| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
Department of Radiology, Université Catholique
de Louvain, Saint-Luc University Hospital, Ave. Hippocrate 10, B-1200
Brussels, Belgium.
2
Center of Biostatistics and Medical Documentation,
Université Catholique de Louvain, Mont-Godinne
University Hospital, Ave. Thérasse 1, B-5530
Yvoir, Belgium.
Received December 11, 2000;
accepted after revision May 16, 2001.
Address correspondence to B. E. Van Beers.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. The study included 72 patients with renal colic who underwent abdominal radiography and multislice helical CT. For each patient, a frontal MIP image was generated, and the stone, when present, was marked with a cross on the transverse CT scan. The cross appeared automatically on the corresponding MIP image. The CT examination was used as the standard of reference. The presence and location of ureteral stones on the abdominal radiographs were assessed during three interpretation sessions. In the first session, the abdominal radiographs were viewed alone. In the second, they were viewed with the transverse CT scans. In the third, the abdominal radiographs were viewed with the CT scans and the MIP images.
RESULTS. Ureteral stones were present in 58 patients. The percentage of stones detected on the abdominal radiographs was 45% when the radiographs were viewed alone, 66% when they were viewed with the CT scans (p = 0.002 vs radiographs alone), and 78% when viewed with the CT scans and MIP images (p = 0.016 vs radiographs with CT scans).
CONCLUSION. The sensitivity of stone detection on abdominal radiographs was greatest when the interpreters viewed the radiographs in conjunction with the CT scans and MIP images.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Prior knowledge of the precise location of the ureteral stone should improve detection of stones on abdominal radiographs. Therefore, viewing the abdominal radiographs in combination with CT scans should be useful [6, 11]. To show the precise location of the ureteral stone at CT in a format similar to that of abdominal radiography, we propose a frontal maximum-intensity-projection (MIP) image of the multislice helical CT data. The purpose of this study was to assess the improvement in the detection of ureterolithiasis on abdominal radiographs when the radiographs are viewed in combination with multislice helical CT scans and MIP images.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Imaging Techniques
The abdominal radiographs were obtained with clinical equipment (Siregraph
T.O.P. 33; Siemens, Erlangen, Germany). The exposure factors were 65-75 kVp,
15-50 mAs, and a 1-mm nominal focal spot. Regular intensifying screens were
used with T-MAT L/RA films (Eastman Kodak, Rochester, NY). The abdominal
radiographs were obtained before CT was performed.
The CT examinations were performed with a multislice helical scanner (MX8000; Marconi Medical Systems, Cleveland, OH). Unenhanced, single breathhold, helical acquisition was obtained from the top of the kidneys to the base of the bladder with 2.5-mm collimation (3.2 mm at full width at half maximum intensity); 120 kVp; 280 mAs; table speed, 16.7 mm/sec; and reconstruction interval, 1.6 mm.
One of three attending radiologists experienced with CT reviewed the images from each examination while the patient was still in the CT suite. The radiologist performed an additional contrast-enhanced CT examination in 13 patients in whom the presence of a ureteral stone or another lesion remained doubtful [15]. He also performed a coronal MIP reconstruction of the unenhanced CT data in each patient using a clinical workstation (MX view; Marconi Medical Systems). If the radiologist detected a stone on the transverse CT scans, it was marked with a cross. This cross automatically appeared in the same position on the MIP image if the "related slice" option of the clinical software was selected. Therefore, in patients with multiple calcifications, the ureteral stone detected on the transverse CT scan could be unequivocally located on the MIP image. A film containing four images—the transverse CT scan obtained at the level of the stone with and without the superimposed cross and the MIP image with and without the cross—was given to the referring physician.
Image Analysis
All abdominal radiographs, CT scans, and MIP images were reviewed in
consensus by two of the authors. The CT scans and MIP images were reviewed on
the same workstation described earlier. For the purpose of the analysis, a
patient was considered to have a ureteral stone if it was unequivocally
visible on the transverse CT scans. In other words, the CT examination was the
standard of reference. The largest transverse diameter and the density of the
stones were measured on the transverse CT scans. The location of the stones
was considered as proximal (above the sacroiliac junction), middle (at the
level of the sacroiliac junction), or distal (beneath the sacroiliac
junction).
Three interpretations of the abdominal radiographs were performed. In the first interpretation session, only the abdominal radiographs were viewed, and the interpreters had no knowledge of the CT findings. However, they were aware of the side of the body in which the patients felt pain. In the second session, the radiographs were reviewed along with the transverse CT scans. In the third session, the interpreters viewed the radiographs along with the transverse CT scans and the MIP images on which the location of the stone was marked with a cross. The interpretations of the abdominal radiographs were classified into three categories: positive for ureterolithiasis, negative for the condition, and indeterminate. Indeterminate results were considered interpretation errors. In accordance with previous studies [13, 16, 17], the ability of the interpreters to detect ureterolithiasis was estimated by determining the fraction of stones correctly detected and located.
The main cause of the interpretation error for the abdominal radiographs was categorized into two classes: poor contrast between ureteral stone and surrounding tissue (because of low stone density or the presence of overlying bone or intestine) and presence of phleboliths or other confounding calcified masses.
Applying the binomial test, we compared the detection rates for interpretations that used radiographs alone, radiographs along with transverse CT scans, and radiographs along with CT scans and MIP images. The diameter and densities of stones were compared using the Wilcoxon's rank sum test.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
|
|
|
|
|
|
|
|
|
The main cause of the errors on the initial interpretation was poor contrast between the ureteral stone and surrounding tissue in 24 patients and presence of phleboliths or other confounding calcified masses in 16 patients. Phleboliths were seen in 32 patients (44%) but were not always a cause of interpretation error.
The stones were located in the proximal ureter in 15 patients (21% of all patients), the middle ureter in five (7%), and the distal ureter in 38 (53%). The stone detection rate on the abdominal radiographs was 60% (9/15) for the proximal ureter, 40% (2/5) for the middle ureter, and 39% (15/38) for the distal ureter during the initial interpretation. During the second interpretation session, the corresponding figures were 87% (13/15), 40% (2/5), and 61% (23/38). During the last interpretation, they were 93% (14/15), 60% (3/5), and 74% (28/38).
The diameter of the stones was 4.6 ± 2.4 mm (mean ± SD). The CT density of the stones was 750 ± 331 H. The diameter of the stones detected during the initial interpretation (5.6 ± 2.3 mm) did not differ significantly from the diameter of the additional stones detected during the last interpretation (4.6 ± 1.8 mm, p = 0.25). Similarly, the density of these two categories of stones did not differ significantly (864 ± 333 H vs 779 ± 321 H, p = 0.41).
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The results of this study show that interpreting the abdominal radiographs in combination with multislice helical CT scans significantly improves the ability of radiologists to detect ureterolithiasis on the radiographs, especially if the CT scans are supplemented with MIP images showing the precise location of the stone in the same format as the abdominal radiographs. In several patients, the precise location of the stones on the abdominal radiograph remained indeterminate when viewed with the transverse CT scans but became obvious when viewed with the MIP image on which the stone had been marked (Figs. 1A,1B,1C,1D and 3A,3B,3C). (The contrast of the figures representing the abdominal radiographs has been electronically enhanced to improve the visibility of the stones.) The better detection of the stones on the abdominal radiographs viewed with the MIP images is in accordance with perception studies showing that prior knowledge of the presence and precise location of a lesion improves the interpreter's ability to recognize it [18, 19].
The two main reasons that ureteral stones are not recognized on abdominal radiographs are poor stone-to-tissue contrast and the presence of visual distractors. By knowing the precise stone location on the MIP reconstruction, we can recognize stones that are only faintly hyperdense relative to the surrounding tissue on the radiograph. Of course, not all urinary stones will be detected. Approximately 10-15% of stones appear radiolucent on abdominal radiographs, usually if they are composed of uric acid, cystine, or xanthine [20]. We think it noteworthy that our 78% detection rate of ureterolithiasis during the third interpretation (with MIP images) was similar to the 80% detection rate of an in vitro study of abdominal radiographs performed by Olcott et al. [20].
The presence of phleboliths or other calcified masses is another potential source of error when reviewing abdominal radiographs. Visual distractors are well-known causes of interpretation errors on medical images. This phenomenon has been called "satisfaction of search" and refers to the fact that the detection of one radiographic finding interfers with that of others [16, 17, 21]. Phleboliths were present in 32 patients in our series. They were not always a cause of diagnostic error because some phleboliths could be recognized by their round shape and radiolucent center [22]. However, phleboliths or other calcified masses were a cause of interpretation error on abdominal radiographs in 16 patients. MIP images were a particularly useful aid in locating a ureteral stone in patients with multiple phleboliths present.
We did not observe any major difference in size or density between the stones detected when abdominal radiographs were viewed alone and the stones revealed only when abdominal radiographs were viewed with the CT scans and MIP reconstructions. Regarding stone location, we found that the rate of detection of stones in the middle and distal ureters was lower than that for proximal stones. Similar results have been reported by Assi et al. [12].
In a previous study comparing the interpretations of abdominal radiographs with and without the use of CT scans, no substantial improvement in the detection of ureterolithiasis was observed when interpreters reviewed the abdominal radiographs along with the CT scans [10]. Several reasons may explain the differences between this previous study and ours. For example, the earlier study did not use MIP images and excluded indeterminate results from the statistical analysis.
Reformatting CT images of the ureter has been recommended in patients with ureteral calculi to obtain images in a format that is well known to clinicians [23]. However, creating curved planar reformats may be cumbersome. In contrast, creating global MIP images, as in our study, is rapid because no segmentation is needed. These MIP reconstructions provide images similar to the familiar conventional radiographs of the abdomen. We believe that these MIP images aid in communicating with clinicians. Nevertheless, acceptance by clinicians was not assessed in this study. We also did not calculate the sensitivity of CT scout radiography for the detection of ureteral stones. This sensitivity has been calculated in previous studies, and the researchers have reported that abdominal radiography was more sensitive than CT scout radiography in revealing ureteral calculi [12, 13].
Our study has limitations. First, the frequency of ureterolithiasis in our population was high (80.5%). The reported frequency of ureteral calculi in patients with acute flank pain is variable. In some large series, a frequency of approximately 50% has been reported [3, 24]. In other series, frequencies of up to 85% have been described [2, 5, 6]. The high frequency of individuals with ureterolithiasis in our patient population may have resulted from accurate patient selection by the physicians of the emergency department.
Second, the patients underwent both radiographic and CT examinations. Only recently has CT been performed systematically in patients presenting with acute renal colic to our emergency department. Because the availability of CT at night was limited, the previous standard policy for patients with renal colic was to perform abdominal radiography and sonography [9, 25]. During the intermediate period after the start of the new policy, patients routinely underwent abdominal radiography before CT scanning, although we are not recommending performing abdominal radiography in patients who will undergo CT.
We, as others [11,12,13, believe that an abdominal radiograph should be obtained after CT only in patients who will need follow-up imaging or lithotripsy if the CT scout radiography does not reveal the calculus. The results of our study suggest that the detection of ureteral stones in these patients may be improved if abdominal radiographs are viewed with helical CT scans and MIP reconstructions. On the basis of these findings, we recommend that abdominal radiography and frontal MIP reconstruction be performed in patients who will need followup imaging or lithotripsy if the ureteral stones is not visible on the CT scout image.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  S J Freeman and H Sells Investigation of loin pain Imaging, March 1, 2008; 20(1): 38 - 56. [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]
|
|
|
|
|
|
S J Freeman and H Sells Investigation of loin pain Imaging, August 1, 2005; 17(1): 19 - 33. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Memarsadeghi, G. Heinz-Peer, T. H. Helbich, C. Schaefer-Prokop, G. Kramer, M. Scharitzer, and M. Prokop Unenhanced Multi-Detector Row CT in Patients Suspected of Having Urinary Stone Disease: Effect of Section Width on Diagnosis Radiology, May 1, 2005; 235(2): 530 - 536. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J K Kim and K-S Cho CT urography and virtual endoscopy: promising imaging modalities for urinary tract evaluation Br. J. Radiol., March 1, 2003; 76(903): 199 - 209. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Tack, S. Sourtzis, I. Delpierre, V. de Maertelaer, and P. A. Gevenois Low-Dose Unenhanced Multidetector CT of Patients with Suspected Renal Colic Am. J. Roentgenol., February 1, 2003; 180(2): 305 - 311. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
