Odontogenic keratocyst and ameloblastoma: radiographic evaluationJira Kitisubkanchana

(1)

https://doi.org/10.1007/s11282-020-00425-2 ORIGINAL ARTICLE

Odontogenic keratocyst and ameloblastoma: radiographic evaluation

Jira Kitisubkanchana¹ · Nor Hidayah Reduwan² · Sopee Poomsawat³ · Suchaya Pornprasertsuk‑Damrongsri¹ · Chanchai Wongchuensoontorn⁴

Received: 20 November 2019 / Accepted: 18 January 2020 / Published online: 6 February 2020

Abstract

Objectives To describe the radiographic features of odontogenic keratocysts (OKCs) and ameloblastomas and to compare the radiographic findings between these 2 lesions.

Methods Radiographs of OKCs and ameloblastomas were retrospectively reviewed. Location, border, shape, association with impacted tooth, tooth displacement, root resorption, and bone expansion were evaluated. Chi-squared or Fisher’s exact tests were used for statistical analysis. A p value < 0.05 was considered to indicate statistical significance.

Results One hundred OKCs and 101 ameloblastomas were reviewed. The ratios of maxilla to mandible were 1:1.4 and 1:9.1 in OKCs and ameloblastomas, respectively. All evaluated features significantly differed between OKCs and ameloblastomas (p ≤ 0.001). Most OKCs showed smooth border (60%) and unilocular shape (82%), while most ameloblastomas showed scalloped border (77.2%) and multilocular shape (68.3%). Association with impacted tooth was found in 47% of OKCs and 18.8% of ameloblastomas. Adjacent tooth displacement was found in 33.7% of OKCs and 55.8% of ameloblastomas. Root resorption was more common in ameloblastomas (66.7%) than in OKCs (7%). Bone expansion was also more common in ameloblastomas (96.3%) than in OKCs (63.6%).

Conclusion A unilocular radiolucent lesion with smooth border, no adjacent tooth displacement, no root resorption and causing mild or no bone expansion is suggestive of an OKC rather than an ameloblastoma.

Keywords Ameloblastoma · Differential diagnosis · Odontogenic keratocyst

Introduction

Odontogenic keratocyst (OKC) and ameloblastoma are common odontogenic lesions. OKC was classified as a tumor (keratocystic odontogenic tumor) by the World Health Organization (WHO) in 2005 due to its aggressive behavior, high recurrence and mutations in PTCH gene [1]. However,

WHO reclassified it again as OKC in 2017 because of insuf- ficient evidence to support the neoplastic origin [2]. OKC accounts for 7–20% of cystic lesions of the jaws [2–4] and ameloblastoma accounts for 13–54% of jaw tumors [5–7].

OKC is the third most common cyst of the jaws [2], while ameloblastoma is the most common odontogenic tumor [8].

Clinically, OKC and ameloblastoma may occur among patients at the same age distribution. The posterior region of the mandible is the common location for both lesions [5, 9].

Radiographically, OKC and ameloblastoma may show some similar radiographic features such as a well-defined unilocular or multilocular radiolucency associated or not associated with an unerupted tooth [9]. Various pathologies showing multilocular radiolucency may mimic OKC or ameloblastoma including glandular odontogenic cyst, traumatic bone cyst, central giant cell granuloma, odontogenic myxoma, and fibro-osseous lesion [10–12]. However, in some instances, radiographic features of OKC or ameloblastoma showing unilocular radiolucency associated with unerupted tooth may similar to those of dentigerous cyst.

* Jira Kitisubkanchana yjira@hotmail.com

1 Department of Oral and Maxillofacial Radiology, Faculty of Dentistry, Mahidol University, No. 6, Yothi Road, Ratchathewi District, Bangkok 10400, Thailand

2 Faculty of Dentistry, Centre of Oral and Maxillofacial Diagnostic and Medicine Studies, Universiti Teknologi Mara, Sungai Buloh Campus, Shah Alam, Malaysia

3 Department of Oral and Maxillofacial Pathology, Faculty of Dentistry, Mahidol University, Bangkok, Thailand

4 Department of Surgery and Oral Medicine, Faculty of Dentistry, Srinakharinwirot University, Bangkok, Thailand

(2)

OKC and ameloblastoma have different biological behav- iors; therefore, management of these lesions differs. How- ever, differential diagnosis of these two lesions is difficult because they share many clinical and radiographic features. Therefore, it is difficult to differentiate these lesions radiographically, and definitive diagnosis is based only on histopathologic examination. However, in many instances incisional biopsy could not reveal a definitive diagnosis especially from large lesions or lesions with inflammation.

Thus, to diagnose such lesions, differences in radiographic findings of these two lesions may play an important role in making the differential diagnosis.

Advanced imaging techniques such as computed tomography and magnetic resonance imaging have been used to differentiate these lesions [13–18]. However, accessibility to these modalities is occasionally limited, and conventional radiography is still the common imaging technique used by most dentists to investigate these lesions. Therefore, information from conventional images may benefit in making differential diagnosis when evaluated meticulously.

It has been suggested that some radiographic findings could be used to differentiate OKC from ameloblastoma [9, 19, 20]. OKC has a propensity to grow along the internal aspect of the mandible causing minimal expansion [9], and is less likely to cause root resorption compared with ameloblastoma [19–21]. Although studies about radiographic features of OKCs and ameloblastomas have been reported [17, 19, 21–34], details of the radiographic features are still limited. Many aspects including displacement of adjacent teeth, root resorption and bone expansion observed on radiographs are available, but these information were derived from different small series. There are only three studies that investigated all these radiographic findings [27, 30, 31]. Gumusok et al. [27] investigated 28 OKCs, MacDonald-jankowski and Li [30] investigated 33 OKCs, and MacDonald-jankowski et al. [31] investigated 61 ameloblastomas. Thus, the aim of the present study was to describe the details of the radiographic features of OKCs and ameloblastomas in a large number of cases. In addition, we aimed to compare the radiographic findings between these lesions. To our knowl- edge, the radiographic features between OKCs and ameloblastomas have never been compared. We believe that this information may help in making differential diagnosis of these lesions.

Materials and methods

Between the years 2003 and 2016, we performed a retrospective review of cases histopathologically diagnosed as OKCs or keratocystic odontogenic tumors and ameloblastomas.

One investigator (S.P.) re-evaluated the hematoxylin and eosin–stained sections and recurred cases were excluded.

Conventional radiographs and/or cone beam computed tomography (CBCT) images of these patients were retrieved.

All images were evaluated by two observers who are special- ists in oral and maxillofacial radiology (J.K. and N.H.R). A consensus was reached after discussion when the observers disagreed. The following features were evaluated: location, border, shape, relationship of the lesion with the impacted tooth, displacement of adjacent teeth, root resorption and bone expansion.

The location was determined by the radiographic mar- gin of the lesion. The maxilla was divided in two anatomic regions on each side: (1) the anterior region, extending from the midline to the distal surface of the canine and (2) the posterior region, extending from the mesial aspect of the first premolar to the distal surface of maxillary tuberosity. The mandible was divided in two anatomic regions on each side:

(1) the anterior region, extending from the midline to the distal surface of the canine, (2) the posterior region, which was subdivided in three regions: (2.1) the body region, extending from the mesial aspect of the first premolar to the angle of the mandible; (2.2) the ramus region, extending from the angle of the mandible to the sigmoid notch and (2.3) the condyle region, extending from the sigmoid notch to the con- dylar region. The border was defined as either (1) smooth, a border showing an even surface free from indentation or (2) scalloped, a border showing a series of contiguous arcs or semicircles (Fig. 1). The shape was defined as unilocular or multilocular (Fig. 2). The impacted tooth associated with the lesion was recorded. The relationship between the lesion and the impacted tooth was classified in three groups: (1) cementoenamel junction (CEJ): lesion surrounds the tooth and is attached to the tooth at the CEJ; (2) root: lesion surrounds the tooth and extends apically along the root beyond the CEJ and (3) entire tooth: lesion surrounds entire tooth.

Tooth displacement and root resorption were recorded as yes or none. Bone expansion was recorded as (1) distinct:

when at least one cortical bone expands more than 5 mm from the normal contour of the bone or maxillary sinus wall;

(2) mild: when at least one cortical bone expands less than or equal to 5 mm from the normal contour of the bone or maxillary sinus wall; and (3) none: when no bone expansion occurred (Fig. 3).

For image acquisition, different imaging systems were used. From 2003 to 2011, Ultraspeed or Insight (East- man Kodak, Rochester, NY) periapical radiographic films were exposed with a GX1000 (Gendex, IL, USA) or Searcher Dx-068 (Takara, Belmont, Osaka, Japan).

Since 2012, intra-oral radiographs were taken with Plan- meca ProX (Planmeca, Finland) using a phosphor plate system (VistaScan^®, Dürr Dental, Bietigheim-Bissingen, Germany). Panoramic and other conventional extra-oral radiographs when available, i.e., postero-anterior skull and postero-anterior mandibular radiographs, were obtained

(3)

with Orthopan Tomograph OP 100 (Trophy, France), PM 2002 EC Proline (Planmeca, Finland) or CS 9000 machine (Carestream Health, Inc., Rochester, NY, USA).

All images which were not taken with a digital system were digitized using a Microtek ScanMaker 9800XL (Microtek Inc. Santa Fe Spring, CA, USA) with a resolution of 300 dpi. Conventional images were assessed using ImageJ Software (NIH, Bethesda, MD, USA, https ://rsb.info.nih.gov/ij). CBCT images were captured using a 3D Accuitomo 170 (J Morita, Kyoto, Japan) or CS 9500 machine (Carestream Health, Inc., Rochester, NY, USA), and images were assessed using the software equipped with the machines. An 18.0-in., light-emitting diode, high- definition screen (resolution 1366 × 768 pixels) was used for image assessment. The examiners were allowed to use the zoom tool and to adjust the brightness and contrast of the images.

Data were analyzed using SPSS, Version 19 (IBM Corp., Armonk, NY, USA). The Chi-squared or Fisher’s exact tests was used to determine evaluated features between OKCs and ameloblastomas. A p value less than 0.05 was considered to be statistically significant.

Results

A total of 89 cases comprised 100 lesions diagnosed as OKC and 101 cases diagnosed as ameloblastoma. All patients were Thai. Patients’ age ranged from 10 to 87 years (mean 31.4 years) and from 3 to 87 years (mean 34.9 years) in OKCs and ameloblastomas, respectively. OKC patients comprised 46 females and 43 males. Among all OKC cases, 6 patients (2 females, 4 males) were associated with nevoid basal cell carcinoma syndrome confirmed by means of

Fig. 1 Cropped panoramic images illustrating border of the lesions. a A lesion with smooth border. b A lesion with scal- loped border

Fig. 2 Images illustrating shape of the lesions. a A lesion with unilocular shape. b A lesion with multilocular shape

(4)

diagnostic criteria [35] and multiple OKCs had developed in all these patients with a total of 17 OKCs. A total of 53 females and 48 males presented ameloblastoma. Each case had at least one conventional radiograph. CBCT images were available in 30 OKCs (16 lesions in the maxilla, 14 lesions in the mandible), 40 ameloblastomas (4 cases in the maxilla, 36 cases in the mandible). All radiographic features were evaluated based on all available radiographs.

The lesion locations are summarized in Table 1. Regard- ing OKCs, 41 lesions were located in the maxilla and 59 lesions were in the mandible. The posterior region was the most common location for OKCs in both maxilla and mandible (27 OKCs and 51 OKCs, respectively). Regard- ing ameloblastoma, 10 lesions were located in the maxilla and 91 lesions were located in the mandible. Most ameloblastomas in the maxilla extended from anterior to posterior regions (8 cases), whereas most ameloblastomas in the mandible were found in the posterior region (52 cases). Lesions crossed midline in 14 cases of OKCs (7 cases in the maxilla and 7 cases in the mandible) and 29 cases of ameloblastomas (all in the mandible).

Radiographic findings of all cases are summarized in Table 2. Statistical analysis revealed significant differences in all evaluated features between OKCs and amelo- blastomas (p ≤ 0.001). In OKCs, 60 of 100 OKCs showed smooth border. Eighty-two OKCs (82%) showed unilocular shape. There were 47 OKCs associated with impacted teeth.

Among these cases, 28 OKCs were located in the maxilla and 19 OKCs were located in the mandible. The relationship between lesion and the associated teeth could be evaluated in 42 OKCs. The cysts attached to the CEJ of the impacted teeth in 17 of 42 lesions (40.5%), extended apically along the root of the impacted teeth in 10 of 42 lesions (23.8%),

Fig. 3 Occlusal cross-sectional images. The distance of expansion was measured from the most expanded point to the point of previ- ous normal contour perpendicular to bone surface. a A lesion with

distinct bone expansion. b A lesion with mild bone expansion. c A lesion at the third molar area without bone expansion (arrows)

Table 1 Distribution of lesions by location in the jaws

op opposite

Location Number of lesions

OKC Ameloblastoma

Maxilla 41 10

Anterior 4 1

Anterior 3 1

Anterior-op anterior 1 0

Posterior 27 1

Anterior–posterior 10 8

Anterior–op posterior 6 0

Mandible 59 91

Anterior 0 3

Anterior 0 1

Anterior-op anterior 0 2

Posterior 51 52

Body 17 10

Ramus 6 1

Body-ramus 18 23

Ramus-condyle 1 0

Body-ramus-condyle 5 9

Body-op body 3 9

Body-op ramus 1 0

Anterior-body 4 15

Anterior-body-ramus 0 2

Anterior-body-ramus-condyle 1 1

Anterior-op body 3 17

Anterior-op ramus 0 1

Total lesions 100 101

(5)

and enclosed the entire teeth in 15 of 42 lesions (35.7%). An example of OKC was illustrated in Fig. 4.

Among the ameloblastomas, 78 cases (77%) showed scalloped border and 69 cases (68%) showed multilocular shape.

Association with impacted tooth was found in 19 of 101 cases (18.8%). Of these cases, only 1 lesion was located in the maxilla and 18 cases were located in the mandible.

Lesions were attached to the tooth at the CEJ of the impacted teeth in 8 lesions (42.1%) and extended apically along the root of the impacted teeth in 5 lesions (26.3%), and enclosed the entire teeth in 6 lesions (31.6%). An example of ameloblastoma was illustrated in Fig. 5.

Adjacent tooth displacement and root resorption were not evaluated in all lesions because some lesions occurred in edentulous areas or areas that were not tooth-bearing.

Hence, adjacent tooth displacement could be evaluated in 86 OKCs and 95 ameloblastomas. Adjacent tooth displacement was found in 33.7% of OKCs and 55.8% of ameloblastomas. Root resorption could be investigated in 85 OKCs and 93 ameloblastomas. It was found in only 7%

of OKCs compared to 66.7% of ameloblastomas. Bone expansion was evaluated in 66 OKCs and 82 ameloblastomas. Of these cases, 30 OKCs and 39 ameloblastomas were studied from CBCT images. The remaining cases were examined with plain radiographs taken by various techniques showing information in three different planes.

Among OKCs, 22 lesions (33.3%) revealed distinct bone expansion, 20 OKCs (30.3%) showed mild bone expansion, and 24 OKCs (36.4%) showed no bone expansion.

While among ameloblastomas, 72 cases (87.8%) showed distinct bone expansion, 7 cases (8.5%) showed mild bone expansion, and only 3 cases (3.7%) showed no bone expansion.

Radiographic findings of different types of ameloblastoma are summarized in Table 3. In this study, the types of ameloblastoma were classified according to WHO 2017 [2].

Seventy-four lesions were conventional type consisting of 62 solid ameloblastomas and 12 desmoplastic ameloblastomas.

Twenty-seven lesions were unicystic type and all of them were located in the mandible.

Table 2 Comparison of the radiographic characteristics between OKC and ameloblastoma

NA data not available

*Significant at p < 0.05 Radiographic

characteristic Number of lesions p value

OKC Ameloblastoma

Maxilla

(n = 41) Mandible

(n = 59) Total (n = 100) Maxilla

(n = 10) Mandible

(n = 91) Total (n = 101) Border

Smooth 33 27 60 5 18 23 < 0.001*

Scallop 8 32 40 5 73 78

Shape

Unilocular 36 46 82 5 27 32 < 0.001*

Multilocular 5 13 18 5 64 69

Relation between radiolucent lesion and impacted tooth

None 13 40 53 9 73 82 < 0.001*

Yes 28 19 47 1 18 19

CEJ 8 9 17 0 8 8

Root 4 6 10 0 5 5

Entire tooth 13 2 15 1 5 6

NA 3 2 5 0 0 0

Adjacent tooth displacement^a

Yes 15 14 29 7 46 53 0.001*

None 25 32 57 3 39 42

Root resorption^b

Yes 1 5 6 4 58 62 < 0.001*

None 39 40 79 6 25 31

Bone expansion

Distinct 10 12 22 4 68 72 0.001*

Mild 8 12 20 0 7 7

None 4 20 24 0 3 3

NA 19 15 34 6 13 19

(6)

Fig. 4 An OKC in a 29-year-old woman—a cropped panoramic image shows a lesion at the right ramus-condyle region with smooth border, unilocular shape. b axial, c sagittal, and d coronal CBCT images clearly depict the lesion with mild bone expansion (white arrows)

Fig. 5 Large ameloblastoma in a 55-year-old woman at the mandible—a cropped pano- ramic image, b axial, c sagittal and d coronal CBCT images display a lesion with scalloped border and multilocular shape.

There is tooth displacement of the right canine and root resorption of the related teeth (aster- isks). Distinct bone expansion is observed (white arrows)

(7)

Discussion

OKC and ameloblastoma might show similar radiographic findings, leading to difficulty in providing differential diagnosis. Many studies have investigated these lesions regarding treatment and recurrence rate. However, details of the radiographic findings of these lesions, including association with impacted tooth, displacement of adjacent teeth, root resorption and bone expansion remain limited as summarized in Tables 4 and 5. Therefore, we aimed to study the image features of OKCs and ameloblastomas to help differentiate between them. The mean age of patients in OKCs and ameloblastomas in this study was in agreement with related reports [30, 32, 33, 36]. It has been reported that a slight male predominance exists regarding both OKC and ameloblastoma [9, 17, 22–25, 28–30, 33, 34, 36–38]. However, no sex predilection of OKCs and ameloblastomas was found in the present study.

In this study, the posterior region of the mandible was the predominant site in both OKCs and ameloblastomas.

These results were similar to many reports [26, 27, 34, 38].

Interestingly, many OKCs in the present study were also found at the posterior region of the maxilla. Our results were in accordance with Neville et al. [12] and Regezi et al.

[39] who suggested that the OKCs in the maxilla were frequently located in the posterior region. A study in Singapore showed slightly different results. Of 19 OKCs in the maxilla, 9 lesions were in the anterior region, 8 lesions were in the posterior region, and 2 lesions were in anterior to posterior region [25]. Among the ameloblastomas, the ratio of the maxilla to the mandible in our study was 1:9.1. This was similar to earlier studies evaluating a large series of ameloblastomas [26, 34].

Our results showed that OKCs and ameloblastomas showed significant differences of all evaluated imaging features. With respect to the shape of the lesions, 82% of OKCs in the present study showed unilocular shape. This result

Table 3 Comparison of the radiographic characteristics among different types of ameloblastomas

NA data not available

a n = 60 solid type, 12 desmoplastic type and 23 unicystic type

b n = 58 solid type, 12 desmoplastic type and 23 unicystic type

Radiographic characteristic Conventional ameloblastoma Unicystic ameloblastoma

Solid type Desmoplastic type

Maxilla

(n = 4) Mandible

(n = 58) Total (n = 62) Maxilla

(n = 5) Mandible

(n = 7) Total (n = 12) Mandible (n = 27) Border

Smooth 2 10 12 2 0 2 9

Scallop 2 48 50 3 7 10 18

Shape

Unilocular 2 15 17 2 0 2 13

Multilocular 2 43 45 3 7 10 14

Relation between radiolucent lesion and associated tooth

None 4 47 51 5 7 12 19

Yes 0 11 11 0 0 0 8

CEJ 0 5 5 0 0 0 3

Root 0 2 2 0 0 0 3

Entire tooth 0 4 4 0 0 0 2

Adjacent tooth displacement^a

Yes 2 34 36 4 5 9 8

None 2 22 24 1 2 3 15

Root resorption^b

Yes 2 39 41 1 2 3 18

None 2 15 17 4 5 9 5

Bone expansion

Distinct 1 42 43 2 6 8 21

Mild 0 5 5 0 0 0 2

None 0 1 1 0 1 1 1

NA 3 10 13 3 0 3 3

(8)

Table 4 Comparison of clinical and radiographic findings of OKCs with previous reports Mand mandibular, NA data not available, Post posterior region, M molar, Max maxillary, C canine a All studied lesions are in the mandible

Author

Number of OK

CsMean age (year)Male: female

Maxilla: mandible

Most common siteUnilocular: multilocular

Association with impacted tooth

Most commonly

associated impacted tooth

Displacement of adjacent teeth

Root resorptionBone expansion Alves et al. [17]934.41:0.8NAMand post1:0.344%Mand third M and CNA13%77% Chirapathom- sakul et al. [21]

6736.91:1.21:2.2Mand post1:0.431.3%Third MNA1.5%NA Apajalahti et al. [22]46461:0.51:1.4Mand molar,

ramus and angle

NA28%Third M17%13%NA Boffano et al. [23]26143.31:0.51: 2.7Mand ramus and angle1:0.2NANANANANA Buckley et al. [24]83NA1:0.7NAMand post1:0.9NANA39%15%NA Chow HT [25]7632.81:0.61:2.5Mand postNA52.8%Mand third MNANANA Gumusok et al. [27]2834.51:11:3.7Mand molar, retromolar and ramus

1:139%Mand third M100%30%26% Habibi et al. [28]8327.11:0.71:2.1Mand postNA33.7NANANANA

MacDonald- Jank

owski and Li [30]

3330.61:0.81:1.5Mand post1:1.156%Max third M69%41%82% Sansare et al. [33]7230.71:0.31:2.6Mand post1:0.6NANANANANA Myoung et al. [36]25630.81:0.71:2.1Mand molarNANANANANANA Simiyu et al. [37]2227.51:0.71:2.5Mand post1:1.49.1%Mand third M and max CNANANA Titinchi and Nortje [38]14534.51:0.61:3Mand post1:0.452.4%Mand third MNA0.7%NA This study10031.41:1.11:1.4Mand post and ramus1:0.342%Max third M33.7%7%64%

(9)

Table 5 Comparison of clinical and radiographic findings of ameloblastomas with previous reports NA data not available, Mand mandibular, Post posterior region, M molar, DC dentigerous cyst a All studied lesions are in the mandible b All cases are unicystic ameloblastoma

Author

Number of ameloblas

tomas

Mean age (y)Male: female

Maxilla: mandible

Most common siteUnilocular: multilocularAsso- ciation with impacted tooth

Most commonly

associated impacted tooth

Displace-

ment of adjacent teeth

Root resorptionBone expansion Reichart et al. [5]3677361:11:5Mand Post1:1NANANANANA Alves et al. [17]931.81:0.8NAMand Post1:0.344%Mand third MNA75%89% Dhanuthai et al. [26]128938.31:11:9.8Mand Post1:110.4%NANANANA Kim et al. [29]7130.41:0.81:6.8Mand Post1:0.4NANANANANA

MacDonald- Jank

owski et al. [31]

6130.51:11:5.1Mand Post1:0.639%Mand third M73%59%100%

Philipsen and R

eich- art [32]

193 Initially diag-

nosed as DC

9016.51:0.71:3–13Mand Post1:0.2NAMand third MNANANA Initially diag-

nosed as non-DC

10335.21:1.81:3–13Mand Post1:0.9––NANANA Siar et al. [34]34030.31:0.71:10.7Mand Post1:1.83.4%NANA6.7%NA This study10134.91:1.11:9.1Mand Post1:2.219%Mand third M55.8%66.7%96.3%

(10)

is in agreement with many previous studies showing that most OKCs showed unilocular shape [17, 21, 23, 33, 38].

In ameloblastomas, the results of shape according to previous studies were still controversial. Some studies reported that the ratio of unilocular to multilocular shape was 1:1 [5, 26], while other studies documented that most ameloblastomas showed unilocular shape [17, 29, 31, 32]. Another study reported that multilocular shape was more prominent in ameloblastomas [34]. The present study found that more cases of ameloblastomas showed multilocular shape (68.3%) than unilocular shape (31.7%).

Regarding the association with impacted tooth, OKCs showed a greater frequency than ameloblastomas in both maxilla and mandible. Notably, almost all ameloblastomas in the maxilla were unassociated with impacted teeth. Only 1 ameloblastoma in the posterior maxilla entirely enclosed an impacted third molar. Therefore, we speculated that a lesion in the maxilla, associated with an impacted tooth was likely to be an OKC rather than an ameloblastoma. Related studies reported that 17–100% of OKCs showed displacement of adjacent teeth [22, 24, 27, 30]. Most studies of ameloblastomas did not evaluate this effect. Only 1 study reported that this effect was found in 73% of ameloblastomas [31]. In the present study, displacement of adjacent teeth was found in more cases of ameloblastomas (55.8%) than in OKCs (33.7%).

Our results showed that 66.7% of ameloblastomas caused root resorption, whereas only 7% of OKCs showed this effect. This was similar to the literature showing that root resorption was more frequently seen among ameloblastomas than among OKCs [17, 19]. With respect to bone expansion in OKCs, the results vary among studies. In our study, 64%

of OKCs caused bone expansion. Three studies reported that bone expansion was found in 26%, 77%, and 82% of OKCs [17, 27, 30]. Among ameloblastomas, a high per- centage of bone expansion as 89% and 100% of cases has been documented [17, 31]. Our study also found that 96.3%

of ameloblastomas caused bone expansion. Among these cases, distinct bone expansion was found in most ameloblastomas (87.8%), whereas this was found in only 33.3%

OKCs. The difference of this effect might be used as one of the radiographic findings to differentiate between 2 lesions as suggested by Ariji et al. [19], who revealed that buccolin- gual bone expansion was a significant feature to differentiate ameloblastomas from OKCs.

There were some limitations to the present study. To evaluate some radiographic features, i.e., relationship between lesion and the impacted tooth as well as bone expansion, conventional radiographs taken with various techniques are required. In this study, some cases had only one available radiograph. Therefore, these radiographic features could not be evaluated in these cases due to inadequacy of radiographs. In addition, although CBCT reveals the details of

the lesion in 3 dimensions, it was not prescribed for all cases due to additional cost and radiation dose to the patients. In cases without CBCT images, all studied radiographic features were investigated from 2 dimensional radiographs with inferior image quality compared to CBCT images.

Conclusion

No gender predilection was observed for OKCs and ameloblastomas. The most common location of both lesions was at the posterior region of the jaws. Almost one half of OKCs were found in the maxilla, while ameloblastomas were predominantly found in the mandible. All evaluated radiographic features showed significant differences between OKCs and ameloblastomas. Most OKCs showed smooth border and unilocular shape, while most ameloblastomas showed scalloped border and multilocular shape. Compared with ameloblastomas, OKCs showed greater frequency to be associated with impacted tooth, and were unlikely to cause tooth displacement and root resorption. We suggest that a radiolucent lesion showing smooth border, unilocular shape, no adjacent tooth displacement, no root resorption, with mild or no bone expansion is likely to be an OKC rather than an ameloblastoma. These radiographic findings might be helpful for differential diagnosis between them.

Compliance with ethical standards

Conflict of interest Jira Kitisubkanchana, Nor Hidayah Reduwan, Sopee Poomsawat, Suchaya Pornprasertsuk-Damrongsri, and Chan- chai Wongchuensoontorn declare that they have no conflict of interest.

Ethical approval All procedures followed were in accordance with the ethical standards of the responsible committee on human experi- mentation (the Institutional Review Board of the Faculty of Dentistry/

Faculty of Pharmacy, Mahidol University, COA No. MU-DT/PY-IRB 2017/007.0902) and with the Helsinki Declaration of 1975, as revised in 2008. According to the Institutional Review Board of our university, for this retrospective study, formal consent is not required.

References

1. Philipsen HP. Keratocystic odontogenic tumour. In: Barnes L, Eveson JW, Reichart P, Sidransky D, editors. World Health Organization classification of tumours. Pathology and genetics:

head and neck tumours. Lyon: IARC Press; 2005. p. 306–307.

2. El-Naggar AK, Chan John KC, Grandis JR, Takata T, Slootweg PJ. World Health Organization classification of head and neck tumours. Lyon: IARC Press; 2017.

3. Manor E, Kachko L, Puterman MB, Szabo G, Bodner L. Cystic lesions of the jaws—a clinicopathological study of 322 cases and review of the literature. Int J Med Sci. 2012;99(1):20–6.

4. Tamiolakis P, Thermos G, Tosios KI, Sklavounou-Andriko- poulou A. Demographic and clinical characteristics of 5294

(11)

jaw cysts: a retrospective study of 38 years. Head Neck Pathol.

2019;13(4):587–96.

5. Reichart PA, Philipsen HP, Sonner S. Ameloblastoma: biological profile of 3677 cases. Eur J Cancer B Oral Oncol.

1995;31B(2):86–99.

6. Siriwardena BSMS, Crane H, O’Neill N, Abdelkarim R, Brierley DJ, Franklin CD, et al. Odontogenic tumors and lesions treated in a single specialist oral and maxillofacial pathology unit in the United Kingdom in 1992–2016. Oral Surg Oral Med Oral Pathol Oral Radiol. 2019;127(2):151–66.

7. Ahire MS, Tupkari JV, Chettiankandy TJ, Thakur A, Agrawal RR.

Odontogenic tumors: a 35-year retrospective study of 250 cases in an Indian (Maharashtra) teaching institute. Indian J Cancer.

2018;55(3):265–72.

8. Worawongwasu R, Tiensuwan M. Odontogenic tumors in Thai- land: a study of 590 Thai patients. J Oral Maxillofac Surg Med Pathol. 2015;27(4):567–76.

9. Lam EWN. Cysts. In: Mallya SM, Lam EWN, editors. White and Pharoah’s oral radiology: principles and interpretation. 8th ed. St.

Louis: Elsevier; 2019. p. 391–395.

10. Chrcanovic BR, Gomez RS. Idiopathic bone cavity of the jaws: an updated analysis of the cases reported in the literature. Int J Oral Maxillofac Surg. 2019;48(7):886–94.

11. Chrcanovic BR, Gomez RS. Glandular odontogenic cyst: an updated analysis of 169 cases reported in the literature. Oral Dis.

2018;24(5):717–24.

12. Neville BW, Damm DD, Allen CM, Chi AC. Oral and maxillofacial pathology. 4th ed. St. Louis: Elsevier; 2016.

13. Kakimoto N, Chindasombatjaroen J, Tomita S, Shimamoto H, Uchiyama Y, Hasegawa Y, et al. Contrast-enhanced multidetec- tor computerized tomography for odontogenic cysts and cystic- appearing tumors of the jaws: is it useful? Oral Surg Oral Med Oral Pathol Oral Radiol. 2013;115(1):104–13.

14. Minami M, Kaneda T, Ozawa K, Yamamoto H, Itai Y, Ozawa M, et al. Cystic lesions of the maxillomandibular region: MR imaging distinction of odontogenic keratocysts and ameloblastomas from other cysts. AJR Am J Roentgenol. 1996;166(4):943–9.

15. Sumi M, Ichikawa Y, Katayama I, Tashiro S, Nakamura T. Dif- fusion-weighted MR imaging of ameloblastomas and keratocystic odontogenic tumors: differentiation by apparent diffu- sion coefficients of cystic lesions. AJNR Am J Neuroradiol.

2008;29(10):1897–901.

16. Crusoe-Rebello I, Oliveira C, Campos PS, Azevedo RA, dos Santos JN. Assessment of computerized tomography den- sity patterns of ameloblastomas and keratocystic odontogenic tumors. Oral Surg Oral Med Oral Pathol Oral Radiol Endod.

2009;108(4):604–8.

17. Alves DBM, Tuji FM, Alves FA, Rocha AC, Santos-Silva ARD, Vargas PA, et al. Evaluation of mandibular odontogenic keratocyst and ameloblastoma by panoramic radiograph and computed tomography. Dentomaxillofac Radiol. 2018;47(7):20170288.

18. Borghesi A, Nardi C, Giannitto C, Tironi A, Maroldi R, Di Bar- tolomeo F, et al. Odontogenic keratocyst: imaging features of a benign lesion with an aggressive behaviour. Insights Imaging.

2018;9(5):883–97.

19. Ariji Y, Morita M, Katsumata A, Sugita Y, Naitoh M, Goto M, et al. Imaging features contributing to the diagnosis of ameloblastomas and keratocystic odontogenic tumours: logistic regression analysis. Dentomaxillofac Radiol. 2011;40(3):133–40.

20. Wakoh M, Okawa Y, Otonari-Yamamoto M, Kamio T, Sakamoto J, Yamamoto A, et al. Reliance on diagnostic elements in panoramic imaging with focus on ameloblastoma and keratocystic odontogenic tumor: psychometric study. Bull Tokyo Dent Coll.

2011;52(1):1–12.

21. Chirapathomsakul D, Sastravaha P, Jansisyanont P. A review of odontogenic keratocysts and the behavior of recurrences. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 2006;101(1):5–9.

22. Apajalahti S, Hagstrom J, Lindqvist C, Suomalainen A. Com- puterized tomography findings and recurrence of keratocystic odontogenic tumor of the mandible and maxillofacial region in a series of 46 patients. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 2011;111(3):e29–37.

23. Boffano P, Ruga E, Gallesio C. Keratocystic odontogenic tumor (odontogenic keratocyst): preliminary retrospective review of epi- demiologic, clinical, and radiologic features of 261 lesions from University of Turin. J Oral Maxillofac Surg. 2010;68(12):2994–9.

24. Buckley PC, Seldin EB, Dodson TB, August M. Multilocularity as a radiographic marker of the keratocystic odontogenic tumor.

J Oral Maxillofac Surg. 2012;70(2):320–4.

25. Chow HT. Odontogenic keratocyst: a clinical experience in Sin- gapore. Oral Surg Oral Med Oral Pathol Oral Radiol Endod.

1998;86(5):573–7.

26. Dhanuthai K, Chantarangsu S, Rojanawatsirivej S, Phattarata- ratip E, Darling M, Jackson-Boeters L, et al. Ameloblastoma: a multicentric study. Oral Surg Oral Med Oral Pathol Oral Radiol.

2012;113(6):782–8.

27. Gumusok M, Toraman Alkurt M, Museyibov F, Ucok O. Evalu- ation of keratocystic odontogenic tumors using cone beam computed tomography. J Istanb Univ Fac Dent. 2016;50(3):32–7.

28. Habibi A, Saghravanian N, Habibi M, Mellati E, Habibi M.

Keratocystic odontogenic tumor: a 10-year retrospective study of 83 cases in an Iranian population. J Oral Sci. 2017;49(3):229–35.

29. Kim SG, Jang HS. Ameloblastoma: a clinical, radiographic, and histopathologic analysis of 71 cases. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 2001;91(6):649–53.

30. Macdonald-Jankowski DS, Li TK. Keratocystic odontogenic tumour in a Hong Kong community: the clinical and radiological features. Dentomaxillofac Radiol. 2010;39(4):167–75.

31. MacDonald-Jankowski DS, Yeung R, Lee KM, Li TK. Ameloblas- toma in the Hong Kong Chinese. Part 1: systematic review and clinical presentation. Dentomaxillofac Radiol. 2004;33(2):71–82.

32. Philipsen HP, Reichart PA. Unicystic ameloblastoma. A review of 193 cases from the literature. Oral Oncol. 1998;34(5):317–25.

33. Sansare K, Raghav M, Mupparapu M, Mundada N, Karjodkar FR, Bansal S, et al. Keratocystic odontogenic tumor: systematic review with analysis of 72 additional cases from Mumbai, India. Oral Surg Oral Med Oral Pathol Oral Radiol. 2013;115(1):128–39.

34. Siar CH, Lau SH, Ng KH. Ameloblastoma of the jaws: a retrospective analysis of 340 cases in a Malaysian population. J Oral Maxillofac Surg. 2012;70(3):608–15.

35. Hasan A, Akintola D. An update of Gorlin–Goltz syndrome. Prim Dent J. 2018;7(3):38–41.

36. Myoung H, Hong SP, Hong SD, Lee JI, Lim CY, Choung PH, et al. Odontogenic keratocyst: review of 256 cases for recurrence and clinicopathologic parameters. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 2001;91(3):328–33.

37. Simiyu BN, Butt F, Dimba EA, Wagaiyu EG, Awange DO, Guthua SW, et al. Keratocystic odontogenic tumours of the jaws and associated pathologies: a 10-year clinicopathologic audit in a referral teaching hospital in Kenya. J Craniomaxillofac Surg.

2013;41(3):230–4.

38. Titinchi F, Nortje CJ. Keratocystic odontogenic tumor: a recurrence analysis of clinical and radiographic parameters. Oral Surg Oral Med Oral Pathol Oral Radiol. 2012;114(1):136–42.

39. Regezi JA, Scuibba JJ, Jordan RCK. Oral pathology clinical patho- logic correlations. 7th ed. St. Louis: Elsevier; 2017. p. 253–257.

Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.