| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Genitourinary Imaging |
1 From the Departments of Radiology (M.M., G.H.P., T.H.H., C.S.P., M.S., M.P.) and Urology (G.K.), University of Vienna, General Hospital of Vienna, Waehringer Guertel 18–20, A-1090 Vienna, Austria. Received March 10, 2004; revision requested May 2; revision received June 28; accepted July 27. Address correspondence to M.M. (e-mail: mazda.memarsadeghi@meduniwien.ac.at).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: This study was approved by the ethics committee of the authors’ university, and written informed consent was obtained from all patients. One hundred forty-seven patients (age range, 11–101 years; mean, 51.4 years ± 18.7 [standard deviation]) suspected of having urinary stone disease underwent unenhanced multi–detector row CT. CT was performed with four detector rows, a section thickness of 1.0 mm, an effective tube current–time product of 100 mAs, and a tube voltage of 120 kVp (CT dose index, 11.4 mGy). From these data, three sets of transverse images were reconstructed with section widths of 1.5, 3.0, and 5.0 mm and approximately 50% of overlap each. Scans were evaluated in varying random orders by two radiologists for the number, size, and location of uroliths and nephroliths and for the presence of phleboliths, renal cysts, and secondary signs of obstruction. The significance of differences between the numbers of detected calcifications and the numbers of associated abnormalities on the scans obtained with varying section widths was tested with the McNemar test at a P level of less than .05. Spearman 
rank correlation coefficients were calculated to assess the correlation between the presence of uroliths and the presence of secondary signs.
RESULTS: Uroliths were found in 72 of 147 (49.0%) patients, and nephroliths were found in 16 patients (10.9%). There was no significant difference between section widths of 1.5 and 3.0 mm with regard to the number of detected stones (264 uroliths and 61 nephroliths for both protocols). Transverse sections 5.0-mm wide revealed significantly fewer uroliths (n = 231; P < .001) and nephroliths (n = 47; P < .016). The final diagnosis was changed in four of 72 patients. All missed renal and ureteral stones measured less than 3 mm in diameter.
CONCLUSION: Overlapping 3-mm sections are sufficient for the detection of urinary stone disease. Small calculi (<3 mm) may be missed on 5.0-mm-thick sections.
© RSNA, 2005
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
In the past decade, computed tomography (CT) has challenged intravenous urography as a tool for evaluating the genitourinary system. Since the first report in 1995 by Smith et al (2) about the use of unenhanced helical CT for identifying urolithiasis, CT has become a well-established diagnostic technique for evaluating patients suspected of having urinary stone disease. The advantages of CT over intravenous urography include a faster examination speed, the avoidance of intravenous contrast material, the ability to diagnose alternative abdominal diseases that mimic the symptoms of renal colic, and the ability to visualize urinary calculi directly as high-attenuation structures with sensitivities ranging from 96% to 100% and specificities ranging from 92% to 100% (1–14).
Multi–detector row CT allows for coverage of a relatively large volume, with near-isotropic resolution in all three dimensions (15). High-quality multiplanar reformations, which have the advantage of displaying the urinary tract in its longitudinal axis, require data acquisition based on thin sections (15,16). The disadvantage of these thin sections, however, is the relatively low signal-to-noise ratio that results when scan parameters are chosen to keep radiation exposure low (15–17). Reconstructing thicker sections will decrease noise but at the expense of spatial resolution and greater partial volume effects. In an attempt to define an optimal examination protocol, we assessed prospectively the effect of multi–detector row CT section width in the evaluation of patients with acute flank pain who are suspected of having or known to have urinary stone disease.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Image Acquisition
All examinations were performed with a four–detector row CT scanner (Somatom Volume Zoom; Siemens, Forchheim, Germany). Unenhanced scans were obtained at these settings: section thickness, 1.0 mm (for a detector configuration of 4 x 1.0 mm); rotation time, 0.5 second; pitch, 1.5; table feed, 6.0 mm per rotation; tube voltage, 120 kVp; and effective tube current–time product, 100 mAs. The resulting volume CT dose index was 11.4 mGy. The scan length was adapted to the length of the patient’s abdomen and pelvis. For image reconstruction, a routine convolution kernel (B30) and a 512 x 512 matrix were used.
We reconstructed three sets of transverse images with section widths of 1.5, 3.0, and 5.0 mm. The reconstruction increment was chosen to be approximately half the section width (0.7, 1.5, and 2.5 mm, respectively).
Image Evaluation
The three sets of images were independently reviewed by one 4th-year resident (M.M.) and one board-certified radiologist (G.H.P.) with 15 years of experience in reading abdominal CT results. The results of any examination from any of the image sets that was rated differently by the two observers were reevaluated by both readers together to reach a consensus opinion; this occurred in 12 cases. The readers were informed about the sites of clinical symptoms.
The readers were asked to determine separately the presence of parenchymal calcifications, pelvicaliceal calculi, and ureteral calcifications. They were instructed to document the location and the size of each calcification with standard metric software devices provided with the workstation (Magic View 1000; Siemens, Forchheim, Germany).
Uroliths were defined as focal calcified structures located within the ureter or the intrarenal pelvicaliceal collecting system. The readers were asked to assign the location of the ureteral calcification to the proximal, middle, or distal third of the ureter.
Ureteral calcifications were identified and differentiated from phleboliths when at least one of the following two criteria were noted: (a) the presence of a soft-tissue rim surrounding the calcification or (b) location of the calcification within the course of the ureter, as identified on an interactive cine-mode display provided with the workstation.
Nephroliths, or renal calculi, were defined as focal calcifications within the renal parenchyma (18). We decided to differentiate between nephroliths and pelvicaliceal calcifications because the latter can cause flank pain (19), while nephroliths are not associated with such pain. In calcifications that could not be classified as parenchymal or pelvicaliceal on the basis of morphologic criteria, the two readers assumed a pelvicaliceal location because all patients had acute clinical symptoms.
The readers were also asked to assess separately the presence or absence of the following secondary signs of acute obstruction, as described by Smith et al (20) and Varanelli et al (21): perirenal stranding, peripelvic stranding, pelvic dilatation, ureteral dilatation, and asymmetric renal enlargement. Similarly, the readers identified additional findings, such as renal cysts or phleboliths.
Three reading sessions were performed for scans obtained at section widths of 1.5, 3.0, and 5.0 mm. Each patient’s scans were presented only once in each reading session. Reading order was randomized with respect to patient order and protocols and varied between the two readers, who were blinded to section thickness during the reading time.
The interval between reading sessions was at least 2 weeks. Readers evaluated the images at a Magic View 1000 workstation with an interactive mouse-driven cine mode. The ambient light was subdued, and there were no time constraints.
Reference Standard
The standard of the truth was determined by consensus of the same two observers who had been involved in the evaluation reading. The diagnosis of urolithiasis or nephrolithiasis was based on unequivocal evidence of calcification on the thinnest transverse reconstruction with a section width of 1.5 mm. For determination of the standard of truth, information from all scans was available, as well as information regarding clinical symptoms, clinical history (including reports of hematuria or pain and treatment charts), and outcome.
Final Clinical Diagnosis
The final clinical diagnosis (other than urolithiasis) was determined by two radiologists (M.M. and M.P.) in consensus by reviewing urology department records, reports of surgical procedures, hospital discharge summaries, hospital charts, and the hospital’s computerized clinical database.
Literature Search
Three radiologists (M.M., C.S.P., M.S.), using the MEDLINE database and other available search algorithms, performed a comprehensive literature search of English-language abstracted studies that involved human subjects or phantoms. The search included articles published between January 1995 and December 2003; only articles with original results were included.
Statistical Analysis
Data were administered and analyzed with the SAS statistical package (SAS I.I., SAS, PROC FREQ SAS/STAT, version 8; SAS Institute, Cary, NC). Results are expressed as absolute numbers and means ± standard deviations. The significance of differences between the number of detected calcifications and the number of associated abnormalities on the images of varying section widths was tested with the McNemar test (significance, P < .05). A power analysis performed before this investigation with nQuery Advisor, version 5.0 (Statistical Solutions, Cork, Ireland) revealed that significant differences (ie, those with P < .05) would be detected for small effect sizes with a power of 0.8.
To assess any correlation between the presence of uroliths and the presence of secondary signs, Spearman correlation coefficients were calculated. These coefficients range from –1 (negatively correlated) to 0 (uncorrelated) to +1 (positively correlated), and the absolute value indicates the strength of the correlation (22).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Seventeen patients had only one stone, and 55 had more than one (Table 1). The stones ranged in size from 1 to 12 mm (mean, 3.12 mm ± 2.11; median, 2 mm). Alternative diagnoses were rendered for 53 of 147 patients (36.1%), and 22 patients (15.0%) had negative CT results. Alternative diagnoses included diverticulitis (n = 31), small-bowel obstruction (n = 8), acute pyelonephritis (n = 3), pancreatitis (n = 6), retroperitoneal abscess (n = 3), and appendicitis (n = 2).
|
There was no difference between section widths of 1.5 and 3.0 mm with regard to the number of detected uroliths (n = 264 for both section widths). With section widths of 5.0 mm, only 231 uroliths were detected in 54 patients, a significant difference (P < .001). The transverse diameters of the missed uroliths ranged from 1 to 3 mm (mean, 1.9 mm ± 0.4; median, 2 mm). The missed uroliths were located in the pelvicaliceal system (n = 18) in 16 patients, in the proximal third of the ureter (n = 2) in one, and in the distal third of the ureter (n = 2) in two (Figs 1–3). Only one urolith was missed in four patients, two were missed in 15 patients, and more than two were missed in the remaining 35 patients. In three patients, the missed uroliths would have led to false-negative interpretations of the examination results, and stone disease would have been missed. All three patients had only one urolith located in the pelvicaliceal system; the sizes of these missed uroliths ranged from 1.5 to 3.0 mm.
|
|
|
|
|
|
|
|
|
There was no difference between section widths of 1.5 and 3.0 mm with regard to the number of detected nephroliths. With 5.0-mm sections, 14 calcifications were missed in eight patients (P < .016). The sizes of missed calcifications ranged from 1 to 3 mm (mean, 2.0 mm ± 0.6). One calcification was missed in four patients, two were missed in two patients, and three were missed in two patients. In one patient, the missed calcifications would have led to a false-negative interpretation of the examination results as showing no nephroliths (Fig 4).
|
|
|
On the right side, perirenal stranding was present in six of the 23 patients (26%), pelvic dilatation in eight (35%), ureteral dilatation in six (26%), and asymmetric renal enlargement in three (13%). On the left side, perirenal stranding was also present in six of the 23 patients (26%), pelvic dilatation in 14 (61%), ureteral dilatation in six (26%), and asymmetric renal enlargement in two (9%). Two patients whose ureteral calculi were missed showed secondary signs of obstruction, such as ureteral dilatation and perirenal stranding, with the 5.0-mm section width. One patient whose pelvicaliceal calculi were missed showed pelvic dilatation with the 5.0-mm section width. We saw secondary signs equally well in all transverse image sets, regardless of section width. The correlation between the presence of a ureteral stone and the presence of secondary signs was low, with Spearman correlation coefficients ranging from +0.14 to +0.33.
Associated Findings
Renal cysts.—A total of 51 renal cysts with sizes ranging from 0.5 to 3.0 cm were equally well seen in all transverse image sets, regardless of section width.
Phleboliths.—There was no difference between transverse section widths of 1.5 and 3.0 mm with regard to the number of detected phleboliths (n = 180 in 128 patients). With transverse section widths of 5.0 mm, only 167 phleboliths were seen in 120 patients, a significant difference (P < .005).
Literature Search
From a total of 773 titles, 56 abstracts and 23 articles (2–7,9,10,12–14,17,21,23–32) were retrieved for data abstraction and evaluation of CT protocols. The readers were not blinded to the origin of publication, journal, or year of publication. Reported protocols were evaluated by three readers (M.M., C.S.P., M.P.), and the following data were recorded for each article: (a) author and year of publication, (b) study collective (human or phantom), (c) CT modality (single– or multi–detector row CT), (d) collimation, (e) pitch, (f) table feed, and (g) reconstruction increment. Reported protocols are summarized in Table 2.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Most of the clinical studies have been performed with 5-mm section collimation and a pitch between 1.0 and 2.0. Some authors even used a transverse section thickness of 7 mm (Table 2) (2–7,9,10,12–14,17,21,23–32). To our knowledge, there have been no clinical studies in which a 3-mm-collimation protocol with a single–detector row scanner was evaluated.
Only two phantom studies involving single–detector row CT systematically evaluated the effect of different scan protocols on image quality (23,31). Rimondini et al (23) compared 5- and 3-mm section collimation, obtained with pitches of 1.5 and 2.0, respectively. Although resolution and signal-to-noise ratio on transverse sections were higher for 5-mm sections, z-axis resolution was superior when thinner collimation was used. Their results, in agreement with those of Saw et al (31), showed that objects with high intrinsic attenuation, such as urinary calculi, would not be affected by the lower signal-to-noise ratio in thinner sections as much as objects of lower intrinsic attenuation would.
Rimondini et al (23) concluded from the results of their phantom study that a section collimation of 3 mm, combined with a pitch of 1.5 or 2.0, provides the best compromise with regard to radiation dose and image quality. Saw et al (31) measured the in vitro value of stone attenuation for 1-, 3-, and 10-mm section collimation obtained with a pitch of 1.0. Their results showed that increasing collimation width deteriorates the ability to differentiate stone compositions. In their study, the attenuation values were consistently lower at larger collimation widths. In addition, detection of smaller stones was not always possible with 10-mm collimation.
Multi–detector row CT now offers faster coverage of a large volume with near-isotropic resolution (15). Previous studies in which multi–detector row CT was used to detect renal calculi have focused on dose issues (17) and the additional diagnostic value of CT as an adjunct to conventional radiography (24). There is no doubt that multi–detector row CT is at least as diagnostically accurate as single–detector row CT for the evaluation of urinary stone disease. Tack et al (17) reported an accuracy of more than 93% for the detection of ureteral stones with a low-dose multi–detector row CT protocol (120 kVp, 30 mAs, effective doses of 1.2 and 1.9 mSv), with excellent inter- and intraobserver agreement.
There is still some controversy, however, regarding the optimal multi–detector row CT protocol for evaluating patients with acute flank pain. Tack et al (17) decided to use a 3-mm section width reconstruction and an effective pitch of 1.5 on the basis of the previously published results of Rimondini et al (23), while Katz et al (25,33) stated in their commentaries that they routinely use 5-mm section widths. Tack et al (17) obtained promising results with an effective dose no higher than that of three-film radiographic examinations, but more studies are needed to confirm the effectiveness of low-dose multi–detector row CT in the diagnosis of urinary stone disease.
The goal of our study was to assess the effect of varying section thicknesses on the detection of renal and ureteral calculi. We compared a 5.0-mm section width with 3.0- and 1.5-mm section widths. All images had been reconstructed in overlapping fashion and were evaluated in cine mode and soft-copy display. Our results showed a significantly decreased sensitivity with 5.0-mm sections, but no difference between 1.5- and 3.0-mm sections, in the detection of calculi. The diameters of stones that were missed ranged from 1 to 3 mm (mean, 1.9 mm ± 0.4; median, 2.0 mm). Most of the missed stones (18 of 22) were located in the pelvicaliceal system. The missed stones would have led to false-negative interpretations of the examinations as not indicating urinary stone disease in three patients.
We found phleboliths in 128 of our 147 patients. The classification of calcifications as phleboliths was based on morphologic criteria, such as a round shape or a radiolucent center (34), or on the precise identification of the ureter in a different position on the section that also contained the phlebolith. There was no difference in the detection rates for phleboliths between the transverse section width of 1.5 and that of 3.0 mm.
In contrast to Smith et al (20), we found secondary signs of obstruction in only 14 of 28 patients (50%) with ureteral stones. The Spearman correlation coefficient of less than 50% (0.33) indicated only a weak correlation, if any, between the presence of ureteral stones and signs of obstruction. The reason for this relatively low frequency of signs of obstruction associated with the presence of ureteral stones in our study may be explained by the fact that most of the ureteral stones were small (mean size, 3.12 mm) and located in the pelvicaliceal system and the proximal third of the ureter.
A limitation of our study was the absence of an absolute standard of truth. We defined the diagnoses of urolithiasis and nephrolithiasis as unequivocal evidence of calcification on the thinnest transverse reconstruction, with a section width of 1.5 mm, as determined by the consensus of two observers. This approach may theoretically increase the number of false-positive findings. However, as pointed out by previous authors (2–7,9,10,12–14,17,21,23–32), CT itself has become the standard method, and no other imaging method is more sensitive in the detection of calcifications. Although it is possible to misinterpret high-attenuation pixels caused by noise as urinary calculi, this effect was not relevant in our study because all calcifications seen on the thinnest and therefore noisiest sections were also found on the 3-mm-thick sections.
In conclusion, given our findings, we consider the reconstruction of overlapping 3-mm-thick transverse CT sections to be sufficient for the identification of urinary stones in patients with acute flank pain. The use of thinner sections increases the number of sections to be evaluated, and such sections suffer from increased noise; the use of thicker sections (
5 mm) decreases the detection rate and leads to inaccurate final diagnoses.
|
FOOTNOTES |
|---|
Author contributions: Guarantors of integrity of entire study, M.M., G.H.P., M.P.; study concepts, M.M., M.P., T.H.H.; study design, M.M., G.H.P., T.H.H., M.P.; literature research, M.M., M.S., C.S.P.; clinical studies, M.M., G.K., M.S., T.H.H.; data acquisition, M.M., M.S., G.K., G.H.P., T.H.H.; data analysis/interpretation, M.M., G.H.P., M.P.; statistical analysis, M.M., M.P.; manuscript preparation, M.M., C.S.P., M.P., M.S.; manuscript definition of intellectual content and final version approval, all authors; manuscript editing, M.M., C.S.P.; manuscript revision/review, M.M., C.S.P., G.H.P., M.P.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  Y. Minami, H. Chung, M. Kudo, S. Kitai, S. Takahashi, T. Inoue, K. Ueshima, and H. Shiozaki Radiofrequency Ablation of Hepatocellular Carcinoma: Value of Virtual CT Sonography with Magnetic Navigation Am. J. Roentgenol., June 1, 2008; 190(6): W335 - W341. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. T. Hsu, Z. J. Wang, A. S. L. Yu, R. G. Gould, Y. Fu, B. N. Joe, A. Qayyum, R. S. Breiman, F. V. Coakley, and B. M. Yeh Physiology of Renal Medullary Tip Hyperattenuation at Unenhanced CT: Urinary Specific Gravity and the NaCl Concentration Gradient Radiology, April 1, 2008; 247(1): 147 - 153. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Memarsadeghi, C. Schaefer-Prokop, M. Prokop, T. H. Helbich, C. C. Seitz, I. M. Noebauer-Huhmann, and G. Heinz-Peer Unenhanced MDCT in Patients with Suspected Urinary Stone Disease: Do Coronal Reformations Improve Diagnostic Performance? Am. J. Roentgenol., August 1, 2007; 189(2): W60 - W64. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. H. Mulkens, S. Daineffe, R. De Wijngaert, P. Bellinck, A. Leonard, G. Smet, and J.-L. Termote Urinary Stone Disease: Comparison of Standard-Dose and Low-Dose with 4D MDCT Tube Current Modulation Am. J. Roentgenol., February 1, 2007; 188(2): 553 - 562. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
