Doi:10.1080/00365520600577940

Scandinavian Journal of Gastroenterology, 2006; 41: 131 Á/137 Laryngeal examination is superior to endoscopy in the diagnosis ofthe laryngopharyngeal form of gastroesophageal reflux disease LAIMAS JONAITIS1, RUTA PRIBUISIENE2, LIMAS KUPCINSKAS1 &VIRGILIJUS ULOZA2 1Department of Gastroenterology, Kaunas University of Medicine, Kaunas, Lithuania, and 2Department of Otolaryngology,Kaunas University of Medicine, Kaunas, Lithuania AbstractObjective. The laryngopharyngeal form of gastroesophageal reflux disease (LF GERD) is a frequent manifestation ofsupraesophageal GERD. Diagnosis of LF GERD is difficult: most of the common diagnostic methods of GERD haveinsufficient accuracy in establishing LF GERD. The purpose of this study was to evaluate the role of endoscopic andlaryngologic examination in the diagnosis of LF GERD and to create a laryngoscopic reflux index (LRI). Material andmethods. A total of 108 LF GERD patients and 90 controls were investigated. The criteria for LF GERD were:complaints, reflux-laryngitis, and esophagitis (endoscopically or histologically proven). Lesions in four laryngeal regionswere evaluated: arytenoids (A), intraarytenoid notch (IAN), vestibular folds (VF), and vocal cords (VC). Three types ofmucosal lesions were evaluated on a points basis: alterations of the epithelium, erythema, and edema. Total LRI wascalculated by summing-up the indices in the separate laryngeal areas. Results. The LRI mean value (11.489/3.78 points)of LF GERD patients was statistically significantly greater than that (1.649/1.93 points) of the controls. The mostsignificant laryngoscopic changes of LF GERD were: mucosal lesions of IAN, mucosal lesions of VC, and edema of VC.
A combination of these three findings reliably distinguishes the LF GERD patients from controls in 95.9% of cases.
The mucosal lesions of IAN have the greatest importance in diagnosing LF GERD: the odds ratio to LF GERD Á/ 21.32,p B/0.001. Endoscopic esophagitis was established in 36 (33.3%) cases. The severity of esophagitis did not correlate with theseverity of the laryngeal findings. Conclusions. Laryngoscopy is superior to endoscopy in diagnosing LF GERD.
Endoscopy has limited value in the diagnosis of LF GERD. Establishing the LRI could be helpful in the differentialdiagnosis of the disease in the everyday clinical practice.
Key Words: GERD, laryngopharyngeal reflux disease, laryngoscopic reflux index, reflux-laryngitis one of the atypical manifestations of supraesopha-geal GERD characterized by morphologic and The prevalence of gastroesophageal reflux disease functional changes in the larynx and pharynx with (GERD) is increasing world-wide and negatively associated clinical symptoms [4]. The diagnostics of interfering with the quality of life of the affectedpopulation [1,2]. It is known that because of LF GERD differs from that of a typical GERD. It pathological reflux of gastric contents the pathologi- depends on a different type of reflux, polymorpho- cal changes in the esophagus or supraesophageal logic manifestation of the disease, and on its inter- region (larynx, upper airways, and mouth) may mittent course [5]. For these reasons, most of the occur [3]. Supraesophageal reflux stimulates the common methods used in the diagnosis of GERD development of pathologic changes in the upper (endoscopy, pH-metry) have insufficient specificity respiratory tract, larynx, and mouth, i.e. causing and sensitivity in diagnosing the laryngopharyngeal manifestation of atypical GERD symptoms and morphological changes in those areas. The laryngo- The most distinct changes induced by supraeso- pharyngeal form of GERD (LF GERD) represents phageal reflux occur in the larynx, making laryngo- Correspondence: Laimas Jonaitis, Department of Gastroenterology, Kaunas University of Medicine, Eiveniu 2, 50009 Kaunas, Lithuania.
Tel: '/37068517472. Fax: '/37037331458. E-mail: [email protected] (Received 11 March 2005; accepted 11 May 2005) ISSN 0036-5521 print/ISSN 1502-7708 online # 2006 Taylor & FrancisDOI: 10.1080/00365520600577940 logic examination an important tool in the diagnos- tics of LF GERD. Modern methods of laryngologic During laryngoscopy, mucosal lesions in four lar- examination facilitate quantitative evaluation of yngeal regions were evaluated (Figure 1): arytenoids morphologic changes in the larynx, establishing (A), the intraarytenoid notch (IAN), vestibular folds associated functional disorders of phonation [7,8].
(VF), and vocal cords (VC). Three types of mucosal There are few data in the literature suggesting how lesions were evaluated: 1) alterations of the epithe- these methods could be used in evaluating the course lium (reduced mucosal light reflex, hypertrophy, and effects of treatment of LF GERD by drawing a roughness, and granuloma), 2) erythema, and 3) comparison of various treatment methods when investigating etiopathogenetic aspects of this form parameters were evaluated by points. Total laryngo- of disease [9,10]. Objective findings in the larynx that are the most characteristic for LF GERD need summing up the indices in the different laryngeal to be established in order to make an accurate areas. The scores of LRI were from 0 to 20 points Therefore our aim in conducting this study was to evaluate the importance of laryngologic and endo- Evaluation of pathological findings in the esophagus scopic examination in the diagnostics of LF GERDand to create a laryngoscopic reflux index (LRI) Endoscopy was performed only in individuals from which could become a convenient and accurate tool the patients’ group. Changes in the esophageal mucosa were evaluated by endoscopy and histo-logical examination of the samples of mucosaobtained from the lower third of the esophagus (two biopsy specimens from 1 to 2 cm above the Z The study was carried out at the Departments of line). Endoscopic assessment and pathological in- Otolaryngology and Gastroenterology of Kaunas vestigation were done by investigators blinded to any University of Medicine (Lithuania) in the years clinical data. Endoscopic evaluation of reflux eso- 2000 to 2003. We investigated 108 patients with phagitis was made using the Los Angeles (1994) LF GERD and 90 healthy persons in the control Reflux esophagitis was established if 1) erosions The patients’ group consisted of 108 persons (40 in the lower oesophagus were found (despite histological investigations) or 2) in the case of consulted an otolaryngologist because of hoarseness absence of erosions the histological esophagitis was lasting for more than 3 months, chronic throat- established. Histological esophagitis was diagnosed clearing, throat itching, and some other GERD on determining intra-epithelial lymphocytes, granu- locytes, hyperplasia of the basal area extending for these patients were found to have changes and more than 15% of epithelium thickness, paraker- symptoms typical for reflux laryngitis, and thepresence of concomitant reflux esophagitis wasconfirmed endoscopically or histologically.
Laryngopharyngeal reflux disease was established if patient fulfilled all three of the following criteria:Complaints (other possible reasons of complaintsexcluded), reflux-laryngitis, and esophagitis (endos-copically or histologically proven) The control group consisted of 90 randomly selected healthy subjects (36 M, 54 F, mean age36.99/11.5 years). Ten persons (11.1%) were cigar-ette smokers. They had neither chronic laryngealdiseases nor any other long-lasting voice disordersand none had ever consulted an otolaryngologistconcerning voice problems. In this respect theyconsidered themselves to be healthy and could serveas controls. The ratio of people in these groups didnot differ much regarding gender, social status, and Figure 1. Reflux laryngitis. Abbreviations: IAN 0/intraarytenoid notch; A 0/arytenoids; VF 0/vestibular folds; VC 0/vocal cords.
Table I. Evaluation of laryngoscopic changes and calculation of laryngoscopic reflux index (LRI).
Mucosal changes0 Á/ no changes1 Á/ reduced mucosa light reflex2 Á/ hypertrophy Intraarytenoid notch (IAN)'/1, if II8 edema* Intraarytenoid notch index (IANI) 0/(0 Á/6) '/2, if III8 edema**Vestibular folds (VF) Vocal cords index (VCI) 0/(0 Á/6)LRI /AI/INI/VFI/VCI *Edematous tissue fills whole notch; **edematous tissue fills whole notch and prolapses to the voice gap (according to J.A. Kaufman, 1994).
atosis, balonization of epitheliocytes of the median b B/0.05) larger than the indices in the control group layer, and augmentation and elongation of vascular papillae for more than 60% of epithelial thickness Owing to the fact that quantitative values (lar- yngoscopic indices) express characteristic laryngo-scopic changes of GERD, limiting (cut-off) values oflaryngoscopic indices were established with the aim of differentiating LF GERD patients from the Statistical analysis of data was performed using control group (Table III). Bearing in mind that the SPSS (Statistical Package for Social Sciences) 10.0 VC index depends on gender difference (the number of males in both groups was significantly higher than Whitney U-test, x2 and two-tailed t -test, as well as that of females) limiting (cut-off) values of males and parametric analysis of variance were used. Differ- females related to this parameter were established ences were considered to be reliable if the level of separately. The smallest (cut-off) value of LRI significance p or a (type I error) was B/0.05 and differentiating LF GERD patients from healthy b (type II error) was 5/0.2. A binary logistic persons is 5 points and the biggest possible LRI regression was carried out for determination of value is 20 points, the laryngoscopic changes of the the most important laryngoscopic parameters in patients are expressed within the limits of 5 Á/20 the diagnostics of LF GERD. By performing an points. Arithmetically dividing LRI values into three analysis of receiver operating characteristic (ROC) shares, we established three grades of reflux-laryngi- curves as presented in the SPSS program, limit (normal/abnormal) scores of quantitative laryngo- correspond to the degree I (I8), those of 11 Á logic parameters were established making it possible points to the degree II (II8) and those of 16 Á to differentiate LF GERD patients from healthy points to the degree III (III8). According to this system I8 of reflux laryngitis was established in 45% Table II. Comparison of mean scores of laryngoscopic indices ofLF GERD patients and the controls (in points).
The most important laryngoscopic signs in diagnosing LFGERD When assessing the rate of laryngoscopic signs, it was found that erythema and edema of the mucosa of the entire larynx of LF GERD patients were significantly (p B/0.001) more common in the pa- tients’ group than in the control group. The same statement applies to reduced mucosal light reflex As, IAN, VF, VC, as well as hypertrophy, roughness and granuloma of the mucosa. The mean values of LRI Abbreviations: LF GERD 0/laryngopharyngeal form of gastroeso- and of indices of separate areas of LF GERD phageal reflux disease; LRI 0/laryngoscopic reflux index.
patients were found to be significantly (a B *Statistically significant difference (p B/0.001).
Table III. Limiting (cut-off) values of the laryngoscopic indices.
IAN increases the odds ratio to be attributed to theLF GERD patient group by 21-fold (OR 0/21.32, Esophageal endoscopy and its relation to laryngeal Endoscopic esophagitis was established in 36 (33%) cases. In all of these cases histological esophagitis was established as well. Esophagitis A (according to the Los Angeles classification) was diagnosed in 26(72%) cases, and esophagitis B was established in 10 Abbreviations: LF GERD 0/laryngopharyngeal form of gastro-esophageal reflux disease; LRI 0/laryngoscopic reflux index.
(28%) cases. The frequency and degree of esopha- gitis did not correlate with the age of the patients(r 0/0.13, p /0.05). Esophagitis was established in of the LF GERD patients, the II8 in 39%, and the the rest of the 72 (67%) patients by histology only.
The histological esophageal changes were not age The following laryngoscopic findings (changes), and gender dependent (r 0/0.15, p /0.05).
established by logistic regression analysis, are the Endoscopic esophagitis (A and B) was established most significant for the diagnosis of LF GERD: 1) in 14 (31.8%) of 44 patients with I8 reflux laryngitis, mucosal lesions of the IAN; 2) mucosal lesions of the in 17 (40.5%) of 42 patients with II8 reflux VC; 3) edema of the VC. A combination of these laryngitis, and in 5 (29.4%) of 17 patients with III8 three laryngoscopic changes reliably distinguishes reflux laryngitis, p /0.05 among groups.
the LF GERD patients from healthy persons in We calculated the scores of the changes in 95.9% of cases. It was established that the mucosal different laryngeal areas as well as the indices of lesions of the IAN (hypertrophy, roughness, granu- these areas and total LRI in the patients without loma) have the greatest importance among the endoscopic esophagitis, with esophagitis A, and with investigated laryngoscopic signs in the diagnostics esophagitis B (Table IV). Despite some differences of LF GERD. The presence of mucosal lesions of the Table IV. Comparison of the mean scores of the findings in different laryngeal regions among non-erosive esophagitis, esophagitis A, andesophagitis B patients.
*Alterations of the epithelium Á/ reduced mucosa light reflex, hypertrophy, roughness, granuloma.
groups, there was no correlation between LRI and uncommon in these patients. There was no clear the degree of severity of endoscopic esophagitis association between baseline suspected reflux laryn-gitis sign or symptom severity and abnormal acidexposure. These investigators concluded that the utility of pH testing in establishing the diagnosis of The number of different manifestations of GERD, reflux laryngitis and response to therapy in this especially atypical forms (laryngopharyngeal symp- population should be re-evaluated. Moreover the pH-metry investigation itself may affect the patient’s behavior and false-negative results are likely [19].
routine clinical practice, such patients are sometimes We established that the combination of three referred to and from general practitioners and laryngoscopic findings (lesions of mucosa of vocal gastroenterologists to the otolaryngologists and pul- cords, lesions of mucosa of the IAN, and edema of monologists and vice versa. Because there are few the vocal cords) reliably distinguishes the LF GERD accurate diagnostic methods of the disease there is a patients from healthy persons in 95.9% of cases.
question of who should diagnose and treat patients Furthermore, we show that the detection of mucosal with suspected laryngopharyngeal disease.
lesions of the IAN (hypertrophy, roughness, granu- It seems that endoscopy has a limited role Á/ in our loma) increases the odds ratio for LF GERD by 21 setting only one-third of patients were found positive for LF GERD by endoscopy, similar finding being It is important to recognize that a comparatively shown in other series [13,14]. This is not surprising, simple, generally available investigation such as a because the antireflux defense in the esophagus is laryngoscopy may be the background for the diag- better developed than in the supraesophageal re- nosis of LF GERD. Some attempts were made to gions. There are no data determining how frequent define the role of laryngoscopic investigations in laryngopharyngeal reflux must be and how long it earlier studies. Habermann et al. investigated 29 must continue, how acidic (or alkaline) the refluxate patients with reflux-laryngitis and also evaluated the has to be to cause the acute or chronic changes in the changes in 4 laryngeal regions using a points system larynx and pharynx [15]. It is obvious that there is and calculated the separate indices of these regions no correlation between the severity of esophageal [20]. But there was no common index created and and laryngeal pathological findings, which is why no correlation found between these changes and endoscopy alone has limited value in predicting or pathological reflux. Belafsky et al. investigated 40 confirming the reflux-caused pathology in the lar- patients (gastroesophageal reflux confirmed by pH- yngopharyngeal area. Endoscopy seems not to be the metry) and 40 healthy controls and evaluated 8 investigation of choice in that clinical setting, there- laryngoscopic signs [21]. They calculated a common fore in our study we sought to elucidate the role of score of laryngoscopic changes (reflux finding score) laryngoscopic investigation in the establishment ofthe diagnosis of laryngopharyngeal reflux disease.
and established the value defining patients as com- Some questions may arise concerning our definition pared with healthy persons. But the drawbacks of the of LF GERD. We excluded the cases where there study were the lack of consistency, older populations was no endoscopic or histologic esophagitis present, were investigated and there might be some mathe- because we did not want to include patients in whom matical uncertainty in establishing their score. Siup- no objective findings of reflux were present, which sinskiene et al. studied 60 patients and established could lead to biased results. It could be speculated three degrees of reflux laryngitis [22]. But there was that we investigated only the most severe cases of LF no confirmation of pathological reflux in this study.
GERD. However, the absence of a correlation Probably we have the highest number of the between the severity of laryngoscopic findings and patients reported and the amount of patients is the degree of esophageal mucosa damage in our sufficient to establish statistically significant results.
study does not support this assumption. It might also We presume to have the mostly extensive investiga- be suggested that a 24-h pH-metry should be carried tion of the larynx and the confirmation of reflux is out, as a gold standard for the GERD [16]. But this also present. Our study is also one of the first to is a gold standard for the typical esophageal GERD.
provide the logistic regression analysis. All this There are conflicting data on the role of long-term suggests that LRI seems to be a useful tool not pH-metry in the establishment of the pathological only for the diagnosis, but also for the quantitative laryngopharyngeal reflux [17]. Richter et al. recently grading of the disease. This could have application in investigated LF GERD patients by using 3-pH- evaluating the efficacy of the treatment. Further- sensors 24-h pH-metry [18]. They established that, more, LRI could be applied as quantitative tool in regardless of symptoms, abnormal pH tests were Another important diagnostic tool is the proton- pharyngoesophageal reflux. Ann Otol Rhinol Laryngol 2001;/ pump inhibitor (PPI) treatment test [23,24], a [7] El Hennawi DN, Iskander NI, Ibrahim IA, Serwa A.
useful option in typical cases of symptomatic Persistent cough: prevalence of gastroesophageal reflux and study of relevant laryngeal signs. Otolaryngol Head Neck PPIs, as for the diagnostic test for LF GERD, remain to be established [25]. In a randomized [8] Malagelada JR. Review article: supra-oesophageal manifes- trial, Vaezi et al. treated patients with suspected tations of gastro-oesophageal reflux disease. Aliment Phar-macol Ther 2004;19(Suppl 1):43 Á laryngopharyngeal reflux disease with esomeprazole [9] Powitzky ES, Khaitan L, Garrett CG, Richards WO, Courey or placebo [26]. Surprisingly, 40 mg esomeprazole M. Symptoms, quality of life, videolaryngoscopy, and twice daily for 16 weeks was no more effective twenty-four-hour triple-probe pH monitoring in patients than placebo in resolving suspected LPR signs and with typical and extraesophageal reflux. Ann Otol Rhinol [10] Garrigues V, Gisbert L, Bastida G, Ortiz V, Bau I, Nos P, et further studies are needed to allow for better al. Manifestations of gastroesophageal reflux and response to characterization of the patient population with omeprazole therapy in patients with chronic posterior laryngitis: an evaluation based on clinical practice. Dig Dis Further studies on the pathogenesis of supraeso- phageal reflux disease are also needed. It is especially [11] Armstrong D. Endoscopic evaluation of gastro-oesophageal important to investigate the events of supraeso- reflux disease. Yale J Biol Med 1999;72:93 Á/100.
[12] Collins BJ, Elliot H, Sloan JM, MacFarlane RJ, Love AH.
phageal reflux during the supine position, especially Oesophageal histology in reflux oesophagitis. J Clin Pathol at night-time. This could be an important point for supraesophageal reflux disease [27]. Therefore [13] Poelmans J, Feenstra L, Demedts I, Rutgeerts P, Tack J. The well-designed controlled studies evaluating long- yield of upper gastrointestinal endoscopy in patients with time pH changes in supraesophageal regions are suspected reflux-related chronic ear, nose, and throatsymptoms. Am J Gastroenterol 2004;99:1419 Á [14] Catalano F, Terminella C, Grillo C, Biondi S, Zappala M, In conclusion, laryngoscopic examination of the Bentivegna C. Prevalence of oesophagitis in patients with larynx is an important and accurate method in the persistent upper respiratory symptoms. J Laryngol Otol diagnosis of LF GERD. We suggest that the majority of patients suspected of having LF GERD must be [15] Kuhn J, Toohill RJ, Ulualp SO, Kulpa J, Hofmann C, Arndorfer R, et al. Pharyngeal acid reflux events in referred immediately to the skilled otolaryngologist, patients with vocal cord nodules. Laryngoscope 1998;108: and afterwards the majority of these patients will not need endoscopy. Establishing the LRI could be [16] Cool M, Poelmans J, Feenstra L, Tack J. Characteristics and helpful in the differential diagnosis of the disease in clinical relevance of proximal oesophageal pH monitoring.
everyday clinical practice and further validation of Am J Gastroenterol 2004;99:2317 Á/23.
the LRI could lead to the establishment of a [17] Sermon F, Vanden Brande S, Roosens B, Mana F, Deron P, Urbain D. Is ambulatory 24-h dual-probe pH monitoring diagnostic algorithm and optimization of the diag- useful in suspected ENT manifestations of GERD? Dig [18] Richter J, Vaezi M, Stasney R, Hwang C, Leathers T, Sostek M, et al. Baseline pH-measurements for patients withsuspected signs and symptoms of reflux-laryngitis [Abstr W938]. Gastroenterology 2004;126(Suppl 2):A537.
[1] Nandurkar S, Talley NJ. Epidemiology and natural history of [19] Vaezi MF. Laryngitis and gastro-oesophageal reflux disease reflux disease. Baillieres Best Pract Res Clin Gastroenterol (GERD): improve laryngoscopic specificity, don’t do pH [2] Klinkenberg-Knol EC. Otolaryngologic manifestations of [20] Habermann W, Eherer A, Lindbichler F, Raith J, Friedrich gastro-oesophageal reflux disease. Scand J Gastroenterol G. Ex juvantibus approach for chronic posterior laryngitis: results of short-term pantoprazole therapy. J Laryngol Otol [3] Jaspersen D, Kulig M, Labenz J, Leodolter A, Lind T, Meyer-Sabellek, et al. Prevalence of extra-oesophageal [21] Belafsky PC, Postma GN, Koufman JA. The validity and manifestations in gastro-oesophageal reflux disease: an reliability of the reflux finding score (RFS). Laryngoscope analysis based on the ProGERD Study. Aliment Pharmacol [22] Siupsinskiene N, Adamonis K. Diagnostic test with ome- [4] Shaker R, Bardan E, Gu C, Kern M, Torrico L, Toohill R.
prazole in patients with posterior laryngitis. Medicina Intrapharyngeal distribution of gastric acid refluxate. Lar- [23] Bate CM, Riley SA, Chapman RW, Durnin AT, Taylor MD.
[5] Ormseth EJ, Wong RK. Reflux laryngitis: pathophysiology, Evaluation of omeprazole as a cost-effective diagnostic test diagnosis, and management. Am J Gastroenterol 1999;94: for gastro-oesophageal reflux disease. Aliment Pharmacol [6] McMurray JS, Gerber M, Stern Y, Walner D, Rudolph C, [24] Johnsson F, Hatlebakk JG, Klintenberg AC, Roman J, Willging JP, et al. Role of laryngoscopy, dual pH probe Toth E, Stubberod A, et al. One-week esomeprazole treat- monitoring, and laryngeal mucosal biopsy in the diagnosis of ment: an effective confirmatory test in patients with suspected gastroesophageal reflux disease. Scand J Gastro- study of acid suppression for the treatment of suspected LPR [Abstr.160]. Gastroenterology 2004;126(Suppl 2):P A Á/22.
[25] Nord HJ. Extraesophageal symptoms: what role for the [27] Fass R, Achem SR, Harding S, Mittal RK, Quigley E.
proton pump inhibitors? Am J Med 2004;117(Suppl 5A): Review article: supra-oesophageal manifestations of gastro- oesophageal reflux disease and the role of night-time [26] Vaezi M, Richter J, Stasney R, Spiegel J, Hwang C, Leathers gastro-oesophageal reflux. Aliment Pharmacol Ther 2004;/ T, et al. A randomized, double-blind, placebo-controlled

Source: http://www.lsmukk.lt/klinika10/publikacijos/Laryngeal%20examination%20is%20superiorto%20endoscopy%20in%20the%20diagnosis%20of%20the%20laryngopharyngeal%20form%20of%20gastroesophageal%20reflux%20disease.pdf