Journal of Gastrointestinal Surgery

, Volume 11, Issue 5, pp 642–647

Recurrent Heartburn after Laparoscopic Fundoplication is Not Always Recurrent Reflux


    • Department of SurgeryUniversity of Adelaide
  • Glyn G. Jamieson
    • Department of SurgeryUniversity of Adelaide
  • Jennifer C. Myers
    • Department of SurgeryUniversity of Adelaide
  • Kin-Fah Chin
    • Department of SurgeryUniversity of Adelaide
  • David I. Watson
    • Department of SurgeryFlinders University
  • Peter G. Devitt
    • Department of SurgeryUniversity of Adelaide

DOI: 10.1007/s11605-007-0163-6

Cite this article as:
Thompson, S.K., Jamieson, G.G., Myers, J.C. et al. J Gastrointest Surg (2007) 11: 642. doi:10.1007/s11605-007-0163-6



A small cohort of patients present after antireflux surgery complaining of recurrent heartburn. Many of these patients have been empirically recommenced on proton pump inhibitors.


The aim of this study was to determine whether patients with symptoms that suggest recurrent reflux had objective evidence of reflux, and to determine predictors of recurrent reflux.


We identified all patients from an existing database who had undergone pH monitoring for “recurrent heartburn” after fundoplication. These patients were then cross-referenced to another database, which recorded the outcomes for patients who had undergone a laparoscopic fundoplication. Patients complaining of dysphagia or other problems without heartburn were excluded from analysis.


Seventy-six patients were identified who met the inclusion criteria. Fifty-six (74%) of these had a normal 24-h pH study. Thirty-five patients (63%) with a normal pH study were on medication for heartburn at the time of referral. Three factors were found to be associated with an abnormal 24-h pH study: a partial fundoplication (P = 0.039), onset of symptoms 6 months or more after surgery (P < 0.001), and a good symptom response when antireflux medication was recommenced (P = 0.015).


Not all patients complaining of recurrent heartburn after fundoplication have evidence of abnormal reflux. Objective evidence of abnormal esophageal acid exposure should be confirmed before recommencing antireflux medication.


Laparoscopic fundoplicationRecurrent heartburnRecurrent reflux24-h pH study


Since the first laparoscopic fundoplication was performed in 1991, the popularity of fundoplication for the treatment of gastroesophageal reflux has escalated.1,2 In 1999, fundoplications accounted for 87 of every 100,000 hospital discharges in the United States. This represented an eighfold increase in the rate of surgery over the previous decade.3 Reports looking at 5-year follow-up results after laparoscopic fundoplication suggest that 86 to 96% of patients are satisfied with the outcome of surgery. However, there are also data that imply “surgical failure” rates of up to 30%.3,4

What determines a “surgical failure” in patients who have undergone a fundoplication? Symptom control is often used as a marker of surgical outcome. In 2001, Spechler et al.5 reported “...62% of surgical patients were taking antireflux medication on a regular basis” with the inference of apparent failure of surgical treatment. However, three recent studies have shown that recurrence of symptoms after fundoplication might be a poor indicator of reflux status. These investigators found that only 23 to 39% of patients who had symptoms suggestive of recurrent reflux had abnormal esophageal acid exposure confirmed when they underwent 24-h pH monitoring.68 Hence, it is possible that many of the patients who use antisecretory medications after fundoplication might not actually have recurrent reflux. These patients might be taking medication unnecessarily.

We undertook this study to determine: 1) whether patients who complained of “recurrent heartburn” after fundoplication had an abnormal 24-h pH study and, 2) whether specific symptoms and/or other patient factors can be identified to predict which patients will have abnormal reflux when investigated by 24-h pH monitoring.

Materials and Methods

Patient Selection and Clinical Follow-up

All patients who underwent pH monitoring in the Department of Surgery at the Royal Adelaide Hospital after a laparoscopic fundoplication for “recurrent heartburn” were identified by comparing a database that is used to store pH study reports, with a clinical database that records the outcome for all laparoscopic fundoplications performed by surgeons associated with the Departments of Surgery at the University of Adelaide, and Flinders University, in Adelaide, South Australia. Patients were included in this study if they had undergone a laparoscopic fundoplication (Nissen or partial) for reflux disease that had been diagnosed before surgery by either an abnormal 24-h pH study (esophageal pH<4 for more than 4% of the study) and/or endoscopy with evidence of esophagitis (minimum Savary–Miller grade I9). Patients were excluded if they had undergone an initial open fundoplication, or if they had undergone postoperative pH monitoring to investigate nonreflux (symptoms without heartburn). Patients with other symptoms as well as heartburn (e. g., acid regurgitation, cough, nasal symptoms, or dysphagia) were not excluded from analysis. In addition to the information that was obtained from the databases, some medical records were reviewed when needed, to review clinic correspondence, endoscopy reports, and operation reports.

The patients identified were divided into two subgroups according to the outcome of the pH study—normal 24-h pH study (pH <4 for <4% of the study duration), abnormal 24-h pH study (pH <4 for> 4%). Clinical follow-up data for these patients were collected prospectively by a research nurse. This was achieved by using a combination of postal questionnaires and telephone interviews at 3 months, 12 months, and yearly after fundoplication.

A range of clinical variables was used to compare the two groups of patients: age, gender, body mass index (greater vs less than 30 kg/m2), preoperative pH study (pH <4 for more vs less than 4% of the study), type of fundoplication (partial vs total), postoperative heartburn score, esophageal motility (decreased motility ≤50% primary peristalsis, increased motility = contraction amplitudes >100 mmHg throughout esophagus and >180 mmHg in distal esophagus), onset of symptoms more than vs less than 6 months after fundoplication, and response to antireflux medication postoperatively. Heartburn was assessed using a 0 to 10 analog scale (0 = no heartburn, 10 = severe heartburn). In addition, a series of yes/no questions were asked to determine the patient’s ability to relieve symptoms of bloating by belching, their ability to belch normally, and whether or not they experienced dysphagia. Patients were also asked about the postoperative use of antireflux medication, including the type, dose, and frequency of use. Patients were asked to rate their improvement after treatment with antireflux medication: no response, mild–moderate response (incomplete eradication of heartburn symptoms), good response (no heartburn symptoms on medication).10,11

Esophageal Manometry

Esophageal manometry was performed using an eight-channel catheter with a 6-cm sleeve (Dentsleeve Pty Ltd, Adelaide, Australia). Data were recorded using an eight-channel Grass polygraph (Model 7D, Grass Instrument Co., Quincy, MA, USA). Esophageal contractility and lower esophageal sphincter (LOS) relaxation were determined by analysis of 10 wet swallows (5 mL each, 30 s apart).12 The basal end-expiratory LOS pressure (mmHg) was measured at 10-s intervals during the rest period (excluding swallow activity).

24-hour pH Study

Acid-suppressing medications were discontinued for 2 days (H2 blockers and prokinetics) to 5 days (proton pump inhibitors) before the study. A single sensor Zinetics antimony pH probe was positioned 5 cm proximal to the LES and pH data were collected for a period of 24 h using an ambulatory pH Digitrapper Mk III (Medtronic Functional Diagnostics, Denmark). Data were analyzed using EsopHogram ver2.01 (Polygram for Windows ver 2.04, Synectics Medical © 1996). A reflux event occurred if the pH dropped below pH 4.0 for longer than 5 s.13 In this study, where all patients had undergone a fundoplication, a cut-off value of 4% was used to define “abnormal reflux”.


Preoperative upper gastrointestinal endoscopy data was available for 60 of 76 (79%) patients. The degree of esophagitis was graded according to the Savary–Miller classification (Grade 0 = no mucosal ulceration, Grade I = single linear ulcer in distal esophageal mucosa, Grade II = multiple noncircumferential ulcerations, Grade III = circumferential ulceration, Grade IV = chronic complicated lesions [deep ulcers, strictures, Barrett’s esophagus]).9 Patients were identified as having Barrett’s esophagus by the presence of visible columnar mucosa in the tubular esophagus.

Statistical Analyses

SPSS for Windows version 11.0 (SPSS Inc., Chicago, IL) was used to perform data analysis. Data were expressed as mean ± range or number (percentage) as appropriate. Pearson’s chi-square tests and Mann–Whitney U tests were used where applicable to compare variables between the two groups of patients. A stepwise forward binary logistic regression analysis was also performed to confirm significant predictors for the presence or absence of abnormal esophageal acid exposure. Differences were considered significant at P < 0.05.

This study was approved by the Clinical Research Ethics Committee of the Royal Adelaide Hospital.


Out of 3,763 pH studies in the Royal Adelaide Hospital pH/manometry database, and 1,717 individual patients who had undergone a laparoscopic fundoplication in our institution, 76 patients were identified who met the inclusion criteria for this study. Some of the characteristics of these patients are summarized in Table 1. Female patients constituted 33 (43%) of the 76 patients, and the mean age at time of study was 57 years (range 28–80). Preoperative body mass index (BMI) values could be calculated for 72 of 76 patients. Twenty-seven (38%) were obese (BMI ≥30 kg/m2). Fifty-three patients (63%) had an abnormal preoperative 24-h pH study, and 55 patients (65%) had a minimum of Savary–Miller grade I9 esophagitis on endoscopy. Twenty-four patients (32%) had undergone a partial fundoplication (anterior 90° [15 patients], 180° [8], or 270° [1]), and the other 52 patients (68%) had undergone a total fundoplication (360°). The time interval between fundoplication and subsequent postoperative 24-h pH testing ranged from 2 months to 13 years, with a mean time interval of 3.7 years.
Table 1

Demographic Data on 76 Patients with Recurrent Symptoms After Fundoplication


Normal pH Study (n = 56)

Abnormal pH Study (n = 20)

P Value

Age, mean ± SD, years

57 ± 12.1

57 ± 13.9


Gender, M/F




BMI ≥30a, no. (%)

24 (43)

3 (15)


pH study >4%b, no. (%)

36 (64)

14 (70)


aBMI = body mass index (kg/m2) calculated on 72/76 patients

bPreoperative pH monitoring

Recurrent symptoms experienced by the study group are listed in Table 2. Heartburn, dysphagia, and acid regurgitation were the most common complaints, followed by cough, nasal symptoms, and chest or abdominal pain. As stated in the “Materials and Methods” section, any patient suffering solely from such symptoms without heartburn was excluded from the study.
Table 2

Symptom Profile of Patients with Recurrent Symptoms After Laparoscopic Fundoplication


Patients (n = 76) no. (%)


76 (100)


52 (68)

Acid regurgitationa

44 (58)


44 (58)

Chest/abdominal pain

38 (50)

Sore throata

2 (3)

aNot sole symptom for any patient included in this study

Postoperative 24-h pH studies were normal in 56 patients (74%) and abnormal in the remaining 20 patients (26%). These formed the two study groups. There was no difference between the study groups with regard to age, gender, or preoperative 24-h pH monitoring outcomes. Patients with a preoperative body mass index ≥30 kg/m2 were significantly more likely to have a normal postoperative 24-h pH study (P = 0.03).

We found that 46 of 56 patients (82%) who had a normal pH study, had a pH <4 for 1% or less of the study duration, and 7 of 56 (13%) had a pH <4 for between 1.1 and 2% of the study duration. The remaining three patients had pH <4 between 2.1 and 3% of the study, with no patients having a pH <4 for 3.1 to 4% of the study. Patients with an abnormal 24-h pH study were more likely to have a strong correlation between symptoms and reflux events identified at pH monitoring (P < 0.0001).14 The converse was true for patients with a normal postoperative pH study (P < 0.0001; Table 3).
Table 3

Symptom-Reflux Event Correlation Between a Positive 24-h pH Study, Symptom Index (SI), and Symptom Sensitivity Index (SSI)


Normal pH Study (n = 56)

Abnormal pH Study (n = 20)

P Value

SI ≥50%a, no. (%)

4 (7)

9 (45)


SSI ≥ 10%b, no. (%)

11 (20)

13 (65)


aSI =% of reflux associated symptom episodes

bSSI =% of symptom associated reflux episodes

Thirty-five of the 56 (63%) patients with a normal postoperative pH study, and 17 of 20 patients (85%) with an abnormal study, were taking antireflux medication at the time of their clinical review. Table 4 summarizes the use of antireflux medication in each group. The majority of patients were taking proton pump inhibitors.
Table 4

Use of Antireflux Medication After Fundoplication


Normal pH Study (n = 56)

Abnormal pH Study (n = 20)

Antireflux medication, no. (%)

35 (63)

17 (85)

H2 blockers, no. (%)

6 (17)

3 (18)

Proton pump inhibitors, no. (%)

29 (83)

14 (82)

Eleven of 20 patients (55%) with an abnormal pH study went on to have revisional surgery. The following diagnoses were made at surgery: slipped wrap (3), disrupted wrap (3), herniated wrap (2), etiology for failure unclear (3). Five of 56 patients (9%) with normal 24-h pH monitoring eventually went on to have revisional antireflux surgery. These were all patients who had troublesome dysphagia and heartburn.

Postoperative variables that were not associated with the 24-h pH study outcome are listed in Table 5. No difference was seen between the two groups for postoperative heartburn score, abnormal esophageal motility, and ability to relieve symptoms of bloating. Three variables were found to be significantly associated with an abnormal postoperative 24-h pH study (Table 6). These were: a partial fundoplication (P = 0.039), onset of symptoms more than 6 months after surgery (P < 0.001), and a good response to antireflux medication (P = 0.015). These postoperative variables were entered into a binary logistic regression model (with the exception of a good response to antireflux medication as a result of incomplete data on all patients) and the same factors (onset of symptoms more than 6 months postoperatively, partial wrap) remained highly significant (P < 0.02).
Table 5

Postoperative Variables Not Significantly Associated with a Positive 24-hr pH Study


Normal pH Study (n = 56)

Abnormal pH Study (n = 20)

P Value

Heartburn score, mean ± SD

5.5 ± 3.5

6.3 ± 3.9


Esophageal motility, no. (%)


44 (79)

16 (80)



2 (4)

0 (0)



6 (11)

4 (20)


 Bloat relief, no. (%)

29 (52)

8 (40)


a>50% primary peristalsis on manometry

bHyperdynamic esophagus on manometry ≥100 mmHg proximal esophageal contraction amplitudes, and distal esophageal contraction amplitudes >180 mmHg

c≤50% primary peristalsis on manometry

Table 6

Intraoperative and Postoperative Variables Significantly Associated with a Positive Postoperative 24-h pH Study


Normal pH Study (n = 56)

Abnormal pH Study (n = 20)

P Value

Partial wrapa, no. (%)

14 (25)

10 (50)


Onset of symptoms > 6 mo postoperativelyb, no. (%)

23 (41)

17 (85)


Good response to anti- reflux medicationc, no. (%)

15 (27)

12 (60)


aPartial wrap includes 90° (15), 180° (8), 270° (1)

bData available on 75/76 patients

cData available on 41/76 patients

Eighteen of the 56 patients who had a normal pH study had an endoscopy performed near the time of the pH study. Four of these patients had an equivalent or improved grade of esophagitis compared to the preoperative endoscopy findings. The other 14 patients had no evidence of esophagitis nor disruption/herniation of their fundoplication.


Only 20 of 76 patients in this study who were assessed for recurrence of “heartburn” after fundoplication had an abnormal 24-h pH study. The remaining 56 patients had no objective evidence of abnormal esophageal acid exposure. In this latter group, 95% had acid in the esophagus (pH <4) for 2% of the study duration or less, suggesting that the negative results were unequivocal findings.

Our results are similar to those reported from three other centers where only 23 to 39% of patients investigated with pH monitoring for recurrent reflux symptoms had an abnormal pH study.68 It seems that many patients are taking antireflux medication unnecessarily because they are prescribed these medications after a fundoplication without objective evidence of reflux.15,16 In our study, 63% of patients with “reflux” symptoms, but a normal pH study, were taking antireflux medications at the time of their assessment. This appears to be inappropriate treatment.

A reasonable question to ask is: are there any factors that can alert the physician that a patient’s symptoms are more likely to be caused by true recurrent reflux? A number of studies have shown that postoperative heartburn, dysphagia, and chest pain can all occur in the absence of abnormal esophageal acid exposure.4,68,17,18 Galvani et al.6 found the symptom of acid regurgitation a reliable indicator of pathologic reflux. However, in our experience, less than 60% of patients complained of regurgitation postoperatively and there was no significant association of a positive 24-h pH study with this symptom.

We assessed nonsymptom-related variables (except postoperative heartburn score, and the ability to relieve bloat symptoms) to determine whether anything can be identified that could guide the physician toward the correct diagnosis. We did not find any significant association between pH outcomes, and age, gender, or esophageal motility. Obesity was significantly associated with a normal pH study, rather than an abnormal study. The lack of association between obesity and pathologic reflux is strongly supported by recent publications.1922 Anvari et al.21 documented no difference in outcome between 70 morbidly obese patients and 70 non-obese patients as measured by postoperative 24-h pH results. Obese patients had a low 1.4% recurrence rate, requiring reoperation during a follow-up period of almost 4 years.21 We have no ready explanation for our finding other than a possible type I statistical error.

Three variables were found to be significantly associated with pathologic reflux: a partial fundoplication, onset of recurrent symptoms 6 months or more after surgery, and a good response to antireflux medication. In particular, 9 of 20 patients with an abnormal pH study had undergone an anterior partial fundoplication. In two randomized trials, we have previously shown that an anterior 90° partial fundoplication is followed by a high rate of patient satisfaction and few adverse effects (dysphagia and wind-related complaints) compared to total fundoplication. However, there is a trade-off with less effective long-term control of reflux.23,24

In contrast to other publications on this topic68, most patients in our study had a long time interval between fundoplication and recurrent symptoms and postoperative pH testing (mean of 3.7 years). Therefore, we were able to differentiate between patients whose recurrent symptoms began in the early period after surgery, and those whose symptoms developed years later. Anecdotally, it has been our experience that patients who develop recurrent heartburn, and who do not have abnormal esophageal acid exposure, tend to report symptoms at the first or second postoperative visit. Our results support this observation. In addition, patients who have a normal pH study often state that antireflux medication only controlled their symptoms “somewhat” or “not at all.” Again, this is supported by the current study. In contrast to this, patients who report a good response to antireflux medication probably have recurrent reflux, and it is likely that this will be confirmed by 24-h pH monitoring.

The more complex issue is how to manage the patient who has a normal pH study. The sparse literature on this topic suggests that if reflux and other pathology is excluded, then a functional diagnosis is likely.3,25 We are now studying these patients in greater detail to rule out other causes of “heartburn”: biliary disease, peptic ulcer disease, gastritis, irritable bowel syndrome, and functional dyspepsia.

Based on the Rome Consensus, functional dyspepsia is defined as persistent or recurrent pain in the upper abdomen in the absence of other pathology for at least 12 weeks in the previous year.26 However, many of the patients who had a normal pH study seem to have symptoms arising from the esophagus itself. This raises the question of whether a condition of “irritable esophagus” exists in a similar fashion to “irritable bowel.” In other words, one could hypothesize that this small subgroup of patients might suffer from an overly sensitive esophagus, and this could be caused by altered sensory receptors on esophageal mucosa or abnormal processing of neurotransmitters.2729 This is a question that requires further study. As none of these patients had moderate to severe dysphagia, it seems unlikely that obstruction caused by a “tight” fundoplication causing intermittent esophageal distension, or spasm, was a cause of the heartburn.

The present study has limitations. First, it does not address recurrence of heartburn in fundoplication patients who have not undergone pH monitoring at our institution, but whose symptoms have been successfully treated by reinstitution of antireflux medication. Therefore, this report could be biased toward patients in whom antireflux medication has not been successful. Second, other studies have established that the reproducibility of 24-h pH monitoring is only of the order of 70 to 80%.14 We defined our cut-off value for abnormal reflux at 4% as we felt that any value above this number is likely to be highly relevant in patients with recurrent symptoms who have previously undergone a fundoplication. Also, we found excellent symptom-reflux event correlation within our 24-h pH recordings, which supports accurate reporting of our results. Nevertheless, had we repeated the test in our negative group, it is possible that a small proportion may have had a positive test.

Third, it is possible that symptoms experienced by patients with a normal pH study could have been caused by nonacid reflux. However, studies have shown that nonacid reflux is very uncommon.30,31 It is therefore unlikely that nonacid reflux was a cause of our patients’ recurrent symptoms.


Fifty-six patients (74%) complaining of recurrent heartburn after laparoscopic fundoplication, who were referred for 24-h pH monitoring, had no evidence of abnormal esophageal acid exposure. Although “62% of surgical patients were taking antireflux medication on a regular basis” in the widely cited JAMA article of Spechler et al.5, it is likely that many of their patients were started on medical therapy without objective testing, and some may not have had abnormal reflux. Furthermore, it appears there is a small group of patients who are proven to have abnormal reflux before surgery, who have an intact fundoplication, and yet continue to have reflux symptoms for reasons that are not clear. Further investigation of this perplexing group of patients is needed.


We would like to thank Carolyn Lally for all her help in managing the Royal Adelaide Hospital fundoplication database.

Copyright information

© The Society for Surgery of the Alimentary Tract 2007