INTRODUCTION — Several questions arise in the care of patients with inflammatory bowel disease (IBD) who want to have children. The effect of the disease and its treatment on fertility and pregnancy and the safety of medications or diagnostic tests for pregnant or nursing women with IBD will be reviewed here. The medical management of IBD is discussed in detail separately. (See appropriate topic reviews.)
FERTILITY — Most studies have demonstrated that men and women with IBD disease do not have decreased fertility compared to the general population. The reduction in birth rates among women with IBD observed in some studies is probably more an effect of patient choice rather than a consequence of disease-related infertility [1]. Sexual dysfunction in men and women with IBD appears largely attributable to depression [2]. However, infertility in patients with IBD may be observed in some circumstances:
GENETIC COUNSELING — Genetic factors play a role in the development of IBD and should be mentioned during preconceptual counseling. (See "Epidemiology and environmental factors in inflammatory bowel disease in adults".)
PREGNANCY AND OBSTETRICAL COMPLICATIONS — Multiple studies have described pregnancy and obstetrical outcomes in women with IBD. These data suggest that women with IBD are at somewhat increased risk for worse obstetrical and pregnancy-related outcomes [14-16]. However, the magnitude of risk for individual patients is related to disease severity before and during pregnancy and probably the type of treatment (described below).
Ulcerative colitis — There are two main questions that need to be addressed in the pregnant woman with ulcerative colitis:
Effect on ulcerative colitis — The course of the ulcerative colitis during pregnancy appears to be determined in part by the activity of the disease at conception. Patients in remission at the time of conception are likely to remain in remission during pregnancy. Approximately one-third of women relapse during the pregnancy (more commonly during the first trimester), which is similar to the rate observed over a nine month period in nonpregnant women [17]. Furthermore, remission achieved during pregnancy is likely to be sustained throughout the remainder of the pregnancy.
In contrast, disease that is active at the time of conception is likely to remain active during pregnancy and may worsen [18]. As a result, women with ulcerative colitis who desire pregnancy should try to conceive at a time when the disease is in remission.
Complications occurring during pregnancy in women whose disease is active are similar to those in nonpregnant patients with ulcerative colitis. (See "Clinical manifestations, diagnosis, and prognosis of ulcerative colitis in adults".) Fulminant colitis can be treated with colectomy with subsequent delivery of a healthy infant [19,20]. However, surgery may be associated with premature labor or spontaneous abortion [21].
The course of ulcerative colitis in the postpartum period is no different from that at other times. Furthermore, the events occurring during one pregnancy do not correlate with the course in subsequent pregnancies [22]. One study suggested that relapses in the three years following pregnancy may be less frequent than in the three years prior to pregnancy [23].
Effect on the pregnancy — In many series, the rate of delivery of healthy offspring in women with ulcerative colitis has been similar to the general population [17,24,25]. However, discordant data have also been reported [1,26,27]. In one study, for example, a significantly greater incidence of low birth weight infants (less than 2500 g) was observed in mothers with ulcerative colitis than in a control population in a large maternity facility [26]. Another population-based study that included 756 women with IBD (both ulcerative colitis and Crohn's disease) found an increased incidence of preterm births and low birth weight compared to the general population [27].
Whether the degree of disease activity accounts in part for the differences among studies is not known. This was indirectly suggested by a Danish cohort study in which the risk of preterm birth was increased when birth occurred after the mother's first hospitalization (odds ratio 1.4), particularly when the first hospitalization for ulcerative colitis took place during pregnancy (odds ratio 3.4) [28]. The risk of congenital abnormalities or stillbirth did not appear to be increased. In another study, severe relapses of colitis during pregnancy were associated with an increased risk of preterm birth and low birth weight [29].
Crohn's disease — The same issues, effect on disease activity and effect on the pregnancy, are important in Crohn's disease.
Effect on Crohn's disease — Similar to ulcerative colitis, Crohn's disease that is active at the time of conception is likely to remain active, and, conversely, disease that is quiescent is likely to remain quiescent [30]. One study suggested that Crohn's disease activity was mildly but significantly lower during pregnancy compared with the year prior to pregnancy [31]. Complications related to Crohn's disease occurring during pregnancy are similar to those seen in nonpregnant patients. (See "Clinical manifestations, diagnosis and prognosis of Crohn's disease in adults".) As an example, fistulas may occur, and a case report documented the occurrence of an enterouterine fistula [32]. In addition, Crohn's disease may rarely present during pregnancy with intraperitoneal sepsis [33].
A concern in patients with active perianal disease has been that vaginal delivery could worsen disease activity. In one series that included five patients with perianal disease (one of whom had a vaginal delivery), vaginal delivery did not exacerbate perianal symptoms, and prophylactic cesarean section in three patients did not avert the subsequent development or recurrence of perianal disease [30]. In another small series, the outcome after vaginal delivery and episiotomy was variable [34]. Considering the available data and clinical experience many obstetricians avoid creating an episiotomy and prefer cesarean section in patients with active perianal disease because of the concern that a fistula will arise from the incision. In patients with quiescent disease, the mode of delivery should be dictated by obstetrical concerns.
Similar to ulcerative colitis, the outcome of pregnancy in patients with Crohn's disease does not predict events during subsequent pregnancies. In one series, relapses in the three years following pregnancy were less frequent than in the three years prior to pregnancy [23].
Effect on the pregnancy — Women with Crohn's disease are at increased risk for low birth weight infants and premature delivery [14,35], but not birth defects [36]. A study based upon a Danish registry of patients and birth records compared the outcomes of 510 newborns to mothers with Crohn's disease with 3018 controls [37]. The average birth weight to mothers with Crohn's disease was significantly lower than for controls (3184 versus 3368 grams). Furthermore, significantly more infants of mothers with Crohn's disease weighed less than 2500 g (10.4 versus 4.7 percent), and were born prior to 37 weeks of gestation (7.3 versus 4.8 percent). The effect of Crohn's disease on birth weight in these studies is similar to that observed in children of mothers who smoke moderately [38].
However, discordant data have also been reported. In a prospective, case-control, European multicenter study of 332 pregnant women with IBD, women with Crohn’s disease or ulcerative colitis had similar pregnancy outcomes (abortions, preterm deliveries, Cesarean sections, congenital abnormalities, and birth weight) compared with non-IBD controls. The favorable outcomes may be attributed to the fact that 86 to 88 percent of patients in this study had quiescent disease at conception/first trimester [24].
Although not specifically addressed in the Danish registry report, other studies have also found that the risk of an abnormal pregnancy outcome in women with Crohn's disease is greatest in those who have active disease at the time of conception, in whom remission may be difficult to achieve during pregnancy [25,39-42].
The significance of these observations for the subsequent health of babies born to women with Crohn's disease is uncertain. Although reduced birth weight is statistically related to subsequent cardiovascular disease, hypertension, and diabetes in adult life (see "Possible role of low birth weight in the pathogenesis of essential hypertension"), the magnitude of the risk in mothers with Crohn's disease may not warrant a change in obstetrical management [43].
However, careful monitoring of fetal growth is advisable in women with Crohn's disease who have become pregnant, particularly if the disease is active. Furthermore, conception should be attempted at a time when the disease is in remission if possible.
SAFETY OF EVALUATION DURING PREGNANCY — Flexible sigmoidoscopy is safe during pregnancy [44]. Colonoscopy has been performed during pregnancy without complications [44,45]. However, experience is relatively limited compared with sigmoidoscopy, and we feel that colonoscopy should generally be avoided unless the information is critical for making treatment decisions. Similarly, radiographic studies involving ionizing radiation should be avoided unless medically necessary. (See "Diagnostic imaging procedures during pregnancy".)
SAFETY OF DRUGS DURING PREGNANCY AND NURSING — The choice of drugs used during pregnancy and lactation should be based upon their relative safety and indications.
Sulfasalazine — Several decades of experience with sulfasalazine suggest that it is safe during pregnancy. Its safety was illustrated in one report, in which no increase in fetal morbidity or mortality was observed among 287 pregnancies of women who had received sulfasalazine compared to 244 pregnancies in women with IBD who had not received sulfasalazine [46].
Sulfasalazine and its sulfapyridine metabolite can be found in umbilical cord blood at similar concentrations to maternal blood. However, these concentrations do not cause significant displacement of bilirubin from albumin [47,48]. As a result, sulfasalazine (unlike sulfonamides) does not have to be stopped during breast feeding to prevent kernicterus.
Antibiotics — Antibiotics are frequently used in the management of patients with Crohn's disease and may also have a role in patients with ulcerative colitis. Metronidazole and ciprofloxacin have been most commonly used. (See "Antibiotics for treatment of inflammatory bowel diseases".)
A number of studies have demonstrated that metronidazole is not associated with an increased risk of birth defects or cancer in humans [49,50]. However, long-term use in pregnancy remains controversial because it is carcinogenic in rodents and mutagenic in bacteria [51,52]. Short-term courses are commonly given. Metronidazole is excreted into breast milk, and the American Academy of Pediatrics classifies the drug as of concern in nursing women [53].
In contrast, ciprofloxacin affects growing cartilage in animals and humans and can cause arthropathy. As a result, it is not recommended for children under age 18 or for pregnant women. However, adverse effects have not been described in the few reports of ciprofloxacin use in pregnancy [54]. Ciprofloxacin is excreted into breast milk, but short courses of ciprofloxacin are probably safe during breastfeeding. The American Academy of Pediatrics classifies the drug as usually compatible with breastfeeding [53].
5-ASA drugs — Experience with the 5-ASA agents, such as mesalamine, during pregnancy and lactation is much more limited than for sulfasalazine. However, a growing body of information suggests that oral and topical 5-ASA agents are safe during pregnancy [55-58]. In one report, for example there was no increase in major malformations in infants born to 165 women exposed to mesalamine during pregnancy (146 of whom had first trimester exposure) compared to a matched control group [58].
Thus sulfasalazine and 5-ASA drugs can be used during pregnancy as in nonpregnant women with IBD. Continued use of these drugs to maintain remission during pregnancy should be encouraged. (See "Sulfasalazine and 5-aminosalicylates in the treatment of ulcerative colitis".)
5-ASAs are excreted in breast milk in low concentrations and there is a report of an infant who developed diarrhea after being breast fed by a mother who was receiving a rectal preparation of a 5-ASA [59]. The American Academy of Pediatrics recommends that infants be monitored for the development of diarrhea and that the agents be given with caution [53].
Glucocorticoids — Because of their importance in the treatment of a variety of inflammatory conditions, systemic glucocorticoids have been used fairly extensively during pregnancy. The risk of steroids in this setting appears to be small, and it is generally agreed that they should not be withheld during pregnancy when they are clinically indicated. (See "Use of antiinflammatory and immunosuppressive drugs in rheumatic diseases during pregnancy and lactation".)
Three potential areas of concern have been raised: congenital malformations, (primarily cleft palate), neonatal adrenal insufficiency, and low birth weight [60]. The combined results of five large studies (two surveillance and three case-control studies) found that the risk of oral clefts is approximately doubled [61-65]. However, the absolute risk is low. For example, in California (where one of the studies was done), the background risk for nonsyndromic oral clefts is about 1-2/1000 livebirths. If systemic glucocorticoids double this risk, the absolute risk would be 0.2 to 0.4 percent. Since palatal closure is usually complete by the 12th week of pregnancy, the risk is limited to administration during the first trimester.
Neonatal adrenal insufficiency following maternal administration of steroids is unusual. The rapid maternal metabolism of prednisolone, binding to serum proteins, and conversion to inactive metabolites by placental 11 beta-hydroxysteroid dehydrogenase results in relatively low fetal compared to maternal concentrations [66]. As a result, the fetal pituitary is rarely suppressed in mothers taking glucocorticoids [67]. However, long-term, high doses will suppress the fetal adrenal glands.
Multiple studies have observed low birth weight in offspring of animals given glucocorticoids during pregnancy. However, this association has been rarely reported in humans [60]. It is difficult to draw conclusions regarding the effects of glucocorticoids on fetal growth because of variability in the dose, duration, and type of steroid and the confounding effects of the underlying maternal disease on the pregnancy. However, repeated standardized courses of dexamethasone or betamethasone (glucocorticoids that are not significantly metabolized by the placenta) given for fetal lung maturation have been associated with small decreases of birth weight in humans [68,69], though the significance of the small decrease in birth weight is likely outweighed by the maternal benefit.
In a number of studies of women with IBD, there were no adverse effects on fetal outcome in mothers who took glucocorticoids during pregnancy [46]. Thus, glucocorticoids should be administered to pregnant women with IBD for the same indications as for nonpregnant patients. However, glucocorticoids have the potential for exacerbating pregnancy-induced hypertension, gestational diabetes, and preterm delivery from premature rupture of membranes [70]. Thus women at risk should be appropriately monitored.
Low levels of prednisone and its metabolite, prednisolone, can be measured in breast milk. A nursing infant of a mother consuming a daily dose of 80 mg of prednisolone would ingest less than 0.1 percent, which is equivalent to less than 10 percent of endogenous cortisol production [71]. As a result, although it may be reasonable to delay breast feeding for several hours after ingesting prednisone, it appears to be safe during breast feeding [71].
6-mercaptopurine and azathioprine — Both 6-mercaptopurine (6-MP) and azathioprine cross the placenta and can be detected in cord blood. The largest experience with these drugs in pregnancy has been derived from patients in whom they were given for non-IBD indications such as solid organ transplantation [72]. These data, combined with experience in IBD, suggest that it is reasonable to continue these drugs during pregnancy in patients who cannot be managed with other types of therapy [73-75]. A detailed discussion on this topic is presented separately. (See "Immunomodulator therapy in Crohn's disease", section on 'Pregnancy'.)
Reports of teratogenicity with these drugs have occurred in fetuses in whom exposure was from either the mother or father. A small retrospective study suggested that males on 6-MP should stop the agent three months before conception to lessen the risks of spontaneous abortion and fetal abnormalities [76]. In another report, six men taking azathioprine fathered seven healthy children [9]. (See "Immunomodulator therapy in Crohn's disease".)
Azathioprine and 6-MP are detectable at low levels in breast milk. Mothers taking these drugs should be discouraged from nursing, although no good data are available regarding risks to the nursing infant. One study of eight women receiving azathioprine (75 to 200 mg daily) found that the majority of the 6-MP was excreted within the first four hours following drug ingestion [77]. Based upon the maximum concentration of 6-MP in breast milk, the infant was estimated to ingest less than 0.008 mg/kg bodyweight/24 hours.
Cyclosporine — Cyclosporine can be successfully used during pregnancy. Teratogenicity appears to be low but premature labor and small for gestational age infants have been reported. However, fetal growth restriction could be due to the mother’s underlying disease. (See "Use of antiinflammatory and immunosuppressive drugs in rheumatic diseases during pregnancy and lactation".)
Methotrexate — Methotrexate is a potent abortifacient, and its use during pregnancy is associated with multiple skeletal abnormalities. Methotrexate-induced developmental toxicity is strongly related to when the drug is given and the dose. For structural anomalies, the critical time is 8 to 10 weeks after the first day of the last menstrual period, and the critical dose is ≥10 mg/week. Women and men taking methotrexate for IBD should discontinue this drug and use contraception for at least three months prior to conception. (See "Use of antiinflammatory and immunosuppressive drugs in rheumatic diseases during pregnancy and lactation".)
Anti-TNF agents — Experience with anti-TNF agents used to treat IBD (ie, infliximab, adalimumab and certolizumab pegol) during pregnancy is limited [78]. (See "Use of antiinflammatory and immunosuppressive drugs in rheumatic diseases during pregnancy and lactation".)
Infliximab — According to the manufacturer's prescribing information, infliximab should be given to a pregnant woman only if clearly needed. (This information is available online at: http://www.remicade.com/remicade/assets/HCP_PPI.pdf). However, no maternal toxicity, embryotoxicity, or teratogenicity to infliximab was observed in a murine toxicity model conducted by the manufacturer. Rates of live births, miscarriages, and therapeutic terminations do not appear to be significantly different in women exposed to infliximab during pregnancy. (See "Use of antiinflammatory and immunosuppressive drugs in rheumatic diseases during pregnancy and lactation".)
An observational study assessed pregnancy outcomes in women with IBD receiving anti-TNF therapy and showed that direct exposure to anti-TNF treatment during pregnancy was not related to a higher incidence of adverse pregnancy outcomes than IBD overall [79].
Infliximab crosses the placenta. The exact timing is unclear but in general, maternal antibodies are transported across the placenta beginning in the early to mid second trimester, with the highest fetal concentration in the third trimester near term. High levels of infliximab (with a prolonged half-life) have been detected in newborns [80]. The clinical significance is unclear but could pose a problem for the fetus/neonate if it develops an infection while infliximab is present.
Whether infliximab is excreted in human milk and whether excretion into breast milk has an adverse effect on a nursing infant are unknown. The elimination half-life is very long (8 to 10 days), so use of the drug close to delivery results in neonatal blood levels and confounds analysis of effects from early breastfeeding. In addition, it is likely that the antibody ingested during breastfeeding would be digested by the infant, thus reducing the potential for adverse effects. As a result, the decision to discontinue its use during nursing should take into account the importance of the drug to the mother.
Infliximab may decrease sperm motility and the number of normal oval forms [81]. Whether these changes translate into impaired fertility is unknown.
Adalimumab — Studies in monkeys have not revealed harm to the fetus when adalimumab was given during pregnancy. There are no well-controlled studies in humans.
Whether adalimumab is excreted into breast milk has not been well studied. As with infliximab, the decision to discontinue its use during nursing should take into account the importance of the drug to the mother.
The manufacturer recommends that providers enroll pregnant patients into a registry by calling 1-877-311-8972.
Certolizumab pegol — There are no well-controlled studies of certolizumab pegol in pregnant or lactating women. Studies in rats showed that pegylated Fab fragments did not cross the placenta and did not reveal evidence of harm to the fetus. However, exposure to certolizumab pegol during pregnancy is too limited to draw any conclusions. As with the other anti-TNFs, administration of certolizumab pegol during pregnancy is not recommended unless the benefit clearly outweighs the risk. However, due to the long elimination half-life (14 days), the drug needs to be discontinued prior to conception. It is not known whether certolizumab pegol is excreted into breast milk and as with infliximab, the decision to discontinue its use during nursing should take into account the importance of the drug to the mother.
Antidiarrheal drugs — Animal studies have suggested that diphenoxylate with atropine (Lomotil®) and loperamide (Imodium®) do not cause teratogenicity during pregnancy. However, case reports in humans have suggested a potential for fetal malformations in infants exposed to diphenoxylate with atropine during the first trimester [82], though the US Food and Drug Administration did not conclude that the available data support a causal relationship [83]. As a result, it is probably best to avoid their use, especially early in pregnancy, unless measures with bulking agents and dietary manipulations fail to control disabling diarrhea. As alternatives, kaolin/pectin and psyllium (Metamucil®) are safe during pregnancy.
As for breastfeeding, the American Academy of Pediatrics (AAP) considers loperamide and atropine to be compatible with nursing [53]. Although there are no data for diphenoxylate, short, infrequent courses of diphenoxylate/atropine can likely be classified similarly.
EFFECT OF BREASTFEEDING ON DISEASE COURSE — Some studies have suggested an association between breast feeding and exacerbation of autoimmune diseases in the mother [84-86]. (See "Rheumatoid arthritis and pregnancy".) A study of postpartum women with IBD found that only 44 percent chose to breastfeed their children either because of a physician recommendation, fear of medication interactions, or personal choice [87]. While the risk of increased IBD activity was slightly higher in women who chose to breastfeed, this association was no longer apparent after adjusting for medication cessation. This implies that any relationship between breastfeeding and disease activity is more of a consequence of discontinuation of IBD therapies rather than the breastfeeding itself.
SURGERY — As discussed above, fulminant colitis can be treated with colectomy with subsequent delivery of a healthy infant [19,20]. However surgery may be associated with premature labor or spontaneous abortion, possibly related to inadvertent uterine manipulation [21]. Therapeutic abortion does not appear to alter the course of ulcerative colitis.
Women who have undergone an ileoanal anastomosis and pouch procedure for ulcerative colitis can expect to have normal pregnancies and deliveries with minimal effects on pouch function. Both vaginal and cesarean deliveries have been reported in these patients [88]. However, some patients may have increased stool frequency during the day and night with occasional patients experiencing incontinence. Ileal pouch perforation due to adhesions to the posterior uterus was described in a case report [89]. (See "Surgical management of inflammatory bowel disease".)
The outcome in women who have undergone proctocolectomy and ileostomy is variable. As an example, one survey included 45 women with ostomies who had recently delivered [90]. One-third had no problems related to the ostomy. Among the others, the following issues were encountered, most of which could be managed medically:
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
SUMMARY AND RECOMMENDATIONS
All topics are updated as new information becomes available. Our peer review process typically takes one to six weeks depending on the issue.