Genitourinary Imaging Original Research

Transcription

1 Genitourinary Imaging Original Research Masch et al. Genitourinary Imaging Original Research William R. Masch 1 Richard H. Cohan 1,2 James H. Ellis 1,2 Jonathan R. Dillman 1,3 Jonathan M. Rubin 1,2 Matthew S. Davenport 1,2,4 Masch WR, Cohan RH, Ellis JH, Dillman JR, Rubin JM, Davenport MS Keywords: artifact, calculus, false-positive, twinkling, urolithiasis DOI: /AJR Received May 16, 2015; accepted after revision June 20, R. H. Cohan and J. H. Ellis have worked as consultants for GE HealthCare regarding nephrogenic systemic fibrosis litigation. M. S. Davenport receives royalties from Wolters Kluwer. J. R. Dillman and J. M. Rubin have received research support from Siemens Healthcare. 1 Department of Radiology, University of Michigan Health System, 1500 E Medical Center Dr, B2-A209P, Ann Arbor MI, Address correspondence to M. S. Davenport (matdaven@med.umich.edu). 2 Department of Radiology, Division of Abdominal Imaging, University of Michigan Health System, Ann Arbor, MI. 3 Department of Radiology, Cincinnati Children s Hospital Medical Center, Cincinnati, OH. 4 Michigan Radiology Quality Collaborative, Ann Arbor, MI. This article is available for credit. AJR 2016; 206: X/16/ American Roentgen Ray Society Clinical Effectiveness of Prospectively Reported Sonographic Twinkling Artifact for the Diagnosis of Renal Calculus in Patients Without Known Urolithiasis OBJECTIVE. The purpose of this study was to determine the clinical effectiveness of prospectively reported sonographic twinkling artifact for the diagnosis of renal calculus in patients without known urolithiasis. MATERIALS AND METHODS. All ultrasound reports finalized in one health system from June 15, 2011, to June 14, 2014, that contained the words twinkle or twinkling in reference to suspected renal calculus were identified. Patients with known urolithiasis or lack of a suitable reference standard (unenhanced abdominal CT with 2.5-mm slice thickness performed 30 days after ultrasound) were excluded. The sensitivity, specificity, and positive likelihood ratio of sonographic twinkling artifact for the diagnosis of renal calculus were calculated by renal unit and stratified by two additional diagnostic features for calcification (echogenic focus, posterior acoustic shadowing). RESULTS. Eighty-five patients formed the study population. Isolated sonographic twinkling artifact had sensitivity of 0.78 (82/105), specificity of 0.40 (26/65), and a positive likelihood ratio of 1.30 for the diagnosis of renal calculus. Specificity and positive likelihood ratio improved and sensitivity declined when the following additional diagnostic features were present: sonographic twinkling artifact and echogenic focus (sensitivity, 0.61 [64/105]; specificity, 0.65 [42/65]; positive likelihood ratio, 1.72); sonographic twinkling artifact and posterior acoustic shadowing (sensitivity, 0.31 [33/105]; specificity, 0.95 [62/65]; positive likelihood ratio, 6.81); all three features (sensitivity, 0.31 [33/105]; specificity, 0.95 [62/65]; positive likelihood ratio, 6.81). CONCLUSION. Isolated sonographic twinkling artifact has a high false-positive rate (60%) for the diagnosis of renal calculus in patients without known urolithiasis. S onographic twinkling artifact becomes evident during color Doppler ultrasound examinations as a mixture of rapidly alternating color signal deep to a strong reflector. It is more commonly observed when the acoustic beam encounters rough surfaces, such as parenchymal or luminal calcification [1]. The presence of sonographic twinkling artifact has been found to be largely independent of ultrasound beam focusing and transducer frequency. This characteristic is purported to improve its use across a range of sonographers and body sizes [2]. Study results have suggested that sonographic twinkling artifact may aid in the detection of renal calculi [2 7] with a variety of reference standard imaging modalities, including abdominal radiography [5], excretory urography [3], gray-scale sonography [3, 4, 6], and CT [8 10]. None of these studies evaluated the clinical effectiveness of sono- graphic twinkling artifact across a range of radiologists in routine clinical practice or isolated the patient populations to patients without known urolithiasis. Few stratified their results by the presence of additional diagnostic features of nephrolithiasis. Given that sonographic twinkling artifact is generally a subjective diagnostic criterion, we were interested in whether the low falsepositive rate observed for sonographic twinkling artifact in research [9, 11] translates into routine clinical practice. The purpose of our study was to assess retrospectively the clinical effectiveness of prospectively reported sonographic twinkling artifact in the diagnosis of renal calculus in patients without known urolithiasis. Materials and Methods Institutional review board approval was obtained, and the requirement for informed consent waived for this HIPAA-compliant retrospective 326 AJR:206, February 2016

2 blinded cohort study in which comparisons with a reference standard were made Ultrasound reports with the word twinkle or twinkling Patients The study population consisted of pediatric and adult patients undergoing abdominal ultrasound in a single health system from June 15, 2011, through June 14, All finalized abdominal ultrasound reports containing the word twinkle or twinkling in reference to suspected urinary tract calculus (n = 4434) were identified by query of the institutional electronic record system. There was one inclusion criterion: availability of images from a reference standard unenhanced thin-section ( 2.5 mm) abdominal CT examination performed within 30 days after the ultrasound examination (n = 158). Exclusion criteria were availability of abdominal CT images (any type) obtained within 3 years before the ultrasound examination (n = 58), known urolithiasis (n = 1), and instances in which twinkle or twinkling referred to suspected calcifications outside the renal collecting system (n = 14). Patients with known urolithiasis and patients with comparison CT performed before ultrasound were excluded to ensure that the radiologists who prospectively dictated the reports of the ultrasound studies were not biased in their assessment of sonographic twinkling artifact. The final study population consisted of 85 patients (mean age, 46 years; range, years; 58 female patients [mean age, 43 years; range, years]; 27 male patients [mean age, 53 years; range, years]) without known urolithiasis who underwent 85 abdominal ultrasound examinations in which sonographic twinkling artifact TABLE 1: Indications for Ultrasound Examinations (n = 85) Indication No. of Examinations Flank pain 28 Generalized abdominal pain 12 Right upper quadrant pain 10 Hematuria 6 Elevated liver function enzyme 5 levels Renal failure 4 Hydronephrosis 4 Urinary tract infection 3 Renal lesion 2 Nausea 2 Cirrhosis 2 Dysuria 2 Other Ultrasound reports for 158 unique patients with reference standard CT performed within 30 days after ultrasound 85 Ultrasound reports for 85 unique patients with reference standard CT performed within 30 days after ultrasound and no preceding CT within 3 years before ultrasound Fig. 1 Study population flowchart. was described in reference to one or more suspected renal calculi and who underwent reference standard thin-section CT of the abdomen within 30 days of the index ultrasound examination. The flowchart of the study population is depicted in Figure 1, and the indications for the ultrasound examinations are shown in Table 1. Ultrasound Technique Ultrasound examinations of the abdomen (i.e., complete abdominal ultrasound, limited abdominal ultrasound, retroperitoneal ultrasound) included in this study were performed with three ultrasound machines: 73 with an iu22 system (Philips Healthcare); 11, a Logiq 9 system (GE Healthcare); and one, an Acuson Antares system (Siemens Healthcare). The highest pulse repetition frequency (PRF) available depending on probe frequency and target depth was used to suppress as much background color signal as possible. Because it produces a random broadband signal, twinkling artifact appears to alias at all PRFs and as such can be differentiated from real flow signals, which at high PRFs either fall into the wall filter or do not alias [2]. These PRF settings resulted in a mean color-flow velocity setting of 85 cm/s (range, cm/s). The mean distance from skin to renal cortex was measured for all patients (all kidneys mean, 38 mm [range, mm]; right kidneys mean, 40 mm [range, mm]; left kidneys mean, 37 mm [range, mm]) Ultrasound reports excluded because reference standard CT was not performed within 30 days after ultrasound 58 Patients excluded because they underwent CT within 3 years before ultrasound 1 Patient excluded because of known history of nephrolithiasis (previous lithotripsy) 14 Patients excluded because twinkle or twinkling referred to suspected calcifications outside the renal collecting system Reference Standard Unenhanced thin-section ( mm) abdominal CT within 30 days or less after the index ultrasound examination was treated as the reference standard. The CT parameters were as follows: variable tube current; kvp; slice thickness, mm; FOV, cm; with or without adaptive statistical iterative reconstruction (30% blend). Two expert abdominal radiologists (30 and 33 years experience) blinded to the ultrasound findings in consensus reviewed the unenhanced CT images and recorded the presence (binary) and maximum size (in millimeters) of renal calculi in each of six sectors (right upper pole, right interpolar, right lower pole, left upper pole, left interpolar, left lower pole). The upper pole sector was defined as above the upper polar line. The lower pole sector was defined as below the lower polar line. The interpolar sector was defined as between the polar lines. The two differences of opinion were resolved by consultation with a third expert abdominal radiologist with 8 years experience. We intentionally chose radiologists with substantial experience to perform the reference standard readings because the reference standard interpretation was being performed de novo and not gleaned from a preexisting report. Concordance with the reference standard was determined in a binary manner. If the ultrasound report described sonographic twinkling artifact in a given location and the reference standard AJR:206, February

3 Masch et al. TABLE 2: Diagnostic Accuracy of Prospectively Reported Sonographic Twinkling Artifact for the Diagnosis of Renal Calculi in Patients Without Known Urolithiasis or Recent ( 3 y) Comparison CT Analysis Sensitivity Specificity showed at least one calculus in the same location, the finding was considered true-positive, irrespective of the number of calculi present. The level of analysis was by kidney (primary) and by adjacent sector (secondary). A true-positive finding for the adjacent sector analysis was considered when there was a matching calculus on the reference standard study either within the same sector or in an adjacent contiguous sector (e.g., an upper pole calculus reported at ultrasound would be considered a true-positive finding if there was a corresponding calculus in the same upper pole or adjacent interpolar region on the reference standard image). An individual sector-based approach was not used because of the possibility of intrarenal stone migration and the variation in scan planes between ultrasound and CT that might affect reporting of the location of a given calculus with respect to the polar lines. Retrospective Ultrasound Review The prospectively dictated ultrasound reports and their corresponding images were reviewed retrospectively by one senior radiology resident (postgraduate year 5) and one abdominal radiologist with 4 years experience who were blinded to the reference standard. Each instance of prospectively described sonographic twinkling artifact was recorded with respect to its sector location. These foci were then identified retrospectively on Positive Predictive Value the source images by use of still images, cine clips, gray-scale imaging, and color Doppler imaging to determine whether there was a corresponding echogenic focus in the same location or a posterior acoustic shadow projected deep to the sonographic twinkling artifact. The reviewers were prohibited from identifying new foci of sonographic twinkling artifact to include in the study. Negative Predictive Value Positive Likelihood Ratio Negative Likelihood Ratio Isolated sonographic twinkling artifact All right kidneys 0.77 (41/53) 0.34 (11/32) 0.66 (41/62) 0.48 (11/23) All left kidneys 0.79 (41/52) 0.45 (15/33) 0.69 (41/59) 0.58 (15/26) All kidneys 0.78 (82/105) 0.40 (26/65) 0.68 (82/121) 0.53 (26/49) Sonographic twinkling artifact and echogenic focus All right kidneys 0.55 (29/53) 0.63 (20/32) 0.71 (29/41) 0.45 (20/44) All left kidneys 0.67 (35/52) 0.67 (22/33) 0.76 (35/46) 0.56 (22/39) All kidneys 0.61 (64/105) 0.65 (42/65) 0.74 (64/87) 0.51 (42/83) Sonographic twinkling artifact and posterior acoustic shadow All right kidneys 0.28 (15/53) 0.94 (30/32) 0.88 (15/17) 0.44 (30/68) All left kidneys 0.35 (18/52) 0.97 (32/33) 0.95 (18/19) 0.48 (32/66) All kidneys 0.31 (33/105) 0.95 (62/65) 0.92 (33/36) 0.46 (62/134) Sonographic twinkling artifact, echogenic focus, and posterior acoustic shadow All right kidneys 0.28 (15/53) 0.94 (30/32) 0.88 (15/17) 0.44 (30/68) All left kidney 0.35 (18/52) 0.97 (32/33) 0.95 (18/19) 0.48 (32/66) All kidneys 0.31 (33/105) 0.95 (62/65) 0.92 (33/36) 0.46 (62/134) Note Values in parentheses are numbers used to calculate performance measure. Statistical Analysis Continuous measures (e.g., patient age, renal calculus size) were summarized with means and ranges. Categoric data (e.g., patient sex) were summarized with counts and percentages. Sensitivity, specificity, positive predictive value, negative predictive value, positive likelihood ratio, and negative likelihood ratio were calculated by renal unit (primary analysis) and adjacent renal sector (secondary analysis) for sonographic twinkling artifact alone, sonographic twinkling artifact with an accompanying echogenic focus, sonographic twinkling artifact with accompanying posterior acoustic shadow, and sonographic twinkling artifact with both an accompanying echogenic focus and a posterior acoustic shadow. The same analysis was repeated for all left kidneys (separately) and all right kidneys (separately) to exclude the effects of repeated measures within patients. The primary analysis was performed by renal unit to account, first, for hypothetical intrarenal stone migration that may have occurred between the ultrasound and CT reference standard examinations and, second, for technical differences between the index ultrasound and CT reference standard that might have influenced the reporting of stone location within a kidney. Results The mean interval between the index ultrasound examination and the reference standard examination was 8 days (range, 0 30 days). All patients had two kidneys, and no patient had a transplant kidney. Calculi were found on the reference standard images in 105 of 170 (62%) renal units and in 64 of 85 (75%) patients. The mean maximum calculus size measured retrospectively on the reference standard images was 3.6 mm (range, 1 20 mm). Twenty radiologists dictated the 85 prospective reports of the ultrasound examinations (mean examinations per radiologist, 4.3; range, 1 11). Sonographic twinkling artifact was prospectively reported alone or in combination with other ultrasound findings in 121 of 170 (71%) renal units. Twenty percent (34 of 170) of renal units had isolated sonographic twinkling artifact with no corresponding echogenic focus or posterior acoustic shadow; 51% percent (87 of 170) of renal units had sonographic twinkling artifact with an accompanying echogenic focus; 21% (36 of 170) of 328 AJR:206, February 2016

4 TABLE 3: Diagnostic Accuracy by Kidney and Adjacent Renal Sector of Prospectively Reported Sonographic Twinkling Artifact for the Diagnosis of Renal Calculi in Patients Without Known Urolithiasis or Recent ( 3 y) Comparison CT Analysis Sensitivity Specificity renal units had sonographic twinkling artifact with an accompanying posterior acoustic shadow; and 21% (36 of 170) of renal units had sonographic twinkling artifact with both an accompanying echogenic focus and a posterior acoustic shadow. All instances of sonographic twinkling artifact with a posterior acoustic shadow also had an accompanying echogenic focus. The diagnostic accuracy of individual foci of sonographic twinkling artifact for the diagnosis of renal calculus is reported in Table 2 (by renal unit) and Table 3 (by adjacent sector). Although sonographic twinkling artifact was somewhat sensitive for the detection of renal calculus (all kidneys, 0.78 [82/105]), it had weak specificity (all kidneys, 0.40 [26/65]), a high false-positive rate (all kidneys, 0.60 [39/65]), and a weak positive likelihood ratio (all kidneys, 1.30). The sensitivity declined (all kidneys, 0.61 [64/105]) and the specificity improved (all kidneys, 0.65 [42/65) Positive Predictive Value Negative Predictive Value when sonographic twinkling artifact was accompanied by an echogenic focus, but the positive likelihood ratio remained suboptimal (all kidneys, 1.72). Only addition of a posterior acoustic shadow resulted in a strong positive likelihood ratio for sonographic twinkling artifact (all kidneys, 6.81), but this combination had poor sensitivity (all kidneys, 0.31 [33/105]) and a weak negative predictive value (all kidneys, 0.46 [62/134]). When the analysis was performed by adjacent sector, the sensitivity of isolated sonographic twinkling artifact declined (all adjacent sectors, 0.65 [117/181]), and the positive likelihood ratio remained weak (all adjacent sectors, 1.54). For both right and left kidneys, the sensitivity and positive predictive value were lower for calculi identified in the upper pole than for those in the lower pole (Table 3). Figure 2 is an example of a false-positive diagnosis of isolated sonographic twinkling artifact for the diagnosis of renal calculus. Positive Likelihood Ratio Negative Likelihood Ratio All kidneys 0.78 (82/105) 0.40 (26/65) 0.68 (82/121) 0.53 (26/49) All right kidneys 0.77 (41/53) 0.34 (11/32) 0.66 (41/62) 0.48 (11/23) Upper pole, interpolar 0.59 (24/41) 0.59 (26/44) 0.57 (24/42) 0.60 (26/43) Interpolar, lower pole 0.71 (36/51) 0.44 (15/34) 0.65 (36/55) 0.50 (15/30) All left kidneys 0.79 (41/52) 0.45 (15/33) 0.69 (41/59) 0.58 (15/26) Upper pole, interpolar 0.52 (22/42) 0.65 (28/43) 0.59 (22/37) 0.58 (28/48) Interpolar, lower pole 0.74 (35/47) 0.62 (24/39) 0.70 (35/50) 0.67 (24/46) All adjacent sectors 0.65 (117/181) 0.58 (93/160) 0.64 (117/184) 0.59 (93/157) Note Values in parentheses are numbers used to calculate performance measure. Discussion Our data show overall suboptimal performance of prospectively reported sonographic twinkling artifact for the diagnosis of renal calculus compared with reference standard CT in patients without known urolithiasis. In these patients, isolated sonographic twinkling artifact had low specificity (0.40 [26/65]), a high false-positive rate (0.60 [39/65]), and a low positive likelihood ratio (1.30). Relying on sonographic twinkling artifact led to a false diagnosis of renal calculus disease in 21 of the 85 (25%) patients included in our study and in 39 of 121 (32%) renal units. Addition of gray-scale features improved specificity but reduced sensitivity. When a corresponding echogenic focus was present, the falsepositive rate decreased to 0.35 (23/65), but the positive likelihood ratio remained low (1.72). A Fig year-old woman who presented to emergency department with right upper quadrant pain. Example of false-positive diagnosis of nephrolithiasis. A and B, Gray-scale (A) and color Doppler (B) ultrasound images show isolated focus of sonographic twinkling artifact in lower pole of right kidney (B) with no corresponding echogenic focus or posterior acoustic shadowing on gray-scale ultrasound (A). Findings were interpreted as suspicious for renal calculus. C, Unenhanced coronal reconstruction (2-mm slice thickness) CT image obtained same day as A and B shows no calculus. B C AJR:206, February

5 Masch et al. A corresponding posterior acoustic shadow was highly specific for the diagnosis of renal calculus (0.95 [62/65]), but this combination of features (twinkling artifact and posterior acoustic shadow) had poor sensitivity (0.31 [33/105]). False-positive diagnoses of renal calculi could lead to an inaccurate explanation for hematuria or pain or prompt unneeded CT. Many patients are evaluated with renal ultrasound in whom urolithiasis is not suspected. Incidental sonographic twinkling artifact interpreted as urolithiasis in these patients has the potential to change management, incur additional radiation exposure (e.g., confirmatory CT or radiography), and increase health care cost (e.g., nephrologic or urologic consultation, additional testing, effect on insurance premiums). This is of particular concern when isolated sonographic twinkling artifact is present in the absence of any grayscale correlate. In our study, the false-positive rate of isolated sonographic twinkling artifact was 60% (39/65). The cause of this false-positive rate is uncertain. Possible explanations include noncalcific strong reflectors within the renal parenchyma or renal sinus [1], renal arterial calcifications [1], and turbulent flow misinterpreted as sonographic twinkling artifact. Sonographic twinkling artifact is a subjective finding. Its presence, detection, and interpretation are subject to variability in ultrasound settings, sonographer skill and technique, and radiologist sensitivity. In this study we tested the clinical effectiveness of prospectively reported sonographic twinkling artifact for the diagnosis of renal calculus in routine clinical practice across a spectrum of radiologists, sonographers, and ultrasound machines. The patients included were not known to have nephrolithiasis at the time ultrasound was performed and the images interpreted, lessening the possibility of review bias. In previous studies of the diagnostic accuracy of sonographic twinkling artifact for the diagnosis of renal calculus, the investigators did not make similar attempts to exclude patients with known nephrolithiasis [8, 9, 11, 12]. Our results can be contrasted to previously published findings. In a prospective cohort study in 2012, Winkel et al. [11] evaluated 105 consecutively registered adult patients referred because of suspicion of nephrolithiasis who were imaged with gray-scale and color Doppler retroperitoneal ultrasound. The reference standard was unenhanced CT with 3-mm-thick reconstruction. On a perpatient basis, those authors reported sensitivity of 81%, specificity of 89%, positive predictive value of 76%, and negative predictive value of 92% for the diagnosis of renal calculus with color Doppler ultrasound and gray-scale ultrasound combined. When the same analysis was conducted sector by sector, sensitivity declined (55%), specificity increased (99%), positive predictive value declined (67%), and negative predictive value increased (98%). The authors concluded that color Doppler ultrasound combined with B- mode ultrasound is valuable in ruling out nephrolithiasis. However, they also reported identical specificity (99%), excellent negative predictive value (97%), and improved positive predictive value (81%) for B-mode ultrasound alone, and all ultrasound examinations included in the study were performed by one of two skilled physicians in a controlled research setting. In a prospective cohort study in 2012, Kielar et al. [9] evaluated 51 patients who presented to the emergency department with acute flank pain and who were imaged with gray-scale and color Doppler retroperitoneal ultrasound. The reference standard was unenhanced CT reconstructed with a 1.25-mm slice thickness. The authors reported overall sensitivity of 83% and a superior positive predictive value of 94% for the diagnosis of renal calculus based on the presence of sonographic twinkling artifact. Unlike in the current study, Kielar et al. evaluated a focused population with active clinical symptoms consistent with renal colic. All patients were imaged by a select group of individuals involved in the research study. The enriched nature of their population coupled with the expertise of the participating practitioners likely explains the difference in reported test characteristics. Dillman et al. [8] in 2011 conducted a retrospective analysis of the records of 74 patients who underwent Doppler sonography of the kidneys followed by reference standard CT (5-mm slice thickness [Dillman JR, verbal communication, 2015]) within 2 weeks of the index ultrasound. They did not target a specific patient population or exclude patients with known urolithiasis. In contradistinction to the investigators in the prospective studies [9, 11], Dillman et al. [8] reported a low true-positive rate (49%) and a high falsepositive rate (51%). They did not stratify their observed diagnostic accuracy by the presence of concomitant gray-scale features, and they used a thicker-section CT reference standard (5 mm) than we did ( 2.5 mm). Our study had several limitations. It was retrospective in design, and we evaluated only patients in whom a sonographic twinkling artifact was detected prospectively; thus, our ability to assess negative predictive value in the general population was limited. Given that 75% (64/85) of the patients had calculi identified with the reference standard, and the required reference standard was unenhanced CT performed within 30 days after the index ultrasound, our study involved a selected population at higher risk of calculous disease, which might increase the positive predictive value of sonographic twinkling artifact reported in this study. We excluded patients who had known nephrolithiasis at the time of index ultrasound to minimize prospective review bias by the interpreting radiologists. Most of the patients in this study were women ( 2:1 ratio), but in the general population men are more likely to have nephrolithiasis [13]. This may have been a result of our sample size or the exclusion of patients with comparison CT. Given that the imaging appearance of renal calculi is not expected to differ between the sexes, it is unlikely that this affected our results. There was a delay between the index ultrasound and the reference standard CT (mean delay, 8 days). We performed our primary analysis by renal unit to account for hypothetical intrarenal stone migration that may have occurred between ultrasound and the reference standard examination. However, this analysis would be unable to account for the hypothetical passage of an intrarenal calculus out of the kidney between the ultrasound and the reference standard examinations. Last, we analyzed only intrarenal sonographic twinkling artifact. Our data may not apply to suspected calculi within the ureter or bladder. In summary, isolated sonographic twinkling artifact that is prospectively reported in routine clinical practice in patients without known urolithiasis has a high false-positive rate for the diagnosis of renal calculus. Specificity improves at the cost of sensitivity when the artifact is accompanied by corresponding gray-scale findings, posterior acoustic shadowing being most predictive. These findings are important because overdiagnosis of renal calculi can lead to misdiagnosis, misdirected care, and resource overutilization. References 1. Rahmouni A, Bargoin R, Herment A, et al. Color Doppler twinkling artifact in hyperechoic regions. Radiology 1996; 199: AJR:206, February 2016

6 2. Shabana W, Bude RO, Rubin JM. Comparison between color Doppler twinkling artifact and acoustic shadowing for renal calculus detection: an in vitro study. Ultrasound Med Biol 2009; 35: Chelfouh N, Grenier N, Higueret D, et al. Characterization of urinary calculi: in vitro study of twinkling artifact revealed by color-flow sonography. AJR 1998; 171: Aytaç SK, Ozcan H. Effect of color Doppler system on the twinkling sign associated with urinary tract calculi. J Clin Ultrasound 1999; 27: Lee JY, Kim SH, Cho JY, et al. Color and power Doppler twinkling artifacts from urinary stones: clinical observations and phantom studies. AJR 2001; 176: Turrin A, Minola P, Costa F, et al. Diagnostic value of colour Doppler twinkling artefact in sites negative for stones on B mode renal sonography. Urol Res 2007; 35: Tchelepi H, Ralls PW. Color comet-tail artifact: clinical applications. AJR 2009; 192: Dillman JR, Kappil M, Weadock WJ, et al. Sonographic twinkling artifact for renal calculus detection: correlation with CT. Radiology 2011; 259: Kielar AZ, Shabana W, Vakili M, et al. Prospective evaluation of Doppler sonography to detect the twinkling artifact versus unenhanced computed tomography for identifying urinary tract calculi. J Ultrasound Med 2012; 31: Sorensen MD, Harper JD, Hsi RS, et al. B-mode ultrasound versus color Doppler twinkling artifact in detecting kidney stones. J Endourol 2013; 27: Winkel RR, Kalhauge A, Fredfeldt KE. The usefulness of ultrasound colour-doppler twinkling artefact for detecting urolithiasis compared with low dose nonenhanced computerized tomography. Ultrasound Med Biol 2012; 38: Korkmaz M, Aras B, Sanal B, et al. Investigating the clinical significance of twinkling artifacts in patients with urolithiasis smaller than 5 mm. Jpn J Radiol 2014; 32: Romero V, Akpinar H, Assimos DG. Kidney stones: a global picture of prevalence, incidence, and associated risk factors. Rev Urol 2010; 12:e86 e96 FOR YOUR INFORMATION This article is available for CME and Self-Assessment (SA-CME) credit that satisfies Part II requirements for maintenance of certification (MOC). To access the examination for this article, follow the prompts associated with the online version of the article. AJR:206, February