Indian Journal of Dental Research

: 2013  |  Volume : 24  |  Issue : 2  |  Page : 277-

Histomorphometric analysis of nuclear and cellular volumetric alterations in oral lichen planus, lichenoid lesions and normal oral mucosa using image analysis software

Sowmya S Venkatesiah1, Alka D Kale2, Seema R Hallikeremath2, Vijayalakshmi S Kotrashetti2,  
1 Department of Oral Pathology & Microbiology, M.S. Ramaiah Dental College and Hospital, M.S.R. Nagar, Bangalore, Karnataka, India
2 Department of Oral Pathology and Microbiology, KLE VK Institute of Dental Sciences and Hospital, Belgaum, Karnataka, India

Correspondence Address:
Sowmya S Venkatesiah
Department of Oral Pathology & Microbiology, M.S. Ramaiah Dental College and Hospital, M.S.R. Nagar, Bangalore, Karnataka


Introduction: Lichen planus is a chronic inflammatory mucocutaneous disease that clinically and histologically resembles lichenoid lesions, although the latter has a different etiology. Though criteria have been suggested for differentiating oral lichen planus from lichenoid lesions, confusion still prevails. Aims: To study the cellular and nuclear volumetric features in the epithelium of normal mucosa, lichen planus, and lichenoid lesions to determine variations if any. Materials and Methods: A retrospective study was done on 25 histologically diagnosed cases each of oral lichen planus, oral lichenoid lesions, and normal oral mucosa. Cellular and nuclear morphometric measurements were assessed on hematoxylin and eosin sections using image analysis software. Statistical Analysis: Analysis of variance test (ANOVA) and Tukey«SQ»s post-hoc test. Results: The basal cells of oral lichen planus showed a significant increase in the mean nuclear and cellular areas, and in nuclear volume; there was a significant decrease in the nuclear-cytoplasmic ratio as compared to normal mucosa. The suprabasal cells showed a significant increase in nuclear and cellular areas, nuclear diameter, and nuclear and cellular volumes as compared to normal mucosa. The basal cells of oral lichenoid lesions showed significant difference in the mean cellular area and the mean nuclear-cytoplasmic ratio as compared to normal mucosa, whereas the suprabasal cells differed significantly from normal mucosa in the mean nuclear area and the nuclear and cellular volumes. Conclusions: Morphometry can differentiate lesions of oral lichen planus and oral lichenoid lesions from normal oral mucosa. Thus, morphometry may serve to discriminate between normal and premalignant lichen planus and lichenoid lesions. These lesions might have a high risk for malignant transformation and may behave in a similar manner with respect to malignant transformation.

How to cite this article:
Venkatesiah SS, Kale AD, Hallikeremath SR, Kotrashetti VS. Histomorphometric analysis of nuclear and cellular volumetric alterations in oral lichen planus, lichenoid lesions and normal oral mucosa using image analysis software.Indian J Dent Res 2013;24:277-277

How to cite this URL:
Venkatesiah SS, Kale AD, Hallikeremath SR, Kotrashetti VS. Histomorphometric analysis of nuclear and cellular volumetric alterations in oral lichen planus, lichenoid lesions and normal oral mucosa using image analysis software. Indian J Dent Res [serial online] 2013 [cited 2021 Apr 11 ];24:277-277
Available from:

Full Text

Oral lichen planus (OLP) is a common mucocutaneous disease characterized by radiating white or gray velvety, thread-like papules distributed in linear, annular, or reticular patches, or in rings and streaks over the buccal mucosa and to a lesser extent on the lips, tongue, and palate. A tiny white elevated dot is frequently present at the intersection of the white lines, which are known as striae of Wickham. [1] OLP presents in reticular, plaque, erosive, bullous, atrophic, hypertrophic, follicular and linear forms. [2]

There are various lesions that resemble lichen planus both clinically and histologically and these are referred to as 'lichenoid lesions.' Oral lichenoid lesions (OLL) encompass several clinical settings, such as oral lichenoid contact lesions (OLCL), oral lichenoid drug reactions (OLDR), and oral lichenoid lesions of graft vs host diseases (OLL-GVHD). [3]

Dulbreuilh, in 1906, was the first to describe the histology of OLP. [4] The essential histologic features of OLP are basal cell liquefaction degeneration, subepithelial band of lymphocytes, and normal epithelial maturation pattern. [5] The OLL, on the other hand, show a diffuse admixture of inflammatory cell infiltrate, which tends to extend into the deeper layers of the corium. [6] But controversy still exists regarding the criteria for diagnosis of these lesionsas there is considerable overlap between the clinical and histopathological features of OLP and OLL. Though histopathological criteria are proposed for the differential diagnosis of OLP from OLL, confusion still prevails. Thus making diagnosis of OLP more complex and challenging. The use of clinical criteria alone might result in false-positive findings for lesions similar to OLP. [7]

Histomorphometry is a quantitative structural technique that allows objective information to be acquired from sections of cells and tissues, and the method possesses the potential to discriminate between the lesions when the diagnosis is ambiguous. [7] The accuracy and objectivity of histomorphometric methods might be useful to discriminate between OLP and OLL and thus facilitate proper patient treatment. The technique has great potential for clarifying the pathogenesis of OLP and possibly for elucidating its role as a premalignant condition. Reports in the literature have suggested that cell and nuclear size parameters may be significantly altered in these lesions. [8] Hence, the aim of this study was to compare the cellular and nuclear volumetric features in epithelium of normal oral mucosa (NOM), OLP, and OLL so as to determine the difference between the three.

 Materials and Methods

In the present retrospective morphometric study, a total of 75 paraffin-embedded tissue blocks consisting of clinically and histologically diagnosed cases of OLP, OLL, and NOM were retrieved from the archives. First, 4-μm thick tissue sections were prepared and stained with Harris's hematoxylin and eosin (H and E). The tissue sections were divided into three groups: Group 1 comprised 25 tissue sections of NOM, and group 2 and group 3 comprised 25 tissue sections each of OLP and OLL diagnosed based on the World Health Organization (WHO) clinical and histological criteria [9] [Figure 1].{Figure 1}

Morphometric technique

The stained sections were subjected to histomorphometric analysis using Leica QWin TM image analysis software (version 3.5.0) along with a research microscope (Leica TM DM2500). The images of five representative fields from each section were captured using a Charge coupled device (CCD) color video camera (Leica™ DFC 320) attached to the research microscope and were stored in a computer. The cellular and nuclear measurements were made using a cursor to outline the basal and suprabasal cells and their nuclei at a magnification of ×400 [Figure 2] and [Figure 3]. Twenty-five basal and suprabasal cells each per slide were analyzed. Basal cells were defined as those cells in contact with the underlying lamina propria. Suprabasal cells were defined as those cells that were just above the basal cells and showed no signs of flattening in the plane perpendicular to the epithelial surface. Cells showing mitosis and those whose outlines were not clearly made out were not considered for counting.{Figure 2}{Figure 3}

Morphometric parameters analyzed were nuclear area (A N ) and diameter (D N ) and cellular area (A C ). From these data, the secondary parameter of nuclear-cytoplasmic ratio (N/C) was generated and was used to calculate nuclear (V N ), cytoplasmic (V CYT ), and cellular volumes (V CELL ).

Nuclear-cytoplasmic ratio (N/C):


Mean nuclear volume (V N ):

A N = D 3 π/6

Mean cytoplasmic volume (V CYT ):


Mean cellular volume (V CELL ):


The means and standard deviations of each parameter of every individual case were obtained for each of the groups investigated. The measurements (in microns) were carried out using the image analysis software. Statistical analysis was performed using the analysis of variance test (ANOVA) and Tukey's post-hoc test.


The clinical data obtained from the department records for all 75 archival tissues is shown in [Table 1]. On comparison of various parameters of basal and suprabasal cells of the study groups, statistically significant difference in basal and suprabasal nuclear and cellular areas, basal N/C ratio, and suprabasal nuclear diameter were observed [Table 2]. Comparison of various parameters between the study groups showed significant difference in a few of the parameters between NOM vs OLP and NOM vs OLL, but no significant difference was found between OLP vs OLL [Table 3]. Similarly, on comparison of nuclear, cellular, and cytoplasmic volumes among the three groups, significant difference was observed between NOM vs OLP in basal and suprabasal nuclear volume, as well as in suprabasal cellular volume. Significant difference was observed in suprabasal nuclear volume in NOM vs OLL [Table 4]. No significant difference was found in cellular, nuclear, and cytoplasmic volumes between OLL vs OLP [Table 5].{Table 1}{Table 2}{Table 3}{Table 4}{Table 5}


It is difficult to arrive at an accurate diagnosis in OLL as differentiation from OLP may not be straightforward [Figure 4]. Confusion arises because the presently available histopathological criteria of OLP is not truly reproducible. [10]{Figure 4}

Various criteria have been put forward: For example, the criteria suggested by Krutchkoff et al., [11] the WHO diagnostic criteria for OLP, and the modified WHO diagnostic criteria for OLP and OLL given by van der Meij and van der Waal. [9] The application of these criteria will exclude a number of patients who actually may have the disease but do not meet all the strict criteria. [9] Thus, OLP and OLL cannot be distinguished on either clinical or histological grounds alone and, at present, no additional diagnostic tools exist. [9] Hence, the present study was designed to find out the cellular and nuclear volumetric features in the epithelium of OLP and OLL and to compare OLP and OLL with NOM using morphometry.

Morphometric analysis of the cellular and nuclear measurements of basal and suprabasal cells of all groups were done. The basal cells of oral stratified squamous epithelium were considered because they represent the progenitor cells that are responsible for the production of the other cells that make up the various layers of the epithelium.Similarly suprabasal cells form the cells of differentiating compartment. So changes in the basal cells and suprabasal cells may have serious implications on future cell behavior, including malignant transformation. The measurement of their size and shape in white lesions (traumatic keratosis, lichen planus, leukoplakia and squamous cell carcinoma) may be an important prognostic marker. [12]

In the present study, the mean nuclear area of basal cells was significantly higher in OLP (16.45 μm 2 ) as compared to NOM (13.23 μm 2 ) (P = 0.03). Shabana et al. have also described increase in nuclear area of basal cells of OLP compared with normal epithelium and with non-neoplastic oral mucosa using morphometric analysis. [13] The mean nuclear area of suprabasal cells was significantly higher in OLP (18.76 μm 2 ) and OLL (18.43 μm 2 ) compared to normal mucosa (15.17 μm 2 ) (P = 0.034 and P =0 .017, respectively). When OLP and OLL were compared, no significant difference was found. The increase in nuclear area in OLP as compared to NOM may be caused by an increase in chromosomes related to abnormal cell division and by an increase in DNA content related to proliferation of the cells. Thus, as proposed by Kinoshita et al., [14] these may be important indices for predicting malignant potential of precancerous lesions.

The mean cellular area of basal and suprabasal cells of OLP and basal cells of OLL were significantly higher compared to normal mucosa, but no significant difference was seen between OLP and OLL in the present study. Similarly, studies done by others have shown significant difference between OLP and normal mucosa, suggesting a risk of malignant change. [12],[15] Studies have shown that the increase in nuclear area in the high-risk lesions and carcinomas may be a reflection of the increase in DNA synthesis. A proportional increase in the nuclear volume and DNA content was found in the epithelial cells in the mucosa of the uterine cervix. [12] Among the main causative factors for the increase in the epithelial cell size seems to be mechanical trauma, repeated friction, and chemical carcinogens. Eveson and MacDonald in their investigation using carcinogens on hamster cheek pouch epithelium, showed progressive increase in the thickness of the progenitor cell compartment due to increase in both size (mean cell area) and number of progenitor cells. Thus, they emphasized the potential diagnostic and prognostic value of the cell area in the deeper layers of the epithelium and the changes occurring in the basal cells for indicating the increased metabolic activity that occurs prior to the invasion of the underlying connective tissues. The increase in basal cellular and nuclear size has been shown to be associated with lesions having high risk of malignant change. [12]

The nuclear-cytoplasmic ratio (N/C) of basal cells of OLP (0.86) and OLL (0.83) was significantly lower than that of NOM (1.01) (P = 0.025 and P = 0.006, respectively). However, there was no significant difference in the N/C ratio between OLP vs OLL in the present study. Similar results were found by Jin et al. who reported a decrease in N/C ratio for fibrous hyperplasias, inflammatory lesions, hyperkeratosis, squamous cell papillomas, lichen planus, leukoplakias with dysplasia, dysplasias from the edge of malignant lesions, as well as in malignant lesions themselves. [16]

There was no statistically significant difference in the N/C ratio of suprabasal cells between the three groups compared. A quantitative study was done by Ramesh et al. on lesional and non-lesional mucosa of OLP, with comparisons to normal buccal mucosa, using cytological smears. They found that in the buccal mucosa, there was no significant difference in cytoplasmic area, nuclear area, or N/C ratio between any of these groups. [17] The present study results showed that nuclear and cellular measurements were invariably lower in the basal cells than in the suprabasal cells, irrespective of the group from which they were derived and irrespective of the parameter considered. The sole exception was the N/C ratio as the N/C ratio of basal cells (0.86) in OLP was higher than that of the suprabasal cells (0.69). OLL and NOM also showed similar findings, with basal cells having N/C ratio of 0.83 and 1.01, respectively, and suprabasal cells having N/C ratio of 0.76 and 0.77 respectively. A number of studies have demonstrated progressive decrease in N/C ratio values during differentiation of normal stratified squamous epithelium. Basal cells, which are the least differentiated, showed a higher value when compared to spinous and granular cell layers which was decreased progressively from basal to granular cell layers. [18],[19] It was found in this study that nuclear and cellular volumes of basal cells in OLP (50.69 μm 3 and 96.30 μm 3 , respectively) were lower than that of suprabasal cells (61.46 μm 3 and 104.18 μm 3 , respectively). Similar findings were found in the OLL and normal mucosa groups. The increase in nuclear and cell volumes between basal and suprabasal layers may be accompanied by an increased synthesis of nuclear and cytoplasmic components, including nuclear and plasma membranes, or it may be the result of intake of fluid producing a passive expansion. [20]

The mean nuclear diameter of basal cells in this study did not show any statistically significant difference between the three groups, whereas the mean nuclear diameter of suprabasal cells of OLP (4.76 μ) was significantly higher compared to NOM (4.34 μ). However, no significant difference was found in the remaining groups in suprabasal cells. A similar study in the literature has shown that the nuclear area, perimeter, and nuclear diameter increased steadily from the normal epithelium through traumatic keratosis, lichen planus, leukoplakia, candidal leukoplakia and the risk group (squamous cell carcinoma patients who had white lesions that underwent malignant change or who had lesions coexisting with carcinomas at first examination) to the highest values in squamous cell carcinoma because of increased metabolic activity. [12]

In this study, the mean nuclear volume of basal and suprabasal cells of OLP (50.69 μm 3 and 61.46 μm 3 , respectively) and suprabasal cells of OLL (58.01 μm 3 ), showed significantly greater values compared to NOM (35.65 μm 3 ). However, no significant difference was observed when compared with other study groups. The mean cytoplasmic volume of basal and suprabasal cells of our study did not show any significant difference between the three groups. The mean cellular volume of basal cells did not differ significantly between the three study groups, but the mean cellular volume of suprabasal cells of OLP (104.18 μm 3 ) showed a significantly higher value than that of normal mucosa (77.98 μm 3 ). However, the other groups did not differ significantly in this parameter. In a quantitative study of the oral epithelium subjected to friction, MacKenzie and Miles found an increase in the both the size and number of the cells, mainly in the stratum spinosum and stratum granulosum rather than the basal cells. It has been suggested that the epithelial responses to a wide range of stimuli may appear similar but that the various cell compartments may contribute to the response in varying degrees. [12] Increase in cellular and nuclear volumes have been described in dysplastic lesions of hamster cheek pouch induced by the carcinogen 7,12-Dimethylbenz (a) anthracene [DMBA]. In all strata, nuclear, cytoplasmic, and cellular volumes were lowest in the normal control group and were increased in the experimental groups by 2-3 times than that of control group, indicating the cellular and nuclear hypertrophy that occurs in conjunction with the epithelial hyperplasia and proliferation induced by the chemical carcinogen. [21] In a study done by Jin et al., nuclear and cellular volumes of both the basal and spinous strata were similar in normal and inflammatory groups but were almost doubled in the lichen planus group. [17] Our results in the present study also showed similar findings, with the nuclear volumes of basal and suprabasal cells and the cellular volume of suprabasal cells being significantly higher than that of normal mucosa. Hence, nuclear and cellular volumes may serve as potential discriminators between benign lesions and premalignant lichen planus, but these parameters cannot be used to differentiate between OLP and OLL.

On review of literature, we found that studies comparing the nuclear and cellular volumes of OLP and OLL are very few. [7] One of the studies has shown that nuclear and cellular volumes were significantly higher in OLP compared to lichenoid lesions in both basal and spinous cell compartments, and the authors have suggested that the actions and effects of the inflammatory infiltrate may also differ between these lesions. However, the present study showed no significant differences between OLP and OLL in both basal and suprabasal strata. The reasons for this nonsignificant difference could be due to shrinkage or swelling of the tissue during processing, osmolality of the fixative, the method used for estimating the mean nuclear diameter, errors in tracing cell outlines by hand (as the assessment of exact cell and nuclear outline by the observer is highly subjective), as well as other factors like age, friction, tobacco smoking, and iron deficiency anemia. However, a definite difference was seen between NOM vs OLP and OLL.This difference may serve as indicator for malignant transformation of OLP and OLL. Controversy still exists regarding the potential of OLP to undergo malignant transformation. There are reports of dysplasia in OLP, but others suggest that existing dysplastic lesions mimic lichen planus, the so-called oral lichenoid dysplasias. [22],[23]

Malignant transformation of OLP may be related to, or dependent on, a series of molecular stimuli originating in the inflammatory infiltrate. Chronic inflammation has been associated with various types of cancer, and it has been widely reported that the inflammatory infiltrate can be a strong risk factor for cancer development in ulcerative colitis, atrophic gastritis, and Barret esophagus. It was recently proposed that OLP could also be included in this group of diseases. Some molecules and radicals generated by inflammatory cells can act as mutagenic agents for epithelial cells or influence important cell cycle regulation mechanisms, e.g. apoptosis, cell cycle arrest, and cell proliferation. [24] In one of the prospective studies of OLP done by Holmstrup et al., all cases (n = 611) were histopathologically investigated prior to the manifestation of oral cancer. During follow-up, nine patients [1.5%; eight females (1.9%) and one male (0.5%)] developed oral cancer. [25] According to the recent study done by van der Meij et al., OLL (according to WHO criteria) has a 142-fold higher risk of development of carcinoma compared to OLP and the authors have strongly recommended that OLL be considered as premalignant lesions. [26] The present study showed similar changes in parameters like nuclear area, cellular area, and nuclear volume between OLP, OLL, and NOM, and no difference was appreciated between OLP and OLL.


From the present study, it can be concluded that morphometry definitely can differentiate lesions of OLP and OLL from NOM and hence may serve as potential discriminators between normal and premalignant lichen planus and lichenoid lesions. OLP and OLL showed increase in nuclear and cellular areas and volumes compared to normal mucosa, suggesting that they have a higher risk for malignant transformation than those lesions that do not show such an increase. Since OLP and OLL did not differ significantly, both the lesions may be similar in their potential for malignant transformation, and morphometry cannot be used to differentiate between OLP and OLL. Thus, OLP and OLL cannot be differentiated easily clinically, histopathologically, or morphometrically.

Further studies using morphometry have to be conducted with larger sample sizes to compare the various clinical and histological stages of OLP and OLP with and without dysplasia. These parameters should be compared with cases of malignancy, lichenoid lesions and lichenoid dysplasias to better understand the behavior of these lesions and to differentiate between OLP and OLL.


Dr. Srinivas Wanaki, Dean, PNMN dental college, Dr. Vijay Wadhwan, Reader, Sharad Pawar dental college and Dr. Punnya V Angadi, Reader, Dept. of Oral Pathology, KLE VK Institute of dental sciences, Belgaum for helping and encouraging for conduct of the study.


1Shafer WG, Hine M, Levy B. Shafer's textbook of Oral Pathology. Diseases of the Skin. Rajendran R, editor. 5 th ed. New Delhi: Reed Elsevier India Private Limited; 2006. p. 1103-7.
2Neville BW, Damm DD, Allenl CM. Oral and Maxillofacial Pathology. Dermatologic diseases. 2 nd ed. Philadelphia: W. B. Saunders Company; 2002. p. 680-5.
3Al-Hashimi I. Oral lichen planus and oral lichenoid lesions: Diagnostic and therapeutic considerations. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2007;103(suppl 1):S25-31.
4Regezi JA, Scuibba JA, Jordan RCK. Oral Pathology: Clinical pathologic correlations. White lesions. 5 th ed. China: Saunders; 2008. p. 90-5.
5Eisenberg E. Oral lichen planus: A benign lesion. J Oral Maxillofac Surg 2000;58:1278-85.
6Odell EW, Morgan PR. Biopsy pathology of the oral tissues. Lichenoid, psoriasiform and other types of stomatitis. 1 st ed. London: Chapman and Hall Medical; 1998. p. 37-59.
7Khoo SP, Primasari A, Saub R. Nuclear and cellular volumetric alterations in oral lichen planus and lichenoid lesions: A histomorphometric study. J Oral Sci 2001;43:151-7.
8White FH, Jin Y, Yang L. Quantitative cellular and nuclear volumetric alterations in epithelium from lichen planus lesions of human buccal mucosa. J Oral Pathol Med 1994;23:205-8.
9Meij V, Waal V. Lack of clinicopathologic correlation in the diagnosis of oral lichen planus based on the presently available diagnostic criteria and suggestions for modification. J Oral Pathol Med 2003;32:507-12.
10van der Waal I. Oral lichen planus and oral lichenoid lesions; a critical appraisal with emphasis on the diagnostic aspects. Med Oral Patol Oral Cir Bucal 2009;1:14:E310-4.
11Krutchkoff DJ, Cutler L, Laskowski S. Oral lichen planus: The evidence regarding potential malignant transformation. J Oral Pathol 1978;7:1-7.
12Shabana AH, EL-Labban NG, Lee KW. Morphometric analysis of basal cell layer in oral premalignant white lesions and squamous cell carcinoma. J Clin Pathol 1987;40:454-8.
13Shabana AH, El-Labban NG, Lee KW, Kramer IR. Morphometric analysis of suprabasal cells in oral white lesions. J Clin Pathol 1989;42:264-70.
14Kinoshita Y, Inoe S, Honma Y. Diagnostic significance of nuclear DNA content and nuclear area in hyperplasia, Dysplasia and carcinoma. J Oral Maxillofac Surg 1992;50:728-33.
15Sugerman PB, Savage NW, Williams SL, Joynson OB. A quantitative cytological study of lesional and non-lesional mucosa in oral lichen planus. Arch Oral Biol 1996;41:117-20.
16Jin Y, Yang LJ, White FH. Preliminary assessment of the epithelial nuclear- cytoplasmic ratio and nuclear volume density in human palatal lesions. J Oral Pathol Med 1995;24:261-5.
17Ramesh T, Mendis BR, Ratnatunga N. The effect of tobacco smoking and of betel chewing with tobacco on the buccal mucosa: Cytomorphometric analysis. J Oral Pathol Med 1999;28:385-8.
18White FH, Gohari K. Variations in the nuclear-cytoplasmic ratio during epithelial differentiation in experimental oral carcinogenesis. J Oral Pathol 1981;10:164-72.
19Bernimoulin JP, Shrouder HE. Quantitative electron microscopic analysis of the epithelium of normal human alveolar mucosa. Cell Tissue Res 1977;180:383-401.
20White FH, Gohari K. Cellular and nuclear volumetric alterations during differentiation of normal hamster cheek pouch epithelium. Arch Dermatol Res 1982;273:307-18.
21White FH, Codd RM, Gohari K. An ultrastructural morphometric study of cellular and nuclear volume alterations during experimental oral carcinogenesis. J Submicrosc Cytol 1985;17:481-93.
22Mac Donald DG, Rennie JS. Oral epithelial atypia in denture induced hyperplasia, lichen planus and squamous cell papilloma. Int J Oral Surg 1975;4:40-5.
23De Jong WF, Albrecht M, Banoczy J, Vander Waal I. Epithelial dysplasia in oral lichen planus. A preliminary report of a Dutch-Hungarian study of 100 cases. Int J Oral Surg 1984;13:221-5.
24Rad M. Correlation between clinical and histopathologic diagnoses of oral lichen planus based on modified WHO diagnostic criteria. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2009;107:796-800.
25Holmstrup P, Thorn JJ, Rindum J. Malignant development of lichen planus-affected oral mucosa. J Oral Pathol 1988;17:219-25.
26Van der Meij EH, Mast H, Van der Waal I. The possible premalignant character of oral lichen planus and lichenoid lesions: A prospective five-year follow- up study of 192 patients. Oral Oncology 2007;43:742-8.