Material and Method: In the present study, 100 cases of oral lichen planus were clinically and histopathologically analyzed. Out of the 100 cases, 50 were prospective and 50 were retrospective cases. Prospective cases were collected based on the clinical diagnosis of oral lichen planus and oral lichenoid lesion. Retrospective cases were collected based on the histopathological diagnosis of oral lichen planus. Both the clinical and histopathological analyses were performed based on a proposal for a set of modified diagnostic criteria of oral lichen planus and oral lichenoid lesion. A final diagnosis of oral lichen planus was made only after the correlation of the clinical diagnosis with the histopathological diagnosis.
Results: The interobserver agreement among three observers for both prospective and retrospective cases in the final diagnosis of oral lichen planus was found to be “good” to “very good” indicating high reproducibility. However, the final diagnoses of true oral lichen planus after clinico-pathological correlation in prospective and retrospective study groups appeared to be 38.0% and 54.0% respectively.
Conclusion: The results of the present study revealed mild to moderate clinico-pathological correlation in the final diagnosis of oral lichen planus for the prospective and retrospective study groups respectively.
Oral Lichen Planus and Oral Lichenoid lesions (OLL) pose a major diagnostic problem since their clinical and histopathological features overlap with each other. Distinguishing OLP and OLL from one another with only limited data may create a major diagnostic challenge, with considerable implications for patient management and follow up. Lichen Planus can be confused with other lichenoid conditions (such as Nonspecific Lichenoid Reactions, Atypical Lichenoid Stomatitis, Graft Versus Host Reactions [GVHRs], Drug Reactions, Lupus Erythematosus [LE], Erythema Multiforme [EM], and Oral Lichenoid Dysplasia [OLD]) both clinically and histopathologically. Hence it is important to familiarize oneself with the clinicopathological patterns of OLP and OLL, so as to develop an accurate diagnostic and prognostic assessment[4].
In 1978, the World Health Organization (WHO) centre for the study of precancerous lesions established a clinical and histopathological definition for the diagnosis of true OLP (Table I)[5]. Ever since, this definition has been used in the diagnosis of OLP and regarded as ‘gold standard’ in the inclusion of patients in studies focusing on several aspects of OLP, but validation of this definition has never been performed[6]. Clinical and histopathological assessment of OLP based on this WHO definition is rather subjective and insufficiently reproducible. Hence, it requires a proposed set of modified diagnostic criteria of OLP and OLL, based on the WHO definition of clinical and histopathological features (Table II)[6-8].
Table I: World Health Organization diagnostic criteria (1978) of oral lichen planus (OLP) (5)
Oral Lichen Planus is a syndrome diagnosis that is based on the presence of several clinical and histopathological criteria. Thus, the diagnostic approach is best described as a method of pattern recognition both clinically and histopathologically[7]. This indicates that diagnosis cannot be achieved solely based on the clinical or histopathological diagnosis. Confirmation of the diagnosis of OLP therefore has to be made after the correlation of the clinical and histopathological diagnoses. However, few data exist on the correlation between clinical and histopathological diagnoses of OLP.
The aim of the present study was therefore to establish a clinical and histopathological correlation in the diagnosis of OLP, based on the modified WHO diagnostic criteria of OLP and OLL.
For the selection of the prospective cases, three different clinicians participated. All the clinicians were well trained and experienced. Out of 50 prospective cases, 40 cases were selected based on the clinician agreement that the clinical picture was diagnostic of OLP and the remaining 10 cases were clinically diagnosed as OLL. These were included as group 1 cases. For group 1 cases, a detailed clinical examination of the patients was performed. All the procedures of biopsy were explained to the patient. After obtaining written consent from the patient, reticular areas of the lesion were selected as the most appropriate site of biopsy and biopsy tissue was obtained under local anesthesia in aseptic conditions. A 3 mm incisional biopsy was obtained. Biopsy specimens were preserved in 10% buffered formalin solution, processed and paraffin embedded. Four micrometer sections were prepared and stained with Hematoxylin and Eosin (H&E).
Fifty retrospective cases were retrieved from the archives based on the histopathological diagnosis of OLP and were considered group 2 cases. For group 2 cases, paraffin blocks of OLP were retrieved and sections of four micrometer were prepared. Sections were stained with routine H&E stains. For these cases, clinical data was noted from the previous case history records and compared with their respective clinical data.
In both the study groups, the reviewing pathologists were blinded to the clinical presentation and site of biopsy. All 100 Group 1 and 2 slides were reviewed by the three different oral pathologists. All the three oral pathologists were given the same 50 prospective and 50 retrospective histopathological slides. Slides were evaluated based on the proposal for a set of modified WHO diagnostic criteria of OLP and OLL[6] (Table II).
The discrete (categorical) clinical and histopahological OLP findings were compared by the chi-square (χ2) test. The unweighted Kappa statistics was used to assess interobserver agreement. A two-sided p value less than 0.05 (p<0.05) was considered statistically significant.
The frequency distributions of prospective OLP findings (true/compatible) of both clinical and histological diagnoses are summarized in Table IV. In clinical diagnosis, there were 40(80.0%) true cases and 10(20.0%) compatible cases.
In contrast, in histological findings, there were 19(38.0%) true OLP cases, 8(16.0%) cases were compatible, 11(22.0%) OLD and 12(24.0%) were diagnosed as others. Comparing the clinico-pathological diagnosis (true/compatible), the χ2 test reveled significantly different OLP findings between the two groups (true/compatible: 80.0%/20.0% vs. 38.0%/62.0%, χ2=18.23, p<0.001). In other words, regarding the findings for true OLP cases, the sensitivity of the clinical diagnosis was higher than the histopathological diagnosis (80.0% vs. 38.0%) while for compatible cases the sensitivity of the histopathological diagnosis was higher than the clinical diagnosis (20.0% vs. 62.0%).
Retrospective findings: The inter-observer agreement of clinical and histopathological diagnosis of OLP among three observers were analyzed by using Kappa statistics and summarized in Table V. The Kappa statistics revealed “good” to “very good” agreement for both clinical and histopathological findings between the observers and agreement was higher in histopathological findings than clinical findings.
The frequency distributions of retrospective OLP findings (true/compatible) of both clinical and histological diagnosis are summarized in Table VI. Clinical diagnosis revealed 27 (54.0%) true OLP and 23 (46.0%) compatible cases. The histopathological diagnosis also showed similar positive findings for true (54.0%) and compatible (46.0%) OLP cases. Thus, both clinical and histopathological diagnosis accounted for 54.0% moderate sensitivity for true OLP and 46.0% mild sensitivity for compatible OLP cases.
In group 1, inter-observer agreement in the final diagnosis appears to be “good” to “very good” but correlation of the clinico-pathological diagnosis resulted in “mild” sensitivity of 38%. Out of 40 (80%) cases in which all clinicians agreed about the clinical diagnosis of OLP, only 19 (38%) cases were histopathologically diagnosed as OLP. This assessment indicates that histopathological confirmation of clinical cases is mandatory for the final diagnosis of true OLP. Similarly in group 2, observer variability was found to be “good” to “very good” but correlation of the clinico-pathological diagnosis appear to be “moderate” sensitivity accounting for 54%. Out of 50 (100%) cases, in which all pathologists agreed about the diagnosis of OLP, only 27 (54%) cases turned out to be clinically true OLP. From the observation of both the study groups, the clinico-pathological similarity between different OLL is evident. Hence we observed that factors contributing for lack of correlation are the inherent nature of OLP and OLL that shows overlapping clinico-pathological features. It is therefore mandatory for the clinician and pathologist to be aware of the clinico-pathological patterns of OLL.
In group 1 and 2, 8 (16%) and 23 (46%) cases were finally diagnosed as OLL. These lesions fail to show strict clinical and histopathological features proposed by modified WHO criterion for OLP. In group 1, cases revealed histopathological features such as extension of inflammatory cells in the deeper stroma, mixed population of inflammatory component, absence of basal cell liquefaction, perivascular cuffing of inflammatory cells and rarely aggregation of inflammatory cells in the form of lymphoid follicles. In group 2, cases were unilateral, solitary and sometimes erythematous without a characteristic reticular pattern. Our observations match with the previous published articles that have mentioned similar clinical and histopathological features of OLL[11]. Segregation of OLP and OLL by clinico pathological assessment is a must, as some OLL cases such as GVHRs and OLL of unknown origin have high propensity for malignant transformation[12,13].
The appropriate selection of the biopsy site has a vital role in the accurate diagnosis of OLP. Previous studies have reported that reticular lesions were histopathologically diagnosed as OLP much more consistently than erythematous and erosive lesions[9,10]. Our observations were in agreement with these findings, as the biopsy was taken from the reticular area of the lesion in 13 (26%) out of 19 (38%) cases diagnosed as true OLP. In a few instances the histopathological features may not be diagnostic as OLP evolves through a cycle of exacerbation and quiescence. Biopsy in any condition helps to differentiate whether the lesion is of inflammatory origin or consists of underlying atypical features in the epithelium. It is always possible that more than one disease process can coexist together. Hence it is prudent to take multiple biopsies. If multiple biopsies are not possible, biopsy of the cancer prone site is more precise[14].
In the present study, 11 (22%) cases in group 1 exhibited epithelial dysplasia along with the characteristic lichenoid infiltrate in the juxta epithelium. These lesions are histopathologically named oral lichenoid dysplasia (OLD). Our results emphasize the significance of assessing the epithelial maturation and cytomorphology in any OLP case as many studies have documented that there are serious flaws in the initial diagnosis of OLP[12-15] only based on the similarly appearing lichenoid infiltrate in the stroma to the extent of ignoring atypical features in the epithelium. OLP and OLD share essentially no pathogenetic relationship but their similarity is the presence of lichenoid inflammatory infiltrate. In OLP, the lichenoid infiltrate represents cell-mediated immune response provoked by different antigens, whereas in OLD, the lichenoid infiltrate represents the immune surveillance mechanism against atypical epithelial cells[4]. Out of 11 (22%) cases, 8 (16%) cases were diagnosed as OLD from clinically true OLP. Only 4(8%) cases were associated with habits and 5 (10%) cases were erosive/ulcerative forms. Hence our observations were not in accordance with the previous studies that have suggested these lesions are predominantly solitary erosive areas occurring on a cancer prone site and associated with habits[12,14,16]. None of the OLP cases in the study were associated with Candida albicans, as it was hypothesized that candida can form carcinogenic N-nitrosobenzylmethylamine that is associated with malignant transformation of OLP[10-16].
In conclusion, the results of the present study show a lack of clinico-pathological correlation in the diagnostic assessment of OLP. To confirm the clinical diagnosis of OLP, histopathological assessment has to be performed and the final diagnosis has to be achieved only after the correlation of clinical and histopathological diagnosis. Assessment of OLP and OLL based on modified WHO criteria displayed a significant role in the accurate differentiation of OLP from OLL and OLD.
ACKNOWLEDGMENT
The authors thank Mr. M.P.S. Negi, Statistician, Institute
for Data Computing and Training (I.D.C.T.), Lucknow, for
providing valuable assistance in data analysis.
1) Scully C, Beyli M, Ferreiro MC, Ficarra G, Gill Y, Griffiths M,
Holmstrup P, Mutlu S, Porter S, Wray D. Update on oral lichen
planus: Etiopathogenesis and Management. Crit Rev Oral Biol
Med. 1998;9:86-122.
2) Moncarz V, Ulmansky M, Lustmann J. Lichen planus: Exploring
its malignant potential. J Am Dent Assoc. 1993;124:102-8.
3) Eisen D, Carrozzo M, Bagan Sebastian JV, Thongprasom K.
Oral lichen planus: Clinical features and management. Oral Dis.
2005;11:338-49.
4) Eisenberg E. Clinico pathologic patterns of oral lichenoid lesions.
OMFS Clin North Am. 1994;6:445-61.
5) K ramer IR, Lucas RB, Pindborg JJ, Sobin LH. Definition of
leukoplakia and related lesions: An aid to studies on oral
precancer. Oral Surg Oral Med Oral Pathol. 1978;46:518-39.
6) Van der Meij EH, van der Waal I. Lack of clinico pathologic
correlation in the diagnosis of oral lichen planus based on
the presently available diagnostic criteria and suggestion for
modification. J Oral Pathol Med. 2003;32:507-12.
7) van der Meij EH, Reibel J, Slootweg PJ, van der Wal JE, de Jong
WF, van der Waal I. Interobserver and intraobserver variability in
the histologic assessment of oral lichen planus. J Oral Pathol Med.
1999;28:274-7.
8) van der Meij EH, Schepman KP, Plonait DR, Axéll T, van der
Waal I. Interobserver and intraobserver variability in the clinical
assessment of oral lichen planus. J Oral Pathol Med. 2002;31:95-8.
9) Xue JL, Fan MW, Wang SZ, Chen XM, Li Y, Wang L. A clinical
study of 674 patients with oral lichen planus in China. J Oral
Pathol Med. 2005;34:467-72.
10) Eisen D. The Clinical features, malignant potential and systemic
association of oral lichen planus: A study of 723 patients. J Am
Acad Dermatol. 2002;46:207-14.
11) Farthing PM, Speight PM. Problems and pitfalls in oral mucosal
pathology. Cur Diag Pathol. 2006;12:66-74.
12) Lodi G, Scully C, Carrozzo M, Griffiths M, Sugerman PB,
Thongprasom K. Current controversies in oral lichen planus:
Report of an international consensus meeting. Part 2. Clinical
management and malignant transformation. Oral Surg Oral Med
Oral Pathol Oral Radiol Endod. 2005;100:164-78.
13) Gonzalez-Molis MA, Scully C, Gil-Montoya JA. Oral lichen
planus: Controversies surrounding malignant transformation.
Oral Dis. 2008;14:229-43.
14) Eisenberg E. Oral lichen planus: A benign lesion. J Oral
Maxillofacial Surg. 2000; 58:1278-85.