| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Head and Neck Imaging |
1 From the Departments of Radiology (L.A.L., D.M.Y.) and Pathology (K.T.M.), and the University of Pennsylvania School of Medicine (R.L.K., P.K., W.H.), University of Pennsylvania Medical Center, 3400 Spruce St, Philadelphia, PA 19104. From the 1998 RSNA scientific assembly. Received August 30, 1999; revision requested October 8; final revision received January 4, 2000; accepted January 12. L.A.L. supported by an RSNA Research and Education Foundation Scholar grant. W.H. supported by the RSNA Research and Education Foundation Medical Student/Scholar Assistant program. Address correspondence to L.A.L. (e-mail: loevner@oasis.rad.upenn.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Computed tomographic (CT) scans (n = 6) and magnetic resonance (MR) images (n = 2) in seven patients with histopathologically proved PTLD of the Waldeyer (lymphoid) ring or cervical lymph nodes were retrospectively reviewed for abnormalities.
RESULTS: The interval between transplantation and PTLD onset was 3.5–108 months (mean, 30 months). All patients had imaging abnormalities involving the Waldeyer ring, and focal 2.0–4.5-cm masses were present in six patients (unilateral oropharyngeal tonsil in two, bilateral oropharyngeal tonsils in one, nasopharyngeal adenoids in three, unilateral pharyngeal tonsil and ipsilateral nasopharynx in one). In three patients, the mass was centrally low attenuating at CT or isointense to fluid at MR imaging, with enhancing solid peripheral lymphoid tissue. Three patients also had nodal findings: one with a 7-cm low-attenuating nodal mass in the right neck and two with numerous bilateral lymph nodes (mostly normal sized) in the anterior and posterior cervical lymph chains. One patient also had a small mass in the upper mediastinum.
CONCLUSION: In the setting of organ transplantation, findings of masses in the Waldeyer ring or an excessive number of cervical nodes should increase the index of suspicion of PTLD.
Index terms: Head and neck neoplasms, 28.34 • Head and neck neoplasms, CT, 28.12111, 28.12112, 28.12115 • Head and neck neoplasms, MR, 28.121411, 28.12143 • Transplantation, 51.459, 60.459, 76.459, 770.459, 81.455 • Lymphatic system neoplasms, 997.34
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
The frequency of PTLD has been reported to range from 1% to 10%, with substantial variability depending on the organ transplanted (3–9). Walker et al (8) reported PTLD in 6.2% of patients after lung transplantation, 5.2% of patients after kidney and pancreas transplantation, 2.0% of patients after heart transplantation, and 1.4% of patients after liver transplantation. The difference in the frequency of PTLD among these groups may be attributed to organ-specific immunosuppressive regimens. The association of Epstein-Barr virus (EBV) infection with the development of PTLD has been well established and is hypothesized to occur through unregulated B-cell proliferation. PTLD encompasses a spectrum of clinical manifestations, in addition to a wide range of histopathologic findings, from B-cell hyperplasia to lymphoma. Lesions primarily occur in the gastrointestinal tract, central nervous system, allografted organ, and, less commonly, lymph nodes (6).
We performed this study to determine the unusual findings of PTLD of the head and neck in adult patients who have undergone solid organ transplantation, with lesions involving the Waldeyer ring (lymphoid ring) in the pharynx, as well as lesions involving the cervical lymph nodes. The magnetic resonance (MR) imaging and computed tomographic (CT) findings will be described.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Histopathologic Findings
The diagnosis of PTLD was histopathologically established with the presence of atypical lymphocytic infiltrates and/or destruction of the normal parenchymal architecture at the biopsy sites in the neck. PTLD was classified according to a system proposed by Nalesnik et al (10) and Knowles et al (11), with subtypes that include plasmocytic hyperplasia, polymorphous B-cell hyperplasia, polymorphous B-cell lymphoma, non-Hodgkin lymphoma, and multiple myeloma.
To determine the frequency of EBV infection in the tissues affected by PTLD, in situ hybridization for the EBV genome was performed with all specimens. Testing included evaluation for EBERI, an RNA sequence that is abundantly expressed during latent EBV infection, and tandem repeats of NotI, a DNA sequence that is amplified during active EBV infection.
Imaging
CT and MR imaging of the neck were performed in six and two patients, respectively. CT was performed at our institution with a helical scanner (model 9800; GE Medical Systems, Milwaukee, Wis), with acquisition of a set of 5-mm-thick transverse images that covered from the skull base to the aortic arch. Two sets of images were obtained during each CT examination, one optimized for bone detail and the other optimized for soft-tissue detail. Only one patient underwent CT after intravenous administration of 100 mL of iodinated contrast material (iohexol, Omnipaque; Nycomed, Princeton, NJ). In the remaining patients, CT scans were obtained without contrast material because of renal insufficiency or to prevent potential complications in transplanted kidneys.
MR images were obtained with a 1.5-T system (Signa; GE Medical Systems) by using an anteroposterior volume neck coil (Medical Advances, Madison, Wis). The imaging protocol consisted of a sagittal T1-weighted conventional spin-echo sequence (repetition time msec/echo time msec = 400–600/11–17) followed by a transverse T2-weighted fast spin-echo sequence (3,500–4,000/85–90 [effective]) and a transverse T1-weighted spin-echo sequence (600/11–17). Images of the neck extended from the level of the cavernous sinus to the upper mediastinum. Then, 0.1 mmol of gadopentetate dimeglumine (Magnevist; Schering, Berlin, Germany) per kilogram of body weight was intravenously administered, and transverse T1-weighted spin-echo images were obtained with imaging parameters similar to those used for the nonenhanced images. Fast spin-echo and contrast-enhanced images were obtained after the application of frequency-selective fat-saturation techniques. For the sagittal T1-weighted localizing sequence, the section thickness was 5 mm with a gap of 1 mm, and the field of view was 30 cm. All other sequences were performed with interleaved images and a section thickness of 5 mm. Other imaging parameters included one or two signals acquired and a 24–26-cm field of view. A 256 x 128 matrix was used for sagittal and contrast-enhanced T1-weighted images, and a 256 x 192 matrix was used for nonenhanced transverse T1- and T2-weighted images.
Images were retrospectively reviewed in consensus by two head and neck radiologists (L.A.L., D.M.Y.), who looked for abnormalities of the Waldeyer ring and cervical lymph nodes. In the patient who underwent both CT and MR imaging, both sets of images were evaluated together. Specifically, images were assessed for the presence of (a) generalized enlargement and/or focal masses of the pharyngeal lymphoid tissue and (b) lymphadenopathy. In addition to size, masses were evaluated for changes in attenuation and signal intensity (relative to that of muscle and cerebrospinal fluid) at CT and MR imaging, respectively, which are suggestive of increased free water concentration. The degree of enhancement was assessed by comparing the attenuation or signal intensity of the masses with that of muscle, the submandibular glands, and the mucosa. Lymph nodes were considered to be abnormal if they were enlarged according to radiologic criteria (diameter > 1.5 cm in the submandibular, submental, and jugulodigastric regions; diameter > 1.0 cm in the remainder of the neck) or if there were an excessive number regardless of their size (multiple groupings of three or more nodes, each at least 5 mm). In addition, the upper mediastinum and thymic region was assessed for focal lesions. Radiologic findings were compared with results of histologic analysis of the specimens.
The medical records of all patients were retrospectively reviewed by one author (R.L.K.) to determine the predisposing disease that led to organ failure, time between transplantation and onset of PTLD, presence of PTLD in other organ systems, initial therapy, and response to therapy.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
|
|
|
|
|
|
|
|
|
The interval between transplantation and the onset of PTLD ranged from 3.5 to 108 months (mean, 30 months; median, 11 months) (Table). All patients initially had PTLD of the neck. Three patients were subsequently found to have PTLD involving other organ systems, including the rectosigmoid and lungs in one patient, the periaortic lymph nodes in another, and the allografted kidney in the third. Patients were treated with combination therapy, including reduction in their immunosuppression regimen (n = 7), surgery (n = 2), chemotherapy (n = 2), and radiation therapy (n = 2) (Table). Outcomes in the five patients with neoplasia included complete remission (n = 1), partial remission (n = 1), and progressive disease (n = 3). At the time of manuscript preparation, two patients had died.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
The system used to classify PTLD is constantly evolving in an attempt to accommodate the wide range and growing number of subtypes of lymphoproliferation. Histopathologic criteria for the diagnosis of PTLD include lymphocytic proliferation with or without destruction of the underlying architecture of the affected tissue (depending on the classification system used). Frizzera et al (12) in 1981, Nalesnik et al (10) in 1989, and Knowles et al (11) in 1995 have developed classifications systems that subdivide PTLD lesions into three distinct categories: hyperplastic, polymorphic, and monomorphic (3). More recently, a T-cell form of PTLD has been reported (13), and this finding represents a rare, late, and aggressive complication of organ transplantation. Unlike its B-cell counterpart, T-cell PTLD is not thought to be associated with EBV infection (13).
The role of EBV infection in the pathogenesis of PTLD has been firmly established (14). EBV infects B lymphocytes, as well as epithelial cells, because the virus attaches to specific receptors located on the membranes of these cells. This receptor specificity may help explain the association of EBV in the development of squamous cell carcinoma of the aerodigestive tract epithelium (6,15).
After the primary infection, latent EBV persists for life. Latent EBV nuclear proteins induce B-cell proliferation. In the immunocompetent patient, this process is kept in check by cytotoxic T lymphocytes, which recognize and lyse EBV-infected cells. In the immunocompromised patient, however, the suppression of this immunosurveillance process may lead to the unregulated proliferation of B cells (16). Tumors are initially polyclonal and gradually progress to monoclonal lesions that harbor malignant transformations (5,16). In the setting of solid organ transplantation, EBV infection may originate from the donor or the recipient, and the infection may represent an acute process or a reactivation process (17,18). Pretransplantation EBV seronegativity is thought to substantially increase the risk for the development of PTLD and, at some institutions, may preclude transplantation (8).
The clinical manifestations of PTLD commonly include a mononucleosis-type syndrome (fever, fatigue, sore throat), regional lymphoid tumors, or disseminated disease (19). Lymphoid tumors may be nodal or extranodal in origin. Extranodal disease is most common and frequently involves the gastrointestinal tract, central nervous system, and/or allografted organ (6). Allograft involvement may clinically mimic organ rejection, and, therefore, biopsy is necessary to prevent therapy that will exacerbate the disease process. Disseminated disease, which is the most severe form of PTLD, manifests as sepsis, generalized adenopathy, and multiorgan failure. Therapeutic interventions are often ineffective.
Because the symptoms of PTLD frequently resemble those of other complications of transplantation, particularly infection and organ rejection, a high index of suspicion for PTLD is crucial to prevent a delay in diagnosis. PTLD should always be considered in the differential diagnosis of organ rejection in the setting of chronic abdominal pain or with the development of cutaneous lesions in a transplant recipient. Oral shedding of EBV and elevation of EBV-positive serum leukocytes may aid in the early diagnosis (7). The work-up of suspected PTLD usually includes cross-sectional imaging of the head, chest, abdomen, and pelvis (20). Routine imaging of the neck has not been part of this protocol. Definitive diagnosis of PTLD is established on the basis of histopathologic analysis of tissue. Specimens are evaluated for histologic characteristics, cellular clonality, EBV infection, and human lymphocyte antigen typing.
The imaging appearance of PTLD has been described in the lungs, gastrointestinal tract, and central nervous system. For intrathoracic PTLD, chest radiography is a simple and accurate surveillance method (21,22). Findings may include pulmonary nodules, airspace consolidation, hilar and mediastinal adenopathy, pleural and pericardial effusions, and thymic enlargement (21,23–25). Although necrosis is uncommon, it has been noted in very large solitary mediastinal nodal masses (21,23).
In intraabdominal PTLD, low-attenuating mass(es) in the liver and spleen may be present at CT. Diffuse infiltration resulting in hepatomegaly or splenomegaly without a discrete mass or masses may also occur (26). Other abdominal findings include thickening or dilatation of the small bowel, mesenteric adenopathy, and disease in the kidney, pancreas, or adrenal glands (27).
Lesions in the central nervous system were more commonly encountered in the era before cyclosporine (25,28). Cerebral lesions typically manifest as peripherally enhancing masses with a central area of low attenuation or with signal intensity similar to that of cerebrospinal fluid, and such lesions must be distinguished from infection (25).
PTLD affecting the neck is reportedly uncommon, and, therefore, the radiologic manifestations have not been well characterized. We have described seven adult transplant recipients in whom the primary manifestation of PTLD began with symptoms in the neck. Six patients had focal masses in the Waldeyer ring, and 50% of these patients had central necrosis and substantial submucosal extension into the parapharyngeal space. These imaging findings differed from those of lymphoma in the immunocompetent patient, which frequently appears as a solid mass that grows exophytically into the airway lumen. It is also noteworthy that in two of the three patients, the necrotic masses were initially misinterpreted as abscesses. Cervical lymphadenopathy was also present in more than 40% (three of seven) of the patients in our study, and two imaging patterns were noted: namely, a large nodal mass and clusters of normal-sized lymph nodes.
There are a few reports of PTLD of the neck described in the pediatric population (27,29,30). Donnelly et al (27) noted that eight of 27 children (all heart transplant recipients) developed PTLD of the neck. Lesions in the Waldeyer ring were present in three patients, and cervical adenopathy was present in five. Similar to our findings, these previously published results showed extensive low attenuation in one of the masses in the Waldeyer ring, as well as in a nodal mass.
It is important to note that microscopic disease may be overlooked at imaging. The presence of histopathologically proved PTLD in nodes that are normal according to imaging criteria emphasizes that the absence of abnormalities at imaging does not rule out PTLD. It further emphasizes that, as our findings suggest, clusters of nodes, although normal in size, should be viewed with suspicion given the appropriate clinical setting.
There are multiple treatment modalities for PTLD, including reduction in immunosuppressive therapy, chemotherapy, radiation therapy, surgical excision, antiviral therapy, and novel immunomodulation techniques. Treatment is determined according to the specific histopathologic finding. Reduction in immunosuppressive medication (or complete cessation in cases of lymphomatous or disseminated disease) is effective and almost always serves as the initial therapeutic option (2,6,19,31). The likelihood of a response is directly related to the clonality of the tumor. Monoclonal tumors, which more closely resemble lymphoma, are less likely to be affected. Reduction in immunosuppressive therapy may be accompanied by organ rejection and requires close clinical follow-up. Failure of tumor regression in the setting of decreased immunosuppression requires other treatment modalities, including chemotherapy and radiation therapy. Efforts in the use of antiviral drugs have shown limited clinical utility in the treatment of PTLD (2,4,20). Immunotherapeutic treatments such as anti–B-lymphocyte monoclonal antibodies, interferon alfa, and cytotoxic T-cell transfusion are currently under investigation (3).
Many factors affect the prognosis in patients with PTLD. Early-onset disease (within 1 year of transplantation) is associated with a better prognosis than is late-onset disease, with mortality rates of 36% and 70%, respectively (32). Patients with early-onset PTLD also are more likely to respond to a reduction in immunosuppressive treatment (32). Heart transplant recipients are reported (27) to have a worse prognosis than those with kidney or liver allografts, perhaps because of the more potent immunosuppressive therapy required after heart transplantation.
In conclusion, patients with PTLD affecting the neck may have focal masses in the Waldeyer ring and lymphadenopathy that manifests as large nodal masses or an excessive number of lymph nodes that are relatively normal in size. Unlike in lymphoma in the immunocompetent patient, necrosis in PTLD lesions may be more common. The presence of a mass or masses in the Waldeyer ring or cervical nodes in patients who have undergone organ transplantation should increase the index of suspicion for the diagnosis of PTLD. Masses with low attenuation on CT scans or signal intensity similar to that of cerebrospinal fluid on MR images should not be mistaken for abscesses, as was initially the case in two of our patients. Because lesions in the Waldeyer ring frequently have a large submucosal component, which makes their identification difficult at physical examination, imaging may be paramount in facilitating detection. Finally, imaging is valuable for monitoring the response to treatment (Fig 4).
|
FOOTNOTES |
|---|
Author contributions: Guarantor of integrity of entire study, L.A.L.; study concepts and design, L.A.L., D.M.Y.; definition of intellectual content, L.A.L., D.M.Y., K.T.M.; literature research, R.L.K., P.K., W.H.; clinical studies, R.L.K.; data acquisition, R.L.K., L.A.L., D.M.Y.; data analysis, L.A.L., D.M.Y., K.T.M., R.L.K.; manuscript preparation, editing, and review, L.A.L., R.L.K., P.K., W.H.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  K. C. Brennan, L. H. Lowe, and G. A. Yeaney Pediatric Central Nervous System Posttransplant Lymphoproliferative Disorder AJNR Am. J. Neuroradiol., August 1, 2005; 26(7): 1695 - 1697. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. R. Gordon, L. A. Loevner, A. I. Sonners, W. E. Bolger, and M. A. Wasik Posttransplantation Lymphoproliferative Disorder of the Paranasal Sinuses Mimicking Invasive Fungal Sinusitis: Case Report AJNR Am. J. Neuroradiol., May 1, 2002; 23(5): 855 - 857. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
