672

OBJECTIVE: To evaluate the diagnostic value of 3-site 24-hour ambulatory pH monitoring in patientswith posterior laryngitis (PL) and the prevalence ofesophageal abnormalities in this patient group.METHODS: Twenty patients with PL and 17 healthyvolunteers were studied as controls. Control subjectshad transnasal esophagogastroduodenoscopy(T-EGD) and ambulatory pH monitoring. Patientsunderwent T-EGD, ambulatory pH monitoring, andbarium esophagram.RESULTS: T-EGD documented no abnormality incontrols. Esophagitis was present in 2 PL patients,and hiatal hernia in 3. Ambulatory pH monitoringshowed that 15 PL patients and 2 controls exhibitedpharyngeal acid reflux. Barium esophagram doc-umented gastroesophageal reflux in 5 PL patients.However, none of these barium reflux eventsreached the pharynx. All PL patients with bariumesophagram evidence of gastroesophageal refluxalso showed pharyngeal acid reflux by pH moni-toring.CONCLUSION: Pharyngeal acid reflux is moreprevalent in patients with PL than in healthy con-trols. Patients with PL infrequently have esophagealsequelae of reflux disease. Ambulatory 24-hoursimultaneous 3-site pharyngoesophageal pH mon-itoring detects gastroesophagopharyngeal acidreflux events in most patients with PL. (OtolaryngolHead Neck Surg 1999;120:672-7.)

Posterior laryngitis (PL) until recently was not wellrecognized. Examiners frequently keyed on the truevocal cords during evaluation. With the advent of

improved flexible and rigid instruments to performdiagnostic laryngoscopy along with videostroboscopy,abnormalities of the posterior larynx have beenincreasingly recognized. These patients may reportchronic or intermittent hoarseness,1-4 voice fatigue andbreaks,3 frequent throat clearing,2,5,6 sore throat,2,3,7

excessive mucus8 or postnasal drip,8-10 cough,3,5 dysp-nea,11 or dysphagia.5 Laryngeal examination mayreveal erythema and edema of arytenoids,8 hyper-trophic mucosa of interarytenoid area,2,3 or pachyder-ma laryngis.1,2 Associated true vocal cord problemssuch as contact ulcer or granuloma,6 edema and ery-thema,2,3 vocal cord nodules,8 Reinke’s edema,8 leuko-plakia,8 and laryngotracheal stenosis11 may accompanythe PL. The typical symptoms of gastroesophagealreflux disease (GERD), such as heartburn, regurgita-tion, and water brash, occur in a minority of patientswith PL.5,10 Therefore the possible role of refluxedgastric acid in the pathogenesis of laryngeal disordersmay easily be overlooked.

Patients with suspected reflux-induced laryngeallesions have been evaluated for reflux disease by variousmodalities such as upper gastrointestinal endoscopy,barium esophagram, and ambulatory pH monitoring. Inthis study we report the combined findings of uppergastrointestinal endoscopy, barium esophagram, and24-hour pharyngoesophageal pH monitoring in a groupof patients with subjective and objective findings sug-gestive of reflux-induced laryngitis.

METHODS

Twenty consecutive patients (13 male, 7 female) with PL,aged 17 to 78 years (47 ± 4 yrs), and 17 healthy volunteers (5male, 12 female), aged 19 to 85 years (46 ± 6 yrs), were stud-ied. Studies were approved by the Human Research ReviewCommittee of the Medical College of Wisconsin, and partici-pants gave written, informed consent before their studies. AllPL patients and healthy volunteers filled out a detailed healthquestionnaire before their studies.

Healthy volunteers were recruited by advertisement anddid not have any esophageal or laryngeal symptoms. In addi-tion, they underwent unsedated transnasal pharyngoesopha-gogastroduodenoscopy (T-EGD) and did not exhibit anypathologic symptoms.

PL patients reported intermittent hoarseness, chronichoarseness, frequent throat clearing, sore throat, dyspnea, or

Pharyngeal pH monitoring in patients with posteriorlaryngitisSECKIN O. ULUALP, MD, ROBERT J. TOOHILL, MD, RAYMOND HOFFMANN, PhD, andREZA SHAKER, MD, Milwaukee, Wisconsin

From the Departments of Medicine (Division of Gastroenterology andHepatology) (Drs Ulualp and Shaker), Otolaryngology and HumanCommunication (Drs Ulualp, Toohill, and Shaker), and Biostati-stics (Dr Hoffmann), Medical College of Wisconsin.

Supported in part by NIH grant no. R01 DK25731.Presented at the Annual Meeting of the American Academy of

Otolaryngology–Head and Neck Surgery, San Francisco, CA,September 7-10, 1997.

Reprint requests: Reza Shaker, MD, Professor of Medicine, Divisionof Gastroenterology and Hepatology, Froedtert Memorial LutheranHospital, 9200 W Wisconsin Ave, Milwaukee, WI 53226.

Copyright © 1999 by the American Academy of Otolaryngology–Head and Neck Surgery Foundation, Inc.

0194-5998/99/$8.00 + 023/1/91774

Otolaryngology–Head and Neck SurgeryVolume 120 Number 5 ULUALP et al 673

chronic cough and exhibited videostroboscopic findings oferythema and edema of the arytenoids, erythema of interary-tenoid tissue and posterior third of the vocal cords, pachyder-ma laryngis, hyperplastic interarytenoid tissue, or erythema ofthe entire larynx. In addition to the above findings, videostro-boscopy also documented vocal cord nodules in 4 patients,vocal fold polyp in 1 patient, and laryngotracheal stenosis in5 patients (Table 1).

Fifteen of the 20 PL patients underwent T-EGD.12 T-EGDcould not be performed in the remaining 5 PL patientsbecause of severe nasal septal deviation (2 patients), technicaldifficulty in intubating the esophagus (1 patient), and vasova-gal reaction (2 patients). All 20 PL patients had a standard bar-ium esophagram findings.

All PL patients and healthy volunteers were studied foroccurrence of pharyngeal acid reflux by 3-site 24-hourpharyngoesophageal pH monitoring.13

Barium esophagram was done after an overnight fasting.Barium swallows were performed with a total of 200 to 250mL of barium and water mixture. Radiographs and videofluo-roscopic recordings were obtained. Spontaneous gastroe-sophageal reflux (GER) events throughout the examinationwere evaluated. If no GER event was observed, the provoca-tive maneuvers were performed while the PL patients were ina supine position. Provocative maneuvers involved exercisesthat increase the intra-abdominal pressure. For this, PLpatients were sequentially asked to perform the followingtasks: (1) coughing, (2) leg raising, and (3) the Valsalvamaneuver. GER was said to be present when barium wasobserved to fill the esophageal lumen at least 3 cm.14 Barium

esophagram reports included the presence or absence of GER,the extent of the barium in the esophagus, and the provocativemaneuver or maneuvers performed to observe GER.

T-EGD was done with an Olympus GIF-N30 endoscope(Olympus Corp, Lake Success, NY) with volunteers or PLpatients sitting upright in a chair. The more patent side of thenasal cavity was determined by nasal examination and wasanesthetized by passing a cotton-tipped swab covered withXylocaine gel. In cases of narrow passage caused by hyper-trophied turbinates, 2 puffs of a nasal decongestant wereapplied. In all cases the pharynx was anesthetized withCetacaine spray. Then the endoscope was passed through thenose to the nasopharynx and pharynx. The larynx wasobserved, and glottal closure during swallowing and phona-tion was evaluated. The scope was then introduced into theesophagus through the upper esophageal sphincter underdirect vision. The gross appearance of esophageal mucosa wasevaluated. Subsequently, direct examination of the stomachand retroflexed examination of the gastroesophageal junctionwere followed by the evaluation of the bulb and second por-tion of the duodenum.

The ambulatory pH studies were done with pH recordingsystems, with a single recording site and another with 2recording sites that were 10 cm apart (Synectics Medical Inc,Irving, TX). pH probes were placed at 3 manometricallydetermined sites: pharynx (2 cm above the upper esophagealsphincter), proximal esophagus (10 cm distal to the pharyn-geal probe, approximately 4 to 5 cm below the upper eso-phageal sphincter), and distal esophagus (5 cm above thelower esophageal sphincter). The probes were secured to the

Table 1. Diagnostic workup results

Patient no. Videostroboscopy Barium esophagram T-EGD esophageal findings Pharyngeal acid exposure

1 PL, VCN GER(–) No abnormality Positive2 PL GER(–) No abnormality Positive3 PL, LTS GER(–) No abnormality Positive4 PL, VCN GER(–) No abnormality Negative5 PL, LTS GER(–) No abnormality Positive6 PL, LTS GER(–) No abnormality Negative7 PL GER(–) No abnormality Negative8 PL GER(+), HH HH Positive9 PL AM No abnormality Positive

10 PL GER(–) No abnormality Negative11 PL AM No abnormality Negative12 PL, LTS GER(–) — Positive13 PL, VCN GER(+), HH HH, Esophagitis Positive14 PL GER(+), AM, HH HH Positive15 PL, LTS GER(+) No abnormality Positive16 PL AM Esophagitis Positive17 PL GER(+), HH — Positive18 PL, VFP GER(–) — Positive19 PL, VCN GER(–) — Positive20 PL GER(–) — Positive

VCN, Vocal cord nodules; LTS, laryngotracheal stenosis; HH, hiatal hernia; AM, abnormal motility; VFP,vocal fold polyp.

674 ULUALP et al

Otolaryngology–Head and Neck Surgery

May 1999

nose. Studies were initiated in the morning after the mano-metric studies and terminated 22 to 24 hours later. Subjects atea standard meal that included the following: (1) breakfast, atotal of 504 kcal (78.6% carbohydrate, 12.1% protein, 9.3%fat); (2) lunch, a total of 822 kcal (67.7% carbohydrate, 11.4%protein, 20.9% fat); and (3) dinner, a total of 1199 kcal (58.2%carbohydrate, 16.2% protein, 25.6% fat). Meals were provid-ed through the Medical College of Wisconsin General ClinicalResearch Center. All subjects kept a detailed diary indicatingthe time of oral intakes and time of going to bed. They alsoindicated position (upright or supine) and events such asheartburn, belching, cough, chest pain, and exercise. For all 3sites, a decrease in pH below 4, which was not related tobelching or to the time of eating or drinking, was consideredto be an acid reflux episode. To be considered a pharyngealreflux event, it had to be simultaneous or preceded by adecrease in pH of similar or larger magnitude in the proximaland distal esophageal sites. Earlier studies have shown that theproximal distribution of refluxate is associated with a declinein pH activity of refluxed material in the esophagus. Also,determination of the temporal relationship between the onsetof pH decline among recording sites differentiated pHdeclines induced by oral intake (in which pharyngeal declineprecedes distal esophageal pH drops) from true gastro-esophagopharyngeal acid reflux episodes (in which pharyn-geal pH decline occurs either simultaneously or after theesophageal pH decline). These strict criteria were applied toavoid counting in spurious readings induced by pharyngealpH probe movement, loss of complete mucosal contact, pHchange caused by aerodigestive tract residue and secretions,and pH change caused by oral intake.

During the study, signals from pH electrodes were storedby the portable data logger, and on completion of each study,they were transferred to a computer for subsequent printingand analysis. In addition, pH recordings were displayed on thescreen, and a computer program was used to create a smallertime scale for determination of the temporal relationshipamong pH declines registered at various sites. The 2 pH elec-trodes were calibrated in buffers of pH 1 and pH 7, before andat the end of each study, and showed no significant drift in thepH signal during the study. Using these techniques, we deter-mined in the pharyngeal site the number of pH declines below4, percent of study time that the pH was below 4, and averageacid clearance time of the acid reflux episodes. Percent acidexposure time was calculated as the percentage of the studyperiod that the pH sensor was exposed to acid. Average acidclearance time was derived by dividing the total acid exposuretime in minutes by the number of reflux episodes. We alsodetermined the presence or absence of hiatal hernia, esophagi-tis, esophageal dysmotility, reflux of barium, and clearanceduring esophagography. Statistical comparison betweengroups for acid reflux event exposure time was performed

Fig 1. Comparison of the number of pharyngeal acidreflux episodes, acid exposure, and acid clearance timebetween PL patients and healthy controls. Data for eachsubject are depicted. As seen, PL patients as a group hadsignificantly more reflux events than controls. In addition,the prevalence of pharyngeal acid reflux events was sig-nificantly higher in the patient group than in controls (*P< 0.01).

Otolaryngology–Head and Neck SurgeryVolume 120 Number 5 ULUALP et al 675

with the nonparametric test (Wilcoxon rank sum test) and forprevalence was performed by the χ2 test.

RESULTSFindings of Barium Esophagram

Esophageal barium studies showed GER in 5 of 20patients with PL. Barium refluxate did not reach thepharynx in any of these PL patients. In 3 of these 5 PLpatients the reflux events occurred spontaneously andreached the thoracic inlet and aortic arch. In the other 2PL patients they occurred with provocative maneuvers;4 of these patients had hiatal hernias. Barium studiesdid not exhibit any detectable structural abnormalities.

Four of 20 PL patients exhibited abnormal eso-phageal motility. These abnormalities included break-ing of the primary wave in the mid to upper esophagus,inadequate clearance of the distal esophagus by sec-ondary peristalsis, and pooling of the barium in the dis-tal esophagus. Among these PL patients, 3 had pharyn-geal reflux episodes during pH monitoring. All PLpatients who exhibited GER on esophageal bariumstudies were subsequently found to have pharyngealacid reflux events during 24-hour pH monitoring (Table1).

Findings of T-EGD

T-EGD in the healthy control group showed normallaryngeal, esophageal, gastric, and duodenal mucosa.Glottal closure during swallowing and phonation wasnormal in all controls. There was no endoscopic evi-dence of hiatal hernia, esophagitis, neoplasm, or stric-ture formation among this group. In the PL patientgroup, T-EGD could not be done in 5 patients, asdescribed in the Methods section. Among the remaining15 PL patients, glottal closure during swallowing andphonation was normal. Macroscopic esophagitis wasobserved in 2 PL patients, and subsequent pH monitor-ing of these PL patients documented the occurrence ofpharyngeal acid reflux. Hiatal hernia was observed in 3of the patients with PL, all of whom demonstrated pha-ryngeal acid reflux during pH monitoring (Table 1).Barium esophagram also documented hiatal hernia inthese PL patients.

Findings of 24-hour pH Monitoring

Pharyngeal acid reflux episodes occurred in 15 of 20patients with PL (Fig 1) and 2 of the 17 healthy controls(1 and 2 episodes, respectively). The number of pharyn-geal reflux episodes among PL patients ranged from 0to 12. In total, 53 pharyngeal acid reflux events wererecorded in PL patients. None of these episodes wasassociated with belching. Except for 1 PL patient, all

pharyngeal acid reflux events in both patients andhealthy controls occurred in the upright position. Noneof the pharyngeal acid reflux events was associated withcough. Four patients with PL reported heartburn (1, 1,2, and 4 episodes, respectively) during the 24-hourstudy period. However, none of these heartburn eventswas associated with a distal esophageal reflux event.The number of pharyngeal acid reflux events in the PLpatient group was significantly more than in controls (P< 0.0005). Similarly, the percent acid exposure time andthe average acid clearance time were significantlygreater in PL patients than in healthy controls (P <0.005,P = 0.001, respectively). The prevalence of pha-ryngeal acid reflux was significantly higher in the PLpatients than in controls (P < 0.05).

DISCUSSION

The association between GER and laryngeal disor-ders was first reported in patients with contact ulcer ofthe larynx by Cherry and Margulies6 in 1968. Sincethen GER has been implicated in the pathogenesis of alarge number of aerodigestive tract disorders. However,the cause-and-effect relationship between the majorityof these disorders and gastric refluxate has not been sys-tematically studied. In practice, to determine the role ofGER in the pathogenesis of these disorders, patients areevaluated by various modalities such as esophagealendoscopy, barium esophagram, and 24-hour pH moni-toring. Except for pharyngeal pH monitoring, eso-phageal occurrence of reflux or its sequela is evaluatedby these techniques, and findings are extrapolated toassess the role of reflux in the pathogenesis of suprae-sophageal lesions.

In this study we report the combined findings of bar-ium esophagram, 24-hour pH monitoring, and endo-scopic evaluation of the esophagus in a group ofpatients with objectively documented PL. We also com-pared the pharyngoesophageal distribution of refluxedgastric acid between these patients and healthy controls.

GER events have been reported to occur sponta-neously or may be provoked during standard bariumesophagography. However, the sensitivity and specifici-ty of this technique for documentation of GERD havebeen reported to range from 20% to 70%14-16 and 74%to 94%,14,16 respectively. Gastroesophagopharyngealreflux of acid barium in patients with PL has beenreported in some studies.1,6 In our study, bariumesophagram revealed GER in 25% of patients, but pha-ryngeal reflux of barium was not observed in anypatients.

Although previous studies using barium have report-ed a 61%5 to 80%2 incidence of hiatal hernia in patients

676 ULUALP et al

Otolaryngology–Head and Neck Surgery

May 1999

with laryngitis, hiatal hernia was documented in only20% of PL patients in our investigation.

Previous studies of patients with laryngeal symp-toms with and without objective findings have reporteda history of recent esophagitis detected by endoscopyin 50% to 67% of cases.17,18 However, our studydemonstrated endoscopic esophagitis in only 10% ofthese PL patients at the time of investigation. Recentuse of acid-suppressive therapy may account for thisdifference.

Ambulatory pH monitoring has been used to docu-ment the role of acid reflux in the pathogenesis of pos-terior acid laryngitis.5,13,17-19Various techniques havebeen used for this purpose. They have recorded fromeither a single site within the esophagus4,18,20or fromthe pharynx and proximal or distal esophagus.5,17,19

These studies reported different degrees of GER. In arecent study using concurrent recording from the phar-ynx, proximal, and distal esophagus, gastroesophago-pharyngeal distribution of refluxed gastric acid inpatients with PL was found to be significantly differentfrom that in patients with GERD and healthy controls.13

Findings of our study are in agreement with this obser-vation.

Although esophageal barium studies are frequentlyperformed, their value in determining patients with PLhas not been systematically evaluated. The findings ofthis study—that only 25% of PL patients demonstrat-ed esophageal reflux events during barium studies,whereas 75% of them exhibited pharyngeal acid refluxdocumented by pH monitoring—suggest the generallyaccepted notion that the role of the barium esopha-gram for documenting GER in patients with suprae-sophageal complications of reflux disease is quite lim-ited.

Although the mechanism of the esophagopharyngealreflux was not studied in this report, previous studieshave shown that in patients with GERD and healthycontrols, esophagopharyngeal acid reflux events occurmost commonly during belching. Esophagopharyngealreflux events may also occur during the transient lowupper esophageal sphincter resting pressure or at theearly stages of swallowing if the esophagus containsgastric acid.21 Similar to prior studies, pharyngealreflux events, documented in our investigation, alsooccurred overwhelmingly in the upright position, butwere not related to belching.

Although in this study the frequency of pharyngealacid reflux was measured, the injurious effect of thegastric refluxate in addition to hydrochloric acid isdependent on its various components, including pepsin,pancreatic enzyme, bile acids, and byproducts of diges-

tion. The role of pepsin, a primary component of gastricsecretion, in inducing esophageal and supraesophageallesions has been reported previously.22,23 However, anacidic pH is required for proteolytic activities of pepsin.Because recording of the reflux of other components ofgastric refluxate is not widely available at this time,documentation of acid reflux is used as a marker orindicator of reflux of gastric content

In conclusion, whereas pharyngeal acid reflux eventsare more prevalent among patients with PL, they arerare among healthy controls. Esophageal complicationsof reflux are rare among patients with posterior acidlaryngitis. Reflux events detected during barium esoph-agography poorly correlate with the existence of PL andpharyngeal acid reflux. Among the 3 modalities used inthis study, pharyngeal acid reflux events recorded by pHmonitoring are more frequently detected in patientswith PL than esophageal sequela of reflux diseasedetected by endoscopy or occurrence of reflux duringesophageal barium studies.

REFERENCES

1. Delahunty JE. Acid laryngitis. J Laryngol Otol 1972;86:335-42.2. Ward PH, Berci G. Observations on the pathogenesis of chronic

nonspecific pharyngitis and laryngitis. Laryngoscope 1982;92:1377-82.

3. Kambic V, Radsel Z. Acid posterior laryngitis: aetiology, histol-ogy, diagnosis and treatment. J Laryngol Otol 1984;98:1237-40.

4. Wilson JA, White A, Von Haacke NP, et al. Gastroesophagealreflux and posterior laryngitis. Ann Otol Rhinol Laryngol 1989;98:405-10.

5. Koufman JA. The otolaryngologic manifestations of gastroe-sophageal reflux disease (GERD): a clinical investigation of 225patients using ambulatory 24 hour pH monitoring and an experi-mental investigation of the role of acid and pepsin in the devel-opment of laryngeal injury. Laryngoscope 1991;101(Suppl 53):1-78.

6. Cherry J, Margulies SI. Contact ulcer of the larynx. Laryngo-scope 1968;78:1937-40.

7. Deveney CW, Benner K, Cohen J. Gastroesophageal reflux andlaryngeal disease. Arch Surg 1993;128:1021-5.

8. Koufman JA. Gastroesophageal reflux and voice disorders. In:Rubin JS, et al, editors. Diagnosis and treatment of voice disor-ders. New York: Igaku-Shoin; 1995. p. 161-75.

9. Hanson DG, Kamel PL, Kahrilas PJ. Outcomes of antirefluxtherapy for the treatment of chronic laryngitis. Ann Otol RhinolLaryngol 1995;104:550-5.

10. Toohill RJ, Mushtag E, Grossman TW, et al. Pharyngeal, laryn-geal, and tracheobronchial manifestations of gastroesophagealreflux. Proceedings of the XXIV World Congress of Otolaryn-gology–Head and Neck Surgery. Berkeley (CA): Kugler andGhedini; 1990. p. 3005-9.

11. Jindal JR, Milbrath MM, Shaker R, et al. Idiopathic subglotticstenosis and gastroesophageal reflux. Ann Otol Rhinol Laryngol1994;103:186-91.

12. Shaker R. Unsedated trans-nasal pharyngoesophagogastroduo-denoscopy (T-EGD): technique. Gastrointest Endosc 1994;40;3:346-7.

13. Shaker R, Milbrath M, Ren J, et al. Esophagopharyngeal distrib-ution of refluxed gastric acid in patients with reflux laryngitis.Gastroenterology 1995;109:1575-82.

14. Thompson JK, Koehler RE, Richter JE. Detection of gastroe-

Otolaryngology–Head and Neck SurgeryVolume 120 Number 5 ULUALP et al 677

sophageal reflux: value of barium studies compared with 24 hrpH monitoring. AJR Am J Roentgenol 1994;192:621-6.

15. Ott DJ, Gelfand DW, Wu WC. Reflux esophagitis: radiographicand endoscopic correlation. Radiology 1979;130:583-8.

16. Sellor RJ, DeCaestecker JS, Heading RC. Barium radiology: asensitive test for gastroesophageal reflux. Clin Radiol 1987;38:303-7.

17. Jacob P, Kahrilas PJ, Herzon G. Proximal esophageal pH-metryin patients with reflux laryngitis. Gastroenterology 1991;100:305-10.

18. McNally PR, Maydonovitch CL, Prosek RA, et al. Evaluation ofgastroesophageal reflux as a cause of idiopathic hoarseness. DigDis Sci 1989;34:1900-4.

19. Wiener GJ, Koufman JA, Wu WC, et al. Chronic hoarseness sec-

ondary to gastroesophageal reflux disease: documentation with 24h ambulatory pH monitoring. Am J Gastroenterol 1989;84:1503-8.

20. Metz D, Childs ML, Ruiz C, et al. Pilot study of the omeprazoletest for reflux laryngitis. Otolaryngology Head Neck Surg 1997;116;1:41-6.

21. Shaker R, Dodds WJ, Hogan WJ, et al. Mechanisms of esopha-go-pharyngeal acid regurgitation [abstract]. Gastroenterology1991;100:A494.

22. Goldberg HI, Dodds WJ, Montgomery C, et al. Controlled pro-duction of acute esophagitis. Experimental animal model. Am JRoentgenol Rad Ther Nucl Med 1970;110:288-94.

23. Dodds WJ, Goldberg HI, Montgomery C, et al. Sequential gross,microscopic and roentgenographic features of acute felineesophagitis. Invest Radiol 1970;5:209-19.

Management of the Tinnitus Patient

This seventh annual conference, sponsored by the University of IowaDepartments of Otolaryngology–Head and Neck Surgery and Speech Pathology andAudiology, will be held September 30–October 2, 1999, at the University of IowaHospitals and Clinics, Iowa City, IA.

For further information, contact the Center for Conferences and Institutes,University of Iowa, 249 Iowa Memorial Union, Iowa City, IA 52242-1317; phone,900-551-9029; fax, 319-335-3533.