Pseudoenhancement of Renal Cysts: Influence of Lesion Size, Lesion Location, Slice Thickness, and Number of MDCT Detectors

Transcription

1 Genitourinary Imaging Original Research Tappouni et al. MDCT of Renal Cysts Genitourinary Imaging Original Research Rafel Tappouni 1 Jennifer Kissane 2 Nabeel Sarwani 1 Erik B. Lehman 1 Tappouni R, Kissane J, Sarwani N, Lehman EB Keywords: CT, pseudoenhancement, renal cyst DOI: /AJR Received November 3, 2010; accepted after revision May 20, Department of Radiology, Penn State Hershey Medical Center, 500 University Dr, Hershey, PA Address correspondence to R. Tappouni (raftappouni@gmail.com). 2 Penn State University College of Medicine, Hershey, PA. AJR 2012; 198: X/12/ American Roentgen Ray Society Pseudoenhancement of Renal Cysts: Influence of Lesion Size, Lesion Location, Slice Thickness, and Number of MDCT Detectors OBJECTIVE. The purpose of this study was to determine the effect of renal cyst size and location on pseudoenhancement in human subjects. MATERIALS AND METHODS. Simple renal cysts obtained with 16-, 40-, and 64- MDCT scanners were analyzed for the presence of pseudoenhancement. Cyst size, location, and attenuation in the unenhanced and nephrographic phases were recorded. Pseudoenhancement was defined as an attenuation increase of 10 HU or more on nephrographic phase compared with unenhanced images. RESULTS. The pseudoenhancement rate was 22% (51/233). There was a statistically significant increase in the pseudoenhancement rate of lesions smaller than 10 mm (38/233) compared with those 10 mm and larger (13/233) (odds ratio, 6.4; p < ). Twelve of 62 cysts measuring mm exhibited pseudoenhancement. There was a statistically significant increase in the pseudoenhancement rate of central (39/53) compared with peripheral (12/51) cysts (odds ratio, 2.7; p < ). The pseudoenhancement rates for the 16-, 40-, and 64-MDCT scanners were 20%, 19%, and 26% with no statistically significant difference between them. CONCLUSION. Pseudoenhancement of renal cysts significantly correlates with size smaller than 1 cm and central location. Although pseudoenhancement increases with larger numbers of detectors, the correlation was not statistically significant. Cysts in the 1- to 1.5-cm range have a 19% likelihood of pseudoenhancement. R enal cysts are common in adults, and the prevalence increases with age to more than 27% in patients older than 50 years [1, 2]. When a cystic lesion meets all the criteria for being a simple renal cyst, no further evaluation or intervention is required. A crucial feature in characterizing cystic renal masses at CT is the presence of contrast enhancement. Simple renal cysts have a characteristic appearance of a spheric, sharply demarcated lesion with a smooth border on CT images. Furthermore, because they are fluid filled, cysts have attenuation similar to that of water (0 20 HU) and do not greatly enhance after IV contrast administration [3, 4]. Although simple cysts are not enhancing, they may have an increase in attenuation after contrast administration because of technical factors [3, 5, 6]. A less than 10 HU increase in attenuation is considered within the technical limits of the study and is not considered to represent enhancement [3, 7, 8]. Renal cyst pseudoenhancement is an artifactual increase in the attenuation of a simple renal cyst after contrast administration, even when the effects of partial volume averaging have been removed [5, 7, 9]. Technical and anatomic variations such as the part of the cyst measured, location of the cyst within the kidney, and adjacent renal or extrarenal structures can affect the attenuation measured [6, 10, 11]. In a retrospective study of the cases of 24 patients [7], attenuation in the renal cysts increased artifactually on contrast-enhanced images, but this pseudoenhancement was not substantial when the cyst was larger than 1.0 cm in diameter. Various authors [9, 12 17] have used phantom model studies to confirm the increased incidence of renal cyst pseudoenhancement with smaller cyst size (< 1.5 cm) and central location. These studies also showed that pseudoenhancement is increasingly a problem with higher numbers of CT detectors and higher voltage. Because of the AJR:198, January

2 Tappouni et al. TABLE 1: MDCT Parameters Scanner Peak Voltage (kvp) Tube Current (mas) Slice Thickness (mm) Collimation (mm) Rotation Time (s) Sensation 16-MDCT (Siemens Healthcare) (varied) Emotion 16-MDCT (Siemens Healthcare) (varied) Sensation 40-MDCT (Siemens Healthcare) (varied) Definition 64-MDCT (Siemens Healthcare) (varied) increasing use of 16-, 40- and 64-MDCT scanners in clinical practice, confirmation of these findings and improved understanding of the means of reducing pseudoenhancement are needed. Although it has been found that size and location of a lesion, peak tube voltage settings, and number of CT detectors affect image contrast and attenuation in phantom studies, clinical studies are lacking. We undertook this study to determine the effect of lesion size and location on renal cyst pseudoenhancement. Materials and Methods Patients The institutional review board approved our study with waiver of informed consent, and the study was compliant with HIPAA. To evaluate the effect of CT detector number, we analyzed data from the first quarter of 2006 and the last quarter of Search of our radiology information system yielded 1321 abdominal CT examinations, of which 597 were triphasic renal mass CT and were included in the study. Triphasic CT scans of renal masses obtained during this period were reviewed for presence of cysts. A cyst was defined as a wellcircumscribed, uniformly low-attenuation spheric lesion without a discernable wall and content with attenuation between 10 and 20 HU in the unenhanced phase. Only scans that had tube current standardized across all phases were included. CT studies of two patients with polycystic kidney disease were excluded because of inability to differentiate individual cysts within the kidneys. Cysts smaller than 6 mm and larger than 5 cm were excluded. CT Technique All CT examinations were performed with 16-, 40-, and 64-MDCT scanners. The unenhanced, arterial, and nephrographic phases were used as part of the triphasic renal mass protocol. The same parameters were used for all phases, as was z-axis current tube modulation. The scans were reconstructed in slice thickness. The matrix was , and all images were viewed in an abdominal window (width, 400 HU; level, 40 HU). The contrast-enhanced images were obtained after IV administration of 100 ml of iohexol (Omnipaque 300, GE Healthcare) injected at 3.5 ml/s and scan delays of 30 and 100 seconds for the arterial and nephrographic phases. All scans were obtained with an abdominal reconstruction kernel and z- axis tube current modulation. The parameters are shown in Table 1. A Image Interpretation One author with 5 years of experience in genitourinary CT and a third-year medical student reviewed the unenhanced and contrast-enhanced images for the presence of renal cysts. Identified cysts with attenuation between 10 and 20 HU were measured at the approximate center of the lesion, and the diameter of the lesion in millimeters was noted. Corresponding circular regionof-interest measurements were obtained for each cyst. For each cyst, the region of interest was a minimum of 66% of the circumference of the lesion (Fig. 1). Attenuation in the unenhanced and nephrographic phases was recorded in similar regions of interest. Cysts were identified in the nephrographic phase. Manual narrowing of the image window, magnification, and lesion localization were used in some cases. Cysts not visualized on the unenhanced scan were excluded. The location of the lesion was assessed by scanning through the entire lesion. Central cysts were defined as a cyst completely surrounded by renal parenchyma. Peripheral cysts were defined as those in which a portion of the cyst caused bulging of the renal capsule or was exophytic (Fig. 1). Other data collected for each lesion included CT scanner model and number of detectors, slice thickness, and age and sex of the patient. Cysts were defined as having displayed pseudoenhancement if they met one of the following criteria: attenuation 20 HU or less in the unenhanced phase (simple cyst) and enhancement of 10 HU or greater in the nephrographic phase. Cysts with 11- to 20-HU attenuation on unenhanced scans remained stable. Stability of a cyst was defined as a 1-mm or less change in diameter at comparison CT performed 1 year or more before or after the CT date. Statistical Analysis All analyses were performed with statistical software (SAS version 9.2, SAS Institute). Descriptive statistics were computed for all variables with frequencies and percentages for categoric variables and means and SD for continuous variables. The relation between lesion size and pseudoenhancement was analyzed by two approaches. The first approach applied a repeated measures analysis with a linear mixed-effects model to compare the mean lesion sizes in the groups with and without Fig. 1 CT attenuation measurement in two patients. Regions of interest are minimum 66% of circumference of lesion for each cyst. A, 63-year-old man with renal cyst abutting renal capsule and therefore considered peripheral. B, 41-year-old man with renal cysts entire circumference of which is surrounded by renal parenchyma and is defined as central. B 134 AJR:198, January 2012

3 MDCT of Renal Cysts TABLE 2: Cysts by Size Category Total No. Percentage of Pseudoenhancing Percentage of All Size Category of Cysts No. of Cysts With Pseudoenhancement Cysts in Size Category Pseudoenhancing Cysts (n = 51) < 10 mm mm TABLE 3: Breakdown of Cysts by Scanner Type (n = 233) MDCT Scanner Model Total No. of Cysts pseudoenhancement. Repeated measures analysis also was used to determine significant differences in degree of pseudoenhancement between lesion size recoded as category and between lesion locations. Tukey adjustment for multiple comparisons was used when means in more than two groups were being compared to maintain a family-wise error rate of Box plots were used to visualize the differences in means and distributions. The second approach was a generalized estimating equations model, which is an extension of logistic regression for repeated measures with a binary outcome to examine the association between lesion size recoded as category and rate of pseudoenhancement. The same approach was used to analyze the association between rate of pseudoenhancement and other variables such as lesion location, slice thickness, and scanner model. Odds ratios were used to quantify the magnitude and direction of any significant associations. Results Effect of Variables on Pseudoenhancement Rate From the 597 triphasic renal mass CT examinations, a total 233 cysts were isolated. The incidence was 21% for cysts in the 6- to 50-mm range, 128 of 597 triphasic CT scans viewed showing cysts, some multiple. Of the 233 cysts included in the study, 51 cysts exhibited pseudoenhancement, for an overall 22% rate of pseudoenhancement. The mean lesion size was significantly (p < 0.01) lower in the group with pseudoenhancement (mean, 8.3 mm) than in the group without pseudoenhancement (mean, 16.1 mm). None of the cysts exhibited hypervascular enhancement in the arterial phase, defined as enhancement equal or greater than that of the renal tissue. Renal cyst size ranged from 6 to 48 mm in diameter (mean, 14.4 [SD, 9.2] mm). The size categories evaluated were smaller than No. of Cysts with Pseudoenhancement 10 mm and 10 mm in diameter or larger. Table 2 shows the breakdown of cyst size category and pseudoenhancement. The 10 mm and larger category was further divided into mm (62 cysts), mm (25 cysts), and mm (50 cysts). Twelve of the 10- to 14-mm cysts (19%) were pseudoenhancing, constituting 24% of the study total that was pseudoenhancing; 0.04% of the 15- to 19-mm cysts were pseudoenhancing, comprising only 0.02% of the study total that was pseudoenhancing. None of the 20- to 49- mm cysts was pseudoenhancing. Only one cyst 1.5 cm or larger was pseudoenhancing compared with 50 cysts smaller than 1.5 cm. Therefore, a cyst smaller than 1.5 cm was 30.4 times (95% CI, ) as likely as a cyst 1.5 cm or larger to be pseudoenhancing. A significant association was found between pseudoenhancement and lesion size (p < 0.001). A lesion smaller than 10 mm was likelier than a lesion 10 mm or larger to be pseudoenhancing (odds ratio, 6.4; 95% CI, ). There were 109 peripheral cysts. In 12 of 109 cysts (11%) pseudoenhancement was seen and accounted for 24% of all pseudoenhancement cases. There were 124 central cysts defined as a cyst completely surrounded by the renal parenchyma. In 39 of 124 cysts (32%) pseudoenhancement was seen and accounted for 76% of all pseudoenhancement cases. There was a statistically significant association between pseudoenhancement and peripheral or central group (p < ). Central cysts were likelier than peripheral cysts to be pseudoenhancing (odds ratio, 2.7; 95% CI, ). Table 3 shows the breakdown of cysts by scanner type. Although there was more pseudoenhancement with the 64-MDCT scanner, there was no statistically significant difference Percentage of Cysts Showing Pseudoenhancement with Scanner Model Percentage of All Pseudoenhancing Cysts (n = 51) 16 Sensation or Emotion (Siemens Healthcare) Sensation (Siemens Healthcare) Definition (Siemens Healthcare) in the rate of pseudoenhancement between the detector number categories (p = 0.74). Degree of Pseudoenhancement Versus Size and Location of Cysts Degree of pseudoenhancement varied between 10 and 42 HU. Table 4 shows the breakdown of cysts by degree of enhancement. Fifty of 51 of the pseudoenhancing cysts (98%) were in the HU attenuation category. The one cyst with HU attenuation was stable at 1- and 2-year followup evaluations. The largest cyst that exhibited pseudoenhancement is shown in Figure 2. The cyst was stable at follow-up evaluation, although when measured on subsequent scans in the same phases, the cyst enhancement decreased from 10 to 6 HU. Figure 3 shows the variations in degree of pseudoenhancement by size criteria. When cyst size was divided into smaller than 1 cm and 1 cm and larger, there was no statistically significant difference between the groups (mean pseudoenhancement, 17 and 12 HU; p = 0.22). The mean degree of pseudoenhancement was 13 HU in the peripheral cysts (n = 12) and 16 HU in the central cysts (n = 39). There was no was no statistically significant difference between the two groups with respect to degree of pseudoenhancement (p = 0.194). Discussion Benign cyst is the most common focal renal lesion seen at abdominal CT. It occurs in as much as 40% of the general population [18]. Although not nearly as common, renal cell carcinoma has a mortality of 25%, making it imperative to accurately differentiate benign cyst from carcinoma when a renal lesion is identified. Since 1986, the accepted method of classifying renal lesions has been the Bosniak system, in which a renal cyst is AJR:198, January

4 Tappouni et al. TABLE 4: Degree of Enhancement No. of Cysts by Location Degree of Pseudoenhancement (HU) Total No. of Cysts (n = 51) Average Size (mm) Peripheral Central NA NA NA Note NA = not applicable. A B Fig year-old man with renal cysts. A, Transverse unenhanced CT scan shows 1.5-cm area of low attenuation (0.41 HU) in lower pole of right kidney. B, Transverse contrast-enhanced nephrographic phase CT scan shows cyst measures HU. C, Follow-up CT scan obtained 18 months after B shows cyst is stable. defined as a lesion with fluid attenuation (0 20 HU) and enhancement of less than 10 HU with administration of contrast material [3]. Enhancement of at least 10 HU after contrast administration is the currently accepted minimum increase indicative of a vascular, and therefore potentially malignant, mass [7, 8, 19, 20]. Most renal cell carcinomas are the clear cell type, which is a heterogeneous, highly vascular lesion with marked enhancement after contrast administration. The next most common renal cell carcinoma is papillary, which is distinct from the clear cell type in that it is closer to homogeneous and less vascular. The attenuation values for papillary renal cell carcinoma on unenhanced images are significantly higher than fluid attenuation. The mean attenuation of papillary renal cell carcinomas on unenhanced CT scans has been found to range from 37 to 38 HU, making it unlikely to be confused with a pseudoenhancing benign simple cyst [21, 22]. A number of technical variables can affect the ability to rely on absolute attenuation values of any given tissue: CT scanner type, peak voltage, scanning time, and positional effects [23, 24]. In addition, studies with various phantom models [9, 12 17] have shown an increased incidence of renal cyst pseudoenhancement with smaller cyst size (< 1.5 cm) and intraparenchymal location. To our knowledge, however, no patient data exist on the influence of these variables on pseudoenhancement. The results of our study confirm the increased incidence of pseudoenhancement in lesions of smaller size. Cysts smaller than 10 mm in diameter were six times as likely as those 10 mm in diameter or larger to be pseudoenhancing. Furthermore, only one cyst in the 1.5- to 1.9- cm group was pseudoenhancing, and none of the cysts 2 cm or larger was pseudoenhancing. Our data also showed that central cysts were 2.7 times as likely as peripheral cysts to be pseudoenhancing. Interestingly, contrary to previous findings [12 14, 17], there was not a statistically significant difference in incidence of pseudoenhancement between CT scanner types or slice thicknesses. All lesions measured in our study were at least twice the minimal slice thickness of the CT image, to prevent undersampling. Although the types of scanners were not equally represented in our study, each CT scanner type yielded a remarkably similar percentage of pseudoenhancement (19 26%). The limitations of our study were a lack of histopathologic confirmation of the cysts imaged and a disproportionate number of cases among scanner types. Future variables to be studied for their effect on the incidence of pseudoenhancement are degree of background enhancement and alterations in tube voltage and Degree of Pseudoenhancement (HU) Size (mm) 10+ Fig. 3 Box plot shows size of lesion in relation to degree of pseudoenhancement, which is 24, 16, and 13 HU for three categories. C 136 AJR:198, January 2012

5 MDCT of Renal Cysts current-time settings in addition to effect of habitus on pseudoenhancement. In addition, dual-energy CT, which can be used to obtain unenhanced images is a promising tool that may facilitate detection of enhancement of renal lesions incidentally found during contrastenhanced CT [25]. Conclusion Pseudoenhancement of renal cysts significantly correlates with size smaller than 1 cm and central location. Although pseudoenhancement increased with higher detector number, the difference was not significant. Pseudoenhancement is exceedingly rare when the renal cyst size is 1.5 cm or larger. Although lesions smaller than 1.0 cm have traditionally been classified as too small to characterize, areas of low attenuation measuring cm have a relatively high likelihood of pseudoenhancement and therefore can fall into a similar indeterminate category. References 1. Tada S, Yamagishi J, Kobayashi H, Hata Y, Kobari T. The incidence of simple renal cyst by computed tomography. Clin Radiol 1983; 34: Laucks SP Jr, McLachlan MS. Aging and simple cysts of the kidney. Br J Radiol 1981; 54: Bosniak MA. The current radiological approach to renal cysts. Radiology 1986; 158: Silverman SG, Lee BY, Seltzer SE, Bloom DA, Corles CL, Adams DF. Small (< or = 3cm) renal masses: correlation of spiral CT features and pathologic findings. AJR 1994; 163: Maki DD, Birnbaum BA, Chakraborty DP, Jacobs JE, Carvalho BM, Herman GT. Renal cyst pseudoenhancement: beam hardening effects on CT numbers. Radiology 1999; 213: Hopper KD, Holabinko JN, Have T, Hartman DS. Variability of Hounsfield unit measurements based on CT of a renal cyst (abstr). Radiology 1995; 197(P): Bae KT, Heiken JP, Siegel CL, Bennett HF. Renal cysts: is attenuation artifactually increased on contrast-enhanced CT images? Radiology 2000; 216: Chung EP, Herts BR, Linnell G, et al. Analysis of changes in attenuation of proven renal cysts on different scanning phases of triphasic MDCT. AJR 2004; 182: Coulam CH, Sheafor DH, Leder RA, et al. Evaluation of pseudoenhancement of renal cysts during contrastenhanced CT. AJR 2000; 174: Davidson AJ, Hartman DS, Choyke PL, Wagner BJ. Radiologic assessment of renal masses: implications for patient care. Radiology 1997; 202: Siegel CL, Fisher AJ, Bennett HF. Interobserver variability in determining enhancement of renal masses on helical CT. AJR 1999; 172: Wang ZJ, Coakley FV, Fu Y, et al. Renal cyst pseudoenhancement at multidetector CT: what are the effects of number of detectors and peak tube voltage? Radiology 2008; 248: Birnbaum BA, Hindman N, Lee J, Babb JS. Renal cyst pseudoenhancement: influence of multidetector CT reconstruction algorithm and scanner type in phantom model. Radiology 2007; 244: Abdulla C, Kaira MK, Saini S, et al. Pseudoenhancement of simulated renal cysts in a phantom using different multidetector CT scanners. AJR 2002; 179: Birnbaum BA, Maki DD, Chakraborty DP, et al. Renal cyst pseudoenhancement: evaluation with an anthropomorphic body CT phantom. Radiology 2002; 225: Gokan T, Ohgiya Y, Munechika H, et al. Renal cyst pseudoenhancement with beam hardening effect on CT attenuation. Radiat Med 2002; 20: Heneghan JP, Spielmann AL, Sheafor DH, et al. Pseudoenhancement of simple renal cysts: a comparison of single and multidetector helical CT. J Comput Assist Tomogr 2002; 26: Carrim ZI, Murchison JT. The prevalence of simple renal and hepatic cysts detected by spiral computed tomography. Clin Radiol 2003; 58: Israel GM, Bosniak MA. How I do it: evaluating renal masses. Radiology 2005; 236: Bosniak MA. The small ( 3.0cm) renal parenchymal tumor: detection, diagnosis, and controversies. Radiology 1991; 179: Tsuda K, Kinouchi T, Tanikawa G, et al. Imaging characteristics of papillary renal cell carcinoma by computed tomography scan and magnetic resonance imaging. Int J Urol 2005; 12: Kim JK, Kim TK, Ahn HJ, et al. Differentiation of subtypes of renal cell carcinoma on helical CT scans. AJR 2002; 178: Levi C, Gray JE, McCullough EX, et al. The unreliability of CT numbers as absolute values. AJR 1982; 139: Bosniak MA, Rofsky NM. Problems in the detections and characterization of small renal masses. Radiology 1996; 198: Neville AM, Gupta RT, Miller CM, Merkle EM, Paulson EK, Boll DT. Detection of renal lesion enhancement with dual-energy multidetector CT. Radiology 2011; 259: AJR:198, January