Outcomes of Pregnancies for Women Undergoing Endoscopy While They Were Pregnant: A Nationwide Cohort Study

Accepted Manuscript Outcomes of Pregnancies for Women Undergoing Endoscopy While They Were Pregnant—a Nationwide Cohort Study Jonas F. Ludvigsson, Benjamin Lebwohl, Anders Ekbom, Ravi Kiran, Peter HR. Green, Jonas Höijer, Olof Stephansson PII: DOI: Reference:

S0016-5085(16)35240-4 10.1053/j.gastro.2016.10.016 YGAST 60779

To appear in: Gastroenterology Accepted Date: 14 October 2016 Please cite this article as: Ludvigsson JF, Lebwohl B, Ekbom A, Kiran R, Green PH, Höijer J, Stephansson O, Outcomes of Pregnancies for Women Undergoing Endoscopy While They Were Pregnant—a Nationwide Cohort Study, Gastroenterology (2016), doi: 10.1053/j.gastro.2016.10.016. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Ludvigsson et al.

ACCEPTED MANUSCRIPT

Outcomes of Pregnancies for Women Undergoing Endoscopy While They Were Pregnant—a Nationwide Cohort Study Authors’ affiliations Jonas F Ludvigsson1, 2, 3, Benjamin Lebwohl1, 4, Anders Ekbom5, Ravi Kiran6, Peter HR Green4, Jonas Höijer7, Olof Stephansson5, 8, 1

M AN U

SC

RI PT

Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden 2 Department of Pediatrics, Örebro University Hospital, Sweden 3 Division of Epidemiology and Public Health, School of Medicine, University of Nottingham, Clinical Sciences Building 2, City Hospital, Nottingham, UK 4 Celiac Disease Center, Department of Medicine, Columbia University College of Physicians and Surgeons, New York, New York, USA 5 Clinical Epidemiology Unit, Department of Medicine Solna, Karolinska Institutet, Sweden 6 Division of Colorectal Surgery New York Presbyterian/Columbia University Medical Center New York, New York, USA 7 Unit of Biostatistics, Institute of Environmental Medicine, Karolinska Institutet, Sweden 8 School of Public Health, University of California, Berkeley, USA

AC C

EP

TE D

Corresponding author: Jonas F Ludvigsson Department of Medical Epidemiology and Biostatistics Karolinska Institutet, Sweden E-mail: [email protected] Phone: +46-19-6021000. Fax: +46-19-187915

1

Ludvigsson et al.

ACCEPTED MANUSCRIPT

Contributors JFL conceived and designed the study with input from the other authors. JFL also wrote the first draft of the paper and supervised the project. JFL and OS funded the study. JH analyzed the data. All authors interpreted the data and contributed to the writing of the paper. All authors revised and approved the final version. OS had full access to all the data. JH takes responsibility for the integrity of the data and the accuracy of the data analyses. JL is the guarantor of the data.

RI PT

Competing interest: None. Disclaimer: This manuscript represents the views of the authors.

SC

Funding and Role of the funding sources All authors have completed the ICMJE uniform disclosure form at www.icmje.org/coi_disclosure.pdf and declare (that): this project was supported by grants from the Swedish Society of Medicine and the Stockholm County Council, and the Swedish Research Council (grant 2013-2429).

M AN U

Details of ethics approval: This project (2008/1182-31/4) was approved by the Research Ethics Committee of the Karolinska Institute, Sweden on September 3, 2008. Data sharing Other researchers can apply for our data through the Swedish National Board of Health and Welfare.

TE D

Transparency The lead author affirms that this manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned (and, if relevant, registered) have been explained.

AC C

EP

License for publication The Corresponding Author has the right to grant on behalf of all authors and does grant on behalf of all authors, a worldwide licence to the Publishers and its licensees in perpetuity, in all forms, formats and media (whether known now or created in the future), to i) publish, reproduce, distribute, display and store the Contribution, ii) translate the Contribution into other languages, create adaptations, reprints, include within collections and create summaries, extracts and/or, abstracts of the Contribution, iii) create any other derivative work(s) based on the Contribution, iv) to exploit all subsidiary rights in the Contribution, v) the inclusion of electronic links from the Contribution to third party material where-ever it may be located; and, vi) license any third party to do any or all of the above.

2

Ludvigsson et al.

ACCEPTED MANUSCRIPT

RI PT

WHAT IS ALREADY KNOWN ON THIS TOPIC • Many pregnant women have gastrointestinal conditions that are usually evaluated through endoscopy. • Guidelines aimed at gastrointestinal disease tend to discourage endoscopy during pregnancy, and suggest that when necessary endoscopy should be carried out during the second trimester.

M AN U

SC

WHAT THIS STUDY ADDS • Although endoscopy during pregnancy may be associated with adverse pregnancy outcome, excess risks are small and likely due to intrafamilial factors or underlying disease activity • Risks of adverse pregnancy outcome do not seem to differ by trimester.

TE D

Running title: Endoscopy and pregnancy Manuscript version: Oct 3, 2016

Word count: 2936. Abstract: 394. Tables: 3. Figures: 2. Character count-title: 90 Appendix: 1

AC C

EP

List of abbreviations: ARR, Adjusted Relative Risk; CI, Confidence Interval; ERCP, Endoscopic Retrograde Cholangiopancreatography; GI, Gastrointestinal; IBD, inflammatory bowel disease; SGA, small for gestational age.

3

Ludvigsson et al.

ACCEPTED MANUSCRIPT

ABSTRACT

RI PT

Background & Aims: Endoscopy is an integral part of the investigation and management of gastrointestinal disease. We aimed to examine outcomes of pregnancies for women who underwent endoscopy during their pregnancy.

M AN U

SC

Methods: We performed a nationwide population-based cohort study, linking data from the Swedish Medical Birth Registry (for births from 1992 through 2011) with those from the Swedish Patient Registry. We identified 3052 pregnancies exposed to endoscopy (2025 upper endoscopies, 1109 lower endoscopies, 58 endoscopic retrograde cholangiopancreatographies). Using Poisson regression, we calculated adjusted relative risks (ARRs) for adverse outcomes of pregnancy according to endoscopy status using 1,589,173 unexposed pregnancies as reference. To consider the effects of disease activity, we examined pregnancy outcomes (preterm birth, stillbirth, small for gestational age [SGA], or congenital malformations) in women who underwent endoscopy just before or after pregnancy. Secondary factors included induction of labor, low birth weight (<2500g), cesarean section, Apgar score below 7 at 5 minutes, and neonatal death within 28 days. To consider intra-familial factors, we compared pregnancies within the same mother.

AC C

EP

TE D

Results: Exposure to any endoscopy during pregnancy was associated with an increased risk of preterm birth (ARR, 1.54; 95% CI, 1.36–1.75) or SGA (ARR, 1.30; 95% CI, 1.07–1.57) but not of congenital malformation (ARR, 1.00; 95% CI, 0.83– 1.20) or stillbirth (ARR, 1.45; 95% CI, 0.87–2.40). None of the 15 stillbirths to women with endoscopy occurred less than 2 weeks after endoscopy. ARRs were independent of trimester. Compared to women with endoscopy less than 1 year before or after pregnancy, endoscopy during pregnancy was associated with preterm birth (ARR, 1.16) but not with SGA (ARR, 1.19), stillbirth (ARR, 1.11), or congenital malformation (ARR, 0.90). Restricting the study population to women having an endoscopy during pregnancy or before/after, and only analyzing data from women without a diagnosis of inflammatory bowel disease, celiac disease, or liver disease, endoscopy during pregnancy was not linked to preterm birth (ARR, 1.03; 95% CI, 0.84–1.27). Comparing births within the same mother, for which only 1 birth had been exposed to endoscopy, we found no association between endoscopy and gestational age or birth weight. Conclusions: In a nationwide population-based cohort study, we found endoscopy during pregnancy to be associated with increased risk of preterm birth or SGA, but not of congenital malformation or stillbirth. However, these risks are small and likely due to intra-familial factors or disease activity. KEY WORDS: ERCP, IBD, fetal, fetus

4

Ludvigsson et al.

ACCEPTED MANUSCRIPT

SC

RI PT

INTRODUCTION Each year more than 12,000 US pregnant women have conditions that are usually evaluated by esophagogastroduodenoscopy (EGD, here denoted “upper endoscopy”), another 6,000 are in need of lower endoscopy, and another 1,000 women have symptomatic choledocholithiasis that in a non-pregnant woman would have entailed ERCP1. Despite these large numbers, research in pregnancy outcome in women undergoing endoscopy during pregnancy is scarce. In addition to some 30 case reports1, we are aware of nine studies2-10 with original data on in total 379 pregnant women undergoing endoscopy. Of these studies, two examined pregnancy outcome in upper endoscopy (n=143)2, 5, two examined pregnancy outcome in sigmoidoscopy or colonoscopy (n=116)3, 4, 10, and four in ERCP (n=120)6-9.

EP

TE D

M AN U

Due to the lack of evidence, endoscopy including ERCP is generally only performed when there are strong indications11. Guidelines aimed at specific GI disease also tend to discourage endoscopy during pregnancy12, 13, even though some organizations note that endoscopy including ERCP may be considered safe13. Despite the lack of knowledge, it is recommended that when endoscopy is necessary it should be performed during the second trimester12, 13. However, so far the total number of reported pregnant women undergoing endoscopy in the second trimester is <100, and consequently the statistical power to detect pregnancy complications that otherwise occur in only a few percent of pregnancies has been extremely low. Few studies have compared pregnancy outcome in women undergoing endoscopy with that in pregnant controls and no endoscopy2, 4, 10, and we are aware of no study calculating the relative risk of adverse pregnancy outcome in pregnant women undergoing endoscopy. Some earlier studies have concluded that endoscopy during pregnancy is safe, but several studies have in fact excluded stillbirths and neonatal deaths from the analyses when not occurring immediately after endoscopy2, 3, 6. Finally other studies have failed to follow-up all pregnant women3, 7, 8, making it difficult to draw any firm conclusions about the safety of endoscopy during pregnancy.

AC C

We carried out a nationwide population-based cohort study examining the risk of adverse pregnancy outcome in pregnant women undergoing endoscopy. We hypothesized that endoscopy during pregnancy would increase the prevalence of adverse pregnancy outcome, especially preterm birth and stillbirth.

5

Ludvigsson et al.

ACCEPTED MANUSCRIPT

MATERIALS AND METHODS

RI PT

Registers The Swedish Medical Birth Registry started in 1973 with data on smoking added in 1983 and body mass index in 1992. Hence this study was restricted to births taking place 1992-2011. Through this Registry we retrieved data on all pregnancy-related variables including maternal age, parity, maternal country of birth, early pregnancy smoking status and delivery year. We divided country of birth into Nordic (Sweden, Denmark, Norway, Finland, and Iceland) versus non-Nordic countries. Smoking status (at the first antenatal visit) is self-reported and was categorized into (i) nonsmoker, (ii) 1-9, (iii) ≥10 cigarettes per day.

M AN U

SC

The Swedish Patient Registry14 started in 1964 with complete national coverage from 1987. This registry supplied information on endoscopy and ERCP, but also for GI disease diagnoses such as inflammatory bowel disease (IBD), celiac disease, and hepatobiliary diseases, we also collected disease data from the Patient Registry. These diseases were defined according to relevant international classification of disease codes (appendix). The positive predictive values of most diagnoses in this registry are between 85 and 95%14.

TE D

Endoscopy and ERCP We defined exposures according to relevant procedure codes, distinguishing between esophagogastroduodenoscopy (upper), colonoscopy and sigmoidoscopy (lower), and ERCP. Together these procedures made up our main exposure “any endoscopy”. We did not include endoscopy limited to the oropharynx or the rectum.

AC C

EP

Pregnancy outcome Our main outcome measures were preterm birth and stillbirth, since we hypothesized that an endoscopy may trigger preterm labor and stillbirth. In Sweden, stillbirth is recorded from 22 completed gestational weeks since mid-2008, and before that from gestational week 28. Gestational age was determined using ultrasound, and when ultrasound data were missing, we used the first day of the last menstrual period for pregnancy start. Routine ultrasound has been offered to pregnant women in Sweden since the 1990s, and about 95% of the women accept this investigation. Preterm birth was defined as <37 completed gestational weeks while very preterm birth was defined as <32 completed gestational weeks, and moderate preterm birth as 32-36 gestational weeks. We also examined small for gestational age (SGA, defined as a birth weight >2 standard deviations below the sex-specific mean for gestational age15 according to Swedish ultrasound-based reference curves) and congenital malformations (using the same definition as in our earlier paper on congenital malformations in IBD16, see also appendix). Finally we examined a number of secondary outcomes: risk of induction of labor, low birth weight (<2500g), cesarean section, Apgar score below 7 at 5 minutes, and neonatal death within 28 days. We also looked at the continuous outcome measures birth weight and gestational age.

6

Ludvigsson et al.

ACCEPTED MANUSCRIPT

This study was approved by the research ethics committee of the Karolinska Institutet, Stockholm, Sweden.

RI PT

Statistical analyses We examined the risk of adverse pregnancy outcomes in women undergoing (i) any type of endoscopy, (ii) upper endoscopy; (iii) lower endoscopy; and (iv) ERCP during pregnancy. For preterm birth, stillbirth, SGA and congenital malformation, we also examined if risk estimates differed between sigmoidoscopy and full colonoscopy. Analyses were adjusted for maternal age, BMI, party, education, smoking, diabetes, civil status, year of birth, and infant sex (see Table 1 for categories).

M AN U

SC

We specifically studied the risk of SGA and congenital malformation according to trimester of endoscopy. In order to rule out that a potentially increased risk of pregnancy outcome was due to underlying GI disease, we examined the risk of adverse pregnancy outcome according to (i) ERCP status in women with a lifetime diagnosis of hepatobiliary disease; (ii) upper endoscopy in women with a diagnosis of celiac disease, and (iii) lower endoscopy in women with IBD, but also (iv) among women who never had any of these diagnoses (“healthy women with endoscopy”). From hereon, hepatobiliary disease, celiac disease, and IBD are abbreviated gastrointestinal disease (“GI disease”).

AC C

EP

TE D

To address bias due to underlying disease activity (women with endoscopy during pregnancy are more likely to have more severe disease than those without endoscopy during pregnancy), we examined risk of adverse pregnancy outcome in women undergoing endoscopy during pregnancy vs. women having an endoscopy <1 year before or after pregnancy (we studied this in “all women”, but also separately in women with or without a lifetime diagnosis of GI disease). Both genetic and environmental factors influence pregnancy outcome, and we cannot rule out that these co-vary with the risk of GI disease and risk of having an endoscopy. To take intrafamilial confounding into account we compared pregnancy outcome within the same mother (sibling pregnancies). Ideally, both intrafamilial confounding and underlying disease activity should be adjusted for in the same analysis. This is difficult for dichotomous outcome measures since only births within the same mother where one child has a different outcome (e.g. preterm birth) than the other (e.g. term birth) contributes to the relative risk calculations and few mothers will have contrasting birth outcomes. We addressed this comparing the continuous variables (i) gestational age and (ii) birth weight in 417 mothers undergoing endoscopy in one of several births within <1 year from their index pregnancy exposed to endoscopy. Finally to estimate the effect of underlying disease activity we compared pregnancy outcome in women having an endoscopy <1 year before or after pregnancy but never 7

Ludvigsson et al.

ACCEPTED MANUSCRIPT

during pregnancy itself vs. those who never had a record of endoscopy.

RI PT

In order to estimate the adjusted relative risks for binary outcomes, we used Poisson regression with cluster robust standard errors17. For the ordinal outcome variable with categories Term birth, Moderately preterm birth, and Very Preterm Birth, we used multinomial regression with cluster robust standard errors. To avoid mother specific and time constant omitted variable bias, we also performed fixed effects Poisson regression with robust standard errors for the binary outcomes. In this way, we estimate the within-mother effect of exposure.

AC C

EP

TE D

M AN U

SC

For the regression analyses, we adjusted for the potential confounders described in Table 1.The analyses were conducted on observations without missing on the covariates. Only singleton births were used in the analyses. P-values<0.05 were regarded as statistically significant. All tests are 2-sided. All analyses were conducted using Stata, version 13.

8

Ludvigsson et al.

ACCEPTED MANUSCRIPT

RESULTS

M AN U

SC

RI PT

Background data We identified 1,592,225 pregnancies (Table 1) in 1,002,604 unique women with complete data on all covariates used in this study. Of these some 0.19% (n=3,052) had been exposed to an endoscopy during pregnancy (upper: n=2,025 (0.13%); lower: n=1,109 (0.07%) and ERCP: n=58 (<0.01%)). Table 1 presents characteristics of the study participants. The mean gestational age was not influenced more than marginally in women who underwent endoscopy (no endoscopy: 39.9 weeks; any endoscopy: 39.5 weeks), while mean birthweight was 3562g and 3479g, respectively (Table 2). Preterm birth was seen in 4.8% of women without endoscopy. This compares with 7.6% in women undergoing any endoscopy, a slightly higher percentage in women with lower endoscopy but lower in those with ERCP (Table 2). SGA was noted in 2.5% of reference women compared to 3.6% for those undergoing endoscopy. No difference was seen with regards to the proportion of pregnancies with congenital malformations.

AC C

EP

TE D

Main results Women undergoing endoscopy during pregnancy were at increased risk of all outcome measures except congenital malformations (Table 3, and Figure 1). In general, risk estimates were higher in women undergoing lower than upper endoscopy. Due to lack of numbers, adjusted relative risks could not be calculated in mothers having an ERCP during pregnancy. None of the crude risk estimates attained statistical significance in ERCP-exposed mothers and most RRs were between 1 and 1.5. Adjusting for covariates, offspring to women undergoing endoscopy during pregnancy had a slightly lower birthweight and shorter gestational age (Table 3). Using full colonoscopy during pregnancy as reference, we found that women undergoing sigmoidoscopy were at increased risk of preterm birth (RR=1.66; 95% CI=1.06-2.60), but at no increased risk of SGA (ARR=1.23; 95% CI=0.57-2.64) or congenital malformation (ARR=0.71; 95% CI=0.37-1.38). No relative risk was calculated for stillbirth due to lack of positive events. Out of the 15 stillbirths to women with endoscopy, none occurred less than 17 days after endoscopy (median interval: 168 days). Examining the time interval between endoscopy in the third trimester and birth, we found no evidence that late endoscopy might trigger childbirth (Figure 2). We found no increased risks of SGA according to trimester of endoscopy (trimesters 1, 2 and 3 respectively: reference, ARR=0.89, ARR=1.04). Neither was there any difference in the risk of congenital malformations (trimesters 1, 2 and 3 respectively: reference, ARR=1.03, ARR=1.05). Pregnancy outcome according to GI disease status Restricting our data to women who never had a diagnosis of IBD, celiac disease or hepatobiliary disease, risk estimates decreased but most often remained statistically

9

Ludvigsson et al.

ACCEPTED MANUSCRIPT

significantly increased (Appendix and Figure 1). When we instead studied women with any of these GI diseases (combined or separately), the risk of adverse pregnancy outcome such as preterm birth was higher in women undergoing endoscopy (Appendix and Figure 1).

SC

RI PT

Comparison with controls having endoscopy <1 year before or after pregnancy Restricting our control population to women with endoscopy <1 year before or after pregnancy, only preterm birth (ARR=1.16) and induction of labor (ARR=1.11) remained significantly increased in pregnancies exposed to endoscopy (Appendix and Figure 1). Limiting our analyses to women without GI disease (where the potential disease activity should play less of a role), endoscopy was not linked to any of our outcomes (Appendix and Figure 1).

M AN U

Within-mother analyses (sibling pregnancies) To address intrafamilial factors, we then examined pregnancy outcome in siblings with the same mother. In this analysis, endoscopy exposure was associated with preterm birth (ARR=1.32), caesarean section (ARR=1.10) and low birth weight (ARR=1.45)(Appendix and Figure 1).

TE D

Within-mother analyses <1 year before or after endoscopy-exposed pregnancy Restricting controls to sibling pregnancies <1 year within the index pregnancy, pregnancies exposed to endoscopy did not differ with regards to gestational age (adjusted mean difference: -0.8 days; p=0.395) or birthweight (adjusted mean difference: +32g; p=0.924). These differences are equivalent to a reduction of -0.3% in pregnancy duration and an increase of +0.7% in birthweight.

AC C

EP

Women with an endoscopy before or after pregnancy Compared to women never having an endoscopy, those women having an endoscopy before or after pregnancy (but never during pregnancy) were also at increased risk of adverse pregnancy outcome, for instance preterm birth (ARR=1.38) and stillbirth (ARR=1.47)(Appendix).

10

Ludvigsson et al.

ACCEPTED MANUSCRIPT

DISCUSSION

RI PT

In this population-based study, we compared the pregnancy outcome in 3,000 pregnancies exposed to endoscopy compared to about 1.6 million pregnancies without endoscopy exposure. While endoscopy was linked to adverse pregnancy outcome, we suggest that most if not all the excess risks are due to disease activity. Also women having an endoscopy outside pregnancy were at increased risk of adverse pregnancy outcome.

M AN U

SC

Earlier literature In 1996, Cappell et al published two papers on pregnancy outcome in women undergoing endoscopy2, 3. They concluded that endoscopy was not linked to adverse fetal outcome, but stillbirths and one involuntary abortion among women with endoscopy were not considered since they took place in high-risk pregnancies. More importantly, the studies only had the statistical power to detect a many-fold increased risk of e.g. preterm birth, while the current study had the power to detect a 20% increase in preterm birth. More recent studies5, 10 have shown conflicting results but study power has been low.

TE D

Strengths and limitations The main strength of this study is the high statistical power and our use of both general population controls and sibling controls. Overall there were 4,852 stillbirths during follow-up, 15 of which occurred in women undergoing endoscopy during pregnancy. Importantly none of these took place just after endoscopy. Other strengths include our adjustment for potential confounders and the control for intrafamilial factors.

AC C

EP

Among the limitations is our lack of knowledge about the duration of the endoscopy, the type of sedation and bowel preparation used, the position of the patient at endoscopy, and the indication for the procedure. There are no Swedish national data on sedation in endoscopy but drugs frequently used are midazolam, fentanyl and propofol. Given the minimal excess risks seen in endoscopy during pregnancy when we considered proxies for disease activity (e.g. through using women undergoing endoscopy before and after pregnancy as reference) and familial factors, it is unlikely that any of the drugs including anesthesia had any major effect on pregnancy outcome including on congenital malformation, but smaller risk increases due to individual drugs cannot be ruled out. Also, we could not examine early spontaneous abortions since stillbirths in Sweden are only registered from gestational week 22 (and until 2008 from week 28), and hence our data on stillbirths are restricted to the second half of pregnancy. We did not correct for multiple comparisons. We aimed to examine the safety of endoscopy during pregnancy and had we corrected for the large number of comparisons (in order to reduce the risk of type 1 error), we would have been highly likely to miss minor excess risks in pregnancies exposed to endoscopies. Finally, our data on smoking were self-reported and may have underestimated the true prevalence of smoking (although adjustment for smoking had no major effect on our

11

Ludvigsson et al.

ACCEPTED MANUSCRIPT

RRs).

TE D

M AN U

SC

RI PT

We used several approaches to tackle the lack of detailed information on disease activity, for instance we used pregnancies taking place <1 year before or after the index pregnancy as reference, and also stratified for presence of GI disease. However we acknowledge the lack of direct information on disease treatment and on actual disease activity indices such as Mayo Score and Crohn’s Disease Activity Index in patients with IBD. Undiagnosed celiac disease and newly diagnosed celiac disease (when the diet may not yet have influenced mucosal healing) have both been linked to adverse pregnancy outcome18, and so has treatment of IBD19. It should also be noted that adverse pregnancy outcome was in fact more common in women having an endoscopy just before or after pregnancy (these women never had an endoscopy during pregnancy). Although we had great power to examine upper and lower endoscopy, the number of women undergoing ERCP during pregnancy was limited, and smaller excess risks cannot be ruled out. Finally, bias is an issue in all observational studies. While selection bias is unlikely (more than 99% of all somatic inpatient care in Sweden is registered in the Patient Registry, and this registry also includes hospital-based outpatient care since 200114), we cannot rule out other forms of bias. We tried to minimize bias through comparing pregnancies to the same mother within <1 year from the index pregnancy. Short of trials where pregnant women are randomly assigned to undergo endoscopy, we believe that our study design offers the best chance examining the effects of endoscopy during pregnancy. That said, we acknowledge that residual confounding is still possible20.

AC C

EP

Clinical implications Earlier recommendations suggest that endoscopy should only be performed during pregnancy if there are strong indications, and if so not during the second trimester13. Our study shows that endoscopy is unlikely to have a more than marginal influence on pregnancy outcome independently of trimester. Neither does it seem that sigmoidoscopy is preferable to a full colonoscopy in the pregnant woman. We cannot explain this latter finding but it is possible that in women with particularly severe gastrointestinal disease where endoscopy is inevitable, the physician will prefer a sigmoidoscopy rather than a full colonoscopy and under such circumstances the sigmoidoscopy will signal a more severe disease.

Conclusions In conclusion, adverse pregnancy outcome is rare in women undergoing endoscopy during pregnancy. Potential excess risks, if any, seem minimal and should be weighed against the need for timely investigation and treatment of women where an underlying GI disease per se may be a more severe threat to pregnancy outcome than the endoscopy.

12

Ludvigsson et al.

ACCEPTED MANUSCRIPT

Figure 1. Risk of stillbirth, preterm birth, small for gestational age and congenital malformations

RI PT

Figure 1. Legend * Upper 95%CI exceeds 3.00. See Table 3 and Appendix for exact data Endo, Endoscopy. GI, Gastrointestinal (represents celiac disease, inflammatory bowel disease or liver disease in Figure 1). IBD, inflammatory bowel disease.

AC C

EP

TE D

M AN U

SC

All women - Endoscopy during pregnancy and relative risk of adverse pregnancy outcome (corresponds to Table 3) Healthy women - Endoscopy during pregnancy and risk for adverse pregnancy outcome in healthy women (corresponds to Appendix: eTable I) GI disease - Endoscopy during pregnancy and risk for adverse pregnancy outcome in women with inflammatory bowel disease, celiac disease or hepatobiliary disease (Appendix: eTable IIa) Endo<1yr - Pregnancies exposed to endoscopy compared to [reference] pregnancies where the woman had an endoscopy <1 year before or after pregnancy (Appendix: eTable III) Healthy + Endo<1yr - Mothers without celiac disease, IBD or liver disease. Pregnancies exposed to endoscopy compared to [reference] pregnancies where the woman had an endoscopy <1 year before or after pregnancy (Appendix: eTable IV) Gi disease + Endo<1yr - Mothers with celiac disease, IBD or liver disease. Pregnancies exposed to endoscopy compared to [reference] pregnancies where the woman had an endoscopy <1 year before or after pregnancy (Appendix: eTable V) Within mother - Sibling comparisons. Risk of adverse pregnancy outcome in the same mother when exposed to endoscopy vs not exposed) (Appendix: eTable VI) Endo before/after pregnancy - Risk of adverse pregnancy outcome in women with endoscopy <1 year before or after pregnancy but not during pregnancy vs. women who have no record of endoscopy (Appendix: eTable VII)

13

Ludvigsson et al.

ACCEPTED MANUSCRIPT

RI PT

Figure 2. Relationship between days from endoscopy to birth (Y axis), according to gestational age at endoscopy (X axis).

AC C

EP

TE D

M AN U

SC

REFERENCES 1. Cappell MS. The fetal safety and clinical efficacy of gastrointestinal endoscopy during pregnancy. Gastroenterol Clin North Am 2003;32:123-79. 2. Cappell MS, Colon VJ, Sidhom OA. A study of eight medical centers of the safety and clinical efficacy of esophagogastroduodenoscopy in 83 pregnant females with follow-up of fetal outcome with comparison control groups. Am J Gastroenterol 1996;91:348-54. 3. Cappell MS, Colon VJ, Sidhom OA. A study at 10 medical centers of the safety and efficacy of 48 flexible sigmoidoscopies and 8 colonoscopies during pregnancy with follow-up of fetal outcome and with comparison to control groups. Dig Dis Sci 1996;41:2353-61. 4. Cappell MS, Fox SR, Gorrepati N. Safety and efficacy of colonoscopy during pregnancy: an analysis of pregnancy outcome in 20 patients. J Reprod Med 2010;55:115-23. 5. Debby A, Golan A, Sadan O, et al. Clinical utility of esophagogastroduodenoscopy in the management of recurrent and intractable vomiting in pregnancy. J Reprod Med 2008;53:347-51. 6. Jamidar PA, Beck GJ, Hoffman BJ, et al. Endoscopic retrograde cholangiopancreatography in pregnancy. Am J Gastroenterol 1995;90:1263-7. 7. Kahaleh M, Hartwell GD, Arseneau KO, et al. Safety and efficacy of ERCP in pregnancy. Gastrointest Endosc 2004;60:287-92. 8. Tang SJ, Mayo MJ, Rodriguez-Frias E, et al. Safety and utility of ERCP during pregnancy. Gastrointest Endosc 2009;69:453-61. 9. Tham TC, Vandervoort J, Wong RC, et al. Safety of ERCP during pregnancy. Am J Gastroenterol 2003;98:308-11. 10. de Lima A, Zelinkova Z, van der Woude CJ. A Prospective Study of the Safety of Lower Gastrointestinal Endoscopy During Pregnancy in Patients with Inflammatory Bowel Disease. J Crohns Colitis 2015;9:519-24. 11. Committee ASoP, Shergill AK, Ben-Menachem T, et al. Guidelines for endoscopy in pregnant and lactating women. Gastrointest Endosc 2012;76:1824. 12. Ludvigsson JF, Bai JC, Biagi F, et al. Diagnosis and management of adult coeliac disease: guidelines from the British Society of Gastroenterology. Gut 2014;63:1210-28. 13. van der Woude CJ, Kolacek S, Dotan I, et al. European evidenced-based consensus on reproduction in inflammatory bowel disease. J Crohns Colitis 2010;4:493-510.

14

Ludvigsson et al.

16.

17.

18.

19.

AC C

EP

TE D

M AN U

20.

RI PT

15.

Ludvigsson JF, Andersson E, Ekbom A, et al. External review and validation of the Swedish national inpatient register. BMC Public Health 2011;11:450. Marsal K, Persson PH, Larsen T, et al. Intrauterine growth curves based on ultrasonically estimated foetal weights. Acta Paediatr 1996;85:843-8. Stephansson O, Larsson H, Pedersen L, et al. Congenital abnormalities and other birth outcomes in children born to women with ulcerative colitis in Denmark and Sweden. Inflamm Bowel Dis 2011;17:795-801. Zou GY, Donner A. Extension of the modified Poisson regression model to prospective studies with correlated binary data. Stat Methods Med Res 2013;22:661-70. Ludvigsson JF, Montgomery SM, Ekbom A. Celiac disease and risk of adverse fetal outcome: a population-based cohort study. Gastroenterology 2005;129:454-63. Mozaffari S, Abdolghaffari AH, Nikfar S, et al. Pregnancy outcomes in women with inflammatory bowel disease following exposure to thiopurines and antitumor necrosis factor drugs: a systematic review with meta-analysis. Hum Exp Toxicol 2015;34:445-59. Szklo M, Nieto FJ. Epidemiology. Beyond the basics (page 273). Burlington, MA, USA: Jones & Bartlett Learning, 2014.

SC

14.

ACCEPTED MANUSCRIPT

15

ACCEPTED MANUSCRIPT

APPENDIX Overview Definitions of hepatobiliary diseases, gastrointestinal diseases, congenital malformations and procedures. The appendix also contains 8 eTables meant for online publication only.

AC C

EP

TE D

M AN U

SC

RI PT

eTABLE I. Endoscopy during pregnancy and risk for adverse pregnancy outcome in healthy women (without any diagnosis of inflammatory bowel disease, celiac disease or hepatobiliary disease) eTABLE IIa. Endoscopy during pregnancy and risk for adverse pregnancy outcome in women with inflammatory bowel disease, celiac disease or hepatobiliary disease (combined) eTABLE IIb. Endoscopy during pregnancy and risk for adverse pregnancy outcome in women with inflammatory bowel disease, celiac disease or hepatobiliary disease (separate categories) eTABLE III. All women. Pregnancies exposed to endoscopy compared to [reference] pregnancies where the woman had an endoscopy <1 year before or after pregnancy eTABLE IV. Mothers without celiac disease, IBD or liver disease. Pregnancies exposed to endoscopy compared to [reference] pregnancies where the woman had an endoscopy <1 year before or after pregnancy. eTABLE V. Mothers with celiac disease, IBD or liver disease. Pregnancies exposed to endoscopy compared to [reference] pregnancies where the woman had an endoscopy <1 year before or after pregnancy. eTABLE VI. Sibling comparisons. Risk of adverse pregnancy outcome in the same mother when exposed to endoscopy vs not exposed) eTABLE VII. Risk of adverse pregnancy outcome in women with endoscopy <1 year before or after pregnancy but not during pregnancy vs. women who have no record of endoscopy.

1

ACCEPTED MANUSCRIPT DEFINITIONS Definition of hepatobiliary diseases. Disease Group Infectious hepatitis Hepatitis A Hepatitis B Hepatitis C Bilirubin disorders Hemochromatosis Wilson Portal vein thrombosis Budd-Chiari Liver alcoholic Liver toxic disease

M AN U

SC

277E 275A 275B 452 453A 571A-D 573D

Liver failure Liver chronic hepatitis Liver fibrosis cirrhosis Primary biliary cirrhosis

570 572C 571E 571 571F 571G

EP

TE D

Liver other inflammatory Liver other Liver not classified Gallstone disease Cholecystitis Gallbladder other Cholangitis Primary sclerosing cholangitis Gallbladder biliary pancreas, other Postcholecystectomy syndrome Ikterus Hepatosplenomegaly Imaging Varices Portal Hypertension Ascites The following operation codes indicated liver disease

AC C

ICD-10 B15-19 B15 B16 + B18.0 + B18.1 B17.1 + B18.2 E80 E83.1 E83.0 I81 I82 K70 K71 K70.4 K71.1 K72 K73 K74 K74.3 K75.4 K75.5 K75.9 K76 K77 K80 K81 K82 K83 K83.0A K87 K91.5 R17 R16.0 R93.2 I85 K76.6 R18

RI PT

ICD-9 70

572 573 571W 571X 574 575 576 576B

782E 789B 456A-C 572D 789F

5350-5359 5388 5394 9014

JKA20 JKA21 JKB11 JKB30 JKE00 JKE02 JKE06 JKE12

2

ACCEPTED MANUSCRIPT 5341

JKE15 JKE18 JKF10

Definition of gastrointestinal diseases IBD: (Crohn: ICD-9: 555; ICD-10: K50; Ulcerative colitis: ICD-9: 556; ICD-10: K51) Celiac disease (ICD-9: 579; ICD-10: K90).

M AN U

SC

RI PT

Definition of congenital malformations. For live born children, information on congenital abnormalities diagnosed during the first year of life was collected from the Birth and Patient Registry in Sweden and the Danish National Patient Registry. The codes used to identify congenital abnormalities were 740–759 in ICD-9 and Q00 to Q89 in ICD-10. Diagnoses of congenital dislocation of the hip (754D in ICD-9 and Q65.0–6 in ICD-10) and undescended testis (752F in ICD-9 and Q53 in ICD-10) were not included because of their poor expected validity. Also excluded were preauricular appendage or tag (744B in ICD-9 and Q17.0 in ICD-10), tongue tie (750A in ICD-9 and Q38.1 in ICD-10), congenital nonneoplastic nevus (Q82.5 in ICD-10), chromosomal abnormalities (758 in ICD-9 and Q90–99 in ICD 10), and congenital malformation syndromes due to known exogenous causes (Q86 in ICD-10).

AC C

EP

TE D

Definition of Procedures. Esophagogastroduodenoscopy: 2861, 2880, 2881, 4480, 4483, 4486, 4487, 4488, 4489, 4490, 9021, 4686, 4687, 9003, 9004, 9021, UJC, UJD. UJF02, UJF05. Colonoscopy and sigmoidoscopy (*): 9011, 9012*, 9023, 4685*, 4688, 4689, 4674, 4684, UJF32, UJF35, UJF42*, UJF45*. ERCP: 9014, 5388, 5394, UJK02, UJK05.

3

ACCEPTED MANUSCRIPT

Very preterm Induction of labour No Yes Cesarean section No Yes Stillbirth No Yes

Ref Ref Ref 1.35; 1.21- 1.33; 1.18- 1.23; 0.981.50 1.50 1.55 Ref Ref Ref 1.11; 1.01- 1.06; 0.95- 1.23; 1.021.22 1.18 1.49 Ref Ref Ref 1.14; 0.74- 1.57; 0.78- 0.61; 0.082.72 3.18 4.49 Ref Ref Ref 1.17; 0.79- 1.25; 0.82- 1.34; 0.651.74 1.90 2.76

AC C

Apgar score<7 No

SC

Moderate Preterm

Ref Ref Ref 1.24; 1.04- 1.27; 1.05- 1.10; 0.751.48 1.55 1.61 1.23; 1.00- 1.28; 1.02- 1.03; 0.661.51 1.60 1.61 1.44; 0.88- 1.41; 0.83- 1.64; 0.682.33 2.39 3.98

M AN U

Yes

Mother with Lower ARR, 95%CI

TE D

Any Preterm birth No

Mother with Upper ARR, 95%CI

EP

Characteristics

Mother with any Endoscopy ARR, 95%CI

RI PT

eTABLE I. Endoscopy during pregnancy and risk for adverse pregnancy outcome in healthy women (without any diagnosis of inflammatory bowel disease, celiac disease or hepatobiliary disease)

Yes

Neonatal death No Yes

Small for gestational age No Yes Low birth weight No

Ref Ref Ref 1.77; 0.74- 2.14; 0.92- 1.19; 0.184.26 4.99 7.84

Ref Ref Ref 1.29; 1.01- 1.36; 1.04- 0.85; 0.471.63 1.76 1.53

Ref

Ref

Ref 4

ACCEPTED MANUSCRIPT

Any major congenital malformation No Yes

Birthweek (w) Birthweight (g)

1.20; 0.95- 1.21; 0.93- 1.01; 0.671.52 1.56 1.79

Ref Ref Ref 1.01; 0.80- 1.06; 0.82- 0.96; 0.591.37 1.55 1.28 -0.2; -0.3to -0.1 -50; -76- to -27

-0.2; -0.3 0.0; -0.2 to to -0.1 0.1 -70; -98 to 28; -21 to -42 078

AC C

EP

TE D

M AN U

SC

ARR, Adjusted Relative Risk

RI PT

Yes

5

ACCEPTED MANUSCRIPT

Very preterm Induction of labour No Yes Cesarean section No Yes Stillbirth No Yes

Ref 2.09; 1.672.61 2.13; 1.612.81 3.51; 1.886.53

Ref 1.29; 1.141.46

Ref 1.09; 0.75-1.58

Ref 1.34; 1.111.61

Ref 1.27; 1.151.41

Ref 1.20; 0.94-1.54

Ref 1.16; 1.001.35

Ref

Ref 1.68; 0.2312.44

Ref

1.27; 0.572.83

SC

Ref 2.03; 1.37-2.99 2.16; 1.32-3.54 2.88; 0.90-9.27

M AN U

Moderate Preterm

Ref 1.67; 1.411.99 1.69; 1.372.09 2.19; 1.383.47

TE D

Yes

Mother with Lower ARR, 95%CI

EP

Any Preterm birth No

Mother with Upper ARR, 95%CI

AC C

Characteristics#

Mother with any Endoscopy ARR, 95%CI

RI PT

eTABLE IIa. Endoscopy during pregnancy and risk for adverse pregnancy outcome in women with celiac disease, inflammatory bowel disease (IBD) or hepatobiliary disease (combined)

1.67; 0.535.28

Apgar score<7 6

Neonatal death No Yes Small for gestational age No

Ref 0.99; 0.25-3.99

Ref 1.30; 0.662.55

Ref

Ref 4.88; 0.6436.94

Ref

0.94; 0.233.80

1.90; 0.2315.91

Ref 1.41; 0.67-2.96

Ref 1.41; 0.922.16

Ref 1.73; 1.382.16

Ref 1.81; 1.05-3.10

Ref 2.20; 1.652.95

Ref 0.95; 0.701.30

Ref 0.83; 0.35-1.96

Ref 0.92; 0.601.42

Birthweek (w)

-0.3; -0.5to -0.2

Birthweight (g)

-93; -129to -57

-0.3; -0.7 to 0.0 -132; 232- to 31

Yes Any major congenital malformation No Yes

EP

Low birth weight No

AC C

Yes

TE D

Ref 1.19; 0.861.64

SC

Yes

Ref 1.12; 0.691.83

M AN U

No

RI PT

ACCEPTED MANUSCRIPT

-0.5; -0.7to -0.3

-119; -174to -65

7

ACCEPTED MANUSCRIPT

eTABLE IIb. Endoscopy during pregnancy and risk for adverse pregnancy outcome in women with celiac disease, inflammatory bowel disease (IBD) or hepatobiliary disease (separate categories)

Very preterm Induction of labour No Yes Cesarean section No Yes Stillbirth No Yes

Ref 1.02; 0.48-2.17 1.13; 0.49-2.60

Ref 2.37; 1.005.63 2.87; 1.057.85

NC

NC

NC

Ref 1.13; 0.751.70

Ref 1.00; 0.61-1.64

Ref 1.67; 0.863.22

Ref 0.79; 0.541.17

Ref 0.91; 0.62-1.34

Ref 2.29; 0.3017.66

SC

Ref 1.36; 0.762.43 1.56; 0.813.00

Mother Mother with any with Upper Endoscopy ARR, ARR, 95%CI 95%CI

M AN U

Moderate Preterm

Mother with Lower ARR, 95%CI

Mother with Lower ARR, 95%CI

Mother Mother with any with Upper Endoscopy ARR, ARR, 95%CI 95%CI

Mother with Lower ARR, 95%CI

Ref 2.19; 1.772.70 2.23; 1.712.91 3.84; 2.156.87

Ref 3.35; 2.195.12 3.79; 2.056.99 7.47; 2.2325.07

Ref 2.09; 1.672.61 2.13; 1.612.81 3.51; 1.886.53

Ref 1.10; 0.791.51 1.06; 0.731.55 1.37; 0.613.07

Ref 1.16; 0.811.67 1.22;0.801.85 0.93; 0.302.90

Ref 1.21; 0.622.32 1.05; 0.472.36 2.37; 0.589.66

Ref 1.34; 1.121.60

Ref 1.38; 0.832.32

Ref 1.34; 1.111.61

Ref 1.34; 1.131.60

Ref 1.30; 1.061.60

Ref 1.63; 1.162.28

Ref 0.51; 0.171.53

Ref 1.20; 1.041.38

Ref 1.62; 1.192.22

Ref 1.16; 1.001.35

Ref 1.23; 1.061.44

Ref 1.19; 1.001.42

Ref 1.39; 0.981.97

Ref 2.88; 0.36-

Ref

Ref 2.09; 0.755.82

Ref 4.43; 0.5933.04

Ref 1.67; 0.535.28

Ref 1.23; 0.403.80

Ref 1.65; 0.545.09

Ref

TE D

Yes

Mother with Upper ARR, 95%CI

EP

Any Preterm birth No

Mother with any Endoscopy ARR, 95%CI

Hepatobiliary disease

AC C

Characteristics#

Inflammatory bowel disease

RI PT

Celiac disease

NC

NC

8

ACCEPTED MANUSCRIPT

22.88

Yes Low birth weight No Yes Any major congenital malformation No Yes

Birthweek (w) Birthweight (g)

Ref 1.30; 0.662.55

Ref 0.90; 0.412.01

Ref 1.02; 0.432.47

Ref 0.76; 0.115.35

Ref

Ref 9.46; 0.89101.12

Ref

Ref

Ref

Ref

Ref

Ref

Ref

NC

1.71; 0.2014.21

NC

1.90; 0.2315.91

NC

NC

NC

Ref 0.26; 0.041.84

Ref 0.34; 0.05-2.41

Ref

Ref 0.65; 0.221.95

Ref 0.84; 0.28-2.52

Ref

Ref 0.48; 0.121.90

Ref 0.64; 0.16-2.54

0.0;-0.3 to 0.3 -42; -129 to 46

0.1; -0.2 to 0.5 -42; -148 to 63

7.43: 0.7573.65

Ref 1.39; 0.922.10

Ref 2.69; 1.196.08

Ref 1.41; 0.922.16

Ref 1.05; 0.631.75

Ref 1.33; 0.78.2.26

Ref 0.74; 0.192.88

Ref 2.23; 1.692.94

Ref 2.97; 1.615.47

Ref 2.20; 1.652.95

Ref 1.15; 0.761.75

Ref 1.36; 0.872.14

Ref 1.06; 0.402.79

Ref 0.93; 0.621.41

Ref 1.11; 0.373.36

Ref 0.92; 0.601.42

Ref 0.83; 0.511.38

Ref 0.85; 0.481.53

Ref 0.26; 0.041.81

-0.6; -0.7 to -0.4 -128; -181 to 75

-1.0; -1.6 to -0.4 -267; -445 to -90

-0.5; -0.7 to -0.3 -119; -174 to -65

-0.2; -03 to 0.0 -44; -95 to 8

-0.1; -0.3 to 0.1 -41; -103 to 21

-0.4; -0.8 to 0.0 -58; -169 to 53

NC

NC

Ref NC

-0.4; -1.0 to 0.2 -46; -185 to 93

RI PT

Ref 1.26; 0.189.01

SC

Small for gestational age No

Ref 1.34; 0.712.55

M AN U

Yes

Ref 2.70; 0.4416.64

TE D

Neonatal death No

Ref 0.82; 0.11-5.93

EP

Yes

Ref 1.29; 0.325.15

AC C

Apgar score<7 No

9

ACCEPTED MANUSCRIPT

ARR, Adjusted Relative Risk. NC, Not calculated due to lack of data

AC C

EP

TE D

M AN U

SC

RI PT

*Includes ERCP but also endoscopies where lower/upper could not be distinguished #Upper endoscopy comparisons restricted to women with a lifetime diagnosis of celiac disease. Lower endoscopy comparisons restricted to women with a lifetime diagnosis of IBD. Any endoscopy comparisons restricted to women with celiac disease, IBD or hepatobiliary disease (see text for definitions).

10

ACCEPTED MANUSCRIPT

Very preterm Induction of labour No Yes Cesarean section No Yes Stillbirth No Yes

Ref Ref Ref 1.11; 1.02- 1.05; 0.95- 1.17; 1.021.22 1.17 1.34 Ref Ref Ref 1.02; 0.94- 0.93; 0.85- 1.18; 1.061.10 1.02 1.33 Ref Ref Ref 1.11; 0.63- 1.15; 0.61- 0.92; 0.341.94 2.15 2.52

AC C

Apgar score<7 No

SC

Moderate Preterm

Ref Ref Ref 1.16; 1.01- 1.00; 0.84- 1.50; 1.241.34 1.19 1.83 1.13; 0.96- 0.99; 0.81- 1.44; 1.141.33 1.21 1.83 1.52; 1.03- 1.06; 0.65- 2.43; 1.502.23 1.73 3.96

M AN U

Yes

Mother with Lower ARR, 95%CI

TE D

Any Preterm birth No

Mother with Upper ARR, 95%CI

EP

Characteristics

Mother with any Endoscopy ARR, 95%CI

RI PT

eTABLE III. All women. Pregnancies exposed to endoscopy compared to [reference] pregnancies where the woman had an endoscopy <1 year before or after pregnancy

Yes

Neonatal death No Yes Small for gestational age No Yes

Ref Ref Ref 1.06; 0.75- 1.05; 0.70- 1.31; 0.801.49 1.58 2.15 Ref Ref Ref 1.08; 0.49- 1.46; 0.60- 0.79; 0.182.56 3.58 3.43

Ref Ref Ref 1.19; 0.96- 1.23; 0.96- 1.12; 0.801.47 1.58 1.58

Low birth weight 11

ACCEPTED MANUSCRIPT

Any major congenital malformation No Yes

Birthweek (w) Birthweight (g)

Ref Ref Ref 0.90; 0.73- 0.92; 0.72- 0.88; 0.641.17 1.21 1.10 -0.1; -0.2- 0.0; -0.1 to 0.1 to 0.0 -34; -57- to -28; -54 to -11 -1

-0.2; -0.4 to -0.1 -37; -74 to 0

RI PT

Yes

Ref Ref Ref 1.18; 0.98- 0.99; 0.79- 1.56; 1.211.24 2.00 1.41

SC

No

AC C

EP

TE D

M AN U

ARR, Adjusted Relative Risk

12

ACCEPTED MANUSCRIPT

Very preterm Induction of labour No Yes Cesarean section No Yes Stillbirth No Yes

Ref Ref Ref 1.07; 0.95- 1.06; 0.93- 1.01; 0.811.20 1.21 1.28 Ref Ref Ref 0.97; 0.87- 0.93; 0.83- 1.11; 0.911.08 1.05 1.34 Ref Ref Ref 1.31; 0.63- 1.42; 0.67- 0.61; 0.082.69 3.02 4.65 Ref Ref Ref 1.10; 0.71- 1.21; 0.76- 1.30; 0.621.71 1.93 2.76

AC C

Apgar score<7 No

SC

Moderate Preterm

Ref Ref Ref 1.03; 0.84- 1.06; 0.86- 1.00; 0.681.32 1.48 1.27 1.01; 0.80- 1.05; 0.82- 0.93; 0.581.28 1.35 1.48 1.19; 0.68- 1.20; 0.66- 1.46; 0.582.09 2.18 3.70

M AN U

Yes

Mother with Lower ARR, 95%CI

TE D

Any Preterm birth No

Mother with Upper ARR, 95%CI

EP

Characteristics

Mother with any Endoscopy ARR, 95%CI

RI PT

eTABLE IV. Mothers without celiac disease, IBD or liver disease. Pregnancies exposed to endoscopy compared to [reference] pregnancies where the woman had an endoscopy <1 year before or after pregnancy.

Yes

Neonatal death No Yes Small for gestational age No Yes

Ref Ref Ref 1.17; 0.44- 1.57; 0.59- 0.81; 0.113.14 4.14 5.82

Ref Ref Ref 1.16; 0.88- 1.22; 0.91- 0.79; 0.441.53 1.63 1.44

Low birth weight 13

ACCEPTED MANUSCRIPT

Any major congenital malformation No Yes

Birthweek (w) Birthweight (g)

Ref Ref Ref 0.92; 0.71- 0.98; 0.74- 0.87; 0.531.29 1.42 1.19 0.0; -0.1 to 0.1 -31; -59 to 3

-0.1; -0.2 0.1; -0.1 to to 0.1 0.2 -49; -79 to 39; -12 to -18 90

RI PT

Yes

Ref Ref Ref 0.97; 0.75- 0.98; 0.74- 0.95; 0.581.29 1.56 1.26

SC

No

AC C

EP

TE D

M AN U

ARR, Adjusted Relative Risk

14

ACCEPTED MANUSCRIPT

Very preterm Induction of labour No Yes Cesarean section No Yes Stillbirth No Yes

Ref 1.18; 1.031.35

Ref Ref 0.99; 1.30; 1.050.68-1.46 1.59

Ref 1.08; 0.961.21

Ref Ref 1.08; 1.04; 0.890.83-1.41 1.22

Ref

Ref Ref 0.90; 1.10; 0.320.083.84 11.37

1.05; 0.442.48

AC C

Apgar score<7 No

Ref Ref 1.43; 1.61; 1.250.95-2.15 2.06 1.47; 1.61; 1.180.87-2.45 2.19 2.03; 2.58; 1.240.57-7.15 5.37

Yes

Neonatal death No Yes Small for gestational age No Yes

SC

Moderate Preterm

Ref 1.33; 1.091.61 1.30; 1.031.64 1.99; 1.173.38

M AN U

Yes

Mother with Lower ARR, 95%CI

TE D

Any Preterm birth No

Mother with Upper ARR, 95%CI

EP

Characteristics#

Mother with any Endoscopy ARR, 95%CI

RI PT

eTABLE V. Mothers with celiac disease, IBD or liver disease. Pregnancies exposed to endoscopy compared to [reference] pregnancies where the woman had an endoscopy <1 year before or after pregnancy.

Ref 0.98; 0.581.67 Ref 1.09; 0.215.65

Ref 1.26; 0.881.79

Ref Ref 1.07; 1.37; 0.640.26-4.48 2.92 Ref

Ref 0.93; 0.09NC 10.13

Ref Ref 1.38; 1.49; 0.930.61-3.11 2.40

Low birth weight 15

No Yes Any major congenital malformation No Yes

Birthweek (w)

Ref Ref 1.37; 1.87; 1.350.78-2.41 2.60

Ref 0.88; 0.631.23

Ref Ref 0.97; 0.96; 0.610.41-2.32 1.52

-0.2; -0.3 to 0.0 -43; -82 to 5

-0.1; -0.4 to 0.3 -70; -174 to 33

-0.3; -06 to -0.1 -76; -134 to -18

SC

Birthweight (g)

Ref 1.50; 1.161.93

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

ARR, Adjusted Relative Risk NC, not calculated due to lack of data. ##Upper endoscopy comparisons restricted to women with a lifetime diagnosis of celiac disease. Lower endoscopy comparisons restricted to women with a lifetime diagnosis of IBD. Any endoscopy comparisons restricted to women with celiac disease, IBD or hepatobiliary disease (see text for definitions).

16

ACCEPTED MANUSCRIPT

Yes Cesarean section No Yes Stillbirth No Yes Apgar score<7 No Yes

Ref Ref 1.13; 0.96- 1.04; 0.831.33 1.31

Ref 1.10; 1.001.21

Ref Ref 1.04; 0.93- 1.18; 1.011.17 1.39

Ref 1.04; 0.502.18

Ref Ref 1.12; 0.46- 0.82; 0.232.72 2.97

Ref 0.98; 0.561.72

Ref Ref 1.15; 0.60- 0.96; 0.422.21 2.21

Ref 0.90; 0.352.33

Ref Ref 1.53; 0.49- 0.70; 0.154.78 3.31

Ref 1.03; 0.731.45

Ref Ref 1.24; 0.82- 0.88; 0.501.89 1.54

Ref 1.45; 1.052.00

Ref Ref 1.21; 0.79- 1.96; 1.221.84 3.15

AC C

Neonatal death No

Ref 1.13; 0.991.29

Yes

Small for gestational age No Yes Low birth weight No Yes

SC

Induction of labour No

Ref Ref 1.05; 0.78- 1.65; 1.161.40 2.35

M AN U

Yes

Ref 1.32; 1.051.64

TE D

Any Preterm birth No

Mother with Lower ARR, 95%CI

EP

Characteristics

Mother with any Mother Endoscop with Upper y ARR, ARR, 95%CI 95%CI

RI PT

eTABLE VI. Sibling comparisons. Risk of adverse pregnancy outcome in the same mother when exposed to endoscopy vs not exposed)

Any major 17

congenital malformation No Yes

Birthweek (w) Birthweight (g)

Ref 0.81; 0.581.13

Ref Ref 0.75; 0.50- 0.87; 0.501.13 1.54

-0.2; -0.3 to -0.1 -41; -65 to -16

0.0; -0.1 to 0.1 -20; -49 to 10

-0.4; -0.5 to -0.2 -64; -107 to -20

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

We did not examine the risk of moderate and very preterm effect since multinomial regressions cannot be calculated using fixed effects. Therefore we only present data on “preterm birth” (yes/no).

18

ACCEPTED MANUSCRIPT eTABLE VII. Risk of adverse pregnancy outcome in women with endoscopy <1 year before or after pregnancy but not during pregnancy vs. women who have no record of endoscopy. Mother with any Endoscop y ARR, 95%CI

Any Preterm birth No

Very Preterm

1.44; 1.38- 1.27; 1.22- 1.48; 1.41- 1.28; 1.22- 1.41; 1.32- 1.20; 1.121.50 1.32 1.56 1.35 1.51 1.28

Yes

1.47; 1.14- 1.35; 1.05- 1.44; 1.04- 1.21; 0.87- 1.55; 1.04- 1.48; 0.981.90 1.74 1.99 1.69 2.30 2.22

1.24; 1.07- 1.16; 1.00- 1.33; 1.11- 1.24; 1.03- 1.05; 0.81- 0.97; 0.751.44 1.34 1.59 1.49 1.35 1.26

AC C

Apgar score<7 No

1.39; 1.34- 1.25; 1.21- 1.30; 1.25- 1.12; 1.08- 1.59; 1.51- 1.40; 1.331.44 1.29 1.36 1.18 1.67 1.47

EP

Stillbirth No

TE D

Cesarean section No Yes

Mother with Lower ARR, 95%CI

1.38; 1.30- 1.35; 1.27- 1.32; 1.22- 1.21; 1.11- 1.55; 1.41- 1.49; 1.361.47 1.44 1.43 1.31 1.70 1.65 1.42; 1.32- 1.39; 1.30- 1.34; 1.22- 1.23; 1.12- 1.64; 1.48- 1.59; 1.421.52 1.50 1.48 1.35 1.83 1.77 1.34; 1.12- 1.28; 1.06- 1.32; 1.04- 1.21; 0.95- 1.24; 0.91- 1.20; 0.871.62 1.55 1.68 1.55 1.69 1.65

Induction of labour No Yes

Mother with Lower RR (crude), 95%CI

SC

Moderate preterm

Mother with Upper ARR, 95%CI

M AN U

Yes

Mother with Upper RR (crude), 95%CI

RI PT

Characteristics

Mother with any Endoscopy RR (crude), 95%CI

Yes

Neonatal death No Yes

1.19; 0.81- 1.30; 0.89- 1.30; 0.82- 1.37; 0.85- 1.02; 0.53- 1.16; 0.591.73 1.91 2.06 2.27 1.96 2.27

Small for gestational age No Yes

1.09; 0.99- 1.09; 0.98- 1.19; 1.05- 1.14; 1.01- 0.98; 0.82- 1.00; 0.841.21 1.20 1.34 1.29 1.16 1.19

Low birth weight No 19

ACCEPTED MANUSCRIPT Yes

1.34; 1.23- 1.30; 1.20- 1.38; 1.24- 1.26; 1.13- 1.40; 1.23- 1.34; 1.171.46 1.42 1.53 1.40 1.59 1.53

Any major congenital malformation No

Birthweek (w)

-0.2; -0.3 to -0.2 -46; -56 to 37

-0.2; -0.2 to -0.2 -42; -52 to -33

-0.2; -0.3 to -0.2 -57; -69 to -44

-0.1; -0.2 to -0.1 -40; -52 to -27

-0.3; -0.4 to -0.3 -47; -63 to -31

-0.3; -0.3 to -0.2 -39; -55 to -24

AC C

EP

TE D

M AN U

SC

Birthweight (g)

1.08; 0.99- 1.09; 1.00- 1.06; 0.96- 1.07; 0.96- 1.06; 0.93- 1.05; 0.921.18 1.18 1.20 1.21 1.21 1.17

RI PT

Yes

20

Ludvigsson et al.

ACCEPTED MANUSCRIPT

Upper endoscopy (1.2)

Preterm birth Any endoscopy (1.4)

Upper endoscopy (1.5)

Lower endoscopy (1.3)

Lower endoscopy (1.6)

EP

TE D

M AN U

SC

RI PT

Stillbirth Any endoscopy (1.1)

Upper endoscopy (1.8)

Lower endoscopy (1.9)

AC C

Small for gestational age Any endoscopy (1.7)

1

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

STROBE Statement—Checklist of items: ACCEPTED MANUSCRIPT Pregnancy outcome in women undergoing endoscopy – A nationwide cohort study (Ludvigsson et al)

Title and abstract

Item No 1

Recommendation (a) Indicate the study’s design with a commonly used term in the title

RI PT

In the title we have indicated that the study was a cohort study. It is clear from the abstract that this was a cohort study.

(b) Provide in the abstract an informative and balanced summary of what was done and what was found 2

Explain the scientific background and rationale for the investigation being reported

Objectives

3

State specific objectives, including any prespecified hypotheses

SC

Introduction Background/rationale

comment: We have expressed our objective in the last paragraph of the introduction, and our hypothesis. 4

Present key elements of study design early in the paper

Setting

5

Describe the setting, locations, and relevant dates, including periods of recruitment,

M AN U

Methods Study design

exposure, follow-up, and data collection Participants

6

(a) Give the eligibility criteria, and the sources and methods of selection of participants. Describe methods of follow-up

(b) For matched studies, give matching criteria and number of exposed and Variables

7

TE D

unexposed

Clearly define all outcomes, exposures, predictors, potential confounders, and effect modifiers. Give diagnostic criteria, if applicable

Data sources/

8*

measurement

For each variable of interest, give sources of data and details of methods of assessment (measurement). Describe comparability of assessment methods if there is

EP

more than one group Bias Study size

9

Describe any efforts to address potential sources of bias

10

Explain how the study size was arrived at

AC C

comment: This is an observational study, where we included all available pregnancies in Sweden 1992-2011 (n=1.6 million). We did not carry out any a priori power analysis, but if the editor requests so we can carry out a post-hoc power analysis.

Quantitative variables

11

Explain how quantitative variables were handled in the analyses. If applicable, describe which groupings were chosen and why

Statistical methods

12

(a) Describe all statistical methods, including those used to control for confounding (b) Describe any methods used to examine subgroups and interactions (c) Explain how missing data were addressed (d) If applicable, explain how loss to follow-up was addressed (e) Describe any sensitivity analyses

Results Participants

13*

(a) Report numbers of individuals at each stage of study—eg numbers potentially eligible, examined for eligibility, confirmed eligible, included in the study, completing follow-up, and analysed (b) Give reasons for non-participation at each stage 1

comment: This was a registry-based study and according to the decision of the ethics ACCEPTED MANUSCRIPT review board of the Karolinska Institutet individual study participants were not contacted (all data were analysed without knowledge of the identity of the study participants), hence we had no “non-participation” at different stages of the study. (c) Consider use of a flow diagram Descriptive data

14*

(a) Give characteristics of study participants (eg demographic, clinical, social) and information on exposures and potential confounders (b) Indicate number of participants with missing data for each variable of interest (c) Summarise follow-up time (eg, average and total amount)

15*

Report numbers of outcome events or summary measures over time

Main results

16

(a) Give unadjusted estimates and, if applicable, confounder-adjusted estimates and

RI PT

Outcome data

their precision (eg, 95% confidence interval). Make clear which confounders were adjusted for and why they were included

(b) Report category boundaries when continuous variables were categorized (c) If relevant, consider translating estimates of relative risk into absolute risk for a Other analyses

17

SC

meaningful time period

Report other analyses done—eg analyses of subgroups and interactions, and sensitivity analyses

18

Summarise key results with reference to study objectives

Limitations

19

Discuss limitations of the study, taking into account sources of potential bias or

M AN U

Discussion Key results

imprecision. Discuss both direction and magnitude of any potential bias Interpretation

20

Give a cautious overall interpretation of results considering objectives, limitations, multiplicity of analyses, results from similar studies, and other relevant evidence

21

Discuss the generalisability (external validity) of the study results

Other information Funding

22

Give the source of funding and the role of the funders for the present study and, if

TE D

Generalisability

applicable, for the original study on which the present article is based

EP

*Give information separately for exposed and unexposed groups. Note: An Explanation and Elaboration article discusses each checklist item and gives methodological background and published examples of transparent reporting. The STROBE checklist is best used in conjunction with this article (freely

AC C

available on the Web sites of PLoS Medicine at http://www.plosmedicine.org/, Annals of Internal Medicine at http://www.annals.org/, and Epidemiology at http://www.epidem.com/). Information on the STROBE Initiative is available at http://www.strobe-statement.org.

2