This class of medication is felt to be compatible with breastfeeding with caution due to potential adverse effects on the fetus [35]. Small amounts of mesalamine are excreted into human milk as well as its metabolite, acetyl-5-aminosalicylate. The estimated daily intake of the neonate is felt to be negligible, and adverse events have not been seen [52]. Diarrhea has been reported in a nursing infant after the maternal exposure to rectal mesalamine [53]. Sulfasalazine is broken down to sulfapyridine and acetyl-5-aminosalicylate. Sulfapyridine is excreted in low levels into breast milk. The low doses that are seen do not increase the risk of kernicterus and that there is no significant displacement of bilirubin from albumin [31]. There is at least one report of bloody diarrhea associated with sulfasalazine use [54]. If diarrhea or bloody diarrhea develops in a breastfed infant on aminosalicylates, either the mother should stop breastfeeding or if clinically reasonable stop the drug.

Very small amounts of prednisone and prednisolone are excreted into breast milk. In a study using radioactively labeled prednisolone in seven patients and examining the levels in breast milk, there was 0.14 % of the administered 5 mg dose found per liter of milk [55]. In a second study, doses of 10–80 mg/day lead to milk concentrations of 5–25 % of serum concentrations. It is estimated that the infant, at even high levels, receives <0.1 % of the dose, which is less than 10 % of the nursing infant’s own endogenous cortisol production [56].

Breastfeeding recommendation: Compatible [35].

There is only a small amount of human data available on maternal use of thiopurines during breastfeeding.

Christensen et al. studied the pharmacokinetics of azathioprine in eight patients—looking at milk and plasma levels after dose of azathioprine from 75 to 200 mg. They found that the highest concentration of azathioprine was seen in the milk with the first 4 h of dosing. Levels seen were very low, and on the basis of maximum concentrations measured, the infant received <0.008 mg/kg body weight over a 24-h period [57].

Angelberger and colleagues followed the babies of 11 mothers taking AZA (median 150 mg/day) for IBD during pregnancy and lactation and compared this group to 12 mothers on no immunosuppressants. Duration of breastfeeding was 6 months in the AZA group vs. 8 months in the non-AZA group [58]. The children were followed to median age of 3.3 years in the AZA group and 4.7 years in the non-exposed group. There were no differences in childhood infections between the AZA exposed and non-exposed groups. Common cold and conjunctivitis were more common in the non-AZA exposed group. This data suggests that exposure to AZA in utero and in breastfeeding does not increase the risk of infection in the neonates [58].

Breastfeeding recommendations: Probably compatible, discard milk 4 h after dosing [35].

Methotrexate is excreted in small amounts into breast milk. Methotrexate may accumulate in neonatal tissues. Breastfeeding is contraindicated for patients on methotrexate [35].

Cyclosporine is excreted into breast milk. In one study, the infant’s blood levels were as high as 78 % of the maternal trough concentrations. The estimated exposure is from 0.2 to 2.1 % of the mother’s weight-adjusted dose [35].

Breastfeeding recommendation: Limited human data—potential toxicity due to detectable levels in the blood of neonates and subsequent immunosuppression [35].

Infliximab and certolizumab have not been detected in breast milk in patients who are receiving these agents for inflammatory bowel disease.

Breastfeeding recommendation: Limited human data—thought to be low risk and compatible with breastfeeding [59].

As discussed above, fertility is decreased in women who undergo a colectomy with IPAA for ulcerative colitis. There is less information available on pouch function in women who do become pregnant with an ileal pouch. Hahnloser and colleagues studied 232 pregnancies in 135 women, 1–16 years after IPAA for ulcerative colitis [60]. They found that daytime stool frequency 7 months after delivery was the same as in the pregravid state (5.4 vs. 5.4) but that at 68 months after delivery, there was a slight increase (5.4 vs. 6.4). In this group of women, there was an increase in occasional fecal incontinence, 21 % pregravid as compared to 36 % at the last post-pregnancy follow-up. Pregnancy did not increase the incidence of stricture formation, pouchitis, or obstruction [60]. In a review of all the available literature on pouch complications associated with pregnancy, Seligman et al. reported that stool frequency and incontinence was not significantly affected by pregnancy or mode of delivery [61]. Antepartum small bowel obstruction was reported in 8 out of 283 pregnancies reviewed [61]. These all resolved nonoperatively after delivery. Other complications seen in this review were postpartum SBO in 6.7 %, pouchitis in 1.8 %, and a perianal abscess in 1 patient [61]. Overall, the literature supports the fact that pregnancy in patients with ileal pouches is well tolerated.

There is an increased rate of Cesarean section as mode of delivery in patients with ulcerative colitis. Nguyen et al. reported a rate of 42 % compared to 30.9 % in a non-IBD population (p = 0.0002) [62]. Lower Cesarean section rates of 13.8 % (all IBD) and 29 % (UC) have been reported in two large community-based practice studies [24, 63]. In general, the decision for a Cesarean section in patients with ulcerative colitis should be made for obstetric indications. Patients who have had a total abdominal colectomy with IPAA can have a successful vaginal delivery as it relates to pouch function [60]. The main delivery concern in this population relates to potential of damage to the anal sphincter. In a patient with borderline continence related to the IPAA, a tear or episiotomy that affects the sphincter may significantly worsen continence issues. The potential risks to the sphincter of vaginal delivery should be discussed with the patient, obstetrician, and surgeon as delivery decisions are being made [12].

Patients with ulcerative colitis typically do quite well during pregnancy, and pregnancy should not be discouraged in this patient population. Fertility in the medically treated patient is no different than in the non-UC population but is affected by surgical therapy with IPAA. Ideally, patients should be in remission prior to conception and should continue to be optimally treated throughout their pregnancy. Compliance with medical regimen should be encouraged to avoid flares of disease which then may impact the pregnancy in terms of preterm delivery and low birth weight infants. Most medications used for the treatment of ulcerative colitis can be used safely during pregnancy. Delivery method should be based on obstetric indications. Collaborative management of the pregnant ulcerative colitis patient with the obstetrician, gastroenterologist, and surgeon (in IPAA patient) is recommended.