Coelho J, Beaugerie L, Colombel JF, et al. Pregnancy outcome in patients with inflammatory bowel disease treated with thiopurines: cohort from the CESAME Study. Gut. 2011;60:198–203.

The study by Coelho et al recently published in Gut1 aimed at evaluating pregnancy outcomes in female IBD patients treated with thiopurines. Patients were part of the CESAME study of 19,486 inflammatory bowel disease (IBD) patients, 11,006 of whom were women, an observational cohort prospectively followed up between 2004–2007 by 680 gastroenterologists in France.2 CESAME's original aim was to assess the risk for malignancies occurring in patients taking thiopurines. In 2005 a substudy assessing pregnancy outcomes was added as well. The study by Coelho et al reports on 86 pregnancies that occurred in thiopurine-treated (group A), compared to 129 pregnancies in IBD control patients. The main finding was that there was no increase in the risks of congenital abnormalities (two in the thiopurine-exposed vs. four in the controls).


  1. Top of page

An IBD diagnosis usually overlaps with the reproductive years. Thus, fertility and pregnancy are issues occupying patients and caregivers. A major question asked is how IBD therapy would affect pregnancy and its outcomes. In that regard, the effect of thiopurines is specifically important. While 5-aminosalicylate (5-ASA) drugs are considered safe and steroids were not associated with major congenital anomalies but with a risk for cleft lip/palate,3 thiopurines, which are the mainstay of IBD (specifically Crohn's disease [CD]) maintenance therapy, including during pregnancy, are defined as pregnancy category D and were associated with negative pregnancy outcomes including congenital anomalies by several authors.

In that sense, Coelho et al are adding a brick to one of the most important foundations of IBD treatment during pregnancy, by reporting that no significant adverse events occur in thiopurine-exposed pregnancies, and that there is no increase in the rate of congenital anomalies. Moreover, as the CESAME cohort is a large, population-based study including 11,006 women, the brick seems to be even more solid.

However, while searching for further supportive data for thiopurine use during pregnancy, one should acknowledge several points that limit the reassurance that seems to be emerging from this study.

A major point is that while this study is a part of the CESAME, it is actually a very small part of it, with a total of only 215 pregnancies that were included, representing 1.79% of the women followed up in the study. Thus, the powerful sample size is significantly decreased. This relates to the design of this substudy, added late as an amendment, while CESAME was already designed and ongoing.

The original aim of CESAME was to assess cancer risk, not pregnancy outcomes. Thus, information regarding fundamental factors that may have affected pregnancy outcomes is missing. Specifically important factors are disease activity that has been previously associated with preterm deliveries and small for gestational age babies, and concomitant (non-IBD) diseases and medications. Regarding thiopurine treatment itself, it is not reported how long prior to pregnancy the patients were treated, or did thiopurine therapy start during pregnancy, what was the duration of therapy, and what was the average dose. The effect of thiopurines in pregnancy may depend on the dosage, and, more importantly, on the duration and timing of exposure.

While the Coelho et al study aimed at assessing pregnancy results in thiopurine-treated women, only group A consisted of women exposed to thiopurines—this group included 86 pregnancies, and only 57 of them were exposed to thiopurines alone. This number is too low for comparison of the congenital anomalies rate between 2 (or 3) groups, due to the expected rarity of the investigated event. The low numbers may also be the reason for the nonsignificance in prematurity rates and rates of birth weights lower than 2500 g that seemed to be higher in group A (thiopurines) compared to group C (no treatment). Noticeably, the numbers of thiopurine-exposed pregnant IBD patients in the reports calling attention to congenital anomalies are usually small. For example, Norgard et al4 found in a Danish population-based study that significantly more congenital anomalies occurred in CD patients exposed to thiopurines during pregnancy compared to unexposed CD patients, with an odds ratio (OR) of 2.9 (95% confidence interval [CI] 0.9–8.9); however, only 26 women were exposed to thiopurines during pregnancy. The numbers in the Coelho et al study are double that.

The low representation of ulcerative colitis (UC) patients, only 13 in group A and approximately a fifth in the whole cohort, is another “numbers issue.” This CD:UC ratio is not intuitive and may have a significant inherent bias. The potential importance of underrepresentation of UC pregnancies is that UC per se was reported by some to be associated with congenital anomalies.5 A cohort like CESAME could have delivered reassuring information to UC mothers, with or without thiopurine treatment.

In conclusion, despite the limitations raised, the study by Coelho et al adds reassurance to physicians using thiopurines for the treatment of pregnant IBD patients. Interestingly, IBD-oriented gastroenterologists already felt reassured: the ECCO guidelines list thiopurines as safe during pregnancy6 and in a recently published survey7 89% of IBD experts reported on continuing azathioprine in their pregnant CD patients. Thus, the (additional) evidence supports the medical practice. As long as these are the data we have, thiopurines should be continued in pregnant IBD patients.


  1. Top of page
  • 1
    Coelho J, Beaugerie L, Colombel JF, et al. Pregnancy outcome in patients with inflammatory bowel disease treated with thiopurines: cohort from the CESAME Study. Gut. 2011; 60: 198203.
  • 2
    Beaugerie L, Brousse N, Bouvier AM, et al. CESAME Study Group. Lymphoproliferative disorders in patients receiving thiopurines for inflammatory bowel disease: a prospective observational cohort study. Lancet. 2009; 374: 16171625.
  • 3
    Park-Wyllie L, Mazzotta P, Pastuszak A, et al. Birth defects after maternal exposure to corticosteroids: prospective cohort study and meta-analysis of epidemiological studies. Teratology. 2000; 62: 385392.
  • 4
    Norgard B, Pedersen L, Christensen LA, et al. Therapeutic drug use in women with Crohn's disease and birth outcomes: a Danish nationwide cohort study. Am J Gastroenterol. 2007; 102: 14061413.
    Direct Link:
  • 5
    Dominitz JA, Young JC, Boyko EJ. Outcomes of infants born to mothers with inflammatory bowel disease: a population-based cohort study. Am J Gastroenterol. 2002; 97: 641648.
    Direct Link:
  • 6
    van der Woude CJ, Kolacek S, Dotan I, et al. European Crohn's Colitis Organisation (ECCO). J Crohns Colitis. 2010; 4: 493510.
  • 7
    Peyrin-Biroulet L, Oussalah A, Roblin X, Sparrow MP. The use of azathioprine in Crohn's disease during pregnancy and in the post-operative setting: a worldwide survey of experts. Aliment Pharmacol Ther. 2011; 33: 707713.