Proximal esophageal pH-metry in patients with ‘reflux laryngitis’

Proximal esophageal pH-metry in patients with ‘reflux laryngitis’

GASTROENTEROLOGY ALIMENTARY 1991;100:30!%310 TRACT Proximal Esophageal pH-Metry in Patients with ‘Reflux Laryngitis’ P. JACOB, P. J. KAHRILAS, and...

796KB Sizes 1 Downloads 96 Views

GASTROENTEROLOGY

ALIMENTARY

1991;100:30!%310

TRACT

Proximal Esophageal pH-Metry in Patients with ‘Reflux Laryngitis’ P. JACOB, P. J. KAHRILAS, and G. HERZON Departments of Medicine Veterans Administration

and Otolaryngology, Northwestern Lakeside Medical Center, Chicago,

Fiberoptic laryngoscopic examinations were performed on 40 patients with gastroesopbageal reflux disease, 25 of whom had persistent laryngeal symptoms (dysphonia, cough, globus sensation, frequent throat clearing, or sore throat) and 15 without laryngeal symptoms who served as disease controls. Ten patients with laryngeal symptoms but none of the controls had laryngoscopic findings consistent with reflux laryngitis. Dual-site ambulatory pH recordings were obtained with the pH electrodes spaced 15 cm apart and with the proximal sensor positioned just distal to the upper esophageal sphincter. Patients in the three groups (disease controls: group 1; patients with symptoms but without laryngoscopic findings: group 2; and patients with both laryngeal symptoms and findings: group 3) were comparable in terms of age, smoking habit, the presence of esophagitis, and distal esophageal acid exposure. Proximal esophageal acid exposure was, however, significantly increased in groups 2 and 3, and nocturnal proximal esophageal acidification occurred in over half of these patients but in none of the group 1 patients. We conclude that the subset of reflux patients who experience laryngeal symptoms show significantly more proximal esophageal acid exposure (especially nocturnally) and often have laryngoscopic findings of posterior laryngitis not observed in control reflux patients.

0

tolaryngologists often see patients with persistent symptoms of cough, sore throat, globus sensation, throat clearing, or hoarseness who have laryngoscopic findings that are either normal or within the spectrum of “acid posterior laryngitis” (1,2). This spectrum includes contact ulcer, interarytenoid erythema, plaques, and nodularity. Contact ulcer is a unilateral or bilateral ulceration over the vocal processes of the arytenoid cartilages with or without granulations. Virchow attributed this lesion to vocal abuse and named it “pachydermia verrucosa laryngis,” a term now used interchangeably with “contact ulcer,”

University Illinois

Medical

School

and

“granuloma,” or “contact ulcer granuloma” (3). Jackson and Jackson identified exposed cartilage at the base of some contact ulcers and elaborated on the “hammer and anvil” theory of pathogenesis whereby the vocal processes clash forcefully with resultant injury and granuloma formation at the point of impact (4). Although this mechanical scenario explains the anatomic selectivity of the lesion, it does not explain the development or the persistence of the lesion in an otherwise healthy person. Cherry and Margulies implicated irritation from refluxed gastric acid as a predisposing factor (5), and numerous clinical series have ensued describing patients recovering from chronic laryngitis following therapy for reflux disease (6-10). The most extensive series to date describes laryngoscopic and histologic findings in 44 cases of “posterior chronic acid laryngitis” (11). Laryngoscopic lesions were limited to the posterior third of the vocal cords and interarytenoid area ranging from reddish swollen mucosa to granulomas of varied size with or without central ulceration. Histologically, the laryngeal epithelium showed hyperplasia of the prickle cell or basal cell layer with abundant lymphocytic and plasma-cell infiltration. Despite the association between posterior laryngitis and reflux disease, overt esophagitis is paradoxically absent in most such individuals. Investigators have used a variety of tests to show that many patients with cervical symptoms have evidence of reflux disease (12-14), but little information exists on what discriminates reflux patients with cervical symptoms from those without. Similarly, little information exists on the specificity of contact ulcer for reflux injury. We thus undertook a comparative study of gastroesophageai reflux disease (GERD) patients with and without laryngeal symptoms to test the specificity of posterior Abbreviations used in this paper: GERLI, gastroesophageal reflux disease: LJES,upper esophageal sphincter. o 1991 by the American Gastroenterological Association 00X6-5065/91/$3.00

306

GASTROENTEROLOGYVol. lOO,No.2

JACOB ETAL.

laryngitis

for symptomatic

patients

and

to look

for

unique features of reflux disease that may account for cervical

morning.

Acid-suppressing

medications

were held for at

least 24 hours before the test.

symptoms. Data Analysis

Materials and Methods Patients Beginning in July 1987, 37 consecutive patients referred to one of the authors (P.J.K.) for symptoms suggestive of “reflux laryngitis” were evaluated. All patients answered a symptom questionnaire regarding the presence and duration of laryngeal symptoms (dysphonia, cough, frequent clearing of the throat, globus sensation, or persistent sore throat) and of gastroesophageal reflux disease (heartburn, dysphagia, or water brash). During the same period 21 nonconsecutive GERD patients without laryngeal symptoms were similarly evaluated and then formed the disease control population.

Patient Evaluation All patients were evaluated by esophagoscopy, esophageal manometry with a Bernstein test, fiberoptic laryngoscopy, and dual-site Z&hour ambulatory esophageal pH-metry. Endoscopic esophagitis was defined as the presence of ulcers, erosions, erythema with exudate, or a biopsy showing increased basal cell layer thickness and vascular papillae extending >30% of the distance to the surface. Esophageal manometry was performed before esophageal pH-metry in all instances, and the positions of both the upper esophageal sphincter (DES) and lower esophageal sphincter (LES) relative to the nares were determined. Mean LES pressure was determined from a 5-minute sleeve recording, and peristaltic function was determined during 10 5-mL water swallows. Peristaltic dysfunction was defined as the occurrence of failed or grossly hypotensive peristalsis (50% test swallows as previously described (15). The Bernstein test was performed by infusing O.lN HCl through the midesophageal port of the manometric assembly at the termination of the esophageal manometry test. Fiberoptic laryngoscopic examinations were performed by a single observer (G.H.) blinded to the results of all other examinations. For the purposes of this study, GERD was defined as either a positive Bernstein test result or recently demonstrated esophagitis as defined above. Ambulatory pH monitoring was performed with a Synectics system (model 93-6200 dual channel Digitrapper; Synectic Medical, Irving, TX) and a dual-site pH probe (5F OD) with antimony crystal sensors spaced 15 cm apart (model 91-0021; Synectic Medical). The proximal sensor was positioned l-3 cm distal to the URS, with the distal sensor assuming the obligatory position 15 cm distal (3-6 cm above the LES). The variability of electrode position was necessary to accommodate esophagi of variable lengths. During the study, patients consumed a normal diet, were taking no medications, and maintained a diary card detailing episodes of heartburn, regurgitation, etc., as well as the times that they went to bed at night and arose in the

Data from the proximal and distal pH sensors were analyzed independently. Reflux events were defined as periods during which the esophageal pH decreased to 14 for 2 5 seconds. Data obtained during supine (nocturnal) periods were evaluated separately. Esophageal pH data were summarized by the group median value and the range of values. Confidence intervals for pH-metry data were determined using Kolmogorov-Smirnov nonparametric analysis. Comparisons among groups were made using the x2 or the Mann-Whitney U test as appropriate. Student’s t test or a x2 analysis was used for comparison of clinical data.

Results Patient Groups

Seventeen of the 58 patients evaluated during the study period were excluded from data analysis because (a) the clinical evaluation revealed no evidence of GERD (n = 8), (b) technical difficulty occurred with the pH monitoring equipment (n = 3), or (c) the patient was unable to tolerate ambulatory pH-metry for a full 24 hours (n = 6). One patient was excluded after laryngoscopy revealed vocal cord polyps determined to be unrelated to reflux disease. After the exclusions, there were 25 GERD patients with laryngeal symptoms in the study population and 15 GERD patients without laryngeal symptoms in the disease control population. Each of the 25 GERD patients in the study population reported the presence of two or more of the following symptoms for 2 4 months: hoarseness (68%), cough (32%), sore throat (20%), frequent clearing of the throat (26%), and globus sensation (56%). Ten of the patients with persistent laryngeal symptoms had abnormal findings from laryngoscopic examinations as follows: interarytenoid erythema (n = S), erythema with nodularity (n = 5), nodularity with contact ulceration (n = l), and leukoplakia (n = 1). None of the disease controls had abnormal results of laryngoscopic examinations. The laryngeal symptom profile was similar among patients with and without laryngeal findings. In subsequent analysis, the patients were divided into three groups: control GERD patients with neither laryngeal symptoms nor laryngoscopic findings (group I), patients with laryngeal symptoms but without laryngoscopic findings (group 2), and patients with both laryngeal symptoms and laryngeal findings (group 3). Table 1 summarizes the clinical and manometric data among the groups. Patients with and without laryngeal symptoms were comparable in terms of age, smoking habit, prevalence of active esophagitis, esophageal length, LES pressure, and peristaltic function.

February 1991

REFLUX LARYNGITIS

Table I. Clinical Data of Gastroesophageal Groups Group 1

RefluxPatient

Group 2

Group 3

Number of patients studied 15 15 10 Age @r; mean * SD) 49 + 14 54 + 18 57 t 17 Current smokers” 13% 10% 13% Recent esophagitis 70% 67% 50% Positive results of Bernstein test 67% 73% 60% LES location (cm] 44 + 2 43 -c 2 45 + 3 LESP (mm Hg; mean f SD) 9.3 ? 8.7 7.8 k 8.0 9.4 k 6.6 Peristaltic dysfunction 2/15 (13%) 3/15 (20%) l/10 (10%) LESP, LES pressure.

Ambulatory pH-Metry Distal esophageal acid exposure was significantly longer than proximal esophageal acid exposure in all patient groups (P < 0.01, paired t test), and on a case-by-case basis every patient showed this pattern. Most commonly, reflux events resulted in distal esophageal acidification without affecting the proximal esophageal pH. Occasionally, a reflux event acidified both the proximal and distal esophagus, as illustrated in the sample recording from a patient with laryngeal nodularity and erythema in Figure 1. None of the patients noted regurgitation or vomiting during the study. The majority of esophageal acid exposure occurred during the day. Figure 2 illustrates the pH data from the distal pH electrode for the three patient groups both in terms of total esophageal acid exposure time and nocturnal acid exposure time during which the subjects were

307

supine. Typical of such data, the values within each patient group are not normally distributed but instead show a skewed distribution. There were no significant differences between the three patient groups either in the total time that the distal esophageal pH was less than 4 or in the duration of nocturnal acid exposure. Data from the proximal esophageal pH electrode are shown in Figure 3. The median values of proximal esophageal acid exposure of both the group 2 patients and the group 3 patients were significantly greater than those of the group 1 patients (P < 0.05). Four of 15 group 2 patients and 5 of the 10 group 3 patients had proximal esophageal acid exposure values greater than the 95% confidence interval for group 1.When examined for nocturnal proximal esophageal acid exposure, none of the group 1 patients had any such events, whereas 6 of 15 patients in group 2 and 7 of 10 patients in group 3 did. The group median value was significantly higher only for the group 3 patients.

Discussion The development of acid reflux injury to the larynx must involve gastroesophageal reflux events forceful enough to reach and traverse the UES. Findings from this study show that the former does occur and that, especially nocturnally, this occurrence is relatively specific for GERD patients with chronic laryngeal symptoms. The second condition required for direct acid irritation of the larynx is that the gastric refluxate traverse the LJES. The present study offers no Supine Period

PH 7 6 5 4

) and distal (-) esophagus in a patient with the symptom of chronic hoarseness Figure 1. Sample pH recording from the proximal (and laryngoscopic findings of erythema and nodularity. The proximal sensor was positioned z cm below the UES and the distal sensor 5 cm above the LES in this patient. The supine (nocturnal) period of the recording is indicated by the arrow above the tracings. Note that there were several reflux events detected by the distal sensor during the upright recording, but that none of these affected the proximal esophageal pH. During the supine recording, one of the many reflux events resulted in proximal as well as distal esophageal acidification.

308

JACOB ET AL.

GASTROENTEROLOGY

Vol. 100, No. 2

750 -

Total Time (minutes) 0 0

600 -

0

Supine Time (minutes)

Group I 0BO(

.

Group II

.

Group III

Figure 2. Distal esophageal acid exposure data during (A) the entire 24-hour study and (B) the supine (nocturnal) period. The shaded area in each panel depicts the 95% confidence interval for the control subjects with reflux disease but without laryngeal symptoms or laryngoscopic findings (group 1). Horizontal dotted lines represent the median values for each group. There were no significant differences among the groups in either the total distal esophageal acid exposure time or supine distal esophageal acid exposure.

Total Time (minutes)

80 -

60

Supine Time (minutes)

60 1

0

0

40 -

,i~~~_~~__~~__--;

(ii;,

Group I

Group II

Group III

0B

Group I

Group II

Group Ill

Figure 3. Proximal esophageal acid exposure data during [A) the entire 24-hour study and (B) the supine (nocturnal) period. The shaded area depicts the 95% confidence interval for the control GERD subjects without laryngeal symptoms or laryngoscopic findings (group 1). Horizontal dotted lines represent the median values for each group. Median proximal esophageal acid exposure was significantly longer (P < 0.05) in group 2 and group 3 patients. None of the group 1 patients showed any supine proximal esophageal acid reflux, but 6 of the group 2 patients and 7 of the group 3 patients did. The median value was significantly higher only in the group 3 patients (P < 0.01).

February

1991

proof of this event. However, recent investigations into physiologic control mechanisms of the UES suggest possible mechanisms by which this might occur. Although distention of the proximal esophagus is known to prompt UES contraction, abrupt distension of a long length of esophagus as occurs during gas reflux, or forceful acid reflux, causes an abrupt complete relaxation of the sphincter (16,17). Because this event occurs independently of swallowing, the larynx is not elevated and the laryngeal inlet is open, leaving the interarytenoid area in direct continuity with the refluxate. Furthermore, resting UES pressure is greatly reduced during periods of sleep or deep relaxation (18,19) and shows no reflex contraction in response to spontaneous gastroesophageal acid reflux (18,20). Again, these observations suggest that the larynx is vulnerable to the forceful nocturnal reflux observed in our patients. Several problems are apparent when ambulatory pH-metry is used to evaluate chronic laryngitis. These include both the lack of relevant normative data and technical limitations. Quantitative analyses of acid reflux have been developed with reference to the distal esophagus, which is continuously bathed by swallowed saliva containing bicarbonate and, hence, is capable of combatting acidification (21). On the other hand, as far as we know, the larynx is without an efficient defense mechanism against acid injury, and experimental data suggest that a single brief exposure of the true vocal cords to acid daily, or even on alternate days, results in the development of a contact ulcer (22). Similarly, subglottic stenosis can be experimentally induced by exposing a tracheal mucosal injury to gastric acid for 1 minute per day in as little as 7 days (23). Thus, it is not surprising that the quantitative aspects of esophageal acid exposure time as usually determined in the distal esophagus have little bearing on laryngeal symptoms or pathology. Another problem with the method of pH recording used in the esophagus is that the technique is not readily applicable to the hypopharynx or larynx. The physical requirements for intraluminal pH recording dictate that a relatively constant impedance be maintained between the intraluminal probe and the reference skin electrode. The environment of the hypopharynx is such that the electrode would be intermittently dry and moist, vastly altering the impedance of the circuit and probably resulting in the “pseudoreflux” described in a recent investigation that attempted hypopharyngeal recordings (13). It was for this reason that we positioned the proximal electrode below rather than above the UES. Gastroesophageal reflux disease is notoriously difficult to define because of the extreme heterogeneity of afflicted patients. Peptic esophagitis is a finding spe-

REFLUX LARYNGITIS

309

cific for reflux disease, but because esophagitis is often a transient condition and is often absent in obviously symptomatic patients, it lacks sensitivity. Similarly, despite the seeming objectivity of &t-hour ambulatory esophageal pH monitoring, this test is subject to the same limitations as older tests in that substantial overlap exists between values obtained from control populations and patients with reflux Furthermore, what little data there disease (24-26). are on the subject suggest that substantial day-to-day variability occurs in ambulatory pH data for a given subject (26). In view of these limitations, we defined our study population of GERD patients by either the presence of endoscopic esophagitis or supportive symptoms along with positive results of a Bernstein test. This definition afforded us a relatively homogeneous population for study. However, we realize that defining GERD according to esophageal symptoms or findings is inadequate when considering extraesophageal manifestations of acid reflux that may very well occur without esophageal findings or symptoms. That, however, will have to be the subject of another investigation. Similarly, significant questions remain to be addressed regarding the prevalence of acid laryngitis, the specificity of the laryngoscopic findings for acid injury, and the responsiveness of acid laryngitis to antireflux medication. We conclude that increased proximal esophageal acid exposure can be shown in a subset of patients with reflux disease who experience chronic laryngeal symptomatology, and that this group of individuals is much more likely to exhibit laryngoscopic findings consistent with acid laryngitis than are GERD patients without laryngeal symptoms.

References 1. Freeland

AP, Ardran GM, Emrys-Roberts E. Globus hystericus and reflux esophagitis. J Laryngol Otol 1974;88:1025-1031. 2. Jackson C. Contact ulcer of larynx. Ann Otol Rhino1 Laryngol 1928;37:227-230. 3. von Leden H, Moore P. Contact ulcer of the larynx, experimental observations. Arch Otolaryngol 1960;72:746-751. 4. Jackson C, Jackson CL. Contact ulcer of larynx. Arch Otolaryngo1 1935;22:1-15. SI. Contact ulcer of the larynx. Laryngo5. Cherry J, Margulies scope 1968;78:1937-1940. J, Siegel CI, Margulies SI, Donner M. Pharyngeal 6. Cherry localization of symptoms of gastroesophageal reflux disease. Ann Otol Rhino1 Laryngol 1970;79:912-915. due to 7. Hallewell JD, Cole TB. Isolated head and neck symptoms hiatus hernia. Arch Otolaryngol 1970;92:499-501. JW, Thomas LE, Schaefer SD. Head and 8. Bain WM, Harrington neck manifestations of gastroesophageal reflux. Laryngoscope 1983;93:175-179. 9. Ward PH, Berci G. Observations

on the pathogenesis

of chronic

310

JACOB ET AL.

non-specific

pharyngitis and laryngitis. Laryngoscope 1982;92:

1377-1362.

10. Chodosh PL. Castro-esophago-pharyngeal reflux. Laryngoscope 1977;87:1418-1427. 11. Kambic V, Radsel Z. Acid posterior laryngitis, aetiology, histology, diagnosis and treatment. J Laryngol Otol 1984;98:12371240. 12. Ossakow SJ, Elta G, Colturi T, Bogdasarian R, Nostrant TT. Esophageal reflux and dysmotility as the basis for persistent cervical symptoms. Ann Otol Rhino1 Laryngol 1987;96:387392. 13. Wiener GJ, Koufman JA, Wu WC, Cooper JB, Richter JE, Caste11 DO. Chronic hoarseness secondary to gastroesophageal reflux disease: documentation with 24-H ambulatory pH monitoring. Am J Gastroenterol1989;84:1503-1508. 14. McNally PR, Maydonovitch CL, Prosek RA, Collette RP, Wong RKH. Evaluation of gastroesophageal reflux as a cause of idiopathic hoarseness. Dig Dis Sci 1989;34:1900-1904. 15. Kahrilas PJ, Dodds WJ, Hogan WJ, Kern M, Arndorfer RC, Reece A. Peristaltic dysfunction in peptic esophagitis. Gastroenterology 1986;91:897-904. 16. Kahrilas PJ, Dodds WJ, Dent J, Wyman JB, Hogan WJ, Arndorfer RC. Upper esophageal sphincter function during belching. Gastroenterology 1986;91:133-140. 17. Davidson GP, Dent J, Willing J, Kocyan P. Transient upper esophageal sphincter relaxations during acid gastroesophageal reflux in children (abstr). Gastroenterology 1988;94(5):ABB. 18. Kahrllas PJ, Dodds WJ, Dent J, Haeberle B, Hogan WJ, Arndorfer RC. Effect of sleep, spontaneous gastroesophageal reflux, and a meal on upper esophageal sphincter pressure in normal human volunteers. Gastroenterology 1987;92:466-471. 19. Jacob P, Kahrilas PJ, Herzon G, Ha T, McLaughlin B. Upper esophageal sphincter tone is augmented by a manometric device and diminished by anesthesia in the dog. Am J Physiol 1990;259:G245-G251.

GASTROENTEROLOGY Vol. 100, No. 2

20. Vakil NB, Kahrilas PJ, Dodds WJ, Vanagunas A. Absence of an upper esophageal sphincter response to acid reflux. Am J Gastroenterol 1989;84:606-610. 21. Helm JF, Dodds WJ, Riedel DR, Teeter BC, Hogan WJ, Arndorfer RC. Determinants of esophageal acid clearance in normal subjects. Gastroenterology 1983;85:607-612. 22. Delahunty JE, Cherry J. Experimentally produced vocal cord granulomas. Laryngoscope 1968;78:1941-1947. 23. Little FB, Koufman JA, Kohut RI, Marshall RB. Effect of gastric acid on the pathogenesis of subglottic stenosis. Ann Otol Rhino1 Laryngol1985;94:516-519. 24. DeMeester TR, Johnson LF, Joseph GY, Toscano MS, Hall AW, Skinner DB. Patterns of gastroesophageal reflux in health and disease. Ann Surg 1976;184:459-470. 25. Schindlbeck NE, Heinrich C, Dendorfer A, Pace F, MullerLissner SA. Influence of smoking and esophageal intubation on esophageal pH-metry. Gastroenterology 1987;92:1994-1997. 26. Shaker R, Helm JF, Dodds WJ, Hogan WJ. Revelations about ambulatory esophageal pH monitoring (abstr). Gastroenterology 1988;94(5):A421.

Received March 221990. Accepted July 31,199O. Address requests for reprints to: P. J. Kahrilas, M.D., Northwestern University Medical School, Gastroenterology Section, Department of Medicine, 1526 Wesley Towers, 250 East Superior Street, Chicago, Illinois 60611. Supported by National Institutes of Health National Research Service Award fellowship 0821201 (PJ.), a grant from the Veterans Administration Medical Research Service (P.J.K.), public health service grant ROl DC00646-OlAl (P.J.K.), and a fellowship from the Schweppe Foundation (P.J.K.). This work was presented at the 1989 meeting of the American Gastroenterological Association and published in part in abstract form (Gastroenterology 1989;96:A232).