Head and Neck Imaging Original Research

Transcription

1 Head and Neck Imaging Original Research García-Ferrer et al. of Mandibular Osteonecrosis Head and Neck Imaging Original Research Luis García-Ferrer 1 Jose V. agán 2 Vicente Martínez-Sanjuan 3 Sergio Hernandez-azan 2 Raquel García 3 Yolanda Jiménez-Soriano 2 Vicente Hervas 1 García-Ferrer L, agán JV, Martínez-Sanjuan V, et al. Keywords: bisphosphonates, jaw,, osteonecrosis DOI: /JR Received ugust 20, 2007; accepted after revision October 27, Department of Radiology, Consorcio Hospital General, Universitario de Valencia, Valencia, Spain. 2 Department of Stomatology, Consorcio Hospital General, Universitario de Valencia, Valencia, Spain. 3 CT and MR ERES Unit, Consorcio Hospital General, Universitario de Valencia, vida. Tres Cruces s/n, Valencia, Valencia, Spain ddress correspondence to V. Martínez-Sanjuan (marsan@comv.es). JR 2008; 190: X/08/ merican Roentgen Ray Society of Mandibular Osteonecrosis Secondary to isphosphonates objective. isphosphonates are a group of drugs used in the treatment of oncology patients with bone metastases. However, in the past few years, osteonecrosis of the jaw has been reported as a serious complication of such treatment. The objective of this study was to examine the use of in the assessment of bone lesions caused by this disease. MTERILS ND METHODS. Fourteen patients were studied who had been treated with IV bisphosphonates and had developed focal lesions of osteonecrosis of the jaw. These patients were referred by the stomatology department of Hospital General Universitario de Valencia. We evaluated both the morphology and the behavior of the lesions in T1, STIR, and after the administration of gadolinium. RESULTS. Twenty-six focal lesions were detected clinically and 36 were detected radiologically. ll the clinically detected focal lesions were visible on. There were 15 focal lesions detected radiologically that were not detected on clinical examination. In all patients, it was possible to assess bone involvement and involvement of the bone marrow, soft tissues, sinuses, and mandibular canal as well as the presence of adenopathy. CONCLUSION. is an effective tool in the assessment of osteonecrosis of the jaw. The significance of focal lesions detectable on radiologic examination but without clinical correlation and their progression over time remains to be determined. isphosphonates are synthetic com pounds with a structure similar to inorganic pyrophosphate, obtained by substituting an oxygen molecule for a carbon molecule between two phosphates (P-C-P). With a great capacity for binding to the bone matrix, this structure makes them more resistant to hydrolysis in an acid medium and to the action of pyrophosphatase and therefore more difficult to metabolize [1]. These drugs are used in the treatment of lytic bone metastases, multiple myeloma, hypercalcemia of malignant origin, osteoporosis, and diseases such as Paget s disease, providing a significant improvement in the symptoms as a result of reducing pain, bone demineralization, and bone fractures, either pathologic or due to insufficiency. Given the prevalence of these diseases, bisphosphonates are one of the most prescribed drug groups in the world [1, 2]. lthough the drugs are well tolerated with a low incidence of adverse effects, in the past few years, a new problem has surfaced associated with the bisphosphonates, which have nitrogen in their structure (, zole- dronic acid, and alendronate): osteonecrosis of the jaw (ONJ). This adverse effect was first described in 2003 by Marx [3], Migliorati [4], and Pogrel [5], with further reports appearing later [1, 6 17]. In September 2004, in view of the growing number of publications, the manufacturer and the U.S. Food and Drug dministration (FD) accepted osteonecrosis as a possible side effect of the use of Zometa (zoledronic acid, Novartis) and redia (, Novartis) [18]. Until 2006, except for a description of the CT images in one case [7], there had been no studies describing the assessment of bisphosphonate-associated osteonecrosis by. series of 11 patients has recently been published [19] with scintigraphy, CT, and studies, but few details are provided about the characteristics on. The aim of this article is to contribute a new series of cases with this disease and its study following a specific protocol. Materials and Methods The patients attended the CT and unit of the Consorcio Hospital General Universitario de JR:190, pril

2 García-Ferrer et al. Valencia. They were referred by the stomatology department with the clinical and histopathologic diagnosis of ONJ. ll signed informed consent for the. The studies were performed with a 1.5-T superconducting magnet (Somatom Sonata, Siemens Medical Solutions) using the 8-channel head antenna. The examination was performed according to the following protocols. Three-plane locator (FLSH): axial T1-weighted (TR/TE, 639/17; number of excitations [NEX], 4; matrix, ; field of view, 240 mm; slice thickness, 5 mm), axial STIR (6,230/93; inversion time, 130 milliseconds; NEX, 4; field of view, 240 mm; matrix, ; slice thickness, 5 mm), axial T1- weighted turbo FLSH fat-saturated (263/4.76; flip angle, 70 ; NEX, 3; field of view, 200 mm; matrix, ; slice thickness, 5 mm), and oblique sagittal of each side T1-weighted (400/17; NEX, 5; field of view, 140 mm; matrix, ; slice thickness, 3 mm). STIR: sagittal of the diseased side that is, of the side in which the T1 images show low-signal lesions and if bilateral, of both sides (2,500/99; inversion time, 130 milliseconds; NEX, 4; matrix, ; field of view, 140 mm; slice thickness, 3 mm), axial diffusion-weighted imaging (DWI) (3,000/83; NEX, 3; field of view, 230 mm; matrix, ; b = 0, 500, and 1,000 s/mm 2 ; slice thickness, 5 mm), and axial perfusion (42/2.47; flip angle, 70 ; field of view, 280 mm; fat satur ation; NEX, 1; matrix, ; slice thickness, 5 mm; repeated 20 times). fter this acquisition, the axial sequence T1- weighted turboflsh fat-saturated, the axial sequence T1-weighted, and the oblique sagittal of the affected side or sides were all repeated. The scanning duration was about 45 minutes. The contrast material was injected with a power injector (Mississippi, Ulrich), at a rate of 3 ml/s at a dose of 0.1 mmol/kg of body weight, followed by 60 ml of normal saline. The most objective way to check the contrast enhancement is to compare images of the same series before and after con trast administration. The images were assessed by two radiologists with head and neck experience who were unaware of the clinical location of the lesions. ll the cases were later reviewed with the dental surgeons and assessed with the findings from the dental views. In our studies, we assessed the presence of areas of change in signal in the bone, primarily in the mandible and the upper maxilla. This signal change shows up as hypointense areas in the T1- weighted sequences before injecting the contrast material and with high-intensity signal in the STIR sequences. These areas were measured, particularly the larger ones, and the independent focal areas found in each subregion were counted. We categorized four quadrants for both clinical and radiologic assessment: upper maxilla, right side (quadrant 1); upper maxilla, left side (quadrant 2); mandible, left side (quadrant 3); and mandible, right side (quadrant 4). We assessed involvement of the cortical bone, the bone marrow, and extension to the soft tissues adjacent to the cortical bone. In the upper maxilla, extension to the maxillary sinus or nasal fossa was also assessed. of the mandibular canal and the mental foramen was assessed in the mandible. We evaluated the enhancement of the paramagnetic contrast material by the inflammatory TLE 1: Characteristics of the Patients with Osteonecrosis of the Jaw (ONJ) Patient No. Sex ge (y) Underlying Disease isphosphonate 1 F 46 reast cancer Zoledronic acid and Treatment Time (mo) Cause of ONJ 48 Extraction 2 F 61 reast cancer Zoledronic acid 36 Extraction 3 M 55 Prostate cancer Zoledronic acid 20 Unknown 4 F 62 reast cancer Zoledronic acid 26 Extraction 5 M 74 Multiple myeloma Zoledronic acid 20 Extraction 6 F 56 Multiple myeloma Zoledronic acid 18 Prosthesis 7 M 53 Multiple myeloma Zoledronic acid 24 Prosthesis 8 F 51 reast cancer Pamidronate 24 Unknown 9 F 39 reast cancer Zoledronic acid and 10 F 59 Multiple myeloma Zoledronic acid and 46 Extraction 59 Extraction 11 F 59 reast cancer Zoledronic acid 27 Extraction 12 M 81 Multiple myeloma Zoledronic acid 24 Unknown 13 M 52 Multiple myeloma Pamidronate 21 Unknown 14 F 70 reast cancer Zoledronic acid and 60 Prosthesis TLE 2: Correlation of Focal Lesions on and Examination Patient No. First Osteonecrosis Second Osteonecrosis Third Osteonecrosis Fourth Osteonecrosis a a Note Gray boxes indicate quadrants in which focal lesions were detected on without clinical correlation. a lesions detected on only after knowledge of the dental surgeon s evaluation. 950 JR:190, pril 2008

3 of Mandibular Osteonecrosis component of the lesions, with enhance ment being observed in the areas of inflammatory tissue in which previous gadolinium injections showed low signal intensity in T1-weighted images. nother element we looked for was the presence of nodal involvement indicating the chain in which we identified adenopathy (sub mandi bular, sub mandibular angle, and jugular digastric). Results Our study contained 14 patients (Table 1), five men (35.7%) and nine women (64.3%), average age ± (± SD) years. The indication for treatment with bisphosphonates was breast cancer bone metastases in seven patients (50%), multiple myeloma in six (42.9%), and prostate cancer bone metastases in one (7.1%). Eight patients (57.1%) re ceived zoledronic acid, two (14.3%) received, and four (28.6%) received plus zoledronic acid. The average time during which they received the treatment with bisphosphonates was ± 14.7 months. The trigger factor was a tooth extraction in seven patients (50%), prosthesis implant in three patients (21.4%), and unknown in four (28.6%). In total, 36 focal lesions were detected radiologically, whereas only 26 were confirmed clinically. ll the clinically detected focal lesions were visible in the study (Table 2), but not all those detected by were visible in the clinical examination, even after the dental surgeon knew where to look. On evaluation of the degree of bone involvement (Table 3), 14 patients (100%) had TLE 3: Distribution of the Osteonecrotic Lesions Detected on Patient No. Cortical one Marrow Soft-Tissue Sinus Mandibular Canal 1 Yes No No No No 2 Yes Yes Yes Yes No 3 Yes Yes Yes No Yes 4 Yes Yes Yes Yes Yes 5 Yes Yes Yes No Yes 6 Yes Yes No No Yes 7 Yes Yes No No Yes 8 Yes Yes No No Yes 9 Yes Yes Yes No Yes 10 Yes Yes No No Yes 11 Yes Yes No No Yes 12 Yes Yes Yes No No 13 Yes Yes Yes No No 14 Yes Yes Yes Yes No cortical lesions, 13 (92.9%) also had bone marrow involvement (Fig. 1), eight (57.1%) patients had soft-tissue involvement (Fig. 2), three (21.4%) also had maxillary sinus lesions (Fig. 3), and in nine (64.3%) patients, there was occupation of the mandibular canal (Fig. 4). ll the patients studied had submandibular adenopathy, 71.4% of them also in the submandibular angle and jugular digastric chain (Figs. 1 and 3). The mandibular osteonecrosis lesion was identified as delimited focal lesions with reduction of the signal in enhanced sequences on T1-weighted (100%) imaging. The STIR study showed an increase in the lesion signal (Fig. 5) in nine patients (64.3%), whereas in four (28.6%), there was virtually no bright signal from the lesion and in one case the lesion showed no signal. In these last four patients, after the IV administration of paramagnetic contrast material, there was a significantly lower enhancement (Fig. 6) than in the lesions that showed increased signal in STIR sequences (Table 4). Fig year-old man with prostate cancer., T1-weighted image shows hypointense area in right mandible (white arrow) that corresponds to focal lesion of osteonecrosis and associated adenopathy (black arrow)., Photograph shows clinical lesion. JR:190, pril

4 García-Ferrer et al. Discussion isphosphonate-associated osteonecrosis is a new disease that is becoming increasingly more common. search on MEDLINE with the terms osteonecrosis, jaw, and bisphosphonates comes up with 127 articles, two having been published in 2003, seven in 2004, 50 in 2005, and 118 in Various texts discuss the help of imaging in the diagnosis of this disease, but only one mentions the use of for classification [19]. This article characterizes the lesions of Fig year-old woman with breast cancer., xial T1-weighted image shows mass of soft tissue (large arrow) that affects masseter muscle and internal pterygoid with extension reaching cheek. Small arrow indicates break of cortical bone in retromolar trigone area., Photograph shows clinical lesion. Fig year-old woman with breast cancer., Oblique sagittal T1-weighted image shows occupation of right maxillary sinus caused by hypointense lesion in upper right maxilla (white arrow), causing lysis of floor of maxillary sinus. ssociated submaxillary adenopathy (black arrow) is seen., Photograph shows clinical lesion (arrow) of osteonecrosis of the jaw. 952 JR:190, pril 2008

5 of Mandibular Osteonecrosis Fig year-old woman with breast cancer., Oblique sagittal T1-weighted image shows focal lesion of osteonecrosis (arrow) affecting mandibular branch and involving mandibular canal., Photograph shows clinical image. 11 patients by orthopantomography, CT,, and scintigraphy, arriving at the conclusion that the first examination to be performed should be orthopantomography; that CT is very useful for the ability to see and characterize the extension of the lesions and in detecting cortical involvement; that should be reserved for those patients who have soft-tissue extension; and that scintigraphy is highly sensitive and can be a useful screening tool for this disease. TLE 4: Mandibular Osteonecrotic Lesion on Sequences Patient No. T1 STIR Gadolinium 1 Low signal Negligible brightness Enhancement 2 Low signal right signal Enhancement 3 Low signal right signal Enhancement 4 Low signal right signal Enhancement 5 Low signal right signal Enhancement 6 Low signal right signal Enhancement 7 Low signal right signal Enhancement 8 Low signal Negligible brightness Poor enhancement 9 Low signal right signal Enhancement 10 Low signal No brightness Poor enhancement 11 Low signal Negligible brightness Poor enhancement 12 Low signal right signal No contrast used 13 Low signal Negligible brightness Poor enhancement 14 Low signal right signal Enhancement In, the osteonecrosis appeared hypointense on T1-weighted images. However, in our series, there was different behavior in STIR and after the administration of gadolinium. The lesions showing very little bright ness in STIR had little or no contrast enhancement, all suggestive of nonviable bone. These findings have also been described by Chiandussi et al. [19], being classed as areas of bone sequestrum. Nevertheless, after review of the orthopantomograms, we found no radiologic data compatible with this diagnosis. DWI and perfusion-weighted imaging (PWI) were acquired in our study. In DWI, the artifacts from echo-planar imaging acquisition were present in all the patients, and image quality was not good enough to provide correct anatomic definition. The perfusionweighted images were acquired with T1- weighted fat saturation (turboflsh), and they provide correct dynamic information about the enhancement of the lesions. This enhancement was maintained in the later sequences after gadolinium injection. For this reason, we considered PWI unnecessary in this pathology. In addition to the behavior of the lesion according to the sequence, we are able to add more information evaluating the involvement of cortical bone, bone marrow, adjacent soft tissues, paranasal sinuses and the mandibular canal, and locoregional adenopathy as well as supply information on the number and location of focal lesions. The treatment for ONJ is controversial and depends to a great extent on the experience of the unit in treating this disease. In our center, débridement of the necrotic bone is performed only on the exposed lesions, leaving them to be treated conservatively in cases that have not exteriorized. lthough the surgical technique involves extracting bone until healthy edges are reached on the basis of bleeding from the JR:190, pril

6 García-Ferrer et al. wound borders, the information we provide about the extension of the lesion, parts affected, and the size are of great assistance when it comes to planning the procedure. In patients treated with bisphosphonates, the development of ONJ is generally preceded by a C Fig year-old woman with breast cancer. C, Typical behavior of osteonecrotic lesion on T1 (), STIR (), and contrast-enhanced T1 (C) sequences. Hypointense lesion with bright signal in STIR and contrast enhancement in quadrants 1 and 2 are seen. rrows show focal lesions of osteonecrosis in quadrant 1 (black arrows) and quadrant 2 (white arrows). D, Photograph shows clinical lesion in quadrant 1. tooth extraction, although in up to 20% there is no identifiable trigger factor. This means that taking a biopsy of a focal lesion with characteristics on of osteonecrosis but with no clinical signs has obvious ethical connotations because of the risk of triggering the disease process. In our series, there were 15 quadrants with radiologically detected lesions compatible with bisphosphonate-associated osteonecrosis but without clinical correlation. The bisphosphonates are distributed over all the bones and follow the same pattern over the jaws. We D 954 JR:190, pril 2008

7 of Mandibular Osteonecrosis believe that these focal lesions are affected areas of the mandible on which a factor (as yet unknown) triggering the process of infection and opening up the focal lesion has not yet acted. Chiandussi et al. [19] described the case of an asymptomatic patient with signs of hyperperfusion on the scintigram but with no findings on either the orthopantomogram or CT (the evaluation of this patient is omitted). This case, even taking into account the omission of the as a negative aspect of the test, supports the existence of silent focal lesions of bisphosphonate-associated osteonecrosis. We believe that there is a need for prospective studies of patients to confirm these data. In conclusion, we believe that is a very useful imaging technique for the assessment of patients with bisphosphonate-associated osteonecrosis, providing information about the number and extension of the focal lesions and anatomic involvement. ecause it could be used as a technique for early detection in patients susceptible to this disease, the lesions diagnosed would not be as advanced as they were in our study, which means probably a better prognosis due to early treatment and less necessity of surgery. References 1. Cheng, Mavrokokki, Carter G, et al. The dental implications of bisphosphonates and bone disease. ust Dent J 2005; 50[4 suppl 2]:S4 S13 Fig year-old woman with multiple myeloma. C, T1 (), STIR (), and contrast-enhanced T1 (C) sequences show atypical behavior of osteonecrotic lesion (arrows). There is hypointense lesion with little brightness on STIR () and with no contrast enhancement (C). 2. Marketos M. The top 200 brand drugs in 2003 (by units). Drug Topics March 22, 2004:76 3. Marx RE. Pamidronate (redia) and zoledronate (Zometa) induced avascular necrosis of the jaws: a growing epidemic. J Oral Maxillofac Surg 2003; 61: Migliorati C. isphosphonates and oral cavity avascular bone necrosis. J Clin Oncol 2003; 21: Pogrel M. isphosphonates and bone necrosis. J Oral Maxillofac Surg 2004; 62: Lugassy G, Shaham R, Nemets, en-dor D, Nahlieli O. Severe osteomyelitis of the jaw in long-term survivors of multiple myeloma: a new clinical entity. m J Med 2004; 117: Vannucchi M, Ficarra G, ntonioli E, osi. Osteonecrosis of the jaw associated with zoledronate therapy in a patient with multiple myeloma. r J Haematol 2005; 128: Schwartz HC. Osteonecrosis and bisphosphonates: correlation versus causation. J Oral Maxillofac Surg 2004; 62: Ruggiero SL, Mehrotra, Rosenberg TJ, Engroff SL. Osteonecrosis of the jaws associated with the use of bisphosphonates: a review of 63 cases. J Oral Maxillofac Surg 2004; 62: Hellstein JW, Marek CL. isphosphonate osteochemonecrosis (bis-phossy jaw): is this phossy jaw of the 21st century? J Oral Maxillofac Surg 2005; 63: Carter G, Goss N, Doecke C. isphosphonates and avascular necrosis of the jaw: a possible association. Med J ust 2005; 182: Durie G, Katz M, Crowley J. Osteonecrosis of the jaw and bisphosphonates. (letter) N Engl J Med. 2005; 353:99 102; discussion Woo S, Hande K, Richardson PG. Osteonecrosis of the jaw and bisphosphonates. (letter) N Engl J Med 2005; 353:99 102; discussion Maerevoet M, Martin C, Duck L. Osteonecrosis of the jaw and bisphosphonates. (letter) N Engl J Med 2005; 353:99 102; discussion agan JV, Murillo J, Jimenez Y, et al. vascular jaw osteonecrosis in association with cancer chemotherapy: series of 10 cases. J Oral Pathol Med 2005; 34: Jimenez-Soriano Y, agan JV. isphosphonates, as a new cause of drug-induced jaw osteonecrosis: an update. Med Oral Patol Oral Cir ucal 2005; 10[suppl 2]:E88 E Pastor-Zuazaga D, Garatea-Crelgo J, Martino- Gorbea R, Etayo-Pérez, Sebastián-López C. Osteonecrosis of the jaws and bisphosphonates: report of three cases. Med Oral Patol Oral Cir ucal 2006; 11:E76 E Hohneker J. Novartis oncology. U. S. Food and Drug dministration Website. medwatch/sfety/2004/zometahcp.pdf. ccessed September 24, Chiandussi S, iasotto M, Dore F, Cavalli F, Cova M, Di Lenarda R. and diagnostic imaging of bisphosphonate-associated osteonecrosis of the jaws. Dentomaxillofac Radiol 2006; 35: C JR:190, pril