- Email: [email protected]
Pregnancy in inflammatory bowel disease: Effect of sulfasalazine and corticosteroids on fetal outcome
GASTROENTEROLOGY
1981;80:72-76
Pregnancy in Inflammatchy Bowel Disease: Effect of Stilfasalazine and Corticosteroids on Fetal Outcome MICHAEL MOGADAM, WILLIAM 0. DOBBINS III, BURTON I. KORELITZ, and SUSAN W. AHMED Georgetown University School of Medicine, Washington, D.C.; University of Michigan Medical School, Ann Arbor, Michigan; Lenox Hill Hospital, New York, New York; and the Division of Biostatistics and Epidemiology, Georgetown University School of Medicine, Washington, DC.
To determine the safety of corticosteroids and sulfascdazine in pregnancy associated with inflammatory bowel disease, we reviewed the outcome of 531 such pregnancies in a national survey. Two hundred eighty-seven pregnancies (172 ulcerative colitis and ~15 Crohn’s disease) were treated with either or both drugs, whereas 244 (237 ulcerative colitis and 107 Crohn’s disease) received neither. The frequency of fetal complications was significantly lower in both “treated” and “untreated” groups, as compared with the reported rates in the general pregnant population. We conclude: Inj7ammatory bowel disease, except for severe active Crohn’s disease, does not seem,to affect the outcome of, concurrent pregnancy adversely. The use of corticosteroid and sulfasalazine in pregnancy associated with ulcerative colitis is unlikely to increase the fetal morbidity or mortality. Patients with severe Crohn’s disease requiring corticosteroid and/or both drugs experience more complications than the “untreated” ones, but still fewer than the prevailing rates in the general The higher complication pregnant population. rate seems to be associated more with disease-related factors than the use of these drugs. In the management of inflammatory bowel disease associated with pregnancy, either or both drugs may be used just as in the nonpregnant patients. Received February 25, 1999. Accepted July 26, 1999. Address requests for reprints to: Michael Mogadam, M.D., 5249 Duke Street, Alexandria, Virginia 22394. This study was supported in part by Pharmacia Inc., Piscataway, N.J. This work was presented in part at the Annual Meeting of the American Gastroenterological Association, May 19-22, 1980, Salt Lake City, Utah. 0 1981 by the American Gastroenterological Association 0016-5085/81/010072-05$02.50
The effectiveness of sulfasalazine and steroids in tbe management of inflammatory bowel disease (IBD) is well established (l-5). However, the safety of these agents in pregnancy is uncertain (6-9). Cleft palate and kernicterus have been suggested as possible complications of sulfasalazine use during pregnancy, yet there are no published reports to support this view (10). In some animal species, large doses of cortisone administered early in pregnancy may produce cleft palate and stillborn fetuses, but their use in human pregnancy is thought to be reasonably safe (11-13). Reinisch et al. (14) reported that prednisone treatment of infertility and its subsequent use for the maintenance of the resulting pregnancy in humans, led to a significant decrease in birthweight of fullterm infants, and an increase in the percentage of newborn infants weighing 2500 grams or less. On the other hand, Papageorgiou et al. (15) and Taeusch et al. (16) demonstrated that corticosteroids were beneficial in markedly reducing the incidence of respiratory distress syndrome without fetal side effects, when administered to mothers with potentially premature deliveries. This study was undertaken to assess the safety of sulfasalazine and corticosteroids in human pregnancy associated with IBD and to determine the effect of IBD on the outcome of pregnancy and the fetus. Patients A detailed
and Methods questionnaire,
dealing
with
pregnancy
IBD, was sent to members of the American Gastroenterological Association and of the American College of Gastroenterology. It consisted of multiple questions relating to pregnancy in women with ulcerative colitis (UC) or Crohn’s disease (CD), and whether during the associated
with
PREGNANCY
January 1981
course
of pregnancy
sulfasalazine,
steroids,
or both
were
used. The fetal outcome was analyzed in relation to treatment during each trimester as well as throughout pregnancy. Each questionnaire contained 440 variables. Two thousand forty questionnaires were mailed, and 976 were returned. Slightly more than 20% of respondents had had no experience with pregnancy in IBD. Nearly threefourths of those with experience submitted data on less than 3 patients each. Many who failed to return the questionnaire were contacted. Some wrote to us voluntarily. Almost all found it too “time consuming” or “very difficult” to retrieve the requested data, particularly if those physicians were based in large institutions. Adequate data were provided on 531 ‘pregnancies in patients with IBD. Two hundred eighty-seven patients (172UC and 115 CD) were treated with either or both drugs, whereas 244 patients (137 UC and 107 CD) did not receive either during the course of pregnancy. Questionnaires containing inadequate supporting data were not included. Patients whose medications were discontinued as soon as the pregnancy was established were assigned to the “untreated” group. The data were then analyzed, using the Z-test for
comparison of two proportions, corrected for continuity. Where the numbers were too small, Fisher’s exact test was performed
(17).
Results Untreated
Group
with ulcerative colitis (UC) The 137 patients and 107 with Crohn’s disease (CD) gave birth to 135 and 105 healthy newborns, respectively. The UC group had one set of twins. There was one each of spontaneous abortion, stillbirth, and prematurity in the UC group. In the CD group, there was one spontaneous abortion, and one infant born with spina bifida. In this group, no increased fetal risk was demonstrated. Indeed, a paucity of fetal morbidity and mortality became apparent (Table 1).
Table
1.
The Number “Untreated” Patients
of Fetal Complications in and Subgroups of “Treated” Treated (288)”
Complications Underweight (zsOO g) Prematurity Spontaneous abortion Stillbirth Developmental defecth
C
S
(102) z 2 1 1 I
(102) 0 0 0 2 0
c+s (84) 1 2 5 0 2
Untreated (245)”
0 1 2 1 1C
C = Corticosteroids; S = Sulfasalazine; C + S = both corticosteroids and sulfasalazine. (The figures in parentheses denote the number of patients in each subgroup.) ’ One set of twins in each group. h One each of congenital deafness, heart defect, and cleft palate. ’ Spina bifida.
Table 2.
IN INFLAMMATORY
BOWEL
“Treated” IBD Pregnancies Birth of Healthy Newborns
Concluded
Ulcerative colitis (n = 172) Treatment
C
s
4 0 1 8 11 38 62
6 0 7 12 5 31b 61’
trimester
1st only 2nd only 3rd only 1st and 2nd 2nd and 3rd throughout Total
c+s 6 0 0 6 5 26 43
73
DISEASE
with the
Crohn’s disease (n = 115) c
s
0 0 1 2 12” 17 32
7 0 5 4 5 18 39
c+s 4 0 1 0 2 24 31
C = Corticosteroids; S = Sulfasalazine; C + S = both corticosteroids and sulfasalazine. a Three patients did not receive S during the last 2 wk of pregnancy. b Two patients also received S in the last trimester. ’ Five patients received C for short periods.
“Treated”
Group
Of 172 pregnancies in UC patients, 166 healthy babies, including one set of twins, were delivered. The 115 pregnancies in patients with CD resulted in 102 healthy newborns (Table 2). The outcome of the remaining pregnancies in each subgroup is summarized in Table 3. One infant was born with cleft palate and microglossia. He had corrective surgery and was well at 1 yr follow-up. His mother had UC, complicated by toxic megacolon during the first trimester which responded to medical care. She received sulfasalazine and steroids during the entire pregnancy. Another baby born to a patient with CD requiring steroid therapy throughout her pregnancy, had left ventricular atresia, and died within a month. One CD patient, receiving steroids and sulfasalazine during the entire course of her pregnancy, gave birth to a newborn with congenital deafness. Only one infant developed deep jaundice, and this cleared with proper management. The mother of this newborn had CD and was treated with steroids during the first and second trimesters. None of the 181 mothers receiving sulfasalazine, including 107 who received this agent throughout pregnancy, had babies with jaundice. With the exception of the baby with left ventricular atresia, the remaining premature births survived. Three miscarriages in the first trimester and two additional spontaneous abortions in the second occurred in CD patients with severe disease requiring treatment with both drugs. Yet, the incidence of spontaneous abortion for the “treated” group as a whole (2%) (p < O.Ol), or for the subgroup with CD (4.3%) (0.01 < p < 0.05), was significantly below the rate of 9%14.4% observed in the general population (Table 4). The proportion of prematurity (1.4%) or underweight offsprings (1%) was a!so significantly less than the predicted values (8.8% and 5.9%, respectively). Since the “untreated” IBD was
74
MOGADAM
ET AL.
Table 3. Fetal Complications
GASTROENTEROLOGY
in the “Treated”
Group
IBD
Ulcerative colitis Treatment
trimester
C
1st only 2nd only 3rd only 1st and 2nd 2nd and 3rd throughout
Vol. 80. No. 1
s
Crohn’s disease c+s
C
S
c+s
1SA + 1P
3SA 1P 1U 1DD IDD 3
1SB
Total fetal “complications”
3
2SA
1SJ
ISB
1
1u lu+lP
1SB”
5
1
1P + 1DD 7
SA = spontaneous abortion. P = prematurity. SB = stillbirth. U = underweight, not including premature ones (<2500 g), SJ = severe jaundice. DD = developmental defects. 0 Steroids were also used in the first trimester. (Figures next to abbreviations denote the number of cases in each group.) C = corticosteroids; S = Sulfasalazine; C+S = both corticosteroids and sulfasalazine.
considered more suitable for comparison as a “control” group, we therefore analyzed the incidence of all complications in the “treated” vs. the “untreated” patients. The outcome of the UC subgroup receiving either or both drugs did not differ significantly from that of the “untreated” IBD (p > 0.10). On the other hand, the CD subgroup treated with steroids and/or both drugs experienced more complications (Table 5). The sulfasalazine subgroup of the CD, however, had no more complications than the “untreated” patients (p > 0.10).
Discussion Sulfasalazine and corticosteroids are the mainstay of pharmacotherapy of inflammatory bowel disease (IBD). Their safety, however, in pregnant women with IBD has not been established. The placenta is highly permeable to small molecules (mol wt < lOOO), and nonionized drugs ‘with a high lipid: water partition coefficient such as alcohol, salicylates, sulfanomides, and corticosteroids are readily transported from maternal plasma to fetal circulation by means of simple diffusion (18). Since the bilirubin displacing effect of sulfasalazine has been
demonstrated to be very low, it is highly improbable that it contributes to kernicterus (19). Although theoretically, an hemolytic process in a GGPD deficient infant might occur, no report of neonatal kernicterus due to maternal use of sulfasalazine has been published. In addition, the crude relative risk for malformations in offspring of mothers exposed to sulfanomides during their pregnancies is similar to the standardized relative risk for nonexposed mothers (20). In the present study, 5 patients receiving sulfasalazine during pregnancy had spontaneous abortions, an incidence of ~3%. Harlap et al, (21) estimated the incidence of first trimester miscarriages in the general population to be 11.8%, and an additional 2.7% in the second trimester, approximating the rate of 9.9% reported by the National Center for Health statistics (22). Underweight or premature newborns were also less frequent in the “treated” group, as compared with the general population (Table 4). There were two stillbirths in the sulfasalazinetreated subgroup (slightly <2%), not significantly different from the expected rate. Comparison of the outcome of the sulfasalazine-“treated” pregnancies Table
5.
Fetal Complications “Untreated”
Table
4.
Fetal Complications General Population
in “Treated” IBD Versus
“Treated”
Complications
(%)
(%)
p-Value
Underweight”
5.9-12.8%(24)
1.0
co.01
0.8 (24) 9.9 (22)
1.4 2.1
1 (22) 1.7 (25) 3.2(26)(27)
1.0 1.0
z-o.10 0.5< p < 0.10
(2590 g) Prematurity Spontaneous abortion Stillbirth Developmental defects
a The lower figure is for the whites and the higher for the blacks (Figures in parentheses denote the reference number.)
Versus
UC Drugs
General population
in “Treated”
IBD
C S c+s Total
“Treated”
“Untreated”
4.6 1.6 6.5 3.5
p-Value
2.2 2.2 2.2 2.2
>O.lO >O.lO >O.lO >O.lO
1.9 1.9 1.9 1.9
to.01 >O.lO
CD C S c+s Total
13.5 2.5 18.5 11.3
C = corticosteroids; S = sulfasalazine; roids and sulfasalazine.
C + S = both corticoste-
January
1981
with that of the “untreated” group also did not yield any increased risk of fetal complications (Table 5), which strongly supported sulfasalazine’s safety. In animal studies, when large doses of corticosteroids are administered early in pregnancy, an increased number of cleft palate, stillborn fetuses, and fetal islet cell degeneration has been reported, but few, if any, adverse effects have been demonstrated in human fetuses. A review of 264 infants born to mothers receiving corticosteroids during pregnancy failed to reveal any evidence of fetal injury (13).Bulmash (ll,12)reported similar observations in patients treated with a variety of disorders requiring steroid therapy. Here too, the relative risk for fetal malformations in pregnant women treated with corticosteroids has been shown to be no different than that for nonsteroid-treated pregnancies (20). Kenny et al. (23) measured cortisol secretion in 8 newborns whose mothers had been treated with steroids. Normal cortisol production was demonstrated, and none of the infants showed symptoms of relative hypocorticism. Since corticosteroids are conjugated to biologically less active sulfates more rapidly in the fetus than in the adult (18),a suppressive fetal blood concentration of active steroids is often not achieved with the therapeutic dosage prescribed during pregnancy. Even with long-term maintenance high dosage therapy, corticosteroid suppression of the newborn hypothalamic pituitary adrenal axis is rare (10). In the present study, there were 185 pregnancies treated with steroids alone or in combination with sulfasalazine. No statistically significant complication arose as a result of the use of this agent, when compared with the outcome in the general pregnant population (21-22,24-27). Our finding of only 3 underweight newborns in these pregnancies is in contrast to the result of Reinisch et al. (14).The overall incidence of infants who are of low birthweight (2500 grams or less) is 12.8% for blacks, as compared with 5.9% for whites. Since the majority of patients with IBD are white, the incidence of 5.9% is more relevant to this group. The low incidence of 1.5% found in patients receiving corticosteroids was similar to that in the subgroup treated with both drugs. Thus, the observation that the offspring of “steroid-induced” and “steroiddependent” pregnancies weighed significantly less at delivery (14) may not be strictly applicable to fertilization and pregnancy in general. The overall incidence of complications in the UC subgroup receiving steroids throughout pregnancy (2.6%)was virtually the same as the “untreated” patients (2.2%). Similarly, no significant difference emerged when UC patients receiving both drugs were analyzed (p > 0.10). Based on our sample sizes and the low incidence of complications, the power of the statistical test, i.e., the sensitivity or ability to
PREGNANCY
IN INFLAMMATORY
BOWEL
DISEASE
75
detect a 4% difference, if there is indeed a difference, is 65%. To obtain a power of 80%, we would have required 1525 patients in each group in order to detect an increase from a 2 to 6%. The available data, however, lend strong support to the safety of corticosteroids alone or in combination with sulfasalazine during the course of pregnancy in these patients. The behavior of the CD subgroup appeared to be distinctly different from the UC patients. They experienced more complications than the UC or the “untreated” patients (p < O.Ol), but fewer than the reported rates in the general pregnant population. The CD patients requiring both agents for the management of their severe disease experienced five spontaneous abortions. Five additional complications occurred in those receiving steroids and/or both drugs throughout pregnancy. Since these agents did not seem to affect the fetal outcome in the UC subgroup adversely, it remains speculative as to why CD patients should do less well. It is quite likely that disease related factors, i.e., the severity and activity of CD (for which these drugs were deemed necessary) and small bowel involvement with malnutrition (or malabsorption), also contributed to fetal complications. Future prospective studies should help clarify this difference. We believe that better and more frequent monitoring of prenatal care in IBD is likely to be the major contributor to the favorable outcome of such pregnancies, particularly those associated with ulcerative colitis. We were impressed with the frequent reluctance and uneasiness of physicians engaged in the care of IBD patients to use either corticosteroids or sulfasalazine during the course of pregnancy. Thus, when a pregnancy resulted in fetal complication, it was readily recalled, and the submitted data consistently met the requirements of our protocol for inclusion in the study. On the other hand, many “treated” pregnancies leading to the birth of healthy newborns, were excluded from this study, a bias mandated by inadequate or anecdotal data. The addition of these cases would have diluted the frequency of fetal complications, giving further support to the safety of these agents in the course of pregnancy. An inherent shortcoming of this type of study is that it lacks the sensitivity of prospective observations. However, to amass a large number of patients with IBD associated with pregnancy is a formidable task, requiring many years of a multicentric, cooperative effort. In the meantime, the findings of this study provide a rational basis for the following conclusions: 1. IBD, except for the severe active CD, does not seem to affect the outcome of coexisting pregnancy adversely. 2. The use of corticosteroid and sulfasalazine in
76
MOGADAM
ET AL.
pregnancy associated with UC is unlikely to increase the fetal morbidity or mortality. 3. The CD patients, with severe disease requiring corticosteroid and/or both drugs experience more complications than the “untreated” IBD, but still fewer than the prevailing rates in the general pregnant population. This higher complication rate seems to be associated more with disease related factors than the use of corticosteroid and sulfasalazine. 4. In the management of IBD associated with pregnancy, either or both drugs many be used just as in the nonpregnant patients.
References 1. Dissanayake A, Truelove SC. A controlled therapeutic trial of longterm maintenance treatment of ulcerative colitis with sulfasalazine. Gut 1973;14:923-26. 2. Khan A, Piris J, Truelove SC. An experiment to determine the active therapeutic moiety of sulfasalazine. Lancet 1977;2:89295. 3. Smith PR, Dawson DJ, Swan HS. Prostoglandin synthetase activity in acute ulcerative colitis: effects of treatment with sulfasalazine, codeine phosphate and prednisone. Gut 1979;20%02-5.
4. Singleton JW, Law DH, Kelley ML Jr., et al. A trial of sulfasalazine as adjunctive therapy in Crohn’s disease. Gastroenterology 1979;77:887-97. 5. Summers RW, Switz DM, Sessions JT Jr., et al. National Cooperative Crohn’s disease study: results of drug treatment. Gastroenterology 1979;77:647-69. 6. Crohn BB, Yarnis H, Korelitz BI. Regional enteritis complicating pregnancy. Gastroenterology 1956;31:615-28. 7. Norton RA, Patterson JF. Pregnancy and regional enteritis. Obstet Gynecol 1972;40:711-12. 8. De Dombal FT, Watts SM, Watkinson G, Goligher JC. Ulcerative colitis and pregnancy. Lancet 1965;2:599-601. 9. De Dombal FT, Burton IL, Goligher JC. Crohn’s disease and pregnancy. Br Med J 1972;3:550-53. 10. Erikson M, Catz CS, Yaffee SJ. Drugs and pregnancy. Clin Obstet Gynecol Annual 1973;16:199-224.
GASTROENTEROLOGY
11. Bulmash
Vol. 89, No. 1
JM. Rheumatoid arthritis and pregnancy. Obstet Gynecol Annual 1979;8:223-76. 12. Bulmash JM. Systemic Lupus erythematosus and pregnancy. Obstet Gynecol Annual 1978;7:153-94. 13. Bongiovanni AM, McPadden AJ. Steroids during pregnancy and possible fetal complications. Fertil-Steril 1960;11:181-86. 14. Reinisch SM, Simon NG, Karow WG, Gandleman R. Prenatal exposure to prednisone in humans and animals retards intrauterine growth. Science 1978;202:436-38. 15. Papageorgiou AN, Desazanges MF, Masson M, et al. The antenatal use of beta-methasone in the prevention of respiratory distress syndrome: a controlled double blind study. Pediatrics 1979;63:73-79. 16. Tauesch HW Jr., Frigoletto F, Kitzmillen J, et al. Risk of respiratory distress syndrome after prenatal dexamethasone treatment. Pediatrics 19796364-72. 17. Colton T. Statistics in Medicine. First edition. Boston: Little, Brown and Company, 1974. 18. Ryan KJ. Endocrine organs of reproduction. The placenta. In: Reid DE, Ryan KJ, Benirschke K, eds. Principles and management of human reproduction. Philadelphia: W.B. Saunders, 1972:87-95. 19. Jarnerot G, Into-malmberg M. Sulfasalazine treatment during breastfeeding. Stand J. Gastroenterol 1979;14:869-71. OP. In: Birth defects and drugs in pregnancy. 20. Heinomen Little, Massachusetts: Publishing Sciences Group, 1977. S, et al. A prospective 21. Harlap S, Shiono MD, Ramcharan study of spontaneous fetal losses after induced abortions. N Engl J Med 1979:301:677-81. 22. Facts of life and death. DHEW publication (PHS) No. 79-1222, 1978. C, Spaulding JS, Migeon CJ. Corti23. Kenny FM, Preeyasombat sol production rate, in infants born of steroid-treated mothers and of diabetic mothers. Pediatrics 1966;37:969-66. 24. Monthly Vital Statistics Report. Final natality statistics, 1977. National Center for Health Statistics 1979;2?No. 1. 25. Shamsi HH, Petzie RH, Steer CM. Changing obstetric practice and amelioration of perinatal outcome in a university hospital. Am J Obstet Gynecol 1979$33:855-f%. and birth defects. 26. Rothman KJ, Louik C. Oral contraceptives N Engl J Med 1978;299:522-4. 27. Bergstrom LV. Congenital deafness. In: Northern JL, ed. Hearing Disorders. Boston: Little, Brown and Company, 1976:1717.
1981;80:72-76
Pregnancy in Inflammatchy Bowel Disease: Effect of Stilfasalazine and Corticosteroids on Fetal Outcome MICHAEL MOGADAM, WILLIAM 0. DOBBINS III, BURTON I. KORELITZ, and SUSAN W. AHMED Georgetown University School of Medicine, Washington, D.C.; University of Michigan Medical School, Ann Arbor, Michigan; Lenox Hill Hospital, New York, New York; and the Division of Biostatistics and Epidemiology, Georgetown University School of Medicine, Washington, DC.
To determine the safety of corticosteroids and sulfascdazine in pregnancy associated with inflammatory bowel disease, we reviewed the outcome of 531 such pregnancies in a national survey. Two hundred eighty-seven pregnancies (172 ulcerative colitis and ~15 Crohn’s disease) were treated with either or both drugs, whereas 244 (237 ulcerative colitis and 107 Crohn’s disease) received neither. The frequency of fetal complications was significantly lower in both “treated” and “untreated” groups, as compared with the reported rates in the general pregnant population. We conclude: Inj7ammatory bowel disease, except for severe active Crohn’s disease, does not seem,to affect the outcome of, concurrent pregnancy adversely. The use of corticosteroid and sulfasalazine in pregnancy associated with ulcerative colitis is unlikely to increase the fetal morbidity or mortality. Patients with severe Crohn’s disease requiring corticosteroid and/or both drugs experience more complications than the “untreated” ones, but still fewer than the prevailing rates in the general The higher complication pregnant population. rate seems to be associated more with disease-related factors than the use of these drugs. In the management of inflammatory bowel disease associated with pregnancy, either or both drugs may be used just as in the nonpregnant patients. Received February 25, 1999. Accepted July 26, 1999. Address requests for reprints to: Michael Mogadam, M.D., 5249 Duke Street, Alexandria, Virginia 22394. This study was supported in part by Pharmacia Inc., Piscataway, N.J. This work was presented in part at the Annual Meeting of the American Gastroenterological Association, May 19-22, 1980, Salt Lake City, Utah. 0 1981 by the American Gastroenterological Association 0016-5085/81/010072-05$02.50
The effectiveness of sulfasalazine and steroids in tbe management of inflammatory bowel disease (IBD) is well established (l-5). However, the safety of these agents in pregnancy is uncertain (6-9). Cleft palate and kernicterus have been suggested as possible complications of sulfasalazine use during pregnancy, yet there are no published reports to support this view (10). In some animal species, large doses of cortisone administered early in pregnancy may produce cleft palate and stillborn fetuses, but their use in human pregnancy is thought to be reasonably safe (11-13). Reinisch et al. (14) reported that prednisone treatment of infertility and its subsequent use for the maintenance of the resulting pregnancy in humans, led to a significant decrease in birthweight of fullterm infants, and an increase in the percentage of newborn infants weighing 2500 grams or less. On the other hand, Papageorgiou et al. (15) and Taeusch et al. (16) demonstrated that corticosteroids were beneficial in markedly reducing the incidence of respiratory distress syndrome without fetal side effects, when administered to mothers with potentially premature deliveries. This study was undertaken to assess the safety of sulfasalazine and corticosteroids in human pregnancy associated with IBD and to determine the effect of IBD on the outcome of pregnancy and the fetus. Patients A detailed
and Methods questionnaire,
dealing
with
pregnancy
IBD, was sent to members of the American Gastroenterological Association and of the American College of Gastroenterology. It consisted of multiple questions relating to pregnancy in women with ulcerative colitis (UC) or Crohn’s disease (CD), and whether during the associated
with
PREGNANCY
January 1981
course
of pregnancy
sulfasalazine,
steroids,
or both
were
used. The fetal outcome was analyzed in relation to treatment during each trimester as well as throughout pregnancy. Each questionnaire contained 440 variables. Two thousand forty questionnaires were mailed, and 976 were returned. Slightly more than 20% of respondents had had no experience with pregnancy in IBD. Nearly threefourths of those with experience submitted data on less than 3 patients each. Many who failed to return the questionnaire were contacted. Some wrote to us voluntarily. Almost all found it too “time consuming” or “very difficult” to retrieve the requested data, particularly if those physicians were based in large institutions. Adequate data were provided on 531 ‘pregnancies in patients with IBD. Two hundred eighty-seven patients (172UC and 115 CD) were treated with either or both drugs, whereas 244 patients (137 UC and 107 CD) did not receive either during the course of pregnancy. Questionnaires containing inadequate supporting data were not included. Patients whose medications were discontinued as soon as the pregnancy was established were assigned to the “untreated” group. The data were then analyzed, using the Z-test for
comparison of two proportions, corrected for continuity. Where the numbers were too small, Fisher’s exact test was performed
(17).
Results Untreated
Group
with ulcerative colitis (UC) The 137 patients and 107 with Crohn’s disease (CD) gave birth to 135 and 105 healthy newborns, respectively. The UC group had one set of twins. There was one each of spontaneous abortion, stillbirth, and prematurity in the UC group. In the CD group, there was one spontaneous abortion, and one infant born with spina bifida. In this group, no increased fetal risk was demonstrated. Indeed, a paucity of fetal morbidity and mortality became apparent (Table 1).
Table
1.
The Number “Untreated” Patients
of Fetal Complications in and Subgroups of “Treated” Treated (288)”
Complications Underweight (zsOO g) Prematurity Spontaneous abortion Stillbirth Developmental defecth
C
S
(102) z 2 1 1 I
(102) 0 0 0 2 0
c+s (84) 1 2 5 0 2
Untreated (245)”
0 1 2 1 1C
C = Corticosteroids; S = Sulfasalazine; C + S = both corticosteroids and sulfasalazine. (The figures in parentheses denote the number of patients in each subgroup.) ’ One set of twins in each group. h One each of congenital deafness, heart defect, and cleft palate. ’ Spina bifida.
Table 2.
IN INFLAMMATORY
BOWEL
“Treated” IBD Pregnancies Birth of Healthy Newborns
Concluded
Ulcerative colitis (n = 172) Treatment
C
s
4 0 1 8 11 38 62
6 0 7 12 5 31b 61’
trimester
1st only 2nd only 3rd only 1st and 2nd 2nd and 3rd throughout Total
c+s 6 0 0 6 5 26 43
73
DISEASE
with the
Crohn’s disease (n = 115) c
s
0 0 1 2 12” 17 32
7 0 5 4 5 18 39
c+s 4 0 1 0 2 24 31
C = Corticosteroids; S = Sulfasalazine; C + S = both corticosteroids and sulfasalazine. a Three patients did not receive S during the last 2 wk of pregnancy. b Two patients also received S in the last trimester. ’ Five patients received C for short periods.
“Treated”
Group
Of 172 pregnancies in UC patients, 166 healthy babies, including one set of twins, were delivered. The 115 pregnancies in patients with CD resulted in 102 healthy newborns (Table 2). The outcome of the remaining pregnancies in each subgroup is summarized in Table 3. One infant was born with cleft palate and microglossia. He had corrective surgery and was well at 1 yr follow-up. His mother had UC, complicated by toxic megacolon during the first trimester which responded to medical care. She received sulfasalazine and steroids during the entire pregnancy. Another baby born to a patient with CD requiring steroid therapy throughout her pregnancy, had left ventricular atresia, and died within a month. One CD patient, receiving steroids and sulfasalazine during the entire course of her pregnancy, gave birth to a newborn with congenital deafness. Only one infant developed deep jaundice, and this cleared with proper management. The mother of this newborn had CD and was treated with steroids during the first and second trimesters. None of the 181 mothers receiving sulfasalazine, including 107 who received this agent throughout pregnancy, had babies with jaundice. With the exception of the baby with left ventricular atresia, the remaining premature births survived. Three miscarriages in the first trimester and two additional spontaneous abortions in the second occurred in CD patients with severe disease requiring treatment with both drugs. Yet, the incidence of spontaneous abortion for the “treated” group as a whole (2%) (p < O.Ol), or for the subgroup with CD (4.3%) (0.01 < p < 0.05), was significantly below the rate of 9%14.4% observed in the general population (Table 4). The proportion of prematurity (1.4%) or underweight offsprings (1%) was a!so significantly less than the predicted values (8.8% and 5.9%, respectively). Since the “untreated” IBD was
74
MOGADAM
ET AL.
Table 3. Fetal Complications
GASTROENTEROLOGY
in the “Treated”
Group
IBD
Ulcerative colitis Treatment
trimester
C
1st only 2nd only 3rd only 1st and 2nd 2nd and 3rd throughout
Vol. 80. No. 1
s
Crohn’s disease c+s
C
S
c+s
1SA + 1P
3SA 1P 1U 1DD IDD 3
1SB
Total fetal “complications”
3
2SA
1SJ
ISB
1
1u lu+lP
1SB”
5
1
1P + 1DD 7
SA = spontaneous abortion. P = prematurity. SB = stillbirth. U = underweight, not including premature ones (<2500 g), SJ = severe jaundice. DD = developmental defects. 0 Steroids were also used in the first trimester. (Figures next to abbreviations denote the number of cases in each group.) C = corticosteroids; S = Sulfasalazine; C+S = both corticosteroids and sulfasalazine.
considered more suitable for comparison as a “control” group, we therefore analyzed the incidence of all complications in the “treated” vs. the “untreated” patients. The outcome of the UC subgroup receiving either or both drugs did not differ significantly from that of the “untreated” IBD (p > 0.10). On the other hand, the CD subgroup treated with steroids and/or both drugs experienced more complications (Table 5). The sulfasalazine subgroup of the CD, however, had no more complications than the “untreated” patients (p > 0.10).
Discussion Sulfasalazine and corticosteroids are the mainstay of pharmacotherapy of inflammatory bowel disease (IBD). Their safety, however, in pregnant women with IBD has not been established. The placenta is highly permeable to small molecules (mol wt < lOOO), and nonionized drugs ‘with a high lipid: water partition coefficient such as alcohol, salicylates, sulfanomides, and corticosteroids are readily transported from maternal plasma to fetal circulation by means of simple diffusion (18). Since the bilirubin displacing effect of sulfasalazine has been
demonstrated to be very low, it is highly improbable that it contributes to kernicterus (19). Although theoretically, an hemolytic process in a GGPD deficient infant might occur, no report of neonatal kernicterus due to maternal use of sulfasalazine has been published. In addition, the crude relative risk for malformations in offspring of mothers exposed to sulfanomides during their pregnancies is similar to the standardized relative risk for nonexposed mothers (20). In the present study, 5 patients receiving sulfasalazine during pregnancy had spontaneous abortions, an incidence of ~3%. Harlap et al, (21) estimated the incidence of first trimester miscarriages in the general population to be 11.8%, and an additional 2.7% in the second trimester, approximating the rate of 9.9% reported by the National Center for Health statistics (22). Underweight or premature newborns were also less frequent in the “treated” group, as compared with the general population (Table 4). There were two stillbirths in the sulfasalazinetreated subgroup (slightly <2%), not significantly different from the expected rate. Comparison of the outcome of the sulfasalazine-“treated” pregnancies Table
5.
Fetal Complications “Untreated”
Table
4.
Fetal Complications General Population
in “Treated” IBD Versus
“Treated”
Complications
(%)
(%)
p-Value
Underweight”
5.9-12.8%(24)
1.0
co.01
0.8 (24) 9.9 (22)
1.4 2.1
1 (22) 1.7 (25) 3.2(26)(27)
1.0 1.0
z-o.10 0.5< p < 0.10
(2590 g) Prematurity Spontaneous abortion Stillbirth Developmental defects
a The lower figure is for the whites and the higher for the blacks (Figures in parentheses denote the reference number.)
Versus
UC Drugs
General population
in “Treated”
IBD
C S c+s Total
“Treated”
“Untreated”
4.6 1.6 6.5 3.5
p-Value
2.2 2.2 2.2 2.2
>O.lO >O.lO >O.lO >O.lO
1.9 1.9 1.9 1.9
to.01 >O.lO
CD C S c+s Total
13.5 2.5 18.5 11.3
C = corticosteroids; S = sulfasalazine; roids and sulfasalazine.
C + S = both corticoste-
January
1981
with that of the “untreated” group also did not yield any increased risk of fetal complications (Table 5), which strongly supported sulfasalazine’s safety. In animal studies, when large doses of corticosteroids are administered early in pregnancy, an increased number of cleft palate, stillborn fetuses, and fetal islet cell degeneration has been reported, but few, if any, adverse effects have been demonstrated in human fetuses. A review of 264 infants born to mothers receiving corticosteroids during pregnancy failed to reveal any evidence of fetal injury (13).Bulmash (ll,12)reported similar observations in patients treated with a variety of disorders requiring steroid therapy. Here too, the relative risk for fetal malformations in pregnant women treated with corticosteroids has been shown to be no different than that for nonsteroid-treated pregnancies (20). Kenny et al. (23) measured cortisol secretion in 8 newborns whose mothers had been treated with steroids. Normal cortisol production was demonstrated, and none of the infants showed symptoms of relative hypocorticism. Since corticosteroids are conjugated to biologically less active sulfates more rapidly in the fetus than in the adult (18),a suppressive fetal blood concentration of active steroids is often not achieved with the therapeutic dosage prescribed during pregnancy. Even with long-term maintenance high dosage therapy, corticosteroid suppression of the newborn hypothalamic pituitary adrenal axis is rare (10). In the present study, there were 185 pregnancies treated with steroids alone or in combination with sulfasalazine. No statistically significant complication arose as a result of the use of this agent, when compared with the outcome in the general pregnant population (21-22,24-27). Our finding of only 3 underweight newborns in these pregnancies is in contrast to the result of Reinisch et al. (14).The overall incidence of infants who are of low birthweight (2500 grams or less) is 12.8% for blacks, as compared with 5.9% for whites. Since the majority of patients with IBD are white, the incidence of 5.9% is more relevant to this group. The low incidence of 1.5% found in patients receiving corticosteroids was similar to that in the subgroup treated with both drugs. Thus, the observation that the offspring of “steroid-induced” and “steroiddependent” pregnancies weighed significantly less at delivery (14) may not be strictly applicable to fertilization and pregnancy in general. The overall incidence of complications in the UC subgroup receiving steroids throughout pregnancy (2.6%)was virtually the same as the “untreated” patients (2.2%). Similarly, no significant difference emerged when UC patients receiving both drugs were analyzed (p > 0.10). Based on our sample sizes and the low incidence of complications, the power of the statistical test, i.e., the sensitivity or ability to
PREGNANCY
IN INFLAMMATORY
BOWEL
DISEASE
75
detect a 4% difference, if there is indeed a difference, is 65%. To obtain a power of 80%, we would have required 1525 patients in each group in order to detect an increase from a 2 to 6%. The available data, however, lend strong support to the safety of corticosteroids alone or in combination with sulfasalazine during the course of pregnancy in these patients. The behavior of the CD subgroup appeared to be distinctly different from the UC patients. They experienced more complications than the UC or the “untreated” patients (p < O.Ol), but fewer than the reported rates in the general pregnant population. The CD patients requiring both agents for the management of their severe disease experienced five spontaneous abortions. Five additional complications occurred in those receiving steroids and/or both drugs throughout pregnancy. Since these agents did not seem to affect the fetal outcome in the UC subgroup adversely, it remains speculative as to why CD patients should do less well. It is quite likely that disease related factors, i.e., the severity and activity of CD (for which these drugs were deemed necessary) and small bowel involvement with malnutrition (or malabsorption), also contributed to fetal complications. Future prospective studies should help clarify this difference. We believe that better and more frequent monitoring of prenatal care in IBD is likely to be the major contributor to the favorable outcome of such pregnancies, particularly those associated with ulcerative colitis. We were impressed with the frequent reluctance and uneasiness of physicians engaged in the care of IBD patients to use either corticosteroids or sulfasalazine during the course of pregnancy. Thus, when a pregnancy resulted in fetal complication, it was readily recalled, and the submitted data consistently met the requirements of our protocol for inclusion in the study. On the other hand, many “treated” pregnancies leading to the birth of healthy newborns, were excluded from this study, a bias mandated by inadequate or anecdotal data. The addition of these cases would have diluted the frequency of fetal complications, giving further support to the safety of these agents in the course of pregnancy. An inherent shortcoming of this type of study is that it lacks the sensitivity of prospective observations. However, to amass a large number of patients with IBD associated with pregnancy is a formidable task, requiring many years of a multicentric, cooperative effort. In the meantime, the findings of this study provide a rational basis for the following conclusions: 1. IBD, except for the severe active CD, does not seem to affect the outcome of coexisting pregnancy adversely. 2. The use of corticosteroid and sulfasalazine in
76
MOGADAM
ET AL.
pregnancy associated with UC is unlikely to increase the fetal morbidity or mortality. 3. The CD patients, with severe disease requiring corticosteroid and/or both drugs experience more complications than the “untreated” IBD, but still fewer than the prevailing rates in the general pregnant population. This higher complication rate seems to be associated more with disease related factors than the use of corticosteroid and sulfasalazine. 4. In the management of IBD associated with pregnancy, either or both drugs many be used just as in the nonpregnant patients.
References 1. Dissanayake A, Truelove SC. A controlled therapeutic trial of longterm maintenance treatment of ulcerative colitis with sulfasalazine. Gut 1973;14:923-26. 2. Khan A, Piris J, Truelove SC. An experiment to determine the active therapeutic moiety of sulfasalazine. Lancet 1977;2:89295. 3. Smith PR, Dawson DJ, Swan HS. Prostoglandin synthetase activity in acute ulcerative colitis: effects of treatment with sulfasalazine, codeine phosphate and prednisone. Gut 1979;20%02-5.
4. Singleton JW, Law DH, Kelley ML Jr., et al. A trial of sulfasalazine as adjunctive therapy in Crohn’s disease. Gastroenterology 1979;77:887-97. 5. Summers RW, Switz DM, Sessions JT Jr., et al. National Cooperative Crohn’s disease study: results of drug treatment. Gastroenterology 1979;77:647-69. 6. Crohn BB, Yarnis H, Korelitz BI. Regional enteritis complicating pregnancy. Gastroenterology 1956;31:615-28. 7. Norton RA, Patterson JF. Pregnancy and regional enteritis. Obstet Gynecol 1972;40:711-12. 8. De Dombal FT, Watts SM, Watkinson G, Goligher JC. Ulcerative colitis and pregnancy. Lancet 1965;2:599-601. 9. De Dombal FT, Burton IL, Goligher JC. Crohn’s disease and pregnancy. Br Med J 1972;3:550-53. 10. Erikson M, Catz CS, Yaffee SJ. Drugs and pregnancy. Clin Obstet Gynecol Annual 1973;16:199-224.
GASTROENTEROLOGY
11. Bulmash
Vol. 89, No. 1
JM. Rheumatoid arthritis and pregnancy. Obstet Gynecol Annual 1979;8:223-76. 12. Bulmash JM. Systemic Lupus erythematosus and pregnancy. Obstet Gynecol Annual 1978;7:153-94. 13. Bongiovanni AM, McPadden AJ. Steroids during pregnancy and possible fetal complications. Fertil-Steril 1960;11:181-86. 14. Reinisch SM, Simon NG, Karow WG, Gandleman R. Prenatal exposure to prednisone in humans and animals retards intrauterine growth. Science 1978;202:436-38. 15. Papageorgiou AN, Desazanges MF, Masson M, et al. The antenatal use of beta-methasone in the prevention of respiratory distress syndrome: a controlled double blind study. Pediatrics 1979;63:73-79. 16. Tauesch HW Jr., Frigoletto F, Kitzmillen J, et al. Risk of respiratory distress syndrome after prenatal dexamethasone treatment. Pediatrics 19796364-72. 17. Colton T. Statistics in Medicine. First edition. Boston: Little, Brown and Company, 1974. 18. Ryan KJ. Endocrine organs of reproduction. The placenta. In: Reid DE, Ryan KJ, Benirschke K, eds. Principles and management of human reproduction. Philadelphia: W.B. Saunders, 1972:87-95. 19. Jarnerot G, Into-malmberg M. Sulfasalazine treatment during breastfeeding. Stand J. Gastroenterol 1979;14:869-71. OP. In: Birth defects and drugs in pregnancy. 20. Heinomen Little, Massachusetts: Publishing Sciences Group, 1977. S, et al. A prospective 21. Harlap S, Shiono MD, Ramcharan study of spontaneous fetal losses after induced abortions. N Engl J Med 1979:301:677-81. 22. Facts of life and death. DHEW publication (PHS) No. 79-1222, 1978. C, Spaulding JS, Migeon CJ. Corti23. Kenny FM, Preeyasombat sol production rate, in infants born of steroid-treated mothers and of diabetic mothers. Pediatrics 1966;37:969-66. 24. Monthly Vital Statistics Report. Final natality statistics, 1977. National Center for Health Statistics 1979;2?No. 1. 25. Shamsi HH, Petzie RH, Steer CM. Changing obstetric practice and amelioration of perinatal outcome in a university hospital. Am J Obstet Gynecol 1979$33:855-f%. and birth defects. 26. Rothman KJ, Louik C. Oral contraceptives N Engl J Med 1978;299:522-4. 27. Bergstrom LV. Congenital deafness. In: Northern JL, ed. Hearing Disorders. Boston: Little, Brown and Company, 1976:1717.