: prevalence, symptom patterns and impact. Aliment Pharmacol Ther 2005;21: Talley NJ. Diagnosing an irritable bowel: does sex matter. Gastroenterology 1991;100: Talley NJ, Zinsmeister AR, Vandyke C, et al. Epidemiology of colonic symptoms and the irritable-bowel-syndrome. Gastroenterology 1991;101: Jones R, Latinovic R, Charlton J, et al. Physical and psychological co-morbidity in irritable bowel syndrome: a matched cohort study using the General Practice Research Database. Aliment Pharmacol Ther 2006;24: Ford AC, Forman D, Bailey AG, et al. Irritable bowel syndrome: a 10-yr natural history of symptoms and factors that influence consultation behavior. Am J Gastroenterol 2008;103: Olafsdottir LB, Gudjonsson H, Jonsdottir HH, et al. Stability of the irritable bowel syndrome and subgroups as measured by three diagnostic criteria: a 10-year follow-up study. Aliment Pharmacol Ther 2010;32: Quigley EM, Shanahan F. The language of medicine: words as servants and scoundrels. Clin Med 2009;9: Leong SA, Barghout V, Birnbaum HG, et al. The economic consequences of irritable bowel syndrome: a US employer perspective. Arch Intern Med 2003;163: Manning AP, Thompson WG, Heaton KW, et al. Towards positive diagnosis of irritable bowel. Br Med J 1978;2: Hammer J, Talley NJ. Diagnostic criteria for the irritable bowel syndrome. Am J Med 1999;107:5S 11S. 13. Longstreth GF, Thompson WG, Chey WD, et al. Functional bowel disorders. Gastroenterology 2006;130: Whitehead WE, Palsson O, Jones KR. Systematic review of the comorbidity of irritable bowel syndrome with other disorders: what are the causes and implications? Gastroenterology 2002; 122: Mikocka-Walus A, Turnbull D, Moulding N, et al. Psychological comorbidity and complexity of gastrointestinal symptoms in clinically diagnosed irritable bowel syndrome patients. J Gastroenterol Hepatol 2008;23: Latimer P, Sarna S, Campbell D, et al. Colonic motor and myoelectrical activity: a comparative study of normal subjects, psychoneurotic patients, and patients with irritable bowel syndrome. Gastroenterology 1981;80: Macsharry J, O Mahony L, Fanning A, et al. Mucosal cytokine imbalance in irritable bowel syndrome. Scand J Gastroenterol 2008;43: O Malley D, Quigley E, Dinan T, et al. Do interactions between stress and immune responses lead to symptom exacerbations in irritable bowel syndrome? Brain Behav Immun 2011;25: Brint EK, MacSharry J, Fanning A, et al. Differential expression of Toll-like receptors in patients with irritable bowel syndrome. Am J Gastroenterol 2011;106: Dinan TG, Cryan J, Shanahan F, et al. IBS: an epigenetic perspective. Nat Rev Gastro Hepat 2010;7: Kellow JE, Azpiroz F, Delvaux M, et al. Applied principles of neurogastroenterology: physiology/motility sensation. Gastroenterology 2006;130: Blackwell S, Romero R, Chaiworapongsa T, et al. Maternal and fetal inflammatory responses in unexplained fetal death. J Matern Fetal Neonatal Med 2003;14: Challis JR, Lockwood CJ, Myatt L, et al. Inflammation and pregnancy. Reprod Sci 2009;16: Iyibozkurt AC, Kalelioğlu I, Gursoy S, et al. Evaluation of serum levels of interleukin-10, interleukin-11 and leukemia inhibitory factor in differentiation of eutopic and tubal ectopic pregnancies. Clin Exp Obstet Gynecol 2010;37: Lahra MM, Gordon A, Jeffery HE. Chorioamnionitis and fetal response in stillbirth. Am J Obstet Gynecol 2007;196:e1 e Epidemiology and Pharmacology Information Centre. The General Practice Research Database: a guide for researchers. London, England, Huerta C, Rodriguez LAG, Wallander MA, et al. Risk of irritable bowel syndrome among asthma patients. Pharmacoepidemiol Drug Saf 2002;11:31 35.

8 August 2012 IBS AND MISCARRIAGE RISK Tata LJ, Card TR, Logan RF, et al. Fertility and pregnancy-related events in women with celiac disease: a population-based cohort study. Gastroenterology 2005;128: Rodríguez LA, Ruigómez A. Increased risk of irritable bowel syndrome after bacterial gastroenteritis: cohort study. BMJ 1999; 318: Office of Deputy Prime Minister. The English Indices of Deprivation (revised). Available at: documents/communities/pdf/ pdf. Accessed September 30, Thompson WG, Heaton KW, Smyth GT, et al. Irritable bowel syndrome in general practice: prevalence, characteristics, and referral. Gut 2000;46: Jellema P, Windt DA, Schellevis FG, et al. Systematic review: accuracy of symptom-based criteria for diagnosis of irritable bowel syndrome in primary care. Aliment Pharmacol Ther 2009; 30: Locke GR 3rd. Natural history of irritable bowel syndrome and durability of the diagnosis. Rev Gastroenterol Disord 2003; 3(suppl 3):S12 S Ruigomez A, Wallander MA, Johansson S, et al. One-year follow-up of newly diagnosed irritable bowel syndrome patients. Aliment Pharmacol Ther 1999;13: Lewis JD, Brensinger C, Bilker WB, et al. Validity and completeness of the General Practice Research Database for studies of inflammatory bowel disease. Pharmacoepidemiol Drug Saf 2002; 11: Mearin F, Baró E, Roset M, et al. Clinical patterns over time in irritable bowel syndrome: symptom instability and severity variability. Am J Gastroenterol 2004;99: Brandt LJ, Bjorkman D, Fennerty MB, et al. Systematic review on the management of irritable bowel syndrome in North America. Am J Gastroenterol 2002;97:S7 S Halder SL, Locke GR, Schleck CD, et al. Natural history of functional gastrointestinal disorders: a 12-year longitudinal population-based study. Gastroenterology 2007;133: Whorwell PJ, Lupton EW, Erduran D, et al. Bladder smooth muscle dysfunction in patients with irritable bowel syndrome. Gut 1986;27: Whorwell PJ, McCallum M, Creed FH, et al. Non-colonic features of irritable bowel syndrome. Gut 1986;27: Scully P, McKernan DP, Keohane J, et al. Plasma cytokine profiles in females with irritable bowel syndrome and extra-intestinal co-morbidity. Am J Gastroenterol 2010;105: Blackmore ER, Moynihan JA, Rubinow DR, et al. Psychiatric symptoms and proinflammatory cytokines in pregnancy. Psychosom Med 2011;73: Bradford K, Shih W, Videlock E, et al. Association between early adverse life events and irritable bowel syndrome. Clin Gastroenterol Hepatol 2012;10: Reprint requests Address requests for reprints to: Ali S. Khashan, PhD, Anu Research Center, Department of Obstetrics and Gynecology, University College Cork, National University of Ireland, Cork University Maternity Hospital, Wilton, Cork, Ireland. a.khashan@ucc.ie; fax: Acknowledgments The authors thank Professor Philip N. Baker, Faculty of Medicine and Dentistry, University of Alberta, for his advice on the study protocol. The authors also thank Ms. Tarita Murray-Thomas, General Practice Research Database, London, UK, for her advice and assistance on the study design. Conflicts of interest The authors disclose no conflicts. Funding This study is based in part on data from the Full Feature General Practice Research Database obtained under license from the UK Medicines and Healthcare Products Regulatory Agency. However, the interpretation and conclusions contained in this study are those of the authors alone. Access to the General Practice Research Database (GPRD) was funded through the Medical Research Council s license agreement with the Medicines and Healthcare Products Regulatory Agency. Dr Khashan and Professor Kenny are funded by the Health Research Board of Ireland.