Abstract
Objectives:
The aim of this study was to assess a three-dimensional (3D) correlation between preoperative 3D bone single photon emission CT (SPECT)/CT, which allows the visualization of radiotracer uptake on 3D volume-rendered CT images, and histopathological characteristics in the medication-related osteonecrosis of the jaw (MRONJ).
Methods:
We conducted a full histopathological assessment of the resected jaws in four patients with Stage 2 or 3 MRONJ. The pathologic results were classified as follows: necrosis without any tissue vascularity (N + V-), necrosis with both vascularity and acute inflammatory cell infiltration due to bacterial infection (N + V+I+), necrosis with regenerative vasculature but no inflammatory cell infiltration (N + V+I-), and chronic inflammation without massive necrosis (N-V +I+). These classifications were correlated with imaging results.
Results:
The N + V- areas visually represented the area of necrotic bone exposed to the oral cavity and were consistent with defect area of radioisotope uptake in SPECT/CT. The N + V- areas were surrounded by the N + V+I + areas where increased radiotracer uptake was clearly seen. Also, abnormal uptake was found in both of the N + V+I- and N-V +I+ areas. The extensive surgical resections from necrotic core to bloody viable margins were performed in all cases, although one had the recurrence of MRONJ at the margin showing abnormal uptake that histologically represented the N + V+I- area.
Conclusions:
Radiologic–pathologic correlation of MRONJ could be achieved using 3D SPECT/CT. The presence of regenerative vascularity with necrosis or inflammation seemed to determine bone metabolism in MRONJ. The recurrence of MRONJ was observed in one case, and 3D SPECT/CT had preoperatively depicted the recurrence site.
Keywords: medication-related osteonecrosis of the jaw, three-dimensional SPECT/CT, vascularity, necrosis, infection
Introduction
Medication-related osteonecrosis of the jaw (MRONJ) arises in cancer or osteoporosis patients who have received antibone resorption agents including bisphosphonates and denosumab. It is regarded that MRONJ is pathologically similar or equal to osteomyelitis which mainly consists of necrosis, vascularity, and inflammation. However, little is known about the distributions of these pathologic components in MRONJ as well as the methods of visualizing them.
Bone scintigraphy is known to be highly sensitive to detect osteonecrosis of the jaw.1 According to a systematic review and international consensus, bone scintigraphy is recommended in addition to cone beam CT or CT in order to further evaluate involved jaw especially when performing surgical intervention for Stage 2 or 3 MRONJ.2 Tomographic imaging techniques such as bone single photon emission CT (SPECT)/CT has recently been used for pre-operative assessment of MRONJ.3,4 Three-dimensional (3D) SPECT/CT is a novel imaging technique that allows to exhibit 3D distribution of bone metabolism, which we reported was useful not only for surgical management in Stage 3 MRONJ5 and but for assessing jaw bone invasion in oral cancer patients.6 In the present study, a radiologic–pathologic study was conducted with an aim of investigating a 3D correlation between pre-operative 3D bone SPECT/CT and histopathological characteristics in MRONJ.
Methods and materials
This study was performed in accordance with ethics protocols approved by the Keio University Ethics Committee (20180127). Data from four patients (cases 1 to 4) diagnosed with either Stage 2 or 3 MRONJ of the mandible were retrospectively enrolled in this study. The 2014 American Association of Oral and Maxillofacial Surgeons criteria was used for deciding clinical stage of MRONJ.7 At the time of surgery, cases 1, 2, 3, 4 were at Stages 3, 3, 2, 3, respectively. In case 3, left mandibular canine and first premolar teeth with surrounding alveolar bone had been removed due to periodontitis long before surgery. The crest was subsequently covered with mucosal walls that were associated with fistula formation persisting over 8 weeks. In case 4, a lump of necrotic bone tissues had dropped off the mandible, and a fistula with mucosal walls had remained until sugery.
These patients underwent bone SPECT/CT using 99mTc hydroxymethylene diphosphonate (99mTc-HMDP) before marginal or segmental mandibulectomy. Coronal and horizontal sections of the resected jaws were obtained at approximately 5 mm intervals, and then stained with Hematoxylin and Eosin (H&E). In order to perform a 3D radiologic–pathologic correlation, an experienced pathologist mapped the pathologic results of the H&E sections with respect to necrosis, vascularity, and inflammatory cell infiltration based on the presence of empty bone lacunae, vascular endothelial cells, and inflammatory cell infiltration, respectively. Then, the sections were color-coded onto the photos of the jaws according to which pathologic feature was dominant within the section (Figure 1).
Figure 1. .
Histopathological results of the jaw in the studied cases were classified according to necrosis (N), vascularity (V), and inflammatory cell infiltration (I).
3D SPECT/CT
The imaging protocol and image generation of 3D SPECT/CT were described elsewhere.5,8 Briefly, 3D volume-rendered CT images of the mandible were reconstructed using data from the CT portion of bone SPECT/CT. Then, the radioisotope uptake based on bone metabolism was projected onto the 3D volume-rendered CT images. The resultant fused images offer anatomic and metabolic information of the jaw at any angle.
Two observers independently scored the abnormal tracer uptake in the areas with the histological features by using a visual 3-point scoring system: +, mild; ++, moderate; and +++, strong. When there were discordant results between the observers, the scoring was determined by a consensus.
Results
Classification of histologic results
The histopathological results showed distinct characteristics, which can be divided into five groups based on necrosis, vascularity, and inflammatory cell infiltration: 1, areas of necrosis without any tissue vascularity (N + V-). 2, areas of necrosis with both vascularity and acute inflammatory cell infiltration due to bacterial infection (N + V+I+). 3, areas of coexistence of necrosis and regenerative vasculature but no inflammatory cell infiltrate (N + V+I-). 4, chronic inflammation without massive necrosis (N-V +I+). 5, no-affected areas (vital bone) (Figure 1).
Cases 1 and 2 had the N + V- areas. The N + V- areas had a large amount of necrotic bone tissues with empty lacunae where vascularity was almost absent. Accordingly, neither inflammatory cell infiltration due to bacterial infection nor regenerative changes such as bone remodeling and angiogenesis was seen in these areas. The N + V- areas were slightly larger than the areas of necrotic bone exposure to the oral cavity. The N + V+I + areas showed vascularity and acute infection characterized by inflammatory cell infiltration including numerous neutrophils in addition to necrotic bone tissues. The N + V+I- areas included a substantial amount of newly formed bone accompanied by regenerative vascularity and necrotic bone with empty lacunae. There was no apparent inflammatory cell infiltration in the N + V+I- areas. The N-V +I+ areas showed predominant lymphocytic infiltration and an increase of areolar connective tissues.
We actually intended to remove necrotic bone tissues from necrotic core to bloody viable margins in all cases, although we found the N-V +I + area in cases 3 and 4, the N + V+I- area in cases 1, 2 and 4 at the margins of bone resection (Figure 2).
Figure 2. .
Gross specimens of the jaws (A, C, E, G) and the corresponding 3D SPECT/CT images (B, D, F, H) in the studied four cases for radiologic–pathologic correlation of MRONJ. The definition of color bars on the gross specimens is described in Figure 1. Red dotted lines on 3D SPECT/CT images indicate the resected lines set on actual surgeries. 3D SPECT/CT revealed that both the N + V+(yellow and blue) and N-V +I+ (green) areas were hypermetabolic whereas the N + V- (black) areas had no apparent bone metabolism in cases 1 and 2 (red arrows). There was no the N + V area in cases 3 and 4. Case 1 had a recurrence of MRONJ at the median side-of the margin (arrowheads) where increased bone metabolism was seen. 3D,three-dimensional; MRONJ, medication-related osteonecrosis of the jaw; SPECT, single photon emission CT.
Radiologic–pathologic correlation of MRONJ
The correlation results between radiotracer uptake and pathologic characteristics were summarized in Figures 2 and 3. The uptake was diverse even in the same classification zone, although the judgements whether abnormally increased tracer uptake was present or not in each zone were completely concordant between the two observers. The N + V- areas were observed as the defect of radioisotope uptake, surrounded by hypermetabolic areas representing the N + V+I + areas. The imaging finding is similar to so-called "cold-in-hot" appearance in the osteonecrosis of the femoral head. The N + V+I- and N-V +I+ areas were also displayed as hypermetabolic areas. There were no particular imaging findings for discriminating between the N + V+I-, N + V+I+, and N-V +I+ areas.
Figure 3. .
Summary of the radiologic-pathologic correlation results. The degree of radiotracer uptake was scored as – (no uptake), + (mild), ++ (moderate), +++ (strong). N.A., not applicable.
Among the four cases, Case one had the recurrence of MRONJ at the median side-of the margin during post-operative follow-up where the N + V+I- area was dominantly occupied. In retrospect, the patient’s 3D SPECT/CT clearly showed abnormal bone metabolism beyond the surgical margin (Figure 2, arrowheads).
Discussion
How much extent should be resected remains a serious problem in patients with MRONJ in terms of both recurrence and quality of life. At present, surgical approaches for MRONJ are divided into conservative and extensive surgical approaches.9 The extensive surgical approach is the removal of necrotic bone from necrotic core to bloody viable margins. However, we failed to achieve complete resection of necrosis and inflammatory lesions in all cases, one of which showed the recurrence of MRONJ; Case 1 had the recurrence of MRONJ at the margin having dominant bone necrosis which had been pre-operatively depicted as abnormal bone metabolism in 3D SPECT/CT. Restow et al mentioned that irrespective of the approach selected (minimally invasive or resective), the delineation between necrotic and viable bone is the crucial step and is a major challenge in this procedure.10 Therefore, although a small number of patients, this study focused on the investigation of 3D radiologic–pathologic correlation and aimed at an improved understanding of the spatial pathologic characteristics in patients with MRONJ.
We found that increased uptake of 99mTc-HMDP was strongly dependent on tissue vascularity that was associated with inflammatory and/or regenerative changes occurred in the necrotic tissues. Interestingly, the area of diminished vascularity was three-dimensionally surrounded by vascularized areas, which clarifies the reason for the common scintigraphic finding in MRONJ.3 The loss of vasculature almost corresponded to the area in which the necrotic bone was exposed to the oral cavity, which is consistent with previous reports.11,12
We found that increased 99mTc-HMDP uptake in the jaw was dependent on the presence of regenerative vascularity with necrosis or inflammation. Bedogni et al reported that osteomyelitis at the margins of bone resection is a risk factor for MRONJ recurrence.13 Especially when considering extended surgery such as segmental mandibulectomy with fibula free flap reconstruction,14 the presence of non-necrotic and non-inflammatory bone at the margins of resection is reportedly important to prevent MRONJ recurrence.15 In this context, to be on the safe side, the resection of both the photopenic area (N + V-) and the areas of increased radiotracer (N + V+I+, N + V+I-, and N-V +I+) might be an option of extensive surgery.
On the other hand, the conservative surgical approach is supposed to resect sequestrum (i.e. the N + V area) and/or to perform the superficial debridement of necrotic bone associated with antibiotic therapy (i.e. the N + V+I + area). Although the extensive surgical approach is better than the conservative surgical approach when prioritizing the cure of the disease, the conservative surgery may be an option in cases of the earlier stage or if the improvement of quality of life is a goal of the treatment.9 Moreover, complete mucosal healing can be achieved about 75% of patients with MRONJ even with the conservative surgery.9 Unfortunately, 3D SPECT/CT with 99mTc-HMDP may not be suitable for surgical planning of the conservative surgery due to the difficulty in discriminating the N + V+I + area from other hypermetabolic areas. However, 3D SPECT/CT can also be applied to inflammation-targeted nuclear imaging such as gallium-67 citrate or indium-111 white blood cell SPECT, which would facilitate further 3D mapping of different pathological characteristics. Moreover, the development of new radiotracers (such as necrosis/apoptosis imaging) will possibly enhance the usefulness of 3D SPECT/CT in the future.
Conclusion
In MRONJ, increased radiotracer uptake in the jaw was dependent on the presence of regenerative vascularity with necrosis or inflammation. Vasculature was absent in the photopenic areas on SPECT that almost corresponded to the area of exposure of necrotic jaw bone to the oral cavity. The clarification of these findings was delivered by a 3D radiologic–pathologic correlation using 3D SPECT/CT. The pre-operative 3D SPECT/CT may help oral oncologists to estimate the surgical margin. In addition, this technique would be helpful for oral surgeons who want to carefully contemplate what extent of jaw resection is optimal when considering the balance of recurrence and quality of life.
Footnotes
Conflict of interest: Tadaki Nakahara received research support by the Development of Medical Devices through Collaboration between Medicine and Industry from Japan Agency for Medical Research and Development, AMED.
REFERENCES
- 1.O'Ryan FS, Khoury S, Liao W, Han MM, Hui RL, Baer D, et al. Intravenous bisphosphonate-related osteonecrosis of the jaw: bone scintigraphy as an early indicator. J Oral Maxillofac Surg 2009; 67: 1363–72. doi: 10.1016/j.joms.2009.03.005 [DOI] [PubMed] [Google Scholar]
- 2.Khan AA, Morrison A, Hanley DA, Felsenberg D, McCauley LK, O'Ryan F, et al. Diagnosis and management of osteonecrosis of the jaw: a systematic review and international consensus. J Bone Miner Res 2015; 30: 3–23. doi: 10.1002/jbmr.2405 [DOI] [PubMed] [Google Scholar]
- 3.Dore F, Filippi L, Biasotto M, Chiandussi S, Cavalli F, Di Lenarda R. Bone scintigraphy and SPECT/CT of bisphosphonate-induced osteonecrosis of the jaw. J Nucl Med 2009; 50: 30–5. doi: 10.2967/jnumed.107.048785 [DOI] [PubMed] [Google Scholar]
- 4.Assaf AT, Zrnc TA, Remus CC, Adam G, Zustin J, Heiland M, et al. Intraindividual comparison of preoperative (99m)Tc-MDP SPECT/CT and intraoperative and histopathological findings in patients with bisphosphonate- or denosumab-related osteonecrosis of the jaw. J Craniomaxillofac Surg 2015; 43: 1461–9. doi: 10.1016/j.jcms.2015.06.025 [DOI] [PubMed] [Google Scholar]
- 5.Miyashita H, Shiba H, Kawana H, Nakahara T. Clinical utility of three-dimensional SPECT/CT imaging as a guide for the resection of medication-related osteonecrosis of the jaw. Int J Oral Maxillofac Surg 2015; 44: 1106–9. doi: 10.1016/j.ijom.2015.05.002 [DOI] [PubMed] [Google Scholar]
- 6.Miyashita H, Nakahara T, Asoda S, Kameyama K, Kawaida M, Enomoto R, et al. Clinical value of 3D SPECT/CT imaging for assessing jaw bone invasion in oral cancer patients. J Craniomaxillofac Surg 2019; 47: 1139–46. doi: 10.1016/j.jcms.2019.03.013 [DOI] [PubMed] [Google Scholar]
- 7.Ruggiero SL, Dodson TB, Fantasia J, Goodday R, Aghaloo T, Mehrotra B, et al. American Association of Oral and Maxillofacial Surgeons position paper on medication-related osteonecrosis of the jaw--2014 update. J Oral Maxillofac Surg 2014; 72: 1938–56. doi: 10.1016/j.joms.2014.04.031 [DOI] [PubMed] [Google Scholar]
- 8.Ogata Y, Nakahara T, Ode K, Matsusaka Y, Katagiri M, Iwabuchi Y, et al. 3D SPECT/CT fusion using image data projection of bone SPECT onto 3D volume-rendered CT images: feasibility and clinical impact in the diagnosis of bone metastasis. Ann Nucl Med 2017; 31: 304–14. doi: 10.1007/s12149-017-1158-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Rupel K, Ottaviani G, Gobbo M, Contardo L, Tirelli G, Vescovi P, et al. A systematic review of therapeutical approaches in bisphosphonates-related osteonecrosis of the jaw (BRONJ. Oral Oncol 2014; 50: 1049–57. doi: 10.1016/j.oraloncology.2014.08.016 [DOI] [PubMed] [Google Scholar]
- 10.Ristow O, Otto S, Geiß C, Kehl V, Berger M, Troeltzsch M, et al. Comparison of auto-fluorescence and tetracycline fluorescence for guided bone surgery of medication-related osteonecrosis of the jaw: a randomized controlled feasibility study. Int J Oral Maxillofac Surg 2017; 46: 157–66. doi: 10.1016/j.ijom.2016.10.008 [DOI] [PubMed] [Google Scholar]
- 11.Marx RE, Tursun R. Suppurative osteomyelitis, bisphosphonate induced osteonecrosis, osteoradionecrosis: a blinded histopathologic comparison and its implications for the mechanism of each disease. Int J Oral Maxillofac Surg 2012; 41: 283–9. doi: 10.1016/j.ijom.2011.12.016 [DOI] [PubMed] [Google Scholar]
- 12.Bedogni A, Blandamura S, Lokmic Z, Palumbo C, Ragazzo M, Ferrari F, et al. Bisphosphonate-Associated jawbone osteonecrosis: a correlation between imaging techniques and histopathology. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2008; 105: 358–64. doi: 10.1016/j.tripleo.2007.08.040 [DOI] [PubMed] [Google Scholar]
- 13.Bedogni A, Saia G, Bettini G, Tronchet A, Totola A, Bedogni G, et al. Long-Term outcomes of surgical resection of the jaws in cancer patients with bisphosphonate-related osteonecrosis. Oral Oncol 2011; 47: 420–4. doi: 10.1016/j.oraloncology.2011.02.024 [DOI] [PubMed] [Google Scholar]
- 14.Caldroney S, Ghazali N, Dyalram D, Lubek JE. Surgical resection and vascularized bone reconstruction in advanced stage medication-related osteonecrosis of the jaw. Int J Oral Maxillofac Surg 2017; 46: 871–6. doi: 10.1016/j.ijom.2017.01.023 [DOI] [PubMed] [Google Scholar]
- 15.Nocini PF, Saia G, Bettini G, Ragazzo M, Blandamura S, Chiarini L, et al. Vascularized fibula flap reconstruction of the mandible in bisphosphonate-related osteonecrosis. Eur J Surg Oncol 2009; 35: 373–9. doi: 10.1016/j.ejso.2008.05.002 [DOI] [PubMed] [Google Scholar]