Comparison of Computed Tomography, Magnetic Resonance Imaging, and Positron Emission Tomography and Computed Tomography for the Evaluation Bone Invasion in Upper and Lower Gingival Cancers

Comparison of Computed Tomography, Magnetic Resonance Imaging, and Positron Emission Tomography and Computed Tomography for the Evaluation Bone Invasion in Upper and Lower Gingival Cancers Young Chan Lee, MD, PhD,* Ah Ra Jung, MD, PhD,y Oh Eun Kwon, MD,z Eui-Jong Kim, MD, PhD,x Il Ki Hong, MD, PhD,k Jung-Woo Lee, DDS, PhD,{ and Young-Gyu Eun, MD, PhD# Purpose:

Preoperative detection of bone invasion is important in cases of gingival cancer. The aim of this study was to compare the diagnostic value of 3 imaging methods for the detection of bone invasion in upper and lower gingival cancer: computed tomography (CT), magnetic resonance imaging (MRI), and positron emission tomography (PET) and CT.

Materials and Methods:

This retrospective cohort study enrolled patients who underwent a maxillectomy or a mandibulectomy for gingival cancer. Each preoperative image (CT, MRI, or PET/CT) was reviewed for the presence of bone invasion, and the possibility for bone invasion was graded. These results were verified with pathology reports. Sensitivity, specificity, positive predictive value, and negative predictive value for the detection of mandibular involvement in alveolar bone were calculated, and a receiver operating characteristics (ROC) curve analysis was performed.

Results:

Forty patients (27 men and 13 women) were enrolled. Pathologic examination disclosed bone invasion in 25 of the 40 patients. Of these patients, 13 had maxillary and 12 had mandibular alveolus involvement. The diagnostic accuracy of CT (90.0%) was highest among the 3 modalities for the detection of bone invasion. In the ROC curve analysis, values for the area under the curve for upper gingival cancer were lower than those for lower gingival cancer.

Received from Kyung Hee University, Seoul, Korea.

supported by a grant of the Korea Health Technology R&D Project

*Assistant Professor, Department of Otolaryngology–Head and

through the Korea Health Industry Development Institute (KHIDI),

Neck Surgery, School of Medicine. yClinical Fellow, Department of Otolaryngology–Head and Neck

funded by the Ministry of Health & Welfare, Republic of Korea [grant number HI18C1039 and HI17C2060].

Surgery, School of Medicine.

Conflict of Interest Disclosures: None of the authors have any

zClinical Fellow, Department of Otolaryngology–Head and Neck

relevant financial relationship(s) with a commercial interest.

Surgery, School of Medicine.

Address correspondence and reprint requests to Dr Eun: Depart-

xProfessor, Department of Radiology, School of Medicine.

ment of Otolaryngology–Head and Neck Surgery, Kyung Hee Univer-

kAssociate Professor, Department of Nuclear Medicine, School of

sity School of Medicine, Seoul, Korea, #1 Hoegi-dong, Dongdaemun-

Medicine.

gu, Seoul 130-702, Republic of Korea; e-mail: [email protected]

{Associate Professor, Department of Oral and Maxillofacial Surgery, School of Dentistry.

Received August 10 2018 Accepted December 11 2018

#Associate Professor, Department of Otolaryngology–Head and Neck Surgery, School of Medicine.

Ó 2018 American Association of Oral and Maxillofacial Surgeons 0278-2391/18/31380-6

This work was supported by the National Research Foundation of

https://doi.org/10.1016/j.joms.2018.12.010

Korea (NRF) Grant funded by the Korea Government (Ministry of Science and ICT) [grant numbers NRF-2018R1D1A1B07050154] and

875.e1

875.e2

DIAGNOSIS OF BONE INVASION IN GINGIVAL CANCER

Conclusions: The 3 imaging methods were less sensitive for the detection of bone invasion in upper gingival cancer than in lower gingival cancer. Cases of upper gingival cancer should be evaluated more carefully for bone invasion before surgery. Ó 2018 American Association of Oral and Maxillofacial Surgeons J Oral Maxillofac Surg 77:875.e1-875.e9, 2019

Gingival cancer is a rare form of head and neck cancer.1 Because primary squamous cell carcinoma (SCC) in the maxillary gingival region is uncommon, there is sparse information about these cancers, especially compared with SCCs of the mandible. According to TNM stage criteria for oral cavity cancer, any tumor invading the underlying bone is designated as stage T4. Because of the proximity of the gingiva to the bone, gingival cancer tends to invade the underlying bone early. Therefore, gingival cancer is graded as T4 at diagnosis, and surgical treatment of gingival lesions often involves removal of the mandible or maxilla. Furthermore, Ogura2 found that maxillary bone invasion is an indicator of cervical metastasis in gingival carcinomas. Bone invasion is one of the most important factors when determining treatment options and prognosis in cases of gingival cancer, because the presence and extent of bone invasion are associated with patient outcomes. Therefore, an accurate preoperative evaluation is important to assess the quality of life of the patient and to establish appropriate bone management (which can vary from periosteal stripping to mandibulectomy or maxillectomy). Previous studies have compared imaging tools used to assess the mandibular invasiveness of oral cancer.3,4 Currently, computed tomography (CT) is the preferred method for the evaluation of bone invasion by lower gingival cancer.2 Previous studies have reported the high specificity and positive predictive value (PPV) of CT for the detection of mandibular invasion.5,6 However, the diagnostic accuracies of the imaging methods usually used in clinical practice (CT, magnetic resonance imaging [MRI], and positron emission tomography [PET] and CT) have not been evaluated for maxillary bone invasion by upper gingival cancer. The purpose of this study was to analyze the diagnostic abilities of imaging modalities to evaluate bone involvement in cases of upper and lower gingival cancer. The authors hypothesized there would be variation between upper and lower gingival cancer because of the anatomic difference. The specific aims of the study were to compare the diagnostic value of 3 imaging methods—CT, MRI, and 18 F-fluorodeoxyglucose (FDG) PET/CT—for the detection of bone invasion by SCC of the gingiva and to compare the diagnostic values between upper and lower gingival cancer.

Materials and Methods STUDY DESIGN AND SAMPLE

To address the research purpose, the authors designed and implemented a retrospective cohort study. The study protocol was approved by the institutional review board at the authors’ institute. The study population was composed of 40 patients who presented for evaluation and management of gingival cancer from January 2013 through December 2017. Inclusion criteria consisted of a diagnosis of SCC originating in the gingiva (upper or lower); gingival cancer that had been confirmed by pathologic examination or preoperative CT, MRI, or PET/CT staging studies; and surgical management. All enrolled patients underwent (marginal or segmental) mandibulectomy or infrastructure maxillectomy. Segmental mandibulectomy was performed for the following cases: 1) imaging evidence of invasion of the medullary space of the mandible, 2) tumor fixation to the occlusal surface of the mandible in an edentulous patient, and 3) presence of a hypoplastic edentulous mandible with substantial loss of vertical height, precluding the safe performance of rim resection. Tumor margins were sent for frozen sectioning. Patients were excluded from the study if they had received prior preoperative treatments, including surgical resection, radiation therapy, or chemotherapy, or if none of the 3 imaging types had been used before the patient’s surgery. VARIABLES

Predictor Variables CT was performed with a GE 9800 scanner (GE Medical Systems, Milwaukee, WI) and a reconstructed slice thickness of 3 mm for axial and coronal images. A 100-mL dose of nonionic iodinated contrast agent (Ultravist1, Schering AG, Berlin, Germany) was administered intravenously at a rate of 2 mL per second. Gingival bone involvement on a CT scan was defined as cortical bone invasion or infiltration by a tumor or infiltration of bone marrow by a lesion. MRI was performed using a 1.5-T MR scanner (Achieva 1.5T, Philips, Best, Netherlands) with a 16-channel sensitivityencoding head-and-neck coil. The sequences included 1) precontrast T1-weighted turbo spin-echo images (acquisition matrix, 288  229; repetition time [TR], 568 ms; echo time [TE], 20 ms) in the axial plane, 2) T2-weighted turbo spin-echo images (acquisition

875.e3

LEE ET AL

matrix, 320  240; TR, 5,074 ms; TE, 100 ms) in the sagittal and coronal planes, 3) T2-weighted turbo spin-echo images (acquisition matrix, 288  223; TR, 6,794 ms; TE, 80 ms) in the axial plane, 4) short tau inversion recovery sequence T2-weighted images (acquisition matrix, 304  240; TR, 7,903 ms; TE, 70 ms) in the coronal plane, 5) postcontrast T1-weighted spectral pre-saturation with inversion recovery (SPIR) images (acquisition matrix, 304  304; TR, 546 ms; TE, 20 ms) in the coronal and sagittal planes, and 6) postcontrast T1-weighted SPIR images (acquisition matrix, 272  215; TR, 530 ms; TE, 16 ms) in the axial plane. Scan volume was set from the skull base to the origin of the supra-aortic vessels with a slice thickness that varied from 3 to 4 mm. On MR images, bone invasion was defined as replacement of cortical bone signals on T1- and T2-weighted images. FDG PET/CT was performed with a Gemini TF16 PET scanner (Philips Healthcare, Cleveland, OH) with an intrinsic resolution of 4.8 mm, full width at half maximum, and simultaneous imaging of 50 contiguous transverse planes with a thickness of 4 mm for a longitudinal field of view of 18 cm. All patients fasted for at least 6 hours before scanning. Imaging was initiated with a planar scout scan to define the axial range of the study, followed by volumetric CT acquisition. The CT parameters were 120 kVp, 50 mAs, and 2-mm slice width and separation. A 1-minute emission scan was performed per bed position after intravenous injection of FDG 370 MBq. All emission scans were performed in 3-dimensional acquisition mode. Attenuation-corrected images were reconstructed into a 144  144 matrix using an ordered subset expectation maximization algorithm incorporating time-offlight information (TOF-OSEM). The reconstruction parameters for TOF-OSEM were 3 iterations and 33 subsets. The possibility of bone invasion on CT, MR, and PET/CT images was graded retrospectively on a 4-point scale (1, definitely absent; 2, probably absent; 3, probably present; 4, definitely present). Each preoperative image was reviewed for the presence of bone invasion by 1 blinded head and neck radiologist or 1 nuclear physician. Outcome Variables All specimens were 5 mm thick and stained with hematoxylin and eosin. The presence of bone involvement was assessed by 1 head and neck pathologist blinded to the patients’ information. Clinicopathologic Variables Demographic data (age and gender), TNM stages, tumor location (upper or lower gingiva), dentate status, presence of metallic artifacts on images, and type of bone surgery performed were obtained from medical records. Patients were classified as edentulous

if they lost at least 1 tooth at the site of the tumor. Bone surgical management was categorized as maxillectomy, marginal mandibulectomy, or segmental mandibulectomy. DATA COLLECTION METHODS

The accuracy of bone involvement in gingival carcinoma diagnoses was evaluated by combining diagnostic tools: CT + MRI, CT + PET/CT, MRI + PET/CT, and CT + MRI + PET/CT. The overall score of each combination of images for the probability of invasion was defined as the sum of scores evaluated for the single image divided by the number of images. Bone invasion was considered present when the score of at least 1 imaging tool was 4 or the total score was higher than 2. The remaining images were considered to show the absence of bone invasion. DATA ANALYSES

Sensitivity, specificity, PPV, and negative predictive value for the detection of mandibular involvement in alveolar bone were calculated based on the readings of each image or combination of images and pathologic results. Differences in sensitivity and specificity among imaging modalities were tested for statistical relevance using the McNemar test. Receiver operating characteristics (ROC) curve analysis was used to assess the diagnostic value of CT, MRI, and PET/CT for the detection of bone invasion. SPSS 12.0 (SPSS, Inc, Chicago, IL) was used for all statistical analyses. A P value less than .05 was considered statistically significant.

Results The median age of the 40 patients (13 women and 27 men) was 60 years (range, 31 to 93 yr). Nineteen patients had been diagnosed with SCC originating from the upper gingiva, and 21 had been diagnosed with SCC originating from the lower gingiva. Pathologic examination reported the presence of bone invasion in 25 patients (62.5%). Histologic evidence of bone invasion was more prevalent in cases of upper gingival cancer (13 of 19; 68.4%) than in cases of lower gingival cancer (12 of 21; 57.1%). The demographic data and TNM staging are presented in Table 1. The number of patients with gingival bone invasion at CT, MRI, and PET/CT according to clinical variables is presented in Table 2. Cases of bone invasion at pathologic examination were analyzed according to clinical variables. Advanced tumor stage (P = .002) and edentulous state (P = .002) were significantly associated with a higher frequency of bone invasion in gingival cancer (Table 3). Sensitivities were 88.0, 88.0, and 80.0% for CT, MRI, and PET/CT, respectively. Specificities were 80.0, 73.3, and 66.6% for CT, MRI, and PET/CT, respectively. Differences between each

875.e4

DIAGNOSIS OF BONE INVASION IN GINGIVAL CANCER

Table 1. DEMOGRAPHIC DATA OF ENROLLED PATIENTS

Variables Age (yr), median (range) Gender Women Men Primary tumor subsite Upper gingiva Lower gingiva Clinical tumor stage T1 T2 T3 T4 Pathologic bone invasion No Yes Bone invasion subsite Maxilla Mandible Dentate status Dentate Edentulous Metallic artifact No Yes Bone surgical management Maxillectomy Marginal mandibulectomy Segmental mandibulectomy

Value

%

60 (31-93) 13 27

32.5 67.5

19 21

47.5 52.5

6 7 3 24

15.0 17.5 7.5 60.0

15 25

37.5 62.5

13 12

32.5 30.0

19 21

47.5 52.5

13 27

32.5 67.5

19 12 9

47.5 30 22.5

Lee et al. Diagnosis of Bone Invasion in Gingival Cancer. J Oral Maxillofac Surg 2019.

imaging modality did not reach statistical significance (P = 1.00 by paired McNemar test). However, the diagnostic accuracy of CT (90.0%) for the detection of bone invasion in gingival cancer was highest among the 3 modalities, and the specificity values of MRI and PET/CT were lower than that of CT (Table 4). Table 5 presents a comparison of the diagnostic values of CT, MRI, and PET/CT for the detection of bone invasion in upper gingival cancer with those of lower gingival cancer. There were no statistically relevant differences between upper and lower gingival cancer for the sensitivity or specificity of the 3 imaging modalities. However, in the ROC curve scoring for the possibility of bone invasion according to subsite, area under the curve (AUC) values of the 3 modalities were lower for upper gingival cancer (CT: AUC, 0.744; P = .096; MRI: AUC, 0.737; P = .105; PET/CT: AUC, 0.808; P = .035) than for lower gingival cancer (CT: AUC, 0.926; P = .001; MRI: AUC, 0.935; P = .001; PET/CT: AUC, 0.824; P = .003; Fig 1). Table 6 presents the sensitivity, specificity, and accuracy of the combinations of imaging modalities for bone invasion in

patients with SCC of the upper and lower gingiva. For lower gingival cancer, the accuracy of the diagnosis for bone invasion was improved when the 3 diagnostic modalities were combined; for upper gingiva cancer, all imaging modalities and combinations exhibited unsatisfactory accuracy. In the present study, most false-positive cases showed discrepancies among CT, MRI, and PET/CT scores for the possibility of bone invasion. Figure 2 shows a false-positive case. The left upper gingiva exhibits a soft tissue mass of approximately 4.6  2.4 cm with heterogeneous enhancement. This lesion appears to involve the inner and outer cortices and medullary bone destruction of the adjacent maxillary bone (Fig 2A). On the T1-weighted enhanced image, tumor invasion is shown by replacement of marrow fat by the tumor (Fig 2B). On the PET/CT image, the left upper gingiva exhibits a hypermetabolic lesion (maximum standardized uptake value, 5.0; Fig 2C). However, the lesion showing increased uptake is clearly separated from the bone. There was no bone involvement on the pathology report.

Discussion The purpose of this study was to analyze the ability of diagnostic tools to evaluate bone involvement in gingival cancer. Gingival cancer is rare; it accounts for only 6.3% of all oral cancers.7 The gingival and alveolar bone is divided into the maxilla and the mandible, and the 2 carcinomas that originate from these subsites exhibit different clinical features because of differences in histologic features and surrounding structures. In the present study, the diagnostic value of 3 imaging methods—CT, MRI, and PET/CT—was compared for the detection of bone invasion by SCC of the gingiva. In addition, the diagnostic values for upper gingival cancer were compared with those for lower gingival cancer. The results indicate that the diagnostic accuracy of CT was highest among the 3 modalities for the detection of bone invasion in gingival cancer. On the ROC curve analysis, all 3 imaging tools showed poorer diagnostic ability for upper gingival cancer than for lower gingival cancer. In a study using CT imaging analysis, Park et al8 found that the density of the cortical bone was greater in the mandible than in the maxilla and exhibited a progressive increase from the incisor to the retromolar pad area. According to a previous study, the path of mandibular invasion is largely explained by 2 patterns: infiltrative and erosive. The erosive disease pattern is associated with shallow mandibular invasion and smaller tumors in the soft tissue, whereas the infiltrative pattern has a poorer prognosis and is related to the size of the primary tumor and the depth of bone

875.e5

LEE ET AL

Table 2. DIAGNOSIS OF BONE INVASION FOR EACH IMAGING MODALITY ACCORDING TO CLINICAL VARIABLES

Patients With Bone Invasion at Imaging, n Variables Gender Women Men Primary tumor subsite Upper gingiva Lower gingiva Clinical tumor stage T1 T2 T3 T4 Dentate status Dentate Edentulous Metallic artifact No Yes

CT

P Value

MRI

.298 10 of 13 15 of 27

P Value .316

10 of 13 16 of 27 .462

13 of 19 12 of 21

<.001

.001

.171 2 of 6 4 of 7 1 of 3 18 of 24

.004 8 of 19 18 of 21

.298 10 of 13 15 of 27

.567 11 of 19 14 of 21

3 of 6 1 of 7 0 of 3 22 of 24

7 of 19 18 of 21

.298

.273

<.001

.060 9 of 19 16 of 21

.316 10 of 13 16 of 27

P Value

10 of 13 15 of 27

14 of 19 12 of 21

2 of 6 1 of 7 0 of 3 22 of 24

PET

.298 10 of 13 15 of 27

Abbreviations: CT, computed tomography; MRI, magnetic resonance imaging; PET, positron emission tomography. Lee et al. Diagnosis of Bone Invasion in Gingival Cancer. J Oral Maxillofac Surg 2019.

invasion.8 Different surgical procedures are available to patients with gingival cancer, depending on the presence or absence of bone involvement. Mucosal and periosteal resection is preferred when gingival Table 3. GINGIVAL BONE INVASION AT PATHOLOGIC EXAMINATION ACCORDING TO CLINICAL VARIABLES

Patients With Bone Invasion at Pathologic Examination, n Gender Women Men Primary tumor subsite Upper gingiva Lower gingiva Clinical tumor stage T1 T2 T3 T4 Dentate status Dentate Edentulous Metallic artifact No Yes

P Value .298

9 of 13 14 of 27 .962 8 of 11 12 of 21 .002 2 of 6 2 of 7 0 of 3 19 of 24 .002 6 of 19 17 of 21 .298 9 of 13 14 of 27

Lee et al. Diagnosis of Bone Invasion in Gingival Cancer. J Oral Maxillofac Surg 2019.

mucosa invasion is present and there is no evidence of bone invasion. For a gingival tumor with bone resorption or destruction indicated at the preoperative evaluation, a segmental resection of varying lengths of bone is chosen, depending on the tumor size. Maxillectomy is required for upper gingival cancer, and segmental mandibulectomy is required for lower gingival cancer. Maxillectomy can result in cosmetic deformities and serious functional problems, such as dental occlusion, masticatory dysfunction, hypernasal speech, and nasal leaks. Brown et al9 classified the resections from partial maxillectomy without a fistula as total alveolar resection. Therefore, the precise evaluation of preoperative bone involvement is important not only to determine the best surgical strategy for treatment but also for the patient’s quality of life after the operation. MRI has been reported to have higher sensitivity but lower specificity than CT for the detection of mandibular invasion, so there is controversy regarding the best diagnostic tools for the assessment of bone involvement.10 According to the results of a recent meta-analysis, most diagnostic imaging tools have high diagnostic accuracy for mandibular involvement, with sensitivity values of 94 and 83% for MRI and CT, respectively, and specificity values of 100% for CT and MRI and 97% for PET/CT.11 Various articles have reported that CT has high sensitivity and specificity, but slice thickness can affect sensitivity. MRI also has high sensitivity and specificity, but these can be

875.e6

DIAGNOSIS OF BONE INVASION IN GINGIVAL CANCER

Table 4. DIAGNOSTIC VALUES OF CT, MRI, AND PET/CT FOR DETECTION OF BONE INVASION IN PATIENTS WITH GINGIVAL CANCER

Bone Invasion at Pathologic Examination Bone Invasion at Imaging

Positive

Negative

22 3

3 12

22 3

4 11

20 5

5 10

CT Positive Negative MRI Positive Negative PET/CT Positive Negative

Sensitivity, %

Specificity, %

Diagnostic Accuracy, %

88.0

80.0

90.0

88.0

73.3

82.5

80.0

66.6

72.5

Abbreviations: CT, computed tomography; MRI, magnetic resonance imaging; PET/CT, positron emission tomography and computed tomography. Lee et al. Diagnosis of Bone Invasion in Gingival Cancer. J Oral Maxillofac Surg 2019.

overestimated if surrounding tissue edema is present.10 In addition, it has been reported that artifacts caused by movement, inflammation around the gingiva, or osteonecrosis can decrease the accuracy of the diagnosis.4 However, in general, MRI has an advantage over CT for the presence of metal arti-

facts.12 In a comparative study of MRI and CT for the evaluation of maxillary and mandibular cancers, MRI was competitive with CT in assessing the loss of cortical margins and bone expansion, and MRI was superior in visualizing the replacement of marrow fat and extraosseous extension of

Table 5. COMPARISON OF DIAGNOSTIC VALUES OF CT, MRI, AND PET/CT FOR DETECTION OF BONE INVASION BETWEEN UPPER AND LOWER GINGIVAL CANCER

Bone Invasion at Pathologic Examination Bone Invasion at Imaging Upper gingiva CT Positive Negative MRI Positive Negative PET/CT Positive Negative Lower gingiva CT Positive Negative MRI Positive Negative PET/CT Positive Negative

Sensitivity, %

Specificity, %

Diagnostic Accuracy, %

84.6

66.6

78.9

3 3

84.6

50.0

73.6

10 3

1 5

76.9

83.3

78.9

11 1

1 8

91.6

88.8

90.4

11 1

1 8

91.6

88.8

90.4

10 2

4 5

83.3

55.5

71.4

Positive

Negative

11 2

2 4

11 2

Abbreviations: CT, computed tomography; MRI, magnetic resonance imaging; PET/CT, positron emission tomography and computed tomography. Lee et al. Diagnosis of Bone Invasion in Gingival Cancer. J Oral Maxillofac Surg 2019.

875.e7

LEE ET AL

FIGURE 1. Receiver operating characteristics curves indicating the possibility of bone invasion in A, upper and B, lower gingival cancer. AUC, area under the curve; CT, computed tomography; MRI, magnetic resonance imaging; PET/CT, positron emission tomography and computed tomography. Lee et al. Diagnosis of Bone Invasion in Gingival Cancer. J Oral Maxillofac Surg 2019.

neoplasm.13 In a recent meta-analysis, Li et al14 found that MRI was superior to CT in terms of sensitivity for the diagnosis of mandibular involvement but inferior to CT in terms of specificity. PET has been reported to have a slightly higher diagnostic accuracy for bone invasion than CT, but the specificity might be lower due to false-positive results caused by inflammatory reactions.4 Furthermore, dentate

status can affect the diagnostic performance of PET/CT.15 For dentate patients with bone invasion in this study, the specificities of MRI and PET/CT were 78 and 100%, respectively. Although not included in the results, the association between dentate status and the diagnostic ability of imaging modalities was analyzed but showed no statistical relevance.

Table 6. COMPARISON OF DIAGNOSTIC VALUES OF CT, MR, AND PET/CT COMBINATIONS FOR DETECTION OF BONE INVASION BETWEEN UPPER AND LOWER GINGIVAL CANCER

Upper gingiva CT + MRI CT + PET/CT MRI + PET/CT CT + MRI + PET/CT Lower gingiva CT + MRI CT + PET/CT MRI + PET/CT CT + MRI + PET/CT

Sensitivity, %

Specificity, %

Diagnostic Accuracy, %

84.6 84.6 84.6 84.6

50.0 66.6 66.6 66.6

73.6 78.9 78.9 78.9

91.6 91.6 91.6 91.6

88.8 66.6 66.6 66.6

90.4 80.9 80.9 80.9

Abbreviations: CT, computed tomography; MRI, magnetic resonance imaging; PET/CT, positron emission tomography and computed tomography. Lee et al. Diagnosis of Bone Invasion in Gingival Cancer. J Oral Maxillofac Surg 2019.

875.e8

DIAGNOSIS OF BONE INVASION IN GINGIVAL CANCER

FIGURE 2. A, Computed tomogram, B, magnetic resonance image, and C, positron emission tomographic computed tomogram for a falsepositive case. Lee et al. Diagnosis of Bone Invasion in Gingival Cancer. J Oral Maxillofac Surg 2019.

In the present results, sensitivity for the detection of bone invasion in upper and lower gingival cancer was 88% for all imaging modalities. Specificity was the highest for CT (80.0%) and lowest for PET/CT

(66.6%). These reported. When gingival cancer, gingival cancer

values are lower than subdivided into upper the diagnostic ability was superior to that

previously and lower for lower for upper

875.e9

LEE ET AL

gingival cancer except for the specificity of PET/CT. However, the difference was not statistically meaningful. A study on the accuracy of radiologic diagnosis of maxillary bone involvement was conducted for maxillary sinus cancer, but not for alveolar bone cancer. El-Hafez et al15 evaluated the ability of MRI and CT to detect bone destruction in the upper jaw from maxillary cancer. In their comparative study of MRI and CT, the ability of MRI to detect bone destruction of each aspect of the maxilla was similar to that of CT. The present results also showed that a combination of imaging modalities improved the sensitivity and specificity for the detection of bone invasion in lower gingival cancer. In contrast, the combination of imaging tools did not provide satisfactory improvement of diagnostic ability in upper gingival cancer. Nevertheless, imaging combinations, including PET/CT, increased the specificity for the detection of bone invasion in upper gingival cancer. A previous report comparing PET/CT and MRI for the detection of bone marrow invasion found that PET/CT was more specific than MRI. In addition, negative PET/ CT findings can be useful for ruling out bone marrow invasion in dentate patients.15 The report of the high specificity of PET/CT is consistent with the present findings in upper gingival cancer. Although there have been many studies on the accuracy of different imaging tests for bone invasion in gingival cancer, there are few studies comparing the accuracy of the tests in the maxilla with that of the mandible. To the authors’ knowledge, this is the first study on the diagnostic value for detection of bone invasion in upper gingival SCC. This study has some limitations. First, the study was confined to carcinomas that developed in the maxilla or mandible, which led to a small number of patients enrolled in the study. Second, for CT, the reconstructed slice thickness was 3 mm for axial and coronal images. DentaScan (GE Healthcare, Chicago, IL) can perform multiple cross-sectional and panoramic reformations from a thin-section (1- to 1.25-mm) axial image. Because CT accuracy is greatly affected by image slice thickness, future studies are needed to compare the accuracy of dental CT using different thicknesses. Third, because this study was retrospective, only the presence or absence of bone involvement was analyzed. Future studies are needed to evaluate the detailed classification of bone invasion. In conclusion, the diagnostic accuracy of CT was highest among the 3 modalities for the detection

of bone invasion in gingival cancer, and the specificity values of MRI and PET/CT were lower than that of CT. Unlike lower gingival cancer, the diagnostic value of CT, MRI, or PET/CT alone was limited for the evaluation of bone invasion in upper gingival cancer. The diagnostic accuracy for bone invasion in lower gingival cancer could be improved through the combination of CT, MRI, and PET/CT. Further studies with larger samples are needed to validate these results.

References 1. Soo KC, Spiro RH, King W, et al: Squamous carcinoma of the gums. Am J Surg 156:281, 1988 2. Ogura I, Kurabayashi T, Sasaki T, et al: Maxillary bone invasion by gingival carcinoma as an indicator of cervical metastasis. Dentomaxillofac Radiol 32:291, 2003 3. Silva M, Zambrini EI, Chiari G, et al: Pre-surgical assessment of mandibular bone invasion from oral cancer: Comparison between different imaging techniques and relevance of radiologist expertise. Radiol Med 121:704, 2016 4. Gu DH, Yoon DY, Park CH, et al: 18F-FDG PET/CT, and their combined use for the assessment of mandibular invasion by squamous cell carcinomas of the oral cavity. Acta Radiol 51: 1111, 2010 5. Goerres GW, Schmid DT, Schuknecht B, et al: Bone invasion in patients with oral cavity cancer: Comparison of conventional CT with PET/CT and SPECT/CT. Radiology 237:281, 2005 6. Hendrikx A, Maal T, Dieleman F, et al: Cone-beam CT in the assessment of mandibular invasion by oral squamous cell carcinoma: Results of the preliminary study. Int J Oral Maxillofac Surg 39:436, 2010 7. Seoane J, Varela-Centelles PI, Walsh TF, et al: Gingival squamous cell carcinoma: Diagnostic delay or rapid invasion? J Periodontol 77:1229, 2006 8. Park H-S, Lee Y-J, Jeong S-H, et al: Density of the alveolar and basal bones of the maxilla and the mandible. Am J Orthod Dentofacial Ortho 133:30, 2008 9. Brown JS, Rogers SN, McNally DN, et al: A modified classification for the maxillectomy defect. Head Neck 22:17, 2000 10. Van den Brekel M, Runne R, Smeele LE, et al: Assessment of tumour invasion into the mandible: The value of different imaging techniques. Eur Radiol 8:1552, 1998 11. Uribe S, Rojas L, Rosas C: Accuracy of imaging methods for detection of bone tissue invasion in patients with oral squamous cell carcinoma. Dentomaxillofac Radiol 42: 20120346, 2003 12. Brown J, Deluca S: Imaging of sinonasal tumors. Am Fam Physician 45:1653, 1992 13. Belkin BA, Papageorge MB, Fakitsas J, et al: A comparative study of magnetic resonance imaging versus computed tomography for the evaluation of maxillary and mandibular tumors. J Oral Maxillofac Surg 46:1039, 1988 14. Li C, Yang W, Men Y, et al: Magnetic resonance imaging for diagnosis of mandibular involvement from head and neck cancers: A systematic review and meta-analysis. PLoS One 9: e112267, 2014 15. El-Hafez YGA, Chen C-C, Ng S-H, et al: Comparison of PET/CT and MRI for the detection of bone marrow invasion in patients with squamous cell carcinoma of the oral cavity. Oral Oncol 47:288, 2011