CT imaging of solid renal masses: pitfalls and solutions

Clinical Radiology 72 (2017) 708e721

Contents lists available at ScienceDirect

Clinical Radiology journal homepage: www.clinicalradiologyonline.net

Review

CT imaging of solid renal masses: pitfalls and solutions S. Krishna a, C.A. Murray a, M.D. McInnes a, R. Chatelain a, M. Siddaiah a, O. Al-Dandan b, S. Narayanasamy a, N. Schieda a, * a b

Department of Medical Imaging, The Ottawa Hospital, University of Ottawa, Ottawa, Canada Department of Radiology, University of Dammam, Dammam, Saudi Arabia

article in formation Article history: Received 8 August 2016 Received in revised form 20 April 2017 Accepted 2 May 2017

Computed tomography (CT) remains the first-line imaging test for the characterisation of renal masses; however, CT has inherent limitations, which if unrecognised, may result in errors. The purpose of this manuscript is to present 10 pitfalls in the CT evaluation of solid renal masses. Thin section non-contrast enhanced CT (NECT) is required to confirm the presence of macroscopic fat and diagnosis of angiomyolipoma (AML). Renal cell carcinoma (RCC) can mimic renal cysts at NECT when measuring <20 HU, but are usually heterogeneous with irregular margins. Haemorrhagic cysts (HC) may simulate solid lesions at NECT; however, a homogeneous lesion measuring >70 HU is essentially diagnostic of HC. Homogeneous lesions measuring 20e70 HU at NECT or >20 HU at contrast-enhanced (CE) CT, are indeterminate, requiring further evaluation. Dual-energy CT (DECT) can accurately characterise these lesions at baseline through virtual NECT, iodine overlay images, or quantitative iodine concentration analysis without recalling the patient. A minority of hypo-enhancing renal masses (most commonly papillary RCC) show indeterminate or absent enhancement at multiphase CT. Follow-up, CE ultrasound or magnetic resonance imaging (MRI) is required to further characterise these lesions. Small (<3 cm) endophytic cysts commonly show pseudo-enhancement, which may simulate RCC; this can be overcome with DECT or MRI. In small (<4 cm) solid renal masses, 20% of lesions are benign, chiefly AML without visible fat or oncocytoma. Low-dose techniques may simulate lesion heterogeneity due to increased image noise, which can be ameliorated through the appropriate use of iterative reconstruction algorithms. Ó 2017 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.

Introduction Renal lesions are common incidental findings detected with cross-sectional imaging. It is estimated that 14.4% of

* Guarantor and correspondent: N. Schieda, Department of Medical Imaging, The Ottawa Hospital, University of Ottawa, Ottawa, Canada. Tel.: þ1 613 697 3473. E-mail address: [email protected] (N. Schieda).

patients undergoing computed tomography (CT) will demonstrate at least one incidental renal lesion >1 cm.1 The vast majority of renal lesions will represent benign cortical cysts1; however, renal cell carcinoma (RCC) is also commonly detected incidentally. Currently, the most common clinical presentation of RCC is as an incidentally discovered renal lesion.2 Overall, CT and magnetic resonance imaging (MRI) both show excellent accuracy for the characterisation of renal lesions. CT is more commonly used at many institutions due to improved accessibility, reduced

http://dx.doi.org/10.1016/j.crad.2017.05.003 0009-9260/Ó 2017 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.

S. Krishna et al. / Clinical Radiology 72 (2017) 708e721

cost, and better tolerance in elderly and sick patients who may be limited in their ability to comply with the necessary breath-holding and prolonged examination times required for renal MRI. Accordingly, CT has a slightly higher rating for the initial characterisation of indeterminate renal masses by the American College of Radiology appropriateness criteria when compared to MRI.3 MRI can also be used as the firstline imaging study for characterisation of renal lesions, particularly in young patients (mainly related to concerns over CT dose), but may also be used as an adjunct test when CT findings are inconclusive or indeterminate.4,5 Dualenergy (DE) CT has numerous described applications in abdominal and pelvic imaging6 and, increasingly DECT may be applied for the characterisation of renal masses.7 Although highly accurate for the characterisation of renal lesions, a variety of pitfalls encountered during the CT imaging of renal masses may result in important errors when unrecognised. An understanding of the limitations of CT for the characterisation of renal masses, as it pertains to the epidemiology and pathophysiology of renal lesions, the various components of a multi-phase CT renal mass protocol and challenges related to attenuation measurement at both non-contrast enhanced CT (NECT) and contrastenhanced CT (CECT) is critical to identify and prevent errors related to pitfalls in renal mass imaging with CT. The purpose of this article is to briefly review the necessary components of a comprehensive multi-phase CT renal mass protocol and to present 10 pitfalls in the CT imaging of renal masses and how to detect and avoid them. This article also presents an up-to-date review of the complimentary and often supplementary role of DECT and MRI for characterisation of lesions that would be considered indeterminate with conventional CT imaging.

Multi-phase renal mass protocol The reference standard for dedicated multi-phase renal mass imaging consists of a triphasic CT protocol composed of: NECT and corticomedullary (CM) and nephrographic (NG) phase CECT.3 NECT is a critical component of a renal mass protocol because it establishes the baseline attenuation of a mass before the administration of iodinated contrast material. The baseline attenuation of a renal mass at NECT is often the initial step in characterisation with CT. NECT is also critically important for detection of areas of internal calcification and macroscopic fat, which may be obscured after contrast medium administration.8 In a solid lesion, calcification is more common in RCC and is generally not seen in angiomyolipoma (AML).8,9 A renal mass with low-density components that measure less than e10 HU should be considered diagnostic of AML.8,10 A renal mass measuring between e10 and þ20 HU is consistent with water attenuation and when combined with findings of a smooth imperceptible wall and internal homogeneity, can be used to confidently diagnose a simple renal cyst.11 Lesions measuring >20 HU on NECT are considered indeterminate (because they overlap in attenuation with RCC) and classically require contrast-enhanced imaging to detect the

709

presence or absence of internal enhancement enabling differentiation between a solid lesion or a haemorrhagic cyst. An attenuation difference of >20 HU comparing NECT to CECT is consistent with enhancement, whereas an attenuation difference of <10 HU is consistent with absence of enhancement.12 These guidelines form the general principles for characterisation of most renal masses using CT. Conventionally, and in the authors’ opinion, CM and NG phase imaging should both be performed because the enhancement pattern may help differentiate between clearcell RCC (which typically enhance avidly on the CM phase and wash-out on the NG phase13) and papillary RCC (which often shows progressive gradual enhancement14), and the CM phase better characterises arterial anatomy and vascular disease entities such as aneurysms or arterialevenous fistulas, which may occasionally mimic tumours.15 Preliminary literature using DECT has shown that for subtyping clear-cell and papillary RCC and differentiating between low-versus high-grade clear-cell RCC, quantitative iodine content analysis may be as accurate as the enhancement pattern, which suggests that when using DECT, multi-phase post-contrast acquisitions may become unnecessary.16,17 The CM phase is timed at approximately 30e40 seconds when the contrast medium is present in the proximal convoluted tubules of the renal cortex and in the Columns of Bertin.18 The CM phase may also be used to image the renal arterial vasculature as well as potential renal vein involvement by enhancing tumour thrombus.19 The CM phase may be timed empirically or by using contrast medium bolus-tracking software. The NG phase is timed at approximately 120 seconds when the contrast medium enters the loop of Henle and the collecting ducts.18 The NG phase may also be used to assess tumour thrombus extent and specifically extent of involvement into the inferior vena cava.20 It is important that technical parameters (e.g., tube voltage, tube current, noise/iterative reconstruction levels) are matched between the NECT and CECT phases of a multiphase study in order to accurately compare quantitative attenuation values and subjective assessment of the appearance of renal lesions, which may be affected by differences in technique.21 The use of a fourth (urographic) phase (performed at approximately 8 minutes or later depending on patient renal excretory function) is important for the detection of urothelial malignancies (particularly in the upper tracts)22 and is also performed as a part of the routine renal mass protocol at some institutions. The routine use of a dedicated urographic phase for the evaluation of renal masses, to the authors’ knowledge, has not been validated for routine renal mass characterisation and in the authors’ opinion, its use should be weighed against the cost of added radiation dose. A detailed explanation of the principles of DECT is beyond the scope of this review, but has been described elsewhere.23e25 Of the current DECT platforms, dual-source ([ds]; Siemens Medical, Malvern, PA, USA) and rapid tube voltage switching ([rs]; General Electric Medical, Milwaukee, WI, USA) are the most well studied, particularly as they apply to genitourinary applications.16,26e29 DE acquisitions

710

S. Krishna et al. / Clinical Radiology 72 (2017) 708e721

can be added to existing renal-mass multi-phase protocols either during the NECT, CM, NG, or both CECT acquisitions. Described applications of DECT in renal mass imaging include: 1) use of a virtual NECT data set to determine baseline attenuation of lesions when only a single phase CECT acquisition is performed or to replace the true NECT dataset30, 2) the use of low-energy monochromatic data to overcome pseudo-enhancement (the artefactual increase in attenuation of a lesion related primarily to beam hardening31),32 and 3) the use of visual analysis of iodine overlay images or quantitative analysis of iodine concentration to confirm or quantify the presence of enhancement in renal lesions.16,27,28 At the authors’ institution, the routine multi-phase renal mass CT protocol consists of: 1) true NECT with thin-section (1.5 mm) reconstructions, 2) CM phase CECT acquired at 30 seconds using bolus-tracking software, and 3) delayed NG phase CECT acquired at 120 seconds. A dual-energy acquisition is now performed during the CM phase of enhancement as a routine component of the renal mass CT protocol. This strategy enables comparison of the radiodensity values on conventional true NECT and NG phase CECT acquisitions while providing the additional information available through DECT acquired during the CM phase of enhancement. One could alternatively acquire a dual-energy acquisition on the NG or both the NG and CM phases of enhancement, and previous authors have also compared attenuation values measured in Hounsfield units on true NECT and 70 keV DECT acquisitions.28,29 We continue to perform a true NECT (despite the ability to generate virtual NECT images from the DECT dataset) because the capacity to measure attenuation in Hounsfield units on virtual NECT using rsDECT has only recently been commercially available and has only been preliminarily validated in normal structures.33 This is concordant with previous studies that have evaluated rsDECT in renal mass imaging.28,29 Although virtual NECT images with the ability to measure Hounsfield units has been available and previously validated using ds virtual NECT,30 elimination of the true NECT from a renal mass protocol when using DECT requires further investigation. For example, the absolute difference between attenuation values measured using true and virtual NECT was >10 HU in up to 25% of cases in a phantom study34 and 7% in both CM and NG phase derivations in a clinical study.33 Moreover, in a recent study by Connolly et al.35 using meta-analysis, the authors concluded that there was a consistent trend towards overestimation of the attenuation in adrenal adenomas using virtual compared to true NECT on dsDECT, when virtual NECT was derived from earlier phases of enhancement. An alternative approach is to perform a true NECT and dual energy NG phase omitting the CM phase.

Ten pitfalls of solid renal mass imaging NECT pitfalls (1) Thin-section (1.5e3 mm) reconstructions of NECT data are required to confirm the presence of intralesional

macroscopic fat; however, in small lesions or minute areas of suspected fat, chemical-shift MRI may outperform NECT. CT histogram analysis is not useful. The presence of macroscopic (gross) fat within a renal lesion is considered diagnostic of benign AML.10 In order to confirm the presence of fat within a renal lesion, region of interest (ROI) measurements are placed and attenuation values of less than e10 HU are required.10 In small lesions or minute areas of suspected internal fat, ROI measurements may be inaccurate because of volume averaging of renal parenchyma into the measurement, artificially increasing the attenuation value (Fig 1).36 The use of thin section (1.5e3 mm) reconstructions improves the accuracy of ROI measurement by reducing volume-averaging effects10; however, in some lesions thin-section CT is still indeterminate (Fig 1). In the authors’ experience, chemical-shift MRI may outperform NECT in these instances for the confirmation of bulk fat because unlike with CT, averaging of fat and renal parenchyma into the same imaging voxel will produce a profound signal intensity (SI) drop, which in small areas/lesions can be reasonably accurately diagnosed as bulk fat (Fig 1) and typically the finding can be confirmed using fat-suppressed T1-weighted imaging.36 The use of CT histogram analysis, by counting the number of lowattenuation pixels in a ROI, is currently not considered valuable for the diagnosis of AML.37 (2) Not all fat-containing renal lesions are AML: a minority of RCC may demonstrate macroscopic fat. Although the presence of internal macroscopic fat is considered diagnostic of AML, rarely RCC may also demonstrate small amounts of internal gross fat38e42 (Fig 2). Differentiating between a benign AML or fat-containing RCC is often not possible unless there are aggressive features (e.g., irregular or invasive margins, rapid growth, metastatic disease) or there is co-existing calcification within the mass, which is more commonly seen in RCC than AML.43,44 A heterogeneous appearance may also favour the diagnosis of RCC in these instances as fat poor AML are more likely to be homogeneous.45 (3) Not all water attenuation lesions are renal cysts/cystic lesions: a minority of solid clear-cell RCC may measure between e10 and þ20 HU. The diagnosis of simple renal cyst can be confidently established when a renal lesion measures between e10 and 20 HU, shows a thin imperceptible wall, and is uniformly homogeneous.11 The latter subjective criteria are critical to emphasise because cystic and necrotic RCC may also demonstrate areas that measure between e10 and 20 HU46 and relying solely on attenuation values for characterisation can be misleading. Recently, Schieda et al.47,48 also demonstrated that a minority of solid clear-cell RCC may also measure of water attenuation at NECT potentially simulating a simple cyst (Fig 3); however, in their study, most RCC measuring water attenuation showed irregular

S. Krishna et al. / Clinical Radiology 72 (2017) 708e721

711

Figure 1 A 59-year-old man who underwent ultrasound for evaluation of elevated creatinine. (a) Axial grey-scale ultrasound image shows an incidental echogenic nodule in the interpolar region of the right kidney (white arrow) suggestive of fat. (b) Axial NECT performed subsequently shows that the nodule has an attenuation of e2 HU, which does not meet the CT attenuation criteria for fat. MRI was performed due to discordance between ultrasound and CT observations. (c) Axial T1-weighted (T1W) in-phase (IP) gradient-echo (GRE) image shows the nodule as hyperintense (white arrow), which is showing profound signal intensity drop on T1W opposed-phase (d) GRE image (white arrow). (e) Axial fat-only image (derived from two-point Dixon technique) also depicts the presence of fat (black arrow) within the nodule. (f) Axial chemical fat-supressed (FS) T1W image demonstrates loss of signal (white arrow) consistent with bulk fat. Features are typical of a subcentimetre triphasic AML.

margins and all were heterogeneous. Therefore, when a complex renal mass is identified at NECT that measures water attenuation, multi-phase imaging should be suggested for further characterisation. (4) Attenuation of haemorrhagic cysts overlap with RCC at NECT; however, a homogeneously hyperattenuating lesion measuring >70 HU is diagnostic of a complex cyst.

Although homogeneous renal masses measuring e10 to þ20 HU can be confidently diagnosed as simple cysts, benign haemorrhagic cysts show increased attenuation at NECT related to internal blood and protein and often measure >20 HU, overlapping in attenuation with RCC. To differentiate between a haemorrhagic cyst and RCC detected at NECT, follow-up imaging (usually with multiphase CT or MRI) is generally performed. Ultrasound may also be

Figure 2 A 76-year-old woman who underwent CT for characterisation of an indeterminate renal nodule detected on ultrasound (not shown). (a) Axial NECT image shows a mass arising from the posterior interpolar region of the left kidney (white arrow). There is a small focus in the centre of the mass, which appears of low attenuation (black arrowhead). (b) Point analysis shows that the focus has a density of e25 HU consistent with macroscopic fat. (c) The CM phase CECT image shows heterogeneous hyperenhancement (white arrow) of the mass, which was resected and was confirmed to be a Fuhrman nuclear grade 2 clear-cell RCC.

712

S. Krishna et al. / Clinical Radiology 72 (2017) 708e721

Figure 3 A 48-year-old man who underwent CT for right upper quadrant pain with an incidental nodule in the right kidney. (a) Axial NECT image demonstrates an incidental 15 mm lesion arising from the anterior interpolar region of the right kidney (white arrow). The lesion measures 17 HU on NECT; however, it appears heterogeneous and demonstrates an irregular wall anteriorly. (b) The nodule enhances on multiphase renal protocol CT (white arrow). MRI was also performed to determine if the nodule was cystic or solid due to water attenuation on NECT. (c) Axial T2W single-shot fast spin-echo (SSFSE) image shows the nodule is hyperintense to renal cortex (white arrow) but less so than fluid. (d) Post-gadolinium enhanced T1W FS GRE image shows solid enhancement of the lesion (white arrow). The mass was resected and was confirmed to be a Fuhrman nuclear grade 2 clear-cell RCC.

considered in this setting; however, the available literature suggests that only 10e50% of haemorrhagic cysts will appear simple on ultrasound.49,50 Not all hyperattenuating homogeneous renal lesions detected at NECT require additional characterisation. Jonisch et al.51 previously demonstrated that when a homogeneous renal mass measures >70 HU at NECT, this level of attenuation is essentially diagnostic of a haemorrhagic cyst (Fig 4). In a recent followup study by Davarpanah et al.52 a lower threshold of 66 HU was proposed for the diagnosis of haemorrhagic cyst at NECT with similar accuracy.52

CECT pitfalls (5) Homogeneous hyperattenuating renal masses may represent RCC or haemorrhagic cysts at uniphasic CECT. DECT analysis of single-phase CECT can accurately differentiate between RCC and complex cysts, without recalling the patient. When a hyperattenuating (>20 HU) renal mass is detected on a single-phase CECT examination, it may not be

possible to discriminate a hyperattenuating renal cyst from an RCC due to overlap in attenuation values. Subjective features such as heterogeneity and invasive or infiltrating margins are differentiating features; however, in a homogeneously hyperattenuating renal mass, distinguishing between RCC and a haemorrhagic cyst is not possible and follow-up imaging with multi-phase CT or MRI is often required. Ultrasound may also be considered as an intermediate step in this setting, although as discussed above, may be limited as haemorrhagic cysts may show low level echoes on ultrasound rendering the diagnosis indeterminate. One caveat to this limitation of single-phase CECT is when a DECT acquisition has been performed. Using data derived from the DE technique, enhancement can be assessed in three ways: 1) through the generation of virtual NECT data, so that NECT attenuation can be measured and compared to CECT attenuation to determine attenuation difference and assess for enhancement; 2) by subjective analysis of iodine overlay images, where structures containing iodine (or conversely not containing iodine) are analysed using a colour (or less optimally grey) scale; and 3)

S. Krishna et al. / Clinical Radiology 72 (2017) 708e721

713

Figure 4 A 74-year-old man who underwent CT to rule out abdominal aortic aneurysm. (a) Axial NECT image shows a hyperdense nodule arising from the posterior interpolar region of the right kidney (white arrow). The nodule is homogeneous and shows a thin smooth wall where it interfaces with retroperitoneal fat. Attenuation of the nodule was 74 HU. Abdominal ultrasound was performed to further characterise. (b) Sagittal grey-scale ultrasound image shows the nodule as a homogeneously anechoic cyst (white arrow). (c) Axial CECT image obtained 8-years later shows that the haemorrhagic cyst has not changed in size or appearance (white arrow).

by measuring iodine concentration within a lesion and comparing iodine concentration to previously described thresholds (discussed later)53 (Fig 5). (6) Nephrographic phase CECT imaging improves the detection of small renal lesions, which can be missed when delayed imaging is not performed. The nephrographic phase of enhancement is the most sensitive phase for detection of renal masses54 because almost all renal masses will enhance less than the renal cortical parenchyma at approximately 120 seconds.55 If only earlier phases are acquired, then some small renal masses may not be detected (Fig 6). This is more prevalent in the medulla, where masses may appear isodense to the adjacent renal parenchyma.56 (7) A proportion of RCC (mainly papillary tumours) are hypoenhancing and may show indeterminate range (10e20 HU) or absent (<10 HU) attenuation change at multiphase CT. NG phase imaging improves detection of enhancement; however, contrast-enhanced ultrasound or MRI may be required to confirm enhancement. A subset of renal masses may be hypovascular and show indeterminate (or absent) enhancement comparing CECT to NECT images. Papillary RCC characteristically shows gradual progressive enhancement.57e60 In a study by Egbert et al.57, it was reported that a substantial minority of papillary tumours showed either indeterminate (10e20 HU) or absent (<10 HU) enhancement at multi-phase CT. A limitation of this study was that a variety of CT protocols were evaluated including a proportion where a dedicated NG phase was not performed. As it can be expected that papillary tumours would show maximal enhancement on delayed phases, a follow-up study was performed by Dilauro et al.61 who confirmed that a majority of papillary tumours showed absent or indeterminate enhancement on CM phase imaging; however, only a minority of papillary tumours show indeterminate enhancement using dedicated NG phase imaging

and no papillary tumour was non-enhancing (<10 HU) by 120 seconds (Fig 7). These results were subsequently confirmed by Al Harbi et al.62 In the study by Dilauro et al.,61 all papillary tumours with indeterminate enhancement on CT showed enhancement on contrast-enhanced MRI, using subjective interpretation of post-gadolinium-enhanced subtraction images and by quantitative signal intensity thresholds. In another study by Bertolotto et al.,63 the use of contrast-enhanced ultrasound was also shown to improve characterisation of indeterminate enhancing renal lesions. (8) Renal cysts may show pseudo-enhancement (artefactual increase in attenuation at CECT), which is more common in small (<3 cm) endophytic cysts. Contrast-enhanced ultrasound, MRI, or DECT can be performed to confirm the artificial increase in attenuation in these lesions. Pseudo-enhancement is defined as the artefactual increase in attenuation of a non-enhancing structure following contrast medium administration, even after the effects of partial volume averaging have been eliminated. This phenomenon is related to inadequate algorithmic correction for beam-hardening artefacts from a polychromatic beam source (Fig 8).31 Pseudo-enhancement is more common in small endophytic cysts.64 As pseudoenhancement cannot be differentiated from true enhancement within a lesion, follow-up imaging is required. Ultrasound has a limited role in this setting, and a previous study by Bertolotto et al.63 showed that conventional ultrasound could not characterise a majority of pseudoenhancing cysts that were accurately diagnosed using contrast-enhanced ultrasound. At the authors’ institution, the use of MRI is preferred to further assess suspected pseudo-enhancement (Fig 8); however, MRI may be limited by poor image subtraction.60 More recently, DECT acquisition with virtual monochromatic image reconstruction of data between 90e140 keV has been shown to accurately characterise and virtually eliminate the problem of pseudo-enhancement as shown by the study by Mileto et al.32 Virtual monochromatic

714

S. Krishna et al. / Clinical Radiology 72 (2017) 708e721

Figure 5 A 64-year-old man who underwent DECT enterography for iron deficiency anaemia. True NECT was not performed. (a) Axial portal venous phase CECT image demonstrates a homogeneous partially exophytic left renal nodule (white arrow), which measures 35 HU. The nodule is indeterminate. Virtual NECT (not shown) was reconstructed from source data and the nodule measured 31 HU, therefore showing no enhancement in keeping with a haemorrhagic cyst. (b) Axial iodine overlay image also demonstrates the absence of enhancement (no iodine uptake) within the nodule by subjective analysis (white arrow). (c) Iodine quantitative spectral analysis similarly confirms the absence of iodine within the nodule as the maximum iodine concentration within the nodule (fuchsia dots) is less than a previously described threshold by Kaza et al.28 of 2 mg/ml (red line) compared to the normal enhancing renal parenchyma (yellow dots).

images use a two basis material decomposition (iodine and water), which enables more accurate beam-hardening correction, allowing for improved linearity of CT attenuation resulting in decreased pseudo-enhancement (Fig 9).65 The use of material decomposition also enables accurate identification of imaging voxels containing iodine, which can then be superimposed onto the virtual non-contrast image set to generate iodine overlay images. This visual representation of data is accurate to identify the presence or absence of iodine (and therefore enhancement) in a renal lesion (Fig 9).66 Multiple previous studies have shown credible results when using quantitative iodine content analysis with DECT for characterisation of enhancement using both dsDECT and rsDECT technologies.16,27e29,67,68 Nevertheless, the threshold of iodine content used to qualify enhancement has varied significantly between studies and requires further analysis (Fig 9).27,28,68 For example, using dsDECT Chandarana et al.68 demonstrated that a threshold of 0.5 mg/ml was accurate to diagnose

enhancement, which was later validated in other studies.16,27 Conversely, when using rsDECT, both Kaza et al.28 and Zarzour et al.29 showed significantly higher threshold levels of iodine content to confirm enhancement. This is concordant with a previous phantom study, which showed significant differences in iodine concentration measurements on ds and rsDECT.69 Moreover, even when using the same technology, different thresholds, which maximised accuracy, were reported in the studies by Kaza et al.28 (>2 mg/ml) and Zarzour et al.,29 who also showed different thresholds on studies performed during CM compared to NG phase (1.22 versus 1.28 mg/ml, respectively). Despite these variations in preliminary studies evaluating the role of iodine content thresholds to define enhancement, the potential for DECT iodine concentration analysis to characterise enhancement was highlighted in the study by Ascenti et al.27 who showed that iodine quantification performed better than traditional standard enhancement measurements.

S. Krishna et al. / Clinical Radiology 72 (2017) 708e721

715

Figure 6 A 45-year-old woman with two incidental lesions in the right kidney detected on abdominal ultrasound. (a) Sagittal oblique grey-scale ultrasound image of the right kidney shows two echogenic nodules in the mid/lower pole (white solid and dotted arrows). CT was subsequently performed for further characterisation. (b) Axial NECT, (c) CM, and (d) NG phase CECT images show a 10 mm fat-containing nodule (white arrows) measuring e56 HU compatible with a triphasic AML corresponding to the echogenic nodule depicted by the solid white arrow in (a). The second nodule (white dotted arrow) in (a) is completely isodense to the renal cortical parenchyma on the (e) NECT and (f) CM phase images, and is only identified by a subtle cortical bulge (white dotted arrows) compared to its visibility as a hypo-attenuating lesion relative to renal cortex on NG phase (g) CECT image (white dotted arrow). This nodule is 10 mm in size, homogeneously enhancing, and shows hyperenhancement with washout of contrast medium. In combination with patient gender and associated classic/triphasic AML, a diagnosis of AML without visible fat was suggested and surveillance was elected.

(9) Not all solid enhancing renal masses are RCC. Twenty percent of <4 cm solid renal masses are benign and are mainly AML without visible fat and oncocytoma. In large surgical series, 20% of small (<4 cm) renal masses were benign and the most common histology of these benign neoplasms were AML without visible fat (AML.wovf) and oncocytoma.70e75 To prevent unnecessary surgery, differentiation of AML.wovf and oncocytoma from RCC using multi-phase CT is a topic of great research interest. Several CT findings in AML.wovf have been consistently reproduced and when these are encountered in a small renal mass, particularly in a female patient, should alert the radiologist to the possibility of a benign diagnosis. AML.wovf are hyperdense on NECT (related to abundant smooth muscle content)9,76e79; however, attenuation values overlap with RCC. AML.wovf most commonly show rapid enhancement on the CM phase with washout of contrast medium on the NG or delayed phases9; however, the enhancement pattern overlaps with clear-cell RCC78 and a minority of AML.wovf will show progressive enhancement on multi-phase CT imaging.9 Finally, AML.wovf are homogeneous and homogeneity can be assessed subjectively or quantitatively using texture analysis to

differentiate AML.wovf from RCC.45 Therefore, a small renal mass in a female patient, which is hyperdense on NECT, homogeneous, and shows rapid enhancement with washout kinetics should prompt the potential diagnosis of AML.wovf (Fig 6). Despite a variety of imaging findings having been described in renal oncocytoma (e.g., central scar, segmental enhancement inversion),80e82 to date, there are no reliable features that can differentiate oncocytoma from RCC.83 The use of more advanced quantitative CT features (including enhancement thresholds and texture features) show promise for diagnosis of oncocytoma; however, lack validation on a large scale and results are limited by number of publications and lesions studied.84 Further study on the potential CT diagnosis of oncocytoma is required. In addition to benign renal neoplasms mimicking RCC, hilar urothelial cell carcinoma (UCC) can also mimic centrally located RCC. Previously described imaging features such as: filling defect in the renal pelvis, extension/growth towards the ureteropelvic junction, tumour centred on the collecting system, preservation of renal shape, homogeneous enhancement, and absence of cystic or necrotic change have been described as favouring a diagnosis of UCC rather than RCC.85

716

S. Krishna et al. / Clinical Radiology 72 (2017) 708e721

Figure 7 A 63-year-old man who underwent CT for characterisation of an incidental renal lesion detected on ultrasound (not shown). (a) Axial NECT, (b) CM phase, and (c) NG phase CECT images demonstrate a 3.5 cm lesion in the interpolar region of the left kidney (white arrows). The lesion is homogeneously isodense to the renal parenchyma on NECT with a relatively smooth margin and no internal calcification. ROI analysis showed indeterminate difference in attenuation comparing CM and NG CECT to NECT images with a maximum difference of 16 HU. MRI was performed for further characterisation. (d) Axial T1W FS post-gadolinium enhanced subtraction image shows enhancement within the nodule (white arrow). Papillary RCC was confirmed at partial nephrectomy.

Canadian guidelines for the evaluation of small renal masses have acknowledged that there is an increasing role of biopsy in patients with small renal masses and biopsy may be considered where a change of management is anticipated.86 Although it may be potentially beneficial to consider biopsy if a benign tumour-like AML.wovf or oncocytoma is suspected radiologically, the overlap of imaging features of these benign tumours with RCC makes this challenging. Active surveillance of small renal masses is another viable option, because 20e25% of small renal masses are benign87 and of the malignant lesions that are RCC, 70e80% are low-grade early-stage lesions with low malignant potential.88 Most tumours show slow indolent growth of 0.2e0.3 cm/year and about 20e30% show zero growth rate and risk of metastatic disease while undergoing active surveillance is about 1%.70,88 Current National Comprehensive Cancer Network (NCCN) guidelines, American Urological Association (AUA) guidelines, and Canadian guidelines recommend that active surveillance be

considered in elderly patients with high surgical risk.2,86,89 Follow-up CT in patients on active surveillance is recommended every 3 months during the first year, every 6 months during the next 2 years, and every year thereafter.86

Dose reduction and iterative reconstruction pitfalls (10) Low-dose CT may simulate heterogeneity due to increased noise. Iterative reconstruction techniques can lessen noise without altering CT attenuation values. Too much iterative reconstruction introduces smoothing into the final images and can make heterogeneous lesions appear homogeneous. Due to growing concerns over radiation exposure to patients, the ALARA (“as low as reasonably achievable”) principle has been liberally applied to CT protocols with the goal of producing diagnostic quality imaging with the lowest possible dose.90 The main disadvantage of low-dose

S. Krishna et al. / Clinical Radiology 72 (2017) 708e721

717

Figure 8 Phantom experiment illustrating the principle of pseudo-enhancement. A 10 mm impermeable water-filled balloon was suspended in a water-bath containing dilute iodine to simulate a renal cyst embedded in renal parenchyma (a). The balloon/cyst measures water attenuation (e5 HU) and the surrounding medium/renal parenchyma measures 65 HU. (b) Iodine was added to the medium to simulate a contrast-enhanced examination and the attenuation of the surrounding medium/renal parenchyma increases to 235 HU. Note that the attenuation of the balloon/ cyst increases to 22 HU, which is a >20 HU change compatible with enhancement. This artefactual increase in density is related to beamhardening artefact and is referred to as pseudo-enhancement. (cef) A clinical example of pseudo-enhancement in a 52-year-old woman. (c) Axial NECT, (d) CM phase, and (e) NG phase CECT images demonstrate a 10-mm well-circumscribed homogeneously hyperattenuating renal nodule arising from the lower pole of the left kidney (white arrows). The nodule increases in density by >20 HU on CECT, which fulfils the criteria for enhancement. Given the small size of the nodule and because it was partially surrounded by renal parenchyma, MRI was performed for confirmation. (f) Axial FS T1W post-gadolinium enhanced subtraction image shows no internal enhancement (white arrow). A diagnosis of pseudo-enhancing haemorrhagic cyst was made.

718

S. Krishna et al. / Clinical Radiology 72 (2017) 708e721

Figure 9 Phantom experiment illustrating how DECT can overcome pseudo-enhancement. (a) Monochromatic low (70 keV) and (b) high (140 keV) energy images demonstrate beam-hardening artefact is reduced at higher energies. (c) Iodine overlay image illustrates the absence of iodine within the balloon (subjectively showing similar iodine concentration to noise). (deg) A clinical example of how DECT can diagnose pseudo-enhancement in a 76-year-old man with a left renal nodule identified at DECT. (d) Axial virtual NECT image and (e) CM phase CECT images demonstrate a nodule (white arrows) arising from the posterior interpolar region of the left kidney. The nodule increases in attenuation by a maximum difference of 13 HU which is indeterminate. (f) Axial iodine overlay image demonstrates complete absence of iodine within the nodule (white arrow). (g) Transverse grey-scale ultrasound image performed 2 years earlier shows that the nodule is a minimally complex Bosniak type 2 cyst with a single thin hair-like internal septum (white arrow).

CT techniques is the increase in image noise.91 Image noise can result in deleterious effects when imaging renal masses by potentially obscuring low contrast structures thereby limiting detection and by simulating heterogeneity in an otherwise homogeneous lesion resulting in limited characterisation (Fig 10). Compared to the standard method of reconstructing raw data using filtered back projection (FBP), complex modelling termed iterative reconstruction can reduce image noise resulting in comparable image quality to a standard-dose CT reconstructed with FPB.92 A concern over iterative reconstruction techniques (compared to FPB) is the effect on quantitative attenuation

measurements. In a study by Shampain et al.,93 there was no difference in attenuation values with the use of iterative reconstruction in small renal masses. It should be noted that with higher levels of iterative reconstruction, there is reduced peak frequency in the noise power spectral curve,94 which results in “over-smoothing”, which has been variably described as image having “blotchy” or “plastic” features. This can result in alteration of tissue texture as well as loss of low-contrast spatial resolution.95 It is possible that this smoothening may simulate homogeneity in a heterogeneous renal lesion, which may potentially impact lesion characterisation.

S. Krishna et al. / Clinical Radiology 72 (2017) 708e721

719

Figure 10 Phantom experiment illustrating the interplay between dose and ability to adequately characterise renal lesions at NECT. (a) Standard dose image obtained with 120 kVp and 200 mAs shows a water balloon suspended in an increased density water-bath to simulate a cyst within renal parenchyma. The balloon/cyst measures of water attenuation and is completely homogeneous with a smooth imperceptible wall. (b) Repeat image obtained with drastically reduced dose (80 kVp, 10 mAs) alters the appearance of the cyst, which now appears heterogeneous with an irregular-appearing wall. Note that the density of the balloon/cyst still remains in the water attenuation range. Post-processing of source data from (b) was obtained by applying iterative reconstruction (adaptive statistical iterative reconstruction, GE healthcare; level of 100%) (c). Note that noise has been dramatically reduced in (c) compared to (b) and the balloon/cyst appears more homogeneous with a sharper margin. Attenuation continues to remain in the water attenuation range.

Conclusion In conclusion, CT is accurate for the characterisation of renal masses and offers several advantages over MRI as the first-line imaging technique for imaging of indeterminate renal lesions including faster examination times, relatively lower cost, and increased accessibility. Interpretive and technical pitfalls of CT are common in renal mass imaging and important to recognise as they may result in errors in diagnosis. This review article highlights pitfalls as they relate to NECT imaging, uniphasic and multi-phasic CECT imaging, and dose-reduction and iterative reconstruction algorithms. The emerging role of DECT in the characterisation of renal masses is presented, particularly as it applies to the improved detection of pseudo-enhancement of renal cysts. A discussion of diagnosing benign small (<4 cm) solid renal masses, including AML without visible fat and oncocytoma, is also presented.

References 1. O’Connor SD, Pickhardt PJ, Kim DH, et al. Incidental finding of renal masses at unenhanced CT: prevalence and analysis of features for guiding management. AJR Am J Roentgenol 2011;197(1):139e45. 2. Novick AC CS, Belldegrun A, et al. Guideline for management of the clinical stage 1 renal mass 2009. Available at: https://www.auanet.org/ documents/education/clinical-guidance/Renal-Mass.pdf. Accessed 19 May 2017. 3. Heilbrun M, Casalino D, Beland M, et al. ACR Appropriateness CriteriaÒ Acute Indeterminate renal mass. Available at https://acsearch.acr.org/docs/ 69367/Narrative/. American College of Radiology. Accessed 19 May 2017. 4. Willatt JM, Hussain HK, Chong S, et al. MR imaging in the characterization of small renal masses. Abdom Imaging 2014;39(4):761e9. 5. Ramamurthy NK, Moosavi B, McInnes, et al. Multiparametric MRI of solid renal masses: pearls and pitfalls. Clin Radiol 2015;70(3):304e16. 6. Marin D, Boll DT, Mileto A, et al. State of the art: dual-energy CT of the abdomen. Radiology 2014;271(2):327e42. 7. Graser A, Becker CR, Staehler M, et al. Single-phase dual-energy CT allows for characterization of renal masses as benign or malignant. Invest Radiol 2010;45(7):399e405.

8. Schieda N, Kielar AZ, Al Dandan O, et al. Ten uncommon and unusual variants of renal angiomyolipoma (AML): radiologicepathologic correlation. Clin Radiol 2015 Feb;70(2):206e20. 9. Hakim SW, Schieda N, Hodgdon T, et al. Angiomyolipoma (AML) without visible fat: ultrasound, CT and MR imaging features with pathological correlation. Eur Radiol 2016;26(2):592e600. 10. Jinzaki M, Silverman SG, Akita H, et al. Renal angiomyolipoma: a radiological classification and update on recent developments in diagnosis and management. Abdom Imaging 2014;39(3):588e604. 11. O’Connor SD, Silverman SG, Ip IK, et al. Simple cyst-appearing renal masses at unenhanced CT: can they be presumed to be benign? Radiology 2013;269(3):793e800. 12. Israel GM, Bosniak MA. How I do it: evaluating renal masses. Radiology 2005;236(2):441e50. 13. Lee-Felker SA, Felker ER, Tan N, et al. Qualitative and quantitative CT features for differentiating clear cell renal cell carcinoma from other solid renal cortical masses. AJR Am J Roentgenol 2014;203(5):W516e24. 14. Young JR, Margolis D, Sauk S, et al. Clear cell renal cell carcinoma: discrimination from other renal cell carcinoma subtypes and oncocytoma at multiphasic multidetector CT. Radiology 2013;267(2):444e53. 15. Schieda N, Isupov I, Chung A, et al. Practical applications of balanced steady-state free-precession (bSSFP) imaging in the abdomen and pelvis. J Magn Reson Imaging 2017;45(1):11e20. 16. Mileto A, Marin D, Alfaro-Cordoba M, et al. Iodine quantification to distinguish clear cell from papillary renal cell carcinoma at dual-energy multidetector CT: a multireader diagnostic performance study. Radiology 2014;273(3):813e20. 17. Zhao J, Zhang P, Chen X, et al. Lesion size and iodine quantification to distinguish low-grade from high-grade clear cell renal cell carcinoma using dual-energy spectral computed tomography. J Comput Assist Tomogr 2016;40(5):673e7. 18. Yuh BI, Cohan RH. Different phases of renal enhancement: role in detecting and characterizing renal masses during helical CT. AJR Am J Roentgenol 1999;173(3):747e55. 19. Welch TJ, LeRoy AJ. Helical and electron beam CT scanning in the evaluation of renal vein involvement in patients with renal cell carcinoma. J Comput Assist Tomogr 1997;21(3):467e71. 20. Shetty C, Lakhar B, Devi B, et al. Dual-phase helical CT of kidney: comparison of corticomedullary and nephrographic phases in detection and characterization of renal masses. Indian J Radiol Imaging 2004;14(3):285e90. 21. Birnbaum BA, Hindman N, Lee J, et al. Multi-detector row CT attenuation measurements: assessment of intra- and interscanner variability with an anthropomorphic body CT phantom. Radiology 2007;242(1):109e19.

720

S. Krishna et al. / Clinical Radiology 72 (2017) 708e721

22. Silverman SG, Leyendecker JR, Amis Jr ES. What is the current role of CT urography and MR urography in the evaluation of the urinary tract? Radiology 2009;250(2):309e23. 23. Johnson TR. Dual-energy CT: general principles. AJR Am J Roentgenol 2012;199(5 Suppl.):S3e8. 24. McCollough CH, Leng S, Yu L, et al. Dual- and multi-energy CT: principles, technical approaches, and clinical applications. Radiology 2015;276(3):637e53. 25. Coursey CA, Nelson RC, Boll DT, et al. Dual-energy multidetector CT: how does it work, what can it tell us, and when can we use it in abdominopelvic imaging? RadioGraphics 2010;30(4):1037e55. 26. Kaza RK, Platt JF, Cohan RH, et al. Dual-energy CT with single- and dualsource scanners: current applications in evaluating the genitourinary tract. RadioGraphics 2012;32(2):353e69. 27. Ascenti G, Mileto A, Krauss B, et al. Distinguishing enhancing from nonenhancing renal masses with dual-source dual-energy CT: iodine quantification versus standard enhancement measurements. Eur Radiol 2013;23(8):2288e95. 28. Kaza RK, Caoili EM, Cohan RH, et al. Distinguishing enhancing from nonenhancing renal lesions with fast kilovoltage-switching dual-energy CT. AJR Am J Roentgenol 2011;197(6):1375e81. 29. Zarzour JG, Milner D, Valentin R, et al. Quantitative iodine content threshold for discrimination of renal cell carcinomas using rapid kVswitching dual-energy CT. NY: Abdom Radiol 2017;42(3):727e34. 30. Graser A, Johnson TR, Hecht EM, et al. Dual-energy CT in patients suspected of having renal masses: can virtual nonenhanced images replace true nonenhanced images? Radiology 2009;252(2):433e40. 31. Maki DD, Birnbaum BA, Chakraborty DP, et al. Renal cyst pseudoenhancement: beam-hardening effects on CT numbers. Radiology 1999;213(2):468e72. 32. Mileto A, Nelson RC, Samei E, et al. Impact of dual-energy multidetector row CT with virtual monochromatic imaging on renal cyst pseudo-enhancement: in vitro and in vivo study. Radiology 2014;272(3):767e76. 33. Kaza RK, Raff EA, Davenport MS, et al. Variability of CT attenuation measurements in virtual unenhanced images generated using multimaterial decomposition from fast kilovoltage-switching dual-energy CT. Acad Radiol 2017;24(3):365e72. 34. Toepker M, Moritz T, Krauss B, et al. Virtual non-contrast in secondgeneration, dual-energy computed tomography: reliability of attenuation values. Eur J Radiol 2012;81(3):e398e405. 35. Connolly M, McInnes M, El-Khodary M, et al. Diagnostic accuracy of virtual non-contrast enhanced dual-energy CT for diagnosis of lipid-rich adrenal adenoma: a systematic review and meta-analysis. In: 102nd Scientific Assembly and Annual Meeting of Radiological Society of North America (RSNA 2016). Chicago; 2016. 36. Schieda N, Avruch L, Flood TA. Small (<1 cm) incidental echogenic renal cortical nodules: chemical shift MRI outperforms CT for confirmatory diagnosis of angiomyolipoma (AML). Insights Imaging 2014;5(3):295e9. 37. Catalano OA, Samir AE, Sahani DV, et al. Pixel distribution analysis: can it be used to distinguish clear cell carcinomas from angiomyolipomas with minimal fat? Radiology 2008;247(3):738e46. 38. Garin JM, Marco I, Salva A, et al. CT and MRI in fat-containing papillary renal cell carcinoma. Br J Radiol 2007;80(957):e193e195. 39. D’Angelo PC, Gash JR, Horn AW, et al. Fat in renal cell carcinoma that lacks associated calcifications. AJR Am J Roentgenol 2002;178(4):931e2. 40. Schuster TG, Ferguson MR, Baker DE, et al. Papillary renal cell carcinoma containing fat without calcification mimicking angiomyolipoma on CT. AJR Am J Roentgenol 2004;183(5):1402e4. 41. Richmond L, Atri M, Sherman C, et al. Renal cell carcinoma containing macroscopic fat on CT mimics an angiomyolipoma due to bone metaplasia without macroscopic calcification. Br J Radiol 2010;83(992):e179e181. 42. Helenon O, Merran S, Paraf F, et al. Unusual fat-containing tumors of the kidney: a diagnostic dilemma. RadioGraphics 1997;17(1):129e44. 43. Chen CL, Tang SH, Wu ST, et al. Calcified, minimally fat-contained angiomyolipoma clinically indistinguishable from a renal cell carcinoma. BMC Nephrol 2013;14:160. 44. Merran S, Vieillefond A, Peyromaure M, et al. Renal angiomyolipoma with calcification: CTepathology correlation. Br J Radiol 2004;77(921):782e3.

45. Hodgdon T, McInnes MD, Schieda N, et al. Can quantitative CT texture analysis be used to differentiate fat-poor renal angiomyolipoma from renal cell carcinoma on unenhanced CT images? Radiology 2015;276(3):787e96. 46. Pooler BD, Pickhardt PJ, O’Connor SD, et al. Renal cell carcinoma: attenuation values on unenhanced CT. AJR Am J Roentgenol 2012;198(5):1115e20. 47. Schieda N. Reply to “Solid renal cell carcinomas with an attenuation similar to that of water on unenhanced CT”. AJR Am J Roentgenol 2016;206(6):W93. 48. Schieda N, Vakili M, Dilauro M, et al. Solid renal cell carcinoma measuring water attenuation (-10 to 20 HU) on unenhanced CT. AJR Am J Roentgenol 2015;205(6):1215e21. 49. Helenon O, Correas JM, Balleyguier C, et al. Ultrasound of renal tumors. Eur Radiol 2001;11(10):1890e901. 50. Kim SHCJ, Kim SY, Moon KC, et al. Ultrasound evaluation of renal masses: gray-scale, Doppler, and more. Ultrasound Clin N Am 2013;8:565e79. 51. Jonisch AI, Rubinowitz AN, Mutalik PG, et al. Can high-attenuation renal cysts be differentiated from renal cell carcinoma at unenhanced CT? Radiology 2007;243(2):445e50. 52. Davarpanah AH, Spektor M, Mathur M, et al. Homogeneous T1 hyperintense renal lesions with smooth borders: is contrast-enhanced MR imaging needed? Radiology 2016;280(1):128e36. 53. Mileto A, Nelson RC, Paulson EK, et al. Dual-energy CT for imaging the renal mass. AJR Am J Roentgenol 2015;204(6):W640e7. 54. Szolar DH, Kammerhuber F, Altziebler S, et al. Multiphasic helical CT of the kidney: increased conspicuity for detection and characterization of small (< 3-cm) renal masses. Radiology 1997;202(1):211e7. 55. Birnbaum BA, Jacobs JE, Ramchandani P. Multiphasic renal CT: comparison of renal mass enhancement during the corticomedullary and nephrographic phases. Radiology 1996;200(3):753e8. 56. Cohan RH, Sherman LS, Korobkin M, et al. Renal masses: assessment of corticomedullary-phase and nephrographic-phase CT scans. Radiology 1995;196(2):445e51. 57. Egbert ND, Caoili EM, Cohan RH, et al. Differentiation of papillary renal cell carcinoma subtypes on CT and MRI. AJR Am J Roentgenol 2013;201(2):347e55. 58. Ruppert-Kohlmayr AJ, Uggowitzer M, Meissnitzer T, et al. Differentiation of renal clear cell carcinoma and renal papillary carcinoma using quantitative CT enhancement parameters. AJR Am J Roentgenol 2004;183(5):1387e91. 59. Kim JK, Kim TK, Ahn HJ, et al. Differentiation of subtypes of renal cell carcinoma on helical CT scans. AJR Am J Roentgenol 2002;178(6):1499e506. 60. Pedrosa I, Sun MR, Spencer M, et al. MR imaging of renal masses: correlation with findings at surgery and pathologic analysis. RadioGraphics 2008;28(4):985e1003. 61. Dilauro M, Quon M, McInnes MD, et al. Comparison of contrastenhanced multiphase renal protocol CT versus MRI for diagnosis of papillary renal cell carcinoma. AJR Am J Roentgenol 2016;206(2):319e25. 62. Al Harbi F, Tabatabaeefar L, Jewett MA, et al. Enhancement threshold of small (< 4 cm) solid renal masses on CT. AJR Am J Roentgenol 2016;206(3):554e8. 63. Bertolotto M, Cicero C, Perrone R, et al. Renal masses with equivocal enhancement at CT: characterization with contrast-enhanced ultrasound. AJR Am J Roentgenol 2015;204(5):W557e65. 64. Tappouni R, Kissane J, Sarwani N, et al. Pseudo-enhancement of renal cysts: influence of lesion size, lesion location, slice thickness, and number of CT detectors. AJR Am J Roentgenol 2012;198(1):133e7. 65. Yamada Y, Yamada M, Sugisawa K, et al. Renal cyst pseudoenhancement: intraindividual comparison between virtual monochromatic spectral images and conventional polychromatic 120-kVp images obtained during the same CT examination and comparisons among images reconstructed using filtered back projection, adaptive statistical iterative reconstruction, and model-based iterative reconstruction. Medicine (Baltimore) 2015;94(15):e754. 66. Song KD, Kim CK, Park BK, et al. Utility of iodine overlay technique and virtual unenhanced images for the characterization of renal masses by dual-energy CT. AJR Am J Roentgenol 2011;197(6):W1076e82. 67. Cha D, Kim CK, Park JJ, et al. Evaluation of hyperdense renal lesions incidentally detected on single-phase post-contrast CT using dualenergy CT. Br J Radiol 2016;89(1062):20150860.

S. Krishna et al. / Clinical Radiology 72 (2017) 708e721 68. Chandarana H, Megibow AJ, Cohen BA, et al. Iodine quantification with dual-energy CT: phantom study and preliminary experience with renal masses. AJR Am J Roentgenol 2011;196(6):W693e700. 69. Mileto A, Barina A, Marin D, et al. Virtual monochromatic images from dual-energy multidetector CT: variance in CT numbers from the same lesion between single-source projection-based and dual-source imagebased implementations. Radiology 2016;279(1):269e77. 70. Pierorazio PM, Hyams ES, Mullins JK, et al. Active surveillance for small renal masses. Rev Urol 2012;14(1e2):13e9. 71. Lim A, O’Neil B, Heilbrun ME, et al. The contemporary role of renal mass biopsy in the management of small renal tumors. Front Oncol 2012;2:106. 72. Kutikov A, Fossett LK, Ramchandani P, et al. Incidence of benign pathologic findings at partial nephrectomy for solitary renal mass presumed to be renal cell carcinoma on preoperative imaging. Urology 2006;68(4):737e40. 73. Phe V, Yates DR, Renard-Penna R, et al. Is there a contemporary role for percutaneous needle biopsy in the era of small renal masses? BJU Int 2012;109(6):867e72. 74. Schieda N, Dilauro M, Moosavi B, et al. MRI evaluation of small (<4 cm) solid renal masses: multivariate modeling improves diagnostic accuracy for angiomyolipoma without visible fat compared to univariate analysis. Eur Radiol 2016;26(7):2242e51. 75. Kang SK, Huang WC, Pandharipande PV, et al. Solid renal masses: what the numbers tell us. AJR Am J Roentgenol 2014;202(6):1196e206. 76. Woo S, Cho JY, Kim SH, et al. Angiomyolipoma with minimal fat and non-clear cell renal cell carcinoma: differentiation on CT using classification and regression tree analysis-based algorithm. Acta Radiol 2013;55(10):1258e69. 77. Jinzaki M, Tanimoto A, Narimatsu Y, et al. Angiomyolipoma: imaging findings in lesions with minimal fat. Radiology 1997;205(2):497e502. 78. Yang CW, Shen SH, Chang YH, et al. Are there useful CT features to differentiate renal cell carcinoma from lipid-poor renal angiomyolipoma? AJR Am J Roentgenol 2013;201(5):1017e28. 79. Schieda N, Hodgdon T, El-Khodary M, et al. Unenhanced CT for the diagnosis of minimal-fat renal angiomyolipoma. AJR Am J Roentgenol 2014;203(6):1236e41. 80. Schieda N, McInnes MD, Cao L. Diagnostic accuracy of segmental enhancement inversion for diagnosis of renal oncocytoma at biphasic contrast enhanced CT: systematic review. Eur Radiol 2014;24(6):1421e9. 81. Schieda N, Al-Subhi M, Flood TA, et al. Diagnostic accuracy of segmental enhancement inversion for the diagnosis of renal oncocytoma using biphasic computed tomography (CT) and multiphase contrast-enhanced magnetic resonance imaging (MRI). Eur Radiol 2014;24(11):2787e94.

721

82. Kim JI, Cho JY, Moon KC, et al. Segmental enhancement inversion at biphasic multidetector CT: characteristic finding of small renal oncocytoma. Radiology 2009;252(2):441e8. 83. Choudhary S, Rajesh A, Mayer NJ, et al. Renal oncocytoma: CT features cannot reliably distinguish oncocytoma from other renal neoplasms. Clin Radiol 2009;64(5):517e22. 84. Sasaguri K, Takahashi N, Gomez-Cardona D, et al. Small (<4 cm) renal mass: differentiation of oncocytoma from renal cell carcinoma on biphasic contrast-enhanced CT. AJR Am J Roentgenol 2015;205(5):999e1007. 85. Raza SA, Sohaib SA, Sahdev A, et al. Centrally infiltrating renal masses on CT: differentiating intrarenal transitional cell carcinoma from centrally located renal cell carcinoma. AJR Am J Roentgenol 2012;198(4):846e53. 86. Jewett MA, Rendon R, Lacombe L, et al. Canadian guidelines for the management of small renal masses (SRM). Can Urol Assoc J 2015;9(5e6):160e3. 87. Leveridge MJ, Finelli A, Kachura JR, et al. Outcomes of small renal mass needle core biopsy, nondiagnostic percutaneous biopsy, and the role of repeat biopsy. Eur Urol 2011;60(3):578e84. 88. Smaldone MC, Kutikov A, Egleston BL, et al. Small renal masses progressing to metastases under active surveillance: a systematic review and pooled analysis. Cancer 2012;118(4):997e1006. 89. NCCN. Clinical practice guidelines in oncology: renal cancer version 2.2016. Available at: https://www.nccn.org/professionals/physician_gls/ PDF/kidney.pdf. Accessed January 2017. 90. Pearce MS, Salotti JA, Little MP, et al. Radiation exposure from CT scans in childhood and subsequent risk of leukaemia and brain tumours: a retrospective cohort study. Lancet 2012;380(9840):499e505. 91. Seibert JA. Tradeoffs between image quality and dose. Pediatr Radiol 2004;34(Suppl. 3):S183e95. discussion S234e141. 92. Sagara Y, Hara AK, Pavlicek W, et al. Comparison of low-dose CT with adaptive statistical iterative reconstruction and routine-dose CT with filtered back projection in 53 patients. AJR Am J Roentgenol 2010;195(3):713e9. 93. Shampain KL, Davenport MS, Cohan RH, et al. Effect of model-based iterative reconstruction on CT number measurements within small (10e29 mm) low-attenuation renal masses. AJR Am J Roentgenol 2015;205(1):85e9. 94. Greffier J, Macri F, Larbi A, et al. Dose reduction with iterative reconstruction: optimization of CT protocols in clinical practice. Diagn Interv Imaging 2015;96(5):477e86. 95. Ehman EC, Yu L, Manduca A, et al. Methods for clinical evaluation of noise reduction techniques in abdominopelvic CT. RadioGraphics 2014;34(4):849e62.