Maximum and mean standardized uptake values of medication-related osteonecrosis of the jaw with bone SPECT/CT: comparison of mandibular pathologies, control and temporomandibular joints

OBJECTIVE
Recently, single-photon emission CT/CT (SPECT/CT) plays an important role in assessing patients with medication-related osteonecrosis of the jaw (MRONJ). The aim of this study was to investigate maximum and mean standardized uptake values (SUVs) of MRONJ with bone SPECT/CT, especially comparison of mandibular pathologies, control and temporomandibular joints.


METHODS
61 mandibular patients with MRONJ who underwent bone SPECT/CT were included in this study. The maximum and mean SUVs of the lesion, right and left sides of the lesion, opposite side of the lesion as control, right and left temporomandibular joints were analyzed using a workstation and software. The SUVs of MRONJ were analyzed using one-way analysis of variance with Tukey's honestly significant difference test. Patient characteristics with MRONJ and SUVs were analyzed using Mann-Whitney U test. P-values less than 0.05 were considered to indicate statistical significance.


RESULTS
The maximum and mean SUVs for opposite side of the lesions (4.4 ± 2.0 and 1.8 ± 0.7) were significantly lower than those for mandibular lesions (18.3 ± 8.1 and 6.3 ± 2.8), right side of the lesions (8.1 ± 3.9 and 2.9 ± 1.3) and left side of the lesions (8.1 ± 3.9 and 2.8 ± 1.4), respectively. The maximum and mean SUVs for right and left sides of the lesions, and opposite side of the lesions, right and left temporomandibular joints were not significant difference. Furthermore, the maximum SUVs of the mandibular lesions were a significant difference for age and staging.


CONCLUSIONS
The maximum and mean SUVs with SPECT/CT can be useful in the quantitative management of MRONJ patients.


Introduction
Medication-related osteonecrosis of the jaw (MRONJ) is a complication of treatment with bisphosphonates and denosumab for patients with osteoporosis and bone metastases from cancer [1].
The term was de ned by American Association of Oral and Maxillofacial Surgeons [2]. Single-photon emission computed tomography/computed tomography (SPECT/CT) plays an important role in Materials And Methods

Patients
The study was approved by the institutional review board of our university (approved no. ECNG-R-318), and informed consent was obtained from all individual participants included in the study. Thirty-nine mandibular patients (11 males and 28 females; mean age, 77.7 years [range, 55-91 years]) with MRONJ underwent bone SPECT/CT prior surgical treatments at our university hospital from January 2020 to December 2021. Patients were considered to have MRONJ by American Association of Oral and Maxillofacial Surgeons [2]. All patients with MRONJ were diagnosed and treated for osteoporosis or bone metastases from cancer at other hospital. Characteristics of the mandibular patients with MRONJ was shown in Table 1.
SPECT/CT imaging SPECT/CT was performed by Optima NM/CT 640 (GE Healthcare, Tokyo, Japan) at 4 hours after injection of 740 MBq of technetium-99m hydroxymethylene diphosphonate (Tc-99m HMDP; Clear Bone Injectable, Nihon Medi-Physics, Tokyo, Japan), following our hospital protocol [6]. The SPECT was acquired using low-energy high-resolution collimation, the 140 keV photo-energy peak for Tc-99m, a 128×128 matrix of 4.2 mm pixel size, and total of 60 projections (30 stops) over 360° with a dwell time of 10 s/stop. Subsequent to the SPECT acquisition, low-dose CT was acquired with 120 kV and 20 mA using a 512×512 matrix. The CT data were generated with a 2.5 mm section thickness.

Data Analysis
The maximum and mean SUVs were obtained using a workstation and software (Xeleris 4DR and Q. Volumetrix MI, GE Healthcare, Tokyo, Japan), following our hospital protocol [15]. SPECT and CT images were displayed on the Xeleris 4DR workstation, volume of interest (VOI) using those imaging was drawn over the mandibular lesions, right, left and opposite sides of the lesions, and right and left temporomandibular joints, and the SUVs in given VOIs was analyzed (Fig. 1). The maximum and mean SUVs in given VOIs were calculated automatically as follows: maximum SUV = (maximum radioactivity / voxel volume) / (injected radioactivity / body weight) and mean SUV = (total radioactivity / VOI volume) / (injected radioactivity / body weight) (Fig. 2).

Statistical analysis
Statistical analyses for the maximum and mean SUVs, such as mandibular lesions, right, left and opposite sides of the lesion, and right and left temporomandibular joints, were compared by Friedman test with Bonferroni method. Statistical analyses for patient characteristics and the SUVs were compared by Mann-Whitney U test. A p value lower than 0.05 indicated signi cant differences by statistical software (IBM SPSS Statistics, version 26, IBM Japan, Tokyo, Japan).
The patient characteristics and SPECT/CT SUVs of the mandibular lesions with MRONJ was summarized in Table 3. The maximum SUV of the mandibular lesions was a signi cant difference for age (p = 0.012), gender (p = 0.002), underlying disease (p = 0.021), and staging of MRONJ (p = 0.025).
Furthermore, the mean SUV of the mandibular lesions was a signi cant difference for age (p = 0.021), gender (p = 0.001), and underlying disease (p = 0.032).
The patient characteristics and SPECT/CT SUVs of the opposite side of the lesions with MRONJ was summarized in Table 4. The maximum and mean SUVs of the opposite side of the lesions was not signi cant difference for age, gender, underlying disease, medication, and staging of MRONJ. Furthermore, the patient characteristics and SPECT/CT SUVs of the temporomandibular joints with MRONJ was summarized in Table 5. The maximum and mean SUVs of the temporomandibular joints was not signi cant difference for age, gender, underlying disease, medication, and staging of MRONJ.

Discussion
In recent years, a new workstation and software "Xeleris 4DR and Q. Volumetrix MI" was developed for the SUVs derived from bone SPECT/CT [15]. Therefore, we investigated maximum and mean SUVs of MRONJ using Q. Volumetrix MI, especially comparison of mandibular pathologies, control and temporomandibular joints.
In this study, the maximum SUV for opposite side of the lesions (4.4 ± 2.2) was signi cantly lower than those for mandibular lesions (17.5 ± 7.7, p 0.001), right side of the lesions (8.1 ± 4.4, p = 0.001) and left side of the lesions (8.2 ± 4.2, p = 0.001). Furthermore, the mean SUV for opposite side of the lesions (1.7 ± 0.8) was signi cantly lower than those for mandibular lesions (6.3 ± 3.0, p 0.001), right side of the lesions (2.9 ± 1.5, p = 0.001) and left side of the lesions (2.9 ± 1.5, p = 0.001). Miyashita et al [3,4] indicated that 3D SPECT/CT is useful not only for detecting MRONJ but also for surgical planning. Modabber et al [16] concluded that SPECT/CT can safely detect different kinds of in ammatory jaw pathologies compared to other conventional imaging modalities; additionally, SPECT/CT assists the surgeon in determining the expansion of the process preoperatively and thereby optimizing surgery planning. We consider that the SPECT/CT SUVs using Q. Volumetrix MI showed potential as a useful parameter for evaluation of mandibular lesions, such as detecting and surgical planning. Suh et al [17] evaluated the diagnostic accuracy of the quantitative parameter SUV at SPECT/CT for the evaluation of temporomandibular joint (TMJ) disorder (TMD), indicated that maximum SUV gradually increased from normal (2.82 ± 0.73) to mild or moderately abnormal (3.56 ± 0.76, p 0.05) and then to severely abnormal (4.86 ± 1.25, p 0.05), however, mean SUV did not vary signi cantly, such as 1.31 ± 0.38 for normal, 1.48 ± 0.36 for mild or moderately abnormal and 1.64 ± 0.46 for severely abnormal. In this study, the maximum SUVs of right and left TMJs were 2.5 ± 0.9 and 2.5 ± 0.8, respectively. Furthermore, the mean SUVs of right and left TMJs were 1.3 ± 0.5 and 1.3 ± 0.4, respectively. Our results seem the same as in the previous reports.
In this study, the maximum SUV of the mandibular lesions with MRONJ was a signi cant difference for age (p = 0.012), gender (p = 0.002), underlying disease (p = 0.021), and staging of MRONJ (p = 0.025). Furthermore, the mean SUV of the mandibular lesions was a signi cant difference for age (p = 0.021), gender (p = 0.001), and underlying disease (p = 0.032). However, the maximum and mean SUVs of the opposite side of the lesions and TMJs were not signi cant difference for age, gender, underlying disease, medication, and staging of MRONJ. We showed the relationship between patient characteristics with MRONJ and SUVs.
There were several limitations of this study. The sample was relatively small. Power calculation because of small sample sizes would strengthen the study, otherwise this is a signi cant limitation. Therefore, further research is necessary to validate these results.

Conclusions
We investigated maximum and mean SUVs of MRONJ with bone SPECT/CT, especially comparison of mandibular pathologies, control and temporomandibular joints. The maximum and mean SUVs using Xeleris 4DR and Q. Volumetrix MI may be useful in clinical practice for the quantitative management of patients with MRONJ.

Declarations
Compliance with Ethical Standards Con ict of Interests The authors have no con ict of interest.     Medication-related osteonecrosis of the jaw of the mandible in a 73-year-old woman with Xeleris 4DR and Q. Volumetrix MI. Using the CT, SPECT and SPECT/CT transaxials, coronals and sagittals as the anatomical reference, the localization and size of volume of interest (VOI) was automatically drawn over the mandibular lesion (sky blue), left (red), right (green), opposite (pink) side of the lesion, right (yellow) and left (blue) temporomandibular joints.