Dentigerous cyst versus unicystic ameloblastoma differential diagnosis in routine histology

Anton Dunsche1 Ortwin Babendererde1 Jutta Lu ttges2 Ingo N. G. Springer1

1

Abstract

Department of Oral and Maxillofacial Surgery, and 2 Department of Pathology, University of Kiel, D-24105 Kiel, Germany

Correspondence to: Dr I. Springer Department of Oral and Maxillofacial Surgery, University of Kiel, Arnold-Heller-Street 16, D-24105 Kiel, Germany Tel.: 49 431 597 2783 Fax: 49 431 597 2950 e-mail: springer@mkg.uni-kiel.de
Accepted for publication January 15, 2003 Copyright Blackwell Munksgaard 2003 J Oral Pathol Med . ISSN 0904-2512 Printed in Denmark . All rights reserved

Background: Unicystic ameloblastomas (UAs) and dentigerous cysts (DCs) have an identical clinical and radiographic appearance. Some subtypes of UAs have a better prognosis than solid or multicystic ameloblastomas, and simple enucleation is the adequate treatment. The present study was designed to test the hypothesis that UAs with small islands of ameloblastomatous epithelium may be misdiagnosed as a DC or keratocyst if no more than two histologic sections are examined. Methods: A total of 101 resection specimens from 22 women and 73 men (mean age: 46.5 years) were selected, all showing the clinical and radiographic features of a DC. Only cysts with a minimum diameter of 15 mm in the panoramic X-ray were considered for the present investigation. The histopathologic diagnosis had been routinely established by examining two sections. For our study, the specimens were investigated by step sections at 50 mm and by staining of 5 mm thin sections with hematoxylin and eosin (H&E) at 1 mm levels. An average of 15 slides were evaluated per case. Results: Microscopic examination of the step sections did not reveal ameloblastomatous epithelium in the cyst lining epithelium of the 101 cases. Thus, every primary diagnosis of a dentigerous cyst was conrmed. In four cases, additional rather large odentogenic cell nests were detected with palisading of basaloid cells, while there was a lack of other signs of ameloblastic differentiation. All lesions were completely resected, and no additional treatment was performed. Conclusions: Step sectioning of larger DCs may reveal associated odontogenic cell nests in some cases but does not lead to the detection of formerly missed ameloblastic cells. Thus, unicystic ameloblastomas are not misdiagnosed if only two slides are prepared for routine diagnosis of DCs. Key words: dentigerous cyst; odontogenic; unicystic ameloblastoma
J Oral Pathol Med 2003: 32: 48691

An inltrative (solid or multicystic) ameloblastoma is a benign epithelial tumor of odontogenic origin showing a strong tendency to recurrence and local aggression (1). Intraosseus, inltrative, peripheral, desmoplastic, or unicystic ameloblastomas are other

486

Dentigerous cyst versus unicystic ameloblastoma

subtypes of ameloblastoma (26). Unicystic ameloblastoma (UA) is a prognostically distinct entity (4). It has a recurrence rate of 6.735.7%, and the average interval to recurrence is approximately 7 years (7). The term `unicystic ameloblastoma' was adopted in the second edition of the international histologic classication of odontogenic tumors (8). Other terms not generally used today are `mural ameloblastoma' (9) and `cystic ameloblastoma' (10). UAs represent 522% of all ameloblastomas (5, 6). The tumor is primarily observed during the second and third decades of life; its preferential location is the mandible, and it can occur inside dentigerous cysts (DCs; 2, 5, 6, 9). In most cases, UAs are associated with tooth impaction, the mandibular third molar being most often involved (6). There are four subtypes of UA (2, 5, 6, 8, 11): 1. A single cystic sac lined by ameloblastomatous epithelium, which may often be seen in focal areas (minimum criterion for diagnosing a lesion as UA); 2. features of subtype 1 plus intraluminal proliferations; 3. features of subtype 1 plus both intraluminal and intramural proliferations; and 4. features of subtype 1 plus intramural proliferations. Enucleation is sufcient for tumors that have proliferated into the lumen (types 1 and 2), whereas subtypes involving the periphery of the brous connective tissue wall of the cyst (types 3 and 4) must be treated radically, i.e. like a solid or multicystic ameloblastoma (2, 47, 10, 1218). UAs and DCs are known to have a similar clinical and radiographic appearance. It appears to be more difcult to differentiate them in cases of dentigerous UAs (associated with an impacted tooth) than in cases of non-dentigerous UAs (not associated with an impacted tooth) (4, 7, 1921). UAs that are not associated with an impacted tooth may mimic a residual cyst or a keratocyst (6). In the Department of Pathology, University of Kiel, two sections are routinely examined in case of a cystic lesion with the clinical and radiologic features of a DC. Considering the need for extensive surgical procedures in cases of type 3 and 4 UAs, we thought it advisable to evaluate the reliability of current concepts in routine histology. In the past, other authors had suggested that in cases of small islands of ameloblastomatous epithelium within the cystic epithelium of a lesion, it might be necessary to examine the entire specimen to be sure of nding these islands (5, 6, 22). Accordingly, we hypothesized that there was a chance that the presence of ameloblastomatous changes in the epithelial cyst lining may be overlooked if cysts that present like DCs clinically are examined by preparing only two sections for routine histology.

Patients and methods

All cystic lesions of the mandibular third molar region treated in the Department of Oral and Maxillofacial Surgery of the University of Kiel, Germany, during a period of 10 years (198594) were evaluated for the present study. All 101 lesions (95 patients) selected for the study had the typical radiographic appearance of a DC. Therefore, only unilocular and no multilocular lesions, as seen in the panoramic X-ray, were included. The resection specimens were macroscopically bisected at their largest diameter and embedded in parafn. A routine histologic evaluation of two slides of the cystic lesions had conrmed the diagnosis in all cases prior to the present study. Two sections were routinely prepared from the parafn blocks at two different levels, one section cut from a supercial portion of the block and one from a deeper portion usually at 100 mm depth. Serial sectioning was only performed during routine histology if any peculiarities were observed, such as abnormalities of the cyst lining epithelium or a high cellularity of the connective tissue that surrounded the cysts. All parafn blocks were stored after routine histology and were available for the present study. Figure 1 shows the age distribution of 115 patients with a DC diagnosed by panoramic X-ray and routine histology. Only 95 patients were included in the study because the parafn blocks from 20 patients were not suitable for serial sectioning. The maximum diameter of the cystic lesions was measured in panoramic radiographs and documented (Fig. 2). Cysts with a diameter of 15 mm (approximately 10.7 mm actual diameter) or more measured in the panoramic radiograph were included (see Discussion). In our department, a magnication of 1.4 is standard for panoramic X-rays and may be used to estimate the actual diameter of a lesion. A minimum of 15 mm was selected as a cut-off point to exclude lesions that were unlikely to represent UAs (e.g. eruption cysts). Ninety-ve patients, 22 females (23.2%) and 73 males (76.8%), with 101 DCs met our criteria. The patients had a mean age of 46.5 years (range 1182 years; SD 16.8 years). For microscopic examination, step sections were prepared, i.e. the parafn block was completely cut into 10 mm thick sections saving every ftieth slide, which was cut to 5 mm and stained with hematoxylin and eosin. This technique resulted in at least one histologic slide per millimeter. Microscopic magnications used were 62.5156.25. The thickness of the epithelium was evaluated. Increased epithelial thickness was dened as more than six layers. The presence of intramural islands of ameloblastoma tissue, odontogenic cell nests with ameloblastomatous differentiation and also lymphocytic inltration were evaluated. In
J Oral Pathol Med 32: 48691

487

Dunsche et al.

Fig. 1. Age distribution of patients with dentigerous cysts. Please compare to the age distribution observed in other studies (39, 40). In the present study, DCs primarily occurred between the ages of 40 and 59 years. This is contrary to other studies showing that DCs primarily occur between the ages of 2039 years. We suppose that this is because of the fact that a minimum size of 15 mm was required for entry into this study. The age was not regarded as a prerequisite for entry into the study as cystic ameloblastomas occur preferably at younger age but, nevertheless, within a wide age range (29). This gure refers to 115 patients, although 95 were examined in the study. The reason for this discrepancy is that the parafn blocks from 20 of the patients were not suitable for serial sectioning.

particular, the criteria proposed by Vickers and Gorlin as indicative of the development of an ameloblastoma were strictly followed (23): 1. Basal cell palisading; 2. basal cell hyperchromatism; 3. polarization of the basal cell nuclei; and 4. vacuolation of the basal cell cytoplasm.

were localized in the left mandible and 43 (42.6%) in the right mandible. The difference is not statistically signicant. The diameters measured in the panoramic radiograph varied from 15 mm (c. 10.7 mm calculated actual diameter) to 66 mm (c. 47.1 mm calculated actual diameter). As measured on the radiograph, 61 cysts (60.4%) were between 15 and 29 mm (c. 10.7 20.7 mm calculated actual diameter) in size. The average diameter measured on the radiograph was 31.12 mm (SD 13.71 mm, n 101); the minimum diameter was 15 mm; and the maximum

Results

diameter was 113 mm. Odontogenic cell nests with palisading of basaloid appearing cells, but lacking other signs of ameloblastomatous differentia-

Forty-one per cent of all cysts analyzed in this study were in patients 4059 years of age (Fig. 1). Fifty-eight (57.4%) of the DCs

tion, were found within the cystic wall in four patients (two males 26 and 52 years of age and two females both 37 years of age; Fig. 3). In none of these patients, the tooth was displaced by the cyst. The cysts were of average size, as measured on the radiograph. The two sections prepared for routine histology prior to the present study did not dissect these palisading cells in any of the four cases. A papillary proliferation of metaplastic squamous epithelium was found in one case after serial sectioning of the specimen (Fig. 4), which was not detectable in the two routine histologic slides. In 17 cases (16.8%), no lymphocytes were found; 56 (55.4%) showed minor inammatory inltrates, 22 (21.8%) moderate inammatory inltrates, and 6 (5.9%) dense inammatory inltrates. Forty-seven (46.5%) of the DCs showed deposits of cholesterol crystals. The epithelial lining appeared regular in 65 cases

Fig. 2. Measurement of the diameter of the cystic lesions. Cysts with a diameter of 15 mm or more measured in the panoramic radiograph were included. Ninety-ve patients, 22 females (23.2%) and 73 males (76.8%), with 101 DCs met our criteria.

488

J Oral Pathol Med 32: 48691

Dentigerous cyst versus unicystic ameloblastoma

Discussion

Various contradictory theories about the development of UAs have been proposed. While some authors suggest that UAs develop by cystic degeneration of solid ameloblastomas, there are certain indications that UAs may develop by mural and/or luminal ameloblastomatous change in a pre-existing cyst (1, 5, 10, 2427). It has also been shown that in UAs a coexistence of nonneoplastic epithelium and neoplastic epithelium is possible (28). Cystic and solid ameloblastomas are supposed to occur at a
Fig. 3. Large odentogenic cell nest exhibiting some palisading of the outer cell layer (left side, large arrow) but without clear ameloblastic differentiation (A). This nding did not become apparent in the course of routine histology. While this nding does not have clinical consequences, it demonstrates that two elective sections may miss certain histologic elements in some cases. Microscopic magnication: 250.

mean age of 36 years (24, 29). UAs are thought to occur primarily in the second and third decade of life (24). To nd denite values, prospective studies were proposed (30). As no such prospective study has been performed yet, we included all lesions that were formerly diagnosed as DC in our department. The age of the patients was disregarded, to be sure not to miss a unicystic ameloblastoma in patients of higher age. In this context, the minimum diameter of 15 mm measured on the panoramic X-ray

(64.4%), but the cells proliferated and the layer was thickened in 36 DCs (35.6%). The epithelium of six cysts showed elongated rete ridges. No ameloblastomatous epithelium in the lining epithelium of these cysts was found. In six cases, basal cell palisading was observed, but no keratinization. Otherwise, none of the Vickers and Gorlin criteria described in the Patients and methods section were observed (23).

as a prerequisite for entrance into the study needs to be discussed. We suggest that two representative sections of a small lesion are more likely to uncover small islands of proliferative tissue in the rst place. Based on our clinical experience, we felt that the larger the lesion the higher is the probability of aggressive behaviour. Other authors found that the radiolucent area in the panoramic X-ray tends to be smaller in cases of dentigerous cysts than in cases of ameloblastoma (31). All lesions studied had the typical clinical and radiographic appearance of DCs. Only unilocular lesions were included in this study. UAs and DCs are supposed to have a similar clinical and radiographic appearance (4, 1921). Moreover, the histologic distinction between UAs and certain non-neoplastic odontogenic cysts can be difcult (20). It has been suggested that six radiographic patterns for UA can be identied ranging from welldened unilocular to multilocular (32). This study aimed at detecting originally misdiagnosed UAs in a group of patients with inconspicuous radiograph and uneventful histology. Prior to this study, all 101 cystic lesions had been diagnosed as DCs by routine histology using two elective sections. The intraluminal epithelial proliferation in plexiform UA may closely resemble hyperplastic odontogenic epithelium (33). In the present study, hyperplastic

Fig. 4. Papillary proliferation of metaplastic squamous epithelium. A papillary proliferation of metaplastic squamous epithelium (P) is seen in this specimen. The two sections prepared for routine histology had not dissected this lesion. Epithelial lining of the cyst (arrow), brous capsule of the cyst (F). This picture illustrates our hypothesis. If this specimen was cut along the two lines (L), this proliferation of metaplastic epithelium (P) would not be visible. Microscopic magnication: 100.

odontogenic epithelium with more than six layers lined the cysts in 35.6% of all cases. However, no signs of ameloblastomatous differentiation were present. Also, no dysplasia was observed, and hence these lesions were simply classied as benign DCs. Many attempts have been made to establish specic immunohistochemical markers for ameloblastomas (20, 3437). For
J Oral Pathol Med 32: 48691

489

Dunsche et al.

example, it has been suggested that differences in the expression of cell surface carbohydrates with blood group specicity may distinguish ameloblastomas from odontogenic cysts (34), although this hypothesis was not conrmed by other authors (33). Another approach aimed at the evaluation of proliferating cell nuclear antigen (PCNA) in the cystic tumor lining of UA and found signicantly more PCNA-positive cells in UA than in dentigerous cyst linings (37). Recently, it has been suggested that calretinin is a specic immunohistochemical marker for neoplastic ameloblastic epithelium and may serve as a diagnostic tool for differentiating cystic odontogenic lesions from ameloblastic tumors (20). However, up to date, no general recommendations exist. Odontogenic cell nests are a frequent nding in the connective tissue that is associated with odontogenic cysts (38). Some of them may show palisading of basaloid cells located at the outer circumference while other signs of true ameloblastic differentiation are lacking, such as basal cell hyperchromatism, polarization of the basal cell nuclei, or vacuolation of the basal cell cytoplasm (23). In our study, extensive step sectioning revealed rather large odontogenic cell nests in four out of 101 cases (3.96%). They, however, lacked further criteria of ameloblastic differentiation. As these cell nests were in the close vicinity of the cysts and did not reach the margins of the specimens, it could be assumed that they were completely resected. In particular, as there was no true ameloblastic differentiation, the patients were not treated further and special follow-up was not thought to be required. We suggest that at present a histologic examination is the most sensitive tool for differentiating between odontogenic cysts and UAs. However, both clinical and radiologic ndings contribute to the diagnosis. Based on a series of 33 cases of the dentigerous variant of UA, it was suggested that the involved tooth crown is displaced by the cystic tumor rather than being projected into the cyst lumen (7). Because certain types of UA require radical resection (2, 46, 10, 1214), we saw a need to test the reliability of the current practices in routine histology. Our results showed that step sectioning of 101 specimens failed to reveal any case of UA not detected with conventional methods. Therefore, we suggest that step sectioning will not improve the reliability of the differential diagnosis of UA versus DC signicantly. In future, the use of specic immunohistochemical markers for UA might be a valuable tool in the differential diagnosis of DC versus UA (20, 3335). After all, if only two sections are examined in cases of cystic lesions that appear to be DCs, certain minor ndings might be missed, as demonstrated in this study: one case with papillary proliferation of metaplastic squamous epithelium (Fig. 4) and four cases with odontogenic islands within the cystic wall (Fig. 3). None of these ndings were visible in the two elective sections

examined in the course of routine histology; however, none of these ndings would have had an impact on the prognosis or the course of therapy. The patients included in this study were between 11 and 82 years of age (mean 46.5 years, SD 16.8 years). This differs from the data of other authors, who have reported that DCs occurred primarily between the ages of 20 and 39 years (39, 40; Fig. 1). We suggest that this may be because of the fact that a minimum size of 15 mm was required for entry into our study. In cases of both DC and UA, inammatory inltrates are a common nding, whereas conventional ameloblastomas rarely develop inammatory inltrates (25). In our study, 68.3% of all cystic lesions showed inammatory inltrates. These consisted of a small number of lymphocytes and a few plasma cells, but did not lead to clinical symptoms. Inammation is not a feature of DCs, but it frequently occurs when there is a connection to the oral cavity, which then leads to secondary inammation. According to our results, the examination of two sections of cystic lesions with the clinical and radiographic appearance of a DC seems to be appropriate because no unicystic amleoblastomas have been misdiagnosed.

References
1. Small I, Waldron C. Ameloblastomas of the jaws. Oral Surg Oral Med Oral Pathol 1955; 8: 28197. 2. Gardner DG. Plexiform unicystic ameloblastoma: a diagnostic problem in dentigerous cysts. Cancer 1981; 47: 135863. 3. Rapidis AD, Angelopoulos AP, Skouteris CA, Papanicolaou S. Mural (intracystic) ameloblastoma. Int J Oral Surg 1982; 11: 16674. 4. Robinson L, Martinez MG. Unicystic ameloblastoma: a prognostically distinct entity. Cancer 1977; 40: 227885. 5. Ackermann GL, Altini M, Shear M. The unicystic ameloblastoma: a clinicopathological study of 57 cases. J Oral Pathol 1988; 17: 5416. 6. Philipsen HP, Reichart PA. Unicystic ameloblastoma. A review of 193 cases from the literature. Oral Oncol 1998; 34: 31725. 7. Li TJ, Wu YT, Yu SF, Yu GY. Unicystic ameloblastoma: a clinicopathologic study of 33 Chinese patients. Am J Surg Pathol 2000; 24: 138592. 8. Kramer IRH, Pindborg JJ, Shear M. Histological Typing of Odontogenic Tumors. Berlin: Springer, 1992; 114. 9. Shteyer A, Lustmann J, Lewin-Epstein J. The mural ameloblastoma: a review of the literature. J Oral Surg 1978; 36: 86672. 10. Leider AS, Eversole LR, Barkin ME. Cystic ameloblastoma. A clinicopathologic analysis. Oral Surg Oral Med Oral Pathol 1985; 60: 62430. 11. Shear M. Cysts of the Oral Regions, 3rd edn. Oxford: Wright, 1992; 7598. 12. Thompson IO, Ferreira R, Van Wyk CW. Recurrent unicystic ameloblastoma of the maxilla. Br J Oral Maxillofac Surg 1993; 31: 1802.

490

J Oral Pathol Med 32: 48691

Dentigerous cyst versus unicystic ameloblastoma

13. Gardner DG, Pecak AM. The treatment of ameloblastoma based on pathologic and anatomic principles. Cancer 1980; 46: 25149. 14. Gardner DG, Corio RL. Plexiform unicystic ameloblastoma. A variant of ameloblastoma with a low-recurrence rate after enucleation. Cancer 1984; 53: 17305. 15. Feinberg SE, Steinberg B. Surgical management of ameloblastoma. Current status of the literature. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1996; 81: 3838. 16. Gardner DG. A pathologist's approach to the treatment of ameloblastoma. J Oral Maxillofac Surg 1984; 42: 1616. 17. Shatkin S, Hoffmeister FS. Ameloblastoma: a rational approach to therapy. Oral Surg Oral Med Oral Pathol 1965; 20: 42135. 18. Marx RE, Smith BH, Smith BR, Fridrich KL. Swelling of the retromolar region and cheek associated with limited opening. J Oral Maxillofac Surg 1993; 51: 3049. 19. Piattelli A, Fioroni M, Santinelli A, Rubini C. Expression of proliferating cell nuclear antigen in ameloblastomas and odontogenic cysts. Oral Oncol 1998; 34: 40812. 20. Coleman H, Altini M, Ali H, Doglioni C, Favia G, Maiorano E. Use of calretinin in the differential diagnosis of unicystic ameloblastomas. Histopathology 2001; 38: 3127. 21. Tozaki M, Hayashi K, Fukuda K. Dynamic multislice helical CT of maxillomandibular lesions: distinction of ameloblastomas from other cystic lesions. Radiat Med 2001; 19: 22530. 22. Gardner DG. Some current concepts on the pathology of ameloblastomas. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1996; 82: 6609. 23. Vickers RA, Gorlin RJ, Ameloblastoma. Delineation of early histopathologic features of neoplasia. Cancer 1970; 26: 699710. 24. Reichart PA, Philipsen HP, Sonner S. Ameloblastoma: biological profile of 3677 cases. Eur J Cancer B Oral Oncol 1995; 31B: 8699. 25. McMillan MD, Smillie AC. Ameloblastomas associated with dentigerous cysts. Oral Surg Oral Med Oral Pathol 1981; 51: 48996. 26. Triantafyllou A, Economopoulou P. Globular hyaline masses in the stroma of ameloblastoma: histopathologic and histochemical study. Ann Dent 1990; 49: 259. 27. Kahn MA. Ameloblastoma in young persons: a clinicopathologic analysis and etiologic investigation. Oral Surg Oral Med Oral Pathol 1989; 67: 70615. 28. Gardner DG, Corio RL. The relationship of plexiform unicystic ameloblastoma to conventional ameloblastoma. Oral Surg Oral Med Oral Pathol 1983; 56: 5460. 29. Rosenstein T, Pogrel MA, Smith RA, Regezi JA. Cystic ameloblastoma behavior and treatment of 21 cases. J Oral Maxillofac Surg 2001; 59: 13116; discussion 13168.

30. Gardner DG. Critique of the 1995 review by Reichart et al. of the biologic profile of 3677 ameloblastomas. Oral Oncol 1999; 35: 4439. 31. Ikeshima A, Ozawa M, Yamamoto H, Araki M, Sairenji E. Differential diagnosis between cyst and tumor. Dentigerous cyst and ameloblastoma containing teeth. J Nihon University Sch Dent 1990; 32: 1926. 32. Eversole LR, Leider AS, Strub D. Radiographic characteristics of cystogenic ameloblastoma. Oral Surg Oral Med Oral Pathol 1984; 57: 5727. 33. Gardner DG, O'Neill PA. Inability to distinguish ameloblastomas from odontogenic cysts based on expression of blood cell carbohydrates. Oral Surg Oral Med Oral Pathol 1988; 66: 4802. 34. Vedtofte P. Distribution of type 1 and 2 blood group chains in normal and pathological odontogenic epithelium defined by monoclonal antibodies specific for Lea and H type 2. Acta Pathol Microbiol Immunol Scand [A] 1985; 93: 26576. 35. Saku T, Shibata Y, Koyama Z, Cheng J, Okabe H, Yeh Y. Lectin histochemistry of cystic jaw lesions: an aid for differential diagnosis between cystic ameloblastoma and odontogenic cysts. J Oral Pathol Med 1991; 20: 10813. 36. Vigneswaran N, Whitaker SB, Budnick SD, Waldron CA. Expression patterns of epithelial differentiation antigens and lectin-binding sites in ameloblastomas: a comparison with basal cell carcinomas. Hum Pathol 1993; 24: 4957. 37. Li TJ, Browne RM, Matthews JB. Expression of proliferating cell nuclear antigen (PCNA) and Ki-67 in unicystic ameloblastoma. Histopathology 1995; 26: 21928. 38. Sciubba JS, Fantasia JE, Kahn LB. Tumors and cysts of the jaws. In: Rosai J, Sobin L (eds). Atlas of Tumor Pathology, 3rd edn. Fascicle 29. Washington, USA: Armed Forces Institute of Pathology, 2001. 39. Roggan R, Donath K. Klinik und Pathomorphologie odontogener rer Zysten. Dtsch Zahna follikula rztl Z 1985; 40: 53640. 40. Shear M. Developmental odontogenic cysts. An update. J Oral Pathol Med 1994; 23: 111.

Acknowledgements
We gratefully acknowledge the nancial support of the Department of Oral and Maxillofacial Surgery and the Department of Pathology of the Unirle for his constant versity of Kiel. We would like to thank Prof. Dr. F. Ha support to our work.

J Oral Pathol Med 32: 48691

491