Downloaded from www.ajronline.org by 37.44.207.115 on 01/28/17 from IP address 37.44.207.115. Copyright ARRS. For personal use only; all rights reserved

Mulkens et al. MDCT of Urinary Stone Disease

Genitourinar y Imaging • Original Research

Urinary Stone Disease: Comparison of Standard-Dose and Low-Dose with 4D MDCT Tube Current Modulation Tom H. Mulkens1 Sofie Daineffe1 Roel De Wijngaert1 Patrick Bellinck1 André Leonard2 Guido Smet2 Jean-Luc Termote1 Mulkens TH, Daineffe S, De Wijngaert R, et al.

OBJECTIVE. The purpose of this prospective study was to compare the performance of standard-dose MDCT with that of low-dose MDCT with tube current modulation in patients with renal colic. SUBJECTS AND METHODS. Three hundred patients underwent 6- and 16-MDCT in 150 standard-dose examinations (6-MDCT effective tube current, 95 mAs at 130 kV; 16MDCT effective tube current, 120 mAs at 120 kV) and 150 low-dose examinations (6-MDCT effective tube current, 51 mAs at 110 kV; 16-MDCT effective tube current, 70 mAs at 120 kV), all performed with 4D tube current modulation. Two experienced radiologists using a clinical workstation and blinded to scan parameters prospectively viewed the images from the 300 examinations. In a second session, one experienced radiologist and two first-year residents using a clinical workstation retrospectively reviewed images from 100 randomly selected standarddose and 100 randomly selected low-dose examinations. RESULTS. Tube current modulation reduced effective tube current 25–31% in all examinations. Mean effective dose was 1.41–1.58 mSv for low-dose examinations, which reached additional dose reduction of 51.2–64.3% in comparison with standard-dose examinations. Excellent correlation existed between mean tube current and body mass index of the patients. Spearman’s correlation coefficient was 0.85–0.88 for all examinations. The sensitivity of low-dose examinations interpreted by two experienced reviewers was 97.3–98.6%; specificity, 93.5%; and accuracy, 95.3%. These findings were comparable with those for standard-dose examinations. Sensitivity, specificity, and accuracy of low-dose examinations of overweight and obese patients reached similar high values: 97–100%, 100%, and 98–100%, respectively. Interobserver agreement for urinary stone detection was excellent between the two reviewers, with kappa values of 0.98 for the low-dose and 0.96 for the standard-dose examinations. An alternative diagnosis was identified in 15% and 16% of cases by two experienced radiologists in the two examinations groups. In the second interpretation session, the residents found an alternative diagnosis in only 10–12% of standard-dose examinations and only 4–5% of low-dose examinations. CONCLUSION. Low-dose MDCT with tube current modulation can be used as standard procedure in evaluation for urolithiasis, even in overweight and obese patients.

Keywords: CT, kidney, MDCT, radiation dose, urinary tract DOI:10.2214/AJR.05.1863 Received October 23, 2005; accepted after revision April 11, 2006. 1Department

of Radiology, Heilig Hart Ziekenhuis, Mechelsestraat 24, B 2500 Lier, Belgium. Address correspondence to T. H. Mulkens ([email protected]).

2Department

of Urology, Heilig Hart Ziekenhuis, Lier,

Belgium. AJR 2007; 188:553–562 0361–803X/07/1882–553 © American Roentgen Ray Society

AJR:188, February 2007

mith et al. [1] in 1995 described unenhanced helical CT as an initial imaging technique for detection of urolithiasis in patients with acute flank pain. At the time, these patients would have been referred for excretory urography. Since then, unenhanced helical CT has become the preferred imaging technique because it has distinct advantages over excretory urography [2–6], including shorter examination time, avoidance of the cost and possible complications of IV administration of contrast material, greater sensitivity and specificity for stone detection, and higher potential for detection of other disease that mimics acute renal colic.

S

A major limitation of unenhanced helical CT compared with excretory urography is the higher radiation dose [2, 6, 7], which has not been altered by the introduction of MDCT [7]. The higher dose is of particular concern because a large number of these examinations are performed because of repeated stone formation in young, otherwise healthy patients who need repeated CT examinations [7]. For this reason, several authors have proposed use of a low-dose technique of single-detector helical CT [8, 9], a combination of single-detector CT and MDCT [10, 11], and MDCT only [12]. A limitation of the low-dose studies was that the protocol was not suitable for obese

553

Downloaded from www.ajronline.org by 37.44.207.115 on 01/28/17 from IP address 37.44.207.115. Copyright ARRS. For personal use only; all rights reserved

Mulkens et al. patients. These patients were excluded [11], normal-dose CT was recommended for them [8, 9], or additional focused CT acquisition at higher tube current was needed for sufficient image quality [12]. Another concern about the use of a low radiation dose is the possibility of missing an alternative diagnosis, which is one of the advantages of the use of CT [13]. In this prospective study, we compared standard-dose with low-dose unenhanced MDCT in examinations of patients with suspected renal colic. All imaging was performed with online tube current modulation whereby the tube current was continuously adapted to the patient’s anatomic characteristics during scanning [14, 15]. We tested the possibility of using low-dose MDCT as a routine examination for all patients with suspected urinary stone disease, including tall patients and obese patients. Subjects and Methods Study Population A prospective study was conducted to compare standard-dose and low-dose MDCT, both with 4D tube current modulation, in patients with urinary stone disease. Four dimensions means the tube current is adapted according to the x-, y-, and z-axes continuously over the scanning time. From June 2004 to May 2005, 300 patients (188 men, 112 women; mean age, 51.39 years; age range, 18–90 years) consecutively underwent imaging examinations on a 6-MDCT (Emotion 6, Siemens Medical Solutions) or 16-MDCT (Sensation 16, Siemens) system. The group was divided in half, 150 patients undergoing a standard-dose scanning protocol and 150 patients undergoing a low-dose protocol. Each protocol was alternately chosen on each CT system. Most of the patients (n = 261) went immediately from the emergency department to imaging because of the presence of acute renal colic. Twelve of the patients had associated macroscopic hematuria. Twenty-four patients underwent CT for follow-up of known urinary stones. Fifteen patients underwent CT because of a history of acute renal failure, with abdominal pain and sonographic findings of hydronephrosis. The height and weight of each patient were recorded, and body mass index (BMI) was calculated (body weight in kilograms divided by the square of height in meters). BMI was used for classification of patients into the following subgroups: underweight, BMI ≤ 18.5; normal weight, BMI = 18.5–24.9; overweight, BMI = 25.0–29.9; obese, BMI = 30.0–39.9; and extremely obese, BMI ≥ 40 [16]. The standard-dose group consisted of 91 men and 59 women (male-to-female ratio, 1.54) with a mean age of 52.55 years, age range of 22–90 years,

554

and mean BMI of 26.71. Three (2%) of the patients were underweight; 52 (34.7%), normal weight; 71 (47.3%), overweight; and 24 (16%), obese. The low-dose group consisted of 97 men and 53 women (male-to-female ratio, 1.83) with a mean age of 50.23 years, age range of 18–87 years, and mean BMI of 24.87. Three (2%) of the patients were underweight; 72 (48%), normal weight; 55 (36.7%), overweight; and 20 (13.3%), obese. The study protocol was approved by the ethics review board of our institution, which waived the requirement for individual patient informed consent.

CT Examinations All patients were examined in the supine position without contrast material. An anteroposterior 52-cm scout image was obtained before helical acquisition. For standard-dose examinations, we used the following scanning protocol for 6-MDCT (n = 70): tube voltage of 130 kV, 6 × 1 mm collimation, 0.6-second rotation, table feed of 8 mm per rotation (pitch factor of 1.33), and effective tube current of 95 mAs, for a CT dose index volume (CTDIvol) of 12.53 mGy. The standard-dose 16-MDCT protocol (n = 80) was as follows: tube voltage of 120 kV, 16 × 1.5 mm collimation, 0.5second rotation, table feed of 24 mm per rotation (pitch factor of 1), and effective tube current of 120 mAs, for a CTDIvol of 8.4 mGy. The low-dose 6-MDCT protocol (n = 73) was as follows: tube voltage of 110 kV, 6 × 1 mm collimation, 0.6-second rotation, table feed of 8 mm per rotation (pitch factor, 1.33), and effective tube current of 51 mAs, for a CTDIvol of 4.43 mGy. The low-dose 16MDCT protocol (n = 77) was as follows: tube voltage of 120 kV, 16 × 1.5 mm collimation, 0.5second rotation, table feed of 24 mm per rotation (pitch factor, 1), and effective tube current of 70 mAs, for a CTDIvol of 4.9 mGy. Overlapping 2-mm-thick images with 1-mm reconstruction increments were reconstructed from the raw data set with a soft (body) filter algorithm on both systems. From this MDCT data set, consecutive axial and coronal images were obtained with 5-mm slice thickness for filming. All examinations were performed with a tube current modulation system (CareDose 4D, Siemens). CareDose 4D online tube current modulation combines two forms of modulation: in the z-axis and in the x–y-axis (angular modulation). Tube current modulation in the z-axis is determined from attenuation values and shape obtained through refined analysis of an anteroposterior or lateral projection radiograph at the start of the examination. In z-axis modulation, tube current is adjusted to maintain a user-selected image quality level in the image data. Noise is regulated on the final image to a level desired by the user in an attempt to render all images of similar noise independent of patient size

and anatomic configuration [14]. In this sense, z-axis modulation is the CT equivalent of the autoexposure control systems used for many years with conventional X-ray systems [14]. Angular (x–y) tube current modulation works differently from z-axis modulation. The tube current is adjusted to minimize X-rays in projections (angles) that are of less importance for the reduction of overall image noise content [14, 15, 17, 18]. Noise on CT scans is dominated by projections in which the attenuation is highest. For a homogeneous object with a circular cross section, attenuation is constant over all projections, and all measured values contribute equally. For a nonhomogeneous object with a noncircular cross section, however, such as the human body, attenuation varies strongly, sometimes by more than three orders of magnitude [17, 18]. Because noise measured in high-attenuation projections (lateral direction) greatly influences the noise level in CT data, the dose for projections with relatively low attenuation (anteroposterior direction) can be reduced substantially without a measurable increase in image noise [17, 18]. Angular tube current modulation is characterized by online monitoring of attenuation and subsequent tuning of the tube current as a function of the projection angle with a delay of 360°. For each slice position, the CT system calculates the average tube current, expressed as average effective tube current throughout the exposure. As described by Mahesh et al. [19], effective tube current is determined by dividing the product of tube current and rotation time by the pitch, which, as reported by Silverman et al. [20], is the ratio between table feed per rotation and X-ray beam width or collimation. The mean effective tube current of the examination is displayed at the scan console at the end of the examination and was recorded in our study. The correlation between BMI and mean effective tube current was calculated for all 300 examinations in the study. Effective dose in millisieverts was calculated with commercially available software (WinDose, Institut für Medizinische Physik, Universität Erlangen) on a PC [21]. This software does not require phantom measurements. Input of the scan parameters (sex, scan region and length, tube voltage, current–time product, pitch, multidetector nature) is represented on a graph of the Monte Carlo phantom model, and calculation of effective dose is computer-simulated according to the recommendations of the International Commission on Radiological Protection (report ICRP-60) [22].

Image Analysis Two reviewers with more than 10 years of experience in abdominal CT used a clinical workstation (Wizard, Siemens) to view all images of each patient when the patient presented. The reviewers were allowed to use all the workstation functions,

AJR:188, February 2007

Downloaded from www.ajronline.org by 37.44.207.115 on 01/28/17 from IP address 37.44.207.115. Copyright ARRS. For personal use only; all rights reserved

MDCT of Urinary Stone Disease

A

B

Fig. 1—23-year-old man (height, 1.79 m; weight, 89 kg; BMI, 27.78) with left-sided renal colic and microscopic hematuria. Standard-dose 6-MDCT examination with mean effective tube current of 93 mAs and calculated effective dose of 5.7 mSv. A, Axial 5-mm MDCT scan shows 1.1-cm stone (arrow) in left pyeloureteral junction. B, Coronal 5-mm MDCT image shows obstructive large stone (arrow) depicted in A.

including multiplanar reformation, curved reformation, and maximal intensity projection. The reviewers received the clinical information, especially the side of the patient’s pain, but were unaware of the final diagnosis, BMI, and scan protocol. The reviewers were asked to record the presence or absence of urinary stones. If stones were present, the reviewers were asked to record number, location (intrarenal, calyces, pelvis, ureter [proximal lumbar, distal pelvic ureter, and junction with bladder], and bladder), and size measured with the standard metric software of the workstation. The following indirect signs also were to be recorded: dilatation of the urinary tract, perirenal or periureteral stranding as described by Smith et al. [2], and presence of urinoma formation. Alternative diagnoses and other unsuspected CT findings also were to be recorded. After an interval of 8 weeks, a second retrospective interpretation session was held. A random 50 examinations performed with each CT system (6and 16-MDCT) with each scan protocol (standard dose and low dose) were loaded and reviewed on a Leonardo workstation (Siemens). Of these 200 examinations, only the consecutive 5-mm-thick axial and coronal images were interpreted by three reviewers: one experienced radiologist and two resi-

AJR:188, February 2007

dents with only 6 months of experience in interpretation of abdominal CT scans. We used only 200 examinations in the second interpretation session and reviewed only 5-mm axial and coronal images rather than all images because we did not have a PACS installation with long-term storage capacity, and storage of all complete examinations (≈ 400–500 images) on CDs did not seem practical. The reviewers were blinded to scan protocol (scan parameters were removed) but were aware of the clinical information. The same parameters as for the first interpretation were recorded: presence, number, location, and size of the urinary stones, indirect signs, and alternative diagnoses. In the second, retrospective interpretation session, the standard-dose group (n = 100) consisted of 61 men and 39 women (male-to-female ratio, 1.56) with a mean age of 51.79 years, age range of 22–87 years, and mean BMI of 26.29. Three of the 100 patients were underweight; 34, normal weight; 48, overweight; and 15, obese. The low-dose group (n = 100) consisted of 64 men and 37 women (maleto-female ratio, 1.73) with a mean age of 49.04 years, age range of 18–87 years, and mean BMI of 24.84. Two of the 100 patients were underweight; 52, normal weight; 34, overweight; and 12, obese. These numbers for both the standard-dose and the

low-dose examinations reflected the similarity between the patient groups for the second interpretation session and for the original 300 examinations.

Final Clinical Diagnosis The final clinical diagnosis (urinary stones and other) was determined by review of the patients’ medical records in the hospital’s computerized clinical database. The diagnosis was confirmed by review of urology department records, reports of surgical procedures, confirmation with additional radiologic examinations (sonography, excretory urography, and MDCT with IV contrast material), hospital discharge records, clinical follow-up information, and confirmation of alternative diagnosis with relief of symptoms after specific medical or surgical treatment.

Statistical Analysis Statistical analysis was performed with commercially available statistical software (InStat, version 3.0, GraphPad). Comparison of calculated effective dose was done with a Kruskal-Wallis test between the four patient groups: standard and low dose for 6-MDCT and for 16-MDCT. Sensitivity, specificity, positive and negative predictive values, and accuracy were calculated for all reviewers. In-

555

Downloaded from www.ajronline.org by 37.44.207.115 on 01/28/17 from IP address 37.44.207.115. Copyright ARRS. For personal use only; all rights reserved

Mulkens et al. terobserver agreement between radiologists for the diagnosis of urinary stones was evaluated with kappa statistics for each interpretation session. Intraobserver agreement was evaluated for the 200 examinations of the same 200 patients (100 lowdose and 100 standard-dose examinations) interpreted by one reviewer in the first and second interpretation sessions. Statistical significance for all tests was set at p < 0.05.

Results Clinical Diagnosis Of the 300 patients, 158 (52.7%) had urinary stone disease. Eighty-four (28%) of the patients had a ureteral stone; 52 (17.3%), a ureteral stone in combination with one or more ipsilateral, contralateral, or bilateral renal stones; 16 (5.3%), only a renal stone or stones; and in six (2%) patients, a stone had already passed into the bladder. For the standard- and low-dose groups, respectively, a ureteral stone was present in 43 and 41 patients, combined ureteral and renal stones in 30 and 22 patients, renal stones in seven and nine patients, and a bladder stone in two and four patients. The difference in number of urinary stones between the standard- and lowdose groups was not significant (Fisher’s exact test, p = 0.42). The location of stones was as follows: 12 (7.6%) in the renal pelvis or at the pyeloureteral junction (Fig. 1), 31 (19.6%) in the lumbar ureter (Fig. 2), 41 (26%) in the pelvic ureter, 52 (32.9%) at the vesicoureteral junction (Fig. 3), and six (3.8%) in the bladder. Other stones (n = 16; 10.1%) were located in the kidney only. Of 142 patients with ureteral (n = 136) or bladder stones (n = 6), 66 (46.5%) had spontaneous passage of the stone, and 54 (38%) needed endoscopic stone extraction, 18 of these after initial unsuccessful extracorporeal shock wave lithotripsy. Twenty (14.1%) of the patients with stones underwent extracorporeal shock wave lithotripsy as the only therapy. Two (1.4%) needed percutaneous extraction of a large (> 10 mm) stone from the renal pelvis (Fig. 1). Of 136 ureteral stones, 70 were left-sided and 66 right-sided. The mean size (maximum diameter) of the ureteral stones was 4.55 mm (4.67 and 4.39 mm in the standard- and lowdose groups, respectively), with a size range of 1–11 mm for the two groups. Dilatation of the urinary tract proximal to the ureteral stone was present in 125 (88%) of the patients. Perirenal stranding was present in 79 (56%) of the patients, and four (3%) of the patients had urinoma formation (Fig. 4).

556

CT findings were normal in 110 (36.6%) of the patients, and no cause of abdominal pain was found. Other imaging studies confirmed the absence of urinary stones in 51 patients (abdominal radiograph with sonography [n = 24] or excretory urography [n = 27]). Fifty-nine patients had complete resolution of symptoms at clinical follow-up. An alternative diagnosis was made in 48 (16%) of the patients: 23 (15.3%) of the patients in the standard-dose group and 25 (16.6%) of those in the low-dose group. Diagnostic Performance of Standard- and Low-Dose MDCT Sensitivity, specificity, predictive values, and accuracy in detection of urinary stone disease for both interpretation sessions are summarized in Table 1. At the first interpretation session, the sensitivity and specificity of two experienced reviewers were high for both standard-dose (at least 97.6% and 94%, respectively) and low-dose (at least 97.3% and 93.5%, respectively) examinations. The overall accuracy was 96.7% for standard-dose and 95.3% for low-dose examinations. In the 300 examinations, the experienced reviewers together had a maximum of 10 cases of false-positive findings. One reviewer had six cases of false-positive findings in the low-dose group and four in the standard-dose group; the other had five cases of false-positive findings in the low-dose group and three in the standard-dose group. Although in six cases a second review of the examinations showed a tiny (≤ 3 mm) stone in the distal ureter, endoscopic examination performed 3 days to 2 weeks after MDCT showed no ureteral stone. The stone probably passed spontaneously between the CT examination and endoscopy, but these examination results were considered false-positive. In two patients, follow-up excretory urography and sonography did not depict a stone, but review of the CT examination confirmed the presence of a tiny distal ureteral stone. Because the patient did not recall spontaneous stone passage, these CT examination results were considered false-positive. In two slim patients, small pelvic phleboliths were misinterpreted as distal ureteral stones. The same two experienced reviewers had a maximum of four cases of false-negative findings in 300 examinations. One reviewer had one case of false-negative findings in the low-dose group and one in the standard-dose group; the other had two cases of false-negative findings in the low-dose group and two

in the standard-dose group. In one patient, a 2-mm stone at the intersection of the ureter and the iliac vessels probably was misinterpreted as a small calcified atheromatous plaque, was overlooked, and passed spontaneously 3 days after the examination (Fig. 5). In another patient, the urologist described grit in the bladder at endoscopy 1 week after CT that had not been identified on CT. In two patients a small pelvic ureteral stone was overlooked by one reviewer. Evaluation of the diagnostic performance of low-dose MDCT in overweight (n = 71) and obese (n = 25) patients, considered separate groups, showed high values. Low-dose examinations of overweight and obese patients interpreted by the two experienced reviewers had a sensitivity of 97% and 100%, specificity of 100% and 100%, and accuracy of 98% and 100%. Thirty-five patients in the low-dose group had a ureteral stone in the pelvic region or at the vesicoureteral junction. Phleboliths were visualized in 24 (68.5%) of these patients. Two slim patients (BMI, 19.81 and 20.11) in this group were the ones with false-positive findings in that small phleboliths were misinterpreted as ureteral stones. In the second interpretation session (n = 200), inexperienced reviewers reached a sensitivity at least 96.0% and a specificity of at least 90.2% for low-dose examinations with overall accuracy of at least 94.0%. In standard-dose examinations, they reached a sensitivity of at least 93.7%, specificity of 94.2% for both reviewers, and accuracy of 94% and 95% (Table 1). Interobserver agreement was excellent at both interpretation sessions. In the low-dose examinations, kappa values were 0.98 between experienced reviewers for the first interpretation session and 0.92 between inexperienced reviewers for the second interpretation session. For the standard-dose examinations, kappa values were 0.96 between experienced reviewers and 0.86 between inexperienced reviewers. Intraobserver agreement for one experienced reviewer of 200 examinations of 200 patients in both interpretation sessions was excellent: a kappa value of 0.96 for the low-dose examinations and of 0.98 for the standard-dose examinations. Dose Reduction With use of the tube current modulation system, mean effective tube current decreased in the standard-dose group to 71.75 mAs (6MDCT) and 88.6 mAs (16-MDCT), reductions of 25% and 26%, respectively (Table 2) compared with the reference levels of 95 and 120

AJR:188, February 2007

MDCT of Urinary Stone Disease

Downloaded from www.ajronline.org by 37.44.207.115 on 01/28/17 from IP address 37.44.207.115. Copyright ARRS. For personal use only; all rights reserved

Fig. 2—40-year-old man (height, 1.80 m; weight, 92 kg; BMI, 28.4) with severe rightsided acute renal colic. Low-dose16-MDCT examination with mean effective tube current of 59 mAs and calculated effective dose of 1.53 mSv. Coronal 5-mm image shows 7-mm stone (arrow) in distal lumbar ureter and obstructive dilatation (asterisk) of more proximal portion of ureter.

A

B

Fig. 3—25-year-old man (height, 1.78 m; weight, 104 kg; BMI, 32.82) with left-sided acute flank pain. Low-dose 16-MDCT examination with mean effective tube current of 66 mAs and effective dose of 1.72 mSv. A, Axial 5-mm MDCT image shows 1-mm distal stone (arrow) at junction of ureter with bladder. B, Coronal 5-mm MDCT image shows stone (arrow) depicted in A.

AJR:188, February 2007

557

Downloaded from www.ajronline.org by 37.44.207.115 on 01/28/17 from IP address 37.44.207.115. Copyright ARRS. For personal use only; all rights reserved

Mulkens et al. 1.58 mSv (6-MDCT: men, 1.41 mSv; women, 1.78 mSv) and 1.41 mSv (16-MDCT: men, 1.29; women, 1.83 mSv) (Table 2). Our low-dose protocols gave a mean dose reduction of 51.2% for 16-MDCT and 64.3% for 6-MDCT compared with our standarddose protocols. A Kruskal-Wallis test showed that these dose differences between the low and standard doses were statistically significant (p < 0.001). There was very good correlation between the mean effective tube current of each examination and BMI. The Spearman’s correlation coefficients were 0.85 and 0.86 for the standard-dose and 0.87 and 0.88 for the low-dose examinations. In addition, a wide, dynamic range of mean effective tube current values in both dose groups followed the range of BMI values. These values and ranges are summarized in Table 2.

Fig. 4—34-year-old woman (height, 1.69 m; weight, 69 kg; BMI, 24.45) with left-sided acute flank pain. Low-dose 6MDCT examination with mean effective tube current of 37 mAs and calculated effective dose of 1.77 mSv. Coronal 5mm image shows small stone (solid arrow) at vesicoureteral junction, secondary renal obstruction (open arrow), and urinoma (asterisk).

TABLE 1: Diagnostic Performance of Four Reviewers Using Standard- and Low-Dose MDCT in Detection of Urinary Stones Interpretation Session 1 Performance Characteristic

Interpretation Session 2a

Reviewer 1

Reviewer 2

Reviewer 3

Reviewer 4

Sensitivity (%)

98.8

97.6

95.8

93.7

Specificity (%)

94.0

95.5

94.2

94.2

Positive predictive value (%)

95.3

96.4

93.9

93.7

Negative predictive value (%)

98.4

97.0

96.1

94.2

Accuracy (%)

96.7

96.7

95.0

94.0

98.6

97.3

98.0

96.0

Standard-dose MDCT

Low-dose MDCT Sensitivity (%) Specificity (%)

93.5

93.5

90.2

92.0

Positive predictive value (%)

93.5

93.4

90.6

92.3

Negative predictive value (%)

98.6

97.3

97.9

95.8

Accuracy (%)

95.3

95.3

94.0

94.0

Note—Reviewers 1 and 2 were experienced radiologists; reviewers 3 and 4 were first-year residents. a Two hundred examinations: 100 low-dose and 100 standard-dose examinations with 5-mm consecutive axial and coronal images.

mAs. The same range of dose reduction was achieved in low-dose examinations: mean effective tube current decreased to 38.77 mAs (6MDCT) and 48.53 mAs (16-MDCT), reductions of 25% and 31% compared with reference levels of 51 and 70 mAs, respectively (Table 2).

558

The calculated mean effective doses for the standard-dose examinations were 4.43 mSv (6-MDCT: men, 4.37 mSv; women, 4.57 mSv) and 2.89 mSv (16-MDCT: men, 2.48 mSv; women, 3.51 mSv). For the low-dose examinations, the mean effective doses were

Alternative Diagnoses In our study, alternative diagnoses were found on MDCT in 23 (15.3%) of the patients receiving the standard dose and in 25 (16.6%) of the patients receiving the low dose. The diagnoses, confirmed at surgery or medical follow-up, were as follows: two cases of acute appendicitis and one of prostate carcinoma in each group, four ovarian masses (Fig. 6) or large (> 4 cm) ovarian cysts in the standard-dose group and six in the low-dose group, one case of pyelonephritis in the standard-dose group and two cases in the low-dose group, four cases of cholecystolithiasis in the standard-dose group and six in the low-dose group, and five cases of colon diverticulitis in the standard-dose group and three in the low-dose group. In the low-dose group there also were two renal tumors (Fig. 7), one abdominal aortic aneurysm, one case of malignant pleuritis, and one case of volvulus. In the standarddose group there also were one inguinal hernia, two cases of severe lumbar disk disease, one case of ureteral diverticulum, one case of rectal carcinoma, and one retroperitoneal tumor. Comparison of the number of alternative diagnoses between standard- and lowdose examinations showed no significant difference (Fisher’s exact test, p = 0.85). The experienced reviewers correctly identified all alternative diagnoses except one case of cholecystolithiasis, which was overlooked by one reviewer. At the second interpretation session, the number of alternative diagnoses found (15% in the standard-dose group and 16% in the low-dose group) was similar to that at the first

AJR:188, February 2007

Downloaded from www.ajronline.org by 37.44.207.115 on 01/28/17 from IP address 37.44.207.115. Copyright ARRS. For personal use only; all rights reserved

MDCT of Urinary Stone Disease Fig. 5—45-year-old man (height, 1.73 m; weight, 89 kg; BMI, 29.74) with leftsided flank pain. Standard-dose 6-MDCT examination at mean effective tube current of 79 mAs and calculated effective dose of 5.16 mSv. Coronal 5-mm image shows 2-mm ureteral stone (arrow) at intersection of ureter and iliac vessels that led to false-negative finding of small calcified atheromatous plaque.

interpretation session. Similar pathologic conditions were found most frequently: two ovarian masses in the standard-dose group and four in the low-dose group; three cases of cholecystolithiasis in the standard-dose group and two in the low-dose group; and two cases of acute appendicitis, two cases of colon diverticulitis, and one case of pyelonephritis in both groups. All alternative diagnoses were identified by the experienced reviewer at both standard-dose (n = 15) and low-dose (n = 16) examinations. Inexperienced reviewers, however, identified only 12 and 10 alternative diagnoses in the standard-dose group (12% and 10%) and five and four alternative diagnoses in low-dose examinations (5% and 4%). These differences in performance between the board-certified radiologists and the two residents were statistically significant in the low-dose examinations (McNemar test, p = 0.002 and 0.001) but only in one comparison in the standard-dose examinations (McNemar test, p = 0.125 and 0.031). Discussion The results of this study show that MDCT performed with reduced tube current resulted

AJR:188, February 2007

in a 51.2–64.3% reduction in radiation dose, compared with our standard-dose protocol, without a significant change in accuracy in the detection of urinary stone disease. With a sensitivity of 96–98.6%, specificity of 92–93.5%, and accuracy of 94–95.3%, the performance of our low-dose protocols was in the same range as that of our standard-dose examinations (Table 1) and was comparable with that found in studies of single-detector helical CT performed at a standard dose [1–5] and single-detector or MDCT performed at a low dose [8–12]. In those studies sensitivity was 93–100%; specificity, 94–100%; and accuracy, 93–98%. A disadvantage of standard unenhanced helical CT in the detection of urinary stone disease is the relatively high radiation dose [2, 7–12], higher than for excretory urography. The reported effective dose ranges from 2.8 to 13.1 mSv for men and from 4.5 to 18.0 mSv for women [10, 12]. This dose is higher than the mean effective dose reported for excretory urography, which is 1.5 mSv for three-film excretory urography [23] and 2.1 mSv for sixfilm excretory urography [24]. This dose level is of particular concern to young patients who

experience repeated stone development and thus may need numerous CT examinations over a lifetime [7, 11, 23]. Previous studies of low-dose single-detector CT and MDCT have shown mean effective dose ranges of 0.97 mSv in men to 1.50 mSv in women for single-detector CT at 120 kV, 70 mA, and a pitch of 2 [8, 9] and 1.2 mSv in men to 1.9 mSv in women with a 4-MDCT at 120 kV, 30 mAs effective tube current, and pitch factor of 1.5 [12]. The calculated mean effective dose values of our low-dose examinations were comparable: 1.21–1.41 mSv for men and 1.78–1.83 mSv for women. A limitation of previous low-dose studies with a fixed tube current was that the protocols were not suited for large or obese patients [8–12]. Heneghan et al. [11] excluded patients weighing more than 90 kg. Hamm et al. [8] and Knopfle et al. [9] recommended use of standard-dose CT for obese patients to achieve adequate image quality. For obese patients, Tack et al. [12] obtained an additional pelvic scan at 60 mAs to supplement the low-dose protocol at 30 mAs. Kalra et al. [25] suggested that there is a linear correlation between patient size and image quality and that the equivalent reduction in radiation dose for lighter and heavier patients may not provide acceptable image quality for the heavier patients. The use of a 4D tube current modulation system reduced mean tube current from 25% to 31% in comparison with protocols of fixed tube current in both standard- and low-dose examinations. This reduction is comparable with a reduction of 32% achieved in a study [15] of this tube current mechanism in routine abdominopelvic CT examinations and reflects the consistency of the mechanism with different protocols in the same anatomic region. The dose reduction of our low-dose examinations with tube current modulation was in the same range (43–66%) as that reported by Kalra et al. [26] with z-axis tube current modulation in 22 patients with urinary stones undergoing 16-MDCT at a low dose. Although 4D tube current modulation yields an overall reduction in mean tube current, mean effective tube current goes above the reference tube current for obese patients. In our study the reference effective tube current was based on subjective good image quality at low dose for normal-sized patients by consensus of two experienced radiologists at the start of the study. The choice of the reference tube current level, and therefore image quality, is user-dependent in tube current modulation systems [14, 15]. For obese pa-

559

Mulkens et al. TABLE 2: Data on Standard-Dose and Low-Dose MDCT Examinations with 4D Tube Current Modulation Effective Tube Current (mAs)

Downloaded from www.ajronline.org by 37.44.207.115 on 01/28/17 from IP address 37.44.207.115. Copyright ARRS. For personal use only; all rights reserved

BMI

Mean

Range

Mean

Range

Correlation Between Effective Tube Current and BMI

Mean

Range

6-MDCT standard dose

130

95

71.75

30–120

4.43

2.03–7.84

0.85

27.70

16.82–39.79

16-MDCT standard dose

120

120

88.6

51–149

2.89

1.49–5.75

0.86

25.99

16.98–44.92

Technique and Type of Dose

Tube Voltage Reference (kV)

Effective Dose (mSv)

6-MDCT low dose

110

51

38.77

23–64

1.58

0.79–2.85

0.87

25.44

17.43–39.43

16-MDCT low dose

120

70

48.53

28–81

1.41

0.73–2.70

0.88

24.37

18.20–35.88

Note—BMI = body mass index (kg/m2).

Fig. 6—37-year-old woman (height, 1.62 m; weight, 84 kg; BMI, 32.01) with left-sided acute flank pain. Low-dose 6-MDCT examination with mean effective tube current of 52 mAs and calculated effective dose of 2.81 mSv. Coronal 5-mm image shows bilateral ovarian teratomas or dermoid cysts (asterisks) and larger, left teratoma compressing left pelvic ureter.

tients, tube current was raised to a mean effective tube current of 64 mAs (46-year-old woman with a BMI of 39.43) for 6-MDCT and to a mean effective tube current of 81 mAs (64-year-old man with a BMI of 35.88) for 16-MDCT, which corresponds to calculated effective doses of 2.85 and 2.7 mSv, respectively. One can discuss whether these values can be called low doses, but we believe that these are low dose levels for obese patients. Huda et al. [27] found that with phantoms of different sizes scanned with constant kilovoltage and effective tube current, the energy imparted with CT increases with patient size, but the corresponding effective dose is higher for smaller phantoms than for larger ones. Because pelvic organs (bladder, colon,

560

and gonads), which account for a considerable part of the effective dose, are close to the center of the pelvis, the effective dose should be lower in obese patients than in underweight ones. Tack et al. [12] therefore concluded that an increase in effective tube current settings in patients with a high BMI may not result in a higher effective radiation dose. Two inexperienced first-year residents reached high sensitivity, specificity, and overall accuracy in the detection of urinary stones at low-dose examinations. This finding reflects a short learning curve in visualization of urinary stones with CT, even at lower doses. This finding is supported by that of Rosser et al. [28] showing a small but real learning curve in CT evaluation of acute flank

pain. Excellent interobserver agreement (κ = 0.86–0.98) in both standard- and lowdose examinations reflects that CT is sensitive in the detection of urinary stones. Almost all stones are of sufficient X-ray attenuation to be readily visible, even with reduced effective tube current and a subsequent increase in image noise [7, 11, 12, 26]. The residents scored worse, however, than experienced reviewers in identifying alternative diagnoses, especially at low-dose examinations, and the difference was statistically significant in these examinations. In a study by Holdgate and Chan [29], emergency clinicians and radiology residents scored well compared with board-certified radiologists in identifying urinary tract calculi but poorly in detection of nonrenal abnormalities. Potentially important pathologic findings were missed in 3% of cases. Review of 1,000 unenhanced CT examinations for urinary stone disease yielded 10% alternative or additional diagnoses [30]. In slim patients without much intraabdominal and intrapelvic fat, the CT signs can be especially subtle and difficult to identify, increasing the risk of missing clinically significant alternative diagnoses [13, 30]. This issue is important because such identification is probably the most powerful advantage of unenhanced CT in the evaluation of suspected renal colic [2, 6, 13, 29, 30]. The presence or absence of a urinary stone is the primary but not the only issue. The delay in diagnosis or potential misdiagnosis that can result from missing an alternative diagnosis does not justify the theoretic benefit of dose reduction [13]. Moreover, there is a tendency toward increased use of unenhanced helical CT in the emergency evaluation of patients with acute abdominal pain or flank pain and those with no typical renal colic [31]. Other urinary diseases and numerous extraurinary abdominal diseases cause symptoms that can be mistaken for urinary colic, and some of these conditions

AJR:188, February 2007

Downloaded from www.ajronline.org by 37.44.207.115 on 01/28/17 from IP address 37.44.207.115. Copyright ARRS. For personal use only; all rights reserved

MDCT of Urinary Stone Disease

A

B

Fig. 7—24-year-old man (height, 1.79 m; weight, 75 kg; BMI, 23.41) with right-sided acute flank pain and macroscopic hematuria. Low-dose 16-MDCT examination with mean effective tube current of 44 mAs and calculated effective dose of 1.15 mSv. A, Coronal 5-mm MDCT image shows evidence of mass lesion (arrows) in region of calyces of upper pole of right kidney. B, Coronal 5-mm conventional renal CT scan after IV administration of iodinated contrast material confirms presence of semisolid tumor mass (asterisk). Right nephrectomy was performed, and pathologic finding was carcinoma of collecting duct (Bellini’s tubule).

may require immediate medical attention [2, 8, 9, 11, 12, 29, 30, 32]. Although experienced radiologists performed well in identifying alternative diagnoses in our low-dose group and in subjects in other studies in which low doses were used [8, 9, 12], unenhanced CT at both standard and low doses remains a limited examination. Without the use of oral or IV contrast material, the ability to diagnose visceral ischemia and infarction, visceral inflammation, mass lesions in solid viscera, and nonopacified bowel is very limited [13, 29, 30]. Additional complete CT examination with gastrointestinal and IV contrast administration is needed in unequivocal cases with persistence of clinical symptoms [29, 30]. A limitation of our study was that we compared standard- and low-dose findings in similar groups of patients and did not compare standard- and low-dose protocols in the same patients at the same time, as in the study of Heneghan et al. [11] or compare low-dose

AJR:188, February 2007

protocols with and without tube current modulation in the same patients. A same-patient method would have been ideal, but it does not seem ethical to use supplemental radiation in a population with a large number of young and otherwise healthy patients. Another limitation of our study was that we selected 200 examinations and did not review all images (only 5-mm axial and coronal images) in the second interpretation session. Therefore, multiplanar reformation, cine viewing, curved reformation, and maximal intensity projection were limited in the second interpretation session. Although this factor did not seem to be a problem for the diagnostic accuracy of the experienced radiologist, these extra resources might have helped the residents in their diagnosis of urinary stones and alternative diagnoses. Although 5-mm-thick images are frequently used in unenhanced CT for urolithiasis [2–5, 7–11], results of one study [32] confirmed that 3-mm slice thickness is better, be-

cause small (< 3 mm) calculi can be missed in 5-mm-thick sections. We did not compare our standard- or lowdose MDCT examinations with other imaging techniques. Catalano et al. [33] reported a high accuracy of 83% with a combination of standard radiography and sonography in the detection of ureterolithiasis. Although it has a somewhat lower accuracy, this combination can be an alternative for young patients and women because of the very low radiation dose and very low false-positive rate. False-negative findings on sonography are mostly caused by small stones, most of which pass spontaneously [5, 6, 33]. A problem with validating the accuracy of low-dose MDCT in evaluation for suspected urinary colic is that CT itself has become the reference standard [12, 13]. For this reason we compared low-dose with standard-dose CT and used tube current modulation in both protocols. Our search of the medical literature yielded no published results of evaluation of

561

Downloaded from www.ajronline.org by 37.44.207.115 on 01/28/17 from IP address 37.44.207.115. Copyright ARRS. For personal use only; all rights reserved

Mulkens et al. this modulation technique in a comparison of low and standard radiation doses in CT for urinary stone disease. In conclusion, our study showed that lowdose MDCT with tube current modulation can be used as a standard procedure for evaluation of patients with suspected acute renal colic. The use of a tube current modulation mechanism adapts the tube current to the patient’s anatomic configuration and size; therefore, even overweight and obese patients can be examined with low doses of radiation. Acknowledgments We thank Steffen Fieuws for his advice and help in statistical evaluation of the data, Stijn Boulanger for help in collecting the patient records, and the CT technicians for performing the examinations and archiving the CT data.

9.

10.

11.

12.

13.

References 1. Smith RC, Rosenfield AT, Choe KA, et al. Acute flank pain: comparison of noncontrast-enhanced CT and excretory urography. Radiology 1995; 194:789–794 2. Smith RC, Verga M, McCarthy S, Rosenfield AT. Diagnosis of acute flank pain: value of unenhanced helical CT. AJR 1996; 166:97–101 3. Miller OF, Rineer SK, Reichard SR, et al. Prospective comparison of unenhanced helical CT and excretory urogram in the evaluation of acute flank pain. Urology 1998; 52:982–987 4. Dalrymple NC, Verga M, Anderson KR, et al. The value of unenhanced helical computerized tomography in the management of acute flank pain. J Urol 1998; 159:735–740 5. Boulay I, Holtz P, Foley WD, White B, Begun FP. Ureteral calculi: diagnostic efficacy of helical CT and implications for treatment of patients. AJR 1999; 172:1485–1490 6. Dalla Palma L, Pozzi-Mucelli R, Stacul F. Presentday imaging of patients with renal colic. Eur Radiol 2001; 11:4–17 7. Spielmann AL, Heneghan JP, Lee LJ, Yoshizumi T, Nelson RC. Decreasing the radiation dose for renal stone CT: a feasibility study of single-and MDCT. AJR 2002; 178:1058–1062 8. Hamm M, Knopfle E, Wartenberg S, Wawroschek

562

14.

15.

16.

17.

18.

19.

20.

F, Weckermann D, Harzmann R. Low dose unenhanced helical computerized tomography for the evaluation of acute flank pain. J Urol 2002; 167:1687–1691 Knopfle E, Hamm M, Wartenberg S, Bohndorf K. CT in ureterolithiasis with a radiation dose equal to excretory urography. Rofo 2003; 175:1667–1672 Meagher T, Sukumar VP, Collingwood J, et al. Low dose CT in suspected acute renal colic. Clin Radiol 2001; 56:873–876 Heneghan JP, McGuire KA, Leder RA, DeLong DM, Yoshizumi T, Nelson RC. Helical CT for nephrolithiasis and ureterolithiasis: comparison of conventional and reduced radiation-dose techniques. Radiology 2003; 229:575–580 Tack D, Sourtzis S, Delpierre I, de Maertelaer V, Gevenois PA. Low-dose unenhanced MDCT of patients with suspected renal colic. AJR 2003; 180:305–311 Katz DS, Venkatarantanan N, Napel S, Graham Sommer F. Can low-dose unenhanced MDCT be used for routine evaluation of suspected renal colic? AJR 2003; 180:313–315 Kalra MK, Maher MM, Toth TL, Hamberg LM, Blake MA, Shepard JA. Strategies for radiation dose optimization. Radiology 2004; 230:619–628 Mulkens TH, Bellinck P, Baeyaert M, et al. Use of an automatic exposure control mechanism for dose optimization in multidetector row CT examinations: clinical evaluation. Radiology 2005; 237:213–223 [No authors listed]. Health implications of obesity: National Institutes of Health Consensus Development Conference Statement. Ann Intern Med 1985; 103:1073–1077 Kalender WA, Wolf H, Suess C, Gies M, Greess H, Bautz WA. Dose reduction in CT by online tube current control: principles and validation on phantoms and cadavers. Eur Radiol 1999; 9:323–328 Greess H, Wolf H, Baum U, et al. Dose reduction in CT by attenuation-based online modulation of tube current: evaluation of six anatomic regions. Eur Radiol 2000; 10:391–394 Mahesh M, Scatarige JC, Cooper J, Fishman EK. Dose and pitch relationship for a particular MDCT scanner. AJR 2001; 177:1273–1275. Silverman FM, Kalender WA, Hazle JD. Common terminology for single and multidetector helical CT. AJR 2001; 176:1135–1136

21. Kalender WA, Schmidt B, Zankl M, Schmidt M. A PC program for estimating organ dose and effective dose values in CT. Eur Radiol 1999; 9:555–562 22. [No authors listed.] ICRP publication 60: 1990 recommendations of the International Commission on Radiological Protection. Ann ICRP 1991; 21:1–3 23. Denton ER, Mackenzie A, Greenwell T, Popert R, Rankin SC. Unenhanced helical CT for renal colic: is radiation dose justifiable? Clin Radiol 2000; 55:409–410 24. Wall BF, Hart D. Revised radiation doses for typical X-ray examinations. Br J Radiol 1997; 70:437–439 25. Kalra MK, Prasad S, Saini S, et al. Clinical comparison of standard-dose and 50% reduced-dose abdominal CT: effect on image quality. AJR 2003; 179:1101–1106 26. Kalra MK, Maher MM, D’Souza RV, et al. Detection of urinary tract stones at low-radiation-dose CT with z-axis automatic tube current modulation: phantom and clinical studies. Radiology 2005; 235:523–529 27. Huda W, Atherton JV, Ware DE, Cumming WA. An approach for the estimation of effective dose at CT in pediatric patients. Radiology 1997; 203:417–422 28. Rosser CJ, Zagoria R, Dixon R, et al. Is there a learning curve in diagnosing urolithiasis with noncontrast helical CT? Can Assoc Radiol J 2000; 51:177–181 29. Holdgate A, Chan T. How accurate are emergency clinicians at interpreting noncontrast CT for suspected renal colic? Acad Emerg Med 2003; 4:315–319 30. Katz DS, Scheer M, Lumerman JH, Mellinger BC, Stillman CA, Lane MJ. Alternative or additional diagnoses on unenhanced helical CT for suspected renal colic: experience with 1000 consecutive examinations. Urology 2000; 56:53–57 31. Chen MY, Zagoria RJ, Saunders HS, Dyer RB. Trends in use of unenhanced helical CT for acute urinary colic. AJR 1999; 173:1447–1450 32. Memarsadeghi M, Heinz-Peer G, Helbich TH, et al. Unenhanced multidetector row CT in patients suspected of having urinary stone disease: effect of section width on diagnosis. Radiology 2005; 235:530–536 33. Catalano O, Nunziata A, Altei F, Siani A. Suspected urinary colic: primary helical CT versus selective helical CT after unenhanced radiography and sonography. AJR 2002; 178:379–387

AJR:188, February 2007