T2006

Measurement of Pharyngeal pH: A New Technology and Normal ValuesShahin Ayazi, John C. Lipham, Jeffrey A. Hagen, Andrew Tang, Jessica M. Leers, ArzuOezcelik, Emmanuele Abate, Steven R. DeMeester, Farzaneh Banki, Tom R. DeMeester

Introduction: Patients with respiratory and laryngeal complaints often have gastroesophagealreflux but classic symptoms are frequently absent and clinical findings are nonspecific.Current methods of measuring pharyngeal acid exposure are often inaccurate due to technicalartifacts or probe malfunction. A newly designed pharyngeal pH probe has been introducedwhich detects aerosolized acid and is placed under direct vision. The aim of this study wasto determine normal values for pharyngeal acid exposure using this catheter. Methods:Seventy eight asymptomatic volunteers were recruited. All had esophageal manometry anda video esophagogram. Dual channel esophageal pH monitoring was performed with thedistal probe 5 cm above the lower sphincter and the proximal probe 1-4 cm below the uppersphincter. Simultaneously, pharyngeal pH was monitored with the new probe positioned 0.5-1 cm below the uvula. Subjects with abnormal distal esophageal acid exposure, a hiatalhernia larger than 2 cm, or a poor technical recording were excluded. Pharyngeal pH wasanalyzed using the standard components of esophageal pH monitoring at pH thresholdsfrom 4 to 6 in 0.5 increments and a composite score was calculated. A separate analysiswas performed for the upright and supine periods and the 95th percentile was determined.Results: The study population consisted of 55 subjects (28 M/27F) with a mean age of 31years (range19-72). The 95th percentile values for pharyngeal acid exposure are shown(Table). In the upright period, 5.5 is the best pH threshold to define abnormal acid exposure.Pharyngeal acid exposure is considerably higher in the supine period and a lower thresholdis necessary. For this period, pH <5.0 would maximize sensitivity and pH < 4.5 wouldmaximize specificity. The 95th percentile values for the composite score were: 9.4 for theupright period, and 4.8 and 6.7 for the supine period at pH < 5 and pH < 4.5 respectively.Conclusion: A newly designed pharyngeal pH probe which detects aerosolized acid mayovercome the shortcomings of current techniques. Using this probe we have defined normalvalues for pharyngeal acid exposure in a large series of normal volunteers. These values cannow be used to determine if patients with laryngeal or respiratory symptoms have abnormalpharyngeal acid exposure.

T2007

Role of Esophageal Histology in the Diagnosis of Gastroesophageal RefluxDisease (GERD) in Patients Without Erosive Disease By Standard Endoscopyor Narrow Band Imaging (NBI)Christopher R. Lynch, Sharad C. Mathur, Ajay Bansal, Amit Rastogi, Krishna Pondugula,Sachin B. Wani, John D. Keighley, April D. Higbee, Prateek Sharma

Background: Most GERD pts do not have erosive disease (NERD) by white light endoscopy(WLE). NBI is a technique that enhances mucosal visualization of the distal esophagusand can reveal subtle erosions (not seen by WLE). There is also emerging evidence thathistopathologic criteria can be of diagnostic value in GERD. Aim: To examine the diagnosticvalue of esophageal histology in patients with reflux symptoms without erosive disease byeither WLE or NBI. Methods: GERD pts on the basis of two validated questionnaires[gastroesophageal reflux questionnaire (GERQ) and reflux disease questionnaire (RDQ)] andan asymptomatic control group were prospectively examined by WLE followed by NBI.Patients were classified as NERD if there was no macroscopic erosive esophagitis (by LAclassification) or subtle erosions by NBI (micro-erosions). NERD pts and controls underwentbiopsy 1 cm above GEJ. Specimens were reviewed by a pathologist blinded to clinicaland endoscopic findings. A validated histological scoring system [accounting for basal cellhyperplasia, papillary elongation, dilation of intracellular spaces (DIS), necrosis/erosion, andneutrophil/eosinophil infiltration—score 0-2 for each] was used with a total score of > 2abnormal. Statistical analysis was done using Fisher's exact test and Spearman's correlationcoefficient. Results: 21 patients had GERD symptoms but no erosive esophagitis on WLE;6 of the 21 pts had micro-erosions on NBI and were excluded. Thus 15 NERD and 35controls were included. There were no significant differences between NERD and controlswith respect to age (62.5 + 11.1 yrs vs. 62.9 + 8.92 yrs; p=NS) or gender (93.3% vs. 97.1%males; p=NS). Mean histology score was significantly higher in NERD group vs. controls(4.1 + 2.2 vs. 1.9 + 1.3; p = 0.003). Also, percentage of pts with a histology score > 2 washigher in NERD than in controls (93.3% vs. 61.8%; p=0.038). There was a significantcorrelation between total histology score and NERD (r=0.55, p=<0.001) with a score of >2having the following characteristics for diagnosing NERD: sens. 93.3%; spec. 38.2%; NPV92.9% and PPV 40.0%. Presence of DIS also correlated significantly with NERD (r=0.34,p=0.001), but DIS alone did not discriminate between NERD and controls (87% vs. 65%,p=NS). Conclusions: In true NERD patients (without erosions by WLE or NBI), biopsies 1cm above SCJ revealed significantly higher histopathological scores compared to controls.A score of > 2 had a high sensitivity and NPV but a low specificity and PPV, suggestingthat esophageal histology may have clinical utility in excluding GERD in patients with refluxsymptoms and a normal WLE and NBI examination.

T : 11501$$CH204-02-08 16:47:13 Page 599Layout: 11501B : o

A-599 AGA Abstracts

T2008

In Refractory GERD Patients Should We Perform Impedance On Therapy orpH Monitoring Off Therapy?Brian S. Smith, James C. Slaughter, Reid M. Ness, Michael F. Vaezi

Introduction One of the most controversial issues in patients with refractory typical orextraesophageal GERD symptoms is whether to perform impedance monitoring on PPItherapy or pH monitoring off therapy. The aim of this study was to assess the probabilityof abnormal test results in either of the above two conditions and to determine if one methodresults in better clinical yield. Methods Patients with refractory GERD symptoms (cough,HB/regurgitation, sore throat, hoarseness) on BID PPI underwent 96-hour wireless (Bravo)capsule pH monitoring 48 hrs off followed by 48 hrs on BID PPI therapy. A similar butseparate group also underwent 24-hour ambulatory multichannel intraluminal impedance(MII) monitoring on BID PPI therapy followed by wireless pH monitoring for 48 hours offtherapy. Acid suppressive therapy was discontinued for at least seven days prior to capsuleplacement 6 cm above the SCJ in both groups. In the first group, pH data collected whileoff acid suppression for the first 48 hours, then on BID PPI therapy for the subsequent 48hours. % total, upright and supine times pH < 4 were measured and considered abnormalif greater than 5.0%, 8.3% and 3.0%, respectively on any two study days. MII catheter waspositioned 5 cm above the manometrically determined LES in the second group. Abnormalnon-acid reflux events on BID PPI therapy were defined as previously reported by >48 totalreflux episodes. Results 35 patients (median age 52 yrs; 72% female) underwent wirelesscapsule study and only 7/35 (20%) had normal reflux parameter off PPI therapy. Theprobability of normalized pH while on therapy was 100% for this group. 43 patients (medianage=54; 56% female) underwent on therapy MII monitoring and 18/43 (42%) had abnormalnon-acid reflux parameters and the probability of abnormal esophageal acid exposure offPPI therapy for this group was 90%. Conclusions In patients with refractory GERD ontherapy, most patients (58%) will have normal impedance findings suggesting lack of symp-tom association with acid or non-acid reflux. However, in the 42% who have abnormalimpedance parameters 90% will have acid reflux parameters off therapy. Thus, both findingshave important clinical implications. However, off therapy pH studies are most helpful ifpH data suggest normal parameters which occur in only 20% of patients. Majority (80%)have abnormal pH and the clinical role for this abnormality will still need to be furtherassessed by additional testing. Therefore, patients with refractory GERD are best studiedwith MII while on therapy.

T2009

Cyp2c19 Genetic Polymorphism Related to the Proton-Pump Inhibitor Test inChinese Patients with Symptoms Suggestive of Gastro-Esophageal RefluxDiseasePing-Huei Tseng, Yi-Chia Lee, Han-Mo Chiu, Hsiu-Po Wang, Jaw-Town Lin, Ming-ShiangWu

Backgrounds & Aims: Proton-pump inhibitor (PPI) test has been proposed as a valuabletool in the diagnosis of gastroesophageal reflux disease (GERD) in Western populations. Weassumed that a higher prevalence of poor metabolizers in Chinese population might confoundthe diagnostic accuracy of a PPI test. Methods: In this open-label, randomized trial, patientswith symptoms suggestive of GERD were randomly assigned to receive a 2-week test withdaily rabeprazole 40mg or daily pantoprazole 80mg after diagnostic endoscopy. Therapeuticresponse was assessed with a five-grade daily record. Genotype of CYP2C19 polymorphismwas determined in each patient. Results: Between April 2006 and August 2007, a total of155 reflux patients completed the study. Eighty (51.6%) patients were diagnosed witherosive esophagitis and the remaining 75 (48.4%) were endoscopy-negative reflux disease.Based on the cut-off value for 50% symptom reduction, the sensitivity, specificity, positivepredictive value, negative predictive value, and accuracy in the prediction of erosive esophag-itis were 90.7%, 55.9%, 72.2%, 82.6% and 75.3% for rabeprazole, and 91.9%, 61.0%,68.0%, 89.3%, and 75.6% for pantoprazole. Regarding the CYP2C19 polymorphic genotypes,52 (38.2%), 68 (50.0%), and 16 (11.8%) patients were homogenous extensive, heterogeneousextensive, and poor metabolizers, respectively. The derived sensitivity, specificity, positivepredictive value, negative predictive value, and accuracy were 90.5%, 63.2%, 73.1%, 85.7%,and 77.5% for extensive metabolizers and 87.5%, 25%, 53.8%, 66.7%, and 56.3% for poormetabolizers. A lower diagnostic accuracy of PPI test in the poor metabolizers was notedwhen compared with that of extensive metabolizers (Fisher's exact test, P = 0.067), barelyto achieve statistical significance. Conclusions: The clinical application of PPI testing inChinese reflux patients might be affected by the CYP2C19 genetic polymorphism, whichwas related to a lower specificity in patients who metabolized PPI poorly.

T2010

Ambulatory 24-H Combined Impedance-pH Monitoring On Acid SuppressiveTherapy - Lessons Learned from Swiss Healthy VolunteersPatrick Janiak, Daniel Pohl, Michael Fried, Radu Tutuian

Background: Combined multichannel intraluminal impedance and pH (MII-pH) is increas-ingly used to monitor patients with persistent reflux symptoms on proton pump inhibitor(PPI) therapy. While normal values for MII-pH monitoring off therapy has been publishedby several groups, there have been only preliminary reports on normal values on PPI bidbased on data from 20 healthy US volunteers (Tutuian et al DDW 2006). Aim: Definenormal values for 24-h impedance pH monitoring on therapy based on data from Swisshealthy volunteers. Methods: Healthy volunteers underwent combined MII-pH monitoringon twice daily esomeprazole 40mg for 7 days. A 8 channel combined MII-pH probe recordedimpedance data at 3, 5, 7, 9, 15 and 17 cm above the LES and pH data 5cm above and10cm below the LES. In each subject we evaluated distal esophageal acid exposure (%timepH<4), the total number of MII-detected reflux episodes, the number of acid reflux episodes(nadir pH <4) and non-acid/weakly acidic (nadir pH 4-7) reflux episodes. Results: Sixteenhealthy volunteers (5F, age mean 24, range 19-42 years) completed impedance-pH mon-itoring on therapy. During dosing with PPI % time pH<4 (median [5th-95th perc]) in healthy

AG

AA

bst

ract

s