- Email: [email protected]
Pregnancy and delivery after antireflux surgery
The American Journal of Surgery 188 (2004) 34 –38
Scientific paper
Pregnancy and delivery after antireflux surgery Rodrigo Gonzalez, M.D., Steven P. Bowers, M.D., Vickie Swafford, R.N., C. Daniel Smith, M.D.* Emory Endosurgery Unit, Department of Surgery, Emory University School of Medicine, 1364 Clifton Rd. N.E., Atlanta, GA 30322, USA Manuscript received May 30, 2003; revised manuscript October 31, 2003 Presented in part at the 44th Annual Meeting of the Society for Surgery of the Alimentary Tract, Orlando, Florida, May 17–22, 2003
Abstract Background: Concerns have been raised that subsequent pregnancy after antireflux surgery (ARS) may predispose to wrap disruption or herniation and adversely affect outcomes. Some surgeons withhold ARS in women of childbearing age for fear of this, but outcomes in this population have not been reported. Methods: All childbearing-age women who underwent ARS for gastroesophageal reflux disease (GERD) between January 1991 and July 2000 were asked to complete a detailed questionnaire. Patients with subsequent pregnancies (SP) after ARS were compared with patients without subsequent pregnancies (NP). Results: Ninety-five of the 118 patients (81%) completed the questionnaire at a mean follow-up of 4.9 years. Fifteen patients had 19 subsequent pregnancies after undergoing ARS, and retching and/or vomiting were reported during 13 of the pregnancies (69%). Preoperative incidence of complicated—GERD including strictures (11% vs. 20%), Barrett’s esophagus (19% vs. 13%), esophagitis (36% vs. 33%), and ulceration (4% vs. 0%)—were similar between the nonpregnant and pregnant groups. Incidence of postoperative moderate to severe esophageal (7% vs. 8%) and extraesophageal symptoms (0% vs. 6%) were similar between the SP and NP groups. Postoperative prevalence of antisecretory medications was similar in SP and NP groups (13% and 23%, respectively). The incidence of fundoplications redone did not reach statistical difference between the NP (11%) and SP (0%) groups. Long-term outcomes and failure rates were similar in both groups, except the SP group reported greater overall satisfaction with ARS. Conclusions: Women of childbearing age have a high incidence of complicated GERD, which may contribute to higher-than-expected rates of symptomatic and anatomic fundoplication failures than first-time ARS. Subsequent pregnancies do not adversely affect outcomes after ARS. © 2004 Excerpta Medica, Inc. All rights reserved. Keywords: Antireflux surgery; Childbearing-age women; Fundoplication failure; Gastroesophageal reflux disease; Increased abdominal pressure; Pregnancy
Gastroesophageal reflux disease (GERD) causes symptoms of heartburn and indigestion in 60 million Americans and accounts for three fourths of all esophageal dysfunction. Antireflux surgery (ARS) is a well-established and successful procedure for the treatment of GERD with long-term control of symptoms in ⬎90% of patients after Nissen fundoplication [1–3]. Since the report of the first laparoscopic Nissen operation by Dallemagne in 1914, laparoscopic ARS has been performed with increased frequency. * Corresponding author. Tel.: ⫹ 1-404-727-1540; fax: ⫹ 1-404-7125416. E-mail address: [email protected]
With the advent of minimally invasive procedures and the development of new surgical techniques, laparoscopic fundoplication has become the standard procedure in the surgical treatment of GERD. Several prospective studies have demonstrated that the operation can achieve excellent symptomatic results and is associated with low rates of morbidity and mortality [1–3]. Increased intraabdominal pressure (IAP) may have adverse effects in patients after fundoplication. Although uncommon, wrap disruption and herniation have been reported after a period of sudden IAP [4]. Wrap herniation has been found to be the most common cause of anatomic failure after laparoscopic fundoplication [4,5]. In our experience,
0002-9610/04/$ – see front matter © 2004 Excerpta Medica, Inc. All rights reserved. doi:10.1016/j.amjsurg.2003.10.024
R. Gonzalez et al. / The American Journal of Surgery 188 (2004) 34 –38
59% of patients with failed fundoplication caused by transdiaphragmatic-wrap herniation were able to identify an acute event (i.e., retching, vomiting, coughing, or straining to lift) that caused the herniation [5]. Gastrointestinal tract physiology is commonly affected by the hormonal changes that occur during pregnancy. Nausea and vomiting during pregnancy (NVP) is a common experience that affecting 50% to 90% of all women [6]. Concerns have been raised that IAP during pregnancy, more specifically NVP, and/or delivery are mechanical consequences of pregnancy that can profoundly impact patients who have undergone antireflux procedures by leading to wrap disruption or herniation and thus adversely affect outcomes. Many physicians who manage GERD frequently think these changes put an antireflux procedure at such risk for failure during or after pregnancy that ARS is contraindicated in women of childbearing age. However, outcomes in this patient population have not been reported. The aim of this study was to investigate the results of ARS in patients who subsequently became pregnant and to evaluate the possible adverse mechanical effects of pregnancy on outcome after ARS.
35
(NP group). Because most NVP occurs during the beginning of the pregnancy, patients who became pregnant but miscarried were also included in the SP group. To establish whether the pregnancy was perhaps causative in postoperative GERD symptoms, we compared within each group the symptoms before ARS with those after surgery but before pregnancy and with those after pregnancy. Symptoms were also compared between the 2 groups. Additionally, we evaluated the severity of preoperative GERD, characteristics of the pregnancy, and type of delivery. Statistical analysis Continuous data were analyzed using Mann-Whitney U-test or analysis of variance for repeated measures as appropriate. Categorical data were analyzed using Fisher’s exact test. Ordinal data were compared using the Wilcoxon matched-pairs signed-ranks test, and P ⬍0.05 was considered statistically significant. Results were reported as mean ⫾ standard error of the mean.
Results Methods A chart review of all women of childbearing age (aged 16 to 40 years) who underwent a fundoplication for GERD from January 1991 to July 2001 was conducted in our prospectively maintained foregut surgery database. Preoperative symptoms in all patients were evaluated by means of a standardized questionnaire. Patients were also asked to complete the same questionnaire during their postoperative clinic visits. The symptoms routinely evaluated include heartburn, dysphagia, regurgitation, chest pain, cough, hoarseness, and asthma. Patients were asked to grade the severity of their symptoms as none (0), mild (1), moderate (2), or severe (3). As part of this study, the same questionnaire was sent by mail to all women of childbearing age identified in our database. If patients did not return the questionnaire, they were contacted by phone. A Social Security Number search was used to ensure current and accurate contact information. In addition to the symptom score questionnaire, inquiries were also made about the use of antacid medication and indications for its use (i.e., subjective vs. objective), type of medication, duration of its use, and symptomatic improvement with the medication. Finally, subjects were questioned about problems during and after their latest pregnancy and delivery, specifically about nausea and vomiting, type of delivery, weight of the baby, and severity of their GERD symptoms during and after pregnancy and delivery. Subjects in this study were divided into 2 groups. The first group consisted of patients who had become pregnant after ARS (SP group). The control group consisted of patients who had not become pregnant after fundoplication
From our database of ⬎1,238 patients undergoing ARS between January 1991 and July 2001, we identified 118 women of childbearing age (9.5%), and 101 (86%) were accounted for on follow-up. Three patients refused to participate in the study, and 3 had died of causes unrelated to the surgery. Herein we report the results in 95 patients. Fifteen patients (16%) had 19 subsequent pregnancies after undergoing ARS and constituted the SP group. The remaining 80 patients constituted the NP group. There was no difference in mean follow-up time between the SP and NP groups (62 ⫾ 6 and 58 ⫾ 3 months, respectively). Patients in the NP group were older than those in the SP group (33 ⫾ 0.6 vs. 28 ⫾ 1.4 years, respectively; P ⬍0.05). The average time interval between ARS and last pregnancy was 34 ⫾ 6 months. Four of the patients became pregnant ⱖ2 times after ARS, and only 1 twin pregnancy was reported. Retching and/or vomiting were reported during 13 of the pregnancies (69%). Patients reported the following incidence of moderate to severe symptoms (scores 2 or 3) during their pregnancies: heartburn (53%), regurgitation (42%), chest pain (16%), and dysphagia (11%). Seven patients required treatment for NVP, and 2 patients were hospitalized because of dehydration caused by severe vomiting. No patient was prescribed antisecretory medication during pregnancy. The pregnancies resulted in 6 miscarriages and 13 deliveries. The type of delivery was vaginal in 9 patients and cesarean section in 4 patients. The mean weight of the babies was 6 lb 12 oz. Only 2 patients (13%) reported changes in GERD symptoms after delivery, and in both of these, medication resulted in complete resolution of symptoms.
36
R. Gonzalez et al. / The American Journal of Surgery 188 (2004) 34 –38
Table 1 Comparison of preoperative GERD complications between groups Complications Stricture Barrett’s Esophagitis Ulceration
NP group (%)
SP group (%)
P*
9 (11) 15 (19) 29 (36) 3 (4)
3 (20) 2 (13) 5 (33) 0
NS NS NS NS
Total (%) 12 (13) 17 (18) 34 (36) 3 (3)
* Fisher’s Exact test. GERD ⫽ gastroesophageal reflux disease; NP ⫽ nonpregnant; SP ⫽ subsequent pregnancy.
Table 1 lists the comparison of the groups with regard to severity of GERD before surgery. The endoscopic evaluation revealed that there was no difference between groups in the preoperative incidence of strictures, Barrett’s esophagus, esophagitis, or esophageal ulcerations. Table 2 lists the groups with regard to objective preoperative testing. The preoperative esophageal (chest pain, heartburn, dysphagia, and regurgitation) and extraesophageal (cough, hoarseness, and asthma) symptom scores were similar in both groups. The esophageal and extraesophageal symptom scores at the last visit were also similar in both groups (Table 3). Office follow-up data is listed in Table 3. Average follow-up at last office visit was 18 ⫾ 2 and 14 ⫾ 5 in the NP and SP groups, respectively (P ⫽ not significant). All GERD symptoms significantly improved after ARS. This improvement resulted in a statistically significant decrease in the number of patients with severe to moderate esophageal symptom scores at the last office visit. There was no difference between the SP and NP groups. The number of patients with moderate to severe extraesophageal symptom scores also improved at the last office visit with the exception of asthma in the NP group. There was also no difference between groups in improvement of extraesophageal symptoms. The number of patients who were currently taking antisecretory medications for reflux in the NP group (23%) was similar to the number in the SP group (13%). In only 20% of these patients (4 and 0 patients in the NP and SP groups, respectively; P ⫽ not significant) was the indication for antisecretory medication determined objectively by either endoscopic evaluation or pH studies. The medications were Table 2 Results of preoperative studies Results
NP group (%)
SP group (%)
P
Total
Hiatal hernia pH ⬍4 (%) Delayed gastric emptying Helicobacter pylori positive
23 (29) 15 ⫾ 2 8/30 (27) 0/45
3 (20) 7⫾1 2/5 (40) 0/8
NS* NS† NS* NS*
26 (27) 14 ⫾ 2 10/35 (29) 0
* Mann-Whitney U test; † Fisher’s Exact test. NP ⫽ nonpregnant; NS ⫽ Not statistically significant; SP ⫽ subsequent pregnancy.
Table 3 Comparison of preoperative and postoperative moderate to severe gastroesophageal reflux symptoms between groups Symptom
Esophageal Heartburn Regurgitation Dysphagia Chest pain Extraesophageal Hoarseness Cough Asthma
Preoperative
Last follow-up
NP group (%)
SP group (%)
NP group (%)
SP group (%)
73 (91) 52 (65) 28 (35) 24 (30)
14 (93) 9 (60) 6 (40) 9 (60)
10 (12) 5 (6) 7 (9) 6 (8)
1 (7) 1 (7) 0 1 (7)
17 (21) 28 (35) 6 (8)
8 (53) 6 (40) 2 (13)
2 (3) 4 (5) 3 (4)
0 0 0
NP ⫽ nonpregnant; SP ⫽ subsequent pregnancy.
taken on a daily basis by 16 of these patients (80%). Nine additional patients (9%) required treatment for nausea or vomiting. Nine patients (9%) required ⱖ dilation after ARS. Although all patients requiring surgery to be redone for recurrent symptoms (9%) were in the NP group, there was no statistical difference between groups (Table 4). Seven of these patients (78%) presented with postoperative transdiaphragmatic-wrap herniation. Six of these patients (67%) could identify an acute episode of vomiting before recurrence of their symptoms. Overall satisfaction after ARS in the SP group (100%) was significantly higher than in the NP group (76%) (P ⬍0.05). Other gastrointestinal symptoms reported by patients after ARS are listed in Table 5. The incidence of esophageal and extraesophageal moderate to severe GERD symptoms were similar between groups. There was also a similar incidence of bloating, nausea, diarrhea, and dumping between the 2 groups. In both groups, overall heartburn was improved or eliminated in 87%, regurgitation in 90%, chest pain in 82%, and dysphagia in 81% of the patients. Cough was improved or eliminated in 85% , hoarseness in 84%, and asthma in 63% of patients (Figs. 1 and 2). Table 4 Additional treatments required by patients in both groups after ARS Treatments Antisecretory medication Symptoms only Positive endoscopy Positive pH study Total Antiemetics Dilations Cholecystectomy Pyloroplasty Redo ARS
NP group (%)
SP group (%)
P*
Total (%)
14 (78) 3 (22) 1 (6) 18 (23) 9 (11) 8 (10) 6 (8) 1 (1) 9 (11)
2 (100) 0 0 2 (13) 0 1 (7) 1 (7) 0 0
NS NS NS NS NS NS NS NS NS
16 (17) 3 (3) 1 (1) 20 (21) 9 (9) 9 (9) 7 (7) 1 (1) 9 (9)
* Fisher’s Exact test. ARS ⫽ antireflux surgery; NP ⫽ nonpregnant; NS ⫽ not statistically significant; SP ⫽ subsequent pregnancy.
R. Gonzalez et al. / The American Journal of Surgery 188 (2004) 34 –38
37
Table 5 Gastrointestinal symptoms reported by patients after ARS Symptoms Moderate/severe symptoms Esophageal Extraesophageal Bloating Nausea Diarrhea Dumping
NP group (%)
SP group (%)
P*
Total (%)
14 (18) 5 (6) 14 (18) 10 (13) 3 (4) 1 (1)
1 (7) 0 3 (20) 2 (13) 1 (7) 0
NS NS 0 NS NS NS
15 (16) 5 (5) 17 (18) 12 (13) 4 (4) 1 (1)
* Fisher’s Exact test. ARS ⫽ antireflux surgery; NP ⫽ nonpregnant; NS ⫽ not statistically significant; SP ⫽ subsequent pregnancy.
Comments Rates of fundoplication failure requiring reoperation reported in the literature range between 2% and 17% [4,5,7,13–17] depending on the definition of failure and the length of follow-up. Hunter et al [5] reported anatomic failure in 7% of patients at a mean follow-up of 28 months. Six percent of the patients experienced postoperative vomiting within 1 week of surgery, and 17% of these patients diagnosed with wrap herniation immediately after the episode of vomiting. Their anatomic fundoplication failures were associated with not dividing short gastric vessels, IAP caused by early postoperative vomiting and other diaphragmatic stressors, and size of the hiatal hernia. Postoperative symptoms—including chest pain (66%), dysphagia (7%), heartburn (5%), and regurgitation (3%)—were present in 26% of patients. GERD symptoms were present despite an intact fundoplication in 3.4% of patients. Based on this and other experiences, many investigators have believed that pregnancy should predict a poor outcome after ARS because of the significant mechanical effects of pregnancy related to nausea and vomiting and increased IAP related to the gravid uterus and delivery. Nothing has
Fig. 1. Comparison between preoperative (Preop) and postoperative (Postop) patients with moderate to severe gastroesophageal reflux disease symptoms in the nonpregnant group. f/u ⫽ follow-up.
Fig. 2. Comparison between preoperative and postoperative patients with moderate to severe gastroesophageal reflux disease symptoms in the subsequent pregnancy group. f/u ⫽ follow-up.
been reported in the literature on this topic. The data reported in this series dispel the concern regarding pregnancy after ARS leading to a high rate of ARS failure. In our experience [4,5], diaphragmatic herniation of the fundoplication is the most common anatomic failure after ARS. Sudden IAP caused by vomiting has been recognized by many as the etiology of wrap disruption or herniation and has been described to adversely affect fundoplication outcomes even during the early postoperative period [4,5,7,8]. Intense efforts have been taken to try to prevent early postoperative nausea and vomiting (PONV), but despite advances in anesthetics and newer antiemetic medications, it still occurs in 20% to 30% of patients undergoing general anesthesia [9]. PONV occurs more frequently in female patients (33%) with 84% of women still having documented PONV despite receiving antiemetic medication [10]. Some of the physiologic changes that occur in the upper gastrointestinal tract during pregnancy and considered to play an important role in the etiology of NVP are the same changes presumed to cause GERD during pregnancy. Esophageal, gastric, and small-bowel motility are impaired as a result of smooth-muscle relaxation promoted by increased levels of female sex hormones. In the esophagus, hormonal changes are postulated to alter lower esophageal sphincter function, thus causing an incompetent sphincter. This alteration may contribute not only to heartburn but also to NVP. Progesterone, a smooth-muscle relaxant, inhibits gastric emptying in early pregnancy when levels are highly increased, which may also be an important factor in NVP. In late pregnancy, the large gravid uterus contributes to the manifestations of upper gastrointestinal symptoms by mechanically compressing the stomach. Clearly, pregnancy and its associated mechanical and hormonal changes are known to have consequences on gastrointestinal function [6,11,12]. Therefore, it would be reasonable to think that IAP during pregnancy, NVP complicating pregnancy, and delivery itself could have a deleterious effect on outcomes after ARS. This was not the case according to the data obtained
38
R. Gonzalez et al. / The American Journal of Surgery 188 (2004) 34 –38
from our study. Our results showed that patients experienced a significant improvement in GERD symptoms after an average of 17 months post-ARS (follow-up at the clinic). This allowed us to assess symptom scores after ARS, but before pregnancy, and compare them with those obtained during the last follow-up visit (after pregnancy). These data are important because they allow us to have 2 different parameters for comparison. To be certain that it was pregnancy that altered ARS outcomes and GERD symptoms, we needed to evaluate if symptoms had improved after ARS and then worsened after pregnancy. We also needed a control group against which to compare long-term changes in symptom scores from patients who did not get pregnant. Heartburn was the only symptom in which the number of patients with moderate to severe scores decreased after ARS and then slightly increased with longer follow-up; this occurred only in the NP group and not the group considered at risk, i.e., the SP group. The age difference between groups is expected because the younger the patient at the time of surgery, the higher the probability for her to become pregnant. Although age in the NP group was statistically significant, a 5-year age difference is not considered to be clinically significant enough to alter outcomes by itself. In our experience, other than the finding that pregnancy did not predict a poor outcome after ARS, the most notable finding is the high rate of ARS failure in women of childbearing age. Despite the groups being matched with regard to severity of GERD and its symptoms, 23% of the NP group were taking antisecretory medications compared with only 13% of the SP group, and 11% of the NP group eventually required ARS to be redone compared with none of the patients in the SP group. This difference did not reach statistical significance because of the small numbers. Perhaps more notable is the fact that in our overall series up to date of ⬎1,600 patients undergoing ARS, only 2.8% overall required ARS to be redone. It appears that young women may be at a higher risk of needing ARS to be redone than others regardless of whether or not they subsequently become pregnant. In conclusion, pregnancy after ARS does not predict a higher-than-expected rate of ARS failure. Therefore, ARS should not be withheld from women of childbearing age for this reason. In contrast, women of childbearing age, regardless of whether they eventually get pregnant, do appear to
have a higher-than-expected rate of repeat ARS. The reason for this is unclear, but it is reasonable to suggest that this be considered when offering this patient group ARS as well as be discussed with the patient. References [1] Hinder RA. Surgical therapy for GERD. Selection of procedures, short- and long-term results. J Clin Gastroenterol 2000;30(suppl): S48 –S50. [2] Lafullarde T, Watson DI, Jamieson GG, et al. Laparoscopic Nissen fundoplication. Five-year results and beyond. Arch Surg 2001;136: 180 – 4. [3] Trus TL, Laycock WS, Branum G, et al. Intermediate follow-up of laparoscopic antireflux surgery. Am J Surg 1996;171:32–5. [4] Hunter JG. Approach and management of patients with recurrent gastroesophageal reflux disease. J Gastrointest Surg 2001;5:451–7. [5] Hunter JG, Smith CD, Branum GD, et al. Laparoscopic fundoplication failures. Patterns of failure and response to fundoplication revision. Ann Surg 1999;230:595– 606. [6] Broussard CN, Richter JE. Nausea and vomiting of pregnancy. Gastroenterol Clin North Am 1998;27:123–51. [7] Soper NJ, Dunnegan D. Anatomic fundoplication failure after laparoscopic antireflux surgery. Ann Surg 1999;229:669 –77. [8] Awad ZT, Anderson PI, Sato K, et al. Laparoscopic reoperative antireflux surgery. Surg Endosc 2001;15:1401–7. [9] Scuderi PE, James RL, Harris L, Mims GR. Antiemetic prophylaxis does not improve outcomes after outpatient surgery when compared to symptomatic treatment. Anesthesiology 1999;90:360 –71. [10] Bradshaw WA, Gregory BC, Finley C, et al. Frequency of postoperative nausea and vomiting in patients undergoing laparoscopic foregut surgery. Surg Endosc 2002;16:777– 80. [11] Chin RK. Nausea and vomiting of early pregnancy and pregnancy outcome. An epidemiological study. Br J Obstet Gyneacol 1990;97: 278. [12] Klebanoff MA, Koslowe PA, Kaslow R, et al. Epidemiology of vomiting in early pregnancy. Obstet Gynecol 1985;66:612. [13] Hinder RA, Filipi CJ, Wetscher G, et al. Laparoscopic Nissen fundoplication is an effective treatment for gastroesophageal reflux disease. Ann Surg 1994;220:472– 83. [14] DePaula AL, Hashiba K, Bafutto M, Machado CA. Laparoscopic reoperations after failed and complicated antireflux operations. Surg Endosc 1995;9:681– 86. [15] Cushieri A, Hunter JG, Wolfe B, et al. Multicenter prospective evaluation of laparoscopic antireflux surgery. Surg Endosc 1995;7: 505–10. [16] Hunter JG, Trus TL, Branum GD, et al. A physiologic approach to laparoscopic fundoplication for gastroesophageal reflux disease. Ann Surg 1996;223:673– 87. [17] Liu JY, Woloshin S, Laycock WS, Schwartz LM. Late outcomes after laparoscopic surgery for gastroesophageal reflux. Arch Surg 2002; 137:397– 401.
Scientific paper
Pregnancy and delivery after antireflux surgery Rodrigo Gonzalez, M.D., Steven P. Bowers, M.D., Vickie Swafford, R.N., C. Daniel Smith, M.D.* Emory Endosurgery Unit, Department of Surgery, Emory University School of Medicine, 1364 Clifton Rd. N.E., Atlanta, GA 30322, USA Manuscript received May 30, 2003; revised manuscript October 31, 2003 Presented in part at the 44th Annual Meeting of the Society for Surgery of the Alimentary Tract, Orlando, Florida, May 17–22, 2003
Abstract Background: Concerns have been raised that subsequent pregnancy after antireflux surgery (ARS) may predispose to wrap disruption or herniation and adversely affect outcomes. Some surgeons withhold ARS in women of childbearing age for fear of this, but outcomes in this population have not been reported. Methods: All childbearing-age women who underwent ARS for gastroesophageal reflux disease (GERD) between January 1991 and July 2000 were asked to complete a detailed questionnaire. Patients with subsequent pregnancies (SP) after ARS were compared with patients without subsequent pregnancies (NP). Results: Ninety-five of the 118 patients (81%) completed the questionnaire at a mean follow-up of 4.9 years. Fifteen patients had 19 subsequent pregnancies after undergoing ARS, and retching and/or vomiting were reported during 13 of the pregnancies (69%). Preoperative incidence of complicated—GERD including strictures (11% vs. 20%), Barrett’s esophagus (19% vs. 13%), esophagitis (36% vs. 33%), and ulceration (4% vs. 0%)—were similar between the nonpregnant and pregnant groups. Incidence of postoperative moderate to severe esophageal (7% vs. 8%) and extraesophageal symptoms (0% vs. 6%) were similar between the SP and NP groups. Postoperative prevalence of antisecretory medications was similar in SP and NP groups (13% and 23%, respectively). The incidence of fundoplications redone did not reach statistical difference between the NP (11%) and SP (0%) groups. Long-term outcomes and failure rates were similar in both groups, except the SP group reported greater overall satisfaction with ARS. Conclusions: Women of childbearing age have a high incidence of complicated GERD, which may contribute to higher-than-expected rates of symptomatic and anatomic fundoplication failures than first-time ARS. Subsequent pregnancies do not adversely affect outcomes after ARS. © 2004 Excerpta Medica, Inc. All rights reserved. Keywords: Antireflux surgery; Childbearing-age women; Fundoplication failure; Gastroesophageal reflux disease; Increased abdominal pressure; Pregnancy
Gastroesophageal reflux disease (GERD) causes symptoms of heartburn and indigestion in 60 million Americans and accounts for three fourths of all esophageal dysfunction. Antireflux surgery (ARS) is a well-established and successful procedure for the treatment of GERD with long-term control of symptoms in ⬎90% of patients after Nissen fundoplication [1–3]. Since the report of the first laparoscopic Nissen operation by Dallemagne in 1914, laparoscopic ARS has been performed with increased frequency. * Corresponding author. Tel.: ⫹ 1-404-727-1540; fax: ⫹ 1-404-7125416. E-mail address: [email protected]
With the advent of minimally invasive procedures and the development of new surgical techniques, laparoscopic fundoplication has become the standard procedure in the surgical treatment of GERD. Several prospective studies have demonstrated that the operation can achieve excellent symptomatic results and is associated with low rates of morbidity and mortality [1–3]. Increased intraabdominal pressure (IAP) may have adverse effects in patients after fundoplication. Although uncommon, wrap disruption and herniation have been reported after a period of sudden IAP [4]. Wrap herniation has been found to be the most common cause of anatomic failure after laparoscopic fundoplication [4,5]. In our experience,
0002-9610/04/$ – see front matter © 2004 Excerpta Medica, Inc. All rights reserved. doi:10.1016/j.amjsurg.2003.10.024
R. Gonzalez et al. / The American Journal of Surgery 188 (2004) 34 –38
59% of patients with failed fundoplication caused by transdiaphragmatic-wrap herniation were able to identify an acute event (i.e., retching, vomiting, coughing, or straining to lift) that caused the herniation [5]. Gastrointestinal tract physiology is commonly affected by the hormonal changes that occur during pregnancy. Nausea and vomiting during pregnancy (NVP) is a common experience that affecting 50% to 90% of all women [6]. Concerns have been raised that IAP during pregnancy, more specifically NVP, and/or delivery are mechanical consequences of pregnancy that can profoundly impact patients who have undergone antireflux procedures by leading to wrap disruption or herniation and thus adversely affect outcomes. Many physicians who manage GERD frequently think these changes put an antireflux procedure at such risk for failure during or after pregnancy that ARS is contraindicated in women of childbearing age. However, outcomes in this patient population have not been reported. The aim of this study was to investigate the results of ARS in patients who subsequently became pregnant and to evaluate the possible adverse mechanical effects of pregnancy on outcome after ARS.
35
(NP group). Because most NVP occurs during the beginning of the pregnancy, patients who became pregnant but miscarried were also included in the SP group. To establish whether the pregnancy was perhaps causative in postoperative GERD symptoms, we compared within each group the symptoms before ARS with those after surgery but before pregnancy and with those after pregnancy. Symptoms were also compared between the 2 groups. Additionally, we evaluated the severity of preoperative GERD, characteristics of the pregnancy, and type of delivery. Statistical analysis Continuous data were analyzed using Mann-Whitney U-test or analysis of variance for repeated measures as appropriate. Categorical data were analyzed using Fisher’s exact test. Ordinal data were compared using the Wilcoxon matched-pairs signed-ranks test, and P ⬍0.05 was considered statistically significant. Results were reported as mean ⫾ standard error of the mean.
Results Methods A chart review of all women of childbearing age (aged 16 to 40 years) who underwent a fundoplication for GERD from January 1991 to July 2001 was conducted in our prospectively maintained foregut surgery database. Preoperative symptoms in all patients were evaluated by means of a standardized questionnaire. Patients were also asked to complete the same questionnaire during their postoperative clinic visits. The symptoms routinely evaluated include heartburn, dysphagia, regurgitation, chest pain, cough, hoarseness, and asthma. Patients were asked to grade the severity of their symptoms as none (0), mild (1), moderate (2), or severe (3). As part of this study, the same questionnaire was sent by mail to all women of childbearing age identified in our database. If patients did not return the questionnaire, they were contacted by phone. A Social Security Number search was used to ensure current and accurate contact information. In addition to the symptom score questionnaire, inquiries were also made about the use of antacid medication and indications for its use (i.e., subjective vs. objective), type of medication, duration of its use, and symptomatic improvement with the medication. Finally, subjects were questioned about problems during and after their latest pregnancy and delivery, specifically about nausea and vomiting, type of delivery, weight of the baby, and severity of their GERD symptoms during and after pregnancy and delivery. Subjects in this study were divided into 2 groups. The first group consisted of patients who had become pregnant after ARS (SP group). The control group consisted of patients who had not become pregnant after fundoplication
From our database of ⬎1,238 patients undergoing ARS between January 1991 and July 2001, we identified 118 women of childbearing age (9.5%), and 101 (86%) were accounted for on follow-up. Three patients refused to participate in the study, and 3 had died of causes unrelated to the surgery. Herein we report the results in 95 patients. Fifteen patients (16%) had 19 subsequent pregnancies after undergoing ARS and constituted the SP group. The remaining 80 patients constituted the NP group. There was no difference in mean follow-up time between the SP and NP groups (62 ⫾ 6 and 58 ⫾ 3 months, respectively). Patients in the NP group were older than those in the SP group (33 ⫾ 0.6 vs. 28 ⫾ 1.4 years, respectively; P ⬍0.05). The average time interval between ARS and last pregnancy was 34 ⫾ 6 months. Four of the patients became pregnant ⱖ2 times after ARS, and only 1 twin pregnancy was reported. Retching and/or vomiting were reported during 13 of the pregnancies (69%). Patients reported the following incidence of moderate to severe symptoms (scores 2 or 3) during their pregnancies: heartburn (53%), regurgitation (42%), chest pain (16%), and dysphagia (11%). Seven patients required treatment for NVP, and 2 patients were hospitalized because of dehydration caused by severe vomiting. No patient was prescribed antisecretory medication during pregnancy. The pregnancies resulted in 6 miscarriages and 13 deliveries. The type of delivery was vaginal in 9 patients and cesarean section in 4 patients. The mean weight of the babies was 6 lb 12 oz. Only 2 patients (13%) reported changes in GERD symptoms after delivery, and in both of these, medication resulted in complete resolution of symptoms.
36
R. Gonzalez et al. / The American Journal of Surgery 188 (2004) 34 –38
Table 1 Comparison of preoperative GERD complications between groups Complications Stricture Barrett’s Esophagitis Ulceration
NP group (%)
SP group (%)
P*
9 (11) 15 (19) 29 (36) 3 (4)
3 (20) 2 (13) 5 (33) 0
NS NS NS NS
Total (%) 12 (13) 17 (18) 34 (36) 3 (3)
* Fisher’s Exact test. GERD ⫽ gastroesophageal reflux disease; NP ⫽ nonpregnant; SP ⫽ subsequent pregnancy.
Table 1 lists the comparison of the groups with regard to severity of GERD before surgery. The endoscopic evaluation revealed that there was no difference between groups in the preoperative incidence of strictures, Barrett’s esophagus, esophagitis, or esophageal ulcerations. Table 2 lists the groups with regard to objective preoperative testing. The preoperative esophageal (chest pain, heartburn, dysphagia, and regurgitation) and extraesophageal (cough, hoarseness, and asthma) symptom scores were similar in both groups. The esophageal and extraesophageal symptom scores at the last visit were also similar in both groups (Table 3). Office follow-up data is listed in Table 3. Average follow-up at last office visit was 18 ⫾ 2 and 14 ⫾ 5 in the NP and SP groups, respectively (P ⫽ not significant). All GERD symptoms significantly improved after ARS. This improvement resulted in a statistically significant decrease in the number of patients with severe to moderate esophageal symptom scores at the last office visit. There was no difference between the SP and NP groups. The number of patients with moderate to severe extraesophageal symptom scores also improved at the last office visit with the exception of asthma in the NP group. There was also no difference between groups in improvement of extraesophageal symptoms. The number of patients who were currently taking antisecretory medications for reflux in the NP group (23%) was similar to the number in the SP group (13%). In only 20% of these patients (4 and 0 patients in the NP and SP groups, respectively; P ⫽ not significant) was the indication for antisecretory medication determined objectively by either endoscopic evaluation or pH studies. The medications were Table 2 Results of preoperative studies Results
NP group (%)
SP group (%)
P
Total
Hiatal hernia pH ⬍4 (%) Delayed gastric emptying Helicobacter pylori positive
23 (29) 15 ⫾ 2 8/30 (27) 0/45
3 (20) 7⫾1 2/5 (40) 0/8
NS* NS† NS* NS*
26 (27) 14 ⫾ 2 10/35 (29) 0
* Mann-Whitney U test; † Fisher’s Exact test. NP ⫽ nonpregnant; NS ⫽ Not statistically significant; SP ⫽ subsequent pregnancy.
Table 3 Comparison of preoperative and postoperative moderate to severe gastroesophageal reflux symptoms between groups Symptom
Esophageal Heartburn Regurgitation Dysphagia Chest pain Extraesophageal Hoarseness Cough Asthma
Preoperative
Last follow-up
NP group (%)
SP group (%)
NP group (%)
SP group (%)
73 (91) 52 (65) 28 (35) 24 (30)
14 (93) 9 (60) 6 (40) 9 (60)
10 (12) 5 (6) 7 (9) 6 (8)
1 (7) 1 (7) 0 1 (7)
17 (21) 28 (35) 6 (8)
8 (53) 6 (40) 2 (13)
2 (3) 4 (5) 3 (4)
0 0 0
NP ⫽ nonpregnant; SP ⫽ subsequent pregnancy.
taken on a daily basis by 16 of these patients (80%). Nine additional patients (9%) required treatment for nausea or vomiting. Nine patients (9%) required ⱖ dilation after ARS. Although all patients requiring surgery to be redone for recurrent symptoms (9%) were in the NP group, there was no statistical difference between groups (Table 4). Seven of these patients (78%) presented with postoperative transdiaphragmatic-wrap herniation. Six of these patients (67%) could identify an acute episode of vomiting before recurrence of their symptoms. Overall satisfaction after ARS in the SP group (100%) was significantly higher than in the NP group (76%) (P ⬍0.05). Other gastrointestinal symptoms reported by patients after ARS are listed in Table 5. The incidence of esophageal and extraesophageal moderate to severe GERD symptoms were similar between groups. There was also a similar incidence of bloating, nausea, diarrhea, and dumping between the 2 groups. In both groups, overall heartburn was improved or eliminated in 87%, regurgitation in 90%, chest pain in 82%, and dysphagia in 81% of the patients. Cough was improved or eliminated in 85% , hoarseness in 84%, and asthma in 63% of patients (Figs. 1 and 2). Table 4 Additional treatments required by patients in both groups after ARS Treatments Antisecretory medication Symptoms only Positive endoscopy Positive pH study Total Antiemetics Dilations Cholecystectomy Pyloroplasty Redo ARS
NP group (%)
SP group (%)
P*
Total (%)
14 (78) 3 (22) 1 (6) 18 (23) 9 (11) 8 (10) 6 (8) 1 (1) 9 (11)
2 (100) 0 0 2 (13) 0 1 (7) 1 (7) 0 0
NS NS NS NS NS NS NS NS NS
16 (17) 3 (3) 1 (1) 20 (21) 9 (9) 9 (9) 7 (7) 1 (1) 9 (9)
* Fisher’s Exact test. ARS ⫽ antireflux surgery; NP ⫽ nonpregnant; NS ⫽ not statistically significant; SP ⫽ subsequent pregnancy.
R. Gonzalez et al. / The American Journal of Surgery 188 (2004) 34 –38
37
Table 5 Gastrointestinal symptoms reported by patients after ARS Symptoms Moderate/severe symptoms Esophageal Extraesophageal Bloating Nausea Diarrhea Dumping
NP group (%)
SP group (%)
P*
Total (%)
14 (18) 5 (6) 14 (18) 10 (13) 3 (4) 1 (1)
1 (7) 0 3 (20) 2 (13) 1 (7) 0
NS NS 0 NS NS NS
15 (16) 5 (5) 17 (18) 12 (13) 4 (4) 1 (1)
* Fisher’s Exact test. ARS ⫽ antireflux surgery; NP ⫽ nonpregnant; NS ⫽ not statistically significant; SP ⫽ subsequent pregnancy.
Comments Rates of fundoplication failure requiring reoperation reported in the literature range between 2% and 17% [4,5,7,13–17] depending on the definition of failure and the length of follow-up. Hunter et al [5] reported anatomic failure in 7% of patients at a mean follow-up of 28 months. Six percent of the patients experienced postoperative vomiting within 1 week of surgery, and 17% of these patients diagnosed with wrap herniation immediately after the episode of vomiting. Their anatomic fundoplication failures were associated with not dividing short gastric vessels, IAP caused by early postoperative vomiting and other diaphragmatic stressors, and size of the hiatal hernia. Postoperative symptoms—including chest pain (66%), dysphagia (7%), heartburn (5%), and regurgitation (3%)—were present in 26% of patients. GERD symptoms were present despite an intact fundoplication in 3.4% of patients. Based on this and other experiences, many investigators have believed that pregnancy should predict a poor outcome after ARS because of the significant mechanical effects of pregnancy related to nausea and vomiting and increased IAP related to the gravid uterus and delivery. Nothing has
Fig. 1. Comparison between preoperative (Preop) and postoperative (Postop) patients with moderate to severe gastroesophageal reflux disease symptoms in the nonpregnant group. f/u ⫽ follow-up.
Fig. 2. Comparison between preoperative and postoperative patients with moderate to severe gastroesophageal reflux disease symptoms in the subsequent pregnancy group. f/u ⫽ follow-up.
been reported in the literature on this topic. The data reported in this series dispel the concern regarding pregnancy after ARS leading to a high rate of ARS failure. In our experience [4,5], diaphragmatic herniation of the fundoplication is the most common anatomic failure after ARS. Sudden IAP caused by vomiting has been recognized by many as the etiology of wrap disruption or herniation and has been described to adversely affect fundoplication outcomes even during the early postoperative period [4,5,7,8]. Intense efforts have been taken to try to prevent early postoperative nausea and vomiting (PONV), but despite advances in anesthetics and newer antiemetic medications, it still occurs in 20% to 30% of patients undergoing general anesthesia [9]. PONV occurs more frequently in female patients (33%) with 84% of women still having documented PONV despite receiving antiemetic medication [10]. Some of the physiologic changes that occur in the upper gastrointestinal tract during pregnancy and considered to play an important role in the etiology of NVP are the same changes presumed to cause GERD during pregnancy. Esophageal, gastric, and small-bowel motility are impaired as a result of smooth-muscle relaxation promoted by increased levels of female sex hormones. In the esophagus, hormonal changes are postulated to alter lower esophageal sphincter function, thus causing an incompetent sphincter. This alteration may contribute not only to heartburn but also to NVP. Progesterone, a smooth-muscle relaxant, inhibits gastric emptying in early pregnancy when levels are highly increased, which may also be an important factor in NVP. In late pregnancy, the large gravid uterus contributes to the manifestations of upper gastrointestinal symptoms by mechanically compressing the stomach. Clearly, pregnancy and its associated mechanical and hormonal changes are known to have consequences on gastrointestinal function [6,11,12]. Therefore, it would be reasonable to think that IAP during pregnancy, NVP complicating pregnancy, and delivery itself could have a deleterious effect on outcomes after ARS. This was not the case according to the data obtained
38
R. Gonzalez et al. / The American Journal of Surgery 188 (2004) 34 –38
from our study. Our results showed that patients experienced a significant improvement in GERD symptoms after an average of 17 months post-ARS (follow-up at the clinic). This allowed us to assess symptom scores after ARS, but before pregnancy, and compare them with those obtained during the last follow-up visit (after pregnancy). These data are important because they allow us to have 2 different parameters for comparison. To be certain that it was pregnancy that altered ARS outcomes and GERD symptoms, we needed to evaluate if symptoms had improved after ARS and then worsened after pregnancy. We also needed a control group against which to compare long-term changes in symptom scores from patients who did not get pregnant. Heartburn was the only symptom in which the number of patients with moderate to severe scores decreased after ARS and then slightly increased with longer follow-up; this occurred only in the NP group and not the group considered at risk, i.e., the SP group. The age difference between groups is expected because the younger the patient at the time of surgery, the higher the probability for her to become pregnant. Although age in the NP group was statistically significant, a 5-year age difference is not considered to be clinically significant enough to alter outcomes by itself. In our experience, other than the finding that pregnancy did not predict a poor outcome after ARS, the most notable finding is the high rate of ARS failure in women of childbearing age. Despite the groups being matched with regard to severity of GERD and its symptoms, 23% of the NP group were taking antisecretory medications compared with only 13% of the SP group, and 11% of the NP group eventually required ARS to be redone compared with none of the patients in the SP group. This difference did not reach statistical significance because of the small numbers. Perhaps more notable is the fact that in our overall series up to date of ⬎1,600 patients undergoing ARS, only 2.8% overall required ARS to be redone. It appears that young women may be at a higher risk of needing ARS to be redone than others regardless of whether or not they subsequently become pregnant. In conclusion, pregnancy after ARS does not predict a higher-than-expected rate of ARS failure. Therefore, ARS should not be withheld from women of childbearing age for this reason. In contrast, women of childbearing age, regardless of whether they eventually get pregnant, do appear to
have a higher-than-expected rate of repeat ARS. The reason for this is unclear, but it is reasonable to suggest that this be considered when offering this patient group ARS as well as be discussed with the patient. References [1] Hinder RA. Surgical therapy for GERD. Selection of procedures, short- and long-term results. J Clin Gastroenterol 2000;30(suppl): S48 –S50. [2] Lafullarde T, Watson DI, Jamieson GG, et al. Laparoscopic Nissen fundoplication. Five-year results and beyond. Arch Surg 2001;136: 180 – 4. [3] Trus TL, Laycock WS, Branum G, et al. Intermediate follow-up of laparoscopic antireflux surgery. Am J Surg 1996;171:32–5. [4] Hunter JG. Approach and management of patients with recurrent gastroesophageal reflux disease. J Gastrointest Surg 2001;5:451–7. [5] Hunter JG, Smith CD, Branum GD, et al. Laparoscopic fundoplication failures. Patterns of failure and response to fundoplication revision. Ann Surg 1999;230:595– 606. [6] Broussard CN, Richter JE. Nausea and vomiting of pregnancy. Gastroenterol Clin North Am 1998;27:123–51. [7] Soper NJ, Dunnegan D. Anatomic fundoplication failure after laparoscopic antireflux surgery. Ann Surg 1999;229:669 –77. [8] Awad ZT, Anderson PI, Sato K, et al. Laparoscopic reoperative antireflux surgery. Surg Endosc 2001;15:1401–7. [9] Scuderi PE, James RL, Harris L, Mims GR. Antiemetic prophylaxis does not improve outcomes after outpatient surgery when compared to symptomatic treatment. Anesthesiology 1999;90:360 –71. [10] Bradshaw WA, Gregory BC, Finley C, et al. Frequency of postoperative nausea and vomiting in patients undergoing laparoscopic foregut surgery. Surg Endosc 2002;16:777– 80. [11] Chin RK. Nausea and vomiting of early pregnancy and pregnancy outcome. An epidemiological study. Br J Obstet Gyneacol 1990;97: 278. [12] Klebanoff MA, Koslowe PA, Kaslow R, et al. Epidemiology of vomiting in early pregnancy. Obstet Gynecol 1985;66:612. [13] Hinder RA, Filipi CJ, Wetscher G, et al. Laparoscopic Nissen fundoplication is an effective treatment for gastroesophageal reflux disease. Ann Surg 1994;220:472– 83. [14] DePaula AL, Hashiba K, Bafutto M, Machado CA. Laparoscopic reoperations after failed and complicated antireflux operations. Surg Endosc 1995;9:681– 86. [15] Cushieri A, Hunter JG, Wolfe B, et al. Multicenter prospective evaluation of laparoscopic antireflux surgery. Surg Endosc 1995;7: 505–10. [16] Hunter JG, Trus TL, Branum GD, et al. A physiologic approach to laparoscopic fundoplication for gastroesophageal reflux disease. Ann Surg 1996;223:673– 87. [17] Liu JY, Woloshin S, Laycock WS, Schwartz LM. Late outcomes after laparoscopic surgery for gastroesophageal reflux. Arch Surg 2002; 137:397– 401.