|
|
||||||||
Genitourinary Imaging |
1 From the Departments of Radiology (M.G.A., F.V.C., H.H.) and Surgery (P.N.B.), University of California, 505 Parnassus Ave, San Francisco, CA 94143. From the 1997 RSNA scientific assembly. Received January 15, 1998; revision requested March 23; final revision received August 5; accepted November 6. Address reprint requests to F.V.C.
| Abstract |
|---|
|
|
|---|
MATERIALS AND METHODS: Clinical and radiologic findings were retrospectively reviewed in 24 patients who underwent MR imaging of a renal allograft because ultrasonographic (US) findings were inconclusive or discordant with clinical findings. The final diagnoses were established with histopathologic analysis (n = 16) or clinical and imaging follow-up (n = 8).
RESULTS: MR imaging diagnoses were correct in 16 patients (67%), whereas US diagnoses were correct in six patients (25%) (P < .05). Five cases of allograft involvement by posttransplantation lymphoproliferative disorder (PTLD) were found at histopathologic analysis; at MR imaging, PTLD appeared as hypointense masses on T1- and T2-weighted images with minimal enhancement. In four of the five cases of PTLD, the masses occurred at the renal hilum and encased hilar vessels. Normal morphology was correctly diagnosed with MR imaging in five patients in whom a mass was suspected at US.
CONCLUSION: MR imaging results are often diagnostic in cases of complex abnormalities of renal allografts. Renal allograft involvement by PTLD appears to have a relatively characteristic MR imaging appearance. Normal MR imaging findings in cases of suspected masses at US may obviate biopsy.
Index terms: Grafts, 81.455 Kidney, MR, 81.1214, 81.12143 Kidney, transplantation, 81.455 Kidney, US, 81.1298, 81.12983
| Introduction |
|---|
|
|
|---|
Magnetic resonance (MR) is an excellent modality for renal imaging, although US and computed tomography (CT) remain the primary modalities for cost reasons (7). The principal current applications of renal MR imaging are in patients with renal failure or complex lesions and in patients who have had a clinically important reaction to intravenously administered iodinated contrast medium. The role of MR imaging in renal allograft disease is not clearly established (1), but MR imaging remains an attractive modality when compared with US or CT because of superior contrast resolution, multiplanar capability, lack of operator dependence, and no need for intravenous iodinated contrast medium. Also, MR imaging has the general potential for increased characterization of abnormalities detected with US or CT; this potential for increased characterization has been shown to have a substantial effect on clinical management in hepatic and gynecologic imaging (8,9).
We undertook a study to assess the efficacy of MR imaging in the evaluation of posttransplantation renal allograft abnormalities when US findings are inconclusive or discordant with clinical findings.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Twenty-four of 40 patients who underwent MR imaging of the renal allograft during the study period met all of the inclusion criteria. The total number of renal transplants studied was 27 because one patient had received three renal allografts, two of which were chronically rejected, and another patient had undergone en bloc transplantation of both kidneys from a pediatric donor (10). There were 14 male and 10 female patients with a mean age of 48 years (range, 1562 years). The interval since transplantation was 3 months or less in six patients and more than 3 months in the remainder. Indications for US were impaired renal function (n = 18), fever (n = 2), and trauma (n = 1). No reason was given in three cases.
The US findings and the reasons for requesting MR imaging were as follows:
1. US demonstrated a definite or possible mass in 11 patients. MR imaging was requested for further assessment.
2. US demonstrated a morphologic abnormality suggestive of rejection in four patients. Abnormalities considered suggestive of rejection at our institution include loss of corticomedullary differentiation, pyramidal swelling, increased parenchymal echogenicity, pelviinfundibular wall thickening, and renal swelling. In three patients, MR imaging was requested because of clinical findings suggestive of sepsis. In the fourth patient, MR imaging was performed as part of a separate study of MR imaging in renal allograft rejection.
3. US demonstrated a nonspecific peritransplant fluid collection in three patients. All three patients had declining renal function. MR imaging was requested because a suspected hematoma at US was not resolving on serial studies (n = 1), because it was unclear whether the fluid collection was separate from adjacent ascites (n = 1), and because of coexistent fever (n = 1).
4. US demonstrated possible renal vein thrombosis in two patients. MR imaging was requested because of lack of certainty about the US diagnosis.
5. US demonstrated a raised resistive index without clinically important abnormal morphology in two patients. MR imaging was requested because of declining renal function that was not adequately explained by the US findings.
6. US demonstrated wedge-shaped hypoechoic lesions suggestive of infarction in one patient. MR imaging was requested because of lack of certainty about the US diagnosis.
7. US demonstrated normal morphology in one patient. MR imaging was requested because of declining renal function that was not explained by the US findings.
The mean interval between US and MR imaging was 2.5 days (range, 030 days). Final diagnoses were established with histopathologic correlation (n = 16) or clinical and imaging follow-up (n = 8). Histopathologic specimens were obtained by means of percutaneous biopsy or aspiration (n = 12) or transplant nephrectomy (n = 4). Diagnoses established with histopathologic correlation were rejection (n = 5), posttransplantation lymphoproliferative disorder (PTLD) (n = 5), infection (n = 3), lymphocele (n = 1), toxic reaction to cyclosporine (n = 1), and normal morphology (n = 1). Diagnoses established with clinical and imaging follow-up were normal morphology (n = 5; minimum follow-up of 1 year), renal vein thrombosis (n = 1; confirmed with conventional angiography), angiomyolipoma (n = 1; follow-up of 2 years), and ascites (n = 1; confirmed with serial US subsequent to MR imaging).
Imaging Technique
US was performed with state-of-the-art equipment, 3.55.0-MHz transducers, and a combined gray-scale and color Doppler technique. All US scans were reviewed and reported by attending radiologists with extensive experience in US. MR imaging was performed with a 1.5-T unit (Signa; GE Medical Systems, Milwaukee, Wis) and phased-array surface coils. T1-weighted spin-echo images (repetition time msec/echo time msec = 600700/1117) and T2-weighted conventional (2,000/80100) or fast (3,0005,000/85120) spin-echo images were acquired in all patients. Intravenous gadolinium contrast medium (0.1 mmol/kg) was administered in 23 of the 24 patients. Gadolinium-enhanced T1-weighted spin-echo images were acquired with the same parameters as the nonenhanced T1-weighted images. The MR imaging findings were reported by attending radiologists experienced in MR imaging who had a special interest in genitourinary radiology. Clinical and prior imaging results were available to the radiologists who reported the MR imaging findings.
Analysis of MR Imaging Findings
The efficacy of MR imaging was established by reviewing the original MR imaging reports. The original impression was compared with the established final diagnosis and categorized as correct (if the leading diagnosis in the report was the final diagnosis) or incorrect (if the leading diagnosis in the report was not the final diagnosis). If the original impression was concordant with the final diagnosis but the report did not specifically give the final diagnosis as the leading diagnosis, the original impression was considered incorrect. The same categorization was performed for the original US reports. The performance of US and that of MR imaging were compared by means of the McNemar test with the Yates correction. In cases of PTLD, the signal intensity and enhancement characteristics were determined with consensus review of the images by three radiologists (M.G.A., F.V.C., H.H.) because, in some cases, these features were incompletely described in the original report. However, the leading diagnosis given in the original report was never altered and remained the sole basis for the analysis of efficacy.
| RESULTS |
|---|
|
|
|---|
|
|
|
|
|
|
|
|
Five of 11 patients with a US diagnosis of a mass or a possible mass in the allograft had normal MR images. In these five patients, there was normal parenchymal signal intensity with all sequences and normal uniform enhancement on postcontrast images. All five patients had normal serial follow-up images, and the final diagnosis was normal morphology in all five (Fig 4).
|
|
|
|
|
|
| DISCUSSION |
|---|
|
|
|---|
Immunosuppression causes a hundredfold increase in the prevalence of cancer in transplant recipients compared with the prevalence in the age-matched general population (11). Non-Hodgkin lymphoma and squamous cell carcinoma of the skin and lips are the most common posttransplantation cancers. PTLD is a potentially fatal, lymphomalike lymphoid proliferation that is associated with Epstein-Barr virus and occurs in approximately 2% of solid-organ transplant recipients (12). PTLD was reported after the introduction of cyclosporine as the principal immunosuppressive agent for transplant recipients (13,14). PTLD differs from classic non-Hodgkin lymphoma in that the proliferating lymphocytes are more polymorphic and may be polyclonal rather than monoclonal. Also, PTLD may regress with reduction or cessation of cyclosporine therapy and may not respond to conventional antilymphoma chemotherapy.
Common sites of involvement by posttransplantation lymphoma and PTLD are the brain, lymph nodes, gastrointestinal tract, and lungs (1416). The renal allograft is an inconstantly described site of involvement. No transplant involvement was found in five renal allograft recipients with cyclosporine-induced PTLD (14). Four renal allografts were involved in 15 patients with posttransplantation lymphoma; two of these renal allografts demonstrated centrally hypoattenuating masses adjacent to the kidney at CT and two demonstrated no radiologic abnormalities (15). Three of six renal allograft recipients with posttransplantation abdominal lymphoma demonstrated poorly defined masses infiltrating the allograft at CT (16).
The MR imaging findings in renal allograft involvement by PTLD have not been described, to our knowledge. We found a relatively typical pattern: a hilar mass that was hypointense on T1- and T2-weighted images, contained traversing renal vessels, and demonstrated minimal enhancement. This pattern was seen in four of the five cases. One case demonstrated multiple solid lesions in the allograft parenchyma; the correct diagnosis was suggested by the signal intensity characteristics. All but one of the five cases were correctly diagnosed with MR imaging; the false-negative diagnosis was due to mischaracterization of a hilar mass as a possible hematoma. This case occurred in the 1st year of the study and was considered suggestive of PTLD when reviewed in the context of our subsequent experience.
The predilection of renal allograft PTLD for the renal hilum is noteworthy. The hepatic hilum was the site of hepatic transplant involvement by PTLD in three of three cases (17); it is possible that the perianastomotic region of a solid-organ transplant is somehow predisposed to the development of PTLD. Diagnosis of allograft involvement by PTLD by means of fine-needle aspiration biopsy may be difficult because the predilection of PTLD to encase the hilar vessels increases the risk of hemorrhage and because the cytologic diagnosis is not straightforward (12). Therefore, recognition of the MR imaging features of PTLD may be helpful in the diagnostic evaluation of peritransplant masses.
US has recognized limitations in the detection of small renal lesions (18). These limitations may result in a lowered threshold for reporting possible masses, especially in patients with an increased risk of malignancy, such as transplant recipients. Our results suggest that MR imaging may be helpful in establishing the normality of suspected masses at US and thus obviate biopsy. In our study, all five cases of normal morphology with a false-positive US diagnosis of a mass had unremarkable clinical and serial follow-up imaging findings.
There are three potential criticisms of our study. First, the study population was relatively small despite an 8-year study period. The smallness of the study population likely reflects the adequacy of clinical and US evaluation, supplemented by biopsy as necessary, in the majority of posttransplantation renal allograft abnormalities. In addition, the smallness of the study population may reflect lack of awareness of the potential of MR imaging and financial constraints on high-technology imaging. Second, the study was performed retrospectively and was based on the original MR imaging reports, which were generated by radiologists with knowledge of the clinical results and other radiologic results. Their knowledge of these additional results explains why some of the leading diagnoses in Table 2 may not intuitively appear to follow directly from the described imaging findings. For example, the small intrarenal fluid collections seen at MR imaging in patient 7 were considered likely to be abscesses rather than cysts because the history of fever was available. Basing the study on the original MR imaging reports may have introduced bias but has the advantage of producing results that represent the real-life incremental benefit of MR imaging, and one of our aims was to investigate the clinical role of MR imaging in this patient population. Third, there was a definite selection bias in our study population because only patients with US findings that were inconclusive or discordant with clinical findings were included. Nonetheless, we specifically wished to evaluate patients with complex lesions that were indeterminate at US. We are not suggesting that MR imaging should be used routinely in the evaluation of renal allograft abnormalities.
In conclusion, MR imaging often provides specific diagnostic information in cases of complex posttransplantation abnormalities of renal allografts. Renal allograft involvement by PTLD appears to have a relatively characteristic MR imaging appearance. Normal MR imaging findings in cases of suspected masses at US may obviate biopsy.
| Footnotes |
|---|
Author contributions: Guarantor of integrity of entire study, H.H.; study concepts, H.H.; study design, M.G.A., F.V.C.; definition of intellectual content, H.H., F.V.C.; literature research, M.G.A., F.V.C.; data acquisition and analysis, M.G.A., F.V.C.; statistical analysis, H.H., F.V.C.; manuscript preparation, F.V.C.; manuscript editing, H.H., F.V.C.; manuscript review, P.N.B.
| References |
|---|
|
|
|---|
3-cm) renal masses: detection with CT versus US and pathologic correlation. Radiology 1996; 198:785-788.This article has been cited by other articles:
![]() |
A. Westphalen, B. Yeh, A. Qayyum, A. Hari, and F. V. Coakley Differential Diagnosis of Perinephric Masses on CT and MRI Am. J. Roentgenol., December 1, 2004; 183(6): 1697 - 1702. [Full Text] [PDF] |
||||
![]() |
D. Yang, Q. Ye, D. S. Williams, T. K. Hitchens, and C. Ho Normal and Transplanted Rat Kidneys: Diffusion MR Imaging at 7 T Radiology, June 1, 2004; 231(3): 702 - 709. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. D. Hohenwalter, C. J. Skowlund, S. J. Erickson, S. Hariharan, W. S. Rilling, M. R. Crain, and P. Drescher Renal Transplant Evaluation with MR Angiography and MR Imaging RadioGraphics, November 1, 2001; 21(6): 1505 - 1517. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Lopez-Ben, J. K. Smith, C. E. Kew II, P. J. Kenney, B. A. Julian, and M. L. Robbin Focal Posttransplantation Lymphoproliferative Disorder at the Renal Allograft Hilum Am. J. Roentgenol., November 1, 2000; 175(5): 1417 - 1422. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. J. Pickhardt, M. J. Siegel, R. J. Hayashi, and M. Kelly Posttransplantation Lymphoproliferative Disorder in Children: Clinical, Histopathologic, and Imaging Features Radiology, October 1, 2000; 217(1): 16 - 25. [Abstract] [Full Text] |
||||
![]() |
T. G. Vrachliotis, K. K. Vaswani, E. A. Davies, E. A. Elkhammas, W. F. Bennett, and J. G. Bova CT Findings in Posttransplantation Lymphoproliferative Disorder of Renal Transplants Am. J. Roentgenol., July 1, 2000; 175(1): 183 - 188. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |