Published online before print March 27, 2008, 10.1148/radiol.2472070798
(Radiology 2008;247:451-457.)
© RSNA, 2008
Small (<2-cm) Upper-Tract Urothelial Carcinoma: Evaluation with Gadolinium-enhanced Three-dimensional Spoiled Gradient-Recalled Echo MR Urography1
Naoki Takahashi, MD,
Akira Kawashima, MD, PhD,
James F. Glockner, MD, PhD,
Robert P. Hartman, MD,
Bradley C. Leibovich, MD,
Anja C. S. Brau, PhD,
Philip J. Beatty, PhD, and
Bernard F. King, MD
1 From the Departments of Radiology (N.T., A.K., J.F.G., R.P.H., B.F.K.) and Urology (B.C.L.), Mayo Clinic, 200 First St SW, Rochester, MN 55905; and GE Healthcare, Global Applied Science Lab, Menlo Park, Calif (A.C.S.B., P.J.B.). Received May 7, 2007; revision requested July 10; revision received August 1; accepted August 16; final version accepted September 28.
Address correspondence to N.T.
 |
ABSTRACT
|
|---|
Purpose: To retrospectively evaluate the detection of small (<2-cm) urothelial tumors by using gadolinium-enhanced three-dimensional (3D) spoiled gradient-recalled echo (GRE) magnetic resonance (MR) urography.
Materials and Methods: This HIPAA-compliant study received institutional review board approval. All patients included had previously consented to the use of their medical records for research purposes. Eleven of 110 patients (10 men, one woman; mean age, 73.5 years) who underwent MR urography were ultimately identified to have 23 upper-tract urothelial carcinomas smaller than 2 cm or carcinoma in situ. Breath-hold coronal T2-weighted single-shot fast spin-echo and breath-hold coronal 3D T1-weighted spoiled GRE images with fat suppression during nephrographic and excretory phases after intravenous injection of gadolinium-based contrast material were obtained in all patients with a 1.5-T imager. Two radiologists reviewed the MR images in consensus for the presence of tumors. Lesion detectability was compared between each sequence by using the McNemar test.
Results: Of 23 tumors, 17 (74%) were detected by using at least one sequence, eight (35%) were detected with T2-weighted imaging, 15 (65%) were detected on nephrographic phase images, and 15 (65%) were detected on excretory phase images. Two lesions each were detected only on either nephrographic or excretory phase images. Detectability was significantly higher on nephrographic and excretory phase images compared with T2-weighted images (P < .05).
Conclusion: Gadolinium-enhanced 3D spoiled GRE MR urography helped detect 74% of small urothelial carcinomas. Nephrographic and excretory phase images are essential for helping detect small urothelial carcinomas.
© RSNA, 2008
|
INTRODUCTION
|
|---|
Urothelial carcinoma of the upper tract accounts for 5% of all urothelial carcinomas and 10% of all renal tumors (1,2). Patients with a history of urothelial carcinoma are at high risk of developing synchronous or metachronous urothelial carcinoma (1). These patients and those with hematuria often require evaluation of the upper and lower tracts. Excretory urography or computed tomographic (CT) urography has been used to evaluate high-risk patients (3,4). However, when the patient has a contraindication to iodinated contrast material, such as renal insufficiency or allergy to iodinated contrast material, retrograde pyelography is often used to image the upper urinary tract. Recent developments in the endoscopic management of upper-tract urothelial carcinoma have further emphasized the need for a noninvasive alternative imaging modality. The most common indication for endoscopic treatment of the upper-tract urothelial carcinoma includes small (<2 cm) low-grade tumor in patients with renal insufficiency (5,6). In addition, patients who have undergone endoscopic treatment for upper-tract urothelial carcinoma have a greater than 30% risk of recurrence; thus, vigorous surveillance is required (5–7).
MR urography is an emerging technique used to help assess the upper urinary tract. MR urography can be performed by using heavily T2-weighted hydrographic sequences without contrast material (8–12) or T1 spoiled gradient-recalled echo (GRE) sequences during the excretory phase after administration of gadolinium-based contrast material (11–13). The clinical usefulness of these sequences has been shown in helping detect upper urinary tract abnormalities, particularly when patients have urinary tract obstruction (11,13). With the recent introduction of a parallel imaging technique and the development of multichannel phased-array coils, breath-hold three-dimensional (3D) T1 spoiled GRE following bolus injection of gadolinium-based contrast agent have been incorporated in many abdominal MR imaging protocols to improve spatial resolution (14).
The role of dynamic gadolinium-enhanced spoiled GRE sequences performed during the nephrographic and excretory phases in helping detect small (<2 cm) urothelial carcinomas of the upper tract has not been assessed previously. Thus, the purpose of our study was to retrospectively evaluate the detection of small (<2 cm) urothelial tumors by using gadolinium-enhanced 3D spoiled GRE MR urography.
|
MATERIALS AND METHODS
|
|---|
Two of the authors (A.C.S.B. and P.J.B.) are employed by GE Healthcare. In this study, we used an MR imager, a prototype MR sequence, and gadolinium-based contrast material manufactured by GE Healthcare (Milwaukee, Wis). The other authors had control over the data and information submitted for publication.
Patients
Our Health Insurance Portability and Accountability Act–compliant study was approved by our institutional review board. All patients had previously consented to the use of their medical records for research purposes. One hundred ten patients underwent MR urography at our institution between May 2005 and January 2007. None of the patients developed nephrogenic systemic fibrosis. Thirteen (11 men, two women; mean age, 71.7 years) of 110 patients were ultimately identified as having 27 upper-tract urothelial carcinomas. Seven patients had more than one lesion (three patients had two lesions each [lesions in one were metachronous], two patients had three lesions each, one patient had four lesions, and one patient had five lesions [lesions were metachronous]). Metachronous lesions were identified at different locations 8 and 12 months after ureteroscopic fulguration of the first lesions.
Four lesions larger than 2 cm (largest diameter) were excluded from further analysis. The size of the tumor was measured on a pathologic specimen when available. If not available, size measured at MR was used. All the lesions that were not detected at MR were considered as less than 2 cm on the basis of ureteroscopic findings and were included in the study. One carcinoma in situ lesion involving 4 cm of ureter at pathologic examination was included in the study. Thus, our final study group consisted of 11 patients (10 men, one woman; mean age, 73.5 years) with 23 lesions. Diagnosis of upper-tract urothelial carcinoma was made by using pathologic examination (n = 12), ureteroscopic findings (n = 10), or a nephrostogram with a typical overhanging border (n = 1). The interval between MR urography and the final diagnosis ranged between 0 and 90 days, with a median of 4 days.
Indications for MR urography were prior history of urothelial carcinoma (n = 9), other imaging studies that showed findings suspicious for upper-tract urothelial tumor (hydronephrosis [n = 4] or soft tissue in the urinary tract [n = 2]) or abnormal urinary cytologic findings (n = 1). Five patients had both prior history of urothelial carcinoma and other imaging studies that showed findings suspicious for upper-tract urothelial tumor. All patients except one had a contraindication to CT urography (renal insufficiency [n = 9] or prior allergy to iodinated contrast material [n = 1]). The other patient had undergone renal transplantation and the requesting physician preferred not to use iodinated contrast material. Seven patients had undergone urinary tract surgery (cystectomy for urothelial carcinoma [n = 3], cystectomy and nephroureterectomy for urothelial carcinoma [n = 2], nephrectomy for renal cell carcinoma [n = 1], and renal transplantation [n = 1]).
MR Urography Technique
MR urography was performed with a 1.5-T imager (Signa Vision; GE Healthcare, Milwaukee, Wis) by using an eight-channel torso phased-array coil. Immediately before and after intravenous administration of 10–20 mg of furosemide (Lasix; Abbott Laboratories, North Chicago, Ill), breath-hold coronal thin-section and thick-slab T2-weighted single-shot fast spin-echo MR images were obtained. Parameters for the thin-section T2-weighted sequence were repetition time msec/echo time msec, 
/90; matrix, 256 x 256; section thickness, 4 mm; and field of view, 40 cm. Parameters for the thick-slab T2-weighted sequence were /500; matrix, 384 x 256; section thickness, 35 mm; and field of view, 40 cm. Five minutes after administration of furosemide, breath-hold coronal dynamic 3D T1 spoiled GRE images with fat suppression were obtained in the nephrographic phase after intravenous injection of 0.1 mmol/kg of gadodiamide (Omniscan; GE Healthcare, Chalfont, St Giles, England) by using a power injector (Spectris; Medrad, Indianola, Pa). The scan delays were 40 and 90 seconds (n = 9) or three consecutive scans after arrival of contrast material to the aorta, as predetermined by the test injection of 2 mL of contrast material (n = 4). Excretory phase images were then obtained by using coronal 3D T1 spoiled GRE with fat suppression, with scan delays of 5 and 10 minutes. Additional delayed excretory phase images were obtained in two examinations, as deemed necessary by the radiologists who monitored the examination. Parameters for 3D T1 spoiled GRE were 5.2/1.8; flip angle, 15°; matrix, 256 x 224; section thickness, 3–4 mm; and field of view, 42 cm. A prototype automatically calibrated reconstruction parallel imaging method was implemented with fat-suppressed 3D T1 spoiled GRE sequence to achieve two times the data acquisition acceleration with minimal motion or aliasing artifact. Acquisition time was less than 22 seconds in all patients. In two patients, 3D T1 spoiled GRE images were obtained without fat suppression owing to technical error (n = 1) and presence of metallic hip prosthesis adjacent to the transplanted kidney (n = 1). In two patients, dynamic 3D T1 spoiled GRE sequence in the nephrographic phase was performed in the transverse plane because the patients had known liver or renal parenchymal masses that were thought to be better evaluated in the transverse plane. Mechanical abdominal compression was not used.
Image Interpretation
Since two patients were identified as having urothelial carcinomas at two different times, findings at each MR urographic examination performed immediately before the diagnosis of urothelial carcinomas were evaluated. Thus, 13 MR urographic examinations in 11 patients were evaluated. Two radiologists (N.T. and A.K., with 6 and 16 years experience with abdominal MR, respectively) reviewed the images in consensus on a workstation (Advantage Windows, version 4.2; GE Healthcare, Milwaukee, Wis). The reviewers were blinded to the location of the tumor but not to the fact that all patients had at least one upper-tract tumor. Reviewers evaluated the presence of the tumor in each sequence (T2-weighted, gadolinium-enhanced 3D T1 spoiled GRE performed in the nephrographic and excretory phases) and identified the sequence in which the lesions were most conspicuous. Presence of urinary obstruction with abrupt cutoff alone was not considered as a sign of tumor presence unless a filling defect or soft tissue was identified. All three sequences were evaluated simultaneously. The lesion was categorized as obstructive or nonobstructive when the lesion was detected by using at least one sequence. The maximal diameter of the lesion was also measured. The signal of the tumor, the urine adjacent to the tumor, and the standard deviation of the air were measured by placing a region of interest on each image (N.T.). Contrast-to-noise ratio (CNR), measured as the signal of the urine subtracted from the signal of the tumor divided by the standard deviation of the signal outside the patient, was calculated. CNR was considered as zero when the lesion was not detected. CNR was not calculated for the carcinoma in situ lesion because it manifested as diffuse thickening of the ureteral wall and it was difficult to place a region of interest on the lesion.
Pathologic and Urinary Cytologic Findings
Tumor size, stage, and grade at pathologic examination were recorded, when available. Tumor size at pathologic examination was correlated with the size measured at MR. Urinary cytologic findings were recorded, when available, within 5 days of MR urographic examination.
Statistical Analysis
Lesion detectability was compared between sequences by using the pairwise McNemar test. Friedman two-way analysis of variance with the rank test was used to compare CNRs of the three sequences followed by pairwise comparisons by using the Wilcoxon signed rank test. Carcinoma in situ lesions were excluded from this analysis. Lesion detectability and CNRs with nephrographic phase images were also compared between examinations that used a test injection and those that did not by using the Fisher exact test and Wilcoxon rank sum test. A P value of less than .05 was considered to indicate a significant difference; a Bonferroni correction was used when multiple comparisons were performed. Statistical analyses were performed by using software (JMP, version 6.0, 2005; SAS Institute, Cary, NC).
|
RESULTS
|
|---|
Lesions Detected
Of 23 small upper-tract urothelial carcinomas in 11 patients, 17 (74%) lesions were detected by using at least one sequence (intrarenal collecting system [n = 7] and ureter [n = 10]). At least one lesion was detected in 10 (77%) of 13 MR urographic examinations (Table 1). Six lesions were not detected with any of the sequences (intrarenal collecting system [n = 4] and ureter [n = 2]). Of 23 lesions, eight (35%) were detected on T2-weighted images, 15 (65%) on nephrographic phase images, and 15 (65%) on excretory phase images. Lesions appeared as low-signal-intensity filling defects or soft tissue within or along the course of the upper tract in the T2-weighted images and excretory phase images, whereas they appeared as focal areas of abnormal enhancement in the nephrographic phase images (Figs 1, 2). The carcinoma in situ lesion appeared as diffuse thickening and enhancement of the ureteral wall and was detected only on the nephrographic phase images. Two lesions each were detected only on nephrographic or excretory phase images; all four of these lesions were located in the ureter (3-, 4-, and 5-mm papillary lesions and one 41-mm carcinoma in situ). None of the lesions were detected only on T2-weighted images. Detectability was significantly higher on nephrographic phase and excretory phase images, compared with T2-weighted images (each P = .008). Eight lesions were most conspicuous on excretory phase images; seven lesions, on nephrographic phase images; and two lesions, on T2-weighted images. Mean CNRs of the lesions were 12.1 ± 25.9 (standard deviation) (range, 0–105.8) for T2-weighted images, 19.0 ± 19.9 (range, 0–70.2) for nephrographic phase images, and 22.0 ± 18.4 (range, 0–52.6) for excretory phase images. No significant difference in the CNR was seen between the sequences (P = .2–.6).
View larger version (150K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1a: Urothelial carcinoma (11 x 7 mm) in left distal ureter of 79-year-old man. Patient had undergone cystectomy for bladder carcinoma. (a) Coronal T2-weighted single-shot fast spin-echo MR image (/90) shows lesion (arrow) as low-signal-intensity filling defect compared with surrounding urine. (b) Coronal gadolinium-enhanced 3D T1 spoiled GRE nephrographic phase MR image (5.2/1.8; flip angle, 15°) shows lesion (arrow) as enhanced mass. (c) Coronal excretory phase MR image (5.2/1.8; flip angle, 15°) shows lesion (arrow) as faint filling defect. Lesion rated most conspicuous on b.
|
|
View larger version (140K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1b: Urothelial carcinoma (11 x 7 mm) in left distal ureter of 79-year-old man. Patient had undergone cystectomy for bladder carcinoma. (a) Coronal T2-weighted single-shot fast spin-echo MR image (/90) shows lesion (arrow) as low-signal-intensity filling defect compared with surrounding urine. (b) Coronal gadolinium-enhanced 3D T1 spoiled GRE nephrographic phase MR image (5.2/1.8; flip angle, 15°) shows lesion (arrow) as enhanced mass. (c) Coronal excretory phase MR image (5.2/1.8; flip angle, 15°) shows lesion (arrow) as faint filling defect. Lesion rated most conspicuous on b.
|
|
View larger version (141K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1c: Urothelial carcinoma (11 x 7 mm) in left distal ureter of 79-year-old man. Patient had undergone cystectomy for bladder carcinoma. (a) Coronal T2-weighted single-shot fast spin-echo MR image (/90) shows lesion (arrow) as low-signal-intensity filling defect compared with surrounding urine. (b) Coronal gadolinium-enhanced 3D T1 spoiled GRE nephrographic phase MR image (5.2/1.8; flip angle, 15°) shows lesion (arrow) as enhanced mass. (c) Coronal excretory phase MR image (5.2/1.8; flip angle, 15°) shows lesion (arrow) as faint filling defect. Lesion rated most conspicuous on b.
|
|
View larger version (111K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2a: Urothelial carcinoma (10 x 10 mm) in left renal pelvis of 60-year-old man. Patient had undergone cystectomy and right nephroureterectomy for urothelial carcinoma of bladder and right kidney. (a) Coronal T2-weighted single-shot fast spin-echo MR image (/90) shows lesion (arrow) as low-signal-intensity filling defect surrounded by high-signal-intensity urine. (b) Coronal gadolinium-enhanced 3D T1 spoiled GRE nephrographic phase MR image (5.2/1.8; flip angle, 15°) shows lesion (arrow) as enhancing mass. (c) Coronal excretory phase (5.2/1.8; flip angle, 15°) MR image shows lesion (arrow) as discrete filling defect. Lesion rated most conspicuous in c.
|
|
View larger version (87K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2b: Urothelial carcinoma (10 x 10 mm) in left renal pelvis of 60-year-old man. Patient had undergone cystectomy and right nephroureterectomy for urothelial carcinoma of bladder and right kidney. (a) Coronal T2-weighted single-shot fast spin-echo MR image (/90) shows lesion (arrow) as low-signal-intensity filling defect surrounded by high-signal-intensity urine. (b) Coronal gadolinium-enhanced 3D T1 spoiled GRE nephrographic phase MR image (5.2/1.8; flip angle, 15°) shows lesion (arrow) as enhancing mass. (c) Coronal excretory phase (5.2/1.8; flip angle, 15°) MR image shows lesion (arrow) as discrete filling defect. Lesion rated most conspicuous in c.
|
|
View larger version (80K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2c: Urothelial carcinoma (10 x 10 mm) in left renal pelvis of 60-year-old man. Patient had undergone cystectomy and right nephroureterectomy for urothelial carcinoma of bladder and right kidney. (a) Coronal T2-weighted single-shot fast spin-echo MR image (/90) shows lesion (arrow) as low-signal-intensity filling defect surrounded by high-signal-intensity urine. (b) Coronal gadolinium-enhanced 3D T1 spoiled GRE nephrographic phase MR image (5.2/1.8; flip angle, 15°) shows lesion (arrow) as enhancing mass. (c) Coronal excretory phase (5.2/1.8; flip angle, 15°) MR image shows lesion (arrow) as discrete filling defect. Lesion rated most conspicuous in c.
|
|
Lesion Findings
Of 17 detected lesions, four were obstructive and 13 were nonobstructive. The mean maximal diameter of the lesions detected at MR was 9 mm (range, 3–17 mm), excluding one carcinoma in situ lesion that measured 41 mm in length. Five lesions were surgically removed and maximal diameters of the tumor at pathologic examination were measured. The differences in the tumor size measurement between pathologic examination and MR images ranged from 0 to 2 mm. All six lesions that were not detected at MR underwent ureteroscopic treatment; thus, none of them had size measurement. None of these six were detected at MR, even with knowledge of ureteroscopic findings and tumor location.
Lesion Detection and Test Injection
Of 23 lesions, three of nine lesions were detected on nephrographic phase images obtained after test injection, whereas 12 of the remaining 14 lesions were detected on nephrographic phase images obtained without using a test injection (Table 2). The difference in detectability was significant (P = .02). Mean CNRs of the lesions seen on nephrographic phase images were 6.4 ± 9.1 (range, 0–21.6) with a test injection and 27.8 ± 20.7 (range, 0–70.2) without. The difference was significant (P = .007).
Pathologic Findings
Pathologic findings were available for 12 lesions. Tumor stage was classified as Ta in six lesions, Tis in one, T1 in three, and T2 in one. Tumor stage was not assessed in one. Tumor grade was rated as grade 1 in two lesions, grade 2 in six, and grade 3 in four. Urine cytologic analysis was available within 5 days of MR urographic examinations in 11 of 13 cases. Urine cytologic findings showed urothelial carcinoma (n = 1), suspicious cells (n = 4), and no suspicious cells (n = 6).
|
DISCUSSION
|
|---|
Our study showed that MR urography can help detect 17 (74%) of 23 small (<2 cm) upper-tract urothelial carcinomas. The two most useful sequences were for nephrographic phase and excretory phase images after administration of gadolinium-based contrast material. T2-weighted images were occasionally helpful.
When a tumor is obstructive, previous studies have shown that MR urography can help detect upper-tract urothelial carcinoma with an accuracy of 88% or higher by using excretory phase MR urography (15) or T2-weighted MR urography (12,16). In one study, the size of urothelial carcinomas diagnosed by using excretory phase MR urography ranged from 16 to 40 mm (13). In our study, the size of the focal lesions ranged from 3 to 17 mm and only four of 24 lesions were obstructive.
Researchers in previous studies emphasized the role of T2-weighted and excretory phase images in helping detect abnormalities in the upper tract. The major shortcoming of T2-weighted and excretory phase images is the lack of specificity. Urinary calculi and blood clots both appear as filling defect on these images and are often indistinguishable from urothelial carcinoma. Moreover, flow-related artifact on T2-weighted images and artifact from concentrated gadolinium-based contrast material on excretory phase images may appear as a filling defect. Presence of enhancement (15) and extension of the lesion outside the urinary tract are specific signs of neoplasm, and these are probably best evaluated on the nephrographic phase images. Obuchi et al (17) showed that contrast-enhanced T1-weighted images obtained at 2–3 minutes demonstrated extension of tumor outside the ureter with high accuracy. Our study showed the importance of the nephrographic phase images in helping detect small urothelial carcinomas. Fritz et al (4) emphasized the importance of images obtained during either the corticomedullary or the nephrographic phase at CT urography in tumor detection and staging.
In our study, a test injection of 2 mL was used to assess the arrival time of contrast material in four patients. Tumor detectability and CNR in nephrographic phase images in patients who underwent test injection were significantly lower compared with those in patients who did not. Previous studies have shown that 2 mL of contrast material is sufficient to opacify the urinary tract (18–20). Increased signal intensity levels in the collecting system or ureter from the small dose of contrast material probably reduced the detectability, CNR, and conspicuity of the lesions on the nephrographic phase images. Currently, it is our routine to obtain nephrographic phase images without the use of a test injection.
Several techniques have been used to avoid T2* effect of concentrated gadolinium-based contrast material during the excretory phase of MR urography, which included the use of intravenous saline hydration or furosemide administration and low-dose gadolinium (18–22). In our study, 10–20 mg of furosemide was given intravenously, and a standard dose of gadolinium (0.1 mmol/kg) was used with 20–40 mL of saline flush. This method resulted in adequate dilution of gadolinium without T2* effect at 10-minute delay from the injection of contrast material in most cases. In addition, the furosemide increases the distension of the collecting systems and ureters for better visualization of pathologic process (22). Use of low-dose gadolinium may allow opacification of the urinary tract for excretory phase of MR urography (18–20); however, it is unlikely to yield diagnostic nephrographic phase images.
The detectability of each imaging sequence was assessed simultaneously with two other imaging sequences. Therefore, the detectability may be higher than that of each sequence evaluated separately. On the nephrographic phase images, a normal urinary tract wall only minimally enhances with contrast material. Thus, it is often difficult to follow the entire course of the urinary tract without excretory phase images. We found that simultaneous evaluation of nephrographic and excretory phase images side by side was helpful in identifying abnormal enhanced lesions.
The prevalence of upper-tract urothelial tumor was 11.8% (13 of 110) in our study population, compared with 4.9% (18 of 370) in a previous report conducted with CT urography (3). Multifocal lesions were seen in 54% (seven of 13) of patients in our study, compared with 39% (seven of 18) in the Caoili et al study (3). The higher prevalence in our study may be explained in part by the selection bias of including higher-risk patients referred for MR urography.
The detectability of urothelial tumor by using CT urography was 89%–100% by using retrospective review of patients with known tumors (3,4), which had a similar study design to ours. The detectability of MR urography in our study was 74% for urothelial tumors less than 2 cm in size and 78% (21 of 27) if all tumors were included. It is difficult to directly compare the detectability of MR urography from our study with that of CT urography from previous studies. However, CT has a better spatial resolution and is less vulnerable to respiration motion artifact. The spatial resolution of CT urography is approximately 0.8 x 0.8 x 2.5 mm (3), whereas that of 3D T1 spoiled GRE was approximately 1.6 x 1.6 x 3.0–4.0 mm in our study. The inherent higher tissue contrast levels seen at MR is not advantageous in excretory phase MR urography because the contrast material in the urinary tract creates high levels of contrast in both CT urography and MR urography. Therefore, the use of MR urography to help detect upper-tract urothelial tumors should be performed only when CT urography is contraindicated.
A limitation of our study is the retrospective nature. Not all lesions had histologic diagnosis, and ureteroscopic or typical nephrostographic findings were used as proof of a tumor in 11 lesions. Follow-up of patients who were not ultimately identified as having upper-tract urothelial carcinomas was incomplete; therefore, the detectability of MR urography for small urothelial tumor could be overestimated in this study. Lesion size measured at MR was used as a definition of small (<2 cm) urothelial tumor in most lesions.
Because of the association between gadolinium-based contrast material exposure and nephrogenic systemic fibrosis in patients with renal insufficiency (23), the risks and benefits of MR urographic examination should be carefully assessed for each patient.
In conclusion, gadolinium-enhanced 3D spoiled GRE MR urography helped detect 74% of small (<2 cm) urothelial carcinomas. Nephrographic phase and excretory phase images are essential for helping detect small urothelial carcinomas. Gadolinium-enhanced 3D spoiled GRE MR urography is a promising technique that may be used for the initial evaluation of patients at high risk for developing upper-tract urothelial carcinoma when CT urography or excretory urography is contraindicated, but it needs validation with a larger study.
|
ADVANCES IN KNOWLEDGE
|
|---|
- Gadolinium-enhanced 3D spoiled GRE MR urography allowed detection of 74% of small (<2 cm) urothelial carcinomas.
- Nephrographic and excretory phase images are essential for helping detect small urothelial carcinomas.
|
IMPLICATION FOR PATIENT CARE
|
|---|
- Gadolinium-enhanced 3D spoiled GRE MR urography is a promising technique that may be used for the initial evaluation of patients at high risk for developing upper-tract urothelial carcinoma when CT urography or excretory urography is contraindicated.
|
ACKNOWLEDGMENTS
|
|---|
The authors thank David W. Stanley, BS, RT(R)(MR), for providing a prototype MR sequence.
|
FOOTNOTES
|
|---|
Abbreviations: CNR = contrast-to-noise ratio • GRE = gradient-recalled echo • 3D = three-dimensional
See Materials and Methods for pertinent disclosures.
Author contributions: Guarantor of integrity of entire study, N.T.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; approval of final version of submitted manuscript, all authors; literature research, N.T.; clinical studies, all authors; statistical analysis, N.T.; and manuscript editing, all authors
|
References
|
|---|
- Hall MC, Womack S, Sagalowsky AI, Carmody T, Erickstad MD, Roehrborn CG. Prognostic factors, recurrence, and survival in transitional cell carcinoma of the upper urinary tract: a 30-year experience in 252 patients. Urology 1998;52:594–601.[CrossRef][Medline]
- Kirkali Z, Tuzel E. Transitional cell carcinoma of the ureter and renal pelvis. Crit Rev Oncol Hematol 2003;47:155–169.[Medline]
- Caoili EM, Cohan RH, Inampudi P, et al. MDCT urography of upper tract urothelial neoplasms. AJR Am J Roentgenol 2005;184:1873–1881.[Abstract/Free Full Text]
- Fritz GA, Schoellnast H, Deutschmann HA, Quehenberger F, Tillich M. Multiphasic multidetector-row CT (MDCT) in detection and staging of transitional cell carcinomas of the upper urinary tract. Eur Radiol 2006;16:1244–1252.[CrossRef][Medline]
- Elliott DS, Segura JW, Lightner D, Patterson DE, Blute ML. Is nephroureterectomy necessary in all cases of upper tract transitional cell carcinoma? long-term results of conservative endourologic management of upper tract transitional cell carcinoma in individuals with a normal contralateral kidney. Urology 2001;58:174–178.[CrossRef][Medline]
- Soderdahl DW, Fabrizio MD, Rahman NU, Jarrett TW, Bagley DH. Endoscopic treatment of upper tract transitional cell carcinoma. Urol Oncol 2005;23:114–122.[Medline]
- Elliott DS, Blute ML, Patterson DE, Bergstralh EJ, Segura JW. Long-term follow-up of endoscopically treated upper urinary tract transitional cell carcinoma. Urology 1996;47:819–825.[CrossRef][Medline]
- Aerts P, Van Hoe L, Bosmans H, Oyen R, Marchal G, Baert AL. Breath-hold MR urography using the HASTE technique. AJR Am J Roentgenol 1996;166:543–545.[Free Full Text]
- Regan F, Bohlman ME, Khazan R, Rodriguez R, Schultze-Haakh H. MR urography using HASTE imaging in the assessment of ureteric obstruction. AJR Am J Roentgenol 1996;167:1115–1120.[Abstract/Free Full Text]
- Tang Y, Yamashita Y, Namimoto T, et al. The value of MR urography that uses HASTE sequences to reveal urinary tract disorders. AJR Am J Roentgenol 1996;167:1497–1502.[Abstract/Free Full Text]
- Sudah M, Vanninen R, Partanen K, Heino A, Vainio P, Ala-Opas M. MR urography in evaluation of acute flank pain: T2-weighted sequences and gadolinium-enhanced three-dimensional FLASH compared with urography—fast low-angle shot. AJR Am J Roentgenol 2001;176:105–112.[Abstract/Free Full Text]
- Chahal R, Taylor K, Eardley I, Lloyd SN, Spencer JA. Patients at high risk for upper tract urothelial cancer: evaluation of hydronephrosis using high resolution magnetic resonance urography. J Urol 2005;174:478–482.[CrossRef][Medline]
- Nolte-Ernsting CC, Bucker A, Adam GB, et al. Gadolinium-enhanced excretory MR urography after low-dose diuretic injection: comparison with conventional excretory urography. Radiology 1998;209:147–157.[Abstract/Free Full Text]
- Rofsky NM, Lee VS, Laub G, et al. Abdominal MR imaging with a volumetric interpolated breath-hold examination. Radiology 1999;212:876–884.[Abstract/Free Full Text]
- Jung P, Brauers A, Nolte-Ernsting CA, Jakse G, Gunther RW. Magnetic resonance urography enhanced by gadolinium and diuretics: a comparison with conventional urography in diagnosing the cause of ureteric obstruction. BJU Int 2000;86:960–965.[CrossRef][Medline]
- Shokeir AA, El-Diasty T, Eassa W, et al. Diagnosis of noncalcareous hydronephrosis: role of magnetic resonance urography and noncontrast computed tomography. Urology 2004;63:225–229.[CrossRef][Medline]
- Obuchi M, Ishigami K, Takahashi K, et al. Gadolinium-enhanced fat-suppressed T1-weighted imaging for staging ureteral carcinoma: correlation with histopathology. AJR Am J Roentgenol 2007;188:W256–W261.[Abstract/Free Full Text]
- Obuchi M, Takahara T, Takahashi M, et al. Breath-hold 3D MR urography with 2 mL Gd-DTPA injection [in Japanese]. Nippon Igaku Hoshasen Gakkai Zasshi 1998;58:163–165.[Medline]
- Szopinski K, Szopinska M, Borowka A, Jakubowski W. Magnetic resonance urography: initial experience of a low-dose Gd-DTPA-enhanced technique. Eur Radiol 2000;10:1158–1164.[CrossRef][Medline]
- Hughes J, Jan W, Goodie J, Lund R, Rankin S. MR urography: evaluation of different techniques in non-dilated tracts. Clin Radiol 2002;57:989–994.[CrossRef][Medline]
- Farres MT, Gattegno B, Ronco P, Flahault A, Paula-Souza R, Bigot JM. Nonnephrotoxic, dynamic, contrast enhanced magnetic resonance urography: use in nephrology and urology. J Urol 2000;163:1191–1196.[CrossRef][Medline]
- Ergen FB, Hussain HK, Carlos RC, et al. 3D excretory MR urography: improved image quality with intravenous saline and diuretic administration. J Magn Reson Imaging 2007;25:783–789.[CrossRef][Medline]
- Kuo PH, Kanal E, Abu-Alfa AK, Cowper SE. Gadolinium-based MR contrast agents and nephrogenic systemic fibrosis. Radiology 2007;242:647–649.[Free Full Text]
TOP
ABSTRACT
INTRODUCTION
