Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online before print August 30, 2002, 10.1148/radiol.2251010242
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2251010242v1
225/1/91    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
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 Tublin, M. E.
Right arrow Articles by Kliewer, M. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Tublin, M. E.
Right arrow Articles by Kliewer, M. A.
(Radiology 2002;225:91-96.)
© RSNA, 2002

Experimental Studies

Conspicuity of Renal Calculi at Unenhanced CT: Effects of Calculus Composition and Size and CT Technique1

Mitchell E. Tublin, MD, Michael E. Murphy, PhD, David M. Delong, MD, Franklin N. Tessler, MD and Mark A. Kliewer, MD

1 From the Department of Radiology, Albany Medical College, NY (M.E.T., M.E.M., F.N.T.); and Department of Radiology, Duke University School of Medicine, Durham, NC (M.E.M., D.M.D., M.A.K.). Received January 2, 2001; revision requested February 16; final revision received February 26, 2002; accepted March 25. Address correspondence to M.E.T., Department of Radiology, University of Pittsburgh Medical Center, 200 Lothrop St, Pittsburgh, PA 15213 (e-mail: tublinme@msx.upmc.edu).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
PURPOSE: To determine the effects of calculus size, composition, and technique (kilovolt and milliampere settings) on the conspicuity of renal calculi at unenhanced helical computed tomography (CT).

MATERIALS AND METHODS: The authors performed unenhanced CT of a phantom containing 188 renal calculi of varying size and chemical composition (brushite, cystine, struvite, weddellite, whewellite, and uric acid) at 24 combinations of four kilovolt (80–140 kV) and six milliampere (200–300 mA) levels. Two radiologists, who were unaware of the location and number of calculi, reviewed the CT images and recorded where stones were detected. These observations were compared with the known positions of calculi to generate true-positive and false-positive rates. Logistic regression analysis was performed to investigate the effects of stone size, composition, and technique and to generate probability estimates of detection. Interobserver agreement was estimated with {kappa} statistics.

RESULTS: Interobserver agreement was high: the mean {kappa} value for the two observers was 0.86. The conspicuity of stone fragments increased with increasing kilovolt and milliampere levels for all stone types. At the highest settings (140 kV and 300 mA), the detection threshold size (ie, the size of calculus that had a 50% probability of being detected) ranged from 0.81 mm + 0.03 (weddellite) to 1.3 mm + 0.1 (uric acid). Detection threshold size for each type of calculus increased up to 1.17-fold at lower kilovolt settings and up to 1.08-fold at lower milliampere settings.

CONCLUSION: The conspicuity of small renal calculi at CT increases with higher kilovolt and milliampere settings, with higher kilovolts being particularly important. Small uric acid calculi may be imperceptible, even with maximal CT technique.

© RSNA, 2002

Index terms: Computed tomography (CT), experimental studies, 81.12111, 81.12115 • Experimental study • Kidney, calculi, 81.811 • Phantoms


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The utility of unenhanced computed tomography (CT) for the evaluation of patients with renal colic is well accepted. Findings in numerous publications (1–4) in the radiology, emergency medicine, and urology literature have indicated the potential of CT to enable rapid identification of renal and ureteral calculi. Moreover, investigators in multiple reports (5–7) have stressed that CT enables detection of radiographically nonopaque calculi and other causes of flank pain besides urinary colic. Other publications (8–11) have largely been geared toward evaluating the effect of ancillary findings of ureteral calcification or obstruction and the effect of differing modes of image review on diagnostic accuracy.

Despite the almost universal acceptance of unenhanced CT for the evaluation of suspected urinary calculi, there has been, to our knowledge, no systematic investigation of the true diagnostic effectiveness of this approach, nor of the effect of differing technical settings on stone conspicuity. Surprisingly, there is no consensus on exactly what the optimal CT protocol should be for detection of small stones.

Investigators in a prior study (12) suggested that the use of decreased kilovolt levels might enhance urinary stone conspicuity. This effect may be influenced by calculus composition and background image noise, however. The purpose of the present study was to determine the effects of calculus size, composition, and technique (kilovolt and milliampere settings) on the conspicuity of renal calculi on unenhanced helical CT images.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Renal calculi were obtained from the Laboratory for Stone Research (Newton, Mass). A phantom model was constructed so that fragments of six stone types were dispersed in layers within a cylinder designed to have attenuation properties similar to those of the human abdomen wall and kidney (Fig 1). The inner surface of a cylindric 10-L Nalgene polypropylene carboy (Nalgene Labware, Rochester, NY) was coated with 1 cm of paraffin wax, leaving an open chamber area of about 400 cm2 in cross section. The wax had an attenuation of -40 HU, a value similar to that of fat in vivo. The carboy was then progressively filled with 1-cm transverse layers (approximately 400 mL each) of gelatin. The gelatin solution contained 3.5% gelatin powder, 0.1% sodium azide (as an antibiotic preservative), and 0.75% diatrizoate meglumine (RENO-60; Bracco Diagnostics, Princeton, NJ). Results of preliminary studies had revealed that this concentration of RENO-60–impregnated gelatin had an attenuation of 40 HU, which approximated the attenuation of kidneys in vivo. Eighteen thin metal wires were attached to the outer surface of the carboy along the long axis of the cylinder, perpendicular to the plane of gelatin layers. On transverse CT images of the phantom, these wires were visible in cross section as points and thus served to unambiguously orient the scanned image and provide reference points for a theoretical honeycomb grid of seven nonoverlapping circles.



View larger version (73K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1a. (a) Schematic diagram of phantom construction. (b) Transverse CT image of renal phantom shows stone clusters (arrows).

 


View larger version (124K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1b. (a) Schematic diagram of phantom construction. (b) Transverse CT image of renal phantom shows stone clusters (arrows).

 
Six stone types were studied: brushite (calcium acid phosphate dihydrate), cystine, struvite (magnesium ammonium phosphate hexahydrate), uric acid, weddellite (calcium oxalate dihydrate), and whewellite (calcium oxalate monohydrate). The density of each type of stone was determined by members of a commercial firm (Excalibur Mineral Company, Peekskill, NY) who measured the specific gravity of fragments by using a Berman balance. Calculi of each chemical type were crushed to obtain fragments in a weight range of 0.4–2.5 mg. The density measurements were used to transform the weight of each stone into a volume, which was then transformed into a diameter on the basis of an ideal sphere equation. Calculated diameters ranged from 0.64 to 1.39 mm.

A total of 188 fragments were individually weighed and photographed and then placed in the phantom, as follows. First, a layer of hot gelatin was poured into the carboy and allowed to fully cool and harden. Second, another layer of hot gelatin was poured into the carboy, but before this hardened, collections of zero to seven calculi of known weights were allocated into each of the seven segments (demarcated by the external wires) (Fig 1a). Third, calculi that had landed less than 5 mm apart were manually separated from one another. Fourth, the gelatin was allowed to fully cool and harden. Fifth, the final position of each calculus was observed and recorded by one of the authors (M.E.M.), both by hand and by obtaining a photograph. Finally, another layer of gelatin was poured over the stone layer to create a separation layer between stone layers. Altogether, eight layers of calculi were set up in the phantom, and each stone layer was divided into seven circular segments, resulting in a total of 56 segments. Each segment contained calculi of only a single chemical type. Several segments did not contain any calculi. The number of calculi in any one segment was random, but the overall number of calculi and the range of weights were comparable for each of the six types of calculi.

The completed phantom was scanned by using a helical CT scanner (Hi-Speed Advantage; GE Medical Systems, Milwaukee, Wis). Scanning was performed with 5-mm collimation and a pitch of 1, without overlapping reconstruction (Fig 1b). For each scan, the kilovolts were sequentially set at 80, 100, 120, or 140, and the milliamperes to 200, 220, 240, 260, 280, or 300, such that a total of 24 image sets were obtained with 24 different combinations of kilovolts and milliamperes.

Two radiologists (M.E.T., F.N.T.) evaluated the CT images at an independent console by using a cine loop; the window (400 HU) and level (20 HU) settings were kept constant for all viewings. To ensure unbiased observations, the radiologists were informed only that the phantom contained eight layers of calculi, that the calculi were distributed within seven circular segments per layer, that calculi were no closer than 4 mm to each other, and that some segments would be empty. The observers did not know the total number of calculi in the phantom, the number or type of calculi in any layer or segment, the maximal number that could occur in any one segment, or the number or position of empty segments. As each observer evaluated the image of a given layer, he recorded the positions of any observed calculi on a blank template. The template displayed an outline of the phantom, including the external wire landmarks to localize calculi within the seven grid segments. The radiologists were permitted to assess only one of the 24 sets of images per day to minimize possible short-term recollection of calculus arrangements. After each observer recorded his or her independent evaluations of a set of images, the images were viewed as a team, and a consensus evaluation was reached. Except where noted, the results presented in this article are based on the consensus evaluation, but in every case, a parallel analysis of the two independent evaluations yielded the same results.

The sites where stones were detected by the radiologists were compared with the true positions of calculi documented on maps and photographs. All observations were categorized as either a true observation of a calculus (a true hit), a failure to observe a calculus (a miss), or a false observation of a calculus (a false hit).

Statistical models were generated to assess the effects of stone type and size (diameter), kilovolt and milliampere settings, stone position, clustering on the detection of stones (true hits), and the rate of false-positive observations (false hits). A logistic regression model was used to produce probability estimates for the detection of a stone and test for the statistical significance of stone characteristics (type and size) and technical factors (kilovolts and milliamperes). For any specified combination of kilovolt and milliampere settings, the model defined the probability estimate for detecting a stone of a given type and size, including the size that had a 50% probability of being observed, which we termed the "detection threshold size." In further statistical analyses, a possible clustering effect was investigated by testing for a relationship between the number of stones in a location and the likelihood of detection. A Poisson model was generated to examine the probability of false-positive observations as a log linear function of stone characteristics, technical factors, and the presence of other stones observed in the area. The number of false observations were counted for each layer of the phantom and added. These data, then, represented an additive count of false signals within a volume, which can be characterized in a Poisson distribution. {kappa} statistics were calculated to estimate interobserver agreement. For all statistical models, pairwise interaction and quadratic terms were generated and tested for significance and magnitude of effect. Results of all statistical tests were considered to indicate a significant difference at a P value of .05 or less.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
The phantom contained 188 stone fragments of six stone types. The number of fragments of each stone type and the means and ranges of sizes are given in Table 1.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Weights, Densities, and Diameters of Calculi Used in the Phantom

 
Interobserver agreement between the two readers was high: the mean weighted {kappa} statistic was 0.86 (standard error, 0.02) across all observations ({kappa} value range, 0.77–0.96). {kappa} values can range from 1 (complete agreement) to 0 (chance agreement) to -1 (complete disagreement).

Results of logistic regression analysis demonstrated the statistical significance of stone type and size, kilovolt and milliampere settings, and layer (all P values < .001) on the conspicuity of stones in the phantom. In the full complement of interaction and quadratic terms tested, only three interaction terms attained statistical significance, and these were stone type by kilovolts (P = .008), stone type by size (P = .005), and kilovolts by milliamperes (P = .03) (Table 2). No significant clustering effect was found: The probability of observing a stone was not increased by considering its proximity to other observed stones. All statistically significant terms—including the three interaction terms—were included in the final model. The test results for the statistical significance of fit were not statistically significant (P = .18), which means there was no significant lack of fit (ie, the model fits the data).


View this table:
[in this window]
[in a new window]

  		TABLE 2. Effect of Kilovolts, Milliamperes, and Stone Type on the 50% Visibility Threshold

 
Weddellite stones were the most easily detected of all stone types, followed by whewellite, brushite, struvite, cystine, and uric acid. At the highest settings (140 kV and 300 mA), the detection threshold size (ie, the size of calculus that had a 50% probability of being detected) ranged from 0.81 mm + 0.03 (weddellite) to 1.3 mm + 0.1 (uric acid). To state this relationship another way, the 50% detection threshold size for a weddellite stone was 0.84 mm at 140 kV and 200 mA, while under these same conditions, there was only a 26% probability of detecting a 0.84-mm whewellite stone, a 13% probability of detecting a 0.84-mm brushite stone, a 5% probability of detecting a 0.84-mm struvite stone, and a 3% probability of detecting a 0.84-mm cystine stone. Uric acid stones (<1 mm) were so poorly detected that there were an insufficient number of true hits to estimate a regression coefficient within the logistic model (Table 3). The largest uric acid stones could only be detected at the highest kilovolt and milliampere settings. A full summary of the probability of detection of the various stone types, as a function of size, is displayed graphically in Figure 2.


View this table:
[in this window]
[in a new window]

  		TABLE 3. Visibility Threshold Sizes by Stone Type at 140 kV and 200 mA

 


View larger version (32K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2. Graph indicates visibility of renal calculi of various diameters imaged at 140 kV and 200 mA. A logistic regression model was used to produce probability estimates for the detection of stones of various sizes. The 95% CI of the threshold visibility of each type of stone (ie, the stone size with a 50% probability of being seen) is indicated by the horizontal lines. The model was based on six independent parameters: stone type, stone size, kilovolt setting, milliampere setting, stone position, and clustering. Br = brushite, Cy = cystine, St = struvite, Ur = uric acid, We = weddellite, Wh = whewellite.

 
The regression model indicated that urinary stones became increasingly more conspicuous at increasing milliampere and kilovolt levels and that this effect was uniform among all types of stones (excluding uric acid) (Table 4). Stated another way, if the probability of detecting a stone is 50% at 200 mA, then an increase to 220 mA increases the probability of detecting that stone to 57%, and an increase to 300 mA increases the probability to 77%. The effect of kilovolts was even greater than that of milliamperes. For example, if the probability of detecting a stone is 50% at 80 kV, then an increase to 100 kV increases the probability of detecting that stone to 78%, and an increase to 140 kV increases the probability to 93%. Figure 3 summarizes the effects of milliampere and kilovolt levels by showing the threshold size of calculi that could be detected at each combination of CT settings. Figure 3 also demonstrates that the detection threshold size was reduced (ie, detection of small calculi became easier) with increasing milliampere values and with increasing kilovolt values within the tested range. Detection threshold sizes were substantially reduced with each increase of kilovolts in the range of 80–140 kV and were also substantially lower when milliamperes were within the range of 260–300 mA, as compared with 200–240 mA. Detection threshold size for each type of calculus increased up to 1.17-fold at lower kilovolt settings and up to 1.08-fold at lower milliampere settings. This enhanced conspicuity, with increased levels of both milliamperes and kilovolts, was uniform among the five types of calculi (excluding uric acid).


View this table:
[in this window]
[in a new window]

  		TABLE 4. Relative Change in Visibility Thresholds at Nonoptimal CT Settings

 


View larger version (29K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3. Graph indicates visibility threshold sizes of renal calculi at various combinations of kilovolt and milliampere settings. A logistic regression model was used to produce estimates of the 50% visibility threshold sizes for all five types of calculi at each combination of CT settings. Because the model indicated that the effects of kilovolt and milliampere levels were uniform among the five types of calculi, the threshold sizes of the five calculi types were averaged, normalized to the threshold sizes at the 140 kV and 300 mA combination, and then plotted as points to highlight the general effects of kilovolts and milliamperes.

 
The probability of detecting a stone when it was not present (false hit) depended on the kilovolt setting (P < .001) and the presence of other stones in proximity (P = .006) but not on milliampere settings. The probability of making a false observation was greater at lower kilovolt settings (80–100 kV) (Fig 4), although it should be noted that this effect was entirely the result of the unusual and unexpected number of false-positive observations at 100 kV. The probability of a false-positive observation increased 3.2-fold when true stones were observed in the same segment. In other words, an observer was more likely to falsely detect stones if true stones were nearby.



View larger version (26K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4. False-positive observations at various kilovolt and milliampere settings. False-positive observations are those in which the recorded position did not correspond to the position of an actual calculus. The number of false-positive observations seen in the entire phantom with a given combination of CT settings was divided by the total number of notations (ie, both true and false), and this ratio was expressed as a percentage. {square} = 300 mA, {blacksquare} = 280 mA, {diamond} = 260 mA, {diamondsuit} = 240 mA, {circ} = 220 mA, {bullet} = 200 mA.

 
Although the final statistical model included all statistically significant terms, the actual effect of the three interaction terms on the predicted probabilities of seeing the various stones was modest in the range of practical interest. For example, when the interaction term of stone type by stone size (ie, different slopes for stone size among the stone types) was included in the full model, the estimated probability of stone detection near the threshold of visibility changed from 52% to 53% for brushite stones and from 50% to 51% for cystine stones. Changes for the other stone types were similar. Likewise, when the statistically significant interaction term of stone type by kilovolts (P < .001) was included in the model, the predicted probability of detection of brushite stones changed only from 52% to 50% at 140 kV.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Investigators of numerous studies (1–4) have documented the superior sensitivity and specificity of unenhanced helical CT when compared with those of intravenous urography for the detection of renal and ureteral calculi. Although this finding has led to the rapid adoption of unenhanced CT techniques, there have been no studies, to our knowledge, designed to directly address the effect of CT technique on the conspicuity of differing types of small calculi. Instead, researchers have tended to concentrate on the relative benefits of differing modes of image review and ancillary signs of ureteral obstruction in cases in which the diagnosis of a ureteral calculus is in doubt (8–11).

Part of the theoretical framework for the present phantom study was laid by the authors of several articles published in the early CT literature (13–19) that showed that differences in attenuation of large calculi are dependent on their composition. In our study, we used calculi that were too small to allow region of interest measurement; nevertheless, the relationship that we found between detection threshold size and chemical composition (Fig 2) is in broad agreement with the results of early CT studies in which attenuation of large calculi was measured directly (Table 5). We confirmed the expectation that the types of calculi with highest attenuation also had the lowest detection threshold sizes. The only exception to this inverse relationship is that the order of conspicuity (brushite more conspicuous than or equal in conspicuity to weddellite and whewellite) is in disagreement with our finding regarding detection threshold sizes (weddellite and whewellite less than brushite).


View this table:
[in this window]
[in a new window]

  		TABLE 5. Comparison of Stone Conspicuity between Studies

 
Findings at prior investigations of the utility of dual-kilovolt CT for pulmonary nodule (20) and renal calculus (17,18) characterization led us to believe that small-calculus conspicuity would increase with decreases in kilovolts. Most of the interactions in soft tissues result from Compton scattering at the high kilovolt level typically used for CT. By decreasing kilovolts closer to the k edge of calcium, photoelectric interactions assume greater importance, and thus, beam attenuation of calcium-containing calculi increases. Indeed, both Mitcheson et al (17) and Mostafavi et al (18) concluded that different types of calculi could be distinguished by comparing attenuations at 80 versus 120 kV. Specifically, both groups reported three salient findings: (a) All calculi had a higher attenuation at 80 versus 120 kV; (b) this difference in attenuation was more dramatic for calcium-containing calculi (eg, brushite and weddellite) than for noncalcified calculi (eg, uric acid and cystine); and (c) therefore, the kilovolt-dependent change in attenuation (ie, Hounsfield units at 80 kV minus Hounsfield units at 120 kV) of any particular stone could be used to help determine its chemical composition. Our results failed to match the expectation of these previous reports. First, we found that small calculi were best detected at the highest kilovolt setting (140), rather than the lowest. Second, we found that the five types of calculi did not differ with respect to the relationship between kilovolts and detection threshold size.

In explaining this discrepancy, it should first be noted that investigators in prior studies (17,18) purposely used calculi that were orders of magnitude larger than our fragments. These large calculi were appropriately chosen so that region of interest measurements could be obtained unambigously from a large uniform area. In addition, assessment of differences in attenuation of the surrounding medium relative to the calculus (ie, contrast) was not a goal of this work. We suspect that the decreased conspicuity of small calculi with decreased kilovolt levels in the present study resulted from increased background noise: Any increase in calculus attenuation that might have occurred with decreases in kilovolts was obscured by image noise. The clustering of false-positive observations at 100 kV remains unexplained; the greater false-positive rate with proximity to true stones makes more clinical and intuitive sense.

Although maximal CT technique improves stone conspicuity, the universal application of high kilovolt and milliampere levels in the clinical setting is neither practical nor safe because of radiation dose and tube cooling concerns. The results of our study indicate that although stone conspicuity decreases with decreasing kilovolt and milliampere levels, detection thresholds are more sensitive to changes in kilovolts. Therefore, a high-kilovolt, low-milliampere technique can be adopted without substantial loss of diagnostic effectiveness but with a large reduction in radiation dose.

Uric acid calculi may not be perceptible, even if they are relatively large and maximal CT technique is used. This was an unexpected finding, considering an earlier report (7) of the effectiveness of CT in the depiction of radiographically nonopaque calculi. In explaining this discrepancy, it should be noted that the calculi evaluated in that study were much larger than those used in our phantom model. Results of the present study suggest that ancillary findings of ureteral obstruction may be more important signs of stone disease for those patients with a history of uric acid calculi, because direct visualization of small stone fragments may not be possible. Although it might be argued that small calculi are not clinically important (many of these small calculi may pass without intervention), confident documentation of stone disease may obviate further work-up of hematuria and/or flank pain.

The use of a phantom model does not allow exact replication of clinical conditions; thus, the resulting limitations of our study deserve mentioning. In contrast to clinical imaging, our phantom facilitated detection of calculi by providing a uniform background density and by eliminating motion artifacts. On the other hand, our phantom was constructed of a gelatin matrix and peripheral wax layer, which closely resemble the size and opacity of normal abdominal viscera and fat. Also, image reading was performed without specific knowledge of the number, position, or composition of calculi. Nevertheless, our results may need clinical verification.

Practical application: In conclusion, the results of the present study demonstrate that conspicuity of small calculi at unenhanced CT is very much dependent on technique. While decreasing kilovolt levels may theoretically increase calculus attenuation, this benefit is likely offset by increased noise. Although the results of this study support the use of high kilovolt and milliampere levels for increasing calculus conspicuity, a more practical approach in actual clinical practice would be to maintain high kilovolt settings and low milliampere settings (eg, 140 kV and 200 mA) to decrease radiation dose and obviate tube-cooling problems. The results of this study also suggest that despite earlier reports to the contrary, small uric acid calculi are imperceptible with even maximal technique. Those patients with a history suggestive of uric acid calculi might benefit from imaging with intravenous contrast material and close clinical follow-up.


   FOOTNOTES
 
Author contributions: Guarantor of integrity of entire study, M.E.T.; study concepts and design, all authors; literature research, M.E.T., M.E.M.; experimental studies, M.E.T., M.E.M.; data acquisition, M.E.T., M.E.M.; data analysis/interpretation, all authors; statistical analysis, M.E.T., D.M.D., M.A.K.; manuscript preparation, all authors; manuscript definition of intellectual content and editing, M.E.T., M.E.M., M.A.K.; manuscript revision/review and final version approval, all authors.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Chen MYM, Zagoria RJ. Can unenhanced helical computed tomography replace excretory urography for evaluation of patients with acute urinary tract colic? J Emerg Med 1999; 17:299-303.[CrossRef][Medline]
  2. Fielding JR, Steele G, Fox LA, Heller H, Loughlin KR. Spiral computerized tomography in the evaluation of acute flank pain: a replacement for excretory urography. J Urol 1997; 15:2071-2073.
  3. Smith RC, Rosenfield AT, Choe KA, et al. Acute flank pain: comparison of non-contrast-enhanced CT and excretory urography. Radiology 1995; 194:789-794.[Abstract/Free Full Text]
  4. Smith RC, Verga M, McCarthy S, Rosenfield AT. Diagnosis of acute flank pain: value of unenhanced helical CT. AJR Am J Roentgenol 1996; 166:97-101.[Abstract/Free Full Text]
  5. Mindelzun RE, Jeffrey RB. Unenhanced helical CT for evaluating acute abdominal pain: a little more cost, a lot more information. Radiology 1997; 205:43-47.[Free Full Text]
  6. Chen MY, Zagoria RJ, Saunders HS, Dyer RB. Trends in the use of unenhanced helical CT for acute urinary colic. AJR Am J Roentgenol 1999; 173:1447-1450.[Abstract]
  7. Federle MP, McAninch JW, Kaiser JA, Goodman PC, Roberts J, Mall JC. Computed tomography of urinary calculi. AJR Am J Roentgenol 1981; 136:255-258.[Abstract/Free Full Text]
  8. Sommer FG, Jeffrey RB, Jr, Rubin GD, et al. Detection of ureteral calculi in patients with suspected renal colic: value of reformatted noncontrast helical CT. AJR Am J Roentgenol 1995; 165:509-513.[Abstract/Free Full Text]
  9. Katz DS, Lane MJ, Sommer FG. Unenhanced helical CT of ureteral stones: incidence of associated urinary tract findings. AJR Am J Roentgenol 1996; 166:1319-1322.[Abstract/Free Full Text]
  10. Kawashima A, Sandler CM, Boridy IC, Takahashi N, Benson GS, Goldman SM. Unenhanced helical CT of ureterolithiasis: value of the tissue rim sign. AJR Am J Roentgenol 1997; 168:997-1000.[Abstract/Free Full Text]
  11. Heneghan JP, Dalrymple NC, Verga M, Rose AT, Smith RC. Soft tissue "rim" sign in the diagnosis of ureteral calculi with the use of unenhanced helical CT. Radiology 1997; 202:707-711.
  12. Chu G, Rosenfield AT, Anderson K, Scout L, Smith RC. Sensitivity and value of digital CT scout radiography for detecting ureteral stones in patients with ureterolithiasis diagnosed on unenhanced CT. AJR Am J Roentgenol 1999; 173:417-423.[Abstract/Free Full Text]
  13. Dretler SP. Stone fragility: a new therapeutic distinction. J Urol 1988; 139:1124-1127.[Medline]
  14. Herremans D, Vandeursen H, Pittomvils G, et al. In vitro analysis of urinary calculi: type differentiation using computed tomography and bone densitometry. Br J Urol 1993; 72(5 pt 1):544-548.[Medline]
  15. Hillman BJ, Drach GW, Tracey P, Gaines JA. Computed tomographic analysis of renal calculi. AJR Am J Roentgenol 1984; 142:549-552.[Abstract/Free Full Text]
  16. Kuwahara M, Kageyama S, Kurosu S, Orikasa S. Computed tomography and composition of renal calculi. Urol Res 1984; 12:111-113.[Medline]
  17. Mitcheson HD, Zamenhof RG, Bankoff MS, Prien EL. Determination of the chemical composition of urinary calculi by computerized tomography. J Urol 1983; 130:814-819.[Medline]
  18. Mostafavi MR, Ernst RD, Saltzman B. Accurate determination of chemical composition of urinary calculi by spiral computerized tomography. J Urol 1998; 159:673-675.[CrossRef][Medline]
  19. Newhouse JH, Prien EL, Amis ES, Jr, Dretler SP, Pfister RC. Computed tomographic analysis of urinary calculi. AJR Am J Roentgenol 1984; 142:545-548.[Abstract/Free Full Text]
  20. Bhalla M, Shepard JA, Nakamura K, Kazerooni EA. Dual KV CT to detect calcification solitary pulmonary nodule. J Comput Assist Tomogr 1995; 19:44-47.[Medline]



This article has been cited by other articles:


Home page
 Am. J. Roentgenol.Home page
B. M. Yeh, J. A. Shepherd, Z. J. Wang, H. Seong Teh, R. P. Hartman, and S. Prevrhal
Dual-Energy and Low-kVp CT in the Abdomen
Am. J. Roentgenol., July 1, 2009; 193(1): 47 - 54.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Roentgenol.Home page
E. K. Paulson, C. Weaver, L. M. Ho, L. Martin, J. Li, J. Darsie, and D. P. Frush
Conventional and Reduced Radiation Dose of 16-MDCT for Detection of Nephrolithiasis and Ureterolithiasis
Am. J. Roentgenol., January 1, 2008; 190(1): 151 - 157.
[Abstract] [Full Text] [PDF]

Home page
 RadiologyHome page
W. C. Chan, B. N. Joe, F. V. Coakley, E. L. Prien Jr, R. G. Gould, S. Prevrhal, W. C. Barber, K. S. Kirkwood, A. Qayyum, and B. M. Yeh
Gallstone Detection at CT in Vitro: Effect of Peak Voltage Setting.
Radiology, November 1, 2006; 241(2): 546 - 553.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2251010242v1
225/1/91    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
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 Tublin, M. E.
Right arrow Articles by Kliewer, M. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Tublin, M. E.
Right arrow Articles by Kliewer, M. A.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
RADIOLOGY RADIOGRAPHICS RSNA JOURNALS ONLINE