Gastroesophageal reflux disease (GERD) often is implicated

Transcription

1 CLINICAL GASTROENTEROLOGY AND HEPATOLOGY 2006;4: ORIGINAL ARTICLES Surgical Fundoplication in Laryngopharyngeal Reflux Unresponsive to Aggressive Acid Suppression: A Controlled Study JASON SWOGER,* JEFF PONSKY, DOUGLAS M. HICKS, JOEL E. RICHTER,* TOM I. ABELSON, CLAUDIO MILSTEIN, MOHAMMED A. QADEER,* and MICHAEL F. VAEZI* *Department of Gastroenterology and Hepatology, Center for Swallowing and Esophageal Disorders; Head and Neck Institute; and the Department of General Surgery, Cleveland Clinic Foundation, Cleveland, Ohio See CME exam on page 396; see editorial on page 431. Background & Aims: In patients with persistent laryngeal symptoms despite aggressive proton pump inhibitor therapy, gastroesophageal reflux disease (GERD) continues to be implicated. The role of surgical fundoplication as the definitive therapy for these patients is uncertain. Methods: In this prospective concurrent controlled study, 72 patients with suspected GERDrelated laryngeal symptoms received aggressive acidsuppressive therapy. Four-month symptomatic nonresponders (<50% improvement) with continued laryngeal inflammation and normalized esophageal acid exposure were offered laparoscopic Nissen fundoplication. The primary outcome was symptom improvement/resolution at 1 year after surgery. Results: Twenty-five of 72 (35%) patients remained unresponsive after 4 months of acid-suppressive therapy. Ten patients (40%) underwent surgical fundoplication (median age, 54 y; men, 4) and 15 patients (60%) continued medical therapy (median age, 52; men, 4). The most common laryngeal symptoms were sore throat, hoarseness, and cough. ph studies at 3 and 12 months were normal in all patients after fundoplication (median % time ph < 4,.0% and.3%; respectively). One of 10 (10%) patients in the surgery group reported improvement of laryngeal symptoms at 1 year compared with 1 of 15 in the control group (6.7%) (P 1.0). Treatment of causes other than GERD improved symptoms in an additional 2 of 10 (20%) patients in the surgical group, and 10 of 15 (66%) patients in the nonsurgical cohort. Conclusions: Surgical fundoplication does not improve laryngeal symptoms reliably in patients unresponsive to aggressive proton pump inhibitor therapy. The argument of low volume or intermittent reflux as the cause of persistent laryngeal symptoms needs to be replaced with evaluation and therapy for other potential non- GERD causes. Gastroesophageal reflux disease (GERD) often is implicated as the cause of chronic laryngeal inflammation, also known as laryngopharyngeal reflux (LPR). Up to 10% of visits to otolaryngologists are thought to result from complaints related to GERD. 1 4 Patients suspected of having LPR present with a variety of chronic symptoms including hoarseness, sore or burning throat, chronic cough, throat clearing, globus, nocturnal laryngospasm, otalgia, postnasal drip, and dysphagia. 2 Laryngeal abnormalities in this group may include edema and erythema, posterior pharyngeal-wall cobblestoning, vocal cord ulcers, interarytenoid changes, medial arytenoid wall edema and erythema, vocal cord granulomas, and subglottic stenosis. 2 Although the pathophysiology is understood poorly, the symptoms and signs of LPR are believed to be caused by microaspiration of gastric contents into the hypopharynx and larynx, causing direct irritation of the affected structures. 2 Medical therapy with proton pump inhibitors (PPIs) may result in symptom and laryngeal improvement in 50% 70% of patients. 2 4 However, many patients continue to have symptoms and laryngeal irritation despite aggressive PPI therapy. 2,4 In this subset of patients, despite normalized esophageal acid exposure on acid-suppressive therapy, 2 GERD continues to be implicated as the probable cause. Laryngeal exposure to intermittent or low-volume acidic or nonacidic gastric reflux or resistance to PPIs is argued as the possible underlying cause. 5 Such contentions ultimately suggest surgical fun- Abbreviations used in this paper: DGER, duodenogastroesophageal reflux; GERD, gastroesophageal reflux; LESP, lower esophageal sphincter pressure; LPR, laryngopharyngeal reflux; PPI, proton pump inhibitor; UESP, upper esophageal sphincter pressure by the American Gastroenterological Association Institute /06/$32.00 doi: /j.cgh

2 434 SWOGER ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 4, No. 4 doplication as the definitive treatment for suspected LPR patients refractory to medical therapy. Surgical fundoplication is an established surgical procedure in preventing both acid and nonacid reflux and is known to be effective in patients with typical symptoms (heartburn and regurgitation) of GERD. 6 However, the role of fundoplication in treating the patient with GERD-suspected laryngeal irritation and chronic symptoms has not been well studied. Thus, the aim of this prospective concurrent control study was to assess the impact of surgical fundoplication on laryngeal symptoms and signs in individuals unresponsive to an aggressive trial of acid-suppressive therapy. Materials and Methods This study was approved by Institutional Review Board of the Cleveland Clinic Foundation and each participant provided written informed consent before inclusion in the study. Patient Population and Study Design Since 1997 an ongoing cohort study at the Cleveland Clinic Foundation s Department of Gastroenterology and Hepatology and Head and Neck Institute has investigated the response of patients with chronic laryngeal symptoms and signs to a 4-month trial of aggressive acid-suppressive therapy. 7 In this study, patients with chronic laryngeal signs and symptoms presumed to be GERD related were tested initially with esophagogastroduodenoscopy, manometry, and 24-hour ambulatory ph monitoring, and, irrespective of findings, were treated aggressively with high-dose PPI therapy (40 mg of omeprazole twice a day or 60 mg of lansoprazole twice a day) for 4 months. Between August 1, 2001, and August 31, 2003, patients unresponsive to the earlier-described therapy ( 50% symptomatic improvement) were screened for potential enrollment in the current study and were offered laparoscopic Nissen fundoplication. Inclusion criteria were age older than 18 years, chronic symptoms (cough, throat clearing, sore/burning throat, hoarseness, and globus), and laryngoscopic findings suggestive of LPR despite aggressive acid suppression. In addition, patients were required to have abnormal ph monitoring or evidence of reflux by barium esophagram off therapy. Patients were excluded from the study if they had Barrett s esophagus, prior surgical therapy for GERD, or if they were unwilling to participate in the study. All patients meeting the inclusion criteria completed a 4-page nurse-administered questionnaire at the start of the study. The questionnaire consisted of demographic information, medical history, tobacco and alcohol use, current medication regimen, history of allergies, presence of other potential laryngeal irritants (asthma, sinusitis, sinus congestion, allergies, postnasal drip, viral illness, and vocal abuse), and a description of specific laryngeal symptoms. Patients were asked to report the presence or absence of symptoms (cough, hoarseness, throat clearing, sore throat, globus sensation, heartburn, regurgitation, problem swallowing, chest pain, and discomfort to talk) by answering yes or no on the questionnaire and score the severity using a scale of 1 to 4 (1 rare: once a month or less; 2 occasional: once a week or less; 3 frequent: several times a week; and 4 all the time: several times daily). Patients electing to undergo surgery underwent baseline evaluation, including esophageal manometry, dual-probe 24- hour ambulatory ph and bilirubin (Bilitec 2000; Synectics, Stockholm, Sweden) monitoring on PPI therapy, and preoperative laryngoscopy. An ambulatory ph study on twice-daily PPI therapy had to show evidence of a normalized esophageal acid pattern (% time ph 4 of less than 5.52 total time) for inclusion in the study. Participants also had a complete preoperative evaluation including chest radiograph, 12-lead electrocardiogram, blood testing (complete blood count, basic metabolic panel, and coagulation panel), and had a consultation with the surgeon ( J.P.) and an anesthesiologist. Postfundoplication patients were monitored for 1 year. All patients kept daily symptom diaries during the study period using the earlier-described severity scoring system. All patients also had blinded symptom assessment using a standard questionnaire at 1, 3, 6, and 12 months after surgery. Patients were asked to score their symptoms (as compared with the previous visit) on a scale of 0% 100% improvement. They underwent objective evaluation including esophageal manometry and 24-hour ph monitoring at 3 months after fundoplication and ph monitoring (Bravo ph capsule; Medtronic Corp, Shoreview, MN) at 12 months after fundoplication. The response of laryngeal signs to surgical fundoplication was evaluated at 6 and 12 months after fundoplication by a single endoscopist blinded to patients symptomatic response status. Clinical (symptomatic and laryngoscopic) response to fundoplication at 12 months was compared with PPI-unresponsive patients who declined surgery and continued on twice-daily PPI therapy for the same length of time. The control group also underwent objective baseline testing including 24-hour ambulatory ph monitoring, esophageal manometry, esophagogastroduodenoscopy, and laryngoscopy. Follow-up symptom assessments were performed in the controls by using the same questionnaire used in fundoplication patients assessing laryngeal symptoms and medication usage. Laryngoscopy Patients underwent a videotaped fiberoptic laryngoscopy at the Cleveland Clinic Head and Neck Institute before surgery and at 6 and 12 months after surgery. The laryngoscopy was performed by 1 specialist (D.M.H.) who was blinded to patient symptoms and group allocation. To ensure technical consistency, all examinations followed a standard protocol of patient instructions and tasks, and used the same equipment. The patients were seated comfortably and asked to sniff 3 sprays of a mixture of topical anesthetic and decongestant (1% neosynephrine; 4% Xylocaine [AstraZeneca, Wilmington, DE]) through 1 nostril. The flexible nasopharyngoscope (Pentax model FNL 13S; Pentax Precision Instrument Corp, To-

3 April 2006 FUNDOPLICATION IN LARYNGOPHARYNGEAL REFLUX 435 kyo, Japan) was coupled to a color video camera (Toshiba model IK-C30A; Toshiba Corp, Tokyo, Japan) and was illuminated by a xenon light source (Karl Storz model 610; Karl Storz Endoscopy American Inc., Culver City, CA). The scope was inserted into the nostril, passed along the nasal floor, and maneuvered to the oropharynx and upper hypopharynx. The patients was asked to perform a series of vegetative tasks (breathing, swallowing, coughing, sniffing, and throat clearing) and vocal tasks (whistling, sustained phonation, alternate phonation/inhalation, and pitch glides) designed to allow visualization of the posterior pharyngeal wall, hypopharynx, and intrinsic larynx. Findings of erythema, edema, redundant tissue, surface irregularities, and other lesions were noted for each structure (posterior pharyngeal wall, posterior commissure [anterior cricoid wall], posterior cricoid wall, arytenoids complex, true and false vocal cords, anterior commissure, epiglottis, and aryepiglottic folds) and were recorded by the examiner on a checklist (see Appendix online at 7,8 Esophageal Manometry Patients underwent esophageal manometry after an overnight fast before surgery and 3 months after surgery. The test was performed using a 4-mm diameter, round polyvinyl catheter (Arndorfer Medical Specialties Inc., Milwaukee, WI). The catheter was perfused continuously with distilled water at a rate of.5 ml/min using a low-compliance pneumohydraulic capillary infusion system. The location and length of the upper-esophageal sphincter (UES) and the lower-esophageal sphincter (LES) were determined by using the stationary pullthrough technique. Esophageal motility was assessed by having the patient perform 10 wet swallows (5 ml each) with the distal recording probe positioned at 5 cm above the LES. The locations and pressure of the LES and the UES were recorded by a computerized motility system (Synectics Gastrosoft Polygram, Milwaukee, WI). Abnormal manometry was defined as follows: hypotensive LES pressure (LESP, 10 mm Hg), hypotensive UES pressure (UESP, 30 mm Hg), ineffective esophageal motility ( 20% low-amplitude esophageal body contractions of 30 mm Hg and/or nonconducted peristalsis), and nutcracker esophagus (body amplitude, 180 mm Hg). 9 Ambulatory ph Monitoring Twenty-four hour ambulatory esophageal ph monitoring was performed immediately after manometry before PPI therapy, while on twice-daily PPI therapy before fundoplication, and 3 months after surgery. Before probe insertion, the patients was asked to sniff 3 sprays of anesthetic (lidocaine hydrochloride 2%; phenylephrine.5%) through 1 nostril. The ph study was performed using 2.1-mm monocrystal line ph catheters with 2 antimony electrodes separated by 15 cm (Medtronic Functional Diagnostics Zinetics, Inc., Salt Lake City, UT). The reference electrode was internalized. The electrodes were calibrated at 37 C in buffer solutions of ph 1 and ph 7 (Fisher Scientific, Fairlawn, NJ) before and after each study. After calibration, the ph probe was passed nasally, with the distal electrode being positioned 5 cm above the proximal border of the LES, and the proximal electrode being positioned just below the UES. The probe apparatus was secured to the nose and cheek to prevent dislodgment of the electrodes. The ph electrodes were connected to a portable digital data recorder (Digitrapper Mark III Gold, Synetics) that was worn around the patient s waist and stored ph data samples every 4 seconds for up to 24 hours. Patients left the office with instructions about recording symptoms, meal times, times of recumbency for sleep, and time of rising in the morning in a diary. Patients also were instructed to perform their normal daily activities. They were told to avoid antacids, coffee, tea, soft drinks, and sticky foods (gum, taffy, jelly), and could not take a bath or a shower during the monitoring period. Patients returned to the clinic after a minimum of 18 hours to have the probes removed and to collect the diaries. The data were downloaded and analyzed using the Gastrosoft computer program (Gastrosoft, Irving, TX). The percentage of time with a ph of less than 4 from both the proximal and distal electrodes was analyzed separately for both upright and supine positioning. An abnormal reflux value was defined as exceeding the 95th percentile for any of the previously determined upright or supine values for healthy volunteers (proximal total, 1.1%; proximal upright, 1.7%; proximal supine,.6%; distal total, 5.5%; distal upright, 8.3%; and distal supine, 3.0%). 10,11 Duodenogastroesophageal Reflux The Bilitec 2000 (Synectics), a fiberoptic sensor developed by Bechi et al, had been used extensively to assess duodenogastroesophageal reflux (DGER). 12 In this study, Bilitec 2000 monitoring was performed simultaneously with the 24-hour ambulatory ph monitoring while patients were on twice-daily PPI therapy before fundoplication. Before probe insertion, the patient was asked to sniff 3 sprays of anesthetic (lidocaine hydrochloride, 2%; phenylephrine,.5%) through 1 nostril. The system is composed of a fiberoptic probe that carries light signals into the esophagus and back to the optoelectronic system using a plastic fiberoptic bundle. The probe head is Teflon (DuPont, Wilmington, DE), and has a 2.0-mm open groove probe across which 2 wavelengths of light are emitted and material is sampled. Two light-emitting diodes (at 470 and 565 nm) represent the source for bilirubin measurement and the reference signal (ie, no bilirubin absorption), respectively. The photodiode system converts light into an electrical signal. The signals are amplified and then are processed by an integrated microcomputer that calculates the differences in absorbances at 470 and 565 nm. This value is directly proportional to the bilirubin concentration in the sample being studied. The sampling time is every 8 seconds, and the apparatus is in place for 24 hours. The probe is passed through the nostril and positioned 5 cm above the manometrically determined LES location. The patients were instructed to follow a diet that consisted only of white foods (ie, turkey, egg whites, white bread, white rice, milk, cottage cheese, vanilla pudding, vanilla ice cream, and so forth). They also

4 436 SWOGER ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 4, No. 4 were provided with 3 cans of Boost (Novartis Medical Health, Inc, Minneapolis, MN) to drink during the study period. For the Bilitec 2000 procedure, the data were downloaded to a computer and analyzed using the Gastrosoft program. Measurements were recorded as percent time of bilirubin absorbance of greater than.14% and were analyzed separately for total, upright, and supine periods. An abnormal DGER value was defined as exceeding the 95th percentile for any of the previously determined upright or supine values for healthy volunteers (total, 1.8%; distal upright, 2.2%; and distal supine, 1.6%). 12 Bravo ph Monitoring All subjects underwent BRAVO ph monitoring 12 months after fundoplication. Patients underwent upper endoscopy after an overnight fast to place the Bravo ph capsule (Medtronic Corp). Before endoscopy, the Bravo ph capsule was calibrated, with confirmation of transmission and reception confirmed. Patients had conscious sedation (1 5 mg of midazolam and mg of meperidine), and upper endoscopy was performed with patients in the left lateral decubitus position. The endoscopist identified the squamocolumnar junction, and the distance from the incisors to the squamocolumnar junction was measured. The endoscope was removed, and the patient was kept in the left lateral decubitus position. The Bravo delivery system was passed orally into the esophagus, and the capsule was attached to the esophageal mucosa as previously described, 13 6 cm proximal to the squamocolumnar junction. Patients were encouraged to perform their normal activities and were not placed under any dietary restrictions. Patients kept a diary documenting food intake, periods of sleep, and symptom occurrence. After 48 hours, patients returned to the Department of Gastroenterology to turn in their diaries and data receivers. Data from the Bravo system were uploaded to a computer using Datalink (Medtronic), which is compatible with Windows 95/98/2000/NT (Microsoft, Redmond, WA). An abnormal reflux value was defined as exceeding the 95th percentile for any of the previously determined upright or supine values for healthy volunteers (total, 5.5%; upright, 8.3%; supine, 3.0%). Surgical Intervention Figure 1. Study protocol. Each patient in the surgical group had their procedure performed by a single surgeon ( J.P.) at the Cleveland Clinic Foundation s Department of General Surgery. The procedure was performed on an elective basis, agreed on by the patient and the surgeon. A standard 360 laparoscopic Nissen fundoplication was performed in all 10 patients. None of the patients required conversion to an open procedure. Surgical technique used the standard 5-port laparoscopic abdominal set-up. The procedure consisted of abdominal insufflation with carbon dioxide gas, retraction of the left lobe of the liver, dissection and identification of the right and left diaphragmatic crura, and takedown of the short gastric vessels using a harmonic scalpel. After the greater curve of the stomach was passed through the retroesophageal window, the diaphragmatic crura were re-approximated using 2 3 interrupted nonabsorbable sutures. A bougie was passed into the stomach to create the fundoplication. All patients then received routine postoperative care. Statistical Analysis Overall descriptive statistics including means, medians, and SDs for continuous variables and frequencies for categoric variables were calculated. All data, unless otherwise noted, are presented as the median (25% and 75% intraquartile ranges). Pearson 2 values and Fisher exact tests were used as appropriate for categoric variables. Symptom improvement was recorded on a scale of 0% 100%. This score was classified further into 1 of 2 categories: more than 50% improvement or 50% or less improvement. A 1-sided Pearson 2 test was used to compare the proportion of patients who reported more than 50% improvement in their symptoms groups. The Wilcoxon rank-sum test was used to study the differences in the global assessment symptom score. A sample size of 10 in the surgical group and 15 in the medical group was determined to be adequate ( 80% power) to detect a larger than 40% difference between the groups. Because of the nonrandomized nature of the protocol, group (surgical vs medical) parameters were compared to assess homogeneity and to identify any potential selection bias. A significance level of.05 was used in the analyses. Calculations were performed in JMP 5.1 software (SAS Institute, Cary, NC). Results Demographics Of the 72 patients treated with aggressive PPIs in our ongoing cohort study, 25 (35%) were unresponsive

5 April 2006 FUNDOPLICATION IN LARYNGOPHARYNGEAL REFLUX 437 Table 1. Demographic Characteristics of the Study Population Characteristic Surgical group (N 10) Controls (N 15) P value Age, y (range) 54 (41 61) 52 (34 67).9 Sex Men 4 (40%) 3 (20%).31 Women 6 (60%) 12 (80%) Race White 10 (100%) 11 (73%).17 Black 4 (27%) Smoking Ever 5 (50%) 5 (33%).43 Current 0 1 (7%) Pack years, mean Current alcohol use 7 (70%) 3 (20%).03 to aggressive acid suppression, reporting less than 50% symptom improvement after 4 months of therapy. Ten of 25 (40%) patients agreed to enroll in this prospective surgical protocol (surgical group), and the remaining 15 patients (60%) declined surgery, continued to receive medical management, and constituted the control group (Figure 1). Baseline patient characteristics (Table 1) were similar between the 2 groups except for current alcohol use, which was more common in the surgically treated group. In addition, the distribution of the primary and secondary reported symptoms were similar between the 2 groups (Table 2). The most common primary symptom in the surgical group was sore/burning throat (44%), followed by hoarseness (33%). In the control group, hoarseness was the most commonly reported primary symptom (31%), followed by chronic cough (25%). The mean ( SE) presenting symptom severity (range, 0 4) was similar (P 1.0) between the groups: the surgical group severity was and the control group severity was Heartburn and regurgitation were present in 5 of 10 (50%) patients in the surgical group and in 7 of 15 (47%) patients in the control group (P.9). On PPI therapy both typical symptoms had improved although throat symptoms persisted. Manometry All patients had normal median (intraquartile) esophageal contraction amplitudes before surgery: surgical group 93 mm Hg ( mm Hg), controls 92 mm Hg ( mm Hg). At baseline, the median (intraquartile) LESP in the surgical group and controls were similar: 16 mm Hg (9 24 mm Hg) and 18 mm Hg (16 24 mm Hg), respectively. After fundoplication, the median (intraquartile) LESP in the surgical group increased to 20.2 mm Hg (18 26 mm Hg). At baseline, there were 4 of 10 (40%) abnormal manometry study results in the surgical group, compared with 4 of 15 (27%) abnormal study results in the control group (P.4). All abnormal manometric diagnoses in the surgical group were caused by hypotensive LESP. At 3 months after fundoplication 0 of 10 patients (0%) had abnormal manometry results. The only significant manometric findings in this group were incomplete relaxation of the LES, which was encountered commonly after fundoplication. Ambulatory ph/bilirubin Monitoring Baseline median (intraquartile) percent time of esophageal acid exposure (total, upright, or supine) was similar between the surgical group and the controls (Table 3). At baseline, 8 of 10 (80%) patients in the surgical group had an abnormal ph study result and the remaining 2 patients had evidence of reflux by barium esophagram showing reflux of barium from the stomach into the esophagus. In the control group, an abnormal baseline esophageal ph was found in 7 of 15 (50%) patients (P.2). Before undergoing fundoplication all patients had normalized their esophageal acid exposure on twice-daily PPIs (Figure 2). At 3 and 12 months after fundoplication, the degree of esophageal acid exposure was significantly (P.001) less than baseline values but similar (P 1.0) to those obtained while on twice-daily PPIs (Table 3 and Figure 2). Bilitec 2000 monitoring was completed in 5 of 10 patients in the surgical group on twice-daily PPIs before surgery to assess the presence of DGER. Four of 5 (80%) patients had normal esophageal exposure to DGER, whereas 1 (20%) patient had abnormal DGER (% time with bilirubin absorbance.14, 3.3% total, 6.5% upright, and 7.7% supine). Table 2. Prevalence of Primary and Secondary Symptoms Symptom Surgical group (N 10) Controls (N 15) Primary Secondary Primary Secondary Chronic cough 20% 10% 25.0% 6.3% Throat clearing 0% 10% 12.5% 31.3% Sore/burning throat 40% 30% 12.5% 25.0% Globus 0% 10% 12.5% 6.3% Hoarseness 30% 20% 31.3% 6.3% Dysphagia 0% 0% 6.3% 0% Other a 0% 0% 0% 18.8% None 0% 10% 0% 0% a Chest pain, shortness of breath, and regurgitation.

6 438 SWOGER ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 4, No. 4 Table 3. Median (Interquartile Range) % Time ph 4 Surgical group (N 10) %ph 4 Baseline 3 months after surgery 12 months after surgery Control group (N 15) at baseline P value a Proximal Total 2.3 ( ).0.4 (.1 1.7).19 Upright 2.6 ( ).0.7 (.2 2.8).32 Supine.0 (0 1.2).0.0 (0.3).4 Distal Total 8.6 ( ).0.3 (.0.7) 5.4 ( ).37 Upright 10.3 ( ).0.1 (.0.4) 5.3 (1.2 15).5 Supine 6.4 ( ) (0 8.4).27 a Comparison for baseline surgical group and the controls. Laryngoscopy Laryngoscopic findings at baseline and posttherapy are presented in Table 4. The most common finding for the surgical group was medial arytenoid wall erythema/edema, which was present in 60% of the patients, followed by interarytenoid erythema (50%) and arytenoid complex erythema/edema (50%). These findings were similar to those in the control group (Table 4). After fundoplication all patients showed improvement and/or resolution of their laryngeal signs during the 1-year follow-up evaluation. Five of 10 patients (50%), were noted to have resolution of their initial findings and had essentially normal laryngoscopic examinations at the 6-month postoperative follow-up evaluation. An additional 3 of 10 patients (30%) were noted to have normalized laryngoscopic examinations at the 12-month follow-up visit. Outcome At the 1-year follow-up evaluation, only 1 of 10 (10%) surgically treated patients and 1 of 15 (7%) controls had laryngeal symptom improvement (Figure 3). Improvement in laryngeal symptoms was achieved in an additional 2 surgically treated patients after beginning treatment for asthma or allergies at the completion of the 12-month study period. In controls, 9 of 15 (60%) patients reported symptom improvement after treating other causes: allergies (n 4), asthma (n 1), discontinuation of alcohol (n 1), allergies plus asthma plus the continued GERD therapy (n 2). Discussion In this prospective concurrent controlled study we found that surgical fundoplication does not produce significant symptom resolution in patients previously unresponsive to aggressive acid suppression. Only 1 of 10 (10%) patients achieved symptom improvement 12 months after surgery, a similar response to those who continued PPI therapy (1 of 15; 7%) (Figure 3). The poor response in the surgical group was despite a technically successful surgery, which resulted in increased LESP and normalized esophageal acid exposure (Figure 2). Thus, the results of our study question the contention that continued nonacid or intermittent acid reflux is the cause of continued chronic laryngeal symptoms while on PPI therapy. Esophageal testing by ph monitoring, esophagoscopy, or barium swallow have poor sensitivity in establishing GERD as the cause of patients laryngeal signs and symptoms. 2 In a recent study, Charbel et al 14 showed that the likelihood of abnormal ph monitoring is only 1% 4% in this group of patients if studied on therapy. Thus, in patients with chronic laryngeal symptoms despite aggressive acid suppression, the driving force for the suspicion of GERD continues to be the laryngoscopic findings. However, recent studies have questioned the specificity of laryngoscopy in accurately identifying Figure 2. Esophageal acid exposure at baseline, after twice-daily PPI therapy, and at 3 and 12 months after fundoplication. Eight of 10 (80%) patients had abnormal esophageal acid exposure by ph monitoring at baseline that was normalized on PPI therapy and postfundoplication despite persistent symptoms.

7 April 2006 FUNDOPLICATION IN LARYNGOPHARYNGEAL REFLUX 439 Table 4. Laryngoscopic Findings at Baseline and at 6 and 12 Months After Surgery Surgery group Baseline 6 months after fundoplication 12 months after fundoplication Controls Posterior cricoid wall erythema/edema 40% 20% 0% 64% Posterior pharyngeal wall erythema 20% 10% 0% 0% True/false vocal cord erythema/edema 40% 10% 0% 57% Arytenoid complex erythema/edema 50% 20% 10% 57% Medial arytenoid wall erythema/granularity 60% 30% 10% 79% Posterior pharyngeal cobblestoning 0% 0% 0% 14% Redundant tissue/pseudosulcus 20% 0% 0% 21% Interarytenoid bar erythema 50% 10% 0% 7% GERD as the cause of laryngeal irritation. 2,8,15 First, LPR diagnosis most often is a result of laryngeal findings of erythema and edema, 16,17 which are highly subjective and are nonspecific for GERD. Second, the intraobserver and interobserver variabilities of laryngoscopy are far from perfect. 17,18 Finally, laryngeal irritation and findings similar to those suspected with LPR can be seen even in healthy normal patients without GERD. 8,15 Thus, laryngoscopic findings are neither specific nor sufficient to diagnose GERD definitively. In fact, in this study we found 80% improvement in laryngeal signs despite poor symptomatic response. This finding further highlights the dissociation between laryngeal signs and patients symptoms. It also suggests a possible multifactorial cause for LPR, further supported by the fact that in most of our patients symptom improvement occurred only after adjunctive therapy for other conditions. Given the lack of specificity of any of the diagnostic tests in this patient group, empiric therapy with aggressive acid suppression with PPIs often is advocated as the initial test to determine if GERD is the cause of patients chronic symptoms. 2,19 Medical therapy in patients with suspected LPR results in an overall response rate of 50% 70%. 2 Because PPI therapy does not suppress all gastroesophageal reflux completely, the lack of response among the unresponsive patients is argued by some to be caused by either nonacid (bile) or intermittent reflux of gastric contents. 5 Surgical fundoplication is thus the only definitive means of assessing the role of these components in persistence of laryngeal signs and symptoms despite aggressive PPI therapy. Based on the result of our study it is unlikely that patients chronic laryngeal symptoms will improve after fundoplication given their previous lack of response to aggressive PPI therapy. Our study prospectively evaluated the role of fundoplication in patients with suspected LPR unresponsive to PPI therapy. Prior observational studies had suggested similar outcomes in patients with classic GERD or other atypical symptoms For example, So et al, 23 studying 150 consecutive patients after fundoplication, found that only 56% of patients had relief of atypical symptoms whereas typical heartburn symptoms resolved in 93%. Furthermore, the only preoperative predictors of relief of atypical symptoms were prior clinical response to pharmacologic acid suppression and abnormal hypopharyngeal ph results. More specifically, a recent uncontrolled case series 26 evaluating the outcome of patients with LPR reported 84% symptom improvement after fundoplication. However, this study again showed a poor outcome in those previously unresponsive to PPI therapy. Therefore, based on the results of our studies, continued symptoms on aggressive acid suppression are most likely an indication for investigating other allergy, sinus, or pulmonary causes for patients residual symptoms. Several aspects of our study design strengthen the conclusions of our findings. First, the study was prospective in design. We used multiple physiologic parameters to document the consequences of the surgical intervention including ph monitoring and esophageal manom- Figure 3. Symptom improvement after fundoplication compared with continued PPI therapy.

8 440 SWOGER ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 4, No. 4 etry. Wireless ph monitoring was used at 12 months in all patients to increase the sensitivity of GERD detection. Blinded laryngoscopic examinations documented changes in laryngeal signs up to 12 months after surgery. An important limitation of our study may be the small sample size. Our study was powered to detect a difference of greater than 40% between the 2 studied groups. This was based on prior surgical data suggesting symptomatic response rates ranging between 80% and 90% For example, Hunter et al 28 reported 80% 91% symptom improvement in patients with atypical GERD symptoms and So et al 23 reported an 80% symptomatic response to fundoplication in patients with chronic laryngitis. Placebo response rate of up to 40% 30 in this group would result in a minimum expected therapeutic gain of 40%. Thus, the estimate of more than 40% difference in our study was the most conservative, yielding a sample size of 10 in the surgical group and 15 in the medical group. In addition, we believe that a response rate of less than 40% for fundoplication would be inadequate to recommend as a treatment option for this group of patients. Nonetheless, should the true difference between the groups be less than 40%, then our study lacks adequate power to detect such a difference. An additional limitation of our study was the lack of randomization. However, given less than convincing historical evidence of success with surgical intervention, and patients reluctance, randomization often is difficult in studies such as ours. However, we used a concurrent control design ensuring similar demographic and physiologic parameters between cases and controls to minimize selection bias. In conclusion, our study suggests that Nissen fundoplication in patients with persistent chronic laryngeal signs and symptoms suspected of having LPR despite aggressive PPI therapy does not result in improved symptom outcome. The findings suggest that GERD alone may not be the cause of the symptoms in many of these patients and that the diagnosis of LPR should be seriously re-evaluated in these patients. Many such patients experienced symptom relief and resolution after investigation and treatment of other causes. LPR may be overdiagnosed as a result of overconfidence in laryngoscopy as a gold standard tool for GERD diagnosis. References 1. Gaynor EB. Laryngeal Complications of GERD. J Clin Gastroenterol 2000;30(Suppl):S31 S Vaezi MF, Hicks DM, Abelson TI, et al. Laryngeal signs and symptoms and gastroesophageal reflux disease (GERD): a critical assessment of cause and effect association. Clin Gastroenterol Hepatol 2003;1: DeVault KR. Overview of therapy for the extraesophageal manifestations of gastroesophageal reflux disease. Am J Gastroenterol 2000;95:S39 S Postma GN, Johnson LF, Koufman JA. Treatment of laryngopharyngeal reflux. Ear Nose Throat J 2002;81(Suppl 2): Amin MR, Postma GN, Johnson P, et al. Proton pump inhibitor resistance in the treatment of laryngopharyngeal reflux. Otolaryngol Head Neck Surg 2001;125: Klaus A, Swain JM, Hinder RA. Laparoscopic antireflux surgery for supraesophageal complications of gastroesophageal reflux disease. Am J Med 2001;111:202S 206S. 7. Park W, Hicks DM, Khandwala F, et al. Laryngopharyngeal reflux (LPR): prospective cohort study evaluating optimal dose of PPI therapy and pre-therapy predictors of response. Laryngoscope 2005;115: Hicks DM, Ours TM, Abelson TI, et al. The prevalence of hypopharynx findings associated with gastroesophageal reflux in normal volunteers. J Voice 2002;16: Richter JE, Wu WC, Johns DN, et al. Esophageal manometry in 95 healthy adult volunteers. Variability of pressures with age and frequency of abnormal contractions. Dig Dis Sci 1987;32: Richter JE, Bradley LA, DeMeester TR, et al. Normal 24-hr ambulatory esophageal ph values. Influence of study center, ph electrode, age, and gender. Dig Dis Sci 1992;37: Sinclair JW, Dalton CB. The performance of ambulatory esophageal ph monitoring in adults. In: Richter JE, ed. Ambulatory esophageal ph monitoring. New York: Igaku-shoin 1991: Vaezi MF, Richter JE. Role of acid and duodenogastric reflux in gastroesophageal reflux disease. Gastroenterology 1996;111: Pandolfino JE, Richter JE, Ours T, et al. Ambulatory esophageal ph monitoring using a wireless system. Am J Gastroenterol 2003;98: Charbel S, Khandwala F, Vaezi MF. The role of esophageal ph monitoring in symptomatic patients on PPI therapy. Am J Gastroenterol 2005;100: Milstein C, Charbel S, Hicks DM, et al. Prevalence of laryngeal irritation signs associated with reflux in asymptomatic volunteers: impact of endoscopic technique (rigid vs. flexible scope). Laryngoscope 2005;115: Ahmed T, Ableson TI, Hicks DM, et al. Chronic laryngitis associated with GERD: a large scale prospective assessment of differences in practice patterns between gastroenterologists and ENT physicians. Gastroenterology 2005;128:M Merati AL, Reider AA. Normal endoscopic anatomy of the pharynx and larynx. Am J Med 2003;115(Suppl 3A):10S 14S. 18. Branski RC, Bhattacharyya N, Shapiro J. The reliability of the assessment of endoscopic laryngeal findings associated with laryngopharyngeal reflux disease. Laryngoscope 2002;112: Koufman JA, Aviv JE, Casiano RR, et al. Laryngopharyngeal reflux: position statement of the committee on speech, voice, and swallowing disorders of the American Academy of Otolaryngology- Head and Neck Surgery. Otolaryngol Head Neck Surg 2002;127: Vela MF, Camacho-Lobato L, Srinivasan R, et al. Simultaneous intraesophageal impedance and ph measurement of acid and non-acid reflux: effect of omeprazole. Gastroenterology 2001; 120: Sifrim D, Dupont L, Blondeau K, et al. Weakly acidic reflux in patients with chronic unexplained cough during 24 hour pressure, ph, and impedance monitoring. Gut 2005;54: Sifrim D, Castell D, Dent J, et al. Gastroesophageal reflux monitoring: review and consensus report on detection and definition of acid, non-acid and gas reflux. Gut 2004;53: So JBY, Zeitels SM, Rattner DW. Outcomes of atypical symptoms attributed to gastroesophageal reflux treated by laparoscopic fundoplication. Surgery 1998;124: Campos GM, Peters JH, DeMeester TR, et al. Multivariate anal-

9 April 2006 FUNDOPLICATION IN LARYNGOPHARYNGEAL REFLUX 441 ysis of factors predicting outcome after laparoscopic Nissen fundoplication. J Gastrointest Surg 1999;3: Leeder PC, Watson DI, Jamieson GG. Laparoscopic fundoplication for patients with symptoms but no objective evidence of gastroesophageal reflux. Dis Esophagus 2002;15: Westcott CJ, Hopkins MB, Bach K, et al. Fundoplication for laryngopharyngeal reflux disease. J Am Coll Surg 2004;199: Oelschlager BK, Eubanks TR, Oleynikov D, et al. Symptomatic and physiologic outcomes after operative treatment for extraesophageal reflux. Surg Endosc 2002;16: Hunter JG, Trus TL, Branum GD, et al. A physiologic approach to laparoscopic fundoplication for gastroesophageal reflux disease. Ann Surg 1996;223: Johnson WE, Hagen JA, DeMeester TR, et al. Outcome of respiratory symptoms after antireflux surgery on patients with gastroesophageal reflux disease. Arch Surg 1996;131: Vaezi MF, Richter JE, Stasney CR, et al. A randomized double blind, placebo controlled study of acid suppression for the treatment of suspected laryngopharyngeal reflux. Gastroenterology 2004;126:A160. Address requests for reprints to: Michael F. Vaezi, MD, PhD, MS epi, Professor of Medicine, Clinical Director, Department of Gastroenterology and Hepatology, Director, Clinical Research and Center for Swallowing and Esophageal Disorders, Vanderbilt University Medical Center, 1501 TVC, Nashville, Tennessee Michael.vaezi@vanderbilt.edu; fax: (615)

10 April 2006 FUNDOPLICATION IN LARYNGOPHARYNGEAL REFLUX 441.e1 APPENDIX Larynx/Pharynx Findings Name: Clinic No: Date: Presence of the Following: Yes No Yes No 1. Posterior Pharyngeal Wall 6. True Vocal Folds a. Erythema a. Erythema b. Edema b. Edema c. Cobblestoning c. Surface Irregularity d. Mass Lesions (polyps) 2. Interarytenoid Bar 7. False Vocal Folds a. Present a. Erythema b. Erythema b. Edema c. Blanched c. Surface Irregularity d. Irregular 3. Posterior Commissure 8. Anterior Commissure a. Erythema a. Erythema b. Edema c. Surface Irregularity 4. Posterior Cricoid Wall 9. Epiglottis a. Erythema a. Erythema b. Edema (laryngeal surface) c. Redundant Tissue b. Erythema (petiole) 5. Arytenoid Complex 10. Aryepiglottic Folds a. Apex Erythema a. Erythema b. Apex Edema b. Edema c. Medial Wall Erythema d. Medial Wall Erosion e. Medial Wall Granularity