Gastroesophageal reflux disease (GERD) is a common chronic

Transcription

1 CLINICAL GASTROENTEROLOGY AND HEPATOLOGY 2009;7: Efficacy of Esophageal Impedance/pH Monitoring in Patients With Refractory Gastroesophageal Reflux Disease, on and off Therapy JASON M. PRITCHETT,* MUHAMMAD ASLAM, JAMES C. SLAUGHTER, REID M. NESS, C. GAELYN GARRETT, and MICHAEL F. VAEZI *Department of Internal Medicine, Department of Biostatistics, Division of Gastroenterology, Hepatology and Nutrition, and Department of Otolaryngology, Voice Center, Vanderbilt University Medical Center, Nashville, Tennessee BACKGROUND & AIMS: Intraluminal impedance monitoring has given new dimensions to the diagnosis of reflux disease. However, there is no defined algorithm for evaluating refractory reflux symptoms. We studied whether combined impedance/ph monitoring in patients on therapy can predict acid reflux in patients off therapy and whether testing should be carried out when patients are on or off therapy. METHODS: Thirty-nine adults (mean age, 50 years; 24 female) with refractory reflux symptoms were evaluated by impedance/ph monitoring while on therapy, followed by wireless ph monitoring while off therapy. Non acid reflux events in patients on therapy were correlated with acid reflux parameters studied off therapy. In addition, the likelihood of test abnormalities on and off therapy was determined. RESULTS: In 25 of 39 patients (64%) on therapy, impedance testing was normal, with a median of 69 events (interquartile, ). The percentage of time at ph 4 was within the normal range for all patients who were on therapy. The ph test results were abnormal in 28 of 39 patients (72%) when studied off therapy. Ninety-three of patients with abnormal impedance on therapy also had abnormal acid reflux off therapy. When both groups were off therapy, the patients with abnormal impedance parameters on therapy had significantly higher median (interquartile) 2-day baseline levels of esophageal acid exposure (8.7%, 6.9% 12.5%), compared with those of patients with normal impedance parameters while on therapy (6.0%, 2.8% 9.4%; P.026). CONCLUSIONS: Abnormal impedance in patients on therapy predicts acid reflux in patients off therapy. In patients with refractory reflux, combined impedance/ph monitoring might provide the single best strategy for evaluation of reflux symptoms. Gastroesophageal reflux disease (GERD) is a common chronic disorder with increasing prevalence in the Western world. 1 Approximately 40% of adults frequently complain of heartburn, 2 and GERD remains the leading outpatient physician diagnosis for gastrointestinal disorders in the United States. 3 The combined direct and indirect cost for managing this disease was estimated to be more than US$14 billion in the year 2004, of which 60% was spent on drugs; proton pump inhibitors (PPIs) claimed more than US$9 billion. 3 Despite recent advances in understanding pathophysiology of GERD, its diagnosis and treatment remain challenging in both primary care and specialty settings. Combined multichannel intraluminal impedance ph (MII-pH) monitoring has given new dimensions to GERD diagnosis. 4,5 Current clinical practice guidelines favor empiric trial of PPI over initial testing to treat presumptive GERD without alarming symptoms. 6,7 Patients who fail to improve on optimal dose PPI are often referred for diagnostic evaluation of suspected GERD. MII-pH monitoring, considered the most sensitive of all available reflux monitoring techniques, can detect various types of esophageal refluxate characteristics: gas, liquid, acid, or nonacid. This test is emerging as a new gold standard for clarifying the mechanisms of PPI-refractory symptoms Multicenter MII-pH studies in patients with PPI-refractory symptoms suggest that approximately one third of patients have weakly acid or nonacid reflux. In the background of potent acid suppression, the clinical significance of these findings currently remains controversial. Recent data suggest that most (40% 50%) patients with persistent symptoms on therapy have no temporal correlation between their symptoms and any type of reflux, 13,14 pointing largely toward searching for a non-gerd cause for patients continued symptoms. 11,13 The presence of nonacidic or weakly acidic reflux in patients on PPI therapy is suggested to imply continued reflux and the need for additional therapies. 10,11 Important controversy in patients with continued symptoms despite aggressive PPI therapy is whether to conduct testing on or off PPI therapy. By using both MII-pH monitoring on therapy and wireless ph monitoring off therapy in the same group of patients with PPI-refractory symptoms, the purpose of the present study was to investigate esophageal refluxate parameters to determine (1) whether weakly acid reflux measured on PPI therapy by impedance-ph is a predictor of acid reflux off therapy and (2) whether on or off therapy testing yields the best diagnostic strategy. Materials and Methods The study protocol was approved by the Institutional Review Board for Human Use at Vanderbilt University. Patient Population Patients presenting to the Vanderbilt Center for Swallowing and Esophageal Disorders with PPI-refractory GERD symptoms (heartburn, cough, chest pain, shortness of breath, hoarseness, or regurgitation) were evaluated. A cohort of 69 patients refractory to twice daily dosage of PPIs who had com- Abbreviations used in this paper: GERD, gastroesophageal reflux disease; LES, lower esophageal sphincter; MII, multichannel intraluminal impedance; PPI, proton pump inhibitor; SAP, symptom association probability; SI, symptom index by the AGA Institute /09/$36.00 doi: /j.cgh

2 744 PRITCHETT ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 7, No. 7 bined MII-pH monitoring on therapy was identified for this study. Thirty-nine patients from this cohort agreed to undergo wireless ambulatory ph monitoring off therapy to assess baseline esophageal acid exposure. These 39 patients with both impedance testing on twice a day PPI and ambulatory wireless ph testing off therapy served as the population for the present study. Esophageal Motility Testing High-resolution manometry (Sierra Scientific Instruments Inc, Los Angeles, CA) was used to measure the location of the lower esophageal sphincter (LES) before placement of the MII-pH. A solid state assembly with 36 circumferential sensors spaced at 1-cm intervals (outer diameter, 4.2 mm) was used. This device detects pressure over a length of 2.5 mm in each of the 12 radially dispersed sectors of the 36 pressure-sensing elements. The sector pressures are averaged, making the sensors a circumferential pressure detector. Before recording, the transducers were calibrated at 0 and 100 mm Hg by using externally applied pressure. With this device LES was measured, and the proximal location was noted for placement of the MII-pH catheter. Combined Multichannel Intraluminal Impedance ph Monitoring Patients underwent MII-pH monitoring while on at least twice daily PPI therapy for 1 month before evaluation. They were instructed to fast for 4 hours before testing. Each patient s primary symptom complaint was recorded as part of the preprocedure evaluation. Patients were given diaries to record time of start and end of meals, position changes (upright or recumbent), and occurrence of symptoms during 24-hour impedance and ph monitoring. Impedance testing was performed by using a combined MII-pH monitoring device (Sandhill Scientific Inc, Highlands Ranch, CO) comprising a data recorder (Sleuth System; Sandhill Scientific Inc) and a 2.1-mm-diameter polyvinyl catheter embedded by 1 ph and 6 impedance sensors at predefined positions. The ph sensors were calibrated before placement by using standardized buffer solutions at ph 4.0 and 7.0 as recommended by the manufacturer. The catheter was placed intranasally so that the esophageal ph sensor was positioned 5 cm above the manometrically defined upper border of the LES. Intraluminal impedance was measured at 3, 5, 7, 9, 15, and 17 cm above the LES. Data sampling frequency for both impedance and ph sensors was 50 Hz. Studies were performed for hours, after which patients returned to the lab for catheter removal and data review. Data were downloaded from the recorder and analyzed by using BioView Analysis software (Sandhill Scientific Inc). Reflux episodes were identified by computerized detection (Autoscan; Sandhill Scientific Inc) of proximally directed decreases in impedance. Tracings were also manually reviewed by an experienced investigator (M.F.V.) to confirm accuracy and correct any errors. Total, upright, and supine reflux events were recorded. Acid reflux events were defined as those occurring with ph less than or equal to 4, and nonacid or weakly acid reflux events were defined as those occurring at ph greater than 4. Wireless ph Monitoring Ambulatory ph monitoring was performed for 48 hours by using Bravo wireless ph monitors (Medtronic, Minneapolis, MN). Study patients were instructed to stop taking all PPIs for 7 days before undergoing evaluation. Wireless capsules were calibrated by submersion in buffer solutions at ph 7.0 and ph 1.0 (Medtronic) and then activated by magnet removal. Patients underwent esophagogastroduodenoscopy with conscious IV sedation for visual anatomic inspection and distance measurements from the incisors to the squamocolumnar junction. Capsules were then placed by using the manufacturer s delivery system at 6 cm above the squamocolumnar junction and attached with vacuum suction of 600 mm Hg. Capsule placement was confirmed at endoscopy. After successful placement, patients were given wireless ph recorders to wear about their waists or to keep within 3 5 feet at all times. Recording devices receive ph data sampling transmitted by the capsule at 433 Hz with 6-second sampling intervals. Patients were instructed to perform their normal daily activities and dietary practices. Distal esophageal ph recording was conducted for total of 48 hours. During this time patients kept diaries of meal times, symptoms, and supine position. After completion of the 48-hour study, data were downloaded from recording devices to dedicated computers by using Datalink software (Medtronic). Patient diary information was manually entered into the computer-based record. Measurements of the total, upright, and supine percentage time when esophageal ph was below 4 were determined during day 1 and day 2 of the wireless study. Acid exposure time (% time ph 4) greater than 5.3% per day was considered abnormal. 15 Statistical Analysis Participants were classified as normal or abnormal reflux parameters on the basis of impedance results, with less than 48 reflux events considered normal and 48 or more events considered abnormal. 16,17 To determine whether there was an association between impedance findings on therapy and wireless ph results off therapy, we used Pearson 2 test. We also examined the association between the percent of time below ph 4 and impedance data by using the Wilcoxon rank sum test. Finally, the predicted probability of being classified as abnormal by wireless ph testing as a function of total number of reflux Table 1. Descriptive Statistics for Study Population as a Function of MII-pH Findings n Normal impedance parameters (n 25) Abnormal impedance parameters (n 14) P value Age (25%, 75%) (47 61) 48 (41 58).23 Gender, % female Chief complaint (%).89 Heartburn Cough Sore throat 12 7 Chest pain 0 7 Shortness of breath 4 7 Hoarseness 0 0 Dysphagia 8 0

3 July 2009 EFFICACY OF IMPEDANCE/pH MONITORING IN GERD 745 Figure 1. Flow diagram of study design and likelihood of normal or abnormal impedance and wireless ph parameters. events was estimated by using logistic regression. Continuous variables were summarized by using the median, 25th, and 75th percentiles, and results were considered significant if P.05. Results The study population consisted of 39 patients with refractory symptoms to twice a day PPI therapy (mean age, 50 years; range, years). There was a female predominance (62%), and cough was the most common presenting symptom (56%), followed by heartburn (18%) and sore throat (10%) (Table 1). There were no significant differences between those with abnormal compared with those with normal impedance parameters on therapy with regard to age, gender, and symptoms (Table 1). In 25 of 39 (64%) patients, impedance testing on twice a day PPI showed normal number of weakly acid or nonacid reflux events. The remaining 14 of 39 (36%) patients had abnormal impedance results (Figure 1). In this on-therapy tested group, the percent time ph 4 was within the normal range for 39 of 39 (100%) patients. On wireless ph testing off therapy, 28 Figure 2. Box plot of % time ph 4 for day 1 and day 2 on the basis of impedance results (range, upper and lower quartile, and median values displayed). of 39 (72%) had abnormal acid reflux parameters, and 11 of 39 (28%) had normal findings (Figure 1). The median (interquartile) number of reflux events was 69.0 ( ) for those with abnormal compared with 28.0 ( ) in those with normal MII-pH parameters (P.001) (Table 2). Reflux events were more common for both groups in the upright position, median (interquartile) of 58.0 ( ) compared with 26.0 ( ), respectively (P.001). Patients with abnormal impedance parameters on PPI therapy had significantly (P.026) higher median (interquartile) 2-day baseline off PPI esophageal acid exposure, 8.7% (6.9% 12.5%), than those with normal impedance parameters, 6.0% (2.8% 9.4%) (Figure 2). There was a statistically significant difference (P.002) in median (interquartile) percent total time of ph 4 on day one, 8.8% (7.2% 18.2%) versus 5.0% (2.5% 9.7%), respectively (Table 2). The degree of acid exposure off PPI therapy was significantly higher in those with abnormal than normal on-therapy impedance parameters in the upright (P Table 2. MII-pH and Wireless ph Parameters Normal a Abnormal a P value Impedance parameters Total no. of reflux episodes (25% 75%) 28.0 ( ) 69.0 ( ).001 No. of upright reflux 26.0 ( ) 58.0 (53 67).001 No. of supine reflux 2.0 ( ) 6.0 ( ).011 Wireless ph parameters Total % of time ph 4 (25% 75%) (average of days 1 and 2) 6.0 ( ) 8.7 ( ).026 Total % of time ph 4 (day 1) 5.0 ( ) 8.8 ( ).002 % of time upright 8.4 ( ) 12.4 ( ).01 % of time supine 0.6 ( ) 6.6 ( ).05 Total % of time ph 4 (day 2) 6.9 ( ) 7.7 ( ).44 % of time upright 8.2 ( ) 9.6 ( ).85 % of time supine 1.5 ( ) 3.0 ( ).18 a Categorized by total number of reflux events by impedance monitoring on therapy.

4 746 PRITCHETT ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 7, No. 7 Table 3. Proportion Abnormal by Wireless ph off Therapy as a Function of MII-pH Findings on Therapy Impedance Wireless ph Normal (n 25) Abnormal (n 14) P value Normal 40% (n 10) 7% (n 1).029 Abnormal 60% (n 15) 93% (n 13).01) and supine (P.05) periods for day 1 but not for day 2 of the monitoring periods. Ambulatory wireless ph study results off therapy were compared as a function of impedance diagnosis (Table 3). In 14 patients with abnormal impedance on therapy, 13 (93%) also had abnormal acid reflux detected by wireless ph off therapy. In comparison, of the 25 with normal impedance on therapy, only 10 (40%) were also normal by wireless ph testing off therapy. This difference was statistically significant, P.029 (Figure 3). The number of reflux events identified by impedance monitoring on therapy correlated with abnormal acid reflux detected by wireless ph monitoring off therapy. As the number of impedance reflux events increased, there was increasing likelihood of having an abnormal wireless ph study off therapy (Figure 4). Discussion By using a stepwise diagnostic approach in the same group of patients with refractory GERD symptoms to twice daily PPI therapy, we assessed the degree of esophageal acid and weakly acid reflux on therapy as a function of esophageal acid reflux off therapy. Our findings suggest that an abnormal number of weakly acid reflux on therapy is a strong predictor of baseline acid reflux. We showed that nearly all (93%) patients with abnormal MII-pH on therapy had abnormal acid reflux when studied by wireless ph monitoring off therapy. Patients with abnormal nonacid or weakly acid reflux by MII-pH were also likely to have more severe degree of acid reflux at baseline than patients with normal impedance results (Figure 2). We showed that higher on-therapy number of reflux events detected by MII-pH predicted higher acid reflux by wireless ph testing off therapy. Collectively these data suggest that MII-pH abnormality on PPI therapy can be considered a surrogate marker of more severe acid reflux events at baseline in this group of patients. Our findings are mechanistically consistent with studies showing that PPIs transform the acidity of reflux with little effect on its frequency. 18,19 Figure 3. Flow diagram of likelihood of off-therapy ph findings as a function of on-therapy impedance data. Figure 4. Probability of being classified abnormal by wireless ph on the basis of total number of impedance events. Our data help clarify the important clinical question of how best to perform the initial diagnostic test in those with refractory symptoms. When patients present with symptoms suggestive of GERD and fail to improve on optimal acid suppression therapy, there is a debate over the most efficient evaluation strategy: do we test on therapy or off therapy? This question, although simple, is complicated by the lack of data on likelihood of detecting abnormal results and their clinical implication. In a given patient with refractory symptoms, the 3 likely causes of treatment failure might be inadequate acid suppression, 20 symptoms caused by weakly acidic reflux events, 4,5,19 or erroneous diagnosis of GERD. 21 Thus, the main goal of the optimal diagnostic test is to determine which of these possibilities are most likely. Our data provide a unique perspective into this important clinical dilemma. We found that using combined MII-pH monitoring on therapy might be the most effective strategy of evaluating PPI-resistant GERD, because it can potentially exclude the greatest number of patients from further testing (Figure 1). All those tested on therapy had normalized esophageal acid exposure, excluding continued acid reflux as the potential contributing factor to their persistent reflux. This finding is consistent with previous data. 22 In addition, the majority (64%) of patients tested with MII-pH on therapy were more likely to have normal impedance parameters (weakly or nonacid), suggesting causes other than GERD. Finally, among the minority of patients (36%) who were found to have abnormal impedance studies on therapy, abnormal esophageal acid exposure off therapy was highly likely (Figure 3), suggesting a possible role for reflux, although cause and effect between nonacid reflux and patients symptoms are still uncertain and await direction from future controlled outcome studies. The alternative strategy of testing patients off therapy first would have identified abnormal acid reflux in 72% of this group of patients (Figure 1), and the etiology for persistent symptoms would then have to be pursued further by impedance monitoring. Therefore, likelihood probabilities from our data suggest that MII-pH monitoring might be the initial test of choice if only 1 test is to be performed.

5 July 2009 EFFICACY OF IMPEDANCE/pH MONITORING IN GERD 747 In contrast to our findings, a recent study suggested that off-ppi MII-pH testing should be preferred to on-therapy testing in the evaluation of PPI-resistant GERD symptoms. 23 However, the study population in this trial was restricted to only those with typical GERD symptoms and only focused on symptom association probability, ignoring the PPI-resistant nature of the patient population. In addition, their results are in conflict with a prior multicenter study of MII-pH monitoring and symptom analysis in 79 patients off PPI and 71 patients on PPI, which concluded that diagnostic yield of MII-pH monitoring was higher on than off therapy. 14 The findings from our study support the latter contention. Ruling out GERD or defining the type of reflux and relationship to symptoms obviously facilitates choosing the right therapy. Normal findings on physiologic testing such as MII-pH monitoring will guide toward searching for other potential non-gerd causes for the continuing symptoms. On the basis of our data, this is the most likely scenario occurring in 64% of patients with refractory symptoms. Documentation of abnormal acid reflux on optimal PPI dosage might indicate lack of compliance, need for reevaluating acid suppression regimen, or, rarely, presence of Zollinger-Ellison syndrome. Prior studies in patients with GERD symptoms refractory to therapy suggested that the likelihood of continued abnormal acid reflux is no more than 1% 4%. 22 In this study we confirmed these findings, showing that the likelihood of continued only acid reflux on twice a day PPI therapy is remote. Presence of abnormal weakly acid and nonacid reflux might guide toward the need for medical (baclofen or metoclopramide/domperidone) or possibly mechanical reflux reduction therapy, 10,11 although there are no controlled outcome studies to date for this assertion. Our data suggesting increased baseline acid reflux in this group are an important step in better understanding the physiology of reflux in this difficult to treat group of patients. Some limitations of our study should be highlighted. First, the results from our study underscore the need for larger outcome studies in those with abnormal MII-pH parameters in patients with refractory symptoms. Second, present analysis discusses the MII-pH findings globally with respect to abnormal number of reflux events in the distal esophagus to address the aim of the study. We did not assess proximal extent and liquid, gas, or mixed nature of the refluxate as some believe might be important in a subgroup of treatment-resistant patients. 5 In addition, we have used number of reflux events as the primary measure as opposed to symptom association probability (SAP) or symptom index (SI). The use of SAP and SI is problematic in this group because patients have already declared lack of clinical response to aggressive acid suppression. Number of reflux events is a more conservative cutoff parameter than either SAP or SI, which is what is needed for this group of patients. This is based on our recent data in 143 subjects that showed that when evaluating patients on therapy, total number of reflux events and not SI or SAP is the most conservative estimate as well as the one with the highest likelihood of encompassing other symptom association parameters. 24 Finally, impedance measurements are based on catheter tests, which can have false-negative results, as compared with wireless ph testing, in which this false-negative test might be less. In conclusion, we recommend MII-pH testing on therapy in patients with persistent symptoms. In this group of patients most will have normal results, suggesting causes other than GERD for patients continued symptoms. In those with abnormal weakly acid or nonacid reflux on therapy, MII-pH has a strong positive predictive value (93%) for baseline acid reflux. It remains to be determined whether the abnormal MII-pH events are due to PPI effect or clinically significant physiologic markers. However, at the very least and given the probability ratios from our data, this test performed on therapy might be the single best strategy in eliminating reflux as the cause of patients persistent symptoms. References 1. Dent J, El-Serag HB, Wallander MA, et al. Epidemiology of gastrooesophageal reflux disease: a systematic review. Gut 2005;54: Locke GR, Talley NJ, Fett SL, et al. Prevalence and clinical spectrum of gastroesophageal reflux: a population-based study in Olmsted County, Minnesota. Gastroenterology 1997;112: Shaheen N, Hansen R, Morgan D, et al. The burden of gastrointestinal and liver diseases, Am J Gastroenterol 2006;101: Tutuian R, Vela M, Hill E, et al. Characteristics of symptomatic reflux episodes on acid suppressive therapy. Am J Gastroenterol 2008;103: Zerbib F, Duriez A, Roman S, et al. Determinants of gastrooesophageal reflux perception in patients with persistent symptoms despite proton pump inhibitors. Gut 2008;57: Kahrilas PJ, Shaheen NJ, Vaezi MF. American Gastroenterological Association Institute technical review on the management of gastroesophageal reflux disease. Gastroenterology 2008;135: Kenneth R. DeVault DOC. Updated guidelines for the diagnosis and treatment of gastroesophageal reflux disease. Am J Gastroenterol 2005;100: Park W, Vaezi M. Esophageal impedance recording: clinical utility and limitations. Curr Gastroenterol Rep 2005;7: Sifrim D, Blondeau K. Technology insight: the role of impedance testing for esophageal disorders. Nat Clin Pract Gastroenterol Hepatol 2006l;3: Tutuian R, Castell D. Review article: complete gastro-oesophageal reflux monitoring combined ph and impedance. Aliment Pharmacol Ther 2006;24(Suppl 2): Agrawal A, Castell D. Clinical importance of impedance measurements. J Clin Gastroenterol 2008;42: Bredenoord A, Tutuian R, André J, et al. Technology review: esophageal impedance monitoring. Am J Gastroenterol 2007; 102: 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: Zerbib F, Roman S, Ropert A, et al. Esophageal ph-impedance monitoring and symptom analysis in GERD: a study in patients off and on therapy. Am J Gastroenterol 2006;101: Pandolfino J, Richter J, Ours T, et al. Ambulatory esophageal ph monitoring using a wireless system. Am J Gastroenterol 2003; 98: Zerbib F, des Varannes S, Roman S, et al. Normal values and day-to-day variability of 24-h ambulatory oesophageal impedance-ph monitoring in a Belgian; French cohort of healthy subjects. Aliment Pharmacol Ther 2005;22: Tutuian R, Mainie I, Agrawal A, et al. Normal values for ambulatory 24-h combined impedance-ph monitoring on acid suppressive therapy. Gastroenterology 2006;130(Suppl 2):A171.

6 748 PRITCHETT ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 7, No Vela MF, Camacho-Lobato L, Srinivasan R, et al. Simultaneous intraesophageal impedance and ph measurement of acid and nonacid gastroesophageal reflux: effect of omeprazole. Gastroenterology 2001;120: Tamhankar AP, Peters JH, Portale G, et al. Omeprazole does not reduce gastroesophageal reflux: new insights using multichannel intraluminal impedance technology. J Gastrointest Surg 2004;8: Furuta T, Shirai N, Watanabe F, et al. Effect of cytochrome P4502C19 genotypic differences on cure rates for gastroesophageal reflux disease by lansoprazole. Clin Pharmacol Ther 2002; 72: Smout AJPM. The patient with GORD and chronically recurrent problems. Best Pract Res Clin Gastroenterol 2007;21: Charbel S, Khandwala F, Vaezi M. The role of esophageal ph monitoring in symptomatic patients on PPI therapy. Am J Gastroenterol 2005;100: Hemmink G, Bredenoord A, Weusten B, et al. Esophageal phimpedance monitoring in patients with therapy-resistant reflux symptoms: on or off proton pump inhibitor? Am J Gastroenterol 2008;103: Trawick E, Slaughter J, Klochan C, et al. Impedance monitoring in refractory GERD: SI, SAP, or number of events? Gastroenterology 2009;136(Suppl 1):S1896. Reprint requests Address requests for reprints to: Michael F. Vaezi, MD, PhD, MSc (Epi), Division of Gastroenterology, Hepatology and Nutrition, Vanderbilt University Medical Center, 1660 TVC, nd Avenue South, Nashville, Tennessee Michael.vaezi@vanderbilt.edu; fax: (615) Conflicts of interest This author discloses the following: Dr Vaezi has research grant support from Takeada, AstraZeneca, and Restech. The remaining authors disclose no conflicts.