The relationship between intravenous bisphosphonate administration and jaw bone osteonecrosis was first reported in 2003 [1]. It is defined as an area of exposed bone in the maxillofacial region for more than eight weeks in a patient who was receiving a bisphosphonate and had not had radiation therapy to the craniofacial region [2]. It is often denoted as bisphosphonate-related osteonecrosis of the jaw (BRONJ). This describes a spectrum of intraoral symptoms, especially seen in cancer patients treated with high levels of intravenous bisphosphonates.

Bisphosphonates are non-metabolic synthetic analogues of pyrophosphate which have potent inhibitory effects on bone resorption [3]. They are the drugs of choice in prevention and treatment of bone disorders characterised by increased bone resorption, such as Paget's disease, osteoporosis, hypercalcaemia of cancer, multiple myeloma and bony metastases [3].

Though there is abundant evidence in the literature of the benefit of bisphosphonates therapy, growing evidence from several recently published reports show that patients treated with bisphosphonates have a potential risk of develop necrosis of the jaw bone [4-10].

The diagnosis of BRONJ is currently based on the patients' medical history of bisphosphonate treatment and clinical evaluation (pain, bone exposure, purulent secretion or swelling). The initial appearance of the disease is variable and thus, it often comes to the attention of the clinician late, when BRONJ has become symptomatic.

The role of imaging in the diagnosis of BRONJ and the imaging findings has previously been described in the literature [11-13], but no distinct clinico-radiological systematization of the disorder has been reported.

Recently, several cone beam computed tomography (CBCT) systems have become available, which are specifically designed to image hard tissues in the maxillofacial region [14].

This results in new diagnostic possibilities with increased diagnostic image quality. With the available CBCT systems, high-contrast structures can be visualized at very low radiation doses, sometimes as low as two panoramic radiographs [15]. Such CBCT data might therefore help to define the radiological range of appearances of BRONJ.

The purpose of the present study was to systematize the clinico-radiological symptoms and course of bisphosphonate-related osteonecrosis of jaw bone and to create guidelines for the most optimal radiological examination methods in each stage of the disease.

Patient recruitment

A series of 22 subjects (17 males and 5 females, age range 54 - 87 years, mean age 69.2 years) with a diagnosis of osteonecrosis development in the jaw bone(s) were consecutively recruited from the patient sample visiting the Department of Paediatric and Special Dental Care, School of Dentistry, Oral Pathology and Maxillofacial Surgery, Faculty of Medicine, Catholic University of Leuven, Leuven, Belgium between October 2007 and March 2008. They all had undergone a biophosphonate treatment based on a previous diagnosis of extraoral malignant disease. All patients gave their informed consent to participate in the study, in accordance with the ethical regulations of the University Hospitals, Catholic University of Leuven.

Nine patients had multiple myeloma, 6 prostate cancer, 3 breast cancer and 4 renal cell cancer. All patients had been treated with either individual intravenous doses of zoledronate or pamidronate and were therefore at risk for developing BRONJ. The diagnosis of BRONJ was confirmed through the clinical evaluation of the exposed bone to the oral cavity in addition to the negative history of radiotherapy to the head and neck.

The presenting symptoms: the localization of the necrosis, the presence of suppuration and the association with previous extractions or traumas as possible triggering factor for the onset of the lesions were documented in all patients.

Radiological examination

As part of a routine clinical examination, a digital panoramic radiograph was taken using a Cranex Tome^® (Soredex, Tuusula, Finland) and the CBCT of the jaws was obtained using a SCANORA 3D^® CBCT unit (Soredex, Tuusula, Finland). The CBCT device uses complementary metal-oxide semiconductor (CMOS) flat panel/CsI detector and is operated at 85 kV and 13 mA, with a full jaw scan with a 75 x 145 mm field of view for a single 360° rotation.

Primary reconstruction of the data provided 1.0 mm axial, coronal, and sagittal slices (300 - 500 slices) as well as the three-dimensional volumetric surface renderings. The data set had a voxel size of 0.25 x 0.25 x 0.25 mm and consisted of contiguous slices with respect to the Z-axis. These were reconstructed secondarily to provide a curved planar two-dimensional multiplanar reconstruction (MPR) 20 mm thick along the arch of the mandible that provides a simulated "panoramic" equivalent. Panoramic reconstruction at narrower defined "focal trough" limits (such as 5 – 10 mm) better depicted the degree of involvement and extent of margins in a projection that was still cross-sectional imaging at 1 mm thick slices, and in 1 mm slice increments also provided orthogonal views to the panoramic curved planar view. Previous radiographs of the patients were also retrieved and correlated with the present ones for possible changes.

The radiographs and CBCT images were evaluated by two dentomaxillofacial radiologists under standardized viewing conditions (40 cm viewing distance, diagnostic screen, dimmed room). Calibration was done beforehand to reduce the intra- and inter-examiner variability regarding the following variables: imaging characteristics of the alveolar and basal bone (lytic/erosive, sclerotic, and mixed), presence of sequestrum, cortical discontinuity, periosteal response, involvement of vital structures (e.g. the mandibular canal and maxillary antrum) and soft tissues involvement. Four stages of the disease progression were proposed based on the clinico-radiologic findings.

Demographic data are presented in Table 1. In most patients, the lesions initially occurred after dental extraction. The body of the mandible was the predominant clinical site of involvement in all the patients. Area of necrosis was discovered on the CBCT image of the mandible in one of the patients with an extensive necrosis of the maxilla.

The staging of the disorder was done based on the clinico-radiological findings (Table 2).

The least changes at first presentation were observed in patients with radiological evidence of osteosclerosis involving the cortical bone, alveolar margin and lamina dura but with clinically intact mucosa and no oral symptoms or discomfort (stage I). This radiological finding was only detected using the periapical radiographs and CBCT (Figures 1 and 2).

In stage II, ulceration was located in the posterior/lingual mandible in the area of the mylohyoid ridge, the exposed bone surface was smooth and patients presented with discomfort as the tongue constantly rub on the ulceration. Thickening of the cortical plate in the affected region was the only radiological findings on CBCT.

Panoramic radiographs did not yield obvious evidence of the disease status in stage I and II of the disease (Table 2).

Increases in sclerotic manifestations correlated with a more sever expression of BRONJ in stage III and IV lesions were detected on both imaging modality, yet much more pathologic characteristics and the entire lesion extent could be visualized on CBCT (Table 2) (Figures 3-5).

A major limitation of this study is the small sample size; however the data presented was representative of the various clinico-pathologic manifestation of BRONJ. Bisphosphonate-associated necrosis of jaw bone remains a challenge with respect to diagnosis and management. This presents a growing concern for oral health care workers since the disorder appears almost exclusively in the jaws [4,7,16]. This is because the diagnosis is made at the late stage when there is bone exposure to the oral cavity. Standard diagnosis based on clinico-radiologic criteria is still lacking and there are no clinico-radiological guidelines for the clinicians to follow. Bone exposure often occurred spontaneously or after dental extractions or oral trauma with progression to sequestration associated with purulent discharge and in some cases fistula formation [4,8,17]. Most reported cases of necrosis of the jaws associated with bisphosphonate use involve the recent nitrogen-containing injectable bisphosphonates such as pamidronate and zoledronate [1,8,18]. Incidentally, all patients in this study were treated with this group of bisphosphonates.

The pathogenesis of BRONJ remains unclear however, many theories have been proposed by various authors [4,6]. Most cases of osteonecrosis in this study occurred in the posterior alveolar region of the mandible where the cortical bone is thickest. The reason for this has been attributed to the high bisphosphonates concentration in the region as a result of high drug uptake and high bone turnover due to constant mastication. [11,19] This condition is said to inhibit osteoclastic activity, thereby decreasing bone remodeling and predisposing the bone to accumulated micro-damage, potentially causing the onset of osteonecrosis.

The most common clinical finding in this series is an area of ulcerated mucosa with exposed devitalized bone. In all patients with exposed necrotic bone, the observed area of ulceration was smaller with wider area of underlying bone necrosis as previously reported by Bedogni et al. [Figure 1), while a focal area of sclerosis was detected on the CBCT of the mandible of the other patient who only had bony exposure on the maxilla (Figure 2). Both patients had intact oral mucosa and did not present with any oral symptoms or discomfort in the sclerotic region. This mild involvement represents the stage I of the disorder. This can easily be detected with periapical radiograph since the bone changes are still limited to the periapical region. Indeed, active bone resorption has been reported to be absent at this early stage [6,10], CBCT provided a good view of this early discrete and subtle bone changes (Figure 2).

Avoiding invasive procedures and institution of preventive measures at this bone pre-exposure stage may delay the clinical progress and possibly provide an environment conducive to resolution of the condition.

Patients with stage II of the disease had an area of ulceration at the posterior/lingual mandible in the area of the mylohyoid ridge with no apparent radiological changes on the panoramic view but presents as cortical thickening on the CBCT. This region had earlier been reported as one of most common site of BRONJ [6]. The ulceration might have been due to the pressure effect of the tongue on the elevated ridge and the bone reaction may still be limited to the periosteum at this stage.

Diagnosis at these vital stages may give room for prompt implementation of the preventive measure recommended for BRONJ in the literature [4,6,7] and adequate follow-up to check the progression of the disease.

Panoramic radiograph may not be useful in the diagnosis of this early lesssion because of its limited resolution, two-dimensional nature with anatomic overlap and tomographic effect which may not allow visualization of the true extent of the bone defects. The present finding correlate well with those earlier described by Arce et al. [12].

Patients with a more advanced stage of the disease presented with bone exposure to the oral cavity at previous tooth extraction sites [8,13]. The exposed bone had a grayish discoloration with irregular surface, which on probing was asymptomatic and avascular. These findings were consistent with avascular osteonecrosis and similar to the documented findings in literature [1,6,8]. The panoramic radiograph in this group of patient showed an area of sclerosis with some radiolucency with poorly defined margin. CBCT was able to provide detailed information about characteristics and extent of the cortical and trabecular bone involvement and the proximity of the alteration in the bony architecture to vital structure such as mandibular canal (Figure 3).

In most severe cases, Stage IV the CBCT and the panoramic radiograph now revealed larger areas of osseous sclerosis involving the entire cross-sectional area of the jaw in all cases, with sclerotic changes encroached on the mandibular canal and the maxillary antrum. Devitalized bone with radiodense appearance and radiolucent rim (Figure 3) denoting bone sequestration were also predominant at this advanced stage of the disease (Figures 3 and 4), while pathologic fracture of the mandible was observed in two patients (Figures 4 and 5).

In all the patients, necrosis involved the alveolar margin and the cortical plate, in line with the findings reported by Phal et al. [11]. This suggests that BRONJ possibly originates in the cortical or alveolar bone with a progression towards the medulla.

CBCT imaging allowed high resolution hard tissue diagnosis, and when offering a field of view, covering both jaws, affected areas could be easily compared to non-affected areas while asymptomatic area can also be detected.

Furthermore, the CBCT data could also be considered valuable for volumetry which helps to determine the extent of the disease. As such it can serve as a guide towards management planning. Indeed, cross-sectional slices allowed identification of the true extent of affected marrow and thus facilitated transfer from the lamina dura three-dimensional images to the surgical topographical landmarks and vice versa.

The observed radiological characteristics suggests a distinct pattern of occurrence and progress of BRONJ, starting from the lamina dura and cortical plate, to involve the alveolar bone and then progress to enter the medullar bone, and finally the entire cross-sectional area of the jaw bone. This sequence is confirmed in subjects with image follow-up over time; showing that the destructive bone changes were often progressive (Figures 1 and 5).

While routine jaw imaging for patients receiving bisphosphonate therapy may not be justified, it is essential that oral health care providers rule out such early bone changes, when a patient at risk presents with a need for jaw bone imaging. This may allow for early diagnosis, better treatment and prevention of new potential case as this can make a significant difference in the outcome of the disease.

The present study showed that bisphosphonate-related osteonecrosis of jaw bone occur in four distinct clinico-radiological stages. For mild cases, panoramic image diagnosis was much less obvious. Whereas cone beam computed tomography was able to fully characterise the bony lesions and describe their extent and involvement of neighbouring structures in all cases. Thus cone beam computed tomography might better contribute to the prevention of bisphosphonate-related osteonecrosis of jaw bone as well to disease management.