- Email: [email protected]
Double-blind, placebo-controlled study of ranitidine for gastroesophageal reflux symptoms during pregnancy
Double-Blind, Placebo-Controlled Ranitidine for Gastroesophageal Symptoms During Pregnancy
Study of Reflux
JANET D. LARSON, MD, EDNA PATATANIAN,
RPh, PHlLlP B. MINER, Jr, MD,
WILLIAM
G. ROBINSON,
F. RAYBURN,
MD, AND MALCOLM
Objective: To determine whether ranitidine (Zantac) taken once or twice daily is effective for relieving symptoms of gastroesophageal reflux among pregnant women who had failed conservative measures. Methods: Volunteers with heartburn despite antacids were sought among our obstetric clinic population for this double-blind, placebo-controlled, triple crossover trial. After a baseline week, 20 patients were randomized to receive the three following weekly regimens: ranitidine 150 mg twice daily, placebo in the morning and ranitidine 150 mg in the evening, or placebo twice daily. Daily scores on symptom diaries, global assessments, and number of antacids taken were compared among the 18 patients completing the study. Results: The twice-daily dosage of ranitidine was the only regimen found to reduce heartburn symptoms when compared with the baseline (P < .OOl) or a placebo (P c .Ol). Compared with ranitidine taken once daily, the twice-daily dosing prompted less need for antacid tablets compared with the placebo (P < .05 versus P > .05) and to the baseline (P < .OOl versus P < .051. The average reduction of heartbum severity using twice-daily ranitidine was 55.6% when compared with baseline (95% confidence interval [CI] 34.8%, 76.5%) and was 44.2% when compared with placebo (95% CI 15.4%, 72.9%). Conclusion: This study indicates the efficacy of ranitidine 150 mg taken twice daily, rather than once daily, for relief of gastroesophageal reflux symptoms during pregnancy. fobstet Gynecol1997;90:83-7. 0 1997 by The American College of Obstetricians and Gynecologists.)
MD
usually vanish shortly after delivery. Reflux symptoms can be exacerbated by factors similar to those in nongravid patients, such as eating certain foods tie, fatty foods, caffeine, chocolate, natural mint, onions, or garlic), smoking, taking medications that reduce the lower esophageal sphincter pressure, or eating before bedtime. Conservative recommendations include avoiding exacerbants, elevating the head of the bed, and consuming antacids. We have found that 35% of our population ingests antacids during the third trimester of pregnanCY.~Severe reflux symptoms may not respond to these measures and may interfere with sleep, alter behavior, and lead to inanition. When this occurs, the clinician must consider the potential risks versus benefits of prescription level medical therapy. Histamine (Hz) receptor blockers have been the most commonly prescribed medications for reflux in nonpregnant patients when conservative measuresfail. The purpose of the present double-blind, triple crossover trial was to compare the efficacy of ranitidine (Zantac; Glaxo Wellcome, Research Triangle Park, NC) taken once or twice daily versus a placebo among pregnant patients with gastroesophageal reflux symptoms who had failed conservative measures.
Materials and Methods Heartburn, or gastroesophageal reflux, affects 30-50% of all gravidas and up to 80% in some populations.‘,’ Symptoms may begin at any time during pregnancy, usually are most problematic in the third trimester, and From the Section cfMaterna/-Fetal Medicine, Department qFObstetrics and Gynecology; University 4 Oklahoma College of Medicine; hwestigational Drug Service, Department qf Pharmacy, University Hospital; Clinical Medicine, Oklahoma Foundation for Digestive Research, Oklahoma City, Oklahoma.
VOL.
90, NO.
1, JULY
1997
After our institutional review board approved the protocol, volunteers for the investigation were sought among obstetric patients at 20 weeks’ gestation or beyond. The patients attended prenatal clinics at our hospital and responded to posted advertisements and descriptive handouts distributed in the clinics. Before enrolling, each patient read and signed the consent form approved by the institutional review board. The form provided a statement that “the general experience
0029.7&H/97/$17.00 PII SOO29-7844(97)00126-9
83
suggests very few if no risks of Zantac, and that pregnant animals given Zantac have been studied and their offspring did not have any birth defects or any problems.” Each enrollee had persistent heartburn associated with pregnancy despite antacid therapy. All failed a l-week trial of conservative measures, which were described on a written sheet given to the patients (ie, elevating the head of the bed, dietary changes, eliminating any ethanol and tobacco products, and continuing antacids). Each patient experienced heartbum on at least 4 of 7 days before beginning the investigation. Exclusions were a documented history of erosive esophagitis, any primary esophageal motility disorder, Zollinger-Ellison syndrome, esophageal disease other than esophagitis, or known esophageal stricture. In addition, known esophageal or gastric varices, active peptic ulcer disease, or para-esophageal or hiatal hernia were exclusionary criteria. Patients were not enrolled if there was a history of liver disease or a chronic psychiatric disorder requiring medication. Early termination from the investigation was by the patient’s request or by delivery before completion of the study. The duration of the study was 4 weeks. The initial week was for baseline measurements with antacids only, followed by three alternating l-week periods of double-blind assignment to either placebo twice daily, placebo in the morning and ranitidine in the evening, or ranitidine twice daily. Each patient recorded her symptoms daily in a diary, using a standard symptom sheet used by the Oklahoma Foundation for Digestive Research for heartburn studies. Symptoms assessed included the intensity of daytime and nocturnal heartburn and of regurgitation (food coming back up from stomach). The diary also asked for the number of antacid tablets taken during the day and the night, and other symptoms, which included belching, epigastric fullness (persistent full feeling after eating a meal), bloating-distention (need to loosen clothes after eating a meal), early satiety (feeling of stomach being over-filled soon after starting to eat a meal), nausea, and upper abdominal discomfort. Intensity was assessed using an g-cm visual analogue scale. An antacid tablet containing calcium carbonate 750 mg (TurnsEX; SmithKline Beecham Corp., Pittsburgh, PA) was supplied during each of the four l-week periods for use, as desired, up to 12 tablets daily. Study capsules were prepared by the investigational pharmacist by crushing a standard 150-mg tablet of ranitidine (Zantac) into a white powder and pouring the powder into unmarked blue capsules. The placebos were prepared by pouring a similar volume of white lactose powder (approximately 150 mg) into the blue capsules. Thus, the ranitidine and placebo containing
84
Larson et
al
Gastroesophageal R+LT
capsules appeared identical. The dosage regimen recommended by the manufacturer is one tablet before meals in the morning and in the evening. All volunteers were interviewed by one of the investigators and filled out an 18-page gastroesophageal reflux questionnaire that was created by the group at the Mayo Clinic and that has been used in multiple trials of nonpregnant heartburn patients.4 The questionnaire assessed heartburn symptoms as related to quality of life and included life-style and general health status questions. After the baseline assessment week, the eligible patients were randomized by the investigational pharmacy service using a computer-generated random number table as to the order in which each regimen was given. Each patient was given a package per week containing the following: 1) seven eight-packs of antacid tablets and 2) two containers of seven capsules each, prepared by the pharmacy, one labeled “for morning” (AM) and the other “for evening” (PM). Physicians and patients were blinded as to the contents of the containers. Each patient took one study capsule from the designated container 30-60 minutes before breakfast and dinner. Diaries were also distributed weekly. Instructions were provided about recording symptoms each day. At the end of each week, each patient completed a global assessment giving impressions of the symptoms for the entire week for heartburn intensity, regurgitation intensity, and improvement compared to her usual heartbum, on a visual analogue scale. Antacid and capsule counts were performed weekly by the principal investigator, and diaries were collected. After each patient delivered, the intrapartum and newborn records were reviewed and their outcomes were recorded. Visual analogue scales were 8 cm long, and patient responses were measured in mm (total 80 mm). A power analysis was calculated using the means and standard deviations of the baseline week heartburn severity visual analogue scales, and it was estimated that 16 patients were necessary to detect a 50% improvement with a power of 0.90 (p = .lO), and 19 patients for a power of 0.95 (/3 = .05). The results of the visual analogue scales were analyzed using means, standard errors of the mean, and standard deviations. These were analyzed using repeated measures analysis of variance, with post-hoc analysis by Tukey-Kramer multiple comparisons test. The treatment heartburn severity results were subtracted from the baseline, and placebo heartburn severity results and a percentage improvement was calculated along with the means and standard errors of the mean. The Wilcoxon signed-rank test was used to analyze the difference between the percentages. Demographic data of the study group
Obstetrics 0 Gynecology
were compared with an historic control group in the University of Oklahoma Perinatal Database, using analysis of variance and Fisher exact test. INSTAT statistical software (GraphPad Software, Inc., San Diego, CA) was used for all calculations. An (Y level of .05 was considered to be significant.
Results Between November 1995 and June 1996,25 (6.1%) of the 409 women in our obstetric population volunteered and met the study criteria. Of these, five withdrew before randomization due to concerns regarding fetal effects of medications during pregnancy (three), early delivery (one), or time constraints (one). Of the 20 remaining, 18 completed 100% of their diaries and returned all of the medication containers. Incomplete data were obtained for two patients; one delivered preterm during the first week of randomization, and the other failed to return three of her four weekly diaries. Demographic data for the 20 patients were compared with those of an historic control group of all deliveries from a l-year period (1995) in our departmental perinatal database. The study patients were older (29.1 + 6.8 versus 24.3 t 6.0 years; P < .OOl by Fisher exact test), less likely to be nulliparous (25.0% versus 29.1%; P < .03), and more likely to be white (85.0% versus 48.0%; P < .Ol) than the general population. Entry into the study was between 26 weeks’ and 36 weeks’ gestation. No patient complained of side effects attributable to the ranitidine. Improvements in heartburn were found either with
Baseline
Placebo
Ran&dine PM
Ranitidiae AM+PM
Figure 1. Global assessment of improvement of heartburn intensity based on end-of-week visual analogue scale C-P < .05 versus baseline. ‘P < ,001 versus baseline, not significant versus placebo. **P < .05 versus placebo, P < ,001 versus baseline; mean 2 standard error of the mean).
VOL.
90, NO.
1, JULY
1997
Baseline
Figure during dine in reduced .05) and reduced but not
Placebo
Rmitidine PM
Rallitidine AM+PM
2. Number of antacid tablets taken per week for heartburn pregnancy when taken with ranitidine in the PM only, ranitithe AM and PM, and placebo only. **Twice-daily ranitidine the need for antacid tablets compared with the placebo (P < with the baseline (P < .OOl). “The evening-only ranitidine dose the use of antacids when compared with the baseline (P < .05) with the placebo. Mean t standard error of the mean.
the placebo or with both dosing regimens of ranitidine. Figure 1 displays the global assessment of improvement. Compared with the baseline group, significant improvements were found using the placebo (P < ,051, evening dose of ranitidine (P < .OOl), and twice daily doses of ranitidine (P < .OOl). Ranitidine needed to be taken twice daily to allow for an improvement over the placebo treatment V’ < .05). The average reduction of heartburn intensity was 55.6% when comparing twicedaily ranitidine with baseline (95% confidence interval [CI] 34.8%, 76.5%). The mean reduction with twicedaily ranitidine, compared with placebo, was 44.2% (95% CI 15.4%, 72.9%). There was no statistical significance between improvement with the single dose of ranitidine over baseline or over placebo (P = .21). As shown in Figure 2, the number of antacid tablets taken was also less with ranitidine when given in the evening only (P < .05) and during the morning and evening (P < .Ol). Regurgitation, or the feeling of “food coming back up from the stomach,” was less intense with twice-daily ranitidine when compared with baseline (P < .05), but this reduction did not reach statistical significance when once- or twice-daily ranitidine was compared with placebo (P > .05). Heartburn intensity during the daytime and at night are compared in Figure 3. The intensity of daytime and nocturnal baseline heartburn did not differ statistically. Both once- and twice-daily regimens of ranitidine showed statistically significant decreases in nighttime heartburn compared with baseline (P < .05 for once daily, P < .OOl for twice daily), but only the twice-daily regimen reduced the intensity when compared with
Larson et al
Gastroesophageal
R@ux
85
Placebo
Ranitidine PM
R&id& Ah4+PM
Figure 3. Daytime versus nighttime heartburn. *P < .05 versus baseline. **P < ,001 versus baseline, P < .05 versus placebo. ***P < ,001 versus baseline, P < .Ol versus placebo, P < .05 versus PM only. Mean t standard error of the mean.
placebo (P < .05). Considering only daytime heartburn scores, the twice-daily regimen was superior to baseline (P < .OOl), placebo (P < .Ol), and once-daily dosing (P < .05X Three patients failed to respond to either ranitidine dose regimen compared with a placebo, although all three treatment arms improved heartburn symptoms compared with the baseline period. Enrollment into the study was at 29 weeks‘ (n = 2) and at 33 weeks’ gestation (n = 11, which was similar to those who responded to the ranitidine. Fifteen (83%) of the study cases were questioned at their 6-week postpartum exam or beyond. The heartburn symptoms subsided after delivery with no further need of antacid or ranitidine therapy in all but one case.
Discussion Heartburn is important when considering the quality of life for gravid women. Of the 25 patients completing our questionnaire, 50% noted heartburn severe enough to change their life-style. Forty-five percent stated that heartburn caused them to alter their daily activities a little or some of the time, while 25% reported that they altered their activities most or all of the time. Although more than one-half (53%) reported that heartburn existed before pregnancy, only 30% believed it was severe enough to require H, blockers, and only 20% consulted their physician regarding heartburn. The pathophysiology of gastroesophageal reflux and symptoms of heartburn during pregnancy is incom-
86 Larson et al
Gastroesophageal R$ux
pletely understood. In nonpregnant adults, gastroesophagealreflux can be traced to the following factors: 1) a decrease of the antireflux barrier, primarily the lower esophageal sphincter, which is influenced by the anatomy of the esophagogastric junction, 2) spontaneous relaxations of the lower esophageal sphincter, 3) increased volume and acidic content of the fluid refluxed, 4) decreased esophageal clearance, and 5) decreased tissue resistance of the esophageal mucosa.’ Physiologic changes due to pregnancy probably result in the increased prevalence of gastroesophageal reflux. The resting pressure of the lower esophageal sphincter is reduced during pregnancy with a decreased positive responsiveness to pharmacologic and physiologic stimulation. The increased abdominal pressure from the enlarging gravid uterus may exacerbate reflux.6 Gastrointestinal mucus secretion protects the mucosa.7In pregnancy, cervical mucus becomes thicker and more viscous in response to progesterone, and there might be a similar alteration in esophagealmucus secretion that may decrease the protective barrier. Ranitidine is a popular histamine (HJ receptor blocker used to treat peptic ulcer disease and gastroesophageal reflux as a twice daily 150 mg dose regimen.’ Ranitidine has been assigned to the Food and Drug Administration Category Bl for pregnancy, meaning that animal studies have not demonstrated a fetal risk but that there are no adequate studies in pregnant women. Preclinical testing in rats and rabbits have shown no adverse effects of ranitidine on reproductive function or pregnancy outcome.’ No antiandrogenie effects of ranitidine have been identified in male rat or human subjects, which distinguishes this agent from cimetidine, another H, blocker.“*” A single case report describes use of ranitidine in a single patient throughout her pregnancy, without any adverse fetal or neonatal sequelae.i2 Those investigators documented serum levels of ranitidine that were comparable to those in the amniotic fluid at 17 weeks, in the cord blood at delivery, and in the neonate’s blood after breastfeeding. The present study shows that twice-daily dosing of ranitidine is necessary for treating the heartburn that is associatedwith pregnancy. Our data do not support the prejudice that evening heartburn is worse because we found no differences in heartburn between night and day when either the symptoms or the number of antacids taken were used as measures of intensity. An evening-only dose of ranitidine is clearly not enough, and the twice-daily dosage was associated with a greater than 50% improvement in heartburn compared with antacids alone. This dose was effective in obtaining relief from other symptoms, such as epigastric fullness, upper abdominal discomfort, and bloating. The reduction of symptoms and antacid use indicates that the
Obstetrics b Gynecology
patients taking ranitidine experience fewer episodes of the heartburn for which they seek relief. Although many patients improved with life-style counseling and antacid therapy, their heartburn intensity was significantly less with pharmacologic treatment. Relief was shortly after ingestion of the medication, and no adverse pregnancy outcomes or drug reactions were noted. While an adequate washout period is always a concern in crossover studies, due to the time constraints in the current study, it was judged to be impractical. Because the plasma half-life of ranitidine is reported as 4.8 hours,r3 we believe the carry-over should be minimal. In addition, the sequence of the 3 trial weeks was random, so that any carry-over effects should cancel out. Three patients failed to respond to either ranitidine dose regimen compared with placebo, although all three improved when compared with the baseline period. Their entry into the study was at 29 weeks‘ (n = 2) and at 33 weeks’ gestation (n = 11, which was similar to those who responded to the ranitidine. Possible explanations for treatment failure include the following observations: 1) these patients had an enhanced placebo effect when compared with treatment successes (49.2% placebo improvement versus 9.5%, P = .02 by Fisher exact test), 2) all three received twice-daily ranitidine during the last week versus only 25% of the treatment successes (P < .04), suggesting an ordering effect, and 3) tobacco use (66.7% versus 12.5%) and less weight gain (5.0 5 0.4 kg versus 10.9 ? 8.8 kg) during pregnancy were more common for the treatment failures than for successes. No differences between the success and failure groups were found as to the presence of heartburn before pregnancy, the degree of obesity, or the intensity of baseline heartburn (7.9 2 3.1 for failures, 5.8 t 2.8 for successes, P = .25). The three patients comprised 60% of the group randomized to receive twice-daily ranitidine treatment at the last week. The trend toward less weight gain in the pregnancy may be an effect either from tobacco use or from the patient’s attempt to ingest less food to reduce reflux symptoms. Other possible explanations could include undiagnosed peptic ulcer disease or psychological pathology. Ranitidine recently was released as an over-thecounter preparation. This nonprescription strength (75 mg) is marketed for relieving and preventing heartburn and acid indigestion when taken as directed. There will probably be many questions from patients regarding the safety and efficacy of ranitidine during pregnancy. Although we did not assessthe nonprescription 75-mg dose, we can inform pregnant patients with severe heartburn symptoms that the 150-mg dose,
VOL.
90, NO.
1, JULY
1997
taken twice symptoms.
daily, is effective for reducing
heartburn
Rqerences 1. Baron
2. 3.
4.
5. 6.
7. 8.
9.
10.
11. 12.
13.
TH, Richter JE. Gastroesophageal reflux disease in pregnancy. Gastroenterol Clin North Am 1992;21:777-91. Olans LB, Wolf JL. Gastroesophageal reflux in pregnancy. Castro intest Endosc Clin North Am 1994;4:699-712. Splinter M, Nightingale B, Sawgraves R, Rayburn W. Medication use during pregnancy by women delivering at a tertiary university hospital. South Med J 1997 fin press). Locke GR, Talley NJ, Weaver AL, Zinsmeister AR. A new questionnaire for gastroesophageal reflux disease. Mayo Clin Proc 1994;69:539-47. Dodds WJ, Hogan WJ, Helm JF, Dent J. Pathogenesis of reflux esophagitis. Gastroenterology 1981;8:376-94. Scott LD. Gastrointestinal disease in pregnancy. In: Creasy RK, Resnik R, eds. Maternal-fetal medicine: principles and practice. Philadelphia (PA): WB Saunders, 1994:1026-39. Orlando RC. Esophageal epithelial defenses against acid injury. Am J Gastroenterol 1994;89:S48-52. Grant SM, Langtry HD, Brogden RN. Ranitidine: An updated review of its pharmacodynamic and pharmacokinetic properties and therapeutic use in peptic ulcer disease and other allied diseases. Drugs 1989;37:846-9. Kamada S, Sakanoue M, Iwata M, Taeuchi M, Shimpo K, Tanabe T. Effects of intravenous administration of ranitidine hydrochloride on the fertility of the male and female rats. J Toxic01 Sci 1984;9 (Suppl 1):13-28. Parker S, Schade RR, Pohl CR, Gavaler JS, Van Thiel DH. Prenatal and neonatal exposure of male rat pups to cimetidine but not ranitidine adversely affects subsequent adult sexual functioning. Gastroenterology 1984;86:675-80. Wang C, Wong KL, Lam KC, Lai CL. Ranitidine does not affect gonadal function in man. Br J Clin Pharmacol 1983;16:430-2. Armentano G, Bracco PL, Di Silverio C. Ranitidine in the treatment of reflux esophagitis in pregnancy. Clin Exp Obstet Gynecol 1989;16:130-3. Physicians’ Desk Reference. Montvale (NJ): Medical Economics Publications, 19971182-4.
Address reprint requests to: William F. Rayburn, MD Section qf Maternal-Fetal Medicine, Department qf Obstetrics and Gynecology University qf Oklahoma Health Sciences Center PO Box 26901, 4SP710 Oklahoma City, OK 73290 E-mail: [email protected]
Received October 7, 1996. Received in revised form February 20, 1997. Accepted February 28, 1997.
Copyright 0 1997 by The American College of Obstetricians Gynecologists. Published by Elsevier Science Inc.
Larsonet al
Gastroesophageal R&LX
and
87
Study of Reflux
JANET D. LARSON, MD, EDNA PATATANIAN,
RPh, PHlLlP B. MINER, Jr, MD,
WILLIAM
G. ROBINSON,
F. RAYBURN,
MD, AND MALCOLM
Objective: To determine whether ranitidine (Zantac) taken once or twice daily is effective for relieving symptoms of gastroesophageal reflux among pregnant women who had failed conservative measures. Methods: Volunteers with heartburn despite antacids were sought among our obstetric clinic population for this double-blind, placebo-controlled, triple crossover trial. After a baseline week, 20 patients were randomized to receive the three following weekly regimens: ranitidine 150 mg twice daily, placebo in the morning and ranitidine 150 mg in the evening, or placebo twice daily. Daily scores on symptom diaries, global assessments, and number of antacids taken were compared among the 18 patients completing the study. Results: The twice-daily dosage of ranitidine was the only regimen found to reduce heartburn symptoms when compared with the baseline (P < .OOl) or a placebo (P c .Ol). Compared with ranitidine taken once daily, the twice-daily dosing prompted less need for antacid tablets compared with the placebo (P < .05 versus P > .05) and to the baseline (P < .OOl versus P < .051. The average reduction of heartbum severity using twice-daily ranitidine was 55.6% when compared with baseline (95% confidence interval [CI] 34.8%, 76.5%) and was 44.2% when compared with placebo (95% CI 15.4%, 72.9%). Conclusion: This study indicates the efficacy of ranitidine 150 mg taken twice daily, rather than once daily, for relief of gastroesophageal reflux symptoms during pregnancy. fobstet Gynecol1997;90:83-7. 0 1997 by The American College of Obstetricians and Gynecologists.)
MD
usually vanish shortly after delivery. Reflux symptoms can be exacerbated by factors similar to those in nongravid patients, such as eating certain foods tie, fatty foods, caffeine, chocolate, natural mint, onions, or garlic), smoking, taking medications that reduce the lower esophageal sphincter pressure, or eating before bedtime. Conservative recommendations include avoiding exacerbants, elevating the head of the bed, and consuming antacids. We have found that 35% of our population ingests antacids during the third trimester of pregnanCY.~Severe reflux symptoms may not respond to these measures and may interfere with sleep, alter behavior, and lead to inanition. When this occurs, the clinician must consider the potential risks versus benefits of prescription level medical therapy. Histamine (Hz) receptor blockers have been the most commonly prescribed medications for reflux in nonpregnant patients when conservative measuresfail. The purpose of the present double-blind, triple crossover trial was to compare the efficacy of ranitidine (Zantac; Glaxo Wellcome, Research Triangle Park, NC) taken once or twice daily versus a placebo among pregnant patients with gastroesophageal reflux symptoms who had failed conservative measures.
Materials and Methods Heartburn, or gastroesophageal reflux, affects 30-50% of all gravidas and up to 80% in some populations.‘,’ Symptoms may begin at any time during pregnancy, usually are most problematic in the third trimester, and From the Section cfMaterna/-Fetal Medicine, Department qFObstetrics and Gynecology; University 4 Oklahoma College of Medicine; hwestigational Drug Service, Department qf Pharmacy, University Hospital; Clinical Medicine, Oklahoma Foundation for Digestive Research, Oklahoma City, Oklahoma.
VOL.
90, NO.
1, JULY
1997
After our institutional review board approved the protocol, volunteers for the investigation were sought among obstetric patients at 20 weeks’ gestation or beyond. The patients attended prenatal clinics at our hospital and responded to posted advertisements and descriptive handouts distributed in the clinics. Before enrolling, each patient read and signed the consent form approved by the institutional review board. The form provided a statement that “the general experience
0029.7&H/97/$17.00 PII SOO29-7844(97)00126-9
83
suggests very few if no risks of Zantac, and that pregnant animals given Zantac have been studied and their offspring did not have any birth defects or any problems.” Each enrollee had persistent heartburn associated with pregnancy despite antacid therapy. All failed a l-week trial of conservative measures, which were described on a written sheet given to the patients (ie, elevating the head of the bed, dietary changes, eliminating any ethanol and tobacco products, and continuing antacids). Each patient experienced heartbum on at least 4 of 7 days before beginning the investigation. Exclusions were a documented history of erosive esophagitis, any primary esophageal motility disorder, Zollinger-Ellison syndrome, esophageal disease other than esophagitis, or known esophageal stricture. In addition, known esophageal or gastric varices, active peptic ulcer disease, or para-esophageal or hiatal hernia were exclusionary criteria. Patients were not enrolled if there was a history of liver disease or a chronic psychiatric disorder requiring medication. Early termination from the investigation was by the patient’s request or by delivery before completion of the study. The duration of the study was 4 weeks. The initial week was for baseline measurements with antacids only, followed by three alternating l-week periods of double-blind assignment to either placebo twice daily, placebo in the morning and ranitidine in the evening, or ranitidine twice daily. Each patient recorded her symptoms daily in a diary, using a standard symptom sheet used by the Oklahoma Foundation for Digestive Research for heartburn studies. Symptoms assessed included the intensity of daytime and nocturnal heartburn and of regurgitation (food coming back up from stomach). The diary also asked for the number of antacid tablets taken during the day and the night, and other symptoms, which included belching, epigastric fullness (persistent full feeling after eating a meal), bloating-distention (need to loosen clothes after eating a meal), early satiety (feeling of stomach being over-filled soon after starting to eat a meal), nausea, and upper abdominal discomfort. Intensity was assessed using an g-cm visual analogue scale. An antacid tablet containing calcium carbonate 750 mg (TurnsEX; SmithKline Beecham Corp., Pittsburgh, PA) was supplied during each of the four l-week periods for use, as desired, up to 12 tablets daily. Study capsules were prepared by the investigational pharmacist by crushing a standard 150-mg tablet of ranitidine (Zantac) into a white powder and pouring the powder into unmarked blue capsules. The placebos were prepared by pouring a similar volume of white lactose powder (approximately 150 mg) into the blue capsules. Thus, the ranitidine and placebo containing
84
Larson et
al
Gastroesophageal R+LT
capsules appeared identical. The dosage regimen recommended by the manufacturer is one tablet before meals in the morning and in the evening. All volunteers were interviewed by one of the investigators and filled out an 18-page gastroesophageal reflux questionnaire that was created by the group at the Mayo Clinic and that has been used in multiple trials of nonpregnant heartburn patients.4 The questionnaire assessed heartburn symptoms as related to quality of life and included life-style and general health status questions. After the baseline assessment week, the eligible patients were randomized by the investigational pharmacy service using a computer-generated random number table as to the order in which each regimen was given. Each patient was given a package per week containing the following: 1) seven eight-packs of antacid tablets and 2) two containers of seven capsules each, prepared by the pharmacy, one labeled “for morning” (AM) and the other “for evening” (PM). Physicians and patients were blinded as to the contents of the containers. Each patient took one study capsule from the designated container 30-60 minutes before breakfast and dinner. Diaries were also distributed weekly. Instructions were provided about recording symptoms each day. At the end of each week, each patient completed a global assessment giving impressions of the symptoms for the entire week for heartburn intensity, regurgitation intensity, and improvement compared to her usual heartbum, on a visual analogue scale. Antacid and capsule counts were performed weekly by the principal investigator, and diaries were collected. After each patient delivered, the intrapartum and newborn records were reviewed and their outcomes were recorded. Visual analogue scales were 8 cm long, and patient responses were measured in mm (total 80 mm). A power analysis was calculated using the means and standard deviations of the baseline week heartburn severity visual analogue scales, and it was estimated that 16 patients were necessary to detect a 50% improvement with a power of 0.90 (p = .lO), and 19 patients for a power of 0.95 (/3 = .05). The results of the visual analogue scales were analyzed using means, standard errors of the mean, and standard deviations. These were analyzed using repeated measures analysis of variance, with post-hoc analysis by Tukey-Kramer multiple comparisons test. The treatment heartburn severity results were subtracted from the baseline, and placebo heartburn severity results and a percentage improvement was calculated along with the means and standard errors of the mean. The Wilcoxon signed-rank test was used to analyze the difference between the percentages. Demographic data of the study group
Obstetrics 0 Gynecology
were compared with an historic control group in the University of Oklahoma Perinatal Database, using analysis of variance and Fisher exact test. INSTAT statistical software (GraphPad Software, Inc., San Diego, CA) was used for all calculations. An (Y level of .05 was considered to be significant.
Results Between November 1995 and June 1996,25 (6.1%) of the 409 women in our obstetric population volunteered and met the study criteria. Of these, five withdrew before randomization due to concerns regarding fetal effects of medications during pregnancy (three), early delivery (one), or time constraints (one). Of the 20 remaining, 18 completed 100% of their diaries and returned all of the medication containers. Incomplete data were obtained for two patients; one delivered preterm during the first week of randomization, and the other failed to return three of her four weekly diaries. Demographic data for the 20 patients were compared with those of an historic control group of all deliveries from a l-year period (1995) in our departmental perinatal database. The study patients were older (29.1 + 6.8 versus 24.3 t 6.0 years; P < .OOl by Fisher exact test), less likely to be nulliparous (25.0% versus 29.1%; P < .03), and more likely to be white (85.0% versus 48.0%; P < .Ol) than the general population. Entry into the study was between 26 weeks’ and 36 weeks’ gestation. No patient complained of side effects attributable to the ranitidine. Improvements in heartburn were found either with
Baseline
Placebo
Ran&dine PM
Ranitidiae AM+PM
Figure 1. Global assessment of improvement of heartburn intensity based on end-of-week visual analogue scale C-P < .05 versus baseline. ‘P < ,001 versus baseline, not significant versus placebo. **P < .05 versus placebo, P < ,001 versus baseline; mean 2 standard error of the mean).
VOL.
90, NO.
1, JULY
1997
Baseline
Figure during dine in reduced .05) and reduced but not
Placebo
Rmitidine PM
Rallitidine AM+PM
2. Number of antacid tablets taken per week for heartburn pregnancy when taken with ranitidine in the PM only, ranitithe AM and PM, and placebo only. **Twice-daily ranitidine the need for antacid tablets compared with the placebo (P < with the baseline (P < .OOl). “The evening-only ranitidine dose the use of antacids when compared with the baseline (P < .05) with the placebo. Mean t standard error of the mean.
the placebo or with both dosing regimens of ranitidine. Figure 1 displays the global assessment of improvement. Compared with the baseline group, significant improvements were found using the placebo (P < ,051, evening dose of ranitidine (P < .OOl), and twice daily doses of ranitidine (P < .OOl). Ranitidine needed to be taken twice daily to allow for an improvement over the placebo treatment V’ < .05). The average reduction of heartburn intensity was 55.6% when comparing twicedaily ranitidine with baseline (95% confidence interval [CI] 34.8%, 76.5%). The mean reduction with twicedaily ranitidine, compared with placebo, was 44.2% (95% CI 15.4%, 72.9%). There was no statistical significance between improvement with the single dose of ranitidine over baseline or over placebo (P = .21). As shown in Figure 2, the number of antacid tablets taken was also less with ranitidine when given in the evening only (P < .05) and during the morning and evening (P < .Ol). Regurgitation, or the feeling of “food coming back up from the stomach,” was less intense with twice-daily ranitidine when compared with baseline (P < .05), but this reduction did not reach statistical significance when once- or twice-daily ranitidine was compared with placebo (P > .05). Heartburn intensity during the daytime and at night are compared in Figure 3. The intensity of daytime and nocturnal baseline heartburn did not differ statistically. Both once- and twice-daily regimens of ranitidine showed statistically significant decreases in nighttime heartburn compared with baseline (P < .05 for once daily, P < .OOl for twice daily), but only the twice-daily regimen reduced the intensity when compared with
Larson et al
Gastroesophageal
R@ux
85
Placebo
Ranitidine PM
R&id& Ah4+PM
Figure 3. Daytime versus nighttime heartburn. *P < .05 versus baseline. **P < ,001 versus baseline, P < .05 versus placebo. ***P < ,001 versus baseline, P < .Ol versus placebo, P < .05 versus PM only. Mean t standard error of the mean.
placebo (P < .05). Considering only daytime heartburn scores, the twice-daily regimen was superior to baseline (P < .OOl), placebo (P < .Ol), and once-daily dosing (P < .05X Three patients failed to respond to either ranitidine dose regimen compared with a placebo, although all three treatment arms improved heartburn symptoms compared with the baseline period. Enrollment into the study was at 29 weeks‘ (n = 2) and at 33 weeks’ gestation (n = 11, which was similar to those who responded to the ranitidine. Fifteen (83%) of the study cases were questioned at their 6-week postpartum exam or beyond. The heartburn symptoms subsided after delivery with no further need of antacid or ranitidine therapy in all but one case.
Discussion Heartburn is important when considering the quality of life for gravid women. Of the 25 patients completing our questionnaire, 50% noted heartburn severe enough to change their life-style. Forty-five percent stated that heartburn caused them to alter their daily activities a little or some of the time, while 25% reported that they altered their activities most or all of the time. Although more than one-half (53%) reported that heartburn existed before pregnancy, only 30% believed it was severe enough to require H, blockers, and only 20% consulted their physician regarding heartburn. The pathophysiology of gastroesophageal reflux and symptoms of heartburn during pregnancy is incom-
86 Larson et al
Gastroesophageal R$ux
pletely understood. In nonpregnant adults, gastroesophagealreflux can be traced to the following factors: 1) a decrease of the antireflux barrier, primarily the lower esophageal sphincter, which is influenced by the anatomy of the esophagogastric junction, 2) spontaneous relaxations of the lower esophageal sphincter, 3) increased volume and acidic content of the fluid refluxed, 4) decreased esophageal clearance, and 5) decreased tissue resistance of the esophageal mucosa.’ Physiologic changes due to pregnancy probably result in the increased prevalence of gastroesophageal reflux. The resting pressure of the lower esophageal sphincter is reduced during pregnancy with a decreased positive responsiveness to pharmacologic and physiologic stimulation. The increased abdominal pressure from the enlarging gravid uterus may exacerbate reflux.6 Gastrointestinal mucus secretion protects the mucosa.7In pregnancy, cervical mucus becomes thicker and more viscous in response to progesterone, and there might be a similar alteration in esophagealmucus secretion that may decrease the protective barrier. Ranitidine is a popular histamine (HJ receptor blocker used to treat peptic ulcer disease and gastroesophageal reflux as a twice daily 150 mg dose regimen.’ Ranitidine has been assigned to the Food and Drug Administration Category Bl for pregnancy, meaning that animal studies have not demonstrated a fetal risk but that there are no adequate studies in pregnant women. Preclinical testing in rats and rabbits have shown no adverse effects of ranitidine on reproductive function or pregnancy outcome.’ No antiandrogenie effects of ranitidine have been identified in male rat or human subjects, which distinguishes this agent from cimetidine, another H, blocker.“*” A single case report describes use of ranitidine in a single patient throughout her pregnancy, without any adverse fetal or neonatal sequelae.i2 Those investigators documented serum levels of ranitidine that were comparable to those in the amniotic fluid at 17 weeks, in the cord blood at delivery, and in the neonate’s blood after breastfeeding. The present study shows that twice-daily dosing of ranitidine is necessary for treating the heartburn that is associatedwith pregnancy. Our data do not support the prejudice that evening heartburn is worse because we found no differences in heartburn between night and day when either the symptoms or the number of antacids taken were used as measures of intensity. An evening-only dose of ranitidine is clearly not enough, and the twice-daily dosage was associated with a greater than 50% improvement in heartburn compared with antacids alone. This dose was effective in obtaining relief from other symptoms, such as epigastric fullness, upper abdominal discomfort, and bloating. The reduction of symptoms and antacid use indicates that the
Obstetrics b Gynecology
patients taking ranitidine experience fewer episodes of the heartburn for which they seek relief. Although many patients improved with life-style counseling and antacid therapy, their heartburn intensity was significantly less with pharmacologic treatment. Relief was shortly after ingestion of the medication, and no adverse pregnancy outcomes or drug reactions were noted. While an adequate washout period is always a concern in crossover studies, due to the time constraints in the current study, it was judged to be impractical. Because the plasma half-life of ranitidine is reported as 4.8 hours,r3 we believe the carry-over should be minimal. In addition, the sequence of the 3 trial weeks was random, so that any carry-over effects should cancel out. Three patients failed to respond to either ranitidine dose regimen compared with placebo, although all three improved when compared with the baseline period. Their entry into the study was at 29 weeks‘ (n = 2) and at 33 weeks’ gestation (n = 11, which was similar to those who responded to the ranitidine. Possible explanations for treatment failure include the following observations: 1) these patients had an enhanced placebo effect when compared with treatment successes (49.2% placebo improvement versus 9.5%, P = .02 by Fisher exact test), 2) all three received twice-daily ranitidine during the last week versus only 25% of the treatment successes (P < .04), suggesting an ordering effect, and 3) tobacco use (66.7% versus 12.5%) and less weight gain (5.0 5 0.4 kg versus 10.9 ? 8.8 kg) during pregnancy were more common for the treatment failures than for successes. No differences between the success and failure groups were found as to the presence of heartburn before pregnancy, the degree of obesity, or the intensity of baseline heartburn (7.9 2 3.1 for failures, 5.8 t 2.8 for successes, P = .25). The three patients comprised 60% of the group randomized to receive twice-daily ranitidine treatment at the last week. The trend toward less weight gain in the pregnancy may be an effect either from tobacco use or from the patient’s attempt to ingest less food to reduce reflux symptoms. Other possible explanations could include undiagnosed peptic ulcer disease or psychological pathology. Ranitidine recently was released as an over-thecounter preparation. This nonprescription strength (75 mg) is marketed for relieving and preventing heartburn and acid indigestion when taken as directed. There will probably be many questions from patients regarding the safety and efficacy of ranitidine during pregnancy. Although we did not assessthe nonprescription 75-mg dose, we can inform pregnant patients with severe heartburn symptoms that the 150-mg dose,
VOL.
90, NO.
1, JULY
1997
taken twice symptoms.
daily, is effective for reducing
heartburn
Rqerences 1. Baron
2. 3.
4.
5. 6.
7. 8.
9.
10.
11. 12.
13.
TH, Richter JE. Gastroesophageal reflux disease in pregnancy. Gastroenterol Clin North Am 1992;21:777-91. Olans LB, Wolf JL. Gastroesophageal reflux in pregnancy. Castro intest Endosc Clin North Am 1994;4:699-712. Splinter M, Nightingale B, Sawgraves R, Rayburn W. Medication use during pregnancy by women delivering at a tertiary university hospital. South Med J 1997 fin press). Locke GR, Talley NJ, Weaver AL, Zinsmeister AR. A new questionnaire for gastroesophageal reflux disease. Mayo Clin Proc 1994;69:539-47. Dodds WJ, Hogan WJ, Helm JF, Dent J. Pathogenesis of reflux esophagitis. Gastroenterology 1981;8:376-94. Scott LD. Gastrointestinal disease in pregnancy. In: Creasy RK, Resnik R, eds. Maternal-fetal medicine: principles and practice. Philadelphia (PA): WB Saunders, 1994:1026-39. Orlando RC. Esophageal epithelial defenses against acid injury. Am J Gastroenterol 1994;89:S48-52. Grant SM, Langtry HD, Brogden RN. Ranitidine: An updated review of its pharmacodynamic and pharmacokinetic properties and therapeutic use in peptic ulcer disease and other allied diseases. Drugs 1989;37:846-9. Kamada S, Sakanoue M, Iwata M, Taeuchi M, Shimpo K, Tanabe T. Effects of intravenous administration of ranitidine hydrochloride on the fertility of the male and female rats. J Toxic01 Sci 1984;9 (Suppl 1):13-28. Parker S, Schade RR, Pohl CR, Gavaler JS, Van Thiel DH. Prenatal and neonatal exposure of male rat pups to cimetidine but not ranitidine adversely affects subsequent adult sexual functioning. Gastroenterology 1984;86:675-80. Wang C, Wong KL, Lam KC, Lai CL. Ranitidine does not affect gonadal function in man. Br J Clin Pharmacol 1983;16:430-2. Armentano G, Bracco PL, Di Silverio C. Ranitidine in the treatment of reflux esophagitis in pregnancy. Clin Exp Obstet Gynecol 1989;16:130-3. Physicians’ Desk Reference. Montvale (NJ): Medical Economics Publications, 19971182-4.
Address reprint requests to: William F. Rayburn, MD Section qf Maternal-Fetal Medicine, Department qf Obstetrics and Gynecology University qf Oklahoma Health Sciences Center PO Box 26901, 4SP710 Oklahoma City, OK 73290 E-mail: [email protected]
Received October 7, 1996. Received in revised form February 20, 1997. Accepted February 28, 1997.
Copyright 0 1997 by The American College of Obstetricians Gynecologists. Published by Elsevier Science Inc.
Larsonet al
Gastroesophageal R&LX
and
87