Ureteral Calculi Detection Using Low Dose Computerized Tomography Protocols is Compromised in Overweight and Underweight Patients

Urolithiasis/Endourology

Ureteral Calculi Detection Using Low Dose Computerized Tomography Protocols is Compromised in Overweight and Underweight Patients Jonathan P. Heldt, Jason C. Smith, Kirk M. Anderson, Gideon D. Richards, Gautum Agarwal, Damien L. Smith, Amy Schlaifer, Nicholas T. Pittenger, Daniel S. Han, Brenton D. Baldwin, Gabriel T. Schroeder and D. Duane Baldwin*,† From the Departments ofUrology and Radiology (JCS), Loma Linda University Medical Center, Loma Linda, California

Abbreviations and Acronyms BMI ⫽ body mass index CT ⫽ computerized tomography mAs ⫽ milliampere-seconds mSv ⫽ millisievert Submitted for publication October 20, 2011. * Correspondence: Department of Urology, Loma Linda University School of Medicine, 11234 Anderson St., Room A560, Loma Linda, California 92354 (telephone: 909-558-4196; FAX: 909-5584806; e-mail: [email protected]). † Financial interest and/or other relationship with Onset Medical.

See Editorial on page 12.

Purpose: Low dose computerized tomography protocols have demonstrated a reduction in radiation exposure while maintaining excellent sensitivity and specificity in the detection of stones in patients of average size. Low dose computerized tomography protocols have not yet been evaluated in subjects in the extremes of weight. We evaluated the effect of body weight when using low dose protocols to detect ureteral calculi. Materials and Methods: Three cadavers of increasing weight (55, 85 and 115 kg) were prepared by inserting 721 calcium oxalate stones (range 3 to 7 mm) in 33 random configurations into urinary tracts. Cadavers were then scanned using a GE LightSpeed® at 7 radiation settings. An independent, blinded review by a radiologist was conducted to generate ROC curves, with areas under the curve compared using a 1-way ANOVA (␣ ⫽ 0.05). Results: Sensitivity and specificity were significantly lower in the low and high weight cadavers compared to the medium weight cadaver at 5 mAs (p ⬍0.001) and 7.5 mAs (p ⫽ 0.048). Differences in sensitivity and specificity at radiation settings of 15 mAs or greater were not significant. Conclusions: The sensitivity and specificity for the detection of ureteral calculi on computerized tomography were decreased for underweight and overweight subjects when using extremely low dose radiation settings (less than 1 mSv). Low dose protocols of 15 mAs (2 mSv) can still be used for these subjects without jeopardizing the ability to identify ureteral stones. Key Words: tomography, x-ray computed; nephrolithiasis; ureteral calculi; diagnostic imaging; radiation dosage

124

www.jurology.com

SINCE its introduction in 1995 as a method of detecting urolithiasis,1 unenhanced helical CT has become the de facto standard in the diagnosis of urinary calculi. Concomitant with the increasing use of CT are increasing concerns regarding the potentially harmful effects of ionizing radiation, most significantly the risk of propagating malignant transformation. In response to these concerns low dose

CT protocols (protocols which deliver an effective dose of 3 mSv or less)2,3 are being evaluated and implemented across a broad range of medical specialties.4 –7 Before the adoption of low dose protocols it is important to consider the potential disadvantages of implementing low dose CT standards. In particular, the clarity of radiographic images, which is inversely proportional to the

0022-5347/12/1881-0124/0 THE JOURNAL OF UROLOGY® © 2012 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION

Vol. 188, 124-129, July 2012 Printed in U.S.A. DOI:10.1016/j.juro.2012.02.2568

AND

RESEARCH, INC.

PATIENT WEIGHT AND STONE DETECTION WITH LOW DOSE COMPUTERIZED TOMOGRAPHY

amount of radiation used in milliamperes,8 could be jeopardized, thereby affecting diagnostic accuracy. However, we previously demonstrated in an average weight cadaver (weight 73 kg, BMI 26.1 kg/m2) that there were no significant differences in the accuracy of ureteral calculi detection between standard dose CT (140 mAs, or approximately 19 mSv effective dose) and ultra low dose protocols (7.5 mAs or 0.95 mSv).9 However, it remains unclear whether similar diagnostic accuracy is applicable across varying patient body types. Increasing adiposity (commonly defined by such metrics as weight, circumference and BMI) has previously been shown to have a significant role in diminishing image quality in abdominal CT.6,10 –13 Additionally, we hypothesized that a small amount of periureteral adipose tissue may be beneficial by isolating the ureter from surrounding structures, thereby improving ureteral identification and stone detection. Therefore, the diagnostic accuracy of low dose protocols for the detection of ureteral calculi must be tested in overweight and underweight body types. While several studies have examined this relationship using a phantom, to our knowledge no previously published study has used a cadaveric model to identify a radiation dose which does not compromise diagnostic accuracy for patients at the extremes of weight. We compared the sensitivity and specificity of ureteral calculi detection using low dose CT protocols across various body sizes in a single-blind, prospective, randomized controlled trial using a cadaveric model.

MATERIALS AND METHODS Three cadaveric vehicles of incrementally increasing weight were acquired from the Department of Anatomic Pathology. The physical attributes of each cadaver were determined. Nine collecting systems (including kidneys, ureters and bladder) were acquired and placed into each of the 3 cadavers. After collecting system placement into the cadaver, the ureters were implanted with 0 to 3 calcium oxalate stones from a collection of 37 different stones. The calculi ranged from 3 to 7 mm in diameter and were placed in 33 random configurations to create 2,162 individual stone patterns for detection. After the placement of each stone into the ureter, water based gel was placed around the ureter and stone to create a stone-fluid interface. All cadavers were handled in accordance with institutional policies for proper conduct and treatment of cadaveric specimens. Abdominal and pelvic CT was performed using a GE LightSpeed VCT 64-slice tomographic scanner with settings of 140 kVp, variable mAs settings, collimation 0.625 mm, 2.5 mm axial reconstruction and pitch 1.375. Cadavers were scanned in natural configurations to ensure the absence of previously existing urinary calculi, metal clips or other objects which would interfere with the accurate diagnosis of implanted stones. Cadavers were positioned

125

using a laser guided gantry to ensure inclusion of the entire urinary system on CT. Scans of each cadaver and each different stone configuration were completed at 7 radiation settings of 140, 70, 50, 30, 15, 7.5 and 5 mAs. Overall 693 CT series were generated with 181 images in each, creating a total of 125,433 reconstructed images. An independent review was conducted by a blinded radiologist using an IMPAX® diagnostic workstation. The stone configuration images at differing radiation settings were randomized before radiologist review. The radiologist’s findings were then compared with known stone size to test the sensitivity and specificity of detection at each radiation setting and in each cadaver. Specificity and sensitivity data were used to generate ROC curves, with the areas under the curve compared using a 1-way ANOVA. In addition, 95% CIs were determined for sensitivity and specificity, and significance was established at ␣ ⫽ 0.05.

RESULTS Patient weight and BMI increased progressively in each of the 3 cadavers. Sensitivity for the detection of stones was significantly different at 5 mAs, at 94% in the 85 kg cadaver vs 85% and 66% in the 55 and 115 kg cadavers, respectively (p ⬍0.001). Specificity was 97% in the 85 kg cadaver vs 89% and 92% in the 55 and 115 kg cadavers, respectively. Sensitivity was not significantly different at the 15 mAs setting, at 96% for the 85 kg cadaver vs 91% and 86% in the 55 and 115 kg cadavers, respectively. Specificity was not significantly affected by increasing body weight (table 1). Sensitivities for 3 and 5 mm stone detection at each of the 7 radiation dose settings are shown in figure 1. The sensitivity was greater than 84% for the average weight cadaver regardless of stone size or radiation setting. However, the underweight and overweight cadaver had lower or equivalent sensitivities for all stone sizes at each radiation setting. The 7 mm calculi were detected with greater than 94% sensitivity at all radiation settings (not illustrated). The sensitivity of stone detection for 5 mm stones at the 7.5 mAs setting was 94%, 97% and 91% for the 55, 85 and 115 kg cadavers, respectively. The greatest variability existed with the 3 mm stones with sensitivities of 68%, 86% and 40% in the 55, 85 and 115 kg cadavers, respectively, at 7.5 mAs. At the highest radiation setting of 140 mAs, sensitivities for 3 mm stone detection in the 55, 85 and 115 kg cadavers were 84%, 95% and 93%, respectively. AUC values for each cadaver stratified by radiation setting are shown in table 2. Overall AUC values were 91.8% for the 55 kg cadaver, 95.9% for the 85 kg cadaver and 89.6% for the 115 kg cadaver (p ⬍0.001). AUC values were significantly lower in the low and high weight cadavers compared to the medium weight cadaver at 5 mAs (p ⬍0.001) and 7.5 mAs (p ⫽ 0.048). There was no significant difference

126

PATIENT WEIGHT AND STONE DETECTION WITH LOW DOSE COMPUTERIZED TOMOGRAPHY

Table 1. Sensitivity and specificity of detection for cadavers of varying body weight as a function of radiation used during unenhanced helical CT Sensitivity (95% CI)

Specificity (95% CI)

Radiation (mAs)

No. Stone Configurations (55, 85, 115 kg)

55 kg

85 kg

115 kg

55 kg

85 kg

115 kg

5 7.5 15 30 50 70 140

192, 174, 198 192, 174, 198 186, 174, 198 192, 174, 198 192, 174, 198 186, 180, 198 192, 180, 198

0.85 (0.77–0.91) 0.86 (0.78–0.92) 0.91 (0.83–0.96) 0.94 (0.87–0.98) 0.96 (0.90–0.99) 0.93 (0.85–0.97) 0.94 (0.87–0.98)

0.94 (0.87–0.98) 0.94 (0.87–0.98) 0.96 (0.89–0.99) 0.96 (0.89–0.99) 0.97 (0.90–0.99) 0.99 (0.94–1.0) 0.98 (0.92–1.0)

0.66 (0.56–0.75) 0.75 (0.65–0.82) 0.86 (0.78–0.92) 0.90 (0.82–0.95) 0.94 (0.88–0.98) 0.90 (0.82–0.95) 0.96 (0.90–0.99)

0.89 (0.80–0.94) 0.91 (0.83–0.96) 0.93 (0.84–0.97) 0.90 (0.81–0.95) 0.94 (0.87–0.98) 0.95 (0.88–0.99) 0.96 (0.88–0.99)

0.97 (0.90–1.0) 0.95 (0.86–0.98) 0.97 (0.90–1.0) 0.96 (0.88–0.99) 0.93 (0.84–0.97) 0.94 (0.85–0.98) 0.94 (0.85–0.98)

0.92 (0.84–0.97) 0.98 (0.92–1.0) 0.97 (0.90–0.99) 0.93 (0.86–0.97) 0.93 (0.86–0.97) 0.93 (0.86–0.97) 0.98 (0.92–1.0)

in the AUC values among the 3 cadavers at 15 mAs or greater (fig. 2). Regardless of body habitus, proximal ureteral stones were detected at a significantly higher rate compared to mid and lower ureteral stones (95.7% vs 92% and 89.5%, respectively). A visual comparison of the quality of the images of a 5 mm stone generated from each cadaver at each of 3 radiation settings (5, 30 and 140 mAs) is shown in figure 3.

DISCUSSION CT comprises 22% of medical imaging, yet it accounts for nearly 50% of the annual cumulative effective radiation dose from all medical imaging procedures.2 As the use of CT continues to increase, there is growing concern of a parallel increasing risk of radiation induced neoplasms.14 These concerns are justified as it is estimated that nearly 50% of all radiation received by the United States population is a direct result of medical imaging.15 Much of this radiation is attributable to computerized tomography, as CT provides the highest radiation dose of any medical imaging modality, with a typical effective dose for abdominal and pelvic CT of approximately 10.0 mSv, roughly the same radiation dose as 1,000 standard chest x-rays.9,15 For patients undergoing repeat scans the risk of malignancy is even greater.16,17 Current estimates place the chance of cancer at 1 in 200 for every 100 mSv received and,

A

indeed, an increased incidence has been noted in epidemiological studies.18 In response to these concerns the International Commission on Radiological Protection has set guidelines for radiation exposure for the general population at 1 mSv per year. Despite these warnings, a significant number (greater than 2%) of patients are still exposed to high (20 to 50 mSv) or very high (more than 50 mSv) doses of radiation on a yearly basis in the medical community.2 Patients being evaluated for urinary stones often undergo multiple CT scans, which places them in a higher risk group for the development of radiation induced neoplasms. In a meta-analysis Niemann et al showed that low dose CT protocols (defined as less than 3 mSv effective dose applied) had a pooled sensitivity of 0.966 (95% CI 0.950 – 0.978) and a specificity of 0.949 (95% CI 0.920 – 0.970) in the diagnosis of urolithiasis, which compares favorably to the 94% to 100% sensitivity and 97% specificity of normal dose CT.3 This finding suggests the safety and efficacy of low dose CT for the diagnosis of ureteral calculi in most patients. Nevertheless, the diagnostic accuracy of low dose CT must be demonstrated across the full spectrum of patients before the widespread adoption of low dose protocols can occur. It has not been previously shown in a direct comparison study that low dose CT can identify urinary calculi in patients at the ex-

B 1.00

1.00 55kg

0.60

86kg

0.40

114kg

0.20

Sensitivity

Sensitivity

0.80

0.80

55kg

0.60

86kg

0.40

114kg

0.20

0.00

0.00 5

7.5

15

30

50

Radiation (mAs)

70

140

5

7.5

15

30

50

70

140

Radiation (mAs)

Figure 1. Sensitivities for detection of 3 mm (A) and 5 mm (B) calculi at incrementally increasing radiation settings

PATIENT WEIGHT AND STONE DETECTION WITH LOW DOSE COMPUTERIZED TOMOGRAPHY

Radiation (mAs)

55 kg

85 kg

115 kg

p Value

5 7.5 15 30 50 70 140

0.88 0.88 0.92 0.92 0.95 0.94 0.94

0.96 0.94 0.97 0.96 0.95 0.96 0.96

0.80 0.85 0.91 0.91 0.94 0.91 0.97

⬍0.001* 0.048* 0.13 0.15 0.75 0.15 0.65

* Statistically significant.

tremes of weight with the same accuracy as in the average sized patient. However, there is limited evidence from prior studies that increased BMI may decrease the diagnostic accuracy of low dose CT. In a study on the use of low dose CT in the evaluation of flank pain Hamm et al reported that obesity appeared to significantly hinder the ability to accurately diagnose stones in 2 patients with a BMI greater than 31 kg/m2.12 Tack et al noted that only 1 of 6 patients with a BMI greater than 35 kg/m2 was accurately diagnosed using low dose (30 mAs) scans.6 In addition, Poletti et al reported that low dose CT achieved 95% sensitivity and 97% specificity for detecting ureteral calculi in patients with a BMI less than 30 kg/m2, while those with a BMI of 30 kg/m2 or greater had only 50% sensitivity and 89% specificity.13 These findings are consistent with the results of the current study, supporting the idea that increased adiposity negatively affects the diagnostic accuracy of CT in detecting ureteral calculi. While patient body habitus is frequently described in terms of BMI, several studies have suggested that alternative metrics of body size may be superior in determining effective radiation doses to ensure diagnostic accuracy. Menke evaluated 8 body size parameters including height, weight, weight-toheight ratio, BMI, body surface area, body diameter (in coronal and sagittal cross sections) and body circumference to determine which metric was most suitable for individual dose adaptation in adult body CT.11 Of these 8 factors body weight and circumference were the most accurate compared to the other parameters. Kalra et al noted a statistically significant difference (p ⫽ 0.0005) in image quality when comparing the standard dose to 50% reduced dose abdominal CT for patients weighing more than 180 lb (81 kg), while no significant difference was found for patients below this threshold of weight.19 In a followup study by the same group, patient weight was noted to have an inverse correlation with image quality (⫺0.46, p ⫽ 0.003).10 Importantly patient weight is a simple metric to use clinically as it is easily obtainable from the patient history, and re-

quires no calculations, additional radiographic imaging or invasive techniques to assess. Therefore, patient body weight was selected as the primary input variable when selecting cadavers for our study. Our results support the existing literature that body weight has a significant role in the radiation dose required for accurate diagnosis of ureteral calculi using unenhanced multidetector CT. At any given radiation level the highest weight cadaver consistently had the lowest AUC for ureteral calculi detection, although this difference was only significant at the lowest radiation settings (5 and 7.5 mAs). However, of particular interest was our finding that the lowest weight cadaver demonstrated lower AUC values than the average weight cadaver, which is contrary to what would be expected if there was a linear relationship between diagnostic accuracy and body weight. While seemingly counterintuitive, it is probable that the lack of perinephric and periureteral fat in this low body fat cadaver could have made the ureters more difficult for the radiologist to trace, thereby negating the beneficial effect of less abdominal fat. In addition, the presence of phleboliths in the pelvis can be mistaken for urinary calculi if the ureter has minimal periureteral adipose tissue to distinguish it from surrounding structures. Further studies will be necessary to validate this mechanistic hypothesis. Our study represents the first prospectively designed randomized controlled trial evaluating the effect of greater or less than average body weight in the detection of ureteral calculi using low dose CT protocols. However, these findings must be viewed in light of the limitations of this study. The most significant limitation of our study is that it was done using cadaveric tissues that had previously been fixed using formalin. While every effort was made to minimize differences in radiographic images between our cadaveric model and live patients, it is

1.00 Area Under ROC Curve

Table 2. Area under the ROC curve for cadavers of varying body weight as a function of radiation used during unenhanced helical CT

127

0.90 0.80

55-kg 85-kg

0.70

115-kg

0.60 0.50 5*

7.5*

15

30

50

70

140

Radiation (mAs)

Figure 2. Comparison of areas under ROC curve for each of 3 cadavers of varying body weight as function of radiation used during unenhanced helical CT to detect ureteral calculi. Asterisk denotes statistically significant differences.

128

PATIENT WEIGHT AND STONE DETECTION WITH LOW DOSE COMPUTERIZED TOMOGRAPHY

Figure 3. Comparison of image quality of 5 mm stone among 3 cadaveric models at increasing radiation settings. Insets show close-ups of stone.

possible that the final images obtained could differ from live tissue. Despite the potential for this form of error, a cadaveric model was used to provide a method of ascertaining exact stone size and location. Additionally, our previously published cadaveric study demonstrated accurate identification of distal ureteral calculi with sensitivity and specificities in accordance with retrospective studies of living patients.9 Another limitation of our study is that we were unable to assess the effect of body weight when using low dose CT to concurrently assess for additional diagnoses, which may be found in up to 10% of CT examinations for urinary stone disease.20 While previous studies have shown that low dose CT was 100% sensitive and specific for depicting nonurinary tract related disorders (6) in patients regardless of body size,13 our study does not address this question as our cadavers were screened specifically to ensure that no additional complicating comorbidities were present. Additionally, our study used only calcium oxalate stones because they are the most common stone seen clinically. Although the effects of stone composition in patients at the extremes of weight remain unknown, previous phantom studies demonstrated that calculi composition did not affect detection rates for calculi larger than 1 mm.21 Despite these limitations our study represents the first prospective randomized single-blind study of ureteral calculi detection using low dose CT in patients of

different body size using a cadaveric model, and provides insight into the use of low dose scans for subjects of greater or less than average body weight.

CONCLUSIONS In this cadaveric study low dose (2 mSv) CT protocols have been shown to diagnose ureteral calculi in the extremes of weight with excellent diagnostic accuracy. This resulted in an effective reduction in radiation of 89% compared to conventional settings. The sensitivity and specificity of stone detection are decreased for underweight and overweight subjects when using extremely low dose studies (less than 1 mSv). Ultra low dose protocols are appropriate for average weight subjects, while low dose protocols are recommended in underweight and overweight subjects presenting for evaluation of ureteral calculi in keeping with the principle of “as low as reasonably achievable.”

ACKNOWLEDGMENTS Drs. Bertha Escobar-Poni and Pedro Nava (Department of Anatomic Pathology, Loma Linda University) provided assistance. Udochukwu Oyoyo (consultant statistician, School of Public Health, Loma Linda University) designed and performed the statistical analysis.

PATIENT WEIGHT AND STONE DETECTION WITH LOW DOSE COMPUTERIZED TOMOGRAPHY

129

REFERENCES 1. Smith RC, Rosenfield AT, Choe KA et al: Acute flank pain: comparison of non-contrast-enhanced CT and intravenous urography. Radiology 1995; 194: 789.

8. Gurung J, Khan MF, Maataoui A et al: Multislice CT of the pelvis: dose reduction with regard to image quality using 16-row CT. Eur Radiol 2005; 15: 1898.

2. Fazel R, Krumholz HM, Wang Y et al: Exposure to low-dose ionizing radiation from medical imaging procedures. N Engl J Med 2009; 361: 849.

9. Jellison FC, Smith JC, Heldt JP et al: Effect of low dose radiation computerized tomography protocols on distal ureteral calculus detection. J Urol 2009; 182: 2762.

3. Niemann T, Kollmann T and Bongartz G: Diagnostic performance of low-dose CT for the detection of urolithiasis: a meta-analysis. AJR Am J Roentgenol 2008; 191: 396. 4. Graser A, Wintersperger BJ, Suess C et al: Dose reduction and image quality in MDCT colonography using tube current modulation. AJR Am J Roentgenol 2006; 187: 695. 5. Heneghan JP, McGuire KA, Leder RA et al: Helical CT for nephrolithiasis and ureterolithiasis: comparison of conventional and reduced radiation-dose techniques. Radiology 2003; 229: 575. 6. Tack D, Sourtzis S, Delpierre I et al: Low-dose unenhanced multidetector CT of patients with suspected renal colic. AJR Am J Roentgenol 2003; 180: 305. 7. Keyzer C, Tack D, de Maertelaer V et al: Acute appendicitis: comparison of low-dose and standard-dose unenhanced multi-detector row CT. Radiology 2004; 232: 164.

10. Kalra MK, Maher MM, Prasad SR et al: Correlation of patient weight and cross-sectional dimensions with subjective image quality at standard dose abdominal CT. Korean J Radiol 2003; 4: 234. 11. Menke J: Comparison of different body size parameters for individual dose adaptation in body CT of adults. Radiology 2005; 236: 565. 12. Hamm M, Knopfle E, Wartenberg S et al: Low dose unenhanced helical computerized tomography for the evaluation of acute flank pain. J Urol 2002; 167: 1687. 13. Poletti PA, Platon A, Rutschmann OT et al: Lowdose versus standard-dose CT protocol in patients with clinically suspected renal colic. AJR Am J Roentgenol 2007; 188: 927. 14. Brenner DJ and Hall EJ: Computed tomography–an increasing source of radiation exposure. N Engl J Med 2007; 357: 2277.

15. Frush DP: Radiation safety. Pediatr Radiol 2009; 39: 385. 16. Katz SI, Saluja S, Brink JA et al: Radiation dose associated with unenhanced CT for suspected renal colic: impact of repetitive studies. AJR Am J Roentgenol 2006; 186: 1120. 17. Jaffe TA, Gaca AM, Delaney S et al: Radiation doses from small-bowel follow-through and abdominopelvic MDCT in Crohn’s disease. AJR Am J Roentgenol 2007; 189: 1015. 18. Ron E: Ionizing radiation and cancer risk: evidence from epidemiology. Pediatr Radiol 2002; 32: 232. 19. Kalra MK, Prasad S, Saini S et al: Clinical comparison of standard-dose and 50% reduced-dose abdominal CT: effect on image quality. AJR Am J Roentgenol 2002; 179: 1101. 20. Katz DS, Scheer M, Lumerman JH et al: Alternative or additional diagnoses on unenhanced helical computed tomography for suspected renal colic: experience with 1000 consecutive examinations. Urology 2000; 56: 53. 21. Tublin ME, Murphy ME, Delong DM et al: Conspicuity of renal calculi at unenhanced CT: effects of calculus composition and size and CT technique. Radiology 2002; 225: 91.