| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 All authors: Department of Radiology, Wake Forest University School of Medicine, Medical Center Blvd., Winston-Salem, NC 27157-1088.
Received July 14, 2000;
accepted after revision October 24, 2000.
Address correspondence to R. J. Zagoria.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. The images of 26 patients whose urinary tract calculi were revealed on unenhanced helical CT and who also underwent digital abdominal radiography were retrospectively reviewed. CT features studied included size and CT attenuation of the calculi. These CT findings were correlated with the ability to detect the same calculi with radiography.
RESULTS. Forty-nine urinary tract calculi were detected with unenhanced helical CT in 26 patients. Twenty-six (53%) calculi were visible on radiography. Most (79%) calculi larger than 5 mm were detectable with radiography (p < 0.01). One (8%) of 13 calculi with CT attenuation below 200 H was detectable on radiographs. Ninety-five percent (21/22) of calculi with CT attenuation exceeding 300 H were visible on radiographs (p < 0.0001). The one remaining calculus was obscured by overlying anatomy.
CONCLUSION. Radiographic surveillance of urinary tract calculi detected with CT may not be useful if the calculi have a CT attenuation below 200 H. Most calculi larger than 5 mm and nearly all calculi with a CT attenuation exceeding 300 H can be seen on abdominal radiographs.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This study was undertaken to determine if calculus detectability with abdominal radiographs could be predicted at the time of diagnosis on the basis of the size and the CT attenuation of urinary tract calculi.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Each CT scan was obtained on either a HiLight Advantage or a LightSpeed CT scanner (General Electric Medical Systems, Milwaukee, WI) without oral or IV contrast media. All scans were completed with 5-mm collimation and a pitch of 1.5-2.0 from the top of the kidneys through the bladder base. Focal calcifications in the kidneys or ureters were diagnosed as calculi.
We retrospectively analyzed each CT scan, using a picture archiving and communication system (PACS) monitor. CT scanners used in this study have been calibrated for contrast scale annually, and their Hounsfield unit measurements have never varied by more than 2 H. CT images containing the urinary tract calculi were magnified on the monitor screen. The dimensions of each calculus were measured using electronic calipers and soft-tissue-viewing settings. The longest dimension of the calculus on the axial image was recorded. Using an ellipseshaped region of interest, which fell completely within the borders of the calculus, we twice measured the average CT attenuation of each calculus. The two mean attenuation measures were averaged. The location of the calculus was also recorded.
Digital abdominal radiography was performed using a computerized radiography system that uses storage phosphor plate technology (ADC 5145; Agfa-Gevaert, Munich, Germany). All radiographs were exposed using a 100-cm focal-to-film distance and phototiming. Radiographs were exposed with kilovoltage set at 70-80 kV, depending on the body habitus of the patient. Digital radiographs and CT scans were reviewed from a PACS with 2000-line-resolution monitors (IMPAX RS 3000; Agfa, Ridgefield, NJ). The resolution of CT images (512 x 512) remains the same when reviewing on this PACS.
Retrospectively, the report from digital abdominal radiography was reviewed. A comparison of the size and CT attenuation for the CT-detected calculi was made with the digital abdominal radiography reports. The reviewers who made the initial interpretation of the radiographs knew the CT results if CT had been performed before radiography. If the finding of a patient's original radiograph was positive, the film was not reviewed for this study. For patients for whom there was a discrepancy or questionable finding on the initial interpretation, the original CT scans and digital abdominal radiographs were reviewed again by two radiologists with expertise in urolithiasis. A consensus was obtained from these radiologists and this information recorded. Patient demographics and indications for the CT scans and digital abdominal radiographs were also recorded.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
|
|
Of patients with calculi, four patients had only one renal calculus, six patients had two renal calculi, and one patient had five renal calculi. Nine patients had one ureteral calculus, and one patient had three ureteral calculi. Of the remaining five patients, one had four renal calculi and one ureteral calculus, one had one ureteral calculus and one renal calculus, one had two ureteral calculi and one renal calculus, and two patients had two renal calculi and one ureteral calculus.
The timing of the studies was also analyzed, and this analysis revealed that 19 of 26 patients had both the digital abdominal radiographs and the unenhanced helical CT scans obtained on the same day. Two patients had the digital abdominal radiograph and the CT scan obtained 1 day apart. One patient had the studies obtained 2 days apart, one patient 3 days apart, one patient 5 days apart, one patient 8 days apart, and one patient 9 days apart. The average number of days separating the two examinations was 1.1 days.
Tables 1 and 2 and Figure 3 summarize the comparisons of calculus size and the CT attenuation for the calculi with detectability on digital abdominal radiographs.
|
|
|
Analysis of the data showed that the correlation of CT attenuation values for calculi below 200 and those greater than 300 H with calculus detectability on digital radiographs was statistically significant (Fisher's exact test, p < 0.0001). Also, there was a significant (Fisher's exact test, p < 0.01) relationship between calculus size and detectability with digital abdominal radiography.
Only one calculus with an attenuation value below 200 H on unenhanced helical CT was revealed on digital abdominal radiography (Fig. 4A,4B). Analysis of these images confirmed that the reported calculus represented on the radiograph was most likely a superimposed density rather than the calculus visible on the unenhanced helical CT image. Only one CT-detected calculus with an attenuation coefficient greater than 300 H was undetected with digital abdominal radiography (Fig. 5A,5B). This patient's images were also retrospectively reviewed, and it was found that the calculus was obscured by overlying bone and stool. These findings were considered to be one false-positive finding and one false-negative finding for the purpose of data analysis.
|
|
|
|
There was a statistically significant correlation between calculus size larger than 5 mm and visibility on digital abdominal radiographs. This correlation was less significant than that for CT attenuation of calculi and their radiographic visibility. Seventy-nine percent of calculi larger than 5 mm were detected with digital radiography. For smaller calculi, size is less of a determinant of detectability, with 63% being undetectable but 37% visible with radiography.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Previous studies have shown that 45-59% of ureteral calculi revealed on unenhanced helical CT are detectable with abdominal radiography [7]. Similarly, 49% of ureteral calculi visible on unenhanced helical CT scans were visible on the routine CT scout radiographs in a recent study [11]. However, because serial unenhanced helical CT scanning, which exposes the patient to approximately 10 times the radiation of a single abdominal radiograph [1], is impractical for following calculus progression in all patients, abdominal radiographs are often obtained after the diagnosis of urinary tract calculi with unenhanced helical CT. Unfortunately, 47% of urinary tract calculi in this study were undetectable with abdominal digital radiography, making this examination useless for calculus surveillance in many patients. This finding is similar to the previously reported series. However, our study indicates that many unnecessary abdominal radiographs obtained for patients whose unenhanced helical CT scans show urinary tract calculi can be avoided. If calculi detected on helical CT are analyzed using CT attenuation coefficients, then unnecessary radiography of many of the radiolucent calculi can be eliminated.
Calculi that are detected with unenhanced helical CT and that have a CT attenuation lower than 200 H will not be detectable on abdominal radiographs. Thirteen calculi in this series measured below 200 H. Of these, 12 were undetectable, invisible on abdominal radiographs. One was reported to be revealed on abdominal radiography, but retrospective review indicated that this radiographic density may have been a superimposed shadow on the side of the radiolucent calculus (Fig. 4A,4B). Of the 22 CT-detected calculi with attenuation higher than 300 H, 21 (95%) were detected with abdominal radiography. The single, high-attenuation calculus that was not detected with abdominal radiography was in the ureter in an area obscured by overlying bone and stool (Fig. 5A,5B).
Fourteen calculi in this series measured between 200 and 300 H on unenhanced helical CT. Of these, most (10/14, 71%; size range, 2-8 mm) were radiolucent and invisible on an abdominal radiograph. However, four (29%; size range, 2-4 mm) of the 14 calculi in this attenuation range were radiopaque with abdominal radiography.
The detection of calculi with radiography depends on several factors, including calculus composition, calculus size, and presence of superimposed structures. Prior studies have shown that CT attenuation coefficients can be used to predict the chemical composition of urinary tract calculi [12,13,14,15] with calculi composed of radiolucent material having lower CT attenuation values. There is also considerable overlap of CT attenuation values for various chemical compositions of calculi [13,14,15]. The CT attenuation of calculi reflects both the composition of the calculus and volume averaging. Small calculi that do not "fill" the thickness of the CT slice will have measured attenuation below their actual density due to averaging with measurements of surrounding soft tissues with lower attenuation. If the calculus is only 2.5 mm (slice thickness, <5 mm), there is a substantial partial volume effect. Although this effect gives a falsely low measurement for the calculus' CT attenuation based on composition, it more truly reflects the radiographic opacity of the calculus, which is also affected by both calculus size and chemical composition. This effect is likely a factor involved in this study because 47% of CT-detected urinary tract calculi were radiolucent. This figure is well above the expected prevalence of radiolucent, uric acid, and xanthine calculi in the general population, which is in the range of 10-20% of all urinary tract calculi [16, 17]. The radiolucency of some of the calculi can be attributed to their diminutive size.
The size of urinary tract calculi measured on CT scans is also predictive of their visibility on abdominal radiographs. Most calculi 5 mm or smaller are not detectable, but 79% of larger calculi are visible on radiographs. Therefore, abdominal radiographs are likely to be useful for calculus detection if calculi are larger than 5 mm on the unenhanced helical CT scans. The fact that the percentage (79%) of larger calculi that are radiopaque is close to the expected percentage of radiopaque calculi based on the prevalence of calcium calculi in prior series [16, 17] likely reflects that partial volume averaging is minimal for larger calculi. This result is expected but helps to explain why a much lower percentage (37%) of smaller calculi are radiopaque. It is likely that many of these are composed of radiopaque material, but they are radiographically undetectable because of their diminutive size.
Size criteria are less useful for predicting radiographic detectability of smaller calculi. Over one third of calculi 5 mm or smaller will be visible with radiography, making size a less selective factor.
There are some limitations to this study. The sample size is relatively small, 49 calculi in 26 patients, and a prospective study will be useful in confirming these results. We evaluated calculi in different locations, both the ureters and the kidneys. Although radiographic visibility might be slightly influenced by calculus location, we did not identify such a relationship. Not all patients from our institution who had a calculus detected had both radiography and CT performed, and this fact could have introduced a selection bias. Because this study only examined the radiographic detectability of calculi in comparison with the CT features of these calculi, the non-randomized design of the study should not influence the results. The need for both studies was mainly determined by the ordering habits of the referring urologists at our institution. Therefore, even though there are potential limitations, the data presented here will likely be useful as a model for future study of a practical method of directing the imaging follow-up, when required, of urolithiasis detected with unenhanced helical CT.
In conclusion, the CT attenuation of the calculi can be used to predict the utility of radiography in revealing these urinary tract calculi (Fig. 6). It may be advisable to measure and report the size and mean CT attenuation of all ureteral calculi detected on unenhanced helical CT scans. Then, if radiologic follow-up of calculi is indicated to assess calculus progression or to plan lithotripsy, the utility of an abdominal radiograph can be accurately predicted without the need for unnecessary radiographs in patients with radiolucent calculi. Calculi that have an average CT attenuation of less than 200 H will most likely be radiolucent and not detectable on abdominal radiographs. Calculi with an attentuation of greater than 300 H will be visible on abdominal radiographs unless they are obscured by superimposed structures. In those cases, oblique views or tomography may be useful in overcoming the obscuration by superimposed structures. Abdominal radiographs should be obtained for calculi that have an attenuation between 200 and 300 H. Although most of these calculi will be radiolucent, some calculi in this range (29%) will be visible on abdominal radiographs. Most calculi larger than 5 mm will be visible with radiography. Nearly half of all urinary tract calculi detected with unenhanced helical CT are undetectable with abdominal radiography because of diminutive size, chemical composition, or superimposed anatomy.
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  R. J. Zagoria Retrospective view of "diagnosis of acute flank pain: value of unenhanced helical CT". Am. J. Roentgenol., September 1, 2006; 187(3): 603 - 604. [Full Text] [PDF]
|
|
|
|
|
|
L. F. Rogers Dot Size, Lead Time, and Other Such Clever Cognomen: The Use of Catch Phrases Am. J. Roentgenol., May 1, 2001; 176(5): 1103 - 1103. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
