| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Genitourinary Imaging |
1 From the Department of Radiology, University of Michigan Medical Center, 1500 E Medical Center Dr, Taubman Center B1 132D, Ann Arbor, MI 48109-9723. Received July 10, 2003; revision requested September 25; revision received August 19, 2004; accepted October 11. Supported by the Radiological Society of North America Research and Education Foundation. Address correspondence to E.M.C. (e-mail: caoili@umich.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Institutional review board approval for reviewing images and medical records of the patients was obtained; informed patient consent was not required. Excretory phase images obtained from multi–detector row CT urography in 85 patients (57 men, 28 women) were reviewed. Examinations were performed by using one of four techniques: abdominal compression and intravenous hydration with 250 mL of normal saline, compression only, intravenous hydration with saline only, and neither compression nor saline hydration. Excretory phase imaging was performed at 300 and 450 seconds for each patient. Two reviewers measured urinary tract distention on transverse images and graded opacification and image quality on volume-rendered images. Effects were compared by using statistical mixed models with repeated-measures analysis of variance.
RESULTS: Saline hydration significantly improved opacification (P = .02) and overall image quality (P < .001) of the intrarenal collecting system and proximal ureter. Delayed excretory phase image acquisition of 450 seconds significantly increased distention of the intrarenal collecting system and proximal ureter (P < .001). No significant effects involving the lower segment of the ureter were seen with any technique; however, there were fewer nonvisualized distal ureteral segments with the longer imaging delay.
CONCLUSION: Compression does not significantly improve distention or opacification of the urinary tract. Saline hydration is effective in improving opacification of the proximal urinary tract. Longer imaging delays improve distention of the proximal urinary tract and may aid in visualization of the lower segment of the ureter.
© RSNA, 2005
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
To date, there are few available data on the assessment of the best way to optimize these features during multi–detector row CT urography. A few investigators (1,4,5) have suggested that abdominal compression during CT urography results in improved opacification of the intrarenal collecting system and ureter and yields comparable opacification when compared with that at excretory urography. Others (6) have noted that the use of intravenous saline hydration yields opacification of the urinary tract comparable to that obtained with excretory urography. This study was performed to retrospectively compare the effects of abdominal compression, intravenous saline hydration, and two imaging delays on both distention and opacification of the intrarenal collecting system and ureter during multi–detector row CT urography.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Imaging
Imaging was performed with multi–detector row CT scanners (LightSpeed QX/i, version 1.3; GE Medical Systems, Milwaukee, Wis). Patients were imaged by using a standardized multi–detector row CT urographic procedure with images obtained during unenhanced, nephrographic, early excretory, and delayed excretory phases. The unenhanced images were obtained through the abdomen and pelvis at 4 x 3.75-mm collimation each (four detector rows at 3.75-mm collimation) and were reconstructed at a 5-mm section thickness. Nephrographic phase images were obtained 100 seconds after the initiation of an intravenous injection of low-osmolality nonionic contrast material (150 mL of Omnipaque 300 [300 mg of iodine per milliliter]; Nycomed, New York, NY) at 3 mL/sec. Nephrographic phase images were obtained from the diaphragm through the kidneys by using 4 x 2.5-mm collimation and were reconstructed at 5-mm section thickness. Early and late excretory phase images were obtained at 300- and 450-second delays, respectively, with the following technique: 4 x 1.25-mm collimation, reconstructed section thickness of 2.5 mm, and 50% (1.25-mm) overlapping intervals. All imaging examinations were performed at 120 kVp and 120–280 mA. Three-dimensional reconstructions of the two excretory phase scans were created at an independent workstation (Advantage Windows 3.1; GE Medical Systems) by CT technologists.
Four imaging procedures were used during the duration of the study. Following the unenhanced phase, abdominal compression was applied to 21 patients immediately prior to contrast material injection. Patients with a history of recent abdominal surgery, abdominal aortic aneurysm, or urinary tract obstruction did not receive abdominal compression and were not included in the analysis. Compression balloons positioned over the anterior abdominal wall and directly below the iliac crests were held in place by a compression belt. The balloons were inflated prior to intravenous contrast agent administration. Compression was maintained during the 100- and 300-second image acquisitions and then was released immediately prior to the 450-second image acquisition. Saline hydration was administered to 23 patients. Each of these patients received an intravenous bolus infusion of 250 mL of normal (0.9%) saline 10–15 minutes immediately prior to multi–detector row CT urography. Both abdominal compression and saline hydration were administered to 19 patients by using the procedures described earlier. Twenty-two patients underwent multi–detector row CT urography without abdominal compression or intravenous saline hydration.
Image Interpretation
Two abdominal radiologists (E.M.C., R.H.C.), both with 52 months of experience with multi–detector row CT urographic interpretation, reviewed the transverse excretory phase images and the three-dimensional reconstructed images at the workstation. Both reviewers were blinded to the imaging procedure used in each patient. To assess urinary tract distention, transverse images were evaluated at the same window settings (window width, 2000 HU; window level, 50 HU). The urinary tract was evaluated at seven segments. The renal collecting system (calyces, infundibula, and renal pelvis) was divided into the upper, middle, and lower pole regions. The upper pole was defined as the most cranial transverse image on which an infundibulum was opacified. The lower pole was defined as the most caudal transverse image on which an infundibulum was opacified. The middle renal collecting system was the portion of the collecting system between the upper and the lower pole. The renal pelvis was not evaluated because of the natural variation in renal pelvic caliber.
The ureter was divided into the proximal, middle, lower, and pelvic segments. The proximal segment was defined as the portion of the ureter that extended from the renal pelvis to the most inferior aspect (lower pole) of the ipsilateral kidney on transverse images (hereafter, proximal ureter). The middle segment was defined as the portion of the ureter located between the lower pole of the ipsilateral kidney and the ipsilateral superior iliac crest (hereafter, middle ureter). The lower segment was defined as the portion of the ureter that extended from the superior iliac crests to the middle of the sacroiliac joint. The pelvic segment was defined as the portion of the ureter distal to the middle of the sacroiliac joint continuing to the ureterovesical junction. At each of these levels, the maximal short-axis diameter of the renal collecting system infundibulum or ureter was measured in consensus on 300- and 450-second transverse excretory phase images (Fig 1).
|
|
|
|
Statistical Analysis
Since patients were not randomly assigned to an imaging procedure, exploratory univariate analyses by using analysis of variance (ANOVA) and 
2 tests were performed by comparing baseline characteristics of patients across the imaging procedures to assess for confounders that might explain any potential differences in results.
We estimated that clinically meaningful differences in the mean short-axis diameter (distention), the opacification score, and the image quality score were 0.25 mm, 0.25, and 0.25, respectively. We calculated that our chosen sample size provided greater than 85% power, with 
= .05 indicating a significant difference.
Consensus measurements of the urinary tract distention were analyzed statistically by using repeated-measures ANOVA, because multiple measurements of each urinary tract segment are not independent variables when obtained in the same patient. For the purpose of this analysis, the portion of the urinary tract from the intrarenal collecting system to the middle ureter constituted the proximal urinary tract. The remainder of the ureter comprised the distal urinary tract. These definitions were chosen on the basis of the expected effect of abdominal compression, which was placed at approximately the middle level of the ureter. Data regarding distention were analyzed with multifactoral repeated-measures ANOVA models by using SAS statistical software (SAS Institute, Cary NC), which calculated the mean short-axis diameter for the included urinary tract segments. The model accounted for abdominal compression, saline administration, and time of image acquisition.
Statistical comparisons were made between patients who received abdominal compression and patients who did not receive abdominal compression, and the comparisons were adjusted for other variable effects such as saline administration and time. Similar statistical comparisons were made between patients who received intravenous saline and patients who did not receive saline and were adjusted for other variable effects such as abdominal compression and time. Finally, statistical comparisons were made for measurements at a 300-second delay and those obtained at a 450-second delay and were adjusted for other variable effects such as abdominal compression and saline administration. A similar analysis of the consensus data was performed for opacification and image quality by using the generalized estimating equations approach, because of the ordinal nature of the dependent variables, the opacification score, and the image-quality grade. In cases of unequal scores or grades, a mean score or grade was assigned to each segment, with equal weight assigned to each reviewer’s score or grade. Again, a mean score or grade was calculated for each variable effect with similar statistical comparisons as aforementioned. P < .05 was considered to indicate a significant difference.
Post hoc analysis was performed with respect to the urinary tract segments that were not visualized (neither distended nor opacified) on the basis of the delay of the excretory phase. Also, to determine if any imaging procedure was associated with improved visualization of the urinary tract, the number of nondistended and nonopacified segments was compared among the four different procedures by using a 2 test.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
|
|
Opacification
The opacification score of the renal collecting systems and proximal ureters, which represented the percentage of a segment filled with contrast material on volume-rendered images, was greater with compression than without, with saline hydration than without, and with the longer excretory phase imaging delays (Fig 4). However, the only statistically significant difference in the renal collecting system and proximal ureteral opacification was that between saline hydration and no saline hydration (P = .02). The difference in opacification between compression and noncompression approached significance, with P = .12. Interestingly, upper tract opacification was not significantly diminished on compression-release images when compared with that on compression images (mean score of 2.96 vs 3.10, P = .41).
|
Post hoc analyses of the data combining all imaging procedures revealed there were more nonvisualized (nondistended, nonopacified) segments in the middle and distal ureters than in the proximal ureter (<10% of proximal segments vs <20% of distal segments) (Fig 5). Although fewer nonvisualized segments were seen with use of both abdominal compression and saline hydration, this was not significantly different than with use of other imaging procedures (P = .08) (Fig 6). Visualization of the middle and distal ureters improved when images were acquired at 450 seconds than when they were acquired at 300 seconds. Nearly 25% of distal ureteral segments were not opacified at 450 seconds; however, a full one-third of distal ureteral segments were not opacified at 300 seconds (Fig 5).
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Abdominal compression during excretory urography was first described in 1930 (7) and gained acceptance by urologists and radiologists (8–10) who experimented with variations of this maneuver, such as the split-film technique developed by Heetderks et al (9) and prone positioning described by Pendergrass (10). Abdominal compression was used to produce a partial obstruction of the middle ureter to improve filling of the urinary tract. Hamby and Kirsch (11) compared excretory urography with ureteral compression versus without ureteral compression. They graded pelvic density, calyceal outline, and ureteral delineation and found that ureteral compression significantly improved the quality of excretory urography with increasing time delays and that compression also reliably provided complete opacification of the urinary tract (11).
Currently, there is no consensus regarding how to perform multi–detector CT urography when abdominal compression is used. While several groups (2,12) have performed excretory phase imaging of the proximal renal collecting systems and ureters during compression and after compression release, others (1,4,5) have used a split-imaging technique reminiscent of the method described by Heetderks et al (9). With the split-imaging technique, initial excretory phase images of the proximal urinary tract are obtained with compression. Subsequently, compression is released, and a set of images through the lower urinary tract is acquired. Authors of studies using both techniques have found compression to be effective. While in some of the studies, CT with abdominal compression was compared with excretory urography with compression (1,4,5), in other studies CT with compression was compared with CT without compression in control groups of patients (2,4).
Heneghan et al (5) compared opacification of the renal collecting system and ureters in 50 patients who underwent CT urography with compression with that in 50 unmatched patients who underwent excretory urography with compression. They reported improved opacification of the pelvicalyceal system and the middle ureter and equivalent opacification of the proximal and distal ureters at CT urography compared with that at excretory urography. However, these investigators did not assess the true effect of compression since there was no control group for comparison.
McNicholas et al (4) compared opacification in 25 patients who underwent supine CT urography without compression with that in 10 patients who underwent supine CT urography with compression. Additional comparisons were made in 10 other patients who underwent CT urography in the prone position and in 25 patients who underwent excretory urography without compression. Prone and CT urography with compression improved opacification of the intrarenal collecting system and the upper ureter compared with excretory urography, whereas supine CT urography without compression did not. CT urography with compression demonstrated greater mean opacification scores of the entire intrarenal collecting system and the entire ureter than did excretory urography. Opacification scores were significantly higher for the middle and distal ureters in patients imaged with abdominal compression than in patients imaged without compression (4). Caoili et al (12) compared urinary tract distention and opacification in 31 patients who underwent helical CT with compression and in 29 patients who did not receive compression. Compression produced a detectable increase in distention of the infundibula, calyces, pelves, and proximal ureters.
As an alternative to compression, several investigators have chosen to administer intravenous saline for hydration either prior to or following contrast material injection. The effects of saline hydration on distention and opacification have also been previously evaluated with excretory urography. Authors of two excretory urography studies (13,14) determined that hydration had the desirable effect of improving filling of the calyces but had an undesirable effect of decreasing density of the contrast-enhanced urine due to the lower concentration of excreted contrast material in the hydrated patients. Because of the superior contrast resolution inherent in CT, excretion of diluted contrast material into the urinary tract is not problematic and may be preferred to prevent streak artifact and obscuration of intraluminal abnormalities (15). Results of the evaluation of the effectiveness of saline hydration on urinary tract visualization with multi–detector row CT urography have been encouraging.
McTavish et al (6) compared mean opacification scores in patients who underwent prone CT urography with a 250-mL infusion of normal saline immediately after the administration of contrast material and in patients without saline hydration. Additional comparisons were made between patients who underwent supine CT urography with saline hydration and patients who underwent excretory urography with compression. Saline hydration improved the mean opacification scores for both prone and supine CT urography in comparison to opacification scores obtained with prone CT urography without saline; however, the only significant improvement was found in the distal ureter. Thus, the advantages of saline hydration with CT urography were limited. McTavish et al also determined that CT urography with saline hydration demonstrated comparable opacification of the urinary tract when compared with excretory urography, with significant improvement of opacification in the distal ureter. The authors concluded that CT urography with saline hydration reliably opacifies the urinary tract (6). Maher et al (16) found that 100 mL of normal saline administered after contrast material injection did not improve opacification of the urinary tract but did improve distention of the pelvis and upper and middle ureters when compared with distention in a control group.
In our study, we sought to compare the relative effects of abdominal compression and saline hydration on both urinary tract distention and opacification. Although earlier reports regarding CT urography have used the terms opacification and distention interchangeably, we believe these are two distinct characteristics. Distention refers to the fullness of the urinary tract segment, whereas opacification refers to the length of the segment filled with contrast material. There is some overlap in that a nonopacified segment is not distended, but the two terms are not equivalent because an opacified segment may have a variable degree of distention. Although maneuvers that might improve one of these characteristics might also improve the other, this is not necessarily the case. Therefore, these characteristics should be evaluated independently.
Perhaps the greatest limitation we encountered was related to the number of lower urinary tract segments that were not visualized, particularly during the earlier of the two excretory phase image acquisitions. Authors of other studies (4,6,7) that evaluated CT urography and excretory urography have also reported the limitation that not all segments are visualized. Daughtridge (8), for example, reported that complete visualization of the ureters at excretory urography is seen in only 65% of patients over 50 years of age. There is no consensus regarding CT urography and the optimal delay for excretory phase image acquisition. For example, Heneghan et al (5) performed excretory phase imaging at a 180-second delay, and McTavish et al (6) used a 480–600 second delay. Given that data show that the diagnostic quality of excretory urography regarding both calyceal and ureteral delineation improves increasingly as time progresses from a 1- to a 15-minute delay after contrast agent injection (11), it may be that longer imaging delays with multi–detector row CT urography are also preferable. Authors of a CT study (12) in which two excretory phases were directly compared found that scanning at 300 seconds rather than at 150 seconds resulted in improved collecting-system distention and more consistent opacification. The results of our current investigation suggest that an even longer imaging delay may be more beneficial. The imaging delay of 450 seconds significantly improved distention of the proximal urinary tract and improved opacification and image quality. It is possible that longer delays may prove even more advantageous.
Our study had several other limitations. Although the subjects in each imaging procedure were relatively homogeneous according to our initial analyses, unforeseen unidentified variations in baseline characteristics could have affected our results. Patient weight, height, body mass index, renal function, or cardiac function may have been confounding variables in the effectiveness of abdominal compression or saline hydration that we could not address. For example, the use of abdominal compression would likely be compromised if a large number of our study population were obese, and this may potentially explain the differences in results in this study versus of those of earlier studies (12). It is also possible that the compression balloons were not positioned correctly or were not inflated enough for a desired effect and thereby reduced the effectiveness of abdominal compression; however, several CT technologists were involved in performing the compression procedure, making this systematic bias less likely.
We selected a study population with a high clinical suspicion of urinary tract disease. Thus, our results can only be applied to those of similar patients and may not be generalizable to other patient populations. Similarly, opacification and perceived image quality were evaluated from volume-rendered images and not images obtained with other techniques, such as maximum intensity projection or average intensity projection. Thus, our results apply specifically to volume-rendered images only. Another limitation is the smaller amount of data on distention and opacification of the lower urinary tract due to the presence of many nonvisualized lower ureteral segments. This smaller amount of data affected statistical analyses, which caused insufficient power to detect relevant differences. The poor visualization may have been caused by too short an imaging delay, particularly since improved visualization was seen with the longer delay of 450 seconds. However, the effectiveness of delays of more than 450 seconds warrants further study. Last, it remains unproven (though widely believed) that improved distention and opacification at multi–detector row CT urography allow for improved detection of collecting-system abnormalities.
In summary, our results indicate that the use of abdominal compression and saline hydration produce few beneficial effects on urinary tract distention, opacification, and overall image quality. Compression was not shown to significantly improve distention or opacification of any portion of the urinary tract. Saline hydration is effective at improving opacification and overall image quality of the upper urinary tract, and use of the longer image delay of 450 seconds enhances distention of the proximal urinary tract and produces fewer nonopacified distal ureteral segments. This knowledge has led us to modify our CT urography procedure by incorporating saline hydration and an even longer imaging delay (450 seconds) in hope of further improving opacification of the distal ureteral segments. It is hoped that this and other future refinements in multi–detector row CT urographic technique will positively affect our ability to detect even the most subtle urinary tract abnormalities.
|
FOOTNOTES |
|---|
Authors stated no financial relationship to disclose.
Author contributions: Guarantors of integrity of entire study, E.M.C., R.H.C.; study concepts, E.M.C., R.H.C., J.H.E.; study design, E.M.C., R.H.C.; literature research, E.M.C., P.I.; clinical studies, P.I., R.H.C., E.M.C.; data acquisition, P.I.; data analysis/interpretation, E.M.C., P.I., R.H.C.; manuscript preparation and revision/review, all authors; manuscript definition of intellectual content and editing, E.M.C., R.H.C., J.H.E.; manuscript final version approval, E.M.C., R.H.C.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. Kim, L. L. Wang, J. P. Heiken, C. L. Siegel, C. F. Hildebolt, and K. T. Bae Opacification of Urinary Bladder and Ureter at CT Urography: Effect of a Log-rolling Procedure and Postvoiding Residual Bladder Urine Volume Radiology, June 1, 2008; 247(3): 747 - 753. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. C. Chow, S. W. Kwan, E. W. Olcott, and G. Sommer Split-Bolus MDCT Urography with Synchronous Nephrographic and Excretory Phase Enhancement Am. J. Roentgenol., August 1, 2007; 189(2): 314 - 322. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. A. Jaffe, L. C. Martin, C. M. Miller, K. M. Franklin, E. M. Merkle, W. M. Thompson, R. C. Nelson, D. M. DeLong, and E. K. Paulson Abdominal Pain: Coronal Reformations from Isotropic Voxels with 16-Section CT--Reader Lesion Detection and Interpretation Time Radiology, January 1, 2007; 242(1): 175 - 181. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. G. Silverman, S. A. Akbar, K. J. Mortele, K. Tuncali, J. G. Bhagwat, and J. L. Seifter Multi-Detector Row CT Urography of Normal Urinary Collecting System: Furosemide versus Saline as Adjunct to Contrast Medium Radiology, September 1, 2006; 240(3): 749 - 755. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
