Survey of findings in patients having persistent heartburn on proton pump inhibitor therapy

Transcription

1 Diseases of the Esophagus (2014) (2016), 29, DOI: /dote Original article Survey of findings in patients having persistent heartburn on proton pump inhibitor therapy R. Mandaliya, 1 A. J. DiMarino, 2 S. Cohen 2 1 Division of Internal Medicine, Abington Memorial Hospital, Abington, and 2 Division of Gastroenterology and Hepatology, Thomas Jefferson University Hospital, Philadelphia, Pennsylvania, USA SUMMARY. In patients with refractory heartburn while on proton pump inhibitor (PPI) therapy, changing drugs or increasing treatment to a twice a day (b.i.d.) dose has become a common practice. This study aims to study patients with persistent heartburn while on PPI therapy and to determine if persistent symptom indicates the need for more aggressive or different therapy. A retrospective review of impedance-ph tracings on PPI therapy (q.d. or b.i.d.) for patients with persistent heartburn was performed. DeMeester score, impedance, and symptom sensitive index (SSI) were used as indices. Statistical analyses were performed using chi-squared test with Yates correction and paired t-test. One hundred consecutive patients, (female 50%, male 50%, mean age 54 [range 16 83] years) were studied on q.d. (n = 45) or b.i.d. PPI (n = 55). Only 20% of the patients had abnormal DeMeester score; 41% had an abnormal impedance score and 56% had abnormal SSI; 29% had all indices normal. There was no difference between patients taking q.d. versus b.i.d. PPI for abnormal DeMeester score (22 vs. 18%), impedance (38 vs. 44%) and SSI (53 vs. 58%); P = 0.80, 0.69, and 0.77, respectively. In 56 patients with positive SSI, symptoms were due to acid reflux in 8 (14%) patients, nonacid reflux in 31 (55%) patients, and combined acid and nonacid reflux in 17 (30%) patients. Patients with persistent heartburn on PPI therapy show a variety of disorders: (i) acid reflux (20%); (ii) nonacid reflux (26%); (iii) positive SSI (56%); (iv) all normal indices (29%). These studies indicate that persistent heartburn on PPI therapy is a complex problem that may not respond to simply increasing acid inhibition. KEY WORDS: DeMeester score, gastroesophageal reflux disease, heartburn, impedance, proton pump inhibitor, symptom sensitivity index. INTRODUCTION Address correspondence to: Professor Sidney Cohen, MD, Division of Gastroenterology and Hepatology, Thomas Jefferson University Hospital, Suite 480 Main Building, 111 S 11th Street, Philadelphia, PA 19107, USA. rohan86m@gmail.com Guarantor of the article: Sidney Cohen. Specific author contributions: Rohan Mandaliya: study concept and design; acquisition, analysis, and interpretation of data; statistical analysis; graphical representation of the results; drafting of the manuscript. Anthony J. DiMarino: administrative, technical of material support; study supervision; critical revision of the manuscript for important intellectual content. Sidney Cohen: study concept and design; analysis and interpretation of the data; editing and critical revision of the manuscript for important intellectual content; statistical analysis; administrative, technical, or material support; study supervision. Conflicts of interest: All authors have no conflicts of interest. Funding: Division of Gastroenterology and Hepatology, Thomas Jefferson University Hospital. Gastroesophageal reflux disease (GERD) remains the most common gastrointestinal diagnosis for outpatient visits based on a recent national survey. 1 Although GERD is a disorder attributed to transient lower esophageal sphincter relaxation (TLESR), symptomatic relief is achieved in most patients with proton pump inhibitor (PPI) therapy aimed at reducing gastric acidity, thus reducing esophageal acid exposure. 2 A subset of patients (10 40%) has persistent symptoms on a standard dose PPI therapy. 3 5 In such patients, simply changing drugs or increasing the dose of PPI has become a reflex strategy. There has been no randomized control trial or large studies supporting this approach. In a survey by Gallup Organization during a period of 7 years ( ), there was a 50% increase in the usage of at least double-dose PPI in patients with GERD. 6 A common presentation of GERD to gastroenterology practice is persistent symptoms on PPI therapy. 3 Patients with persistent heartburn on empiric PPI therapy may next undergo ph monitoring, which 27 1

2 28 2 Diseasesof of the Esophagus does not detect nonacid gastroesophageal reflux episodes. Multichannel intraluminal impedance monitoring detects all reflux episodes, liquid, gas, or mixed. When used in combination with conventional ph monitoring, it characterizes reflux episodes as acid or nonacid. It is a sensitive tool to detect and characterize gastroesophageal reflux. 7 The purpose of this retrospective analysis was to study the profile of consecutive patients with persistent heartburn on PPI therapy in terms of DeMeester score, impedance score, and symptom sensitivity index (SSI) to characterize the etiology of persistent symptom while on PPI therapy. MATERIALS AND METHODS The present study consists of a retrospective review of impedance-ph monitoring data in consecutive patients with persistent heartburn suggestive of or potentially related to GERD on PPI therapy at Thomas Jefferson University Hospital between January 2008 and September The institutional review board of Thomas Jefferson University approved the retrospective review of data. Exclusion criteria were patients under the age of 18 years, noncompliant to medication (not taking PPI on a daily basis), prior antireflux surgery, achalasia, gastroparesis, and scleroderma. Patients having atypical GERD symptoms like cough, chest pain, sore throat, hoarseness, and asthma were not included. Persistent heartburn was defined as patients seeking medical help for unsatisfactory control of heartburn symptom while on PPI therapy, either once a day (q.d.) or twice a day (b.i.d) for at least 4 weeks. Different PPIs were used depending on the preference of the referring physician. Consecutive patients were selected having persistent heartburn while on PPI therapy for at least 4 weeks at the time of impedance-ph testing. Patients were also subgrouped on the basis of PPI dose: q.d. or b.i.d. On the day of the combined impedance-ph study, the patients came to the motility laboratory with a fasting period of at least 4 hours. Combined impedance-ph was performed with a 2.1-mm diameter catheter (Sandhill Scientific, Inc., Highlands Ranch, CO, USA) with six impedance and two ph measuring electrodes. The six impedance-measuring sites were located at 3, 5, 7, and 9 cm (distal esophagus) and at 15 and 17 cm (proximal esophagus) above the lower esophageal sphincter (LES). Two ph sensors were located at 5 cm above and 10 cm below the LES for measurement of ph in distal esophagus and stomach, respectively. As per the specifications by the manufacturer (Sandhill Scientific Inc.), before the study, the esophageal and gastric ph electrodes were calibrated in buffer solutions of ph 4 and 7. After LES location by esophageal manometry, the impedance-ph catheter was passed transnasally into the esophagus and stomach and positioned in relation to manometric measurements. The catheter was connected to a portable data recorder provided by the company, which was worn around the waist. Patients were discharged and were advised to maintain normal activities, sleep schedule, and eat their usual meals at usual times. They were asked to remain upright during the day, and lie down only during their sleep. They were given a data logger device where they pressed buttons at various events like meal times, posture changes, and occurrence of symptom (e.g. heartburn). On the next day, after removal of the impedance ph catheter, data from the portable recorder was uploaded on a personal computer and analyzed with special software (Bioview Analysis, Sandhill Scientific Inc.). An acid reflux was defined as a decrease in the distal esophageal ph to less than 4 for at least 5 seconds occurring within 10 seconds after impedance detected a reflux event. distal esophageal acid exposure time when ph <4 in upright and recumbent position was defined as >1.6 and >0.0%, respectively. 8 DeMeester score was considered as an index of acid exposure. It is a composite score that has six components ([i] percent total time ph < 4; [ii] percent upright time ph < 4; [iii] percent supine time ph < 4; [iv] number of reflux episodes; [v] number of reflux episodes 5 min; [vi] longest reflux episode [minutes]). A value of DeMeester score > 14.7 was considered abnormal and indicative of abnormal esophageal acid exposure. 9 Impedance was defined as the total number of reflux episodes (acid and nonacid) for the duration of the entire study. value of impedance on PPI therapy was defined as total number of reflux episodes of more than 48 during the entire study. 10 SSI was used as an index to measure symptom associated with reflux. There are two other indices also used in research and clinical practice: symptom association probability (SAP) and symptom index (SI). SSI was defined as the number of symptompositive reflux of the total reflux episodes. SAP is defined as 1-P where P is the probability value that reflux and symptom episodes were unrelated. SI is defined as the number of reflux-related symptoms of the total symptom episodes. In a study by Taghavi et al. in patients with GERD, SSI had the highest positive and negative predictive values and sensitivity compared with SAP and SI. The specificity of SSI and SAP was equal and lower than SI. A value of SSI 1 was defined as positive based on a recent study. 11 Taghavi et al. demonstrated using the receiver operating characteristic curve that SSI cutoff of 1.3 should be used for considering abnormal values. 11

3 Persistent heartburn 29 3 The above three indices (DeMeester score, impedance, and SSI) were compared between the two groups of patients: those taking q.d. PPI and those taking b.i.d. PPI. Statistical analyses were performed using SPSS version 20 software (IBM Corp., Armonk, NY, USA). The data were normally distributed, except the number of acid, the nonacid reflux episodes, and the total reflux events. The chi-squared test with Yates correction with the help of 2 2 contingency table was used to compare the relative frequencies of the two study groups. Wilcoxon signed rank test was used to calculate difference in frequencies of acid and nonacid reflux episodes in a single patient. A P-value of 0.05 or smaller was used to indicate statistical significance. RESULTS The study included 100 patients (50 [50%] females and 50 [50%] males; mean age 54 [range years]) who underwent ambulatory-combined impedance-ph monitoring while on PPI therapy from January 2008 until September There were 45 (45%) patients studied on q.d. PPI and 55 (55%) patients on b.i.d. PPI. Patients had tried multiple PPI medications, but at the time of the study, they were on: esmoprazole (n = 31%), ompeprazole, (n = 22%), lansoprazole (n = 19%), pantoprazole (n = 17%), and rabeprazole (n = 11%). Only 20 (20%) patients had an abnormal DeMeester score with persistent heartburn on PPI therapy. Of the 45 patients taking q.d. PPI, 10 (22%) patients had an abnormal DeMeester score, whereas 10 (18%) patients of 55 patients had an abnormal DeMeester score on b.i.d. PPI; χ 2 with Yates correction = 0.063, P = (Fig. 1). There were 41 (41%) patients who had an abnormal impedance score. There was no significant difference in the abnormal impedance score between patients taking q.d. PPI (17/45 [38%]) and those taking b.i.d. PPI (24/55 [44%]); χ 2 with Yates correction = 0.151, P = (Fig. 1). Figure 2 shows the median number of acid and nonacid reflux events in 100 patients. In the 100 patients, 5440 impedance events were recorded; 1385 (25%) were acid reflux and 4055 (75%) were nonacid reflux. Total number of reflux episodes of 48 or more was considered abnormal. The median number of total reflux episodes in a patient was 40 (25th 75th percentile 27 63). The median number of acid reflux episodes was 5 (25th 75th percentile 2 14), and the median number of nonacid reflux episodes was 26 (25th 75th percentile 21 50). While on therapy, more nonacid reflux episodes were reported compared with acid reflux episodes in a patient; W = 4071, P < (Wilcoxon signed rank test). Fig. 1 Comparison between once a day and twice a day proton pump inhibitor (PPI) therapy in two groups of patients with persistent heartburn symptoms in terms of abnormal acid reflux (DeMeester score), abnormal impedance score, and abnormal symptom sensitive index (SSI). There is no significant difference in the above mentioned parameters (P > 0.05 in all three comparisons) while on once a day PPI or twice a day PPI therapy when measured in patients with persistent heartburn symptoms. There were 56 (56%) patients who had abnormal symptom sensitivity index. And 44 (44%) patients did not have any symptom-positive reflux, meaning that the symptom occurrence was not associated with the reflux. We categorized the patients with positive SSI into three groups: (i) symptoms due to acid reflux only; (ii) symptoms due to nonacid reflux only; and (iii) symptoms due to both acid and nonacid reflux. Of the 56 patients with positive SSI, symptoms were due to acid reflux in 14%, nonacid reflux in 55%, and combined acid and nonacid reflux in 30% of the patients (Fig. 3). Comparing q.d. to b.i.d. dose PPI in patients with abnormal SSI, 24/45 (53%) had abnormal SSI on q.d. PPI and 32/55 (58%) on b.i.d. PPI, χ 2 with Yates correction = 0.08, P = Also, on considering positive SSI for acid reflux, there was no difference between q.d. and b.i.d. PPI (11/45 [24%] vs. Fig. 2 Median number of reflux episodes (total episodes, acid reflux episodes, and nonacid reflux episodes) in patients on proton pump inhibitor (PPI) therapy. Note the significantly higher number of nonacid reflux episodes compared with acid reflux episodes due to PPI therapy with the number of total reflux episodes in normal range (<48). Values in brackets indicate 1st and 3rd quartiles of the median value.

4 30 4 Diseasesof of the Esophagus Fig. 3 Distribution of the type of reflux episode associated with symptoms in patients with positive symptom sensitivity index. Of the 56 patients with positive SSI, 56% had symptoms due to nonacid reflux, 30% had symptoms due to both acid and nonacid reflux, whereas only 14% had symptoms due to acid reflux. Fig. 5 Distribution of patients having persistent heartburn on proton pump inhibitor therapy. SSI, symptom sensitive index. 14/55 [25%]); χ 2 with Yates correction = 0.013, P = 0.9. Of the 56 patients with abnormal SSI, 45% had normal acid and nonacid reflux parameters. On considering a cut-off for SSI 1, there were seven patients with SSI for nonacid reflux of 1, and all seven patients had high impedance scores, suggesting the role of nonacid reflux in refractory symptoms. Of the patients, 29 (29%) had all the three indices (DeMeester score, impedance, and SSI) as normal, suggesting some other causes of their symptoms. Figure 4 shows the distribution of patients with one or more abnormal indices with amount of respective overlap using a quantitative Venn diagram. Thus, as shown in Figure 5, patients with persistent esophageal symptoms of heartburn on PPI therapy showed a variety of findings: (i) persistent acid reflux (20%); (ii) nonacid reflux (26%); (iii) positive SSI (56%); many of these patients (45%) had normal acid and nonacid reflux parameters; and (iv) all indices as normal (29%). Patients with one or more indices abnormal: 71% Patients with abnormal impedance 41% impedance alone 9% SSI and impedance 17% DeMeester s score, impedance, and SSI 11% SSI alone 25% Patients with abnormal SSI 56% Patients with all indices normal: 29% DeMeester s score and impedance 4% DeMeester s score alone 2% Patients with abnormal DeMeester s score 20% DeMeester s score and SSI 3% Patients with normal DeMeester s score, impedance, and SSI 29% Fig. 4 Quantitative Venn diagram showing the distribution of the patients in terms of the three abnormal parameters with their respective overlap among each other. The center of all three circles indicates the percentage of patients with all three parameters abnormal. The lower single circle indicates patients with all three parameters being normal. Of all patients, 71% had a single or multiple parameters being abnormal and 29% had no abnormality. SSI, symptom sensitive index.

5 Persistent heartburn 31 5 DISCUSSION The current study demonstrates the difficulty in defining persistent esophageal symptoms in patients on PPI therapy. A combination of simultaneous ph, impedance, and SSI evaluations show at least four possible groups of patients. First, a small group of patients (20%) has continued acid reflux; second, a somewhat larger group (26%) has nonacid reflux; third, some patients had positive symptom score with modest or nonpathological quantities of acid or nonacid events; fourth, some patients (29%) had no measurable abnormality. These findings would explain the clinical frustration of treating all patients resistant to PPI therapy in the same manner. Only 20% of the patients with persistent heartburn had abnormal parameters of acid reflux on PPI therapy. This suggests that majority of patients have other causes of persistent heartburn. Impedance measurements detected additional 26% patients having nonacid reflux. Fifty-six percent of the patients had abnormal SSI. More than half (55%) had symptoms solely to nonacid reflux, whereas 14% of the patients had symptoms due to acid reflux. This result is similar to a multicentre study by Mainie et al., where 37% of the patients on b.i.d. PPIs had a positive SI for nonacid reflux, and 11% of the patients had positive SI for acid reflux. 12 The median number of nonacid reflux events was significantly higher than that of acid reflux in our patients on PPI therapy. Consistent with the results of studies in the past, patients on PPI therapy have statistically significant relative frequency of nonacid reflux episodes than acid reflux episodes during monitoring. 13 Tamhankar et al. reported no difference in the frequency of total reflux episodes on impedance-ph monitoring in patients on and off PPI. 14 These results may be expected because PPI does suppress acid production, but does not have any effect on TLESRs. Another approach to treating GERD would be to alter TLESRs, the major mechanism of gastroesophageal reflux. TLESR is the result of a complex neural pathway with sensory afferent fibers arising from the muscle layers of the stomach, terminating in the nucleus tractus solitaries and dorsal motor vagal nucleus in the brainstem The efferent fibers project to the enteric nervous system of the LES. 19 The various neurotransmitters involved in the above neuronal pathway include acetylcholine, cholecystokinin, gamma aminobutyric acid, glutamate, nitric oxide, and opiods. 20 Primary peristalsis is a major determinant of esophageal acid clearance, and patients with GERD frequently exhibit prolonged esophageal acid exposure due to delayed esophageal clearance. 21,22 Prokinetic drugs such as cisapride and metochlorpromide may be of benefit along with acid suppressants as they act upon the motility disturbances underlying GERD. 23,24 Holloway suggested that the beneficial effects of cisapride were due to improved esophageal acid clearance with the help of its prokinetic effect. 25 Cisapride is a 5HT4 agonist and acts by enhancing the release of acetylcholine from postganglionic nerve endings of the myentric plexus in gastrointestinal smooth muscle. Unlike metachlorpromide, cisapride has no antidopaminergic nor direct cholinomimetic effects. Tegaserod, a 5-HT4 receptor partial agonist in one randomized study, was shown to cause a significant decrease in postprandial esophageal acid exposure. 26 TLESR may represent a potential target for pharmacological treatment of GERD. Mittal et al. using atropine showed that reducing TLESR in humans resulted in a decrease in reflux. 27 Baclofen, a gamma aminobutyric acid b (GABAb) receptor agonist has shown to be the most promising class of agent to reduce the rate of TLESRs. In a prospective randomized cross-over study by Zhang et al. in 2002, GABAb agonist baclofen significantly inhibited gastroesophageal reflux episodes by the inhibition of TLESR. 28 In other small randomized studies, baclofen has been shown to decrease the number of postprandial acid and nonacid reflux events, 29 and nocturnal reflux activity. 30 Motility drugs may be considered in the future to alter reflux provided that safety is shown. Of the patients, 25% had positive SSI without abnormal acid or nonacid reflux parameters. One possible explanation is esophageal hypersensitivity. 31 The mechanisms underlying this condition are not clear; however, one of the few explanations is the dilatation of intercellular spaces within the esophageal epithelium and exposure of subepithelial nerves to acid. 32 Visceral pain modulators like tricyclic antedepressants (TCAs) and serotonin reuptake release inhibitors may be an approach in such patients. Viazis, in a randomized placebo-controlled study, showed that citalopram 20 mg q.d. for 6 months was effective in patients with acid hypersensitive esophagus and refractory gastroesophageal reflux symptoms. 33 TCAs have also been found to be beneficial in patients with noncardiac chest pain. In a double-blind, placebo-controlled trial in 60 patients, imipramine significantly reduced chest pain episodes in 52% of the patients. 34 The results of our study show that there is no significant difference in the patients taking q.d. or b.i.d. PPI in terms of acid control, number of reflux, and SSI. These results indicate that persistent heartburn on PPI therapy is a complex problem with other possible underlying causes that may not respond to simply increasing the acid control. Charbel et al. compared patients with GERD with persistent symptoms on q.d. and b.i.d. PPI therapy with ambulatory ph monitoring. 35 They concluded that the likelihood of an abnormal esophageal ph for symptomatic GERD patients on b.i.d. PPI

6 32 6 Diseasesof of the Esophagus therapy is very small. They also concluded that the number of patients with typical GERD symptoms having abnormal total % time ph of <4 (>5.5%) was significantly higher in patients taking q.d. PPI than those taking b.i.d. PPI, (31 vs 7%). However, there was no significant difference in acid control in patients with typical GERD symptoms on q.d. and b.i.d. PPI using cut-off value for total time ph of <4 to be 1.6%. For patients on PPI, the cut-off value for total time ph of < 4 be 1.6% rather than 5.5%. Kuo and Castell studied 19 normal subjects on PPI and reported that the 95% upper limit of normal total % time of distal esophageal acid exposure was 1.6%. 8 There are several limitations in this study. It should be noted that different patients having persistent heartburn were studied on q.d. and b.i.d. PPI. The findings are observational and were not subject to controlled conditions such as a within-subjects design, taking patients on q.d. PPI therapy, evaluating them, then putting them on b.i.d. PPI, and studying their reflux characteristics. This was a nonrandomized study with a retrospective review of patients who were referred to a tertiary referral center for evaluation of persistent reflux symptom. Due to the small sample size of the study (n = 100), it is subject to the risk of type II error. A large multicenter randomized control study could be done to increase the statistical significance of the study. The results from our study show that the patients with persistent esophageal symptoms of heartburn on PPI therapy show a variety of disorders. These studies indicate that persistent symptoms while on PPI therapy is a complex problem that may not respond to simply increasing acid inhibition. References 1 Peery A F, Dellon E S, Lund J et al. Burden of gastrointestinal disease in the United States: 2012 update. Gastroenterology 2012; 143: Dodds W J, Dent J, Hogan W J et al. Mechanisms of gastroesophageal reflux in patients with reflux esophagitis. N Engl J Med 1982; 307: Fass R, Sifrim D. Management of heartburn not responding to proton pump inhibitors. Gut 2009; 58: Klinkenberg-Knol E C, Festen H P, Jansen J B et al. Longterm treatment with omeprazole for refractory reflux esophagitis: efficacy and safety. Ann Intern Med 1994; 121: Leite L P, Johnston B T, Just R J, Castell D O. Persistent acid secretion during omeprazole therapy: a study of gastric acid profiles in patients demonstrating failure of omeprazole therapy. Am J Gastroenterol 1996; 91: Organization TG. The 2000 Gallup Study of Consumers Use of Stomach Relief Products. Princeton, NJ: Gallup Organization, Sifrim D, Castell D, Dent J, Kahrilas P J. Gastro-oesophageal reflux monitoring: review and consensus report on detection and definitions of acid, non-acid, and gas reflux. Gut 2004; 53: Kuo B, Castell D O. Optimal dosing of omeprazole 40 mg daily: effects on gastric and esophageal ph and serum gastrin in healthy controls. Am J Gastroenterol 1996; 91: Johnson L F, DeMeester T R. Twenty four hour ph monitoring of the distal esophagus. A quantitative measure of gastroesophageal reflux. Am J Gastroenterol 1974; 62: Tutuian R, Mainie I, Agrawal A, Freeman J, Castell D O. Normal values for ambulatory 24-hour combined impedance-ph monitoring on acid suppressive therapy. Gastroenterology 2006; 130 (Suppl. 2): A Taghavi S A, Ghasedi M, Saberi-Firoozi M et al. Symptom association probability and symptom sensitivity index: preferable but still suboptimal predictors of response to high dose omeprazole. Gut 2005; 54: Mainie I, Tutuian R, Shay S et al. Acid and non-acid reflux in patients with persistent symptoms despite acid suppressive therapy: a multicentre study using combined ambulatory impedance-ph monitoring. Gut 2006; 55: Clayton S B, Rife C C, Singh E R, Kalbfleisch J H, Castell D O. Twice-daily proton pump inhibitor therapy does not decrease the frequency of reflux episodes during nocturnal recumbency in patients with refractory GERD: analysis of 200 patients using multichannel intraluminal impedance-ph testing. Dis Esophagus 2012; 25: Tamhankar A P, Peters J H, Portale G et al. Omeprazole does not reduce gastroesophageal reflux: new insights using multichannel intraluminal impedance technology. J Gastrointest Surg 2004; 8: Neuhuber W L. Sensory vagal innervation of the rat esophagus and cardia: a light and electron microscopic anterograde tracing study. J Auton Nerv Syst 1987; 20: Altschuler S M, Bao X M, Bieger D, Hopkins D A, Miselis R R. Viscerotopic representation of the upper alimentary tract in the rat: sensory ganglia and nuclei of the solitary and spinal trigeminal tracts. J Comp Neurol 1989; 283: Kalia M, Sullivan J M. Brainstem projections of sensory and motor components of the vagus nerve in the rat. J Comp Neurol 1982; 211: Barber W D, Burks T F. Brain stem response to phasic gastric disention. Am J Physiol. Am J Physiol 1983; 245: G Rinaman L, Card J P, Schwaber J S, Miselis R R. Ultrastructural demonstration of a gastric monosynaptic vagal circuit in the nucleus of the solitary tract in the rat. J Neurosci 1989; 9: Hirsch D P, Tytgat G N, Boeckxstaens G E. Transient lower oesophageal sphincter relaxations a pharmacological target for gastro-oesophageal reflux disease? Aliment Pharmacol Ther 2002; 16: Allen M L, Castell J A, DiMarino A J Jr. Mechanisms of gastroesophageal acid reflux and esophageal acid clearance in heartburn patients. Am J Gastroenterol 1996; 91: Lin S, Ke M, Xu J, Kahrilas P J. Impaired esophageal emptying in reflux disease. Am J Gastroenterol 1994; 89: McCallum R W. Cisapride: a new class of prokinetic agent. The ACG Committee on FDA-related matters. The American College of Gastroenterology. Am J Gastroenterol 1991; 86: McCallum R W. Review of the current status of prokinetic agents in gastroenterology. Am J Gastroenterol 1985; 80: Holloway R H, Downton J, Mitchell B, Dent J. Effect of cisapride on postprandial gastro-oesophageal reflux. Gut 1989; 30: Kahrilas P J, Quigley E M, Castell D O, Spechler S J. The effects of tegaserod (HTF 919) on oesophageal acid exposure in gastro-oesophageal reflux disease. Aliment Pharmacol Ther 2000; 14: Mittal R K, Holloway R, Dent J. Effect of atropine on the frequency of reflux and transient lower esophageal sphincter relaxation in normal subjects. Gastroenterology 1995; 109: Zhang Q, Lehmann A, Ridga R, Dent J, Holloway R H. Control of transient lower oesophageal sphincter relaxations and reflux by the GABA(B) agonist baclofen in patients with gastro-oesophageal reflux disease. Gut 2002; 50: Vela M F, Tutuian R, Katz P O, Castell D O. Baclofen decreases acid and non-acid post-prandial gastro-oesophageal reflux measured by combined multichannel intraluminal impedance and ph. Aliment Pharmacol Ther 2003; 17:

7 Persistent heartburn Orr W C, Goodrich S, Wright S, Shepherd K, Mellow M. The effect of baclofen on nocturnal gastroesophageal reflux and measures of sleep quality: a randomized, cross-over trial. Neurogastroenterol Motil 2012; 24: Galmiche J P, Clouse R E, Balint A et al. Functional esophageal disorders. Gastroenterology 2006; 130: Knowles C H, Aziz Q. Visceral hypersensitivity in non-erosive reflux disease. Gut 2008; 57: Viazis N, Keyoglou A, Kanellopoulos A K et al. Selective serotonin reuptake inhibitors for the treatment of hypersensitive esophagus: a randomized, double-blind, placebo-controlled study. Am J Gastroenterol 2012; 107: Cannon R O 3rd, Quyyumi A A, Mincemoyer R et al. Imipramine in patients with chest pain despite normal coronary angiograms. N Engl J Med 1994; 330: Charbel S, Khandwala F, Vaezi M F. The role of esophageal ph monitoring in symptomatic patients on PPI therapy. Am J Gastroenterol 2005; 100: