Prospective , controlled , multicentre study of loperamide in pregnancy

Can J Gastroenterol Vol 14 No 3 March 2000 185 The Motherisk Program, The Hospital For Sick Children, University of Toronto, Toronto, Ontario; Birth Defects Unit and Telefono Rosso, lnstitute of Pediatrics, Catholic University, Rome, ltaly; Laboratory of Teratology, Hebrew University, Jerusalem, lsrael; Mario Negri Institute, Milan, Italy; The Family Federation of Finland, Department of Genetics, Helsinki, Finland Correspondence: Adrienne Einarson, The Motherisk Program, Division of Clinical Pharmacology, The Hospital For Sick Children, 555 University Avenue, Toronto, Ontario M5G 1X8. Telephone 416-813-4927, fax 416-813-7562, e-mail einarson@sickkids.on.ca Received for publication November 3, 1998. Accepted January 25, 1999 ORIGINAL ARTICLE

L operamide (Imodium, Janssen-Ortho Inc, North York, Ontario/McNeil Consumer Products, Guelph, Ontario) is a synthetic piperidine derivative used for the treatment of both acute and chronic diarrhea.It exerts its action via cholinergic, noncholinergic, opiate and nonopiate receptor-mediated mechanisms.After oral administration, little systemic absorption takes place.About 40% of a 2 mg dose is excreted in the feces, with about 30% being unchanged drug.Reproductive studies performed in the rat and rabbit revealed no evidence of impaired fertility or harm to the fetus at dose levels up to 30-fold the therapeutic dose for humans.This drug is commonly used for both acute and chronic conditions, such as travellers' diarrhea and inflammatory bowel disease (1).There are limited human data available on the use of this drug in pregnancy.
In a surveillance study of Michigan Medicaid patients, 108 pregnant women were exposed to this drug during the first trimester.There were six major birth defects, three of which were cardiovascular abnormalities (2).Because of the limitations of this study, such data are not sufficient to counsel women on the safety and risk of this drug in pregnancy.Because at least 50% of pregnancies are unplanned, it is likely that this drug may be used in early pregnancy (3).Because of this paucity of information, we decided to carry out this study to ascertain further the risk and safety for women consuming this drug during pregnancy.

PATIENTS AND METHODS
The Motherisk Program is a counselling service for pregnant and lactating mothers, and their health professionals.Information about the safety or risk of drugs, chemicals, radiation and infectious diseases is given to approximately 35,000 callers per year.The other centres in Italy, Israel and Finland have similar services.The authors attempted to follow all the women who had called enquiring about the safety of this drug in pregnancy.All women who had used the drug during pregnancy and agreed to participate were included in the study, with no exclusions.Upon contacting these women, history of exposure and pregnancy outcome were obtained, as well as other end points of interest, with the aid of a structured questionnaire (4).Exposure history included medical indication for drug used, dosage, frequency of administration and timing of exposure.Outcomes were confirmed whenever possible by the child's primary care physician.The primary outcome of interest was the presence or absence of a major malformation.Major malformation was defined as the presence of an anomaly that has an adverse effect on either the function or social acceptability of the individual (5).Secondary outcomes of interest were spontaneous or therapeutic abortions, stillbirths, and presence or absence of a minor abnormality.Exposure was said to be during organogenesis if it occurred between the fourth and 14th week of gestation, when rates of 1% to 3% for major malformations, 10% to 15% for minor malformations and up to 15% for spontaneous abortions are expected.In addition to comparing mean birth weights between the study and control groups, birth weights were compared between the babies whose mothers took the drug for only a few days for an acute case of diarrhea and those who took the drug throughout the pregnancy for chronic bowel disease.
Each woman was matched to a subsequent woman who had either called the authors about loperamide but had not needed to use it, or a woman who had been exposed to cisapride (a nonteratogenic drug for gastrointestinal motility problems) (6).They were also matched for age, smoking and alcohol use.To assess the possibility of selection bias, women lost to follow-up were compared with those participating for age, parity, smoking and alcohol use.Outcome end points of interest were compared between the study and control groups with the Student's t test, c 2 or the Fisher's exact test, whenever appropriate.These follow-up procedures were approved by The Hospital for Sick Children's Research Ethics Committees.

RESULTS
A total of 105 women from five centres were followed prospectively.There were 58 cases from Toronto, Ontario; 25 from Rome, Italy; 16 from Jerusalem, Israel; four from Milan, Italy; and two from Helsinki, Finland.Eighty-nine (85%) women took loperamide during organogenesis, with 21 (20%) taking it throughout their pregnancies.Indications for use were short term, for an acute case of diarrhea, or chronic, for bowel disease such as Crohn's disease or irritable bowel syndrome.The doses varied greatly, from 4 to 6 mg in total, to 2 to 6 mg/day throughout the pregnancy.
In the study group, there were 95 live births, four therapeutic abortions and six spontaneous abortions.There were no major malformations and three minor malformations reported.The malformations consisted of a heart murmur, a mild right pelviectasis (no symptoms) and a hypospadia (minor).The mean birth weight was 3368 g (Table 1).
A subanalysis comparing the women who took the drug throughout pregnancy for chronic bowel disease with the women who took only a few doses for acute diarrhea was performed; the mean birth weight was 3261 g among women who took the drug throughout their pregnancy compared with 3441 g in those who took the drug for acute diarrhea.However, this difference was not statistically significant, although in the chronic use group, half of the babies weighed less than 3000 g (11 of 21).There were no statistical differences in the preterm delivery rates between the acute users and chronic users -four of 74 (5%) and two of 21(9%) ba-Can J Gastroenterol Vol 14 No 3 March 2000  bies, respectively, were born before 36 weeks' gestation (Table 2).
In the control group, there were 94 live births, two therapeutic abortions and nine spontaneous abortions.There was a baby with Down's syndrome and one major malformation: cleft palate with cupped ears.There were three minor malformations: bilateral twisted feet (no surgery required), microfacelia and pulmonary stenosis.The mean birth weight was 3407 g.Comparison between the two groups (Table 2) found no statistical differences in pregnancy outcome, including spontaneous and therapeutic abortions, rates of major and minor malformations, and mean birth weights.Maternal characteristics were also compared among the study group, the control group and the group lost to follow-up; no differences were found.

DISCUSSION
This is the first prospective, controlled study of loperamide in pregnancy to date.The results do not support the findings in the previous study with the Michigan Medicaid patients (2).In that study, which was a surveillance study with no personal interviews of patients (108), six major malformations were found in the first trimester, three of which were cardiac anomalies, compared with our group, which found no major malformations in the first trimester.This illustrates the importance of prospective, controlled studies of exposures in pregnancy, which include history and other specifics, such as concurrent drug use, smoking and alcohol use.With that being the only report available before the present study, it is probable that women are advised not to take loperamide during pregnancy.This is reasonable, except that many women are not aware that they are pregnant when taking medications.A large number of the women who took the drug on an acute basis did not consider whether they were pregnant, with a substantial number reporting that they were on holiday in a foreign country.In previous studies and during the counselling process, women have informed us that they have had a therapeutic abortion of an otherwise wanted pregnancy based on fear of the consequences of the drug exposure on their fetus (7).We did not find an increase in the rates of preterm delivery, nor were the birth weights statistically significantly different, although the babies of chronic users were 200 g smaller than babies in the control group.However, this fact, in addition to the finding that half of the babies born to mothers who were exposed throughout the pregnancy weighed less than 3000 g, supports the results, as in other studies, that women with chronic bowel disease have smaller babies and should be encouraged to treat their condition with the appropriate medication during pregnancy, without fear of harming their babies (8,9).

CONCLUSIONS
This first prospective controlled study of loperamide in pregnancy of 105 cases, with 89 exposures in the first trimester, suggests that this drug does not increase the baseline risk of major malformations.However, an increase in teratogenic risk of some rare malformations cannot be ruled out.