Original Research—Laryngology and Neurolaryngology
Narrow-Band Imaging of Laryngeal Images and Endoscopically Proven Reflux Esophagitis
Otolaryngology– Head and Neck Surgery 2015, Vol. 152(5) 874–880 Ó American Academy of Otolaryngology—Head and Neck Surgery Foundation 2015 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/0194599814568285 http://otojournal.org
Wen-Hung Wang, MD, PhD1,2,3, and Kai-Yu Tsai, MD1
No sponsorships or competing interests have been disclosed for this article.
Abstract Objective. To compare the difference between white light (WL) and narrow-band imaging (NBI) endoscopy in evaluating patients who had reflux laryngitis and esophagitis. Study Design. Retrospective review of medical records and endoscopic images. Setting. Outpatient clinic. Subjects and Methods. There were 102 consecutive patients with reflux esophagitis (mean age, 48 6 11 years) who had office-based transnasal esophagoscopy (TNE) with WL and NBI views, including 60 men (59%) and 42 women (41%). We compared WL and NBI endoscopy in observing the laryngeal and esophageal epithelium. The nasopharynx, base of the tongue, epiglottis, hypopharynx, larynx, esophagus, gastroesophageal junction, and stomach were examined, and all procedures were digitally recorded. All patients were evaluated with WL and NBI views to determine the reflux finding score (RFS) in the larynx and Los Angeles (LA) classification grade in the esophagus. Results. The NBI views were more sensitive than the WL views in the erythema/hyperemia, vocal cord edema, and global RFS scores. The NBI view facilitated the identification of the erythema/hyperemia change representing dilation or proliferation of microvessels caused by epithelial inflammation. The global RFS score was significantly associated with severity of LA grade only with the NBI view. Conclusion. The endoscopic findings with the NBI view permit an easier identification of the RFS parameters of laryngeal erythema/vocal cord edema, which have a stronger correlation with the severity of reflux esophagitis, than the WL view. The importance of NBI in the evaluation of reflux laryngitis and gastroesophageal reflux disease deserves further study.
Keywords endoscopy, gastroesophageal reflux disease, outpatient, reflux laryngitis
Received August 25, 2014; revised December 5, 2014; accepted December 23, 2014.
E
ar, nose, and throat physicians often encounter patients complaining of the sensation of a foreign body in the throat. This may be caused by diseases, including acute laryngitis, hypopharyngeal cancer, sinusitis, tonsillar hyperplasia, cervical abnormalities, and globus hystericus or pharyngeus. Globus hystericus or pharyngeus may be related to acid reflux disease, which may cause inflammation of the esophagus, termed reflux esophagitis.1 Gastroesophageal reflux disease (GERD) is a prevalent chronic disorder caused by the reflux of gastric contents into the esophagus. The reported incidence is 10% to 20% in Western countries and 5% in Asia.2,3 The most common symptoms include regurgitation and heartburn. Moreover, GERD may cause pulmonary problems, including asthma, bronchitis, microaspiration, and pulmonary fibrosis.4 The main findings of reflux esophagitis on endoscopy are mucosal breaks in the esophagus.3 The severity of esophagitis may be graded with the Los Angeles (LA) classification: grade A, 1 mucosal break (length 5 mm) that does not extend between the tops of 2 mucosal folds; grade B, 1 mucosal break (length .5 mm) that does not extend between the tops of 2 mucosal folds; grade C, 1 mucosal break that is continuous between the tops of 2 mucosal folds but involves \75% esophageal circumference; or grade D, 1 mucosal break that involves 75% esophageal circumference.5 Laryngopharyngeal reflux occurs when the height of the reflux exceeds the hypopharynx or larynx. Laryngeal inflammation is caused by stomach acid, and the patient 1 Department of Otolaryngology, Cathay General Hospital, Taipei, Taiwan, Republic of China 2 Department of Otolaryngology, Sijhih Cathay General Hospital, New Taipei City, Taiwan, Republic of China 3 School of Medicine, Fu-Jen Catholic University, New Taipei City, Taiwan, Republic of China
Corresponding Author: Wen-Hung Wang, MD, PhD, Department of Otolaryngology–Head and Neck Surgery, Cathay General Hospital and School of Medicine, Fu-Jen Catholic University, No. 280, Jen-Ai Road, Sec. 4, Taipei 10630, Taiwan. Email: [email protected]
Downloaded from oto.sagepub.com at WESTERN OREGON UNIVERSITY on May 20, 2015
Wang and Tsai
875
may experience a sensation of a foreign body in the throat, difficulty with swallowing, throat clearing, coughing, fatigue with talking, and hoarse voice. Esophageal inflammation caused by acid reflux may cause complaints of heartburn. Patients with laryngopharyngeal reflux do not necessarily experience heartburn, but they may have hoarseness, cough, laryngitis, subglottic stenosis, laryngeal cancer, sinusitis, pharyngitis, or sleep apnea.4 The reflux finding score (RFS) that is based on laryngoscopy is a clinical severity scale that standardizes the laryngeal findings of laryngopharyngeal reflux for better diagnosis and evaluation of therapeutic efficacy.6 The RFS, which is highly reproducible between different observers, involves 8 items: subglottic edema with pseudosulcus, ventricular obliteration, erythema/ hyperemia, vocal cord edema, diffuse laryngeal edema, posterior commissure hypertrophy, granuloma/granulation, and thick endolaryngeal mucus. The RFS may range from 0 (no abnormal findings) to 26 (worst), and RFS .7 predicts laryngopharyngeal reflux with 95% confidence.6 Conventional white light (WL) endoscopy for GERD has low sensitivity but excellent specificity (90%-95%).3 Major limitations of WL endoscopy in GERD include difficulties in visibly distinguishing and diagnosing nonerosive reflux disease (NERD) and Barrett esophagus.7 Laryngeal and esophageal mucosa may be observed with endoscopic techniques that use a full spectrum of visible light. Narrow-band imaging (NBI) endoscopy equipment, developed during the past decade, has better depth of penetration and increases the diagnostic sensitivity of endoscopy to characterize tissues. The NBI endoscopy uses narrow-bandwidth filters in a sequential red-green-blue illumination system and may detect superficial pharyngeal mucosal lesions.8-10 The red wavelength (540 nm) may penetrate the deepest layers and reveal dark larger vessels. The blue wavelength (415 nm) in the superficial layer shows microvascular red. The green wavelength (440 nm) is displayed between the red and blue colors. Light filtering through the filter palate controls the intensity of the 3 colors and adjusts color contrast. Therefore, the system reduces the unnecessary intermediate colors to increase the contrast of blood vessels. Angiogenesis, the formation of new blood vessels from preexisting vessels, is an important phenomenon in various physiological processes, including acute and chronic inflammation.11 Superficial mucosal lesions that usually cannot be detected by regular WL endoscopy can be identified with blue light in NBI endoscopy because of the vasculature pattern of neoangiogenesis. The inflammatory response involves dilation of capillaries to increase blood flow and microvascular structural changes.12 Vascular damage is common to inflammatory lesions following a wide variety of sublethal insults and is associated with characteristic changes in the inflamed tissues.13 Inflamed mucosa may be accompanied by dilation or proliferation of microvessels, and we may easily observe the severity of epithelial inflammation with NBI endoscopy. The relation between laryngeal findings with NBI endoscopy and reflux esophagitis is not well established. The purpose of this study was to compare the difference between
WL and NBI endoscopy in evaluating patients who had reflux laryngitis and esophagitis.
Methods Participants Consecutive patients (102 patients; mean age, 48 6 11 years) with reflux esophagitis who had office-based transnasal esophagoscopy (TNE) with WL and NBI views were included in this retrospective study. There were 60 men (59%) (mean age, 48 6 11 years) and 42 women (41%) (mean age, 47 6 11 years). We compared WL and NBI endoscopy in observing the laryngeal and esophageal epithelium. Included subjects were outpatient adult patients (age range, 21-65 years) who had endoscopically proven reflux esophagitis. Patients were excluded because of acute laryngitis, hypopharyngeal cancer, sinusitis, tonsillar hyperplasia, autoimmune diseases (xerostomia), or cervical abnormalities. The institutional ethics committee approved this study (Chang Gung Memorial Hospital Internal Review Board no. 100-1237B).
Narrow-Band Imaging System and Endoscopy The TNE with NBI system was equipped with a light source and central video system (CLV-180; Olympus Medical Systems, Tokyo, Japan). The system switched between conventional and NBI views with a button on the control section of the videoscope. All endoscopic examinations were performed in an outpatient clinic by 1 experienced otolaryngologist (W.-H.W.) who was blinded to the subject’s prior evaluation or diagnosis. The patient received simethicone (20 mg/mL; dose, 20 mL) 30 minutes before endoscopy, and the nasal cavity was anesthetized with lidocaine spray (2%). The TNE was performed with the patient in the seated position, first in the WL mode and then using the NBI system. The nasopharynx, base of the tongue, epiglottis, hypopharynx, larynx, esophagus, gastroesophageal junction, and stomach were examined, and all procedures were digitally recorded. All patients were evaluated to determine the RFS in the larynx (both using WL and NBI), the LA classification grade in the esophagus (both using WL and NBI), and the presence of a brownish area (using NBI). The TNE examinations were performed by 1 physician who was blinded to the subject’s prior diagnosis. We retrospectively reviewed the medical records of all 102 patients. A blinded reviewer read the TNE images and determined the RFS in all patients. The LA classification was used in the esophagus to avoid subjective differences and interobserver variability in image interpretation. In addition, we compared the severity of the image interpretation and association between both WL and NBI views.
Statistical Analysis We compared the outcome differences between the WL and NBI views in the larynx and esophagus using a paired t test. Spearman rank correlation was used to determine comparative relevance of laryngeal and esophageal images. Several conditions (inflammation, trauma, or infection) could cause
Downloaded from oto.sagepub.com at WESTERN OREGON UNIVERSITY on May 20, 2015
876
Otolaryngology–Head and Neck Surgery 152(5)
Figure 1. Endoscopic finding showed negative reflux (control) under white light (WL) (A) and narrow-band imaging (NBI) views (B). A patient with arytenoid erythema and diffuse edema under WL (C) and NBI views (D). A patient with a more apparent brownish area and edema (arrows) under the NBI view (F) than the WL view (E).
Table 1. Comparison of RFS under WL and NBI Views.a RFS Variable Subglottic edema with pseudosulcus Ventricular obliteration Erythema/hyperemia Vocal cord edema Diffuse laryngeal edema Posterior commissure hypertrophy Granuloma/granulation Thick endolaryngeal mucus Global RFS score
WL View
NBI View
161
0.9 6 1
P Valueb .158
2.9 6 1 2.6 6 0.9 1.760.6 2.1 6 0.6 2.2 6 0.6
2.9 6 1 3.161 1.8 6 0.6 2.2 6 0.5 2.2 6 0.6
1.000 \.001 \.001 .014 1.000
0.4 6 0.8 0.4 6 0.8 13.2 6 3.7
0.4 6 0.8 0.4 6 0.8 13.5 6 3.8
1.000 1.000 \0.001
Abbreviations: NBI, narrow-band image; RFS, reflux finding score; WL, white light. a The RFS scores are reported as mean 6 standard deviation. The figures in boldface represent a significant difference between WL and NBI. b Paired t test.
similar laryngeal reflux findings such as erythema, hyperemia, or vocal cord edema; therefore, statistical significance was defined by P .001.
Results RFS from WL and NBI Views We examined the 8 items of RFS in the larynx. In 1 patient, there was arytenoid erythema and diffuse edema noted with
WL and NBI views, in contrast with image findings of negative reflux (control) with WL and NBI views (Figure 1). We could identify the brownish area over bilateral arytenoid and interarytenoid folds. The inflamed mucosa often was accompanied by dilation or proliferation of microvessels, and we could easily observe the range and severity of epithelial inflammation with NBI views. Another patient had a brownish area and edema over bilateral arytenoids and the posterior part of the false vocal cords that were more apparent with the NBI than the WL view (Figure 1). The NBI views were more sensitive than the WL views in erythema/hyperemia, vocal cord edema, and global RFS scores (Table 1). The NBI view facilitated the identification of the erythema/hyperemia change representing dilation or proliferation of microvessels caused by epithelial inflammation. However, there was no significant advantage of the NBI compared with the WL view in detecting subglottic edema with pseudosulcus, ventricular obliteration, posterior commissure hypertrophy, granuloma/granulation, or thick endolaryngeal mucus.
Laryngeal RFS with WL and NMI Views and the LA Classification We compared the laryngeal signs of RFS between the WL and NBI views and the LA classification in the esophagus using Spearman correlation. The endoscopic images of granuloma/granulation and thick endolaryngeal mucus in the larynx with both WL and NBI views had higher correlations with the severity of LA grade (Table 2). When granuloma/ granulation and thick mucus in the larynx were identified
Downloaded from oto.sagepub.com at WESTERN OREGON UNIVERSITY on May 20, 2015
Wang and Tsai
877
Table 2. Correlation between Laryngeal Signs of RFS under WL and NBI Views and LA Classification.a RFS Parameter vs LA Classification Subglottic edema with pseudosulcus Ventricular obliteration Erythema/hyperemia Vocal cord edema Diffuse laryngeal edema Posterior commissure hypertrophy Granuloma/granulation Thick endolaryngeal mucus Global RFS score
WL View P Valueb
NBI View P Valueb
0.087 (.383)
0.112 (.2860)
0.063 0.123 0.195 0.124 0.279
0.063 (.531) 0.151 (.130) 0.166 (.096) 0.150 (.132) 0.279 (.005)
(.531) (.217) (.049) (.213) (.005)
0.440 (\.001) 0.513 (\.001) 0.168 (.092)
0.440 (\.001) 0.513 (\.001) 0.382 (\.001)
Abbreviations: LA, Los Angeles; NBI, narrow-band image; RFS, reflux finding score; WL, white light. a The values reported are the Spearman correlation coefficients (Spearman r values). The figures in boldface represent a significant association with the LA classification. b Spearman correlation test.
on endoscopy (WL or NBI views), a higher grade of esophagitis was more likely. Patients who had endoscopic findings of posterior commissure hypertrophy and granuloma with WL and NBI views typically had a higher grade of reflux esophagitis such as LA grade C with WL and NBI views (Figure 2). However, the global RFS score was significantly associated with severity of LA grade only with the NBI view (Table 2 and Figure 3).
Discussion The TNE used in this study included a flexible endoscope that was passed transnasally under topical anesthesia.14 The advantages of TNE compared with conventional esophagoscopy under general anesthesia included increased patient safety, increased practice efficiency, cost savings, and early detection of esophageal cancer.15 The use of office-based TNE by otolaryngologists is available for patients who are at risk for Barrett esophagus, esophageal adenocarcinoma, hypopharyngeal cancer extended to the esophagus, and GERD.16 The TNE can be safely performed with topical anesthesia in an office setting for diagnostic and therapeutic procedures. Previous studies showed that TNE had no major complications such as esophageal perforation, and the minor complications included epistaxis (most common), vasovagal reaction, and laryngospasm.14,17 Most patients tolerate TNE well, and patient tolerance to TNE on a 10-point scale (1, no discomfort; 10, severe discomfort) is 2.0 6 1.2 (range, 1-4).17 Acute laryngitis usually is caused by infection, but chronic laryngitis may result from acid reflux, smoking, muscular imbalance, or dehydration.18 Most (50%-60%) chronic laryngitis and chronic pharyngitis may be related to GERD.19 The most common symptoms associated with
Figure 2. Endoscopic finding for a patient showing posterior commissure hypertrophy with granuloma and Los Angeles grade C esophagitis under white light (A and C) and narrow-band imaging views (B and D).
laryngopharyngeal reflux include throat cleaning (98.3%), chronic cough (96.6%), heartburn (95.7%), and hoarseness (94.9%).20 Animal studies have shown that small amounts of gastric acid may cause marked laryngeal mucosal inflammation and edema.21 Extraesophageal gastric acid may cause vagal-mediated reflexes such as bronchospasm, laryngospasm, and cough and may cause chronic laryngitis.22 In addition to detailed medical history and laryngeal observations, diagnosis of these diseases may require a 24-hour pH recording, but this study may be limited because of high cost and reluctance of patients to perform the study. Furthermore, a negative pH study does not exclude extraesophageal reflux because reflux episodes may be intermittent.23 Therefore, we sought to improve general outpatient techniques to facilitate examination and diagnosis. Most patients with GERD have no evidence of esophageal damage with standard WL endoscopy. Nonerosive subtypes of GERD (NERD) account for 70% patients with GERD.24 Standard WL endoscopy fails to detect GERD endoscopically in 60% to 70% patients. The sensitivity of standard WL endoscopy for the diagnosis is low.25,26 A previous study showed that NBI may increase the sensitivity for the diagnosis of GERD.27 In a previous study of patients with and without GERD symptoms who completed 2 validated GERD questionnaires, the distal esophagus was examined by standard WL and NBI endoscopy, and the features observed only with NBI were compared between patients with GERD and control subjects. On multivariate analysis, increased number
Downloaded from oto.sagepub.com at WESTERN OREGON UNIVERSITY on May 20, 2015
878
Otolaryngology–Head and Neck Surgery 152(5)
Figure 3. A visual representation of the data presented in Table 2. LA, Los Angeles; NBI, narrow-band image; RFS, reflux finding score; WL, white light.
and dilation of intrapapillary capillary loops were the best predictors for diagnosing GERD.28 The maximum, minimum, and average numbers of intrapapillary capillary loops/ field were significantly greater in the GERD group than in control subjects (P \ .0001). Although the interobserver agreement for the various NBI findings was very good, the intraobserver agreement was modest.28 Another study of 3 subtypes of GERD (NERD, reflux esophagitis, and Barrett esophagus) using NBI magnifying endoscopy showed that a significantly higher proportion of patients with GERD had an increased number of intrapapillary capillary loops, microerosions, and nonround pit patterns below the squamocolumnar junction than did healthy control subjects, and the numbers of intrapapillary capillary loops/field were significantly greater in patients with GERD.29 Another comparative study of NBI and convention endoscopy in the diagnosis of GERD (107 subjects: NERD, 36 subjects; erosive reflux disease, 41 subjects; controls, 30 subjects) showed that microerosions, increased vascularity, and pit patterns at the squamocolumnar junction not seen on conventional endoscopy were well seen with NBI; interobserver agreement was good for increased vascularity (k = 0.95) and microerosions (k = 0.89) but lower for pit patterns (k = 0.59).30 Therefore, NBI views improved the endoscopic diagnosis of mucosal morphology at the squamocolumnar junction and GERD.30 A study that evaluated intra- and interobserver variations in the endoscopic scoring of esophagitis with conventional imaging with and without NBI showed that intra- and interobserver reproducibility in grading esophagitis could be improved when NBI was applied with conventional imaging because of improved depictions of small erosive foci.31 These previous studies examined the lining of the esophagus with NBI magnifying endoscopy (esophagogastroduodenoscopy with 80-fold optical magnification) to easily observe pit patterns and intrapapillary capillary loops, but these were not easily performed with TNE because of the absence of a magnification function. In a study of 111 patients with laryngopharyngeal reflux who underwent TNE by the same otolaryngologist,
biopsy-proven Barrett esophagus was observed in 13.5% patients, including 7 of 58 patients (12.1%) with WL and 8 of 53 patients (15.1%) with NBI; 3 patients (2.7%) had dysplasia on biopsy, and all 3 cases were detected with NBI (5.7%).32 It was concluded that NBI may be a useful adjunct to increase the diagnostic sensitivity of TNE by the otolaryngologist.32 Our preliminary results (data not shown) for diagnosis of GERD by TNE also showed that we could easily observe microerosions and Barrett esophagus, except pit patterns, number, and dilation of intrapapillary capillary loops, with the NBI view better than the WL view. Several studies have shown the benefit of NBI in the early detection of recurrent laryngeal and hypopharyngeal cancer and early diagnosis of laryngeal cancer.33-35 A previous study showed that the NBI laryngoscope is highly valuable for the diagnosis and treatment of laryngopharyngeal reflux disease and that laryngopharyngeal reflux disease has characteristic findings on NBI laryngoscopy, including erythema and edema between the arytenoid cartilages (71.7%); epiglottis congestion (67.4%); pharyngeal isolation/integration erythema (65.2%); pharyngeal pebble-like changes (65.2%); hypertrophy of the posterior commissure (52.2%); vocal cord erythema and edema (47.8%); vocal nodules or vocal polyps with erythema or edema (39.1%), arytenoid cartilage edema and erythema (21.7%); ventricular edema, erythema, and laryngeal ventricle disappearance (17.4%); granuloma (6.5%); ulcers (4.3%); and false vocal cord pitch (4.3%).36 Several studies have investigated the association between laryngopharyngeal reflux and GERD. In 1 study, the prevalence rate of laryngopharyngeal reflux in subjects with reflux esophagitis was 23.9%.37 Another study showed that 24-hour multichannel intraluminal impedance and pH esophageal monitoring analysis confirmed GERD diagnosis in \40% patients with a previous diagnosis of laryngopharyngeal reflux, most likely because of the low specificity of the laryngoscopic findings.38 However, the expression of Helicobacter pylori positivity and the degree of GERD correlate with laryngopharyngeal reflux, and Helicobacter pylori positivity and the degree of GERD are higher in patients with RFS 7.39
Downloaded from oto.sagepub.com at WESTERN OREGON UNIVERSITY on May 20, 2015
Wang and Tsai
879
Another study showed that the severity of laryngeal mucosal lesions in patients with GERD is significantly greater than in control patients, and a higher degree of laryngeal mucosal injury may be documented in patients for whom GERD is associated with more advanced esophageal lesions.40 Both the global RFS score and scores of all RFS parameters (except for presence of granulomatous tissue) were significantly higher in patients with GERD than in control subjects; patients in whom GERD was associated with more severe esophageal lesions (LA group B) had a significantly higher global RFS score and scores of all analyzed parameters of laryngeal injury (except subglottic edema) than individuals in whom the degree of esophageal involvement was classified as group A.40 In our study, when the RFS parameters of granuloma/granulation and thick endolaryngeal mucus were identified in the larynx with either WL or NBI endoscopy, a higher grade of esophagitis was present. The granulation and thick mucus were important predictors for GERD and erosive esophagitis. In contrast with the previous study, we observed that the global RFS score was significantly associated with severity of the LA grade only with NBI, and traditional WL endoscopy was not sufficiently reliable to reflect the severity of reflux esophagitis.40 Differences between the studies include the fewer patients with GERD in the previous (46 patients with GERD and 46 controls) than in present study (GERD, 102 patients), and the grading interpretation of GERD may be inconsistent between WL and NBI views, which may affect the calculation of the correlation with laryngopharyngeal reflux.40 The mean global RFS scores in our study were 13.2 6 3.7 in WL and 13.5 6 3.8 in NBI views, consistent with previous findings.6 The endoscopic findings in the larynx showed that NBI has higher sensitivity for detecting erythema, hyperemia, and edema in the mucosa, possibly because of the ability of NBI to increase the contrast of blood vessels and improve the identification of the neoangiogenic vasculature in inflamed mucosa. Limitations of the present study include the lack of a standard 24-hour pH recorder for the definite diagnosis of GERD. There was a lack of a patient reflux symptom index41 measurement to determine laryngeal signs and symptoms that most significantly correlated. There were no treatment data for patients who had higher scores based on NBI and who did better with antireflux treatment. Therefore, further studies with longer follow-up are required to determine the usefulness of NBI in evaluating laryngopharyngeal reflux and identifying correlations with GERD.
Conclusion Literature review showed no previous comparison of the endoscopic images of laryngopharyngeal reflux and GERD with WL and NBI views. Although previous studies showed that NBI with and without magnification may increase the sensitivity for the diagnosis of GERD and observation of microerosions and Barrett esophagus, there was limited previous study to investigate the association between reflux laryngitis and esophagitis with NBI.
In addition, the imaging technology for the improved detection of laryngopharyngeal reflux appears promising. The endoscopic findings with the NBI view permit an easier identification of the RFS parameters of laryngeal erythema/vocal cord edema, which have a stronger correlation with the severity of reflux esophagitis, than the WL view. Among the RFS variables, granuloma/granulation and thick endolaryngeal mucus might be 2 important laryngeal imaging signs for predicting the severity of reflux esophagitis with either the WL or NBI view. The importance of NBI in the relation between reflux laryngitis and GERD deserves further study. A large prospective study with more patients, reflux symptom index, and outcome data about antireflux treatment are needed for the future study of NBI in reflux esophagitis. Author Contributions Wen-Hung Wang, study concept and design, acquisition of data, analysis and interpretation of data, drafting of the manuscript, critical revision of the manuscript for important intellectual content, statistical analysis; Kai-Yu Tsai, acquisition of data, analysis and interpretation of data, drafting of the manuscript, critical revision of the manuscript for important intellectual content, statistical analysis.
Disclosures Competing interests: None. Sponsorships: None. Funding source: None.
References 1. Lee BE, Kim GH. Globus pharyngeus: a review of its etiology, diagnosis and treatment. World J Gastroenterol. 2012;18: 2462-2471. 2. Lynch CR, Wani S, Rastogi A, et al. Effect of acid-suppressive therapy on narrow band imaging findings in gastroesophageal reflux disease: a pilot study. Dis Esophagus. 2013;26:124-129. 3. Kasap E, Zeybel M, Asxık G, Ayhan S, Yu¨ceyar H. Correlation among standard endoscopy, narrow band imaging, and histopathological findings in the diagnosis of nonerosive reflux disease. J Gastrointest Liver Dis. 2011;20:127-130. 4. Sansak I, Leelakusolvong S, Charatcharoenwitthaya P, et al. Esophageal reflux episodes and endoscopic findings in laryngopharyngeal reflux patients with persistent symptoms despite acid suppression therapy. Thai J Gastroenterol. 2010;11:97-103. 5. Lundell LR, Dent J, Bennett JR, et al. Endoscopic assessment of oesophagitis: clinical and functional correlates and further validation of the Los Angeles classification. Gut. 1999;45:172-180. 6. Peter C. Belafsky, Gregory N. Postma, James A. Koufman. The validity and reliability of the reflux finding score (RFS). Laryngoscope. 2001;111:1313-1317. 7. Banerjee R, Reddy DN. Enhanced endoscopic imaging and gastroesophageal reflux disease. Indian J Gastroenterol. 2011; 30:193-200. 8. Wang WH, Lin YC, Weng HH, et al. Narrow-band imaging for diagnosing adenoid hypertrophy in adults: a simplified grading and histologic correlation. Laryngoscope. 2011;121:965-970.
Downloaded from oto.sagepub.com at WESTERN OREGON UNIVERSITY on May 20, 2015
880
Otolaryngology–Head and Neck Surgery 152(5)
9. Lin YC, Wang WH, Lee KF, et al. Value of narrow band imaging endoscopy in early mucosal head and neck cancer. Head Neck. 2012;34:1574-1579. 10. Tan NC, Herd MK, Brennan PA, Puxeddu R. The role of narrow band imaging in early detection of head and neck cancer. J Oral Maxillofac Surg. 2012;50:132-136. 11. Sua´rez Y. Microregulation of plaque neovascularization. Arterioscler Thromb Vasc Biol. 2010;30:1500-1501. 12. Young-Joon S. Oxidative Stress, Inflammation, and Health. Boca Raton, FL: CRC/Taylor & Francis; 2005. 13. Moses JM, Ebert RH, Graham RC, Brine KL. Pathogenesis of inflammation. J Exp Med. 1964;120:57-82. 14. Sabirin J, Abd Rahman M, Rajan P. Changing trends in oesophageal endoscopy: a systematic review of transnasal oesophagoscopy. ISRN Otolaryngol. 2013;2013:586973. 15. Aviv JE. Transnasal esophagoscopy: state of the art. Otolaryngol Head Neck Surg. 2006;135:616-619. 16. Banerjee R, Reddy DN. Enhanced endoscopic imaging and gastroesophageal reflux disease. Indian J Gastroenterol. 2011; 30:193-200. 17. Aviv JE, Takoudes TG, Ma G, Close LG. Office-based esophagoscopy: a preliminary report. Otolaryngol Head Neck Surg. 2001;125:170-175. 18. Hopkins C, Yousaf U, Pedersen M. Acid reflux treatment for hoarseness. Cochrane Database Syst Rev. 2006;(1):CD005054. 19. Vaezi MF. Sensitivity and specificity of reflux-attributed laryngeal lesions: experimental and clinical evidence. Am J Med. 2003;115(suppl 3A):97S-104S. 20. Vaezi MF. Laryngitis and gastroesophageal reflux disease: increasing prevalence or poor diagnostic tests. Am J Gastroenterol. 2004; 99:786-788. 21. Ludemann JP, Manoukian J, Shaw K, et al. Effects of simulated gastroesophageal reflux on the untraumatized rabbit larynx. J Otolaryngol. 1998;27:127-131. 22. Qua CS, Wong CH, Gopala K, Goh KL. Gastro-oesophageal reflux disease in chronic laryngitis: prevalence and response to acid-suppressive therapy. Aliment Pharmacol Ther. 2007;25: 287-295. 23. Richter JE. Ambulatory esophageal pH monitoring. Am J Med. 1997;103:130S-134S. 24. Azumi T, Adachi K, Furuta K, et al. Esophageal epithelial surface in patients with gastroesophageal reflux disease: an electron microscopic study. World J Gastroenterol. 2008;14:57125716. 25. Frazzoni M, DeMicheli E, Savarino V. Different patterns of oesophageal acid exposure distinguish complicated reflux disease from either erosive reflux oesophagitis or non-erosive reflux disease. Aliment Pharmacol Ther. 2003;18:1091-1098. 26. Dent J, Armstrong D, Delaney B, et al. Symptom evaluation in reflux disease: workshop background, processes, terminology recommendations, and discussion outputs. Gut. 2004;53:1-24.
27. Assirati FS, Hashimoto CL, Dib RA, Fontes LH, NavarroRodriguez T. High definition endoscopy and ‘‘narrow band imaging’’ in the diagnosis of gastroesophageal reflux disease. Arq Bras Cir Dig. 2014;27:59-65. 28. Sharma P, Wani S, Bansal A, et al. A feasibility trial of narrow band imaging endoscopy in patients with gastroesophageal reflux disease. Gastroenterology. 2007;133:454-464. 29. Lv J, Liu D, Ma SY, Zhang J. Investigation of relationships among gastroesophageal reflux disease subtypes using narrow band imaging magnifying endoscopy. World J Gastroenterol. 2013;19:8391-8397. 30. Fock KM, Teo EK, Ang TL, Tan JY, Law NM. The utility of narrow band imaging in improving the endoscopic diagnosis of gastroesophageal reflux disease. Clin Gastroenterol Hepatol. 2009;7:54-59. 31. Lee YC, Lin JT, Chiu HM, et al. Intraobserver and interobserver consistency for grading esophagitis with narrow-band imaging. Gastrointest Endosc. 2007;66:230-236. 32. O’Brien TJ, Parham K. Transnasal endoscopy: white light versus narrowband image. Ann Otol Rhinol Laryngol. 2009; 117:886-890. 33. Zabrodsky M, Lukes P, Lukesova E, Boucek J, Plzak J. The role of narrow band imaging in the detection of recurrent laryngeal and hypopharyngeal cancer after curative radiotherapy. Biomed Res Int. 2014;2014:175398. 34. Kraft M, Fostiropoulos K, Gu¨rtler N, et al. Value of narrow band imaging in the early diagnosis of laryngeal cancer [published online July 3, 2014]. Head Neck. doi:10.1002/hed.23838. 35. Piazza C, Del Bon F, Peretti G, Nicolai P. Narrow band imaging in endoscopic evaluation of the larynx. Curr Opin Otolaryngol Head Neck Surg. 2012;20:472-476. 36. He N, Si Y, Yong Y, Weng J, Qin Y. The value of narrow band imaging laryngoscope for laryngopharyngeal reflux disease. Lin Chung Er Bi Yan Hou Tou Jing Wai Ke Za Zhi. 2012;26:776-778. 37. Lai YC, Wang PC, Lin JC. Laryngopharyngeal reflux in patients with reflux esophagitis. World J Gastroenterol. 2008;14:4523-4528. 38. Bortoli N, Nacci A, Savarino E, et al. How many cases of laryngopharyngeal reflux suspected by laryngoscopy are gastroesophageal reflux disease-related? World J Gastroenterol. 2012; 18:4363-4370. 39. Tezer MS, Kockar MC, Kocxkar O, Celik A. Laryngopharyngeal reflux finding scores correlate with gastroesophageal reflux disease and Helicobacter pylori expression. Acta Otolaryngol. 2006;126:958-961. 40. Krawczyk M, Scierski W, Ryszkiel I, et al. Endoscopic evidence of reflux disease in the larynx. Acta Otolaryngol. 2014; 134:831-837. 41. Mesallam TA, Stemple JC, Sobeih TM, Elluru RG. Reflux symptom index versus reflux finding score. Ann Otol Rhinol Laryngol. 2007;116:436-440.
Downloaded from oto.sagepub.com at WESTERN OREGON UNIVERSITY on May 20, 2015
Narrow-Band Imaging of Laryngeal Images and Endoscopically Proven Reflux Esophagitis
Otolaryngology– Head and Neck Surgery 2015, Vol. 152(5) 874–880 Ó American Academy of Otolaryngology—Head and Neck Surgery Foundation 2015 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/0194599814568285 http://otojournal.org
Wen-Hung Wang, MD, PhD1,2,3, and Kai-Yu Tsai, MD1
No sponsorships or competing interests have been disclosed for this article.
Abstract Objective. To compare the difference between white light (WL) and narrow-band imaging (NBI) endoscopy in evaluating patients who had reflux laryngitis and esophagitis. Study Design. Retrospective review of medical records and endoscopic images. Setting. Outpatient clinic. Subjects and Methods. There were 102 consecutive patients with reflux esophagitis (mean age, 48 6 11 years) who had office-based transnasal esophagoscopy (TNE) with WL and NBI views, including 60 men (59%) and 42 women (41%). We compared WL and NBI endoscopy in observing the laryngeal and esophageal epithelium. The nasopharynx, base of the tongue, epiglottis, hypopharynx, larynx, esophagus, gastroesophageal junction, and stomach were examined, and all procedures were digitally recorded. All patients were evaluated with WL and NBI views to determine the reflux finding score (RFS) in the larynx and Los Angeles (LA) classification grade in the esophagus. Results. The NBI views were more sensitive than the WL views in the erythema/hyperemia, vocal cord edema, and global RFS scores. The NBI view facilitated the identification of the erythema/hyperemia change representing dilation or proliferation of microvessels caused by epithelial inflammation. The global RFS score was significantly associated with severity of LA grade only with the NBI view. Conclusion. The endoscopic findings with the NBI view permit an easier identification of the RFS parameters of laryngeal erythema/vocal cord edema, which have a stronger correlation with the severity of reflux esophagitis, than the WL view. The importance of NBI in the evaluation of reflux laryngitis and gastroesophageal reflux disease deserves further study.
Keywords endoscopy, gastroesophageal reflux disease, outpatient, reflux laryngitis
Received August 25, 2014; revised December 5, 2014; accepted December 23, 2014.
E
ar, nose, and throat physicians often encounter patients complaining of the sensation of a foreign body in the throat. This may be caused by diseases, including acute laryngitis, hypopharyngeal cancer, sinusitis, tonsillar hyperplasia, cervical abnormalities, and globus hystericus or pharyngeus. Globus hystericus or pharyngeus may be related to acid reflux disease, which may cause inflammation of the esophagus, termed reflux esophagitis.1 Gastroesophageal reflux disease (GERD) is a prevalent chronic disorder caused by the reflux of gastric contents into the esophagus. The reported incidence is 10% to 20% in Western countries and 5% in Asia.2,3 The most common symptoms include regurgitation and heartburn. Moreover, GERD may cause pulmonary problems, including asthma, bronchitis, microaspiration, and pulmonary fibrosis.4 The main findings of reflux esophagitis on endoscopy are mucosal breaks in the esophagus.3 The severity of esophagitis may be graded with the Los Angeles (LA) classification: grade A, 1 mucosal break (length 5 mm) that does not extend between the tops of 2 mucosal folds; grade B, 1 mucosal break (length .5 mm) that does not extend between the tops of 2 mucosal folds; grade C, 1 mucosal break that is continuous between the tops of 2 mucosal folds but involves \75% esophageal circumference; or grade D, 1 mucosal break that involves 75% esophageal circumference.5 Laryngopharyngeal reflux occurs when the height of the reflux exceeds the hypopharynx or larynx. Laryngeal inflammation is caused by stomach acid, and the patient 1 Department of Otolaryngology, Cathay General Hospital, Taipei, Taiwan, Republic of China 2 Department of Otolaryngology, Sijhih Cathay General Hospital, New Taipei City, Taiwan, Republic of China 3 School of Medicine, Fu-Jen Catholic University, New Taipei City, Taiwan, Republic of China
Corresponding Author: Wen-Hung Wang, MD, PhD, Department of Otolaryngology–Head and Neck Surgery, Cathay General Hospital and School of Medicine, Fu-Jen Catholic University, No. 280, Jen-Ai Road, Sec. 4, Taipei 10630, Taiwan. Email: [email protected]
Downloaded from oto.sagepub.com at WESTERN OREGON UNIVERSITY on May 20, 2015
Wang and Tsai
875
may experience a sensation of a foreign body in the throat, difficulty with swallowing, throat clearing, coughing, fatigue with talking, and hoarse voice. Esophageal inflammation caused by acid reflux may cause complaints of heartburn. Patients with laryngopharyngeal reflux do not necessarily experience heartburn, but they may have hoarseness, cough, laryngitis, subglottic stenosis, laryngeal cancer, sinusitis, pharyngitis, or sleep apnea.4 The reflux finding score (RFS) that is based on laryngoscopy is a clinical severity scale that standardizes the laryngeal findings of laryngopharyngeal reflux for better diagnosis and evaluation of therapeutic efficacy.6 The RFS, which is highly reproducible between different observers, involves 8 items: subglottic edema with pseudosulcus, ventricular obliteration, erythema/ hyperemia, vocal cord edema, diffuse laryngeal edema, posterior commissure hypertrophy, granuloma/granulation, and thick endolaryngeal mucus. The RFS may range from 0 (no abnormal findings) to 26 (worst), and RFS .7 predicts laryngopharyngeal reflux with 95% confidence.6 Conventional white light (WL) endoscopy for GERD has low sensitivity but excellent specificity (90%-95%).3 Major limitations of WL endoscopy in GERD include difficulties in visibly distinguishing and diagnosing nonerosive reflux disease (NERD) and Barrett esophagus.7 Laryngeal and esophageal mucosa may be observed with endoscopic techniques that use a full spectrum of visible light. Narrow-band imaging (NBI) endoscopy equipment, developed during the past decade, has better depth of penetration and increases the diagnostic sensitivity of endoscopy to characterize tissues. The NBI endoscopy uses narrow-bandwidth filters in a sequential red-green-blue illumination system and may detect superficial pharyngeal mucosal lesions.8-10 The red wavelength (540 nm) may penetrate the deepest layers and reveal dark larger vessels. The blue wavelength (415 nm) in the superficial layer shows microvascular red. The green wavelength (440 nm) is displayed between the red and blue colors. Light filtering through the filter palate controls the intensity of the 3 colors and adjusts color contrast. Therefore, the system reduces the unnecessary intermediate colors to increase the contrast of blood vessels. Angiogenesis, the formation of new blood vessels from preexisting vessels, is an important phenomenon in various physiological processes, including acute and chronic inflammation.11 Superficial mucosal lesions that usually cannot be detected by regular WL endoscopy can be identified with blue light in NBI endoscopy because of the vasculature pattern of neoangiogenesis. The inflammatory response involves dilation of capillaries to increase blood flow and microvascular structural changes.12 Vascular damage is common to inflammatory lesions following a wide variety of sublethal insults and is associated with characteristic changes in the inflamed tissues.13 Inflamed mucosa may be accompanied by dilation or proliferation of microvessels, and we may easily observe the severity of epithelial inflammation with NBI endoscopy. The relation between laryngeal findings with NBI endoscopy and reflux esophagitis is not well established. The purpose of this study was to compare the difference between
WL and NBI endoscopy in evaluating patients who had reflux laryngitis and esophagitis.
Methods Participants Consecutive patients (102 patients; mean age, 48 6 11 years) with reflux esophagitis who had office-based transnasal esophagoscopy (TNE) with WL and NBI views were included in this retrospective study. There were 60 men (59%) (mean age, 48 6 11 years) and 42 women (41%) (mean age, 47 6 11 years). We compared WL and NBI endoscopy in observing the laryngeal and esophageal epithelium. Included subjects were outpatient adult patients (age range, 21-65 years) who had endoscopically proven reflux esophagitis. Patients were excluded because of acute laryngitis, hypopharyngeal cancer, sinusitis, tonsillar hyperplasia, autoimmune diseases (xerostomia), or cervical abnormalities. The institutional ethics committee approved this study (Chang Gung Memorial Hospital Internal Review Board no. 100-1237B).
Narrow-Band Imaging System and Endoscopy The TNE with NBI system was equipped with a light source and central video system (CLV-180; Olympus Medical Systems, Tokyo, Japan). The system switched between conventional and NBI views with a button on the control section of the videoscope. All endoscopic examinations were performed in an outpatient clinic by 1 experienced otolaryngologist (W.-H.W.) who was blinded to the subject’s prior evaluation or diagnosis. The patient received simethicone (20 mg/mL; dose, 20 mL) 30 minutes before endoscopy, and the nasal cavity was anesthetized with lidocaine spray (2%). The TNE was performed with the patient in the seated position, first in the WL mode and then using the NBI system. The nasopharynx, base of the tongue, epiglottis, hypopharynx, larynx, esophagus, gastroesophageal junction, and stomach were examined, and all procedures were digitally recorded. All patients were evaluated to determine the RFS in the larynx (both using WL and NBI), the LA classification grade in the esophagus (both using WL and NBI), and the presence of a brownish area (using NBI). The TNE examinations were performed by 1 physician who was blinded to the subject’s prior diagnosis. We retrospectively reviewed the medical records of all 102 patients. A blinded reviewer read the TNE images and determined the RFS in all patients. The LA classification was used in the esophagus to avoid subjective differences and interobserver variability in image interpretation. In addition, we compared the severity of the image interpretation and association between both WL and NBI views.
Statistical Analysis We compared the outcome differences between the WL and NBI views in the larynx and esophagus using a paired t test. Spearman rank correlation was used to determine comparative relevance of laryngeal and esophageal images. Several conditions (inflammation, trauma, or infection) could cause
Downloaded from oto.sagepub.com at WESTERN OREGON UNIVERSITY on May 20, 2015
876
Otolaryngology–Head and Neck Surgery 152(5)
Figure 1. Endoscopic finding showed negative reflux (control) under white light (WL) (A) and narrow-band imaging (NBI) views (B). A patient with arytenoid erythema and diffuse edema under WL (C) and NBI views (D). A patient with a more apparent brownish area and edema (arrows) under the NBI view (F) than the WL view (E).
Table 1. Comparison of RFS under WL and NBI Views.a RFS Variable Subglottic edema with pseudosulcus Ventricular obliteration Erythema/hyperemia Vocal cord edema Diffuse laryngeal edema Posterior commissure hypertrophy Granuloma/granulation Thick endolaryngeal mucus Global RFS score
WL View
NBI View
161
0.9 6 1
P Valueb .158
2.9 6 1 2.6 6 0.9 1.760.6 2.1 6 0.6 2.2 6 0.6
2.9 6 1 3.161 1.8 6 0.6 2.2 6 0.5 2.2 6 0.6
1.000 \.001 \.001 .014 1.000
0.4 6 0.8 0.4 6 0.8 13.2 6 3.7
0.4 6 0.8 0.4 6 0.8 13.5 6 3.8
1.000 1.000 \0.001
Abbreviations: NBI, narrow-band image; RFS, reflux finding score; WL, white light. a The RFS scores are reported as mean 6 standard deviation. The figures in boldface represent a significant difference between WL and NBI. b Paired t test.
similar laryngeal reflux findings such as erythema, hyperemia, or vocal cord edema; therefore, statistical significance was defined by P .001.
Results RFS from WL and NBI Views We examined the 8 items of RFS in the larynx. In 1 patient, there was arytenoid erythema and diffuse edema noted with
WL and NBI views, in contrast with image findings of negative reflux (control) with WL and NBI views (Figure 1). We could identify the brownish area over bilateral arytenoid and interarytenoid folds. The inflamed mucosa often was accompanied by dilation or proliferation of microvessels, and we could easily observe the range and severity of epithelial inflammation with NBI views. Another patient had a brownish area and edema over bilateral arytenoids and the posterior part of the false vocal cords that were more apparent with the NBI than the WL view (Figure 1). The NBI views were more sensitive than the WL views in erythema/hyperemia, vocal cord edema, and global RFS scores (Table 1). The NBI view facilitated the identification of the erythema/hyperemia change representing dilation or proliferation of microvessels caused by epithelial inflammation. However, there was no significant advantage of the NBI compared with the WL view in detecting subglottic edema with pseudosulcus, ventricular obliteration, posterior commissure hypertrophy, granuloma/granulation, or thick endolaryngeal mucus.
Laryngeal RFS with WL and NMI Views and the LA Classification We compared the laryngeal signs of RFS between the WL and NBI views and the LA classification in the esophagus using Spearman correlation. The endoscopic images of granuloma/granulation and thick endolaryngeal mucus in the larynx with both WL and NBI views had higher correlations with the severity of LA grade (Table 2). When granuloma/ granulation and thick mucus in the larynx were identified
Downloaded from oto.sagepub.com at WESTERN OREGON UNIVERSITY on May 20, 2015
Wang and Tsai
877
Table 2. Correlation between Laryngeal Signs of RFS under WL and NBI Views and LA Classification.a RFS Parameter vs LA Classification Subglottic edema with pseudosulcus Ventricular obliteration Erythema/hyperemia Vocal cord edema Diffuse laryngeal edema Posterior commissure hypertrophy Granuloma/granulation Thick endolaryngeal mucus Global RFS score
WL View P Valueb
NBI View P Valueb
0.087 (.383)
0.112 (.2860)
0.063 0.123 0.195 0.124 0.279
0.063 (.531) 0.151 (.130) 0.166 (.096) 0.150 (.132) 0.279 (.005)
(.531) (.217) (.049) (.213) (.005)
0.440 (\.001) 0.513 (\.001) 0.168 (.092)
0.440 (\.001) 0.513 (\.001) 0.382 (\.001)
Abbreviations: LA, Los Angeles; NBI, narrow-band image; RFS, reflux finding score; WL, white light. a The values reported are the Spearman correlation coefficients (Spearman r values). The figures in boldface represent a significant association with the LA classification. b Spearman correlation test.
on endoscopy (WL or NBI views), a higher grade of esophagitis was more likely. Patients who had endoscopic findings of posterior commissure hypertrophy and granuloma with WL and NBI views typically had a higher grade of reflux esophagitis such as LA grade C with WL and NBI views (Figure 2). However, the global RFS score was significantly associated with severity of LA grade only with the NBI view (Table 2 and Figure 3).
Discussion The TNE used in this study included a flexible endoscope that was passed transnasally under topical anesthesia.14 The advantages of TNE compared with conventional esophagoscopy under general anesthesia included increased patient safety, increased practice efficiency, cost savings, and early detection of esophageal cancer.15 The use of office-based TNE by otolaryngologists is available for patients who are at risk for Barrett esophagus, esophageal adenocarcinoma, hypopharyngeal cancer extended to the esophagus, and GERD.16 The TNE can be safely performed with topical anesthesia in an office setting for diagnostic and therapeutic procedures. Previous studies showed that TNE had no major complications such as esophageal perforation, and the minor complications included epistaxis (most common), vasovagal reaction, and laryngospasm.14,17 Most patients tolerate TNE well, and patient tolerance to TNE on a 10-point scale (1, no discomfort; 10, severe discomfort) is 2.0 6 1.2 (range, 1-4).17 Acute laryngitis usually is caused by infection, but chronic laryngitis may result from acid reflux, smoking, muscular imbalance, or dehydration.18 Most (50%-60%) chronic laryngitis and chronic pharyngitis may be related to GERD.19 The most common symptoms associated with
Figure 2. Endoscopic finding for a patient showing posterior commissure hypertrophy with granuloma and Los Angeles grade C esophagitis under white light (A and C) and narrow-band imaging views (B and D).
laryngopharyngeal reflux include throat cleaning (98.3%), chronic cough (96.6%), heartburn (95.7%), and hoarseness (94.9%).20 Animal studies have shown that small amounts of gastric acid may cause marked laryngeal mucosal inflammation and edema.21 Extraesophageal gastric acid may cause vagal-mediated reflexes such as bronchospasm, laryngospasm, and cough and may cause chronic laryngitis.22 In addition to detailed medical history and laryngeal observations, diagnosis of these diseases may require a 24-hour pH recording, but this study may be limited because of high cost and reluctance of patients to perform the study. Furthermore, a negative pH study does not exclude extraesophageal reflux because reflux episodes may be intermittent.23 Therefore, we sought to improve general outpatient techniques to facilitate examination and diagnosis. Most patients with GERD have no evidence of esophageal damage with standard WL endoscopy. Nonerosive subtypes of GERD (NERD) account for 70% patients with GERD.24 Standard WL endoscopy fails to detect GERD endoscopically in 60% to 70% patients. The sensitivity of standard WL endoscopy for the diagnosis is low.25,26 A previous study showed that NBI may increase the sensitivity for the diagnosis of GERD.27 In a previous study of patients with and without GERD symptoms who completed 2 validated GERD questionnaires, the distal esophagus was examined by standard WL and NBI endoscopy, and the features observed only with NBI were compared between patients with GERD and control subjects. On multivariate analysis, increased number
Downloaded from oto.sagepub.com at WESTERN OREGON UNIVERSITY on May 20, 2015
878
Otolaryngology–Head and Neck Surgery 152(5)
Figure 3. A visual representation of the data presented in Table 2. LA, Los Angeles; NBI, narrow-band image; RFS, reflux finding score; WL, white light.
and dilation of intrapapillary capillary loops were the best predictors for diagnosing GERD.28 The maximum, minimum, and average numbers of intrapapillary capillary loops/ field were significantly greater in the GERD group than in control subjects (P \ .0001). Although the interobserver agreement for the various NBI findings was very good, the intraobserver agreement was modest.28 Another study of 3 subtypes of GERD (NERD, reflux esophagitis, and Barrett esophagus) using NBI magnifying endoscopy showed that a significantly higher proportion of patients with GERD had an increased number of intrapapillary capillary loops, microerosions, and nonround pit patterns below the squamocolumnar junction than did healthy control subjects, and the numbers of intrapapillary capillary loops/field were significantly greater in patients with GERD.29 Another comparative study of NBI and convention endoscopy in the diagnosis of GERD (107 subjects: NERD, 36 subjects; erosive reflux disease, 41 subjects; controls, 30 subjects) showed that microerosions, increased vascularity, and pit patterns at the squamocolumnar junction not seen on conventional endoscopy were well seen with NBI; interobserver agreement was good for increased vascularity (k = 0.95) and microerosions (k = 0.89) but lower for pit patterns (k = 0.59).30 Therefore, NBI views improved the endoscopic diagnosis of mucosal morphology at the squamocolumnar junction and GERD.30 A study that evaluated intra- and interobserver variations in the endoscopic scoring of esophagitis with conventional imaging with and without NBI showed that intra- and interobserver reproducibility in grading esophagitis could be improved when NBI was applied with conventional imaging because of improved depictions of small erosive foci.31 These previous studies examined the lining of the esophagus with NBI magnifying endoscopy (esophagogastroduodenoscopy with 80-fold optical magnification) to easily observe pit patterns and intrapapillary capillary loops, but these were not easily performed with TNE because of the absence of a magnification function. In a study of 111 patients with laryngopharyngeal reflux who underwent TNE by the same otolaryngologist,
biopsy-proven Barrett esophagus was observed in 13.5% patients, including 7 of 58 patients (12.1%) with WL and 8 of 53 patients (15.1%) with NBI; 3 patients (2.7%) had dysplasia on biopsy, and all 3 cases were detected with NBI (5.7%).32 It was concluded that NBI may be a useful adjunct to increase the diagnostic sensitivity of TNE by the otolaryngologist.32 Our preliminary results (data not shown) for diagnosis of GERD by TNE also showed that we could easily observe microerosions and Barrett esophagus, except pit patterns, number, and dilation of intrapapillary capillary loops, with the NBI view better than the WL view. Several studies have shown the benefit of NBI in the early detection of recurrent laryngeal and hypopharyngeal cancer and early diagnosis of laryngeal cancer.33-35 A previous study showed that the NBI laryngoscope is highly valuable for the diagnosis and treatment of laryngopharyngeal reflux disease and that laryngopharyngeal reflux disease has characteristic findings on NBI laryngoscopy, including erythema and edema between the arytenoid cartilages (71.7%); epiglottis congestion (67.4%); pharyngeal isolation/integration erythema (65.2%); pharyngeal pebble-like changes (65.2%); hypertrophy of the posterior commissure (52.2%); vocal cord erythema and edema (47.8%); vocal nodules or vocal polyps with erythema or edema (39.1%), arytenoid cartilage edema and erythema (21.7%); ventricular edema, erythema, and laryngeal ventricle disappearance (17.4%); granuloma (6.5%); ulcers (4.3%); and false vocal cord pitch (4.3%).36 Several studies have investigated the association between laryngopharyngeal reflux and GERD. In 1 study, the prevalence rate of laryngopharyngeal reflux in subjects with reflux esophagitis was 23.9%.37 Another study showed that 24-hour multichannel intraluminal impedance and pH esophageal monitoring analysis confirmed GERD diagnosis in \40% patients with a previous diagnosis of laryngopharyngeal reflux, most likely because of the low specificity of the laryngoscopic findings.38 However, the expression of Helicobacter pylori positivity and the degree of GERD correlate with laryngopharyngeal reflux, and Helicobacter pylori positivity and the degree of GERD are higher in patients with RFS 7.39
Downloaded from oto.sagepub.com at WESTERN OREGON UNIVERSITY on May 20, 2015
Wang and Tsai
879
Another study showed that the severity of laryngeal mucosal lesions in patients with GERD is significantly greater than in control patients, and a higher degree of laryngeal mucosal injury may be documented in patients for whom GERD is associated with more advanced esophageal lesions.40 Both the global RFS score and scores of all RFS parameters (except for presence of granulomatous tissue) were significantly higher in patients with GERD than in control subjects; patients in whom GERD was associated with more severe esophageal lesions (LA group B) had a significantly higher global RFS score and scores of all analyzed parameters of laryngeal injury (except subglottic edema) than individuals in whom the degree of esophageal involvement was classified as group A.40 In our study, when the RFS parameters of granuloma/granulation and thick endolaryngeal mucus were identified in the larynx with either WL or NBI endoscopy, a higher grade of esophagitis was present. The granulation and thick mucus were important predictors for GERD and erosive esophagitis. In contrast with the previous study, we observed that the global RFS score was significantly associated with severity of the LA grade only with NBI, and traditional WL endoscopy was not sufficiently reliable to reflect the severity of reflux esophagitis.40 Differences between the studies include the fewer patients with GERD in the previous (46 patients with GERD and 46 controls) than in present study (GERD, 102 patients), and the grading interpretation of GERD may be inconsistent between WL and NBI views, which may affect the calculation of the correlation with laryngopharyngeal reflux.40 The mean global RFS scores in our study were 13.2 6 3.7 in WL and 13.5 6 3.8 in NBI views, consistent with previous findings.6 The endoscopic findings in the larynx showed that NBI has higher sensitivity for detecting erythema, hyperemia, and edema in the mucosa, possibly because of the ability of NBI to increase the contrast of blood vessels and improve the identification of the neoangiogenic vasculature in inflamed mucosa. Limitations of the present study include the lack of a standard 24-hour pH recorder for the definite diagnosis of GERD. There was a lack of a patient reflux symptom index41 measurement to determine laryngeal signs and symptoms that most significantly correlated. There were no treatment data for patients who had higher scores based on NBI and who did better with antireflux treatment. Therefore, further studies with longer follow-up are required to determine the usefulness of NBI in evaluating laryngopharyngeal reflux and identifying correlations with GERD.
Conclusion Literature review showed no previous comparison of the endoscopic images of laryngopharyngeal reflux and GERD with WL and NBI views. Although previous studies showed that NBI with and without magnification may increase the sensitivity for the diagnosis of GERD and observation of microerosions and Barrett esophagus, there was limited previous study to investigate the association between reflux laryngitis and esophagitis with NBI.
In addition, the imaging technology for the improved detection of laryngopharyngeal reflux appears promising. The endoscopic findings with the NBI view permit an easier identification of the RFS parameters of laryngeal erythema/vocal cord edema, which have a stronger correlation with the severity of reflux esophagitis, than the WL view. Among the RFS variables, granuloma/granulation and thick endolaryngeal mucus might be 2 important laryngeal imaging signs for predicting the severity of reflux esophagitis with either the WL or NBI view. The importance of NBI in the relation between reflux laryngitis and GERD deserves further study. A large prospective study with more patients, reflux symptom index, and outcome data about antireflux treatment are needed for the future study of NBI in reflux esophagitis. Author Contributions Wen-Hung Wang, study concept and design, acquisition of data, analysis and interpretation of data, drafting of the manuscript, critical revision of the manuscript for important intellectual content, statistical analysis; Kai-Yu Tsai, acquisition of data, analysis and interpretation of data, drafting of the manuscript, critical revision of the manuscript for important intellectual content, statistical analysis.
Disclosures Competing interests: None. Sponsorships: None. Funding source: None.
References 1. Lee BE, Kim GH. Globus pharyngeus: a review of its etiology, diagnosis and treatment. World J Gastroenterol. 2012;18: 2462-2471. 2. Lynch CR, Wani S, Rastogi A, et al. Effect of acid-suppressive therapy on narrow band imaging findings in gastroesophageal reflux disease: a pilot study. Dis Esophagus. 2013;26:124-129. 3. Kasap E, Zeybel M, Asxık G, Ayhan S, Yu¨ceyar H. Correlation among standard endoscopy, narrow band imaging, and histopathological findings in the diagnosis of nonerosive reflux disease. J Gastrointest Liver Dis. 2011;20:127-130. 4. Sansak I, Leelakusolvong S, Charatcharoenwitthaya P, et al. Esophageal reflux episodes and endoscopic findings in laryngopharyngeal reflux patients with persistent symptoms despite acid suppression therapy. Thai J Gastroenterol. 2010;11:97-103. 5. Lundell LR, Dent J, Bennett JR, et al. Endoscopic assessment of oesophagitis: clinical and functional correlates and further validation of the Los Angeles classification. Gut. 1999;45:172-180. 6. Peter C. Belafsky, Gregory N. Postma, James A. Koufman. The validity and reliability of the reflux finding score (RFS). Laryngoscope. 2001;111:1313-1317. 7. Banerjee R, Reddy DN. Enhanced endoscopic imaging and gastroesophageal reflux disease. Indian J Gastroenterol. 2011; 30:193-200. 8. Wang WH, Lin YC, Weng HH, et al. Narrow-band imaging for diagnosing adenoid hypertrophy in adults: a simplified grading and histologic correlation. Laryngoscope. 2011;121:965-970.
Downloaded from oto.sagepub.com at WESTERN OREGON UNIVERSITY on May 20, 2015
880
Otolaryngology–Head and Neck Surgery 152(5)
9. Lin YC, Wang WH, Lee KF, et al. Value of narrow band imaging endoscopy in early mucosal head and neck cancer. Head Neck. 2012;34:1574-1579. 10. Tan NC, Herd MK, Brennan PA, Puxeddu R. The role of narrow band imaging in early detection of head and neck cancer. J Oral Maxillofac Surg. 2012;50:132-136. 11. Sua´rez Y. Microregulation of plaque neovascularization. Arterioscler Thromb Vasc Biol. 2010;30:1500-1501. 12. Young-Joon S. Oxidative Stress, Inflammation, and Health. Boca Raton, FL: CRC/Taylor & Francis; 2005. 13. Moses JM, Ebert RH, Graham RC, Brine KL. Pathogenesis of inflammation. J Exp Med. 1964;120:57-82. 14. Sabirin J, Abd Rahman M, Rajan P. Changing trends in oesophageal endoscopy: a systematic review of transnasal oesophagoscopy. ISRN Otolaryngol. 2013;2013:586973. 15. Aviv JE. Transnasal esophagoscopy: state of the art. Otolaryngol Head Neck Surg. 2006;135:616-619. 16. Banerjee R, Reddy DN. Enhanced endoscopic imaging and gastroesophageal reflux disease. Indian J Gastroenterol. 2011; 30:193-200. 17. Aviv JE, Takoudes TG, Ma G, Close LG. Office-based esophagoscopy: a preliminary report. Otolaryngol Head Neck Surg. 2001;125:170-175. 18. Hopkins C, Yousaf U, Pedersen M. Acid reflux treatment for hoarseness. Cochrane Database Syst Rev. 2006;(1):CD005054. 19. Vaezi MF. Sensitivity and specificity of reflux-attributed laryngeal lesions: experimental and clinical evidence. Am J Med. 2003;115(suppl 3A):97S-104S. 20. Vaezi MF. Laryngitis and gastroesophageal reflux disease: increasing prevalence or poor diagnostic tests. Am J Gastroenterol. 2004; 99:786-788. 21. Ludemann JP, Manoukian J, Shaw K, et al. Effects of simulated gastroesophageal reflux on the untraumatized rabbit larynx. J Otolaryngol. 1998;27:127-131. 22. Qua CS, Wong CH, Gopala K, Goh KL. Gastro-oesophageal reflux disease in chronic laryngitis: prevalence and response to acid-suppressive therapy. Aliment Pharmacol Ther. 2007;25: 287-295. 23. Richter JE. Ambulatory esophageal pH monitoring. Am J Med. 1997;103:130S-134S. 24. Azumi T, Adachi K, Furuta K, et al. Esophageal epithelial surface in patients with gastroesophageal reflux disease: an electron microscopic study. World J Gastroenterol. 2008;14:57125716. 25. Frazzoni M, DeMicheli E, Savarino V. Different patterns of oesophageal acid exposure distinguish complicated reflux disease from either erosive reflux oesophagitis or non-erosive reflux disease. Aliment Pharmacol Ther. 2003;18:1091-1098. 26. Dent J, Armstrong D, Delaney B, et al. Symptom evaluation in reflux disease: workshop background, processes, terminology recommendations, and discussion outputs. Gut. 2004;53:1-24.
27. Assirati FS, Hashimoto CL, Dib RA, Fontes LH, NavarroRodriguez T. High definition endoscopy and ‘‘narrow band imaging’’ in the diagnosis of gastroesophageal reflux disease. Arq Bras Cir Dig. 2014;27:59-65. 28. Sharma P, Wani S, Bansal A, et al. A feasibility trial of narrow band imaging endoscopy in patients with gastroesophageal reflux disease. Gastroenterology. 2007;133:454-464. 29. Lv J, Liu D, Ma SY, Zhang J. Investigation of relationships among gastroesophageal reflux disease subtypes using narrow band imaging magnifying endoscopy. World J Gastroenterol. 2013;19:8391-8397. 30. Fock KM, Teo EK, Ang TL, Tan JY, Law NM. The utility of narrow band imaging in improving the endoscopic diagnosis of gastroesophageal reflux disease. Clin Gastroenterol Hepatol. 2009;7:54-59. 31. Lee YC, Lin JT, Chiu HM, et al. Intraobserver and interobserver consistency for grading esophagitis with narrow-band imaging. Gastrointest Endosc. 2007;66:230-236. 32. O’Brien TJ, Parham K. Transnasal endoscopy: white light versus narrowband image. Ann Otol Rhinol Laryngol. 2009; 117:886-890. 33. Zabrodsky M, Lukes P, Lukesova E, Boucek J, Plzak J. The role of narrow band imaging in the detection of recurrent laryngeal and hypopharyngeal cancer after curative radiotherapy. Biomed Res Int. 2014;2014:175398. 34. Kraft M, Fostiropoulos K, Gu¨rtler N, et al. Value of narrow band imaging in the early diagnosis of laryngeal cancer [published online July 3, 2014]. Head Neck. doi:10.1002/hed.23838. 35. Piazza C, Del Bon F, Peretti G, Nicolai P. Narrow band imaging in endoscopic evaluation of the larynx. Curr Opin Otolaryngol Head Neck Surg. 2012;20:472-476. 36. He N, Si Y, Yong Y, Weng J, Qin Y. The value of narrow band imaging laryngoscope for laryngopharyngeal reflux disease. Lin Chung Er Bi Yan Hou Tou Jing Wai Ke Za Zhi. 2012;26:776-778. 37. Lai YC, Wang PC, Lin JC. Laryngopharyngeal reflux in patients with reflux esophagitis. World J Gastroenterol. 2008;14:4523-4528. 38. Bortoli N, Nacci A, Savarino E, et al. How many cases of laryngopharyngeal reflux suspected by laryngoscopy are gastroesophageal reflux disease-related? World J Gastroenterol. 2012; 18:4363-4370. 39. Tezer MS, Kockar MC, Kocxkar O, Celik A. Laryngopharyngeal reflux finding scores correlate with gastroesophageal reflux disease and Helicobacter pylori expression. Acta Otolaryngol. 2006;126:958-961. 40. Krawczyk M, Scierski W, Ryszkiel I, et al. Endoscopic evidence of reflux disease in the larynx. Acta Otolaryngol. 2014; 134:831-837. 41. Mesallam TA, Stemple JC, Sobeih TM, Elluru RG. Reflux symptom index versus reflux finding score. Ann Otol Rhinol Laryngol. 2007;116:436-440.
Downloaded from oto.sagepub.com at WESTERN OREGON UNIVERSITY on May 20, 2015
