In Vivo Evaluation of the Chemical Composition of Urinary Stones Using Dual-Energy CT

Transcription

1 Genitourinary Imaging Original Research Manglaviti et al. Evaluation of Urinary Stones With Dual-Energy CT Genitourinary Imaging Original Research Giuseppina Manglaviti 1 Silvia Tresoldi 2 Chiara Stefania Guerrer 3 Giovanni Di Leo 4 Emanuele Montanari 3,5 Francesco Sardanelli 4,6 Gianpaolo Cornalba 2,7 Manglaviti G, Tresoldi S, Guerrer CS, et al. Keywords: crystallography, dual-energy CT, dual-source CT, kidney stones, stone composition DOI: /AJR Received June 24, 2010; accepted after revision December 19, Dipartimento di Radiologia, Centro Diagnostico Italiano, Milan, Italy. 2 Dipartimento di Radiologia Diagnostica ed Interventistica, Azienda Ospedaliera San Paolo, Via di Rudinì 8, Milan, Italy. Address correspondence to S. Tresoldi (silvia.3soldi@gmail.com). 3 Urologia, Dipartimento di Chirurgia, Azienda Ospedaliera San Paolo, Milan, Italy. 4 Unità di Radiologia, IRCCS Policlinico San Donato, San Donato M.se (MI), Italy. 5 Dipartimento di Medicina, Chirurgia e Odontoiatria, Università degli Studi di Milano, Milan, Italy. 6 Dipartimento di Scienze Medico-Chirurgiche, Università degli Studi di Milano, Milan, Italy. 7 Dipartimento di Scienze e Tecnologie Biomediche, Sezione di Scienze Radiologiche, Università degli Studi di Milano, Milan, Italy. WEB This is a Web exclusive article. AJR 2011; 197:W76 W X/11/1971 W76 American Roentgen Ray Society In Vivo Evaluation of the Chemical Composition of Urinary Stones Using Dual-Energy CT OBJECTIVE. The purpose of this article is to evaluate in vivo the chemical composition of urinary stones using dual-source and dual-energy CT, with crystallography as the reference standard. MATERIALS AND METHODS. Forty patients (mean [± SD] age, 49 ± 17 years) with known or suspected nephrolithiasis underwent unenhanced abdominal CT for urinary tract evaluation using a dual-energy technique (tube voltages, 140 and 80 kvp). For each stone 5 mm or larger in diameter, we evaluated the site, diameter, CT density, surface (smooth vs rough), and stone composition. Patients were treated with extracorporeal shock wave lithotripsy (n = 34), percutaneous nephrolithotomy (n = 4), or therapeutic ureterorenoscopy (n = 2). Collected stones underwent crystallography, and the agreement with the results of dual-energy CT was calculated with the Cohen kappa coefficient. The correlation among stone composition, diameter, and CT density was estimated using the Kruskal-Wallis test. RESULTS. Thirty-one patients had a single stone and nine had multiple stones, for a total of 49 stones. Forty-five stones were in the kidneys, and four were in the ureters; 23 had a smooth surface and 26 had a rough surface. The mean stone diameter was 12 ± 6 mm; mean CT density was 783 ± 274 HU. According to crystallography, stone composition was as follows: 33 were calcium oxalate, seven were cystine, four were uric acid, and five were of mixed composition. Dual-energy CT failed to identify four stones with mixed composition, resulting in substantial agreement between dual-energy CT and crystallography (Cohen κ = 0.684). Stone composition was not correlated with either stone diameter (p = 0.920) or stone CT density (p = 0.185). CONCLUSION. CT showed excellent accuracy in classifying urinary stone chemical composition, except for uric acid hydroxyapatite mixed stones. U rolithiasis is a common cause of acute and chronic pain and, consequently, of outpatient visits or hospitalization. Its incidence in the white population is about 5 12% [1]. The risk of recurrence of a calcium oxalate stone after treatment is about 10% at 1 year and 50% at 10 years [1]. Many factors contribute to the etiopathogenesis of urolithiasis: diet (increased animal proteins and refined sugar and salt consumption), sex (the disease is more common among men than among woman), age (peak incidence occurs at ages years), low fluid intake, genetic factors, and geographic factors [2]. Urinary stones can be considered the consequence of crystallization and aggregation of highly concentrated urinary components. There are two main categories of urinary stones: calcium stones and noncalcium stones. The most common stone composition is calcium oxalate (40 60%), followed by uric acid (5 10%), hydroxyapatite (2 4%), and cystine (1 3%) [1]. Lithiasis can involve each tract of the excretory system and can be associated with metabolic or anatomic alterations. Clinical presentation is variable, from occasional abdominal pain to renal failure (about 3% of renal failures are a direct consequence of lithiasis) [3, 4]. Moreover, clinical manifestations and renal damage are not always related to the stone size. Recently, MDCT without contrast agent has supplanted excretory urography, which was previously considered as the reference standard, for the evaluation of urolithiasis [5, 6]. In fact, MDCT provides accurate determination of stone location, size, number, shape, and CT density; can detect the presence of hydroureteronephrosis or inflammation; and also permits the creation of 3D reconstructions. Dual-energy CT also allows the evaluation of urinary stones chemical composition, W76 AJR:197, July 2011

2 Evaluation of Urinary Stones With Dual-Energy CT which is clinically relevant for treatment planning [7 11]. In fact, it permits both early drug therapy to dissolve uric acid stones and medical prophylactic measures to prevent the stones recurrence. Moreover, knowledge about the stone s chemical composition guides the choice of the appropriate treatment approach among noninvasive techniques (e.g., extracorporeal shock wave lithotripsy), minimally invasive techniques (e.g., percutaneous nephrolithotomy or therapeutic ureterorenoscopy), or invasive techniques (e.g., open surgery) [12, 13]. The purpose of our study was to evaluate in vivo the chemical composition of urinary stones using dual-energy CT, with stone crystallography as the reference standard. Materials and Methods Population This retrospective study was approved by the local institutional review board. Between March and July 2008, 52 patients with known or suspected urolithiasis underwent dual-energy CT examination for evaluation of the urinary tract. Forty patients (32 men and eight women; mean [± SD] TABLE 1: Classification of Renal Stones According to Chemical Composition as Detectable by Dual- Energy CT Stone Type, Stone Subtype Calcium stones, calcium oxalate Noncalcium stones Uric acid Cystine Hydroxyapatite (infective stones) Mixed stones Cystine and hydroxyapatite Uric acid and hydroxyapatite age, 49 ± 17 years; age range, years) subsequently treated with extracorporeal shock wave lithotripsy (n = 34), percutaneous nephrolithotomy (n = 4), or therapeutic ureterorenoscopy (n = 2) entered the analysis. For these 40 patients, crystallography of the collected stones was performed. The remaining 12 patients were excluded because of the lack of the reference standard. MDCT Protocol and Image Analysis All examinations were performed with a dualsource MDCT (Somatom Definition, Siemens Healthcare). The imaging protocol consisted of an unenhanced spiral scan with a single x-ray tube on the whole abdomen acquired along the craniocaudal direction with the patient in the supine position, followed by a dual-energy acquisition focused on the site of the stone previously detected. Technical parameters for the unenhanced abdominal scan were as follows: tube voltage, 120 kvp; reference tube current, 250 ma with automatic exposure control; pitch factor, 0.9:1; acquisition slice thickness, 5 mm; reconstruction slice thickness, 1.5 mm; reconstruction increment, 1.5 mm; gantry rotation time, 0.5 second; filter kernel, B30f (mediumsmooth); field of view, 35 cm; and detector configuration, mm. Technical parameters for the dual-energy scan were as follows: tube voltage, 80 kvp and 140 kvp; reference tube current, 96 ma and 400 ma with automatic exposure control; pitch factor, 0.7:1; acquisition slice thickness, 5 mm; reconstruction slice thickness, 0.75 mm; reconstruction increment, 0.5 mm; gantry rotation time, 0.5 second; filter kernel, B30f (mediumsmooth); field of view, 26 cm; and detector con- A Fig. 1 Calcium stone and uric acid stone. A and B, Left panels show axial CT images centered on stones in left kidney. Calcium stone (A) in 26-year-old man is oxalate and oval shaped, with smooth surface, maximum diameter of 1.6 cm, and mean CT density greater than 1000 HU (evaluated with region of interest in area > 50% of stone surface). Uric acid stone (B) in 53-yearold man has irregular shape, rough surface, maximum diameter of 1.5 cm, and CT density of 574 HU. Right panels show elaboration with Kidney Stones software (Siemens Healthcare), including three views of stone in sagittal, coronal, and axial planes and settings dialog. In diagram, four stone types (calcium oxalate, hydroxyapatite, cystine, and uric acid) are described by four small circles (CT density at 140 kvp on x-axis and CT density at 80 kvp on y-axis). Blue line splits plane in two parts; if line is left in its default position (Siemens Standard) and orientation (A), lower part contains noncalcium stones (hydroxyapatite, cystine, and uric acid) and upper part contains calcium stones (calcium oxalate). Stone shown in three views is automatically painted red by software if its CT density lies in lower part of diagram (i.e., under line) or in blue if its CT density lies in upper part of diagram. When parameters are changed, blue line varies its orientation, and, consequently, software switches color of stone in three views. For example, in B, line orientation has been changed to easily differentiate, in three views, uric acid stones (red) from cystine, calcium oxalate, and hydroxyapatite (blue). (Figure 1 continues on next page) AJR:197, July 2011 W77

3 Manglaviti et al. Fig. 1 (continued) Calcium stone and uric acid stone. A and B, Left panels show axial CT images centered on stones in left kidney. Calcium stone (A) in 26-year-old man is oxalate and oval shaped, with smooth surface, maximum diameter of 1.6 cm, and mean CT density greater than 1000 HU (evaluated with region of interest in area > 50% of stone surface). Uric acid stone (B) in 53-yearold man has irregular shape, rough surface, maximum diameter of 1.5 cm, and CT density of 574 HU. Right panels show elaboration with Kidney Stones software (Siemens Healthcare), including three views of stone in sagittal, coronal, and axial planes and settings dialog. In diagram, four stone types (calcium oxalate, hydroxyapatite, cystine, and uric acid) are described by four small circles (CT density at 140 kvp on x-axis and CT density at 80 kvp on y-axis). Blue line splits plane in two parts; if line is left in its default position (Siemens Standard) and orientation (A), lower part contains noncalcium stones (hydroxyapatite, cystine, and uric acid) and upper part contains calcium stones (calcium oxalate). Stone shown in three views is automatically painted red by software if its CT density lies in lower part of diagram (i.e., under line) or in blue if its CT density lies in upper part of diagram. When parameters are changed, blue line varies its orientation, and, consequently, software switches color of stone in three views. For example, in B, line orientation has been changed to easily differentiate, in three views, uric acid stones (red) from cystine, calcium oxalate, and hydroxyapatite (blue). figuration, mm. Images acquired with the dual-energy modality were postprocessed using a dedicated remote workstation (Leonardo, Siemens Healthcare) and dedicated software (Syngo Dual Energy Viewer, Siemens Healthcare) for the evaluation of the stone chemical composition (Table 1). Images were evaluated in consensus by two radiologists with 2 and 30 years of experience in urogenital radiology. All the examinations were visualized on the axial, coronal, and sagittal planes, and urinary stone CT density was measured using a region of interest smaller than 50% of the maximal diameter of each stone. For each patient, we evaluated the number, location (kidney, ureter, or bladder), maximal diameter, and CT density (expressed in Hounsfield units [HU]) of stones. Using 3D reconstructions, we also evaluated the stone surface, distinguishing between smooth or rough surface. Referring to the manufacturer s standard settings, the software displayed calcium stones in blue and noncalcium stones in red. By varying these settings, we were also able to differentiate three types TABLE 2: Characteristics of 49 Stones in 40 Patients Evaluated With Dual-Energy CT Stone Composition at Dual-Energy CT of stones among noncalcium stones uric acid, hydroxyapatite, and cystine (Fig. 1). Stones smaller than 5 mm in diameter were excluded from analysis, because such stones usually are ejected spontaneously and do not require treatment. Statistical Analysis Continuous variables were expressed as mean ± SD, whereas categoric variables were expressed as frequencies or percentages. Agreement between dual-energy CT and crystallography was estimated Diameter (mm) CT Density (HU) Surface Site < 1000 > 1000 Smooth Rough Kidney Ureter Bladder Calcium oxalate 15 (31) 16 (33) 2 (4) 20 (41) 13 (27) 17 (35) 16 (33) 29 (60) 4 (8) 0 (0) 33 (68) Cystine 3 (6) 3 (6) 1 (2) 7 (14) 0 (0) 5 (10) 2 (4) 7 (14) 0 (0) 0 (0) 7 (14) Uric acid 0 (0) 3 (6) 1 (2) 4 (8) 0 (0) 0 (0) 4 (8) 4 (8) 0 (0) 0 (0) 4 (8) Hydroxyapatite 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) Mixed stones 1 (2) 3 (6) 1 (2) 5 (10) 0 (0) 1 (2) 4 (8) 5 (10) 0 (0) 0 (0) 5 (10) Total 19 (39) 25 (51) 5 (10) 36 (73) 13 (27) 23 (47) 26 (53) 45 (92) 4 (8) 0 (0) 49 (100) Note Data are no. (%) of stones. Total B W78 AJR:197, July 2011

4 Evaluation of Urinary Stones With Dual-Energy CT Fig year-old man with single mixed (uric acid and hydroxyapatite) stone correctly classified by dual-energy CT. A, Standard CT views are shown. Left panel shows axial image, and right panel shows elaboration with Kidney Stones software (Siemens Healthcare), including colored views of stone in sagittal, coronal, and axial planes and dual-energy diagram. Dendritic stone (with diameters of 3 and 4 cm and mean CT density of 530 HU) mostly composed of uric acid is easily seen because it is mostly colored in red and only uric acid circle lies under diagram line. B, 3D view shows stone characterized by irregular shape and rough surface. Patient was successfully treated with percutaneous nephrolithotomy. using the Cohen kappa coefficient [14]. The correlation among stone composition at crystallography, the diameter, and the CT density at dual-energy CT was estimated using the Kruskal-Wallis test. The effect of the pretreatment CT density and of the stone surface on the treatment efficacy was evaluated with the chi-square test. All calculations were performed using SPSS (version 17.0, SPSS). Results Of 40 patients, 31 had a single stone and nine had multiple stones, for a total of 49 stones, the characteristics of which are summarized in Table 2. The mean stone size was 12 ± 6 mm (range, 6 40 mm), and the mean stone CT density was 783 ± 274 HU (range, HU). Forty-five stones (92%) were located in the kidneys (including two patients with dendritic stones) and four (8%) were in the ureters; 23 (47%) stones had a smooth surface, and 26 (53%) had a rough surface. According to dual-energy CT, stones were predicted to be composed of calcium oxalate (n = 33), cystine (n = 7), uric acid (n = 4), and mixed composition (n = 5). None of the stones was predicted to be composed purely by hydroxyapatite. Of the five stones with mixed composition, one was predicted to be composed of uric acid and hydroxyapatite, and four were predicted to be hydroxyapatite and cystine. Table 3 shows the agreement between dual-energy CT and crystallography (Cohen κ = 0.684, substantial agreement). Dual-energy CT failed to correctly identify the chemical composition of four stones; all of them had a mixed composition that was A B AJR:197, July 2011 W79

5 Manglaviti et al. determined to be uric acid and hydroxyapatite by crystallography but was misclassified as cystine and hydroxyapatite by dual-energy CT. In one case of a mixed stone (uric acid and hydroxyapatite) with a diameter larger than 2 cm, the two techniques were in agreement (Fig. 2). Thirty-four patients underwent extracorporeal shock wave lithotripsy, four patients underwent percutaneous nephrolithotomy, and two patients underwent therapeutic ureterorenoscopy. Table 4 shows the summarized criteria used by urologists in choosing patient treatment. Patients with stones 20 mm or larger in diameter were not suitable for extracorporeal shock wave lithotripsy and underwent percutaneous nephrolithotomy or therapeutic ureterorenoscopy. The patients who underwent percutaneous nephrolithotomy were two patients with dendritic stones and two with cystinuria; the patients who underwent therapeutic ureterorenoscopy were one patient with a pelvic single kidney and cystinuria and one with a ureteral stone. Twenty-four (71%) patients who underwent extracorporeal shock wave lithotripsy had a good response to treatment consisting of fragmentation and expulsion of the stones. In 10 (29%) patients who underwent extracorporeal shock wave lithotripsy, no response such as rupture or fragment clearance was observed, making necessary another treatment that was chosen according to the stone site: percutaneous nephrolithotomy for kidney stones (five patients) and ureterorenoscopy for ureteral or renal pelvis stones (five patients) (Fig. 3). All these patients had oxalate stones with a smooth surface and CT density greater than 1200 HU; nine of the stones were larger than 1 cm. Table 5 summarizes extracorporeal shock wave lithotripsy efficacy in stone fragmentation based on CT density and stone surface in the 34 patients treated. Correlation of the stone composition at crystallography with diameter at dual-energy CT (p = 0.920, Kruskal-Wallis test) or density at dual-energy CT (p = 0.185, Kruskal-Wallis test) was not statistically significant. Discussion At present, because of its high sensitivity and temporal resolution and because the use of IV contrast material and bowel preparation is not necessary, MDCT has replaced excretory urography. This technique permits submillimetric evaluation of the size and site of stones but cannot evaluate their chemical composition. Several studies tried to predict stone TABLE 3: Agreement Between Dual-Energy CT and Crystallography in Evaluation of Stone Chemical Composition Dual-Energy CT Calcium Oxalate Cystine Uric Acid composition using CT density measurements (Hounsfield units) in vivo but were only able to differentiate uric acid from non uric acid stones, both in vivo and in vitro [15, 16]. A high accuracy in detecting stone composition in vitro was described by Bellin et al. [15], whereas Zarse et al. [16] found that high-resolution CT can identify an attenuation value characteristic of each kind of stone using the right window level to localize homogeneous regions inside the stones. Even if this CT approach can be helpful, it is not well established enough to be used in clinical practice. Dual-energy CT has been shown to be very effective for characterizing urinary stones. However, in contrast to studies that examined stones ex vivo [8 10, 17], we tried to predict stone composition in vivo (i.e., before treatment). Because uric acid stones are made of low-molecular-weight elements (hydrogen, carbon, nitrogen, and oxygen), their Crystallography Cystine and Hydroxyapatite Uric Acid and Hydroxyapatite Calcium oxalate Cystine Uric acid Cystine and hydroxyapatite Uric acid and hydroxyapatite Total Note Cohen kappa = 0.684, indicating substantial agreement [14]. Total x-ray attenuation properties at high and low voltages are different from those of other stone types (composed of calcium oxalate, hydroxyapatite, or cystine), which are made of high-molecular-weight elements (phosphorus, calcium, and sulfur). As a consequence, uric acid stones have a higher Hounsfield unit value at higher voltages, whereas other kinds of stones (calcified) have a higher Hounsfield unit value at lower voltages. Some studies in the literature determined the accuracy of dual-energy CT in discriminating uric acid stones from non uric acid stones using an ex-vivo model with human stones inserted in porcine kidneys; in all cases, crystallography confirmed the main chemical composition [7]. Other studies confirmed a high accuracy of dual-energy CT in differentiating in vitro uric acid stones from non uric acid stones [10, 11]. Our study, in agreement with the literature, showed high TABLE 4: Potential Treatment Strategies Based on Stone Size and Chemical Composition Stone Size (mm), Stone Composition Treatment < 5, All No treatment 5 15 Uric acid Medical treatment and extracorporeal shock wave lithotripsy Non uric acid Extracorporeal shock wave lithotripsy Uric acid Medical treatment and extracorporeal shock wave lithotripsy Calcium oxalate (dihydrate) Extracorporeal shock wave lithotripsy Hydroxyapatite Extracorporeal shock wave lithotripsy Cystine Percutaneous nephrolithotomy or therapeutic ureterorenoscopy Calcium oxalate (monohydrate) Percutaneous nephrolithotomy or therapeutic ureterorenoscopy > 20, All Percutaneous nephrolithotomy or therapeutic ureterorenoscopy Note Open surgery is the chosen treatment in case of anatomic anomalies, complex calculosis, ectopic kidney, or unsuccessful extracorporeal shock wave lithotripsy or percutaneous nephrolithotomy. W80 AJR:197, July 2011

6 Evaluation of Urinary Stones With Dual-Energy CT Fig year-old man with single oxalate stone in left ureter correctly diagnosed by dual-energy CT. Left panel shows axial CT image of stone. Right panel shows elaboration with Kidney Stones software (Siemens Healthcare), including views of stone in sagittal, coronal, and axial planes and settings dialog. Stone had diameter of mm, CT density of 1200 HU, and smooth surface. Stone was treated with two extracorporeal shock wave lithotripsy sessions and was not ejected completely, making second procedure necessary. Therapeutic ureterorenoscopy was performed because stone was in distal tract of ureter. accuracy of dual-energy CT in determining the chemical composition of urinary stones. In the last 20 years, better knowledge of the etiopathogenesis of urolithiasis and the development of more elaborate tools led to great changes in the therapy choice. In particular, endoscopic surgery and extracorporeal lithotripsy have supplanted open surgery in most cases. Unfortunately, at present, there is no clear standardization of indications, and the therapeutic decision often falls to the specialist s preferences, surgical ability, and technical means. Most commonly, the urologist chooses to use the least invasive method; however, it is known that this is not always the most useful approach to obtain the optimal result, which is complete stone removal without damage to the urinary tract and to renal function. In the late 90s, the concept that stone fragility depends on chemical composition was introduced and studied in association with the efficacy of extracorporeal lithotripsy. In the same years, a few authors reported that cystine and monohydrate calcium oxalate stones were more resistant to fragmentation than were stones composed of hydroxyapatite, dihydrate calcium oxalate, and uric acid [11, 18]. Thus, the chemical composition of stones affects the type of fragmentation and, as a consequence, its elimination with extracorporeal treatment; for example, cystine and monohydrate calcium oxalate stones tend to yield large residual fragments that are difficult to eliminate [19 21]. In our study, because the accuracy of dual-energy CT in the characterization of the stones chemical composition had to be assessed, treatment strategies were based only on stone size (for stones 20 mm, extracorporeal shock wave lithotripsy was used; for stones > 20 mm, percutaneous nephrolithotomy or therapeutic ureterorenoscopy was used). If we had known the stones composition in advance, we could have changed management technique for four patients with oxalate monohydrate stones (diameter, mm) who unsuccessfully underwent extracorporeal shock wave lithotripsy; instead, they would have been treated first with percutaneous nephrolithotomy or therapeutic ureterorenoscopy. In addition, the stone s shape can influence whether adequate fragmentation is achieved; stones of irregular aspect, with spikes or cutoff edges, seem to be more fragile. In particular, two kinds of stones have been classified: smooth and rough. Rough stones are definitely more fragile than smooth ones and are more suitable for extracorporeal treatment, probably because they are less compact and easier to break [12, 13, 19, 22]. In the present study, we found that MDCT can provide useful information on stone shape, as shown by the significant (p < 0.001) differences in CT density and surface between ejected and unejected stones after extracorporeal shock wave lithotripsy. It is important to know the stones chemical composition to plan therapies and to prevent recurrences; the risk of recurrence for calcium oxalate stones is 10% at 1 year and 50% at 10 years after treatment [1]. Moreover, some of these therapies can be curative if they are performed before treatment; for example, urinary alkalization with citrates can dissolve uric acid stones and can be used much earlier in association with allopurinol therapy, thus eliminating the need for extracorporeal shock wave lithotripsy, a procedure that is expensive and subject to complications (e.g., renal hemorrhage, fibrosis, or hypertension) [1 4]. In contrast, alkalization is contraindicated in infective calculosis (hydroxyapatite) because high urinary ph induces bacterial proliferation [3]. We should also emphasize that metabolic analyses are usually performed only for recurrent stones, and crystallography can be performed only if stone fragments are collected. Knowing the stone composition in vivo before treatment appears to be a more practical approach. With regard to patients with cystinuria and dendritic AJR:197, July 2011 W81

7 Manglaviti et al. TABLE 5: Efficacy of Extracorporeal Shock Wave Lithotripsy in Stone Fragmentation According to Stone CT Density and Surface in 34 Patients Parameter Complete Fragmentation Incomplete Fragmentation CT density (HU) a < > Surface Smooth 6 10 Rough 18 0 Note p < for both CT density and surface. HU = Hounsfield units. a CT density was measured before treatment. stones, MDCT can be useful for evaluating the extent of kidney involvement before percutaneous nephrolithotomy. In our study, in four cases, we found disagreement between the chemical composition as predicted by dual-energy CT and that found by crystallography; all of the stones were mixed uric acid and hydroxyapatite according to crystallography but were misclassified as cystine and hydroxyapatite according to dual-energy CT. All these stones were less than 1 cm in diameter. Thus, it is probable that the combination of mixed composition and diameter less than 1 cm led to a decrease in the ability of the software to characterize the stones. In fact, as stated before, the dual-energy technique is based on the different CT densities of stone components at 80 and 140 kv. The difficulty of dual-energy CT in correctly classifying small mixed stones is probably related to the impossibility of identifying a uniform region of interest on which we could measure CT density. In fact, when inhomogeneous areas of different components (with different electronic densities) are adjacent to each other, the measuring process is less reliable [23]. This was confirmed by the fact that a mixed stone larger than 2 cm, composed of uric acid and hydroxyapatite according to crystallography, was correctly classified by dual-energy CT. In clinical practice, management of mixed stones is chosen on the basis of the prevalent stone type, thus confirming the importance, even in these cases, of a correct classification. The reason why we only evaluated stones larger than 5 mm in diameter is that smaller stones, according to the International Urology Guidelines, are not considered for extracorporeal shock wave lithotripsy [21, 22, 24]. According to Miller and Kane [25], stones smaller than 4 mm in diameter are ejected spontaneously in 80% of cases, whereas stones larger than 7 mm in diameter are rarely ejected spontaneously and need intervention. A correlation between stone CT density as measured in Hounsfield units and response to lithotripsy treatment is already known: calcium oxalate stones with smooth surface, diameter greater than 1 cm, and a CT density greater than 1200 HU rarely get fragmented by extracorporeal shock wave lithotripsy, whereas calcium oxalate stones with a CT density less than 1000 HU can be treated successfully. In contrast, stones with CT density greater than 1000 HU made of cystine are not treated with extracorporeal shock wave lithotripsy but with percutaneous nephrolithotomy or ureterorenoscopy [18 21]. Furthermore, with regard to uric acid stones, because CT is able to precisely establish the stone site, it can be very useful to point the lithotriptor because this kind of stone is transparent and not visible on radiographs and, if such a stone is in the ureter, it often is not detectable with ultrasound [26 28]. Moreover, the possibility of detecting chemical composition of urinary stones in the pretreatment phase obviates the collection of stone fragments, a procedure that often fails in patients who are not compliant or that is impossible in cases in which stones are pulverized (i.e., laser treatments). It is worth noting that, in our sample, 25% (10/40) of patients underwent two procedures within a short time, which increased costs and the risks of complications. Several limitations of this study should be considered. First, the sample size was relatively small. In vivo studies of a larger population are warranted to better clarify the potential of dual-energy CT in this setting. A second limitation is the fact that we did not specifically evaluate the x-ray dose for this dual-energy study. However, although the mean radiation dose for excretory urography is about 5 msv, the dose for a two-acquisition contrast-en- hanced CT urography is about 10 msv [5, 6]. Other studies showed the possibility of further decreasing radiation dose for CT urography [29]. At our center, radiation dose in a phantom model permitted us to estimate the mean radiation dose for a dual-energy CT examination, with care dose active, a little lower than that for an unenhanced spiral abdominal CT examination. Finally, regarding clinical application of our results, we should consider the small number of MDCT units with dualenergy capabilities available for clinical use, even though a dual-energy approach could be possible using a non-dual-source standard 64- MDCT scanner [30]. In conclusion, MDCT with a dual-energy technique had excellent accuracy in classifying urinary stone chemical composition except for small uric acid hydroxyapatite mixed stones. The technique should be further validated on larger populations, and the clinical impact of this approach needs to undergo randomized clinical trials showing improved treatment of patients with urolithiasis. References 1. Moe OW. Kidney stones: pathophysiology and medical management. Lancet 2006; 367: Sellaturay S, Fry C. The metabolic basis for urolithiasis. Surgery 2008; 26: Park S. Medical management of urinary stone disease. Expert Opin Pharmacother 2007; 8: Ajayi L, Jaeger P, Robertson W, et al. Renal stone disease. Medicine 2007; 35: Stacul F, Rossi A, Cova MA. CT urography: the end of IVU? Radiol Med 2008; 113: Williams JC, Kim SC, Zarse CA, et al. Progress in the use of helical CT for imaging urinary calculi. J Endourol 2004; 18: Graser A, Johnson TRC, Bader M, et al. Dual energy CT characterization of urinary calculi: initial in vitro and clinical experience. Invest Radiol 2008; 43: Yeh BM, Shepherd JA, Wang ZJ, Teh HS, Hartman RP, Prevrhal S. Dual-energy and lowkvp CT in the abdomen. AJR 2009; 193: Boll DT, Patil NA, Paulson EK, et al. Renal stone assessment with dual-energy multidetector CT and advanced postprocessing techniques: improved characterization of renal stone composition pilot study. Radiology 2009; 250: Primak AN, Fletcher JG, Vrtiska TJ, et al. Noninvasive differentiation of uric acid versus non-uric acid kidney stones using dual-energy CT. Acad Radiol 2007; 14: Matlaga BR, Kawamoto S, Fishman E. Dual source computed tomography: a novel technique W82 AJR:197, July 2011

8 Evaluation of Urinary Stones With Dual-Energy CT to determine stone composition. Urology 2008; of dual energy and limitations due to respiratory mination of urinary stone composition. Urology 72: motion. AJR 2008; 190: ; 58: Pearle MS, Nadler R, Bercowsky E, et al. Prospective 18. Pareek G, Armenakas NA, Panagopoulos G, et al. 24. Preminger GM, Tiselius HG, Assimos DG, et al. randomized trial comparing shock wave lithotripsy Extracorporeal shock wave lithotripsy success 2007 Guideline for the management of ureteral and ureteroscopy for management of distal ureteral based on body mass index and Hounsfield units. calculi. Eur Urol 2007; 52: calculi. J Urol 2001; 166: Tiselius HG, Ackermann D, Alken P, et al. Guidelines on urolithiasis. Eur Urol 2001; 40: Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977; 33: Bellin MF, Renard-Penna P, Conort P, et al. Helical CT evaluation of the chemical composition of urinary tract calculi with a discriminant analysis of CT-attenuation values and density. Eur Radiol 2004; 14: Zarse CA, McAteer JA, Tann M, et al. Helical CT accurately reports urinary stone composition using attenuation values: in vitro verification using highresolution micro-computed tomography calibrated to fourier transform infrared microspectroscopy. Urology 2004; 63: Grosjean R, Sauer B, Guerra R, et al. Characterization of human renal stones with MDCT: advantage Urology 2005; 65: Weld KJ, Montiglio C, Morris MS, et al. Shock wave lithotripsy success for renal stones based on patient and stone computed tomography characteristics. Urology 2007; 70: Pearle MS, Lingeman JE, Leveillee R, et al. Prospective randomized trial comparing shock wave lithotripsy and ureteroscopy for lower pole caliceal calculi 1 cm or less. J Urol 2008; 179:S69 S Katz G, Lencovsky Z, Pode D, et al. Place of extracorporeal shock-wave lithotripsy (ESWL) in management of cystine calculi. Urology 1990; 36: Pareek G, Armenakas NA, Fracchia JA. Hounsfield units on computerized tomography predict stone-free rates after extracorporeal shock wave lithotripsy. J Urol 2003; 169: Motley G, Dalrymple N, Keesling C, Fischer J, Harmon W. Hounsfield unit density in the deter- 25. Miller OF, Kane CJ. Time to stone passage for observed ureteral calculi: a guide for patient education. J Urol 1999; 162: LeRoy AJ. Diagnosis and treatment of nephro lithiasis: current perspectives. AJR 1994; 163: Dhar M, Denstedt JD. Imaging in diagnosis, treatment, and follow-up of stone patients. Adv Chronic Kidney Dis 2009; 16: Sandhu C, Anson KM, Patel U. Urinary tract stones. Part I. Role of radiological imaging in diagnosis and treatment planning. Clin Radiol 2003; 58: Poletti PA, Platon A, Rutschmann OT, et al. Lowdose versus standard-dose CT protocol in patients with clinically suspected renal colic. AJR 2007; 188: Ho LM, Yoshizumi TT, Hurwitz LM, et al. Dual energy versus single energy MDCT: measurement of radiation dose using adult abdominal imaging protocols. Acad Radiol 2009; 16: AJR:197, July 2011 W83