Narrow-Band Imaging for Diagnosing Adenoid Hypertrophy in Adults: A Simplified Grading and Histologic Correlation

Transcription

1 The Laryngoscope VC 2011 The American Laryngological, Rhinological and Otological Society, Inc. Narrow-Band Imaging for Diagnosing Adenoid Hypertrophy in Adults: A Simplified Grading and Histologic Correlation Wen-Hung Wang, MD, PhD; Yen-Chun Lin, MD; Hsu-Huei Weng, MD, MPH, PhD; Kam-Fai Lee, MD Objectives/Hypothesis: To investigate the use of narrow-band imaging (NBI) endoscopy to detect the appearance of a light crest (LC) on the epithelial surface of the nasopharyngeal mucosa, which is suggested to be a distinctive endoscopic finding associated with the presence of adenoid hypertrophy. Study Design: Cross-sectional study. Methods: A total of 79 consecutive adults with a high suspicion of malignancy underwent NBI endoscopy and nasopharyngeal biopsy to validate the diagnostic accuracy of the novel endoscopic technique. The degree of correlation between the LC grading and the histologic examinations of lymphoid hyperplasia, including the number of mucin-producing cells and lymphoid follicles, was then assessed. Results: The appearance of an LC on NBI endoscopy correlated with the histologic evidence of lymphoid hyperplasia with a sensitivity of 92.1%, a specificity of 95.1%, a positive predictive value of 94.6%, a negative predictive value of 92.9%, a false-positive value of 4.9%, a false-negative value of 7.9%, and an accuracy of 93.7%. The screening performance of NBI endoscopy for the presence of adenoid hypertrophy is significantly superior to that of conventional endoscopy (P ¼.0003). The LC grading was significantly correlated with the number of mucin-producing cells and lymphoid follicles (P <.001). Conclusions: In NBI endoscopy, observation of an LC on the epithelial surface of the nasopharyngeal mucosa is a highly accurate predictor of the presence of adenoid hypertrophy. Key Words: Lymphoid hyperplasia, adenoid, nasopharyngeal carcinoma, narrow-band imaging, endoscopy. Level of Evidence: 2c Laryngoscope, 121: , 2011 INTRODUCTION Nasopharyngeal lymphoid hyperplasia (adenoid) is a vegetative mass in the posterior wall of the nasopharynx. Although adenoidal tissue undergoes regression during the adolescent period, it may present as the chief cause of nasal obstruction in adults and even mimic nasopharyngeal carcinoma (NPC). NPC is a common cancer in Southeast Asia, 1,2 and it typically affects middle-aged adults. 3 However, with routine conventional endoscopy screening, it is occasionally difficult to differentiate between NPC and adenoid in adults. According to a previous report, diagnosis of primary NPC by using conventional endoscopy resulted in a false-positive rate From the Department of Otolaryngology Head and Neck Surgery, Chang Gung Memorial Hospital at Chiayi and the Graduate Institute of Clinical Medical Sciences, Chang Gung University, College of Medicine (W.-H.W., Y.-C.L.) Taiwan; the Departments of Diagnostic Radiology (H.-H.W.), Pathology (K.-F.L.) Chang Gung Memorial Hospital at Chiayi and College of Medicine, Chang Gung University, Taiwan; and the Department of Respiratory Care, Chang Gung Institute of Technology, Taiwan (H.-H.W.). Editor s Note: This Manuscript was accepted for publication December 16, The authors have no funding, financial relationships, or conflicts of interest to disclose. Wen-Hung Wang, MD, PhD, and Yen-Chun Lin, MD, contributed equally to this report Send correspondence to Wen-Hung Wang, MD, PhD, Department of Otolaryngology Head and Neck Surgery, Chang Gung Memorial Hospital at Chiayi and College of Medicine, Chang Gung University, 6 West Sec, Chia-Pu Road, Pu-Tzu City, Chiayi County 613, Taiwan. ent.taiwan@gmail.com DOI: /lary of 38%, a false-negative rate of 0.4%, a sensitivity of 62%, a specificity of 99.6%, a positive predictive value of 98%, and a negative predictive value of 88.9%. 4 Therefore, the major limitation of the application of conventional white-light endoscopy for mass screening of NPC is a higher false-positive rate due to adenoids, which are occasionally mistaken for NPCs and lead to many nonessential biopsies. Narrow-band imaging (NBI) is a novel optical technique that enhances the diagnostic sensitivity of endoscopes for characterizing tissues by using narrowbandwidth filters in a sequential red-green-blue illumination system. The central wavelengths of each band are 415 nm and 540 nm. 5 The narrow-band blue light, which has a short wavelength (415 nm), penetrates the mucosa and highlights the superficial vasculature. Furthermore, the blue filter is designed to correspond to the peak absorption spectrum of hemoglobin and thus enhances the image of the capillary vessels on the surface mucosa. As a result, superficial mucosal lesions that usually cannot be detected by regular white-light endoscopy can be identified on the basis of their neoangiogenetic vasculature pattern by using blue light in NBI. 6,7 It is now generally accepted that NBI is of great benefit in detecting superficial mucosal lesions over the pharyngeal mucosa However, to date, no reports have documented its application to the nasopharynx. There are several possible reasons: 1) NPC is a common cancer in Southeast Asia but is relatively rare in other areas, and hence it may be difficult to recruit enough 965

2 Fig. 1. In a normal nasopharynx under narrow-band imaging (NBI) view, abundant microvessels and a reticular subepithelial capillary network pattern could be found. For lymphoid hyperplasia under NBI view, the degree of light crest was graded as none, mild, moderate, or marked, which corresponded to histologic mucin producing cell findings. *The arrows refer to a fine white line on the crests of the surface, called light crest. **The arrows refer to mucin-producing cells. patients for analysis in most other countries; and 2) unlike other pharyngeal mucosa, proliferation of lymphoid cells with extremely irregular shapes and variable sizes frequently occurs within the nasopharynx and may hinder observation. We previously reported a case of early recurrent NPC that was easily detected by NBI endoscopy. 12 Because irradiation can lead to the destruction of hypertrophied lymphoid tissue, 13 adenoid tissue is seldom found in patients with NPC after radiotherapy. Therefore, it is important to define and classify the configuration of the commonly encountered nasopharyngeal lymphoid proliferations. According to our daily clinical observation, adenoid often presents a fine white line on the crests of the nasopharyngeal epithelial surface under NBI view; this is called the light crest (LC) (Fig. 1). The aim of the present study was to examine the diagnostic value of LC as a distinctive endoscopic finding for adenoids and investigate the histologic appearance of the mucosa showing an LC. 966 MATERIALS AND METHODS A total of 79 adults (58 males, 21 females) with a mean age (standard deviation) of 52.9 (21.8) years underwent nasopharyngeal biopsy due to nasopharyngeal tumor, asymmetric nasopharyngeal wall, elevated serum Epstein-Barr virus titer, NPC family history, or neck mass of unknown primary tumor. Previously, they also underwent detailed endoscopic examination, which was performed by using both the conventional white-light and NBI systems. The ethics committee of our hospital approved the study, and written informed consent was obtained from all of the patients before the endoscopic examinations and nasopharyngeal biopsy. NBI SYSTEM AND ENDOSCOPIC PROCEDURES The NBI system used was equipped with an ENF- V2 or VQ rhinolarynx videoscope (Olympus Medical Systems, Tokyo, Japan), a light source (CLV-160B; Olympus Medical Systems), and a central video system (CV-160B; Olympus Medical Systems). A button on the control

3 section of the videoscope enabled switching between conventional and NBI views. All endoscopic examinations were performed by one experienced otolaryngologist (w.-h.w.) in an outpatient clinic. The patients were examined while in the seated position. Before the endoscopic procedure, the nasal cavity of each patient was anesthetized with a 4% lidocaine hydrochloride spray. We performed transnasal endoscopy, first in the white-light mode and then using the NBI system. The NBI images were recorded before and after biopsy, and all of the procedures were digitally videotaped to verify the biopsy site. A normal nasopharynx under NBI view can be identified as having abundant microvessels and a reticular subepithelial capillary network pattern (Fig. 1, control). LC is defined as a fine white line on the crests of the nasopharyngeal epithelial surface under NBI endoscopy. We examined all patients to determine the presence and degree of LC. The degree of LC was graded as 1) none ( ): unable to see any white fine line; 2) mild (þ): white fine line approaches the parallel direction and appears similar to falling raindrops; 3) moderate (þþ): the fine white line interlocks the furcation; or 4) marked (þþþ): fine white line forms many circles and a honeycomb shape (Fig. 1, LC). Examination of the Diagnostic Value of LC and Histologic Assessment From January 2008 through December 2009, one experienced otolaryngologist (w.-h.w.) prospectively validated the accuracy of LC for diagnosing adenoid in 79 consecutive patients, followed by nasopharyngeal biopsy due to suspected malignancy. The diagnostic accuracy of LC for adenoid was evaluated in relation to the histologic findings of the biopsy samples. All of the biopsy specimens were immersed in formalin and then embedded in paraffin. Sections cut from the paraffin blocks were stained with hematoxylin and eosin. The high-power field (HPF) was defined as the area visible under the 400 magnification level of the objective being used. The low-power field (LPF) was defined as the 100 magnification level image field. We determined the average cell numbers by examining three different HPFs or LPFs. A visual analogue scale graded as none (0/ HPF), mild (1 9/HPF), moderate (10 19/HPF), or marked (>20/HPF) was used to grade mucin-producing cells (MPCs). Another scale graded as none (0/LPF), mild (1 2/LPF), moderate (3 4/LPF), or marked (>5/LPF) was used to grade lymphoid follicles. Lymphoid hyperplasia was regarded as being present histologically when an increase in the number of lymphocytes with/without lymphoid follicles in the nasopharyngeal mucosa was seen in sections of the biopsy. To assess the diagnostic accuracy, staff members who were unaware of the endoscopic findings evaluated the histologic findings. The grades of LC and histologic examinations (MPC and lymphoid follicles) of lymphoid hyperplasia were compared. Statistical Analyses Spearman s rank correlation test was used to compare the LC grading with the histologic markers of TABLE I. Results of the Accuracy of Diagnoses of Lymphoid Hyperplasia and Non Lymphoid Hyperplasia by Means of Conventional White- Light View and Narrow Band Imaging View (Light Crest Detection) on the Basis of Histopathologic Examination. Histopathology (N ¼ 79) LH Non-LH White light view LH Non-LH 4 22 NBI view LH 35 2 Non-LH 3 39 LH ¼ lymphoid hyperplasia; Non-LH: non lymphoid hyperplasia; NBI ¼ narrow-band imaging. lymphoid hyperplasia. The sensitivity, specificity, and positive and negative predictive values for the diagnosis of lymphoid hyperplasia were calculated. Statistical differences were compared by applying the v 2 test for dichotomous variables. P values of less than.05 were considered significant. The StatView program, version 5.0 (SAS Institute, Cary, NC), was used for data analysis. RESULTS The subjects were divided into three groups on the basis of the histopathologic examination: 1) the lymphoid hyperplasia group included 38 subjects (24 males, 14 females; mean age [standard deviation], 49.2 [18.5] years) with lymphoid hyperplasia; (2) the non lymphoid hyperplasia group or control group included seven subjects (5 males, 2 females; mean age, 51.1 [22.5] years) who were negative for malignancy; and (3) the NPC group included 34 adults (29 males, 5 females; mean age, 57.9 [24.3] years) of which 13 had nonkeratinizing carcinoma, and 21 had undifferentiated carcinoma. As determined by pathology (Table I), the results of lymphoid hyperplasia diagnosis by conventional whitelight endoscopy were as follows: false positive, 35.3% (12 of 34 patients) (95% CI, 19.7%-53.5%); false negative, 8.9% (4 of 45 patients) (95% CI, 2.5%-21.2%); sensitivity, 91.1% (41 of 45 patients) (95% CI, 78.8%-97.5%); specificity, 64.7% (22 of 34 patients) (95% CI, 46.5%-80.3%); positive predictive value, 77.4% (41 of 53 patients) (95% CI, 63.8%-87.7%); negative predictive value, 84.6% (22 of 26 patients) (95% CI, 65.1%-95.6%); and accuracy, 79.7% (63 of 79 patients) (95% CI, 69.2%-87.9%). Under the NBI view, the appearance of LC correlated with histologic evidence of lymphoid hyperplasia with a sensitivity of 92.1% (95% CI, 84.6%-99.8%), a specificity of 95.1% (95% CI, 81.5%-98.4%), a positive predictive value of 94.6% (95% CI, 77.6%-98.1%), a negative predictive value of 92.9% (95% CI, 87.6%-99.8%), a false positive value of 4.9% (95% CI, 1.6%-18.5%), a false negative value of 7.9% (95% CI, 0.2%-15.4%), and an accuracy of 93.7% (95% CI, 87.5%-98.6%). The screening performance of NBI endoscopy for the presence of lymphoid hyperplasia is significantly superior to that of 967

4 TABLE II. Screening Performance and the Corresponding 95% Confidence Intervals of Standard Endoscopy With and Without the Adjuncts of Narrow-Band Imaging. Modalities Sensitivity, % (95% CI) Specificity, % (95% CI) PPV, % (95% CI) NPV, % (95% CI) Accuracy, % (95% CI) SE 91.1 ( ) 64.7 ( ) 77.4 ( ) 84.6 ( ) 79.7 ( ) SE þ NBI 92.1 ( ) 95.1 ( ) 94.6 ( ) 92.9 ( ) 93.7( ) FP, % (95% CI) FN, % (95% CI) P value SE 35.3 ( ) 8.9 ( ).0003* SE þ NBI 4.9 ( ) 7.9 ( ) *Comparison between SE and SE þ NBI, significantly different when P <.05. CI ¼ confidence interval; PPV ¼ positive predictive value; NPV ¼ negative predictive value; SE¼ standard endoscopy; NBI ¼ narrow-band imaging; FP ¼ false positive; FN ¼ false negative. conventional endoscopy (P ¼.0003) (Table II). On the other hand, there were no signs of LC found on the surface of NPC (Fig. 2E). None of the patients with cancer-positive biopsies had LC. None of the patients with LC later developed cancer during at least 2 years of follow-up. Under histologic assessment, LC grading demonstrated a significantly positive correlation with the number of MPCs (P <.001) (Fig. 1 and Fig. 2D) and lymphoid follicles (P <.001) (Table III); however, both of these markers were difficult to find in the epithelium of NPC (Fig. 2F). On the other hand, when we calculated Fig. 2. Gross and microscopic findings of (A, B) normal nasopharynx, (C, D) lymphoid hyperplasia, and (E, F) nasopharyngeal carcinoma under narrow-band imaging view and histologic examination. (D) There are histologically abundant mucin-producing cells found in the epithelium of rich lymphoid proliferation. (F) The epithelium of nasopharyngeal carcinoma shows dysplastic change with no mucin-producing cells found. 968

5 TABLE III. Relationship Between Light Crest Grading and Numbers of Histologic Mucin-Producing Cells (4003) and Lymphoid Follicles (1003). Hematoxylin and Eosin Staining Grade Non-LC Mucin-producing cells Lymphoid follicles LC þ þþ þþþ None(0/HPF ) <.001 Mild(1 9/HPF) Moderate(10 19/HPF) Marked(>20/HPF) Total mean number of cells None(0/LPF ) <.001 Mild(1 2/LPF) Moderate(3 4/LPF) 4 3 Marked(>5/LPF) 4 Total mean number of cells P* *P values by Spearman rank correlation for the grades of light crest and numbers of histologic mucin-producing cells and lymphoid follicles. HPF refers to the area visible under the 400 magnification level of the objective being used. The numbers of histologic mucin-producing cells were calculated under HPF. LPF refers to 100 magnification level. The numbers of histologic lymphoid follicles were calculated under LPF. LC ¼ light crest; HPF ¼ high-power field; LPF ¼ low-power field. the mean number of cells corresponding to each level of LC grading by the total sum of the number of cells for each case divided by the total sum of the number of cases, the mean number of MPCs under HPF corresponding to LC grading was 0.08 for LC ( ), 1.33 for LC (þ), 1.54 for LC (þþ), and 2.33 for LC (þþþ); the mean number of lymphoid follicles under LPF corresponding to LC grading was for LC ( ), 0.67 for LC (þ), for LC (þþ), and 1.75 for LC (þþþ) (Fig. 3). The results showed that the number of MPCs and lymphoid follicles increased as the LC grade increased. DISCUSSION In clinical practice, the common diagnostic modalities for nasopharyngeal masses include endoscopy and radiology. Bohman et al. 14 investigated the computed tomography (CT) approach by using the physical characteristics of the fascial planes of the nasopharynx to provide a basis to categorize growth patterns of the more common benign nasopharyngeal masses. The authors concluded that lymphoid hyperplasias are confined to the surface by the very dense pharyngobasilar fascia that lies beneath the submucosa. It takes a very aggressive process to cross this fascial plane and move laterally throughout the paranasopharyngeal space; the loose areolar nature of the buccopharyngeal fascia permits benign tumors in this space to assume a spherical configuration. However, using CT to screen every patient with a possible benign nasopharyngeal mass is inconvenient and costly and necessitates radiation exposure. Conversely, fiberoptic endoscopic examination is a convenient, timesaving screening tool in daily outpatient clinics. Our results of lymphoid hyperplasia diagnosis by conventional white-light endoscopy showed a sensitivity of 91.1%, a specificity of 64.7%, and an accuracy of 79.7%. The 12 false-positive cases diagnosed with white light were mainly due to the inability to identify early NPC lesions that might be mistaken for adenoids. Because we were able to easily identify the fine structure and appearance of LC on the nasopharyngeal mucosa by means of NBI endoscopy, we could reduce the number of false-positive cases (Table II); specificity and accuracy in diagnosing adenoids were 92.1% and 93.7%, respectively. Moreover, the four false-negative cases diagnosed by means of white light were due to lymphoid hyperplasias of a large size or with easy touch bleeding that were mistaken as NPC. The three false-negative cases diagnosed by means of NBI were due to abnormal vessels Fig. 3. Comparison of the correlation between light crest grading and numbers of histologic mucin-producing cells (mean numbers were calculated under high-power field, 400) and lymphoid follicles (mean numbers were calculated under low-power field, 100). MPC ¼ mucin-producing cells; LF ¼ lymphoid follicles. 969

6 on the lymphoid tissue surface that were mistaken for NPC. Earlier studies that used the NBI system with magnifying endoscopy in the gastric mucosa showed that the appearance of a light blue crest (LBC) in the mucosa is a distinctive endoscopic finding suggesting an increased likelihood of detecting intestinal metaplasia, that is, paraneoplastic lesions in the stomach. 15 The LBC was defined as a fine, blue-white line on the crests of the epithelial surface or gyri as visualized by magnification endoscopy with NBI. This appearance is speculated to be caused by the reflection of the short and narrow wavelength light ( nm) at the surface of the ciliated tissue structure. 15 From our observation, it is interesting that nasopharyngeal lymphoid proliferation often presents a similar fine white line on the crests of the nasopharyngeal epithelial surface under NBI view. Lines of similar appearance, which were more irregular with branching or linear distribution (Fig. 1) than the LBC, were found in most of the adenoid group (92.1%) and some of the control group (28.6%) but none of the NPC group (0%). We think that the LCs were caused by the reflection of light at the surface of the adenoid, and, when lymphoid hyperplasia was destroyed by NPC, the LCs might disappear. Generally, the adenoidal tissue undergoes regression during the adolescent period. According to a previous study, 16 the involution of the adenoid is associated with a decreased germinal center reaction and relative accumulation of mature B cells in the adenoidal tissue, as analyzed by three-color flow cytometry. Our preliminary results also revealed fewer histologic MPCs/lymphoid follicles under histologic examination and fewer white line furcation distributions under NBI view on the surface of the adenoid in adults as compared with children. However, we could not reach a conclusion about the age-associated changes in the histologic epithelium composition of the human adenoid because only a few cases of adenoid in children were obtained. This is the first study to document the utility of NBI screening as an adjunctive technique to investigate the reliability of NBI endoscopy for nasopharyngeal lymphoid hyperplasia in adults. NBI significantly improves the specificity of lymphoid hyperplasia in the nasopharyngeal mucosa by detecting the LC pattern. We strongly suggest that NBI be used for routine surveillance of the nasopharyngeal mucosa in high-risk populations to reduce the rate of nonessential biopsy. CONCLUSION In NBI endoscopy, observation of an LC on the epithelial surface in the nasopharyngeal mucosa is a highly accurate predictor of the presence of histologic lymphoid hyperplasia and may be treated as a distinctive endoscopic finding. The cost of availability for the system would be reasonable and meaningful when avoiding many nonessential biopsies. However, a large-scale prospective study with greater case numbers is still needed to verify the accuracy of NBI in lymphoid hyperplasia detection in the future. BIBLIOGRAPHY 1. WHO International Agency for Research on Cancer. Epstein-Barr virus. In: IARC Monographs on the Evaluation of Carcinogenic Risks in Humans, publ. 70. Lyon, France: IARC Press; 1997: Department of Health. Leading causes of death from cancer in the Taiwan area from 1991 to Health and Vital Statistics of the Republic of China. Taipei, Republic of China: Department of Health; 1997: Chen YP, Tsang NM, Tseng CK, Chen WC, Leung WM, Tang SGJ. Nasopharyngeal Cancer Registry in sixteen years. Ther Radiol Oncol 1996;3: Hao SP, Tsang NM, Chang KP. Screening nasopharyngeal carcinoma by detection of the latent membrane protein 1 (LMP-1) gene with nasopharyngeal swabs. Cancer 2003;97: Gono K, Yamazaki K, Doguchi N, et al. Endoscopic observation of tissue by narrowband illumination. Opt Rev 2003;10: Piazza C, Dessouky O, Peretti G, Cocco D, De Benedetto L, Nicolai P. Narrow-band imaging: a new tool for evaluation of head and neck squamous cell carcinomas. Review of the literature. Acta Otorhinoparyngol Ital 2008;28: Gono K, Obi T, Yamaguchi M, et al. Appearance of enhanced tissue features in narrow-band endoscopic imaging. J Biomed Opt 2004;9: Muto M, Nakane M, Katada C, et al. Squamous cell carcinoma in situ at oropharyngeal and hypopharyngeal mucosal sites. Cancer 2004;101: Watanabe A, Tsujie H, Taniguchi M, Hosokawa M, Fujita M, Sasaki S. Laryngoscopic detection of pharyngeal carcinoma in situ with narrowband imaging. Laryngoscope 2006;116: Nonaka S, Saito Y. Endoscopic diagnosis of pharyngeal carcinoma by NBI. Endoscopy 2008;40: Ugumori T, Muto M, Hayashi R, Hayashi T, Kishimoto S. Prospective study of early detection of pharyngeal superficial carcinoma with the narrowband imaging laryngoscope. Head Neck 2009;31: Lin YC, Wang WH. Narrow band imaging for detecting early recurrent nasopharyngeal carcinoma. Head Neck 2009 Dec 22. In press. 13. Irwin JB. Irradiation treatment of lymphoid hyperplasia of the nasopharynx. Calif Med 1951;74: Bohman L, Mancuso A, Thompson J, Hanafee W. CT approach to benign nasopharyngeal masses. AJR Am J Roentgenol 1981;136: Uedo N, Ishihara R, Iishi H, et al. New diagnostic modality of gastric intestinal metaplasia: narrow band imaging system with magnifying endoscopy [abstract]. Gastrointest Endosc 2005;61:AB Mattila PS, Tarkkanen J. Age-associated changes in the cellular composition of the human adenoid. Scand J Immunol 1997;45: