Effective staining method with iodine for leukoplakia and lesions surrounding squamous cell carcinomas of the tongue assessed by colorimetric analysis

Transcription

1 J Med Dent Sci 2009; 56: Original Article Effective staining method with iodine for leukoplakia and lesions surrounding squamous cell carcinomas of the tongue assessed by colorimetric analysis Keiko Maeda 1, Masashi Yamashiro 1, Yasuyuki Michi 1, Tetsuo Suzuki 1, Yoshio Ohyama 1, Norihiko Okada 2, and Teruo Amagasa 1, 1) Maxillofacial Surgery, Maxillofacial Reconstruction and Function, Division of Maxillofacial and Neck Reconstruction, Graduate School, Tokyo Medical and Dental University, Japan 2) Diagnostic Oral Pathology, Oral Restitution, Oral Health Science, Graduate School, Tokyo Medical and Dental University, Tokyo, Japan To determine whether staining with iodine solution provides an efficient criterion for determining the area of resection for the lesions surrounding squamous cell carcinoma (SCC) and leukoplakia of the tongue, we determined the optimum density of iodine solution and staining procedure and analyzed the color of lightly stained lesions (LSLs) in relation to the histopathologic findings. Sixtyfive patients with SCC or leukoplakia of the tongue were divided into two groups: lesions stained with 3% Lugol solution and restained with either 5% Lugol solution (n=38) or 10% iodine glycerin (n=27). Among the lesions stained with 5% Lugol solution, significant differences were found in all color values. Color difference values (ΔE*ab) using 3% and 5% Lugol solutions were significantly different between epithelial hyperplasia/mild epithelial dysplasia and moderate to severe dysplasia (P < 0.05). According to the evaluations of five clinicians in 46 LSLs, a distinctive boundary was most often obtained using 5% Lugol solution. These results suggest that the most effective method for obtaining a clear boundary and distinguishing moderate to severe dysplasia from mild or no epithelial dysplasia according to Corresponding Author:Keiko Maeda Maxillofacial Surgery, Maxillofacial Reconstruction and Function, Division of Maxillofacial and Neck Reconstruction, Graduate School, Tokyo Medical and Dental University, Tokyo, Japan Yushima, Bunkyo-ku, Tokyo , Japan TEL:+81 (3) FAX:+81 (3) keimaeda0209.mfs@tmd.ac.jp Received May 7;Accepted September 18, 2009 the measured color value was to stain with 3% followed by 5% Lugol solution. Key words: squamous cell carcinoma of the tongue; leukoplakia of the tongue; vital iodine staining; iodine lightly stained lesions; color measurement Introduction Vital iodine staining, known as the Schiller iodine test, was first performed to detect early cervical cancer in the 1930s, and this technique can be used immediately before resection as a useful and minimally invasive test to determine the area of resection for esophageal cancer 1,2. Many studies have also reported the usefulness of iodine staining of oral lesions using current detection techniques such as narrow band imaging (NBI) and magnifying endoscopy, which are routinely used by clinicians 3-6. However, there is still no consensus on the optimal density or staining sequence of iodine solutions for oral lesions.in this study, we performed vital iodine staining with 3% Lugol solution followed by either 5% Lugol solution or 10% iodine glycerin and determined the density and staining sequence of iodine solutions that provided the clearest boundary for lightly stained lesions (LSLs) in order to more clearly determine the area of resection of squamous cell carcinoma (SCC) with the surrounding lesions or leukoplakia of the tongue. The color and shape of the LSLs were often irregular as viewed subjectively by each operator; however, our method could distinguish each histopathologic group of LSLs using differences in the objective values of lightness

2 124 K.Maeda et al. J Med Dent Sci Figure 1 : A : A case showing staining with 3% and 5% Lugol solutions (SCC, T2N0M0) (a) Before staining. Lesions that were clinically deemed cancerous were marked with a dotted line using methylrosanilium chloride solution (Pyoktanin Blue ), (b) stained with 3% Lugol solution, (c) stained with 5% Lugol solution. The histopathologic diagnosis of the lightly stained lesions was mild epithelial dysplasia (lesions marked with*). B: A case showing staining with 3% Lugol solution and 10% iodine glycerin (SCC, T2N0M0) (a) Before staining, (b) stained with 3% Lugol solution, (c) stained with 10% iodine glycerin The histopathologic diagnosis of the lightly stained lesions was moderate epithelial dysplasia (lesions marked with*). (L *) and hue (a *, b *). Here, we evaluated the color of LSLs by colorimetry and determined the relationships between the color of LSLs and the histopathologic diagnosis, density of iodine solution, and sequence of staining. The values obtained were compared with those derived by subjective clinical judgment. Patients and methods Patients Sixty-five patients (31 men and 34 women) who had conditions diagnosed as SCC (n=45) or leukoplakia of the tongue (n=20) and who underwent biopsy under local anesthesia or surgery under general anesthesia at the Maxillofacial Surgery, Tokyo Medical and Dental University Hospital Faculty of Dentistry (Tokyo, Japan), from July 2007 to April 2009 were included in this study. This study was approved by the Ethics Review Board of the Faculty of Dentistry (No. 277). Details of the study were explained to the patients, and informed consent was obtained prior to the start of the investigation. The mean age of the patients was 59.2 years (range: years). The selection criteria were as follows: patients with SCC whose lesions were classified as T1 (n=23) or T2 (n=22); patients with leukoplakia who underwent biopsy for definite diagnosis or were selected for surgical treatment; and patients whose lesions could be measured and illuminated by an artificial sunlight lamp. The patients were divided into two groups: one group underwent staining of the lesions with 3% followed by 5% Lugol solution (n=38), and the other group underwent staining using 3% Lugol solution followed by 10% iodine glycerin (IG) (n=27). Staining solutions and procedures Solutions of 3% Lugol, 5% Lugol, and 10% IG were used for staining according to the following procedure. First, the untreated lesion was photographed, and a colorimeter (CS-100, KONICA MINOLTA HOLDINGS, INC., Tokyo, Japan) was calibrated in the oral cavity. Then, 3% Lugol solution was applied, and the area was rinsed with water and dried with gauze. Color measurement was performed after 2 minutes, then the LSLs were photographed and sodium thiosulfate was applied, and restaining with 5% Lugol or 10% IG was performed using the same steps (Fig. 1). The

3 Effective iodine staining method for oral lesions 125 Figure 2 : (a) Before staining (SCC, T2N0M0). (b) Stained with 5% Lugol solution. The color measuring point is indicated by *. (c) Before resection. The incision line and boundary line between SLs and LSLs are indicated as a solid line and a dotted line using methylrosanilium chloride solution (Pyoktanin Blue ), respectively. The color measuring point is indicated by *. (d) Re-stained specimen with 5% Lugol solution. The boundary line after resection corresponded to the dotted line marked before resection. A shallow incision (yellow line) was made with a scalpel into the boundary line between LSLs and SLs, and the areas to be measured were marked with a pin (white point). (e) Specimen fixed in 10% buffered formalin. Sections of the specimen were made through the shallow incision line (arrow) and pin hole (white dotted line). (f) Arrow indicates the shallow incision line in the specimen showing the macroscopical boundary line between SLs and LSLs (H-E staining 40). Left side of the incised line shows LSLs. boundary line between stained lesions (SLs) and LSLs after staining was marked with a dotted line using methylrosanilium chloride solution (Pyoktanin Blue ) (Fig. 2abc) Color measurement Color was measured using CS-100 according to the following method described by Yamashiro7: an artificial sunlight lamp (SOLAX XC-100, SERIC LTD., Tokyo, Japan) was used to illuminate the lesion at a 45 angle and a distance of 1 m, and color was measured in the vertical direction. A close-up lens No. 122 (measurement diameter: mm) was attached to the colorimeter, and color was measured within the wavelength range of mm. Numerical values were recorded using the data processor DP-101 (KONICA MINOLTA HOLDINGS, INC.). Areas for color measurement in LSLs were determined on the basis of the following criteria: 1) lesions deemed to surround tumorous lesions or leukoplakia clinically; 2) the size of the LSL was more than 5 mm (measurement diameters of close-up lens: mm); and 3) areas within approximately 10 mm from the clearer boundary line between the LSLs and SLs. Moreover, when the LSLs were small, one area was selected as the measuring points, whereas when the LSLs were large, several lesions were selected. In each test, five measurements were taken to eliminate marked variations, and an average value was calculated. For the light source, D55 was used and data were calculated in the Y, x, y mode. Using a dental mirror-type standard white board, an arbitrary standard calibration must be carried out for each measurement under the same conditions as those used for the test areas. The CIE L*a*b* color system was used to describe the developed color, as recommended by the Commission Internationale dl Eclairage in In this system, L* indicates lightness, and a* and b* indicate hue. The following formulas are used for converting Y, x, y into L*, a*, b * 8:

4 126 K.Maeda et al. J Med Dent Sci L* = 116 (Y/Yn) 1/3-16 a* = 500 [(X/Xn) 1/3 - (Y/Yn) 1/3 ] b* = 200 [(Y/Yn) 1/3 - (Z/Zn) 1/3 ] Color differences refer to the quantitative differences in color perception and were calculated using the following formula: ΔE*ab = [(ΔL*) 2 + (Δa*) 2 + (Δb*) 2 ] 1/2 Tissue processing and evaluation Resected specimens were re-stained, and it was confirmed that the boundary line between SLs and LSLs after resection corresponded to the dotted line marked before resection using methylrosanilium chloride solution (Pyoktanin Blue ) (Fig. 2d). A shallow incision was made with a scalpel into the boundary line between LSLs and SLs and the areas to be measured were marked with a pin and then the specimens were fixed in 10% buffered formalin. Tissue sections (4 μ m thick) were prepared and stained with H-E by routine methods. Sections of the specimens were made through the shallow incision line and pin hole (Fig. 2e), and all of the final histopathologic diagnoses were made on the basis of these stained sections (Fig. 2f). For histopathologic evaluation of LSLs, SCC and epithelial dysplasia were classified as mild, moderate, or severe according to World Health Organization (WHO) standards 9, and if atypical findings were not observed, those with hyperkeratosis and acanthosis of the simple elongated epithelial layer were grouped into epithelial hyperplasia (HYP). When there were measured areas of LSLs that differed from the histopathologic diagnosis, the more severe histopathologic diagnosis of the lesion was taken as the final diagnosis. Clinical evaluation and statistical analysis Clinical staging for cancer of the tongue was performed according to the International Union Against Cancer 10, and leukoplakia was classified according to WHO criteria 9. Clinical evaluations as to the clarity of the boundary line between SLs and LSLs were scored as follows: Category Ⅰ, unclear boundary line; Category Ⅱ, partially clear boundary line; Category Ⅲ, clear boundary line all around the lesion. When the clinical evaluations did not coincide among three or more of five clinicians, the clarity of the boundary line was classified as None. Scores that coincided among three or more of five clinicians evaluations were used as the average score, and an evaluation-concordance ratio (%) was defined as the average evaluationconcordance over four of five clinicians evaluations in this study. Color difference (ΔE*ab) was assessed on the basis of National Bureau of Standards (NBS) units, as reported by Koksal 11 and Seghi 12. The criteria for categorization were as follows: trace (0-0.5), slight ( ), noticeable ( ), appreciable ( ), strong ( ), and very strong (> 12.0). It was possible for five oral surgeons having more than 10 years of clinical experience to evaluate the clarity of the boundary lines with digital photography in 46 cases (5% Lugol solution: 26 cases; 10% IG: 20 cases). For the statistical analysis, the cases were divided into three groups according to the different densities of iodine solution: 3% Lugol, 5% Lugol, and 10% IG. Data were analyzed using SPSS 16.0 J statistical software (SPSS Inc., Chicago, IL, USA), and values were expressed as mean ± standard deviation. P-values of the color values (L*, a*, b*) and color differences (ΔE*ab) are from Dunnett s multiple comparison test. Color values were compared between SLs and each histopathologic group of LSLs, and color differences were compared between moderate to severe epithelial dysplasia (MOD/SED) and HYP or mild epithelial dysplasia (MLD). Fisher s exact test was used for analysis of clinical evaluation of the clarity of the boundary line of the LSLs. The level of significance was set at P < 0.05 (two-sided). Results Classification of LSLs LSLs were observed in 57 of 65 patients (87.7%) using 3% Lugol solution alone. The histopathologic diagnoses for the 57 measured areas with detected LSLs were HYP in 7 patients (12.3%), MLD in 16 patients (28.1%), and MOD/SED in 34 patients (59.6%). LSLs were not observed in 8 patients; the histopathologic diagnoses for these 8 patients were SCC without surrounding lesions in 6 patients, leukoplakia and ulcer in 1 patient each. When staining with 5% after 3% Lugol solution, LSLs were observed in 35 of 38 patients (92.1%). The histopathologic diagnoses for the 35 patients with detected LSLs were HYP in 4 patients (11.4%), MLD in 11 patients (31.4%), and MOD/SED in 20 patients (57.1%). LSLs were not observed in 3 patients, and the histopathologic diagnosis was SCC in all three. When staining with 10% IG after 3% Lugol solution, LSLs were observed in 26 of 27 patients (96.3%). The histopathologic diagnoses for the 26 patients with detected LSLs were HYP in 3 patients (11.5%), MLD in 7 patients (26.9%), and MOD/SED in 16 patients

5 Effective iodine staining method for oral lesions 127 Table 1. Color values and differences after staining (61.5%). LSLs were not observed in 1 patient and the histopathologic diagnosis was SCC. LSLs color values The results of the color values and differences in the colors of the LSLs are summarized in Table 1. P-values of the color values (L*, a*, b*) and color differences (ΔE*ab) are from Dunnett s multiple comparison test. Color values were compared between SLs and each histopathologic group of LSLs, and the color differences (ΔE*ab) were compared between MOD/SED and HYP or MLD. After staining with 3% Lugol solution, the L* (lightness) values in each histopathologic group of LSLs and the b*value in the MOD/SED group showed significant differences compared with the SLs (P < 0.05). When staining using 5% after 3% Lugol solution, all color values (L*, a*, b*) showed a significant difference between SLs and each histopathologic group of LSLs, whereas when staining using 10% IG after 3% Lugol solution, there was no significant difference except in the L* value between the MOD/SED group of LSLs and SLs. The mean values of color difference (ΔE*ab) between SLs and HYP or MLD using 3% Lugol solution were 6.39 and 8.86, respectively, and that between SLs and HYP using 5% after 3% Lugol solution was 11.36, and these values were assessed as strong according to the NBS units. Other color difference values were >more than 12.0 and were assessed as very strong. When using the 5% after 3% Lugol solutions, there were significant differences between the MOD/SED and MLD groups (P = and P = 0.042, respectively) and the HYP group (P = and P = 0.036, respectively), whereas with 10% IG after 3% Lugol solution, there was no significant difference between the MOD/SED and MLD groups (P = 0.818) or between the MOD/SED and HYP groups (P = 0.328). Clinical evaluation of LSLs The clinical evaluations of the clarity of the boundary lines between LSLs and SLs were scored as follows: with 3% Lugol solution there were 5 cases (10.9%) in category Ⅰ, 12 cases (26.1%) in category Ⅱ, 23 cases (50.0%) in category Ⅲ, and 6 cases (13.0%)

6 128 K.Maeda et al. J Med Dent Sci on which the clinicians could not agree. With the 5% Lugol solution, 2 cases (7.7 %) were in category Ⅰ, 3 cases (11.5%) in category Ⅱ, and 21 cases (80.8%) in category Ⅲ. After staining with 10% IG, 1 case (5.0%) was in category Ⅰ, 2 cases (10.0%) in category Ⅱ, and 17 cases (85.0%) in category Ⅲ. When staining with 3% Lugol solution followed by 5% Lugol solution or 10% IG was compared with staining with 3% Lugol solution, the clarity of the boundary line in category Ⅲ was increased in 34.6% (9/26) and 30.0% (6/20) of the cases, respectively, after the second staining (Table 2). Six lesions stained with 3% Lugol solution fell into the category of None for clarity. The evaluation-concordance ratios (%) among 4 of 5 clinicians were 45.7% (21/46 cases) for 3% Lugol solution, 73.1% (19/26 cases) for 5% Lugol solution, Table 2. Clinical evaluation of LSLs a and 70.0% (14/20 cases) for 10% IG. P-values were determined using two-tailed Fisher s exact test, and significant differences were found between 3% Lugol solution and 5% Lugol solution/ 10% IG groups (P < 0.001). Discussion Most studies of vital iodine staining using endoscopy in the upper gastrointestinal tract used 1-3% iodine solutions, whereas 2-3% Lugol solutions or 10% IG has been used for oral lesions. The densities and compositions of iodine solutions vary among authors (Table 3) 3-6, The technical term of unstained lesions (USLs) was used in previous reports, but the regions called USLs were lightly stained by iodine in our colorimetric study. Therefore, we termed unstained lesions as lightly stained lesions. To the best of our knowledge, although malignant lesions have been detected and LSLs have been histopathologically classified, there have been few studies objectively quantifying the color of LSLs and comparing different densities of iodine solutions and staining sequences to optimize the clarity of staining of LSLs 21,22. In this study, significant differences were found in the L* values between MOD/SED and SLs with each density of iodine solution. When stained with 3% Lugol solutions or 5% after 3% Lugol solutions, there were significant differences between the color difference (ΔE*ab) values of MLD and MOD/SED, but none was found in staining with 10% IG after 3% Lugol solutions. When staining with 5% Lugol solution, significant differences were found in all color values and differences. These results suggest that it might be possible to distinguish each histopathologic group of LSLs on the basis of the values of lightness (L*) and hue (a*, b*) by staining with 3% Lugol solution followed by 5% Lugol solution. In contrast, the use of 10% IG would not produce a distinguishable color difference between each histopathologic group of LSLs and SLs because of the very strong black-brown staining. The number of cases with a clear boundary line increased by 30% after the second staining. Such an increase in clarity would improve the evaluationconcordance ratios and provide a more accurate area of minimum resection. One of the probable reasons for the increased clarity might be that the shapes of the LSLs were roughly visible after the first staining, and staining for the second time enhanced the sharpness of the boundary line. The evaluation-concordance ratios (%) of the boundary lines using 3% Lugol solution, 5% Lugol solution, and 10% IG were 45.7%, 73.1%, and 70.0%, respectively. Staining with 5% Lugol solution had the highest concordance ratio, and it was clear that the staining sequence with different densities of iodine reduced the inconsistency of the evaluations by each clinician regarding the determinations for resection. Kurita et al. have reported favorable staining with 10% IG 16. However, in our experiments, LSLs are stained strongly dark over time with 10% IG, making it difficult to differentiate LSLs from SLs, and it is often difficult to draw a resection line without bleaching. Moreover,

7 Effective iodine staining method for oral lesions 129 Table 3. Reports of incidence of LSLs* by vital staining when iodine staining is performed in patients, high-dose iodine and glycerin remain in the mucosa and even after gargling with sodium thiosulfate and water, many patients complain of irritation and pain, except when the patients are under general anesthesia. Iodine staining should be avoided in conscious patients with wounds, such as ulcers, or after radiotherapy. When iodine staining is performed macroscopically in the oral cavity without an endoscope, and if the LSLs borders are not clearly seen using 3% Lugol solution, it is necessary to sequentially increase the iodine concentration. The rates of malignant transformation of oral leukoplakia were reported to be 7.9% (n = 444) in our institution over the last 30 years 23, and Petti reported annual malignant transformation rates from 0.69% to 2.03% (95% CI) 24. In order to reduce recurrence and malignant transformation when resection is selected as the treatment for oral leukoplakia, it is important to resect the LSLs with epithelial dysplasia with a sufficient safety margin. The oral cavity has various functions such as mastication, speech, and swallowing and is also esthetically important. Therefore, it is desirable to minimize resection in order to obtain a favorable postoperative quality of life. However, there is no appropriate standard for the distance between the surgical margin and the border of the LSLs. The results of this study indicate that the evaluation of the boundary line between SLs and LSLs will differ by each clinician if it is indistinct after iodine staining, and that the evaluation-concordance ratio will increase if the boundary line is distinct. From the above results, it may be concluded that the most distinct boundary line could be observed by restaining with 5% Lugol solution after staining with 3% Lugol solution. Consequently, it is expected that this method will minimize the deviation in excision range among clinicians. Moreover, this method may enable clinicians to distinguish the MOD/SED group from the MLD or HYP group in LSLs. References 1. Schiller W. Early diagnosis of carcinoma of the cervix. Surg Gynecol Obstet 1933;56: Voegeli R. Schiller s iodine test in the diagnosis of esophageal disease. Preliminary report. Pract Otorhinolaryngol 1966;28: Kerawara CJ, Beale V, Reed M. The role of vital tissue staining in the marginal control of oral squamous cell carcinoma. Int J Oral and Maxillofac Surg 1999;29: Yajima Y, Noma H, Furuya Y, et al. Quantification of telomerase activity of regions unstained with iodine solution that surround oral squamous cell carcinoma. Oral Oncol 2004;40: Yokoo K, Noma H, Inoue T, et al. Cell proliferation and tumour suppressor gene expression in iodine unstained area surrounding oral squamous cell carcinoma. Int J Oral Maxillofac Surg 2004;33:75-83.

8 130 K.Maeda et al. J Med Dent Sci 6. Nomura T, Matsubara S, Ro Y, et al. Usefulness of vital staining with iodine solution in resection of early tongue carcinoma. J Jpn Stomatol Soc 2008;57: (in Japanese) 7. Yamashiro M. A study on colorimetry of oral mucosal lesions. J stmatol Jpn 1996; 63(1): (in Japanese) 8. Hunt R.W.G.MEASURING COLOUR.3rd ed. Kingston-uponthemes, England: Fountain Press; 1998.p Pindborg JJ, Reichart PA, Smith CJ, et al. World Health Organization International Histological Classification of Tumours. Histological typing of cancer and precancer of the oral mucosa. Berlin: Springer; Sobin LH, Wittekind C. TMN classification of Malignant Tumors. 6th ed. New York: John Wiley & Sons; Koksal T, Dikbas I. Color stability of different denture teeth materials against various staining agents. Dent Mater J 2008;27: Seghi RR, Hewlett ER, Kim J. Visual and instrumental colorimetric assessments of small color differences on translucent dental porcelain. J Dent Res 1989;68: Shiozaki H, Yano H, Yano T, et al. Endoscopic screening of early esophageal cancer with the lugol dye method in patients with head and neck cancers. Cancer 1990;66: Epstein JB, Scully C, Spinelli J. Toluidine blue and Lugol's iodine application in the assessment of oral malignant disease and lesions at risk of malignancy. J Oral Pathol Med 1992;21: Ina H, Shibuya H, Ohashi I, et al. The frequency of a concomitant early esophageal cancer in male patients with oral and oropharyngeal cancer. Cancer 1994;73: Kurita K, Kurashina K. Vital staining with iodine solution in delineating the border of oral dysplasia. Oral Surg Oral Med Oral Pathol Oral Radiol Oral Endod 1996;81: Nakanishi Y, Ochiai A, Yoshimura K. The clinicopathologic significance of small areas unstained by lugol's iodine in the mucosa surrounding resected esophageal carcinoma. Cancer 1998;82: Dawsey SM, Fleischer DE, Wang GQ, et al. Mucosal iodine staining improves endoscopic visualization of squamous dysplasia and squamous cell carcinoma of the esophagus in Linxian, China. Cancer 1998;83: Freitag CP, Barros SG, Kruel CD, et al. Esophageal dysplasias are detected by endoscopy with Lugol in patients at risk for squamous cell carcinoma in southern Brazil. Dis Esophagus 1999;12: Tincani AJ, Brandalise N, Altemani A, et al. Diagnosis of superficial esophageal cancer and dysplasia using endoscopic screening with a 2 % lugol dye solution in patients with head and neck cancer. Head Neck 2000;22: Ishihara R, Yamada T, Iishi H, et al. Quantitative analysis of the color change after iodine staining for diagnosing esophageal high-grade intraepithelial neoplasia and invasive cancer. Gastrointest Endosc 2009;69: Du GF, Li CZ, Chen HZ, et al. Rose bengal staining in detection of oral precancerous and malignant lesions with colorimetric evaluation: a pilot study. Int J Cancer 2007;120: Amagasa T, Yamashiro M, Ishikawa H. Oral leukoplakia related to malignant transformation Oral Science Int 2006;3: Petti S. Pooled estimate of world leukoplakia prevalence: a systematic review. Oral Oncol 2003;39: