BJR
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
British Journal of Radiology (2003) 76, 696-703
© 2003 British Institute of Radiology
doi: 10.1259/bjr/33169417
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Walter, C
Right arrow Articles by Landwehr, P
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Walter, C
Right arrow Articles by Landwehr, P

Full Paper

Imaging of renal lesions: evaluation of fast MRI and helical CT

C Walter, MD M Kruessell, MD A Gindele, MD H G Brochhagen, MD A Gossmann, MD and P Landwehr, MD

Department of Diagnostic Radiology, University Hospital of Cologne, Joseph-Stelzmann-Str. 9, D-50931 Köln, Germany

Correspondence: Dr med Christof Walter, Department of Diagnostic Radiology, Krankenhaus der Barmherzigen Brüder, Nordallee 1, D-54292 Trier, Germany


   Abstract
Top
 Abstract
Introduction
 Patients and methods
 Results
 Discussion
 References
 
The purpose of this study is to compare triphasic helical CT and fast MRI with respect to detection, characterization and staging of suspected renal masses. To achieve this triphasic helical CT (plain, corticonephrographic and tubulonephrographic phase) and MRI with fast T1 weighted and T2 weighted sequences were performed in 29 patients with a suspected renal lesion. Image quality, lesion characterization and lesion extent were assessed for both methods in all patients. The acquisition phase for CT and the image sequence for MRI offering the best image quality and best diagnostic information regarding renal parenchyma, renal vessels, detection of enlarged lymph nodes, and other abdominal organs were determined. Histologically confirmed renal cell carcinomas (n=18) were staged based on the Robson classification. Quantitative data were obtained from operator-defined regions of interest (ROIs) in all acquisition phases (CT) and all image sequences (MRI). For most criteria the rating of image quality for helical CT was generally higher as compared with fast MRI. CT and MRI detected all 24 histologically proven masses, while no false positive solid tumour was diagnosed with both imaging modalities. All three acquisition phases in CT and all applied image sequences in MRI were regarded as necessary in order to gain important diagnostic information. Altogether, 12 of 18 renal cell carcinomas (67%) were correctly staged by CT and MRI. Helical CT and fast MRI allow the correct detection and characterization of suspicious renal lesions. Both imaging modalities can be recommended for clinical routine application. Although the correct histological staging of renal cancer remains difficult for both imaging methods, both are excellent in providing the critical staging information needed before surgery. Helical CT offers a significantly shorter acquisition time to cover the entire abdomen.


   Introduction
Top
 Abstract
 Introduction
Patients and methods
 Results
 Discussion
 References
 
Technical advances in CT – and particularly in the helical scan mode [1] – and MRI with fast sequences [2] have improved the detection and characterization of renal lesions due to better image quality and spatial resolution [3, 4]. Different studies have independently explored the role of helical CT and fast MRI in the examination of renal lesions [5–9]. In the clinical routine, helical CT is commonly used to evaluate suspicious findings detected with ultrasound or excretory urography, or to exclude a malignancy. Previous reports have proposed that MRI may be equivalent to CT in the detection and characterization of cystic and solid renal masses [10–12]. However, these studies only compared conventional CT without helical scan mode with new MRI techniques. To the best of our knowledge, helical CT and fast MRI have not yet been compared in renal imaging. The purpose of this study was to compare both imaging methods with respect to their ability to detect and characterize renal masses and to evaluate and recommend an appropriate imaging strategy for CT and MRI.


   Patients and methods
Top
 Abstract
 Introduction
 Patients and methods
Results
 Discussion
 References
 
Patients
41 consecutive patients with a suspected renal lesion were referred to our department for helical CT between September 1998 and March 1999. Eight patients could not be included in the study because only a monophasic or biphasic CT scan was available. One patient was excluded from the study because histology of a renal lesion could not be obtained, and three further patients because MRI could not be performed due to contraindications for an MRI examination. Thus 29 patients (16 men, 13 women, ranging from 32 years to 85 years of age, mean: 63 years) participated in this study. All patients gave their informed consent for the study protocol, which had previously been approved by the institutional ethics commission. Seven of these 29 patients had also been included in a previous study [13]. 22 of the patients underwent surgery based on the reported imaging findings, and in one patient a CT-guided biopsy was performed. In 24 masses from 23 patients, histology revealed 18 renal cell carcinomas, 2 transitional cell carcinomas, 1 low grade renal cell carcinoma, 1 oncocytoma, 1 tuberculosis and 1 echinococcus cyst. Tumour sizes ranged from 1.5 cm to 10 cm. In the other six patients, no pathological lesions were found. In 18 of the 29 patients, simple cysts (1 to >6 per patient) corresponding to the Bosniak classification (category I) [14] were found. Confirmation of these cysts was based on clinical follow-up, laboratory findings (blood and urine samples), or imaging follow-up (ultrasound, helical CT or MRI). All patients were followed up clinically for at least 1 year.

Imaging technique
Helical CT
All studies were performed with a Somatom plus 4 helical scanner (Siemens, Erlangen, Germany). A multiphase imaging protocol was applied using a triphasic acquisition comprising the plain pre-contrast, corticonephrographic, and finally the tubulonephrographic phase. Oral barium contrast medium (1 l Micropaque oral, Guerbet, Sulzbach, Germany) was administered to discriminate intestinal structures. All scans were obtained at 240 mAs and 120 kV. The CT protocol consisted of volumetric data acquisition beginning at the level of the diaphragm and continuing to the right lower pole of the kidney. Imaging parameters were 5 mm slice collimation, 7 mm table speed per rotation and 4 mm reconstruction increment before and after intravenous administration of the contrast material bolus (volume: 130 ml, flow rate: 3 ml s-1, delay 45 s and 120 s) (Solutrast 300, Schering, Berlin, Germany). After the pre-contrast scans (5 min before start of the contrast bolus), 20 ml of contrast medium were administered to enhance the renal pelvis and the ureters. The last scan was performed from the diaphragm to the symphysis. The scan time for one rotation of the X-ray tube was 0.75 s (acquisition time 18–24 s per scan for the plain and corticonephrographic phases, and 30–35 s for the tubulonephrographic phase). Images were obtained with standard soft tissue window and level settings (450 HU and 50 HU).

MRI
MRI was performed on a 1.5 T superconducting imaging system (ACS NT, Philips Medical Systems, Best, Netherlands) using the standard body coil. All patients received butylscopolaminium bromide (0.1 ml kg-1 body weight (bw)) intravenously to reduce bowel motion. The imaging protocol included T2 weighted sequences as previously described [15]. The important parameters of the applied sequences were as follows:

1. Fast spin-echo sequence with shortened echo spacing (ultrashort turbo spin echo, UTSE), 8 mm slice thickness in axial plane (repetition time (TR): 4186 ms; echo time (TE): 100 ms; turbo factor (TF): 21; number of signal averages (NSA): 4; scan time: 3:37 min). To make the data set comparable with the CT examination, the number of slices was set to 40 covering the whole abdomen.

2. The same UTSE sequence was performed in the coronal plane with 20 slices and a slice thickness of 4 mm to cover the kidneys in a second plane (TR: 2109 ms; TE: 100 ms; TF: 21; NSA: 6; scan time: 2:44 min).

3. The third T2 weighted sequence was a UTSE sequence with frequency-selective fat suppression (selective pre-saturation using inversion recovery, SPIR) in the same plane and with the same slice thickness as described under 2 (TR: 2390 ms; TE: 100 ms; TF: 21; NSA: 6; scan time: 2:58 min).

For the T1 weighted images, a gradient-echo-sequence was performed in the axial plane before application of the contrast medium (TR: 145 ms; TE: 4.6 ms; flip angle (FA): 70 degrees; slice thickness: 6 mm; 21 slices; NSA: 1; scan time: 0:43 min). After administration of the intravenous contrast medium (0.1 mmol kg-1 bw Gd-DTPA, Schering, Berlin, Germany), a T1 weighted UTSE SPIR sequence was applied (TR: 503 ms; TE: 9 ms; TF: 5; slice thickness: 6 mm; 21 slices; NSA: 1; scan time: 0:54 min). The T1 weighted sequence started 4 min to 5 min after the contrast medium application.

Image analysis
Qualitative CT evaluation
All images were read by two experienced radiologists (PL, AGo) retrospectively without knowledge of clinical, other imaging, or histological data. The interpretation was made with consensus between the two. Lesion detection images from all three phases were analyzed together. The image quality regarding differentiation between renal parenchyma, renal pelvis and perirenal space, and the degree of visualization of renal vessels, lymph node status and the assessment of other abdominal organs was assessed using a four-point scale (1=excellent, 2=good, 3=sufficient, 4=insufficient). The readers were asked to determine the acquisition phase that showed the best image quality and provided the best diagnostic information. Size, location, contrast enhancement and enhancement pattern of each detected lesion were recorded for each patient. Cystic renal lesions were classified using the Bosniak criteria [14]. Solid lesions with contrast enhancement (>20 HU) were recorded as being suspicious. Histologically proven renal cell carcinomas (n=18) were staged based on the Robson classification (Table 1Go) [16, 17].


View this table:
[in this window]
[in a new window]
  		Table 1. Staging of renal cell carcinoma with Robson's classification

 
Quantitative CT evaluation
The CT values for attenuation of the renal cortex, medulla, pelvis, vein, solid renal lesions (n=20, 17 renal cell carcinomas, 1 transitional cell carcinoma, 1 oncocytoma, 1 low grade renal cell carcinoma), and cysts (n=18) were also measured. Measurements of attenuation were obtained from operator-defined regions of interest (ROIs) with an area of more than 10 mm2 identical in size and position in all three acquisition phases for all patients. The quantitative values were expressed as mean±standard deviation (SD).

Qualitative MRI evaluation
All images were read by two experienced radiologists (CW, AGi) retrospectively using the criteria described above for image quality and lesion detection by CT. Corresponding to the CT evaluation the interpreters were asked to determine the sequence that showed the best image quality and gave most diagnostic information according to the criteria described above (see: Qualitative CT evaluation). For each patient lesion detection was performed by recording size, location, contrast enhancement and enhancement pattern on MRI. Two groups of renal lesions were defined. The first group consisted of simple renal cysts, which showed intermediate to low signal intensity on post-contrast T1 weighted images without contrast enhancement and high signal intensity on T2 weighted images. The diagnosis of a renal cyst was additionally made by the following previously published criteria: absence of a discernible wall and presence of a sharp delineation from surrounding parenchyma [10, 14]. The second group comprised solid lesions with inhomogeneous or heterogeneous signal intensity in different sequences with contrast enhancement. Solid lesions without significant enhancement, e.g. angiomyolipoma, were not observed in the present patient sample. Renal cell carcinomas (n=18) were subsequently staged based on the Robson classification.

Quantitative MRI evaluation
The signal intensity of the renal cortex, medulla, pelvis, vein and renal lesions, and background noise was determined. Intensity values were obtained from operator-defined ROIs with more than 20 pixels identical in size and position in all evaluated images for all patients. In renal lesions (tumours or cysts) measurements from two ROIs were averaged. Background noise was measured with the ROI placed outside the tissue.

Renal tumours may consist of different tissue portions (solid, cystic, necrotic, haemorrhagic, etc.) with different relaxation times, leading to corresponding signal intensities that depend on image weighting. Thus, renal masses, especially renal cell carcinomas, can show up from predominantly hypointense to hyperintense in all sequences. However, on T1 weighted images after contrast application [18] renal carcinomas have the tendency to show hypointensity compared with renal parenchyma. Since in this study all tumours on T1 weighted images were hypointense, these tumours were summarized in one group. On T2 weighted sequences tumours can either show more or less hypointense as well as hyperintense structures, thus two groups of tumours were defined for these sequences. Lesion/renal parenchyma contrast-to-noise ratio (CNR) was calculated as the difference between the signal intensity of the lesion and renal parenchyma, respectively, divided by the standard deviation of the background noise. CNR was calculated for 20 histologically confirmed solid lesions (17 renal cell carcinomas, 1 transitional cell carcinoma, 1 oncocytoma, 1 low grade renal cell carcinoma) and 18 cysts. The quantitative values were expressed as mean±SD.

Statistical analysis
All statistical analyses were performed with the SPSS software package (SPSS Inc., Chicago, IL).

A first analysis focused on the qualitative MR and CT data. Separately for both imaging modalities, Friedman's rank repeated-measures analysis of variance (ANOVA) was performed on the image quality parameters. This analysis examines whether the image quality parameters differ within each modality. In case of a significant result (alpha <0.05), follow-up tests were performed: each pair of parameters was compared with a Wilcoxon test. Since 5 parameters result in a total of 10 comparisons, a Bonferoni correction was applied (alpha <=0.005 for each individual W-test).

Statistical analysis of the quantitative data was performed with paired Student t-tests and Wilcoxon Signed-Rank test comparing mean differences in objective ROI measurements among the three helical CT series and the different image MRI sequences. The sensitivity and specificity of both methods were not calculated because of the number of patients (see results, lesion detection and characterization). The accuracy between helical CT, MRI and histology was determined. The McNemar tests and Kappa coefficients were used to evaluate the degree of difference between both methods in staging renal cell carcinomas.


   Results
Top
 Abstract
 Introduction
 Patients and methods
 Results
Discussion
 References
 
Qualitative and quantitative analysis
The Friedman rank ANOVA detected no differences between the parameters of CT image quality [c2(df=3)=2613; p>0.45)]. However, the analysis was significant for the 5 parameters of the MRI image quality [c2(df=3)=3369; p<0.001)]. Pair wise comparison with Wilcoxon tests showed that the assessment of the other abdominal organs was inferior to all other parameters (p<0.001). The four other parameters did not differ (all p>0.35).

Helical CT
The differentiation between perirenal space and renal parenchyma as well as the assessment of the other abdominal organs was excellent. Only the differentiation between renal pelvis and renal parenchyma was sometimes slightly impaired. Visualization of renal vessels and detection of lymph nodes were rated excellent except for a few individual cases.

There were no relevant differences between the cortical and tubular acquisition phase with regard to differentiating between renal pelvis and renal parenchyma. In 10 out of 21 of the cases, the demarcation of renal pelvis to renal parenchyma was best in the corticonephrographic phase, in the other cases the tubulonephrographic phase proved to be better. A similar ratio was observed for the demarcation of perirenal space to renal parenchyma (corticonephrographic: 15, tubulonephrographic: 14). The renal vessels were better visualized in 16 patients in the corticonephrographic and in 13 patients in the tubulonephrographic phase. In most cases lymph nodes could be detected best in the tubulonephrographic phase (24 of 29) and fewer (5 of 29) in the corticonephrographic phase. Without exception, the best assessment of other abdominal organs, especially the liver, was achieved using the tubulonephrographic phase. Conversely, the plain phase failed to yield superior diagnostic information in any of the investigated patients.

Quantitative analysis yielded an average attenuation of the kidneys in unenhanced images of 36 HU±4 for the renal cortex and the medulla (Table 2Go). In the cortical acquisition phase, mean cortical attenuation was significantly higher than the attenuation of the renal medulla (p<0.01). In contrast, the attenuation of both regions reached the same level in the tubulonephrographic phase (Table 2Go). Attenuation in the renal pelvis increased from 23 HU±11 to 328 HU±196 and 276±155 in the corticonephrographic and tubulonephrographic phase, respectively (p<0.01). However, no significant differences were observed between the two contrast-enhanced acquisition phases with regard to the renal pelvis. The increase of the attenuation measured in the left renal vein was slightly diminished compared with the renal cortex (Table 2Go). Nonetheless, the changes in attenuation were significantly different compared with the non-enhanced images (p<0.01). Attenuation in the late phase (tubulonephrographic) was significantly decreased in the left renal vein (p<0.01, Table 2Go). Renal simple cysts showed no contrast enhancement and therefore the attenuation ranged from 18 HU (native phase) to 20 HU (tubulonephrographic phase). The measured attenuation of the tumours showed an increase from 39 HU±8 to 101 HU±45 in the corticonephrographic phase and remained at this level in the tubulonephrographic phase.


View this table:
[in this window]
[in a new window]
  		Table 2. Quantitative analysis of spiral CT

 
Fast MRI
Image quality in MRI generally received lower ratings than helical CT for most image quality criteria, especially in the assessment of the abdominal organs. The T2 weighted UTSE sequences appeared to be most effective for diagnosis, due to less marked pulsatility and fewer bowel motion artefacts as compared with the T1 weighted sequences. Only in a few cases (7 of 29) the demarcation between renal space and renal parenchyma was better in the T1 weighted sequence. Moreover, assessment of other abdominal organs also gave equal or better diagnostic information in 5 cases.

In the quantitative analysis, the CNR of the T2 weighted and T1 weighted sequences were within the same range with respect to the hyperintense renal tumours (Table 3Go). For the hypointense tumours in the T2 weighted sequence the CNR was slightly, but not significantly, decreased. The same levels of CNR were observed for the cysts in T1 weighted and T2 weighted sequences (Table 3Go).


View this table:
[in this window]
[in a new window]
  		Table 3. Quantitative analysis of renal lesions in MRI

 
Lesion detection and characterization
All 24 histologically proven renal masses in 23 patients were detected by helical CT and MRI using a triphasic helical CT protocol and fast T1 weighted and T2 weighted sequences. A differentiation between solid benign and malignant lesions, e.g. renal adenoma and oncocytoma versus renal cell carcinoma, was not possible. Compared with malignant tumours, benign solid lesions showed a similar signal intensity on the T1 weighted and T2 weighted MRI images, comparable attenuation with CT and a similar contrast enhancement pattern with both imaging modalities. In 18 of the 29 patients simple renal cysts could be diagnosed using either one of the two imaging methods. There was no problem in differentiating simple renal cysts and cystic renal tumours using helical CT or MRI, although it may be noted that there were no patients with complicated renal cysts in the present sample. Nonetheless, no false positive solid tumour was diagnosed with either imaging modality.

Staging of renal cell carcinomas with MRI and CT
Histology revealed 8 carcinomas at Robson stage I, 6 renal cell carcinomas at stage III, and 4 renal cell carcinomas at stage IV (n=18). Altogether, 12 of 18 renal cell carcinomas (67%) were staged correctly by both CT and MRI (Table 4Go, Figure 1Go). Five of 8 renal cell carcinomas, histologically confirmed to be Robson stage I, were classified correctly on CT, 2 were incorrectly identified as stage II and 1 as stage IV (Table 4Go). The tumours of Robson stage III were classified correctly in 4 of 6 cases. One of these 6 cases showed histological evidence of microscopic invasion of the renal vein, but no macroscopic tumour thrombus. One of the remaining 2 Robson stage III cases was underestimated as Robson stage II, while the other was overestimated as stage IV with CT. In 3 of 4 cases of Robson stage IV, CT allowed a correct staging of the tumours. One tumour was falsely classified as stage III.


View this table:
[in this window]
[in a new window]
  		Table 4. MRI and CT staging (Robson's classification)

 


View larger version (115K):
[in this window]
[in a new window]
  		Figure 1. Renal cell carcinoma (histology: Robson stage I) in the lower part of the left kidney in a 57-year-old patient. Comparison of helical CT and fast MRI. (a) Helical CT in plain acquisition phase the tumour was slightly hypodense compared with the renal parenchyma. (b) Helical CT after application of contrast medium the renal cell carcinoma showed inhomogeneous enhancement in the tubulonephrographic phase. (c) Fast MRI in the axial T2 weighted ultrashort turbo spin echo (UTSE) sequence the normal renal parenchyma had intermediate signal intensity and the renal tumour showed predominantly hyperintense signal changes, indicating a cystic or necrotic component of the tumour. (d) Fast MRI after injection of contrast agent a similar enhancement pattern of the carcinoma was found in the axial T1 weighted UTSE sequence with fat suppression (SPIR) compared with the enhancement in helical CT. The renal cell carcinoma was correctly staged in both modalities (Robson stage I).

 
With MRI, 7 of 8 renal cell carcinomas that were histologically reported as Robson stage I were correctly diagnosed (Table 4Go). Only one stage I tumour was falsely classified as stage II. However, only 1 of 6 tumours of Robson stage III was correctly categorized. One tumour which showed only histological evidence of microscopic invasion of the renal vein was diagnosed as Robson stage II, just as 3 other stage III tumours. One of these 6 tumours was overestimated and classified as Robson stage IV. All 4 histologically confirmed stage IV tumours were correctly staged with MRI. The accuracy was 67% for either method (confidence interval 45–85%) with regard to staging renal cell carcinomas (Table 4Go).


   Discussion
Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
References
 
The finding of incidental lesions using ultrasound in clinical practice and in the evaluation of haematuria leads to an increasing number of CT and MRI examinations. Thus, detection and characterization of renal masses, as well as the staging of malignant renal tumours are essential for appropriate management, especially if nephron-sparing surgery is considered. The last few years have brought about remarkable technical progress in helical CT and fast MRI. Different imaging protocols have been proposed for both imaging modalities. However, in previous studies conducted to compare MRI and CT in renal imaging, examination protocols have varied and did not reflect the available technical standard, particularly when conventional incremental CT was compared with high-speed MRI. Even the protocol of the present study may be further improved owing to technical progress such as multislice helical CT and new sequence and coil developments in MRI. Nonetheless, the present study focused on clinical routine protocols of helical CT and fast MRI because these methods are often recommended in the literature and optimized for renal imaging [5, 6, 13, 15].

Helical CT
In previous studies, different acquisition phases have repeatedly been evaluated [6, 19, 20]. Most authors propose a biphasic imaging protocol consisting of a corticonephrographic and a tubulonephrographic phase. However, detection and characterization of small renal lesions and the evaluation of contrast enhancement pattern additionally requires the plain phase [5]. Some authors even recommend a fourth acquisition phase to evaluate the excretory phase and to visualize the collecting system [21]. In the present study, this fourth scan of the excretory phase was considered unnecessary. Our approach yielded very good visualization of the collecting system, and the certain diagnosis of a transitional cell carcinoma restricted to the collecting system was reliably obtained (Figure 2Go).



View larger version (115K):
[in this window]
[in a new window]
  		Figure 2. Transitional cell carcinoma of a 70-year-old man. (a) Helical CT after administration of 20 ml contrast medium 3–5 min before the bolus injection, the renal pelvis was sufficiently filled and the tumour could be detected (transverse slice in the tubolonephrographic phase). Fast MRI the tumour can be seen without contrast medium, using a T2 weighted sequence in (b) transverse and (c) coronal plane.

 
In the present investigation the beginning of the post-contrast acquisition was chosen at 45 s and 120 s in order to achieve optimal contrast enhancement in abdominal parenchymal organs, e.g. the liver [22], and vessels as well as in the kidneys. The ratings of the image quality (abdominal organs and renal veins) and the attenuation values supported this approach. In the renal vein, the contrast did not decrease, as had been observed in another study in which the second spiral scan was performed 180 s after contrast medium injection, which resulted in a significant decrease of attenuation in the renal vein [5].

The attenuation values of the renal parenchyma separated into renal cortex, medulla, renal tumours, and cysts were comparable with the numbers published in the literature [5, 20].

The data on image quality and attenuation of the different renal regions recommend the use of the presented imaging protocol for the detection and characterization of renal lesions (see imaging technique). All 24 histologically confirmed masses could be detected using this protocol, without any difficulties in differentiating solid and cystic masses. The imaging parameters used and the chosen delay seemed to offer the best compromise for depicting small renal masses [5], while still enabling reliable staging especially when assessing the abdominal organs and lymph nodes. Comparing the CT acquisition phases with regard to best diagnostic information, our findings suggest that all three phases should be used. If any of them is excluded from the protocol important diagnostic information may be lost.

Fast MRI
In the past, image quality in MRI was the critical factor in detection and characterization of renal masses. Rapid progress of technical equipment and pulse sequences offered increasingly shorter acquisition times and better image quality. However, acquisition time still remains a critical issue. None of the published studies comparing CT and MRI provide information about the scan time necessary to cover the whole abdomen and to achieve a spatial resolution comparable with the one offered by helical CT [10, 11]. In recent reports superior image quality and shorter acquisition time of fast T1 weighted and T2 weighted sequences in renal imaging was demonstrated in comparison with conventional spin echo sequences [13, 15]. Nonetheless, in order to image the whole abdomen from the diaphragm to the symphysis in a reasonable length of time, it was necessary to increase slice thickness up to 8 mm (whereas a slice thickness of 4 mm was possible with CT), which resulted in a loss of spatial resolution in MRI. As described previously, it is necessary to combine different T1 weighted and T2 weighted sequences to obtain all necessary diagnostic information for correct diagnosis and characterization of renal masses [23]. To obtain the same diagnostic information, the total duration of the MRI examination is three to four times longer than with CT.

Overall, the image quality received lower ratings with MRI as compared with CT, due to more bowel and motion artefacts. The assessment of the abdominal organs, e.g. the liver, was particularly impaired by these image artefacts. The CNR was comparable with T1 weighted and T2 weighted sequences in other studies [13, 15] and offered sufficiently high contrast between renal parenchyma to renal tumours or cysts (Table 3Go). The evaluation of the MRI sequences offering the best diagnostic information showed the necessity to include T2 weighted sequences. Except in very few cases all image criteria required to make a diagnosis were fulfilled by the applied T2 weighted UTSE sequence. T1 weighted images were mainly used to determine the contrast enhancement pattern and to confirm or exclude the malignancy of a detected renal lesion. Nevertheless, the image information regarding differentiation of solid or cystic lesions were similar on MRI and CT. All 24 renal masses were correctly diagnosed and characterized, and as with CT, no false positive solid lesion was detected (Figure 1 and 2GoGo).

Staging of renal cell carcinomas: comparison of MRI and CT
Accurate pre-operative staging affects the therapeutic approach and aids in planning surgery [24]. The results of the presented study showed that MRI is similar to helical CT in staging renal cell carcinomas (67% accuracy for both imaging modalities), but as in other studies the number of examined patients was small. Nevertheless, the results are comparable with those described in earlier reports [25, 26]. As described previously [27, 28], the differentiation of stage I and stage II is difficult no matter what imaging technique is used. The invasion of perirenal fat tissue is often microscopic and no image criteria exist to diagnose such invasion. The staging of stage I was slightly more accurate with MRI compared with CT.

The accuracy of detecting venous extension has previously been reported to be as high as 100% for renal thrombosis [29]. In our series of patients, CT was superior to MRI in detecting venous infiltration. The difficulty with histological Robson stage III is that the venous involvement may only be microscopic. Accordingly, one histologically proven stage III tumour in our patient sample only had microscopic tumour invasion. This patient was staged as Robson stage III on CT without having a macroscopic tumour or thrombus in the renal vein. MRI correctly detected no macroscopic vessel invasion. The difference between the results in the presented study and the findings in the literature in assessing renal venous involvement may be partially due to the chosen imaging protocol. The highest sensitivity and specificity to assess venous involvement can be achieved with a gradient-echo sequence. In other studies that used spin-echo sequences the detection rate of venous invasion was comparable with our data [26, 30].

In summary, our study qualitatively and quantitatively compared the diagnostic potential of triphasic acquisition helical CT and fast multisequential MRI for the detection and characterization of renal masses. The presented findings demonstrate that the optimized examination protocols allowed excellent detection and correct characterization of all suspicious renal lesions. Therefore, both imaging modalities can be recommended for clinical routine application in these cases. The correct staging of renal cancer still remains difficult for both imaging methods. Especially for renal vein involvement, high contrast has to be achieved with CT, and for MRI the gradient-echo sequence should probably be used [31]. In general, both helical CT and fast MRI can provide answers to the important clinical questions in order to ensure the right treatment regarding suspicious renal masses.


   Acknowledgments
 
The authors gratefully acknowledge Dr E Naumann, Department of Statistical Methodology, University of Trier, for his assistance with statistical analysis.


   Footnotes
 
Supported in part by a Köln Fortune grant (86/1997). Back

Received for publication August 6, 2002. Revision received March 27, 2003. Accepted for publication July 3, 2003.


   References
Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 

  1. Kalender WA, Seissler W, Klotz E, Vock P. Spiral volumetric CT with single-breath-hold technique, continuous transport, and continuous scanner rotation. Radiology 1990;176:181–3.[Abstract/Free Full Text]
  2. Outwater EK, Mitchell DG, Vinitski S. Abdominal MR imaging: Evaluation of a fast spin-echo sequence. Radiology 1994;190:425–9.[Abstract/Free Full Text]
  3. Catasca JV, Mirowitz SA. T2 weighted MR imaging of the abdomen: fast spin-echo vs. conventional spin-echo sequences. AJR Am J Roentgenol 1994;162:61–7.[Abstract/Free Full Text]
  4. Riderer SJ. Recent technical advances in MR imaging of the abdomen. J Magn Reson Image 1996;6:822–32.[CrossRef]
  5. Szolar DH, Kammerhuber F, Altziebler S, et al. Multiphasic helical CT of the kidney: increased conspicuity for detection and characterization of small (<3 cm) renal masses. Radiology 1997;202:211–7.[Abstract/Free Full Text]
  6. Kopka L, Fischer U, Zoeller G, et al. Dual-phase helical CT of the kidney: value of the corticomedullary and nephrographic phase for evaluation of renal lesions and preoperative staging of renal cell carcinoma. AJR Am J Roentgenol 1997;169:1573–8.[Abstract/Free Full Text]
  7. Silverman SG, Lee BY, Seltzer SE, et al. Small (<3 cm) renal masses: correlation of spiral CT features and pathologic findings. AJR Am J Roentgenol 1994;163:597–605.[Abstract/Free Full Text]
  8. Melki PS, Argaud C, Suminski M, Helenon O, et al. Comparison of single-breath-hold fast spin-echo sequences with routine non-breath-hold techniques: application to MRI of renal masses. J Magn Reson Image 1996;6:608–14.[CrossRef]
  9. Semelka RC, Hricak HH, Stevens S, Finegold R, Tomei E, Carrol PR. Combined gadolinium-enhanced and fat-saturation MR imaging of renal masses. Radiology 1991;178:803–9.[Abstract/Free Full Text]
  10. Semelka RC, Shoenut JP, Kroeker A, MacMahon RG, Greenberg HM. Renal lesions: controlled comparison between CT and 1.5T MR imaging with non-enhanced and gadolinium-enhanced fat suppressed spin-echo and breath-hold FLASH-techniques. Radiology 1992;182:425–30.[Abstract/Free Full Text]
  11. Kreft B, Steudel A, Textor J, et al. Magnetresonanztomographie der Nieren: Erprobung neuer Pulssequenzen und Vergleich mit der CT in der Differentialdiagnose von Raumforderungen. Fortschr Röntgenstr 1996;164:212–7.[CrossRef]
  12. Nishimura K, Hida S, Okada K, Yoshida O, Nishimura K. Staging and differential diagnosis of renal cell carcinoma: a comparison of magnetic resonance imaging (MRI) and computed tomography (CT). Acta Urol Jpn 1988;34:1323–31.
  13. Walter C, Kruessell M, Heindel W, Kugel H, Jung G, Gindele A. Fast and ultrafast sequences with fat suppression and breath hold mode: new standard in renal imaging? Eur Radiol 2001;11:2092–8.[CrossRef][Medline]
  14. Bosniak MA. The small (=3.0 cm) renal parenchymal tumor: detection, diagnosis, controversies. Radiology 1991;179:307–17.[Free Full Text]
  15. Walter C, Jockenhöfer A, Heindel W, Kugel H, Jung G, Krahe T, et al. Fast and ultrafast magnetic resonance imaging in renal lesions. J Magn Reson Image 1998;8:1219–27.[CrossRef]
  16. Patel SK, Stack CM, Turner DA. Magnetic resonance imaging of renal cell carcinoma. RadioGraphics 1987;7:703–28.[Abstract]
  17. Robson CJ, Churchill BM, Anderson W. The results of radical nephrectomy for renal cell carcinoma. J Urol 1969;101:297–301.[Medline]
  18. Sussmann SK, Glickstein MF, Krzymowski GA. Hypointense renal cell carcinoma: MR imaging with pathologic correlation. Radiology 1990;177:495–7.[Abstract/Free Full Text]
  19. Cohan RH, Sheman LS, Korobkin M, Bass JC, Francis IR. Renal masses: assessment of corticomedullary-phase and nephrographic-phase CT scans. Radiology 1995;196:445–51.[Abstract/Free Full Text]
  20. Birnbaum BA, Jacobs JE, Ramchandant P. Multiphasic renal CT: comparison of renal mass enhancement during the corticomedullary and nephrographic phases. Radiology 1996;200:753–8.[Abstract/Free Full Text]
  21. Zeman RK, Zeiberg A, Hayes WS, et al. Helical CT of renal masses: the value of delayed scans. AJR Am J Roentgenol 1996;167:771–6.[Abstract/Free Full Text]
  22. Heiken JP, Brink JA, Vannier MW. Spiral (helical) CT. Radiology 1993;189:647–56.[Abstract/Free Full Text]
  23. Krestin GP, Hauser M, Huch Böni RA, Knecht M, Debatin J. Dynamic contrast-enhanced breathheld GRE sequences in detecting and characterizing renal lesions. Proc ISMRM, 5th Meeting, Vancouver: ISMRM 1997:812.
  24. Jakobson E, Eickhoff JH, Andersen JP, Ottesen M. Prognosis after nephrectomy for renal cell carcinoma. Scand J Urol Nephrol 1994;28:229–36.[Medline]
  25. Fein AB, Lee JKT, Balfe DM, et al. Diagnosis and staging of renal cell carcinoma: a comparison of MR imaging and CT. AJR Am J Roentgenol 1987;148:749–53.[Abstract/Free Full Text]
  26. Gschwend JE, Vogel U, Bader C, Mattfeldt T, Hautmann RE. Predictive value of magnetic resonance imaging and computed tomography for conservative renal surgery in an ex vivo tumor enucleation study followed by step-sectioning. J Urol 1996;155:451–4.[CrossRef][Medline]
  27. Johnson CD, Dunnick NR, Cohan RH, Illescas FF. Renal adenocarcinoma: CT staging of 100 tumors. AJR Am J Roentgenol 1987;148:59–63.[Abstract/Free Full Text]
  28. Zagoria RJ, Bechtold RE, Dyer RB. Staging of renal adenocarcinoma: role of various imaging procedures. AJR Am J Roentgenol 1995;164:363–70.[Abstract/Free Full Text]
  29. Roubidoux MA, Dunnick N, Sostman HD, Leder RA. Renal carcinoma: Detection of venous extension with gradient-echo MR imaging. Radiology 1992;182:269–72.[Abstract/Free Full Text]
  30. Kabala JE, Gillat DA, Persad RA. Magnetic resonance imaging in the staging of renal cell carcinoma. Br J Radiol 1991;64:683–8.[Abstract/Free Full Text]
  31. Laissy JP, Menegazzo D, Debray MP, Toublanc M, Ravery V, Dumont E, et al. Renal carcinoma: diagnosis of venous invasion with Gd-enhanced MR venography. Eur Radiol 2000;10:1138–43.[CrossRef][Medline]



This article has been cited by other articles:


Home page
 Am. J. Roentgenol.Home page
K. E. Maturen, H. V. Nghiem, E. M. Caoili, E. G. Higgins, J. S. Wolf Jr., and D. P. Wood Jr.
Renal Mass Core Biopsy: Accuracy and Impact on Clinical Management
Am. J. Roentgenol., February 1, 2007; 188(2): 563 - 570.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Roentgenol.Home page
A. J. Beer, M. Dobritz, N. Zantl, G. Weirich, J. Stollfuss, and E. J. Rummeny
Comparison of 16-MDCT and MRI for characterization of kidney lesions.
Am. J. Roentgenol., June 1, 2006; 186(6): 1639 - 1650.
[Abstract] [Full Text] [PDF]

Home page
 RadioGraphicsHome page
R. F. J. Browne, C. P. Meehan, J. Colville, R. Power, and W. C. Torreggiani
Transitional Cell Carcinoma of the Upper Urinary Tract: Spectrum of Imaging Findings
RadioGraphics, November 1, 2005; 25(6): 1609 - 1627.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Walter, C
Right arrow Articles by Landwehr, P
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Walter, C
Right arrow Articles by Landwehr, P

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
BJR DMFR IMAGING  ALL BIR JOURNALS