MR imaging of rectal cancer: Radiomics Analysis to Assess Treatment Response after Neoadjuvant Therapy.
Natally Horvat, Harini Veeraraghavan, Monika Khan, Ivana Blazic, Junting Zheng,Marinela Capanu, Evis Sala, Julio Garcia-Aguilar, Marc J. Gollub, Iva Petkovska. Radiology 2018: https://doi.org/10.1148/radiol.2018172300
Dr. Carole Allimant – Centre Hospitalier Universitaire de Montpellier, Department of Radiodiagnostic and Interventional Radiology, Montpellier, France
Dr. Clarisse Dromain – Centre Hospitalier Universitaire Vaudois (CHUV), Department of Radiodiagnostic and Interventional Radiology, Lausanne, Switzerland.
The standard treatment for patients with MR imaging–staged locally advanced rectal cancer (LARC) is neoadjuvant chemotherapy–radiation therapy (CRT) followed by total mesorectal excision. After neoadjuvant CRT, approximately 50% to 60% of patients are downstaged and 15% to 27% show pathologic complete response (pCR)(1).
Selected patients with cCR can be safely treated with CRT (2). However a nonsurgical “watch-and-wait” strategy is not currently supported by the National Comprehensive Cancer Network guidelines (3) and there is an increasing need for evaluation of tumor response.
Radiomics analysis involves computer-based extraction of a large number of quantitative features representing voxel heterogeneity(5) that appears to be an interesting emergent tool for predicting response after CRT (6–8).
This is a retrospective study of patients with LARC who underwent neoadjuvant-CRT followed by surgery at Memorial Sloan Kettering Cancer Center from March 2012 to February 2016. Rectal MR imaging was performed less than 3 months before surgery. Among others, mucinous tumor, recurrent rectal cancer and poor image quality MRI were excluded.
Their final study population consisted of 114 patients. The minimum sequence required was high-spatial-resolution axial oblique T2WI perpendicular to the long axis of the tumor and/or area after treatment, with section thickness of 3 mm. DWI (section thickness of 5mm, 0/800 b-values) was available in only 106 patients.
Two radiologists (5 years and 9 years of experience) reviewed T2WI and DWI, blinded to endoscopic and histopathologic results, to classify the area after treatment as either cCR or cPR in consensus. They had the primary staging MR images at their disposal to guide the delimitation of the tumor bed. For combined assessment, the patient was classified as cCR when cCR was obtained on both T2 or DW images. When combined imaging features at T2WI and DWI were used to diagnose cCR, high sensitivity (84%) and NPV (94%) were achieved; however, it resulted in low specificity (56%) and PPV (30%).
One of the radiologists manually segmented the entire area after treatment, excluding equivocal normal rectal wall and mucosal edema on the T2WI image to provide the volume of interest for texture analysis. Thirty-four texture features consisting of first order (mean, standard deviation, kurtosis, and skewness), second-order Haralick (9) (energy, entropy, correlation, contrast, and homogeneity), Gabor edges at four different orientations and bandwidth (g = 1.4), and second-order texture measures on the Gabor edges were computed.
They used Random Forests (RF) classifier(10) which is a complex model of multiple decision tree–based classifiers of those features. At the end they computed the relative importance of each feature by using the Gini index (10). Second-order texture measures of energy, homogeneity, and contrast were among features that resulted in the largest decrease in the Gini index ie that resulted in the least misclassification (cCR vs pCR). Patients with pCR had significantly higher energy, higher homogeneity, and lower contrast compared with patients with pPR. This indicates that tumors with a homogenous appearance were associated with pCR.
Fourteen of the 34 features were significantly different between pCR and pPR. Their best RF classifier produced high diagnostic performance with an area under the curve of 93% (95%CI: 0.87-0.97) for differentiating pCR from pPR, with a sensitivity of 100% (95% CI: 0.84, 1), specificity of 91% (95%CI: 0.84-0.96), PPV of 72% (95%CI:0.53-0.87),and NPV of 100% (95%CI: 0.96-1).
When compared with the assessment of both T2WI and DWI, radiomics had significantly higher specificity and PPV; however, sensitivity and NPV were not statistically different.
This is an interesting preliminary study showing that radiomic measures can offer better classification performance compared with the qualitative assessment for diagnosing pCR.
They postulate that small clusters of residual tumor cells within the residual area after treatment in pPR may explain the difference in texture in pCR and pPR; however, additional studies with direct correlation to whole-specimen histopathologic analysis are necessary to investigate this further.
Sixty-two (55%) patients were imaged with 1.5-T units, whereas the remaining 52 (45%) were imaged with 3.0-T units. Most computed texture measures showed no significant difference by magnet strength. This was an important finding given the large number of MR imagers available in the market to allow the widespread use of these new imaging techniques.
Although promising and because radiomics classification follows data augmentation it could potentially lead to overoptimistic results, these results require validation on a larger and independent data set to assess the potential for clinical translation.
Moreover, manual segmentation of volume of interest is a time-consuming process; therefore, it is important to develop a user-friendly tool to encourage the use of radiomic measures in daily clinical practice.
1. van de Velde CJH, Boelens PG, Borras JM, Coebergh J-W, Cervantes A, Blomqvist L, et al. EURECCA colorectal: multidisciplinary management: European consensus conference colon & rectum. Eur J Cancer Oxf Engl 1990. 2014 Jan;50(1):1.e1-1.e34.
2. Renehan AG, Malcomson L, Emsley R, Gollins S, Maw A, Myint AS, et al. Watch-and-wait approach versus surgical resection after chemoradiotherapy for patients with rectal cancer (the OnCoRe project): a propensity-score matched cohort analysis. Lancet Oncol. 2016 Feb;17(2):174–83.
3. NCCN Guidelines. National Comprehensive Cancer Network. NCCN Clinical Practice Guidelines in Oncology. Rectal Cancer version [Internet]. 2017 [cited 2018 Mar 18]. Available from: www.nccn.org/pro- fessionals/physician_gls/pdf/rectal.pdf
4. Patel UB, Brown G, Rutten H, West N, Sebag-Montefiore D, Glynne-Jones R, et al. Comparison of magnetic resonance imaging and histopathological response to chemoradiotherapy in locally advanced rectal cancer. Ann Surg Oncol. 2012 Sep;19(9):2842–52.
5. Nie K, Shi L, Chen Q, Hu X, Jabbour SK, Yue N, et al. Rectal Cancer: Assessment of Neoadjuvant Chemoradiation Outcome based on Radiomics of Multiparametric MRI. Clin Cancer Res Off J Am Assoc Cancer Res. 2016 Nov 1;22(21):5256–64.
6. De Cecco CN, Ciolina M, Caruso D, Rengo M, Ganeshan B, Meinel FG, et al. Performance of diffusion-weighted imaging, perfusion imaging, and texture analysis in predicting tumoral response to neoadjuvant chemoradiotherapy in rectal cancer patients studied with 3T MR: initial experience. Abdom Radiol N Y. 2016 Sep;41(9):1728–35.
7. Cusumano D, Dinapoli N, Boldrini L, Chiloiro G, Gatta R, Masciocchi C, et al. Fractal-based radiomic approach to predict complete pathological response after chemo-radiotherapy in rectal cancer. Radiol Med (Torino). 2018 Apr;123(4):286–95.
8. Liu Z, Zhang X-Y, Shi Y-J, Wang L, Zhu H-T, Tang Z, et al. Radiomics Analysis for Evaluation of Pathological Complete Response to Neoadjuvant Chemoradiotherapy in Locally Advanced Rectal Cancer. Clin Cancer Res Off J Am Assoc Cancer Res. 2017 Dec 1;23(23):7253–62.
9. Haralick RM, Shanmugam K, Dinstein IH. Textural features for image classi cation. Trans Syst Man Cybern 1973;SMC- 3(6):610–621. 1973;
10. Breiman L. Random Forests. Mach Learn. 2001 Oct 1;45(1):5–32.
Dr. Carole Allimant works at the Centre Hospitalier Universitaire de Montpellier, more specifcially at the Department of Radiodiagnostic and Interventional Radiology. In 2017, Dr. Allimant finished her residency of which she spent an inter-university exchange of 7 months at the CHUV, Pr Denis, in Lausanne/CH. At this year's ECR, she held an oral presentation on «Tumor targeting and 3D voxel-based dosimetry to predict tumor response, toxicity and survival after Y-90 resin microsphere radioembolization in HCC»
Comments may be sent to Dr. Carole Allimant.
Long-term Risk of Pancreatic Malignancy in Patients With Branch Duct Intraductal Papillary Mucinous Neoplasm in a Referral Center
Ilaria Pergolini, Klaus Sahora, Cristina R. Ferrone, Vicente Morales-Oyarvide, Brian M. Wolpin, Lorelei A. Mucci, William R. Brugge, Mari Mino-Kenudson, Manuel Patino, Dushyant V. Sahani, Andrew L. Warshaw, Keith D. Lillemoe, and Carlos Fernández-del Castillo. Gastroenterology 2017;153:1284-1294.
Dr. Clarisse Dromain and Dr. Rafael Duran – Centre Hospitalier Universitaire Vaudois (CHUV), Department of Radiodiagnostic and Interventional Radiology, Lausanne, Switzerland.
Intraductal papillary mucinous neoplasm (IPMN) can be classified into three types: branch duct IPMN (BD-IPMN), main duct IPMN and mixed type. Because of their increased likelihood to develop malignancy and/or symptoms, main duct and mixed-type IPMNs more often undergo surgical therapy (1). The management of BD-IPMN is more challenging and long-term outcomes are uncertain with limited knowledge of the natural evolution and malignancy potential of pancreatic cysts beyond 5 years (2).
In the study by Pergolini et al, long-term outcomes of a large cohort of patients undergoing primary surveillance for BD-IPMN were evaluated to determine the risk of pancreatic cancer development after 5 years of follow-up (2). In addition, the authors sought to identify whether any subset of low-risk BD-IPMNs can safely forego surveillance after 5 years of follow-up.
This is a retrospective study in which consecutive patients with a clinical diagnosis of BD-IPMN followed and treated at the Massachusetts General Hospital with a minimum of 6 months of follow-up were included. Patients had at least one cross-sectional imaging (CT, MRI and/or endoscopic ultrasound) done 3 months or longer after the initial diagnosis. Diagnosis of BD-IPMN was based on the presence of unilocular or multilocular cysts of the pancreas and a non-dilated main pancreatic duct (MPD) (<5 mm). All patients with a dilation of the MPD > 5 mm or with cysts suspicious for another diagnosis rather than BD-IPMN (e.g., serous cystadenoma, mucinous cystic neoplasm, pseudocyst) or prior diagnosis of pancreatic cancer were excluded. Imaging and medical charts were reviewed. Patients were classified at the initial diagnosis as having “suspected BD-IPMN” when communication between the cyst and the MPD was documented or when cysts were multifocal or had a cyst fluid carcinoembryonic antigen >192 ng/mL. All remaining cysts were classified as “presumed BD-IPMN” (3). Charts of patients were reviewed for high-risk stigmata (HRS) and worrisome features (WFs). HRS were defined by the presence of obstructive jaundice in a patient with cystic lesion of the head of the pancreas, MPD ≥10 mm, enhancing solid component within the cyst, or cytology positive for high-grade dysplasia or adenocarcinoma. WFs were acute pancreatitis, cyst size ≥ 3 cm, thickened/enhancing cyst walls, non-enhancing mural nodules, MPD size of 5-9 mm, abrupt change in caliber of the MPD and lymphadenopathy (1).
577 patients with suspected or presumed BD-IPMN were included. Median follow-up for the entire cohort was 82 months (range, 6-329 months). 363 patients (63%) underwent surveillance for > 5 years, and 121 (21%) were followed for > 10 years. At diagnosis, a majority of the patients were asymptomatic (n=479, 83%) and of these, 48 (10%) developed symptoms during follow-up. Median cyst size at diagnosis was 14 mm (range 2-54 mm) and 53 patients (9%) had a cyst ≥3 cm. During follow-up, an additional 76 patients developed a lesion >3cm and the median cyst diameter had increased to 20 mm (range, 3-90 mm). At the end of the observational period, 319 patients (55%) had increased cyst size; median increase was 0.9 mm per year of follow-up. Patients who developed malignancy had a significantly larger cyst size at diagnosis compared with those without malignant transformation (19 vs. 14 mm; P =.047), as well as a greater maximum diameter during follow-up (32 vs. 19 mm; P < .001).
Overall, 7.8% of the entire patient population (45/577) developed pancreatic malignancy during the study period; in 36 patients malignancy arose within the pancreatic cyst and 9 patients malignancy arose away from the cyst (distinct or concomitant pancreatic cancer). When looking at the risk of malignancy within 5 years of follow-up, 4.3% (25/577) of the patients developed pancreatic malignancy.
Of the 363 patients followed for > 5 years, malignancy occurred in 5.5% (20/363) and was still present (4.1%) after 10 year follow-up. 22% of patients who were followed for > 5 years had either WFs or HRS by the 5-year time point, and 10% of them developed pancreatic malignancy subsequently. Importantly, the absence of WFs or HRS at a 5-year time point did not exclude the development of malignancy and these patients (n=282) had an 18.8 times higher age-standardized incidence rate of pancreatic malignancy than expected in the general population.
In the analysis, a subgroup of cysts appeared to have a significantly lower risk of development of malignancy. Among the 363 patients followed for > 5 years, 108 (30%) had a cyst that was ≤1.5 cm in size, and only 1 patient in this group (0.9%) developed malignancy. In contrast, 19 patients (7.5%) with cysts >1.5 cm developed malignancy (P=.01). The 1.5-cm cutoff showed a negative predictive value for malignancy of 99% and a sensitivity of 95%.
BD-IPMN is increasingly diagnosed on imaging. The main challenge is to identify which of these BD-IPMNs are susceptible to malignant transformation to define adequate surveillance programs. The overall risk of malignancy in the present study, almost 8%, lasted for 10 years or more for patients with BD-IPMNs. Although it is important to stress that this study cohort, composed of selected patients referred to surgeons and gastroenterologists, may not be representative of the population of incidentally discovered cysts, presented results support continued surveillance after 5 years from the initial diagnosis. This information is relevant because recently published American Gastroenterological Association guidelines recommend stopping surveillance after 5 years (4) while the International Consensus Guidelines 2012 provides recommendations regarding indications for surgical resection and frequency of follow-up on patients managed non-operatively, but does not give an end point to surveillance (1). Pancreatic cysts that remain ≤ 1.5 cm for more than 5 years might be considered low-risk for progression to malignancy.
1. Tanaka M, Fernandez-del Castillo C, Adsay V, Chari S, Falconi M, Jang JY, et al. International consensus guidelines 2012 for the management of IPMN and MCN of the pancreas. Pancreatology : official journal of the International Association of Pancreatology. 2012;12(3):183-97.
2. Pergolini I, Sahora K, Ferrone CR, Morales-Oyarvide V, Wolpin BM, Mucci LA, et al. Long-term Risk of Pancreatic Malignancy in Patients With Branch Duct Intraductal Papillary Mucinous Neoplasm in a Referral Center. Gastroenterology. 2017;153(5):1284-94 e1.
3. Mukewar S, de Pretis N, Aryal-Khanal A, Ahmed N, Sah R, Enders F, et al. Fukuoka criteria accurately predict risk for adverse outcomes during follow-up of pancreatic cysts presumed to be intraductal papillary mucinous neoplasms. Gut. 2017;66(10):1811-7.
4. Vege SS, Ziring B, Jain R, Moayyedi P, Clinical Guidelines C, American Gastroenterology A. American gastroenterological association institute guideline on the diagnosis and management of asymptomatic neoplastic pancreatic cysts. Gastroenterology. 2015;148(4):819-22; quize12-3.
Rafael Duran is a junior attending in Radiology specialized in Interventional Radiology, more specifically Interventional Oncology. Dr. Duran works on the development of novel tumor response criteria and imaging biomarkers following locoregional therapies and novel drugs such as agents targeting tumor metabolism or hypoxia. His current main research focuses on liver cancer molecular biology and immnuo-oncology in the setting of image-guided locoregional therapies.
Comments may be sent to Dr. Rafael Duran.
Non-invasive measurement of liver iron concentration using 3-Tesla magnetic resonance imaging: validation against biopsy
Gaspard d’Assignies, Anita Paisant, Edouard Bardou-Jacquet, Anne Boulic, Elise Bannier, Fabrice Lainé, Martine Ropert, Jeff Morcet, Hervé Saint-Jalmes, Yves Gandon.
Eur Radiol. 2017 Nov 24. doi: 10.1007/s00330-017-5106-3.
Dr. Lara Sammut, Higher Specialist Trainee, Mater Dei Hospital, Msida/MT
Dr. Kelvin Cortis, Consultant Radiologist, Mater Dei Hospital, Msida/MT
Liver iron overload is the histological hallmark of hereditary hemochromatosis, transfusional hemosiderosis, as well as certain acquired liver diseases such as non-alcoholic steatohepatitis (NASH) and dysmetabolic iron overload syndrome (DIOS). Excess accumulation of iron in the liver is toxic, and can lead to complications such as cirrhosis and hepatocellular carcinoma. The liver iron content (LIC) (mg Fe/g dry tissue) correlates strongly with total body iron, thus LIC can be used as a surrogate biomarker of body iron stores [1,2].
MRI and genetic testing has drastically reduced the need for iron quantification through liver biopsy , which is both invasive and prone to sampling bias. The two main LIC MRI quantification methods are the signal- intensity- ratio (SIR) and the relaxometry (R2*) method. The SIR method has certain drawbacks, including failure to account for fat effects and tendency to overestimate LIC. This is not the case for the R2* method, which is based on the calculation of T2/R2 or T2*/R2* from signal decay at various echo times.
The aim of the study carried out by d’Assignies et al. was to evaluate the ability of the R2* method at 3T field strength to detect and quantify liver iron, using liver biopsy with biochemical LIC results as a reference method. The authors also compared both SIR and R2* methods at the 3T field strength.
A total of 105 patients who underwent a liver biopsy with estimation of biochemical LIC (LICb) were included prospectively in this study. The LICb concentration ranged from 0 to 630 μmol/g, and 49 patients (47%) had normal LICb values (less than 36 μmol/g). 56 patients had liver iron overload, this was due to genetic haemochromatosis in 31 patients, dysmetabolic syndrome in 22 patients and to other causes (alcoholic or viral hepatitis) in 3 patients.
All of the patients underwent a breath-hold 3T multiple-echo gradient-echo sequence (mGRE) using a body coil and the LIC was calculated using both the SIR (LICSIR) and the R2* (LICR2*) algorithm respectively. Results were correlated with LICb from liver biopsies.
Results showed a strong positive correlation between LICb with both LICR2* (r = 0.95, p < 0.001) and LICSIR (r = 0.92, p < 0.001) methods.
In comparison to LICb, LICR2* and LICSIR were found to detect liver iron overload with a sensitivity/specificity of 96%/93% and 92%/95% respectively. LICR2* presented the lowest differences for patients with LICb values under 130 μmol/g. Above this value, LICSIR was found to have the lowest differences. The authors thus suggest using relaxometry for iron overload below 130 μmol/g and SIR for values above 130 μmol/g because of its better correlation with LIC.
The authors state that the R2* calculation is well-established at 1.5T, however no such validation with biopsies has been done before with at 3T, making the study highly practical since a significant proportion of new MRI scanners use a 3T field strength.
Dedicated DICOM freeware and detailed sequence parameters for scanners from different manufacturers are available on their website www.mrquantif.org. This software uses both methods jointly and selects the best one, and outputs reports and curves that can be imported in the hospital PACS system.
1. Hernando D, Levin YS, Sirlin CB, Reeder SB. Quantification of Liver Iron with MRI: State of the Art and Remaining Challenges. J Magn Reson Imaging 2014; 40(5): 1003–1021.
2. Blachier M, Leleu H, Peck-Radosavljevic M et al. The burden of liver disease in Europe: a review of available epidemiological data. J Hepatol 2013; 58:593–608
3. Brissot P, Troadec MB, Bardou-Jacquet E et al. Current approach to hemochromatosis. Blood Rev. 2008;22(4):195-210.
Dr. Lara Sammut is a fourth-year radiology resident at Mater Dei Hospital, Malta. An active member of ESGAR, she regularly attends workshops organized by the society as well as had the opportunity to present scientific studies at the society’s annual meetings, amongst others. She completed her undergraduate medical degree at the University of Malta in 2011. Her main interests are breast and diagnostic abdominal radiology. She aspires to take part in research studies abroad in the near future.
Comments are to be sent directly to Dr. Cortis or Dr. Sammut.
Magnetic Resonance Elastography is as Accurate as Liver Biopsy for Liver Fibrosis Staging
Hiroyuki Morisaka, Utaroh Motosugi, Shintaro Ichikawa, Tadao Nakazawa, Tetsuo Kondo, Satoshi Funayama, Masanori Matsuda, Tomoaki Ichikawa, Hiroshi Onishi. J Magn Reson Imaging 2017 Oct 14. doi: 10.1002/jmri.25868
Dr. Nathania Bonanno, Higher Specialist Trainee, Mater Dei Hospital, Msida/MT
Dr. Kelvin Cortis, Consultant Radiologist, Mater Dei Hospital, Msida/MT
The importance of diagnosing hepatic fibrosis at an early, potentially reversible, stage is that of preventing complications arising from progression to cirrhosis (including aggravated risk of having hepatocellular carcinoma and deranged liver function that may lead to liver failure), and also preventing the complications of portal hypertension (including ascites and variceal bleeding). This introductory concept set the tone for the study carried out by Morisaka et al., which has just been published in the September issue of the Journal of Magnetic Resonance Imaging (impact factor 3.083).
Percutaneous image-guided liver biopsy is an imperfect gold standard, with limitations that include sampling errors and intra/inter-observer variability. Various image-based methods have been employed for the non-invasive estimation of liver fibrosis, including T1 and T2 measurements, morphological assessment of hepatic architecture, contrast-enhancement, and elastography. Ultrasonography (US)-based elastography1 and magnetic resonance elastography (MRE)2 have already been well-validated for evaluating liver fibrosis. In fact, prior studies have shown that liver MRE has a high repeatability3-5 and has been cross-validated with biopsy and US-based elastography [6,7].
The study conducted by Morisaka et al. is novel, in that the authors ventured beyond the boundaries of former studies and used resected liver specimens as reference standards for liver fibrosis. The authors did so with the aim of comparing the diagnostic accuracy of fibrosis staging by MRE and liver biopsy. This is in contrast to earlier studies wherein MRE could not be directly compared to liver biopsy because biopsy was used as a reference standard [8-10].
The study population was retrospectively identified. In all, 200 patients who underwent preoperative MRE (using 2D liver MRE with gradient-echo sequence on a 1.5 or 3T scanner) and subsequent surgical liver resection were included in this study. Next, the patient cohort was distributed to two datasets, the preliminary set (n=80) and the validation set (n=120). Data from the former were used to estimate cut-off and distributions of liver stiffness values measured by MRE for each of the five liver fibrosis stages as determined by the METAVIR scoring system (F0-F4). In the remaining 120 patients – the study group – liver biopsy specimens were obtained from the resected liver tissues using a standard biopsy needle and then examined to compare the diagnostic accuracies of MRE- and liver biopsy-based fibrosis staging methods.
Two experienced abdominal radiologists, blinded to the histological results, independently measured the liver stiffness value on MRE. Two types of MRE-based liver fibrosis staging methods (threshold and Bayesian prediction method) were applied to the assembled data. Two pathologists, also blinded, evaluated all biopsy samples independently to stage liver fibrosis using surgically resected whole tissue specimens as the reference standard.
The results from the first radiologist and pathologist were used for the main analyses and that from the second observer was used for the assessment of inter-observer agreement.
For the analytic part of the study, the authors compared statistical accuracy for liver fibrosis staging (graded as significant fibrosis ≥F2, severe fibrosis ≥F3, and cirrhosis F4) between liver biopsy and the two MRE-based methods using a modified McNemar’s test (which is, simply put, a test to determine whether there is statistical difference between two paired populations from which the samples are drawn).
The overall diagnostic accuracy of MRE marginally surpassed that of liver biopsy – 59.1% (71/120) and 51.6% (62/120), respectively. Diagnostic accuracies of MRE for significant fibrosis, severe fibrosis, and cirrhosis were all statistically equivalent to that of liver biopsy (p ≤ 0.017). The proportions of under- and over-diagnosis between liver biopsy and MRE were also statistically comparable. The proportion of precise staging, i.e., not over- and under-staging, was 50-60% overall for both MRE- and liver biopsy-based staging.Morisaka et al. set out to prove that MRE can be an alternative to liver biopsy for fibrosis staging; they have successfully done so by demonstrating that the diagnostic accuracy of MRE-based methods for liver fibrosis staging is equivalent to that of liver biopsy.
This article is presently one of a kind and its results may reshape current practice –the need for liver biopsy will be alleviated and MRE can make its way to the forefront as the leading method of liver fibrosis staging. It offers a non-invasive, less-costly alternative to liver biopsy sampling for assessing fibrosis in patients with chronic liver disease.
The article is exceptionally-written, comprehensive, and certainly worth the read.
1. Castera L, Vergniol J, Foucher J, et al. Prospective comparison of transient elastography, Fibrotest, APRI, and liver biopsy for the assessment of fibrosis in chronic hepatitis C. Gastroenterology 2005;128: 343–350.
2. Venkatesh SK, Yin M, Ehman RL. Magnetic resonance elastography of liver: technique, analysis, and clinical applications. J Magn Reson Imaging 2013;37:544–555.
3. Hines CD, Bley TA, Lindstrom MJ, Reeder SB. Repeatability of magnetic resonance elastography for quantification of hepatic stiffness. J Magn Reson Imaging 2010;31:725–731.
4. Shire NJ, Yin M, Chen J, et al. Test-retest repeatability of MR elastography for noninvasive liver fibrosis assessment in hepatitis C. J Magn Reson Imaging 2011;34:947–955.
5. Venkatesh SK, Wang G, Teo LL, Ang BW. Magnetic resonance elastography of liver in healthy Asians: normal liver stiffness quantification and reproducibility assessment. J Magn Reson Imaging 2014;39:1–8.
6. Oudry J, Chen J, Glaser KJ, Miette V, Sandrin L, Ehman RL. Cross-validation of magnetic resonance elastography and ultrasound-based transient elastography: a preliminary phantom study. J Magn Reson Imaging 2009;30:1145–1150.
7. Motosugi U, Ichikawa T, Amemiya F, et al. Cross-validation of MR elastography and ultrasound transient elastography in liver stiffness measurement: discrepancy in the results of cirrhotic liver. J Magn Reson Imaging 2012;35:607–610.
8. Venkatesh SK, Wang G, Lim SG, Wee A. Magnetic resonance elastography for the detection and staging of liver fibrosis in chronic hepatitis B. Eur Radiol 2014;24:70–78.
9. Batheja M, Vargas H, Silva AM, et al. Magnetic resonance elastography (MRE) in assessing hepatic fibrosis: performance in a cohort of patients with histological data. Abdom Imaging 2015;40:760–765.
10. Yin M, Glaser KJ, Talwalkar JA, Chen J, Manduca A, Ehman RL. Hepatic MR elastography: clinical performance in a series of 1377 consecutive examinations. Radiology 2015:142141.
Dr. Nathania Bonanno is a third year radiology resident at Mater Dei Hospital, Malta. She completed her undergraduate medical degree at the University of Malta in 2011. Her main interests are in diagnostic and interventional abdominal and pelvic radiology, with special regard to oncological imaging. She has presented multiple scientific studies at numerous international and national radiology conferences.
Comments are to be sent directly to Dr. Cortis or Dr. Bonanno.
Early peribilary hyperenhancement on MRI in patients with Primary Sclerosing Cholangitis: significance and association with the Mayo Risk Score.
Ni Mhuircheartaigh JM, Lee KS, Curry MP, Pedrosa I, Mortele KJ. Abdom Radiol (NY). 2017 Jan;42(1):152-158.
Dr. Christopher Tang, Radiology Registrar, Guy’s and St Thomas’ NHS FT, London/GB
Dr. Sofia Gourtsoyianni, Consultant Radiologist, Guy’s and St Thomas’ NHS FT, London/GB
This paper published by Mhuircheartaigh et al in Abdominal Radiology focused on the use of contrast enhanced MRI in patients with Primary Sclerosing Cholangitis (PSC). Two main aims were evaluated: (1) Whether there is any relationship between peribiliary hyperenhancement on MRI and the Mayo risk score, and (2) Whether peribiliary hyperenhancement on MRI correlates with overall clinical outcome.
MRI is well established in the diagnosis of PSC, however there has been limited evidence characterising its role in terms of predicting disease severity and outcome. The Mayo risk score is a validated clinical model calculating estimated survival points for PSC patients, based on patient age, history of variceal bleeding, and blood markers including bilirubin, albumin and aspartate aminotransferase (AST).
This study was performed retrospectively, including all patients with known or suspected PSC who underwent contrast-enhanced MRI of their liver over a 4 year period. As part of the MRI protocol, a gadolinium based contrast was used, with post-contrast images obtained during the arterial, portal venous (45s after the arterial phase), and delayed venous phase (90s after the arterial phase). Two abdominal imaging radiologists independently reviewed the MRI studies, evaluating for the absence or presence of peribiliary hyperenhancement on each phase of imaging. If hyperenhancement was present, then it was classified as localised, segmental or diffuse.
A total of 62 patients were identified, who had their diagnosis of PSC confirmed by a combination of clinical assessment, ERCP and MRCP findings and liver biopsy (in a minority). These patients were initially stratified into low (n=41), medium (n=14) and high (n=7) risk groups based on their Mayo risk score.
The interobserver agreement for peribiliary hyperenhancement was good (k=0.919), with a consensus reached for those cases where there was initial disagreement. Accounting for this, the authors demonstrated a significant difference in the prevalence of arterial peribiliary hyperenhancement between the intermediate and low risk groups (p<0.05), with a difference in prevalence of arterial peribiliary hyperenhancement approaching significance between the high and low risk groups (p=0.059). No significant association was noted with later phase enhancement. Furthermore, there was a significant association between the extent of peribiliary hyperenhancement with the Mayo risk group (p<0.01), with more diffuse enhancement associated with a high Mayo risk.
A time to event analysis was also performed based on the presence of arterial peribiliary hyperenhancement, using a combined end point of death or liver transplantation. This demonstrated a statistically significant difference in survival times between patients who exhibited arterial peribiliary hyperenhancement and those who did not.
The study finding of a statistically significant association between arterial peribiliary hyperenhancement and higher Mayo risk scores and thus poorer prognosis is interesting, since the role of MRI in monitoring PSC is currently unclear. Although no causality was demonstrated, the authors postulate that arterial peribiliary hyperenhancement is a reflection of ongoing acute inflammation and cholangitis, which would intuitively suggest a worse prognosis. PSC is a slowly progressive disease, and current management relies on continual monitoring of disease progression. Although the Mayo risk score already provides a way to prognosticate PSC patients, one potential advantage of using MRI to assess for arterial peribiliary hyperenhancement is the ability to assess for disease heterogeneity within an individual and potentially monitor treatment response.
The authors note some study limitations. Firstly the study only demonstrated an association of arterial peribiliary hyperenhancement with the Mayo score, which is a surrogate marker of progression, rather than a true assessment of outcome. Secondly, the study cohort only included a relatively small number of patients who had high Mayo risk scores. Finally, the study only evaluated for peribiliary hyperenhancement, and other clinical features of PSC such as biliary strictures and cirrhosis were not assessed. The role of these other factors could also be evaluated in future research. Correlation with T2 weighted signal intensity and/or DWI characteristics of peribiliary hyperenhancement areas might also be worth investigating in the future.
This is certainly an interesting study that highlights that arterial phase is of clinical value when assessing PSC patients with contrast enhanced liver MRI, and it is something that should be commented on.
1. Kim WR, Therneau TM, Wiesner RH, et al. (2000) A revised natural history model for primary sclerosing cholangitis. Mayo Clin Proc 75(7):688–694
2. Petrovic BD, Nikolaidis P, Hammond NA, et al. (2007) Correlation between findings on MRCP and gadolinium-enhanced MR of the liver and a survival model for primary sclerosing cholangitis. Dig Dis Sci 52(12):3499–3506
3. Ruiz A, Lemoinne S, Carrat F, et al. (2014) Radiologic course of primary sclerosing cholangitis: assessment by three-dimensional magnetic resonance cholangiography and predictive features of progression. Hepatology 59(1):242–250
Dr. Christopher Tang is a second year radiology registrar at Guy’s and St Thomas’ Hospital, London. He graduated from the University of Oxford before completing his undergraduate medical training with distinction at King’s College London. He currently has a broad range of interests within diagnostic radiology and has presented at international radiology and nuclear medicine conferences. Comments may be sent to Central ESGAR Office at email@example.com
Ureteral stones: Implementation
of a Reduced-Dose CT Protocol in Patients in the Emergency Department with Moderate to High Likelihood of Calculi on the Basis of STONE Score
Christopher L. Moore, MD; Brock Daniels, MD; Dinesh Singh, MD; Seth Luty, MS; Gowthaman Gunabushanam, MD; Monica Ghita, PhD; Annette Molinaro, PhD; Cary P. Gross, MD. Radiology. 2016 Sep; 280(3)
Dr. Raquel Madaleno, Resident, Medical Imaging Department, University Hospital of Coimbra, Portugal
Prof. Dr. Luís Curvo Semedo, Assistant Professor of Radiology, Faculty of Medicine and Medical Imaging Department, University Hospital of Coimbra, Portugal
The main goal of this prospective study by Moore, C. et al (published in last September issue of Radiology) was to determine the sensitivity of reduced-dose CT in the diagnosis of ureteral stones in moderate to high risk patients presenting to the emergency department, requiring urologic intervention.
Renal stones have an incidence of about 0,1-0,4% per year in the United States and in Europe. They usually reoccur, nearly in 75% of the cases during the following 20 years (1).
CT is a method with high diagnostic accuracy in the detection of ureteral stones. The American College of Radiology Appropriateness Criteria emphasizes that, when CT is performed for stone disease, the reduced-dose CT protocols should be preferred, usually yielding an effective dose of about 3mSv. This is of special interest since, frequently, many imaging revaluations are needed (2). However, low-dose protocols are poorly implemented due to the concern of reducing the accuracy in detecting both ureteral stones and alternative diagnoses, which is one of the advantages of CT compared with other imaging techniques (1). In this study the authors used a reduced-dose CT protocol which was the subject of a previous publication, reaching an average radiation dose reduction of 88%.
The patients were stratified based on STONE score in high (90% likelihood of ureteral stone), moderate (50% likelihood of ureteral stone) and low (<10% likelihood of ureteral stone) risk groups. The STONE score uses five criteria (sex, timing, origin, nausea and erythrocytes) that showed to consistently predict the likelihood of ureteral stones in previous studies. Reduced-dose CT was initially performed in 165 patients, including all high risk patients and 73% of patients with moderate risk. After that acquisition, a standard-dose CT was performed in 32% of the patients, when the radiologist determined that image quality was limiting the diagnostic accuracy. This value was slightly higher than expected by the authors, probably due to the radiologist discomfort with low-dose images and the decrease in diagnostic confidence. The reduced-dose CT images from patients who were rescanned with standard-dose CT were blindly reread, resulting in 30 true-positive and 18 true-negative findings for ureteral stone, and five false-negative findings for ureteral stone, without any false-positive findings, resulting in a sensitivity of 86% and a specificity of 100%. The false-negative cases represented five small stones that were expelled without need of intervention. Previous studies have shown that the reduced-dose CT is sensitive enough to detect large stones (> 5 mm) that may require urologic intervention, while it possesses a lower sensitivity in detecting small stones that usually do not require interventional procedures (2).
Of all patients that were studied with reduced-dose CT alone, 25 procedures were necessary, namely 24 urologic procedures (lithotripsy, stent, surgical stone extraction or nephrostomy), and one non-urologic surgery for acute appendicitis.The clinical conditions that required these procedures were all correctly identified with reduced-dose CT, and therefore the study reached a sensitivity of 100% to detect abnormalities that required intervention. However, there was one case of mild diverticulitis identified on rescan with standard-dose CT only, that did not require surgical intervention.
The results of this paper are similar to previous studies that achieved high sensitivity and specificity for diagnosing urolithiasis (3,4).The reduced-dose CT protocols were also evaluated in the management of acute appendicitis, with similar results comparably to standard-dose CT with regard to negative appendectomy rates (5). It would be interesting to see more studies about low-dose CT in the evaluation of the most relevant differential diagnoses for acute flank pain, namely cholecystolithiasis, diverticulitis, pelvic tumours and aortic aneurysms.
In conclusion, this paper shows that reduced-dose CT protocols has a high diagnostic accuracy in the detection of ureteral stones. However, definition of an adequate level of noise that allows both the detection of urolithiasis and alternative diagnoses should be defined in order to allow a widespread implementation, specially in the emergency department.
1. Niemann, T., Kollmann, T., Bongartz, G. Diagnostic Performance of Low-Dose CT for the Detection of Urolithiasis: A Meta-Analysis. American Journal of Roentgenology 2008 191:2, 396-401
2. Drake, T., Jain, N., Bryant, T., Wilson, I., Somani, B. Should low-dose computed tomography kidneys, ureter and bladder be the new investigation of choice in suspected renal colic?: A systematic review. Indian Journal of Urology 2014 Apr-Jun; 30(2): 137–143.
3. Tack, D., Sourtzis, S., Delpierre, I., Maertelaer, V., Gevenois, P. Low-Dose Unenhanced Multidetector CT of Patients with Suspected Renal Colic. American Journal of Roentgenology 2003 180:2, 305-311.
4. Poletti, P., Platon, A., Rutschmann, O., Schmidlin, F., Iselin, C., Becker, C. Low-Dose Versus Standard-Dose CT Protocol in Patients with Clinically Suspected Renal Colic. American Journal of Roentgenology 2007 188:4, 927-933.
5. Kim, K., Kim, Y., Kim, S., Kim, S., Lee, Y., Kim, K., et al. Low-Dose Abdominal CT for Evaluating Suspected Appendicitis. The New England Journal of Medicine 2012; 366:1596-160.
Dr. Raquel Madaleno is a third year resident working at the Medical Imaging Department, University Hospital of Coimbra, Portugal. She completed her undergraduate medical degree at the NOVA Medical School of the New University of Lisbon in 2013. She has her main interest in abdominal radiology, with particular regard to liver and bowel imaging.
Comments may be sent to Central ESGAR Office at firstname.lastname@example.org
Reducing Artifacts during Arterial Phase of Gadoxetate Disodium-enhanced MR Imaging: Dilution Method versus Reduced Injection Rate
Kim YK, Lin WC, Sung K, Raman SS, Margolis D, Lim Y, Gu S, Lu D. Radiology, 2016, Ahead of Print, 10.1148/radiol.2016160241
Dr. Domenico De Santis, "Sapienza" - University of Rome, Department of Radiological Sciences, Oncology and Pathology, Italy
Gadolinium ethoxybenzyl dimeglumine (Gd-EOB-DTPA), labeled as Primovist in Europe and Japan and Eovist in the United States, is a liver-specific magnetic resonance (MR) contrast media characterized by a biliary excretion up to 50% in the normal liver. After it is injected intravenously, Gd-EOB-DTPA distributes throughout the vascular and extravascular spaces during the arterial, portal-venous and delayed phases. The key feature of this contrast media is its ability to be progressively taken up by the hepatocytes via the OATP1B1 and OATP1B3 transporters and excreted into the biliary duct via the MRP2 transporter during the hepatobiliary phase. This phase is typically acquired 20 minutes after the intravenous injection .
These aforementioned features allow Gd-EOB-DTPA to have characteristics of a traditional gadolinium chelate contrast agent during the dynamic phases and to provide specific functional liver information, improving the detection of liver lesions and diffuse liver diseases, specifically liver metastases, hepatocellular carcinomas and liver cirrhosis [2, 3].
An increasing number of institutions are adopting Gd-EOB-DTPA in clinical practice to evaluate patients with known or suspected focal liver disease. However, every radiologist who has performed MR examinations with Gd-EOB-DTPA has experienced the presence of ghosting artifacts during the arterial phase (AP), resulting in suboptimal or non-diagnostic AP. This issue, also documented in a number of publications [4, 5], should not be underestimated since the AP is of paramount importance in the characterization of liver lesions [6, 7].
Different strategies have been adopted to reduce ghosting artifacts, such as diluting the contrast media or performing a slow-rate injection. However, no unanimous consent regarding the optimal injection strategy has been obtained thus far.
The authors of this manuscript performed a retrospective study including 318 consecutive patients (mean age 62.7 years, 205 men) who underwent clinically indicated Gd-EOB-DTPA MRI examination. These patients were placed into two comparable cohorts:
- Dilution cohort ([Gd-EOB-DTPA was diluted 1:1 with saline and injected at a rate of 2.0 mL/sec] n= 159);
- Slow injection cohort ([Gd-EOB-DTPA was not diluted and was injected at a rate of 1.0 mL/sec] n= 159).
The volume of Gd-EOB-DTPA injected was either 10 mL and 20 mL, the latter dose was used in patients who weighed more than 90.7 kg (200 lbs). In order to control potential bias and to assess intra-individual analysis, a sub-cohort of patients requiring imaging follow-up underwent MR examination with both methods. MR images were acquired with either 1.5 or 3 Tesla scanners. A fat-suppressed T1-weighted three-dimensional spoiled gradient echo sequence was performed in breath holding during the AP and the image analysis was subjectively performed by means of a 5-point scale by two independent readers blinded to the injection protocol (Table 1).
In the patient-based analysis, the dilution cohort had a significantly lower mean artifact score compared to the slow injection cohort (1.46 VS 1.95; P <0.001) and was characterized by less frequent severe ghosting artifacts (3.8% VS 15.1%, P <0.001) (Fig 1).
In the image-based analysis, the estimated log odds of mean and severe artifact score would decrease by 0.605 and 1.780, respectively, with the dilution method when compared with that attained with the slow injection method (all P <0.001).
In the sub-cohort of patients who underwent both injection protocols there were significant differences in the mean artifact score changes and frequency of severe artifact according to the order of the two injection methods within each patient (Fig 2).Concerning the relationship of artifacts with contrast media dose, a higher incidence of severe motion artifact was reported when using 20mL of Gd-EOB-DTPA as opposed to 10mL (both injected at a rate of 2 mL/sec). However, in a subsequent analysis of dilution versus slow injection methods for the frequency of severe artifact, the effect of contrast media volume was significant only in the slow injection group, where the severe artifacts were reported in 16.8% of cases with 10mL of Gd-EOB-DTPA and 36% of cases with 20mL of Gd-EOB-DTPA (P =0.030).With this study the authors demonstrated that a 50% contrast agent dilution method outperforms a 50% injection rate reduction in minimizing ghosting artifact on AP images associated with administration of Gd-EOB-DTPA. Therefore, the dilution of Gd-EOB-DTPA with saline represents the best method to reduce the ghost artifact in AP and achieve a diagnostic MR examination. Thus, to the best of our knowedge, this approach is highly advisable in every institution dealing with advanced liver imaging.
1. Leonhardt M, Keiser M, Oswald S, Kuhn J, Jia J, Grube M, et al. Hepatic uptake of the magnetic resonance imaging contrast agent Gd-EOB-DTPA: role of human organic anion transporters. Drug Metab Dispos. 2010;38:1024-8.
2. Reimer P, Schneider G, Schima W. Hepatobiliary contrast agents for contrast-enhanced MRI of the liver: properties, clinical development and applications. Eur Radiol. 2004;14:559-78.
3. Kanki A, Tamada T, Higaki A, Noda Y, Tanimoto D, Sato T, et al. Hepatic parenchymal enhancement at Gd-EOB-DTPA-enhanced MR imaging: correlation with morphological grading of severity in cirrhosis and chronic hepatitis. Magn Reson Imaging. 2012;30:356-60.
4. Davenport MS, Viglianti BL, Al-Hawary MM, Caoili EM, Kaza RK, Liu PS, et al. Comparison of acute transient dyspnea after intravenous administration of gadoxetate disodium and gadobenate dimeglumine: effect on arterial phase image quality. Radiology. 2013;266:452-61.
5. Pietryga JA, Burke LM, Marin D, Jaffe TA, Bashir MR. Respiratory motion artifact affecting hepatic arterial phase imaging with gadoxetate disodium: examination recovery with a multiple arterial phase acquisition. Radiology. 2014;271:426-34.
6. Cruite I, Tang A, Sirlin CB. Imaging-based diagnostic systems for hepatocellular carcinoma. AJR Am J Roentgenol. 2013;201:41-55.
7. Wald C, Russo MW, Heimbach JK, Hussain HK, Pomfret EA, Bruix J. New OPTN/UNOS policy for liver transplant allocation: standardization of liver imaging, diagnosis, classification, and reporting of hepatocellular carcinoma. Radiology. 2013;266:376-82.
Dr. Domenico De Santis is a fourth year Radiology Resident at "Sapienza" University of Rome, Italy. He is currently working as a research fellow at the Department of Radiology, Division of Cardiovascular Imaging at the Medical University of South Carolina, Charleston, USA. His clinical practice in Italy is primarily focused on oncological imaging with special interest in abdominal radiology-specifically liver imaging and virtual colonography-under the mentorship of Prof. Andrea Laghi. He has given scientific presentations at many international radiology meetings and he has his work has been publishing in several radiological journals.
Comments may be sent directly to Dr. De Santis.
Andrew D. Smith, Cody R. Branch, Kevin Zand, Charu Subramony, Haowei Zhang, Katherine Thaggard, Richard Hosch, Jason Bryan, Amit Vasanji, Michael Griswold, Xu Zhang.
Liver Surface Nodularity Quantification from Routine CT Images as a Biomarker for Detection and Evaluation of Cirrhosis. Radiology. 2016 Sep; 280(3):771-81
Cedric Bohyn – Resident, Department of Radiology, AZ Delta Hospital Roeselare & University Hospitals Leuven, Belgium
Philippe Lefere – Department of Radiology, AZ Delta Hospital Roeselare, Belgium
The goal of this retrospective study was to assess the feasibility of distinguishing a cirrhotic from a non-cirrhotic liver on routine CT-images by the use of a semi-automated scoring system that quantifies the liver surface nodularity (LSN).
Performing a liver biopsy remains the gold standard in determining the cirrhotic stage of a liver. The sensitivity and specificity for accurate diagnosis of cirrhosis and its etiology ranges from 80 to 100 percent (1). However, biopsy should only be considered after non-invasive serologic and radiologic evaluation failed to confirm a diagnosis. The reason is that biopsy has many inherent risks like its invasive nature, sampling error, bleeding, infection and possible (but rare) even death. Because of this, non-invasive methods are necessary and they are to be examined whether they can be used for routine assessment of liver cirrhosis.
The most frequently used non-invasive methods to diagnose compensated cirrhosis are serum laboratory tests and ultrasound (US) or MRI-guided elastography, but they have their drawbacks. Therefore, morphometric analyses are also evaluated such as liver volume ratios, splenic volume, or subjective assessment of LSN on US, CT, and MRI. None of these mentioned are proven to be adequate enough to replace the gold standard. Only severe surface nodularity in the setting of chronic liver disease strongly suggests cirrhosis and can warrant a biopsy.
The author (A.D.S.) developed a semi-automated post-processing software able to quantify LSN on CT or MRI-images (Fig. 1). The hypothesis was that with this software LSN could reproducibly be quantified on routine CT-images, allowing accurate diagnosis and staging of the severity of HCV-induced cirrhosis. At the moment of the study, there were no known objective quantitative methods to determine LSN on routine CT imaging.
The study group consisted of 94 patients of which 24 patients with a healthy liver and 70 with a spectrum of hepatitis C chronic liver disease (HCV CLD). HCV is the leading cause of cirrhosis and hepatocellular carcinoma in the US and it affects 200 million people worldwide. Its estimated that 10 – 20% of patients with HCV will develop liver cirrhosis and 1 - 5% will develop HCC within 20-30 years (2).
The authors evaluated the accuracy, reproducibility, and intra- and interobserver variance of this semi-automated LSN-measurement for non-enhanced and enhanced (portal venous phase), and thin and thick-section CT-images for both acquisitions. The comparison was made with splenic volumes, LLS-to-TLV ratios (*), and total liver volume (TLV) for the different stages of CLD. (*) ratio of left lateral segment (LLS) to total liver volume (TLV).
The accuracy of the LSN score for differentiating cirrhotic from non-cirrhotic livers was high and consistently higher compared to splenic volume and LLS/TLV (Fig 2). The sensitivity of LSN was higher than splenic volume and LLS-to-TLV ratios and had an equal specificity compared to LLS-to-TLV ratios.
Furthermore, only LSN scores were higher for F4 decompensated versus compensated cirrhosis, suggesting an increase of LSN in size and heterogeneity with increasing severity of HCV cirrhosis.
This accuracy is comparable to the accuracy of liver stiffness measurements by US or MR elastography. Using the dedicated software, LSN scoring obtained a high reproducibility across different section thicknesses, with or without IV contrast. This may serve as proof that it can be reliably used on different types of CT acquisitions.
Intra- and interobserver agreement was very good with an intraclass correlation coefficient of 0.963 and 0.899, respectively.
LSN quantification was also less time consuming (median total time 1.9 min) than measuring splenic and liver volumes or calculating the LLS-to-TLV ratios (10 – 25 min).
Precirrhotic fibrosis was not investigated in this study. Pickhardt et al. investigated if this LSN scoring tool allows an accurate staging of hepatic fibrosis. They found accurate staging is possible of both significant (≥ F2) and advanced (≥ F3) levels of hepatic fibrosis. These stages are emerging as important clinical indicators that may influence management decisions. In addition, the diagnostic performance shows to be favourable in comparison to US and MRI elastography (3).
A. D. Smith et al. did show in another study of patients with cirrhosis from a variety of causes, that there is a direct linear correlation between the LSN score and the clinical stage of cirrhosis. Moreover, the LSN score proved to be an independent predictor of cirrhosis decompensation and death (4). They suggest LSN could be a complementary tool in combination with the well-known MELD-score (Model End-stage Liver Disease) in evaluating cirrhosis from different causes.
We can conclude that LSN quantifying is a promising tool to evaluate different stages of liver cirrhosis and even precirrhotic fibrosis. The main advantages are its reproducibility, time efficiency and ability to be retrospectively applied on routine CT-images.
1. Heidelbaugh, Joel J., and Michael Bruderly. "Cirrhosis and chronic liver failure: part I. Diagnosis and evaluation." Am Fam Physician 74.5 (2006): 756-62.
2. Sebastiani, Giada, Konstantinos Gkouvatsos, and Kostas Pantopoulos. "Chronic hepatitis C and liver fibrosis." World journal of gastroenterology: WJG 20.32 (2014): 11033.
3. Pickhardt, Perry J., et al. "Accuracy of Liver Surface Nodularity Quantification on MDCT as a Non-invasive Biomarker for Staging Hepatic Fibrosis." American Journal of Roentgenology 207.6 (2016): 1194-1199.
4. Smith, Andrew D., et al. "Liver Surface Nodularity Score Allows Prediction of Cirrhosis Decompensation and Death." Radiology (2016): 160799.
Dr. Cedric Bohyn is a second year resident under the University of Leuven training scheme, currently working at the AZ Delta Hospital in Roeselare (Belgium). He has interest in a wide variety of radiologic topics, yet with some predilection for abdominal radiology. He completed his undergraduate medical degree at the University of Leuven in 2015. Appointed secretary of the Young Radiologist Section (YRS – Belgian resident association as part of the Belgian Society of Radiology), he fully supports and promotes the training of residents. Moreover, he actively participated in organizing the BSR's annual conference of 2016 where he also presented various topics as he did at JFR 2016. As a junior resident, he aspires to continuously learn and strengthen his knowledge on the challenging radiological matter.
Comments may be sent directly to Dr. Cedric Bohyn.
Francois Cornelis, Vlasios Storchios, Elena Violari, Constantinos T. Sofocleous, Heiko Schoder, Jeremy C. Durack, Robert H. Siegelbaum, Majid Maybody, John Humm, and Stephen B. Solomon.18F-FDG PET/CT Is an Immediate Imaging Biomarker of Treatment Success After Liver Metastasis Ablation. J Nucl Med 2016; 57:1052–1057
Myrte de Boer – PhD-student (interventional radiology), The Netherlands Cancer Institute / GROW School for Oncology and Developmental Biology - Maastricht University
Doenja Lambregts – Abdominal Radiologist, The Netherlands Cancer Institute/ Antoni van Leeuwenhoek Hospital
The role of thermal ablation techniques like radio frequency ablation (RFA) and microwave ablation (MWA) is emerging for patients with metastatic colorectal cancer. In the European Society for Medical Oncology (ESMO) consensus guidelines for the management of patients with metastatic colorectal cancer from 2016 it is recommended to consider local ablation techniques in patients with unresectable liver metastases. Moreover, it can be used in addition to surgery to eradicate all visible lesions . Although liver ablation is a potentially curative treatment, up to 30-55% of patients show recurrence in the ablation zone, also known as local tumor progression within 1 year. Re-ablation may still result in complete cure in these patients, but only if the (recurrent) metastases are detected in an early stage so that re-ablation is still technically feasible [2, 3]. There is thus an important role for imaging in the early detection of residual and recurrent disease after liver ablation.
The 2010 Cardiovascular and Interventional Radiological Society of Europa (CIRSE) guidelines for RFA of liver tumours advise contrast-enhanced CT or MRI as the standard modalities to assess treatment outcome . An important limitation of these modalities is the difficulty to distinguish between reactive changes in the ablation zone and viable tumour shortly after the ablation. In areas with benign peri-ablational enhancement, small areas of residual tumour can be easily missed. To overcome these limitations, several studies are ongoing that are examining new modalities and methods to detect residual disease or local tumor progression early after ablation in order to allow for timely re-ablation [2, 5]. The currently reviewed paper is an example of such a study, focusing on PET/CT.
The aim of this retrospective study by Cornelis et al., published in the journal of nuclear medicine in July 2016, was to investigate the value of 18F-FDG PET/CT to predict local treatment failure immediately after percutaneous ablation of liver metastases. Twenty-one patients together underwent percutaneous ablation of a total of 25 metastases (the majority of which from colorectal cancer). Contrast-enhanced CT and FDG-PET/CT were performed immediately after the ablation and scored by two readers (in consensus) who assessed the patterns of enhancement (none, ringlike, or nodular) and FDG uptake (none, indeterminate, obvious residual uptake) around the enhancement zone. In addition the PET/CT was quantitatively analyzed as the ‘tissue radioactivity concentration’ (TRC), which was measured for the whole ablation zone, the ablation margins and normalized using the background liver uptake. In total 11/25 lesions recurred locally within one year.
The results of the study showed that the CT enhancement pattern immediately after ablation was a poor predictor of recurrence with a sensitivity of 27%, specificity of 93%, and overall accuracy of 64%, while the uptake pattern on PET/CT showed much better results with a sensitivity of 100%, specificity 86%, and accuracy of 92%. Moreover, the quantitative PET measurements (in specific the normalized TRC measurements) were found to be 100% accurate in predicting a recurrence. As such, the authors concluded that 18F-FDG PET/CT immediately after ablation is superior to CT and appears to be useful for the early detection of residual disease.
A limitation of the study by Cornelis and colleagues – as also acknowledged by the authors in their paper – is that it is a small retrospective study with a potential patient selection bias because the study population mainly consisted of patients who were referred for ablations in difficult salvage situations. Moreover, there was no histopathological validation to confirm recurrent lesions. Nevertheless, the results of this study are promising and it will be interesting to await results of further prospective studies validating the value of PET/CT – as well as other novel imaging techniques that are currently being investigated such as PET-MRI – in this specific setting.
Suggestions for relevant lecture and poster from the ESGAR e-Education Portal:
1. Van Cutsem, E., et al., ESMO consensus guidelines for the management of patients with metastatic colorectal cancer. Ann Oncol, 2016. 27(8): p. 1386-422.
2. Nielsen, K., et al., The use of PET-MRI in the follow-up after radiofrequency- and microwave ablation of colorectal liver metastases. BMC Med Imaging, 2014. 14: p. 27.
3. Poulou, L.S., et al., FDG-PET for detecting local tumor recurrence of ablated liver metastases: a diagnostic meta-analysis. Biomarkers, 2012. 17(6): p. 532-8.
4. Crocetti, L., T. de Baere, and R. Lencioni, Quality improvement guidelines for radiofrequency ablation of liver tumours. Cardiovasc Intervent Radiol, 2010. 33(1): p. 11-7.
5. Cornelis, F., et al., 18F-FDG PET/CT Is an Immediate Imaging Biomarker of Treatment Success After Liver Metastasis Ablation. J Nucl Med, 2016. 57(7): p. 1052-7.
Myrte de Boer recently graduated as a medical doctor at the University of Amsterdam. During her medical studies she co-authored a paper about critically ill patients and gave a lecture at an international conference about the effect of physical activity on breast cancer. Currently she is working as a PhD student at the (interventional) radiology department of the Netherlands Cancer Institute in Amsterdam (the Netherlands) under the supervision of Prof. Regina Beets-Tan and Dr. Monique Maas. The main focus of her thesis is novel imaging strategies for patients undergoing liver ablation (particularly for colorectal liver metastases). Comments may be sent directly to Ms. Myrte de Boer.
Katrijn L M Michielsen, Ignace Vergote, Raphaëla Dresen, Katya Op de Beeck, Ragna Vanslembrouck, Frédéric Amant, Karin Leunen, Philippe Moerman, Steffen Fieuws, Frederik De Keyzer and Vincent Vandecaveye.
Whole-body diffusion-weighted magnetic resonance imaging in the diagnosis of recurrent ovarian cancer: a clinical feasibility study. Br J Radiol. 2016 Nov;89(1067):20160468.
Dr. Max Lahaye – Abdominal Radiologist, The Netherlands Cancer Institute, Amsterdam/NL
The aim of this paper by Michielsen et al. published in the British Journal of Radiology in November 2016 was to assess the clinical feasibility of whole-body diffusion-weighted MRI (WB-DWI/MRI) for diagnosis and prediction of complete tumour resection in patients with suspected recurrent ovarian cancer.
Complete tumour resection (CTR) is essential to maximize the survival of patients with recurrent ovarian cancer. Whether a CTR is feasible is determined by the amount and the localization of disease in the abdomen. The role of CT and fluorine-18 fludeoxyglucose positron emission tomography (18F-FDG-PET)/CT for the detection of recurrent ovarian cancer has been acknowledged.
However, their value for predicting complete resection remains debatable because usually disease load is underestimated [1-3]. Alternative non-invasive methods to reliably determine the surgical resectability are thus desperately needed.
MRI imaging with diffusion-weighted images (DWI-MRI) might be able to determine whether a complete tumour reduction can be achieved. DWI measures the mobility of water molecules to study tissue cellularity. In tissues with increased cellularity (i.e. tumour) the movement of water molecules in the extracellular tissue space is restricted, resulting in an increase in signal on DWI. DWI is a very powerful technique to detect malignant disease, as has been shown by numerous studies in recent years. Whole-body diffusion-weighted MRI (WB-DWI/MRI) has been reported as a promising diagnostic modality for depicting the amount and localization of disease in ovarian cancer, but also in other gynaecological and gastrointestinal (mainly colorectal ) malignancies. However these studies have often small study populations and include a mix of cancer types (colorectal /ovarian) [4-5].
The paper of Michielsen et al. studied a homogenous group of 51 patients with a clinical suspicion of ovarian cancer recurrence. All patients received a CT of thorax/abdomen and WB-DWI/MRI. In this manner the additional value of WB-DWI/MRI over CT could be evaluated in the same group of patients.
WB-DWI/MRI showed 94% accuracy for detecting ovarian cancer recurrence, compared with 78% for CT (p=0.008). In addition, WB-DWI/MRI correctly predicted complete resection in 33 of 35 (94%) patients eligible for salvage surgery compared with only 17 of 35 (49%) for CT (p=0.001). The better prediction of operability by WB-DWI/MRI over CT was mainly attributable to the higher sensitivity for detection of multifocal or diffuse serosal intestinal disease spread, metastases around the central mesenteric vessels and unresectable distant metastases.
These findings are in line with similar studies on primary ovarian cancer, which also showed a high sensitivity for detecting metastases allowing to predict a complete tumor resection [4,6,7]. The current paper by Michielsen et al. adds to these previous data by demonstrating that WB-DWI/MRI could also play an important role in optimizing treatment planning for recurrent ovarian cancer. By improving patient selection, surgeries can thus be omitted in patients who will not benefit from these extensive surgeries with a relative a high rate of morbidity and even mortality.
It would be very interesting to see whether these good results can be reproduced by other centers and also in other cancer types. In order to select the most optimal treatment for advanced cancer patients, intimate knowledge of the extent of disease is paramount. Consequently, the radiologist plays a vital role in the multidisciplinary team of the oncological patient. Upcoming promising techniques like WB-DWI/MRI must therefore be further investigated.
Suggestions for relevant lecture and poster from the ESGAR e-Education Portal:
ESGAR 2016 - LS 7.2 - Whole body MRI for distant staging - advanced stage ovarian, V. Vandecaveye (Leuven, Belgium)
ESGAR 2015 - EE-143 Peritoneal cancer resections - What the radiologist needs to know, 10.5444/esgar2015/EE-143, E. Aherne, H. Fenlon, C.G. Cronin; Dublin/IE
1. Harter P, Sehouli J, Reuss A, Hasenburg A, Scambia G, Cibula D, et al. Prospective validation study of a predictive score for operability of recurrent ovarian cancer: the Multicenter Intergroup Study DESKTOP II. A project of the AGO Kommission OVAR, AGO Study Group, NOGGO, AGO-Austria, and MITO. Int J Gynecol Cancer 2011; 21: 289–95.
2. Lenhard MS, Burges A, Johnson TR, Stieber P, Kumper C, Ditsch N, et al. PET-CT in recurrent ovarian cancer: impact on treatment planning. Anticancer Res 2008; 28: 2303–8.
3. Dragosavac S, Derchain S, Caserta NM, DE Souza G. Staging recurrent ovarian cancer with (18)FDG PET/CT. Oncol Lett 2013; 5: 593–7.
4. Michielsen, K, Vergote, I, Op de Beeck, K, Amant, F, Leunen, K, Moerman P, et al. Whole-body MRI with diffusion-weighted sequence for staging of patients with suspected ovarian cancer: a clinical feasibility study in comparison to CT and FDG-PET/CT, Eur Radiol 2014 24, 889-901.
5. Soussan M, Des Guetz G, Barrau V, Aflalo-Hazan V, Pop G, Mehanna V, et al. Comparison of FDG-PET/CT and MR with diffusion-weighted imaging for assessing peritoneal carcinomatosis from gastrointestinal malignancy, Eur Radiol 2012 22, 1479-1487.
6. Fujii S, Matsusue E, Kanasaki Y, Kanamori Y, Nakanishi J, Sugihara S, et al. Detection of peritoneal dissemination in gynecological malignancy: evaluation by diffusion-weighted MR imaging. Eur Radiol 2008; 18: 18–23.
7. Espada M, Garcia-Flores JR, Jimenez M, Alvarez-Moreno E, De Haro M, GonzalezCortijo L, et al. Diffusion-weighted magnetic resonance imaging evaluation of intraabdominal sites of implants to predict likelihood of suboptimal cytoreductive surgery in patients with ovarian carcinoma. Eur Radiol 2013; 23: 2636–42.
Dr. Max Lahaye is an abdominal radiologist at The Netherlands Cancer Institute, Amsterdam, The Netherlands. He completed his PhD on the prediction of risk factors for a local recurrence in rectal cancer in 2009 and is still involved in research projects on imaging and treatment of colorectal cancer. He has presented multiple scientific talks and lectures at many conferences and has published in high impact radiology journals such as Radiology. Currently, the main focus of his research is on WB-DWI/MRI for staging colorectal and ovarian cancer.
Comments may be sent directly to Dr. Max Lahaye.
This paper published in Clinical Gastroenterology and Hepatology by Kwong W.T. et all. in June 2016  aims to determine the rate of pancreatic cancer development from neoplastic pancreatic cysts after 5 years of surveillance.
Dr. Dario Picone (University of Palermo/IT)
Prof. Giuseppe Brancatelli (University of Palermo/IT)
Cystic lesions of the pancreas are increasingly recognized because of the improvement of imaging techniques during the last decades. The prevalence of cystic lesions in the pancreas has been estimated up to 3% using computed tomography (CT) and up to 20% using magnetic resonance (MR) imaging [2-4].
The management of asymptomatic pancreatic cysts is a growing challenge. In almost all cases, pancreatic cysts are incidentally discovered during abdominal imaging for unrelated indications [5-7], and most of them are small, asymptomatic and benign. But they have an established malignant potential [8,9] and need a surveillance which results in significant health care costs [10, 11]. The 2015 American Gastroenterological Association (AGA) guidelines recommend discontinuation of surveillance at 5 years in the absence of significant changes on imaging for asymptomatic neoplastic cysts. The AGA high-risk features are: cystic size > 3 cm; dilated main pancreatic duct > 6 mm; mural nodule.
The most common pancreatic cysts are intraductal papillary mucinous neoplasm (IPMN), mucinous cystic neoplasm and serous cystadenomas.
IPMN represents 20% of cystic lesions of the pancreas  and it is more commonly in men with a mean age of occurrence of 65 years . It is classified on the basis of site of origin of the pancreatic ductal system: main-duct IPMN; side-branch IPMN and mixed IPMN . Its imaging findings are correlated with the subtype of tumor. The main-duct IPMN is characterized by a diffuse or segmental dilatation of the main pancreatic duct; the side-branch IPMN is characterized by a macro or microcyst connected with the main pancreatic duct. While the mixed IPMN can have a combination of features and can be encountered in the main duct and side branches. The presence of mural nodules, focal solid components, enhancement of the wall and main pancreatic duct diameter larger than 18 mm are features suggestive of malignant degeneration within main-duct IPMN. While other malignant feature of side-branch IPMN is the size of the cyst more than 3 cm in diameter .
Mucinous cystic neoplasms have malignant potential and are associated with mucin production  and its histopathological characteristic is the presence of ovarian stroma . It represents 10% of cystic lesion of the pancreas, and it occurs almost exclusively in females, with a mean age of 47 years . The presence of nodules, eggshell calcification, increased wall thickness or irregularity, obstruction or displacement of the main pancreatic duct and a diameter greater than 6 cm suggest malignant transformation .
Serous cystoadenoma represents 20% of cystic lesions of the pancreas  and it is more common in female with a mean age of occurrence of 61 years . The typical imaging finding is a microcystic form, like a honeycomb configuration separated by fibrous septa. Classically, it has been thought that malignant transformation from serous cystadenoma is rare, but the tumor diameter and location of the tumor in the pancreatic head are factors that should merit special consideration when considering clinical management.
In this paper, authors analyzed radiographic and clinical data of 310 patients from four medical centers from 2002 to 2010. The median follow-up was 87 months and the median number of imaging studies per patient was 5. In according to the AGA high-risk features, they calculated the risk of pancreatic cancer for patients: 212 patients 0/3 high-risk features; 85 patients 1/3 high-risk features; 13 patients 2/3 high-risk features; 0 patient 3/3 high-risk features.
The rate of malignant transformation on long-term surveillance (>5 years) among those with 0, 1, and 2 high-risk features was 0%, 1%, and 15%, respectively. During the surveillance (>5 years) there were 29 deaths, and of these deaths, only 3 were correlated to pancreas cancer. Authors compared the disease-specific mortality of pancreatic cysts beyond 5 years to the mortality from other causes, and the results demonstrated that the mortality related to non-pancreatic causes was 8 times higher than mortality owing to pancreatic cancer among patients with neoplastic pancreatic cysts after more than 5 years of surveillance.
In conclusion there is a very low risk of malignant transformation of asymptomatic neoplastic pancreatic cysts after 5 years.
Overall the authors conclude that the discontinuation of surveillance can be considered for patients with pancreatic lesions, and 0 or 1 high-risk feature have a less than 1% risk of developing pancreatic cancer; while surgery or continued surveillance should be considered for patients with neoplastic pancreatic cysts with 2 high-risk features because they have a 15% risk of developing pancreatic cancer.
1. Kwong W .T., Hunt G.C., Fehmi S.E. et all. Low Rates of Malignancy and Mortality in Asymptomatic Patients With Suspected Neoplastic Pancreatic Cysts Beyond 5 Years of Surveillance. Clinical Gastroenterology and Hepatology 2016;14:865–871. doi: 10.1016/j.cgh.2015.11.013. Epub 2015 Dec 2.
2. Del Chiaro M., Verbeke C., Salvia R. the European Study Group on Cystic Tumours of the Pancreas et all . European experts consensus statement on cystic tumours of the pancreas. Digestive and Liver Disease 45 (2013) 703–711.
3. Ip IK, Mortele KJ, Prevedello LM, et al. Focal cystic pancreatic lesions: assessing variation in radiologists management recommendations. Radiology 2011;259:136–41.
4. Laffan TA, Horton KM, Klein AP, et al. Prevalence of unsuspected pancreatic cysts on MDCT. American Journal of Roentgenology 2008;191:802–7.
5. ZhangXM,MitchellDG,DohkeM,etal.Pancreaticcysts:depictiononsingle- shot fast spin-echo MR images. Radiology 2002;223:547–53.
6. Olga R. Brook, MD Peter Beddy, MD Jay Pahade, MD Corey Couto, MD Ian Brennan, MD Payal Patel, MD Alexander Brook, PhD Ivan Pedrosa, MD. Delayed growth in incidental Pancreatic cysts: Are the Current American College of Radiology Recommendations for Follow-up Appropriate?. Radiology 2016; 278:752–761.
7. Laffan TA, Horton KM, Klein AP, et al. Prev- alence of unsuspected pancreatic cysts on MDCT. AJR Am J Roentgenol 2008;191(3): 802–807.
8. Lee KS, Sekhar A, Rofsky NM, Pedrosa I. Prevalence of incidental pancreatic cysts in the adult population on MR imaging. Am J Gastroenterol 2010;105(9):2079–2084.
9. Lee SH, Shin CM, Park JK, et al. Out- comes of cystic lesions in the pancreas after extended follow-up. Dig Dis Sci 2007;52(10):2653–2659.
10. Lahat G, Lubezky N, Haim MB, et al. Cystic tumors of the pancreas: high malignant po- tential. Isr Med Assoc J 2011;13(5):284–289.
11. Tanaka M, Chari S, Adsay V, et al. Interna- tional consensus guidelines for management of intraductal papillary mucinous neoplasms and mucinous cystic neoplasms of the pan- creas. Pancreatology 2006;6(1-2):17–32.
12. Adsay NV. Cystic neoplasia of the pancreas: pathol- ogy and biology. J Gastrointest Surg 2008;12(3): 401–404.
13. Procacci C, Megibow AJ, Carbognin G, et al. Intraductal papillary mucinous tumor of the pan- creas: a pictorial essay. RadioGraphics 1999;19(6): 1447–1463.
14. Manfredi R, Graziani R, Motton M, et al. Main pancreatic duct intraductal papillary mucinous neo- plasms: accuracy of MR imaging in differentiation between benign and malignant tumors compared with histopathologic analysis. Radiology 2009;253 (1):106–115.
15. Campbell F, Azadeh B. Cystic neoplasms of the exocrine pancreas. Histopathology 2008;52(5): 539–551.
16. Al-Haddad M, Schmidt MC, Sandrasegaran K, De- witt J. Diagnosis and treatment of cystic pancreatic tumors. Clin Gastroenterol Hepatol 2011;9(8): 635–648.
17. Compagno J, Oertel JE. Mucinous cystic neoplasms of the pancreas with overt and latent malignancy (cystadenocarcinoma and cystadenoma): a clinico- pathologic study of 41 cases. Am J Clin Pathol 1978; 69(6):573–580.
18. Goh BK, Tan YM, Chung YF, et al. A review of mu- cinous cystic neoplasms of the pancreas de ned by ovarian-type stroma: clinicopathological features of 344 patients. World J Surg 2006;30(12):2236–2245.
19. Crippa S, Salvia R, Warshaw AL, et al. Mucinous cystic neoplasm of the pancreas is not an aggressive entity: lessons from 163 resected patients. Ann Surg 2008;247(4):571–579.
Dr. Dario Picone is a final year resident in radiology at the University of Palermo. His main interest is abdominal radiology, with particular regard to liver and bowel imaging. He has spent part of his training at the Radiology Service of the University College London Hospital and at the Radiology Service of the Beaujon Hospital in Clichy where he focused on prostate cancer imaging and liver imaging respectively.
Comments may be sent directly to Dr. Dario Picone.
European Association for the Study of the Liver (EASL). EASL Clinical Practice Guidelines on the management of benign liver tumours. J Hepatol. 2016 Apr 11. [Epub ahead of print] doi: 10.1016/j.jhep.2016.04.001
Dr. Federica Vernuccio (University of Palermo/IT)
Prof. Giuseppe Brancatelli (University of Palermo/IT)
With this paper, the aim of the European Association for the Study of the Liver (EASL) was to critically review the existing literature on hepatic hemangioma, focal nodular hyperplasia (FNH) and hepatocellular adenoma (HCA), providing official practical recommendations for diagnosis and management of benign focal liver lesions (BLLs).
Hepatic hemangiomas, which are the most common primary liver tumors , may be diagnosed at US if typical findings occur (less of 3 cm in diameter, homogeneously hyperechoic, with sharp margins, posterior enhancement and no halo sign) . In case of oncologic background, underlying liver diseases or atypical features at US, contrast enhanced imaging, possibly through MR, is required for further characterization: it will show a peripheral globular progressive enhancement, with complete filling on delayed phases in cavernous hemangiomas . Using gadoxetic acid on MR, hemangioma shows the so called pseudo wash-out during the transitional and hepatobiliary phases. When imaging is inconclusive, percutaneous biopsy is indicated .
No clinical advices, follow-up or therapy is indicated for hemangiomas except in case of unusual evolutions or complications such as Kasabach-Merrit syndrome, growing lesions or symptomatic lesions due to compression of surrounding organs.
The differential diagnosis of FNH and HCA is mandatory due to their different management. Both these lesions usually occur in young females that are often under oral contraceptives, and can be single or multiple. However, FNH is more common , usually asymptomatic and does not imply any laboratory test abnormality or require any lifestyle change, follow-up, unless there is underlying vascular liver disease, or treatment. On the other hand, HCA may be symptomatic, determine elevation of alcaline phosphatase and transaminase, and patients’ management requires discontinuation of oral contraceptive and weight loss, follow-up  (for women with HCA smaller than 5 cm, an MR examination after 6 months of lifestyle changes, and thereafter at 1 year and annual imaging; in pregnant women, US every 6-12 weeks is indicated) and therapy . Specifically, resection is indicated for HCA in men, for nodules more than 5 cm in diameter or increasing in size more than 20% and in case of beta-catenin mutation because of the higher risk in these cases of potential complications such as hemorrhage and malignant transformation; embolization for bleeding HCA with haemodynamic instability.
For these different approaches, the role of the radiologist is prominent in detection and characterization of these lesions and in the differential diagnosis. Both FNH and HCA do not share unique features on conventional US, and require contrast enhanced imaging (CEUS is preferred for lesions smaller than 3,5 cm , whereas MR is preferred in larger lesions). The main imaging features of FNH at imaging are: lesion homogeneity except the central scar; absence of a capsule; lobulated contours; central scar; slight difference compared to the surrounding liver parenchyma in precontrast imaging; strong and homogeneous enhancement in arterial phase, that becomes similar to that of the adjacent liver parenchyma during portal-venous and delayed phases . Atypias as the presence of steatosis, pseudocapsule and wash out may be encountered. Thus, in a small proportion of patients imaging is inconclusive and liver biopsy is indicated .
The imaging features of HCAs are more variable because HCAs encompass four main different molecular subtypes, as shown in table 2 :
-HCA with inactivation of HNF-1alfa that is responsible for the absence of expression of the liver fatty acid binding protein, resulting in extensive steatosis at pathology and diffuse and homogeneous signal dropout on opposed-phase in chemical shift T1 sequences, moderate enhancement on arterial phase often with washout on portal and/or delayed phases at imaging-
-Inflammatory HCAs, showing inflammatory infiltration and sinusoidal dilatation at pathology and telangectatic features at imaging, that enclose strong hyperintensity on T2-weighted sequences (diffuse or as a rim-like band in the periphery, known as atoll sign), hypervascularity on arterial phase with persistent enhancement on delayed phase. Some atypias may be encountered, as the presence of fat within the lesion, although the signal drop out on opposed phase is heterogenous and moderate.
-HCA with activation of the beta catenin that is responsible of cellular atypias, pseudoglandular formations and cholestasis at pathology.
-Unclassified HCAs, with no specific imaging features.
In conclusion, the EASL guidelines reported in this article provide practical clinical, diagnostic and therapeutical keys for the management of benign liver lesions, that should be known by radiologists in their everyday practice.
1. Kaltenbach TE, Engler P, Kratzer W, Oeztuerk S, Seufferlein T, Haenle MM, Graeter T. Prevalence of benign focal liver lesions: ultrasound investigation of 45,319 hospital patients. Abdom Radiol (NY). 2016;41:25-32.
2. Quaia, E., Bertolotto, M. & Dalla Palma, L. Characterization of liver hemangiomas with pulse inversion harmonic imaging. Eur. Radiol. 2002;12: 537–544
3. Caldironi MW, Mazzucco M, Aldinio MT, Paccagnella D, Zani S, Pontini F, et al. Echo-guided fine-needle biopsy for the diagnosis of hepatic angioma. A report on 114 cases. Minerva Chir 1998;53:505–509
4. Chun YS, Parker RJ, Inampudi S, Ehrenwald E, Batts KP, Burgart LJ, Schumacher CW, Mehling JA, Engstrom BI, Hill MJ, Reddy SK, Sielaff TD. Imaging Surveillance of Hypervascular Liver Lesions in Non-Cirrhotic Patients. J Gastrointest Surg. 2016;20:564-7. doi: 10.1007/s11605-015-2942-9
5. Roche V, Pigneur F, Tselikas L, Roux M, Baranes L, Djabbari M, et al. Differentiation of focal nodular hyperplasia from hepatocellular adenomas with low-mechanical-index contrast-enhanced sonography (CEUS): effect of size on diagnostic confidence. Eur Radiol 2015;25:186–195
6. Buetow PC, Pantongrag-Brown L, Buck JL, Ros PR, Goodman ZD. Focal nodular hyperplasia of the liver: radiologic-pathologic correlation. Radiographics. 1996;16:369-88.
7. Katabathina VS, Menias CO, Shanbhogue AK, Jagirdar J, Paspulati RM, Prasad SR. Genetics and imaging of hepatocellular adenomas: 2011 update. Radiographics. 2011;31:1529-43.
Additional educational material
EuroRad case Nr. 5528 Hepatic Adenoma (HA) and Focal Nodular Hyperplasia (FNH) Grazioli L
Eurorad case Nr 2814 Rapidly filling hemangioma in a cirrhotic patient. Iannaccone R, Marin D, Celestre M, Guerrisi A
Dr. Federica Vernuccio is a third year resident in radiology at the University of Palermo. She has already developed interest toward abdominal imaging, and liver specifically. She has spent 6-months at the Radiology Service of the Beaujon Hospital in Clichy where she is conducting clinical research on hepatocellular adenomas under the mentorship of Prof. Vilgrain.
Comments may be sent directly to Federica Vernuccio.
D. Tsurumaru, M. Miyasaka, Y. Nishimuta et al. Differentiation of early gastric cancer with ulceration and resectable advanced gastric cancer using multiphasic dynamic multidetector CT. Eur Radiol. 2016; 26: 1330-1337.
Dr. Ruolei Chen (Radiology Registrar, Guy’s and St Thomas’ Hospitals, London, GB)
Dr. Sofia Gourtsoyianni (Consultant Radiologist, Guy’s and St Thomas’ Hospitals, London, GB)
This paper by D. Tsurumaru and colleagues was published in the European Radiology Journal in May 2016. The objectives were to (a) determine whether multiphasic dynamic multidetector CT (MDCT) can help differentiate early gastric cancer with ulceration (EGC-U) from advanced gastric cancer (AGC) and (b) to evaluate the diagnostic performance on differentiating these tumours.
Accurate differentiation of EGC-U from AGC is crucial: the deeper depth of tumour invasion of gastric cancer is associated with poorer survival rates , whilst treatment options for these are different. On endoscopy, EGC-U and AGC have similar appearances .
A retrospective study was performed for the period January 2006 to December 2012. A total of 16 patients with EGC-U with the ulcer stage of Ul-III or IV and 24 patients with AGC with the tumour stage of T2 to T4a were analysed. All these patients had pathologically confirmed gastric adenocarcinoma after gastrectomy, with the tumour being visible on both preoperative gastroscopy and contrast-enhanced MDCT.
A very meticulous imaging protocol has been adopted/proposed . Prior to CT imaging and after an overnight fast, each patient ingested an effervescent agent with water and had an injection of intramuscular Buscopan to suppress peristalsis. Images were obtained at 40s (arterial phase), 70s (portal phase) and 240s (delayed phase) after an infusion of intravenous contrast. Two radiologists independently reviewed the CT images on a workstation using virtual endoscopy. The lesion size (largest dimension using multiplanar reconstruction) and its attenuation value on each phase of contrast were recorded. Two endoscopists interpreted the endoscopic images independently and had to diagnose the lesion as EGC-U or AGC according to the Japanese Classification of Gastric Carcinoma .
The authors found that the mean lesion size was 35 mm for the EGC-Us and significantly larger for the AGCs at 45 mm (p=0.033). On MDCT analysis, the mean attenuation values of EGC-U cases were significantly lower than those of the AGC cases in both the arterial and portal phases (both p<0.0001). Delayed phase attenuation values were equivocal between the two groups (Table 1). Peak enhancement for EGC-U was in the delayed phase and for AGC was in the arterial phase.
Different cut-off values were used for each reader to assess diagnostic performance of arterial, portal venous and peak enhancement value, respectively.
An explanation for the increased enhancement demonstrated by AGC compared to EGC-U in the arterial and portal phases is related to gastric cancers showing neovascularity during the arterial to capillary phase in vivo [5, 6]. The peak enhancement of EGC-U in the delayed phase in this study may be due to EGC-U containing varying degrees of fibrous tissue associated with ulceration, similar to benign gastric ulcers.
It is interesting to note that this study found the endoscopic evaluation to have lower accuracy values (70.0 % and 72.5 %) in identifying EGC-U and AGC compared to portal venous phase MDCT (92.5 % and 90.0 %). However, this may be partially attributable to the endoscopists being only able to view selected images from the initial gastroscopy.
The authors note several study limitations. First, it is a retrospective study with a relatively small patient population. Second, their CT protocol did not include an unenhanced CT phase with possible greater accuracy for gastric cancer detection being achieved if this was available. Third, all the cases in their series were resectable gastric cancers and it is unclear whether the results would be applicable to non-resectable gastric cancers. However, even with these potential limitations, their findings suggest that the attenuation value measurements help differentiate EGC-U from AGC.
In conclusion, the finding that EGC-U had significantly lower attenuation values compared to the AGC in both the arterial and portal phases may be useful as a diagnostic feature for differentiating these two on MDCT, which in conjunction with gastroscopy, can improve diagnostic accuracy rates and patient outcome.
1. Kim JP, Lee JH, Kim SJ, Yu HJ, Yang HK. Clinicopathologic characteristics and prognostic factors in 10,783 patients with gastric cancer. Gastric Cancer. 1998; 1: 125-133.
2. Kitamura K, Yamaguchi T, Nishida S, Yamamoto K, Okamoto K, Taniguchi H et al. Early gastric cancer mimicking advanced gastric cancer. Br J Cancer. 1997; 75: 1769-1773.
3. Komori M, Asayama Y, Fujita N, Hiraka K, Tsurumaru D, Kakeji Y, Honda H. Extent of arterial tumor enhancement measured with preoperative MDCT gastrography is a prognostic factor in advanced gastric cancer after curative resection. AJR. 2013; 201(2): 253-61.
4. Hwang SW, Lee DH, Lee SH, Park YS, Hwang JH, Kim JW et al. Preoperative staging of gastric cancer by endoscopic ultrasonography and multidetector-row computed tomography. J Gastroenterol Hepatol 2010; 25: 512-518.
5. Efsen F, Fischerman K. Angiography in gastric tumours. Acta Radiol Diagn. 1974; 15: 193-197.
6. Shibata S, Iwasaki N. Angiographic findings in diseases of the stomach. Am J Roentgenol Radium Ther Nucl Med. 1970; 110: 322-331.
Additional educational material:
Proposed e-poster (ESGAR 2015)EE-038 - Gastrointestinal Wall Lesions : Identified on Routine Transabdominal Ultrasonography.
EuroRad case Nr. 3146
Belo-Oliveira Pedro, Rodrigues Henrique, Belo-Soares Pedro (2005, Jul 13). Polypoid gastric cancer
Comments may be sent directly to Dr. Roulei Chen.
Nougaret S, Vargas HA, Lakhman Y, Sudre R, Do RK, Bibeau F, Azria D, Assenat E, Molinari N, Pierredon MA, Rouanet P, Guiu B. Intravoxel Incoherent Motion-derived Histogram Metrics for Assessment of Response after Combined Chemotherapy and Radiation Therapy in Rectal Cancer: Initial Experience and Comparison between Single-Section and Volumetric Analyses. Radiology 2016 Feb 26 [ePub ahead of print]. DOI: dx.doi.org/10.1148/radiol.2016150702
Dr. Andrea Delli Pizzi – Radiology Resident, Gabriele d’Annunzio University, Chieti/IT (currently at The Netherlands Cancer Institute for research internship)
Prof. Regina Beets-Tan – Radiologist (head of department of Radiology), the Netherlands Cancer Institute, Amsterdam/NL
In this paper published in Radiology in February 2016, Nougaret et al. assessed the diagnostic performance of intravoxel incoherent motion (IVIM) parameters and apparent diffusion coefficient (ADC) to assess response to chemoradiotherapy (CRT) in rectal cancer by means of histogram analysis.
Functional imaging for rectal cancer response evaluation is a hotly debated topic and in particular the use of diffusion-weighted imaging has received much attention over the last years. The growing interest for rectal tumor response evaluation is probably related to the upcoming concept of organ-preserving treatment for good or complete responders after CRT. In order to select patients for organ-preservation (i.e. local excision or ‘watchful waiting’) an accurate assessment of treatment response becomes crucial.
Apart from visual assessment of diffusion images, many studies have focused on quantitatively analyzing tumor diffusion by measuring the tumor ‘Apparent Diffusion Coefficient’ (ADC). However, reports so far have shown inconsistent results and several studies found no clear benefit in measuring tumor ADC values to assess treatment response.
Nougaret and colleagues postulate that these conflicting results may be attributed to technical differences between DWI sequences used in literature, but also to failure to distinguish perfusion effects from true diffusion with the typically applied mono-exponential fitting used to calculate ADCs as well as the inability to capture tumor heterogeneity when evaluating only mean ADC values. In order to overcome these issues the authors applied extensive b-value sampling (34 b values) and a bi-exponential curve fit analysis with the IVIM model combined with histogram analysis to study tumor heterogeneity. IVIM is a relatively novel approach that uses a bi-exponential function to analyze DWI-data. It takes into account two components of the diffusion signal decay curve, which are related to tissue perfusion and tissue diffusivity.1 This approach was recently applied in abdominal imaging to investigate other tumors such as hepatocellular carcinoma and pancreatic cancer.2,3 Currently the paper by Nougaret et al. is one of the few concerning IVIM application in rectal cancer and it is so far the only one assessing its diagnostic performance in treatment response evaluation. The study included 31 patients with rectal cancer who underwent MRI including DWI before and after chemoradiotherapy. ADC, perfusion-related diffusion fraction (f), slow diffusion coefficient (D), and fast diffusion coefficient (D*) were calculated and compared with histopathologic findings. For each parameter, histogram analysis was performed. The results of the study showed that none of the ADC and IVIM metrics could predict response prior to treatment. After CRT, IVIM-derived diffusion coefficient D as well as ADC increased significantly and their respective post-treatment values were significantly higher in good versus poor responders. Diagnostic performance of post-CRT D and ADC histogram metrics to predict a good response are shown in the table 3. Best results were obtained for median D values post-CRT (area under the curve of 0.98, sensitivity 94%, specificity 100%). Histogram metrics did not yield better results than median values for assessment of response. The authors therefore conclude that histogram analyses may not be required in everyday clinical practice.
The authors also found no benefit for the IVIM-derived f and D* parameters in assessing response. This is in line with previous reports that focused on IVIM for the assessment of response in liver tumors.3-5 Moreover in a study published in 2013, Bäuerle et al. used an IVIM approach to analyze two groups of rectal carcinoma patients (with and without chemoradiotherapy).6 In the group without CRT they showed a correlation between f and the postoperative histological vascular area fraction, but no significant correlation was found in the group after CRT. A reasonable explanation for these incongruent results is that D* value is affected by high uncertainty and poor reproducibility and f value is widely influenced by T2 relaxation times of blood and tissues. Hence, as described by Lemke et al., authors underline that a specific T2 correction may be essential for a correct interpretation of results.7 Moreover it would be interesting to see how IVIM-derived perfusion related parameters correlate with perfusion parameters derived from dynamic contrast enhanced (DCE) -MRI acquisitions. A recent study by Hötker et al. showed that both DWI-MRI and DCE-MRI parameters were significantly associated with the degree of tumor regression in the resected specimen.8 It would be interesting to see how IVIM-derived perfusion related parameters may fit into this picture. Further studies in larger patient groups are required to establish the correlation between DCE perfusion parameters, ADC and IVIM metrics and to draw definite conclusions regarding their potential (complementary) value in predicting response.
1. Koh DM, Collins DJ, Orton MR. Intravoxel incoherent motion in body diffusion-weighted MRI: reality and challenges. AJR Am J Roentgenol 201;196(6):1351–1361.
2. De Robertis R, Tinazzi Martini P, Demozzi E, Dal Corso F, Bassi C, Pederzoli P, D'Onofrio M. Diffusion weighted imaging of pancreatic cancer. World J Radiol. 2015 Oct 28;7(10):319-28.
3. Woo S, Lee JM, Yoon JH, Joo I, Han JK, Choi BI. Intravoxel incoherent motion diffusion-weighted MR imaging of hepatocellular carcinoma: correlation with enhancement degree and histologic grade. Radiology 2014;270(3):758–767.
4. Andreou A, Koh DM, Collins DJ, et al. Measurement reproducibility of perfusion fraction and pseudodiffusion coefficient derived by intravoxel incoherent motion diffusion-weighted MR imaging in normal liver and metastases. Eur Radiol 2013;23(2):428–434.
Additional educational material:
ESGAR 2015, Recommended Poster: Various kinds of neoplastic and non-neoplastic lesions in the rectal MRI encountered in the daily practice, J.-H. Yoon, S. Lee, S.H. Kim, Y. Lee, Y.-J. Lim, J.W. Ryu, H.-D. Kim; Busan/KR
ESGAR 2015, WS 26 - Rectal cancer: assessment of treatment respons, L. Curvo-Semedo (Coimbra, Portugal)
EURORAD Case Nr. Case 12099 - Ovarian metastases from colorectal cancer, Serpa, Sara; Silva, David; Amaral, Rui; Simões, Marta; Fernandes, Otília (2014, Sep 22)
Comments may be sent directly to Dr. Andrea Pizzi.
Huh J, Kim SY, Yeh BM, Lee SS, Kim KW, Wu EH, Wang ZJ, Zhao LQ, Chang WC.
Troubleshooting Arterial-Phase MR Images of Gadoxetate Disodium-Enhanced Liver
Korean J Radiol. 2015 Nov-Dec;16(6):1207-15. doi:10.3348/kjr.2015.16.6.1207. Epub 2015 Oct 26
Dr. Daniel Ramos Andrade, Resident, Medical Imaging Department, University Hospital of Coimbra, Portugal
Prof. Dr. Luís Curvo Semedo, Assistant Professor of Radiology, Faculty of Medicine and Medical Imaging Department, University Hospital of Coimbra, Portugal
Gadoxetate disodium is widely used nowadays for characterization of focal liver lesions, as well as for detection of liver metastases and hepatocellular carcinoma in cirrhotic patients, because of its combined properties of a conventional non-specific extracellular and a hepatocyte-specific contrast agent .
Unfortunately, arterial-phase images are sometimes of very poor quality due to artifacts and weak arterial enhancement.
The purpose of this review article, published by the Korean Journal of Radiology in late 2015 by Huh et al. was to identify these two challenges and to suggest possible solutions for overcoming them.
A review of the literature on this subject is very confounding because some authors believe degradation of arterial phase image quality is due to Gibbs phenomenon / ringing / truncation artifacts, related to rapid changes in the contrast concentration during the obtainment of the data in the center of the k-space [2,3,4], while others postulate that transient severe motion / respiratory motion artifact / acute transient dyspnoea is the main culprit [5,6]. It can be very difficult to distinguish both artifacts. Ringing artifacts usually manifest as duplicates of strongly enhanced vessels that are centered on the vessels and fade further away from the vessels, being typically confined to the abdomen, whereas motion artifacts appear as duplicates of all the abdominal organs and abdominal wall and typically extend both anteriorly and posteriorly to the abdomen .
This article ultimately concludes that transient motion is the main source of artifacts in the clinical practice and not the ringing artifacts. After this review was published, another study by Motosugi et al further added that severe artifacts in the arterial phase were mainly associated with breath-hold failure (28 of 33 patients with severe artifacts) .
It is believed that transient severe motion artifacts occur in about 10.7-18% of patients but the pathophysiology and causes remain elusive . While the risk factors are yet to be identified, this article states that we can attempt to decrease these artifacts by using a single-breath-hold multiple arterial-phase acquisition  or using motion-insensitive sequences like CAIPIRINHA VIBE and radial-VIBE.
 Campos JT, Sirlin CB, Choi JY. Focal hepatic lesions in Gd-EOB-DTPA enhanced MRI: the atlas. Insights Imaging. 2012 Oct;3(5):451-74. doi: 10.1007/s13244-012-0179-7. Epub 2012 Jun 15.
 Tanimoto A, Higuchi N, Ueno A. Reduction of ringing artifacts in the arterial phase of gadoxetic acid-enhanced dynamic MR imaging. Magn Reson Med Sci. 2012;11(2):91-7.
 Motosugi U, Ichikawa T, Sou H, Sano K, Ichikawa S, Tominaga L, Araki T. Dilution method of gadolinium ethoxybenzyl diethylenetriaminepentaacetic acid (Gd-EOB-DTPA)-enhanced magnetic resonance imaging (MRI). J Magn Reson Imaging. 2009 Oct;30(4):849-54. doi: 10.1002/jmri.21913.
 Haradome H, Grazioli L, Tsunoo M, Tinti R, Frittoli B, Gambarini S, Morone M, Motosugi U, Colagrande S. Can MR fluoroscopic triggering technique and slow rate injection provide appropriate arterial phase images with reducing artifacts on gadoxetic acid-DTPA (Gd-EOB-DTPA)-enhanced hepatic MR imaging? J Magn Reson Imaging. 2010 Aug;32(2):334-40. doi: 10.1002/jmri.22241.
 Davenport MS, Viglianti BL, Al-Hawary MM, Caoili EM, Kaza RK, Liu PS, Maturen KE, Chenevert TL, Hussain HK. Comparison of acute transient dyspnea after intravenous administration of gadoxetate disodium and gadobenate dimeglumine: effect on arterial phase image quality. Radiology. 2013 Feb;266(2):452-61. doi: 10.1148/radiol.12120826. Epub 2012 Nov 28.
 Ringe KI, Husarik DB, Sirlin CB, Merkle EM. Gadoxetate disodium-enhanced MRI of the liver: part 1, protocol optimization and lesion appearance in the noncirrhotic liver. AJR Am J Roentgenol. 2010 Jul;195(1):13-28. doi: 10.2214/AJR.10.4392.
 Bashir MR, Castelli P, Davenport MS, Larson D, Marin D, Hussain HK, Jaffe TA. Respiratory motion artifact affecting hepatic arterial phase MR imaging with gadoxetate disodium is more common in patients with a prior episode of arterial phase motion associated with gadoxetate disodium. Radiology. 2015 Jan;274(1):141-8. doi: 10.1148/radiol.14140386. Epub 2014 Aug 25.
 Motosugi U, Bannas P, Bookwalter CA, Sano K, Reeder SB. An Investigation of Transient Severe Motion Related to Gadoxetic Acid-enhanced MR Imaging. Radiology. 2015 Oct 16:150642. [Epub ahead of print]
 Kim SY, Park SH, Wu EH, Wang ZJ, Hope TA, Chang WC, Yeh BM. Transient respiratory motion artifact during arterial phase MRI with gadoxetate disodium: risk factor analyses. AJR Am J Roentgenol. 2015 Jun;204(6):1220-7. doi: 10.2214/AJR.14.13677.
 Pietryga JA, Burke LM, Marin D, Jaffe TA, Bashir MR. Respiratory motion artifact affecting hepatic arterial phase imaging with gadoxetate disodium: examination recovery with a multiple arterial phase acquisition. Radiology. 2014 May;271(2):426-34. doi: 10.1148/radiol.13131988. Epub 2014 Jan 21.
 Tamada T, Ito K, Sone T, Yamamoto A, Yoshida K, Kakuba K, Tanimoto D, Higashi H, Yamashita T. Dynamic contrast-enhanced magnetic resonance imaging of abdominal solid organ and major vessel: comparison of enhancement effect between Gd-EOB-DTPA and Gd-DTPA. J Magn Reson Imaging. 2009 Mar;29(3):636-40. doi: 10.1002/jmri.21689.
 Tamada T, Ito K, Yoshida K, Kanki A, Higaki A, Tanimoto D, Higashi H. Comparison of three different injection methods for arterial phase of Gd-EOB-DTPA enhanced MR imaging of the liver. Eur J Radiol. 2011 Dec;80(3):e284-8. doi: 10.1016/j.ejrad.2010.12.082. Epub 2011 Feb 5.
 Boll DT, Merkle EM. Imaging at higher magnetic fields: 3 T versus 1.5 T. Magn Reson Imaging Clin N Am. 2010 Aug;18(3):549-64, xi-xii. doi: 10.1016/j.mric.2010.08.008.
 An C, Park MS, Kim D, Kim YE, Chung WS, Rhee H, Kim MJ, Kim KW. Added value of subtraction imaging in detecting arterial enhancement in small (<3 cm) hepatic nodules on dynamic contrast-enhanced MRI in patients at high risk of hepatocellular carcinoma. Eur Radiol. 2013 Apr;23(4):924-30. doi: 10.1007/s00330-012-2685-x. Epub 2012 Nov 9.
In addition to these artifacts, arterial phase MR images with Gadoxetate Disodium are also suboptimal due to weak arterial enhancement . The standard injection dose and the content of gadolinium for this hepatocyte-specific contrast is half and a quarter, respectively, of the conventional extracellular agent dose. This shortens the enhancement duration of the contrast bolus and lowers the magnitude of the peak enhancement compared with extracellular contrast agents. This article reviewed three solutions for this problem, namely:
1- Decreasing the injection rate to 1ml/sec. By slowing the injection rate, the bolus is stretched and the enhancement peak is higher [4,12]
2- Increasing the amount of Gadoxetate Disodium. Increasing the amount of the contrast, as it is done in several centers with a fixed dose of 10mL, also prolongs the peak arterial perfusion time 
3- Diluting the Gadoxetate Disodium with saline. Dilution of Gadoxetate Disodium with saline has also been proposed as an option to stretch the contrast bolus, but the evidence is still very limited and there is a risk of contamination [3,12].
From the technical standpoint, the article also recommends four modalities:
1- A bolus detection method. It helps to achieve a precise timing for acquisition of arterial images. 
2- Multiple arterial phase image acquisition with a fixed delay. The best images can be chosen and the bad ones discarded. This option also helps avoiding transient severe motion. 
3- 3T imaging. Theoretically, 3T provides a two-fold increase in signal gain in signal-to-noise ratio and better lesion-to-liver contrast, although Gadoxetate Disodium enhanced liver MRI studies comparing 3T with 1,5T are lacking. 
4- Subtraction images. This technique is more beneficial for depicting arterial enhancement than visual comparison only. 
All of the challenges, causes and possible solutions for obtaining high-quality arterial-phase MR images with Gadoxetate Disodium are summarized in the table below.
In conclusion, this article is probably the most important review yet published regarding possible solutions for the poor image quality of arterial-phase MR images of Gadoxetate Disodium-enhanced examinations. Nevertheless, additional large studies are necessary so that, once and for all, a standard protocol for Gadoxetate-disodium administration can be adopted.
Comments may be sent directly to Dr. Daniel Ramos Andrade.
Additional educational material:
RANZCR-AOCR 2012 Poster No.: R-0206
Y. X. Kitzing, S. McCormack, B. Ng. Tips and pitfalls of liver imaging with gadoxetic acid: a pictorial review
ECR 2012 Poster No.: C-0824
S. Suzuki, T. Kobayashi, M. Ozaki, H. Evaluation of simultaneous saline injection methods in Gd-EOB-DTPA-enhanced magnetic resonance imaging for reducing truncation artifacts
RSNA 2014 Poster No.: GIE205 Jimi Huh, So Yeon Kim, Benjamin M. Yeh, Seung Soo Lee, Kyoung Won Kim, En-Haw Wu, Z. Jane Wang, Li-qin Zhao, Wei Chou Chang. Trouble Shooting for Arterial Phase Images of Gd-EOB-DTPA-enhanced Liver MR
ECR 2015 Poster No.: C-0046
J. Zhao, C.-M. Li, W. chen, L. Zhang, P. CAI, J. Wang; Chongqing/CN. Comparison of three different slow injection rates of Gd-EOB-DTPA using MR fluoroscopic triggering technique for arterial phase MR imaging of the liver
Lee YJ, Lee JM, Lee JS et al. Hepatocellular Carcinoma: Diagnostic Performance of Multidetector CT and MR Imaging — A Systematic Review and Meta-Analysis. Radiology. 2015; 275: 97-109
Dr. Davide Bellini, "Sapienza" - University of Rome, Department of Radiological Sciences, Oncology and Pathology, Italy
Dr. Riccardo Ferrari, "Sapienza" - University of Rome, Department of Radiological Sciences, Oncology and Pathology, Italy
This paper published in Radiology by Lee et al in April 2015 (1) aims to provide evidence based advices about which imaging modality to prefer in the setting of hepatocellular carcinoma (HCC).
Early detection of HCC is extremely important given that the diagnosis of HCC in an early stage allows curative treatment strategies and improves 5-years survival rate by more than 50% (2). Due to the lack of large randomized controlled trial assessing the diagnostic accuracy of different imaging modalities for detection of HCC, the present paper provides the best level of evidence and should be considered the most clinically relevant article on this topic.
Authors performed a systematic review and meta-analysis of the literature published in the past decade (from January 2000 to December 2012) in the attempt to obtain updated diagnostic performance values of CT and MR imaging for detection of HCCs in patients affected by chronic liver disease, on per patient and per lesion basis. They also evaluated the influence of hepatobiliary contrast agents on diagnostic performance.
Methods for analysis and inclusion criteria were based on the recommendations established by the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA)(3). The Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2 tool)(4) was used to evaluate quality of included studies. Heterogeneity was quantified by using the I2, sub-group analysis and meta-regression analysis were used to evaluate factors that could affect heterogeneity.
Forty studies (six on CT, 22 on MR and 12 on both CT and MR imaging) were included in the quantitative analysis, evaluating 1135 patients and 1332 HCCs with CT, 2489 patients and 2320 HCCs with MR. The mean size of the HCCs was approximately 2 cm (ranging from 0.9 to 4.6 cm), CT scanner used was at least 16 row MDCT in most of the studies (11 of 18), 1.5-Tesla magnetic field and hepatobiliary contrast agents were used in most of the articles on MR.
As major result, authors suggest that MR imaging should be considered the imaging modality of choice for evaluating HCCs in patients with chronic liver disease due to the higher per-lesion sensitivity compared to CT (80%.vs 68%; P = .0023) (Figure 1). For HCC lesions smaller than 1 cm, sensitivity decrease by 40% (Se = 48 %) for MR and by 54% (Se = 31%) for CT.
No significant differences were found in cumulative values of Sensitivity and Specificity on per patient basis between MR and CT (CT: Se 88%, Sp 74-100%; MR: Se 94%, Sp 81-100%). Moreover, promising evidences of improved sensitivity using hepatobiliary contrast agent have been reported. In particular, the use of Gadoxetic acid -enhanced MR imaging showed significantly higher per lesion sensitivity than MR imaging performed with other contrast agents (87% vs 74%, P = .03). This result further strengthens the importance of gadoxetic acid–enhanced MR imaging in the evaluation of patients suspected to have HCC.
Two limitations of the study should be mentioned. Cumulative value of sensitivity could be overestimated due to the retrospective nature of studies included in the analysis. The heterogeneity observed among studies included for per-lesion analysis reduces the strength of results.
1. Lee YJ, Lee JM, Lee JS, et al. Hepatocellular carcinoma: diagnostic performance of multidetector CT and MR imaging-a systematic review and meta-analysis. Radiology. 2015;275(1):97-109.
2. Forner A, Llovet JM, Bruix J. Hepatocellular carcinoma. Lancet. 2012;379(9822):1245-55.
3. Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. Ann Intern Med. 2009;151(4):W65-94.
4. Whiting PF, Rutjes AW, Westwood ME, et al. QUADAS-2: a revised tool for the quality assessment of diagnostic accuracy studies. Ann Intern Med. 2011;155(8):529-36.
Dr. Davide Bellini is a final year radiology resident on the “Sapienza, University of Rome” training scheme in Italy. He has a main interest in abdominal radiology, with particular regard to CT colonography. He has a vivid interest in Evidence Based Medicine, in particular meta-analysis and guidelines development. He has presented lectures and scientific talks at a number of international radiology meetings and has published in several radiological and gastroenterological journals.
Comments may be sent directly to Dr. Davide Bellini.
Raghavan K, Jeffrey RB, Patel BN, DiMaio MA, Willmann JK, Olcott EW. MDCT Diagnosis of Perineural Invasion Involving the Celiac Plexus in Intrahepatic Cholangiocarcinoma: Preliminary Observations and Clinical Implications. AJR. 2015; 205:578-584.
Ruben Vandenbulcke – Resident, Department of Radiology, AZ Delta Hospital Roeselare & University Hopitals Leuven, Belgium
Philippe Lefere – Department of Radiology, AZ Delta Hospital Roeselare, Belgium
The purpose of this retrospective study, published in the American Journal of Roentgenology in December 2015 by Raghavan et al., was to test if soft-tissue infiltration along the celiac plexus and delayed enhancement exceeding two-thirds of the tumor area on preoperative Multi Detector Computed Tomography (MDCT) correlate with histologic evidence of perineural invasion in resected intrahepatic cholangiocarcinomas. Both conditions could possibly indicate irresectability of these malignant intrahepatic lesions.
Intrahepatic cholangiocarcinoma (ICC) is the second cause of primary liver cancer after hepatocellular carcinoma, representing 10-20 % of all primary malignant liver tumors. ICCs have an aggressive tumor biology with 5-year survival rates of 5-10% for unresectable disease and 20-35% for patients where a curative treatment is chosen with radical surgical resection.1 These rates are indicating an unmet need for multimodal treatment strategies to improve current survival and recurrence-free survival rates.
Twenty patients who underwent resection for intrahepatic cholangiocarcinoma were analyzed in this single centre retrospective study. Preoperative MDCT image sets were reviewed by two radiologists for perineural invasion involving the celiac plexus. Perineural invasion was diagnosed when contiguous soft-tissue infiltration was noted extending from the intrahepatic tumor directly along the extrahepatic perineural pathways to ultimately involve the fat planes around the celiac artery. The two known extrahepatic pathways include the neural plexus along the common hepatic artery within the hepatoduodenal ligament for right liver lobe tumors and the neural plexus along the left gastric artery within the gastrohepatic ligament for left liver lobe tumors (Figure 1). Also tumor size and delayed enhancement of tumor volume were noted.
Each resected intrahepatic cholangiocarcinoma specimen was histologically examined with perineural invasion defined as presence of neoplastic cells within the epineurium, perineurium or endoneurial space.
Preoperative MDCT showed infiltration along the celiac artery in 6 out of 20 patients with in 5 out of these 6 cases histologic examination revealing intratumoral perineural invasion. In this small group, soft-tissue infiltration along the celiac artery on preoperative MDCT was significantly associated with histologic evidence of perineural invasion (Table 1). No significant association was found on perineural infiltration and tumor size or delayed enhancement more than two-thirds of tumor area.
Despite the small population size, this is the first study comparing histologic findings of perineural invasion with MDCT appearance of soft-tissue infiltration along the perineural celiac axis in the setting of intrahepatic cholangiocarcinomas.
Only few patients with intrahepatic cholangiocarcinoma present with resectable disease at the time of presentation because of the aggressive behavior of these tumors. Resectability of intrahepatic cholangiocarcinoma is defined as the ability to resect tumor to negative margins while preserving adequate functioning liver. Patients demonstrating intrahepatic metastases, visceral peritoneum perforation, macrovascular or periductal invasion, lymphatic spread and extrahepatic metastases should not undergo resection because of the poor recurrence-free survival.2,3
In similarity to pancreatic adenocarcinoma where perineural invasion is associated with a poor prognosis, perineural invasion in intrahepatic cholangiocarcinoma is a major factor contributing to the high incidence of local recurrence after resection with curative intent.3,4 This finding on MDCT is a negative prognostic factor and a contraindication for major hepatic resection with curative intent due to distant extrahepatic perineural and lymphovascular tumor spread.4,5
The high incidence of recurrence is specifically associated with this perineural en lymphovascular invasion, highlighting the need for more effective adjuvant therapies after radical tumor resection with curative intent.5
Given the relatively poor response to systemic chemotherapy (gemcitabine and cisplatin) for unresectable cholangiocarcinoma, local transarterial therapies are developed such as transarterial chemoembolization (TACE) and selective internal radiotherapy (SIRT). These therapies are showing anti-tumor effects with acceptable toxicities.1,2 Results with SIRT are more pronounced in patients with solitary tumors and absent extrahepatic tumor spread, which could be an effective downstaging method for secondary surgical treatment.6
More studies with larger sample size are needed to more extensively explore correlations between preoperative MDCT findings and the behavior of intrahepatic cholangiocarcinoma. MDCT diagnosis of perineural invasion of the celiac plexus may assist in determination of prognosis and treatment planning.
1. Mavros MN, Economopoulos KP, Alexiou VG, Pawlik TM. Treatment and prognosis for patients with intrahepatic cholangiocarcinoma: systematic review and meta-analysis. JAMA Surg. 2014; 149:565-574.
2. Bridgewater J, Galle PR, Khan SA, et al.: Guidelines for the diagnosis and management of intrahepatic cholangiocarcinoma. J Hepatol 2014; 60:1268–1289.
3. Shirai K, Ebata T, Oda K, et al. Perineural invasion is a prognostic factor in intrahepatic cholangiocarcinoma.World J Surg 2008; 32:2395–2402.
4. Deshmukh SD, Willmann JK, Jeffrey RB. Pathways of extrapancreatic perineural invasion by pancreatic adenocarcinoma: evaluation with 3D volume-rendered MDCT imaging. AJR 2010; 194:668–674.
5. Fisher SB, Patel SH, Kooby DA, et al. Lymphovascular and perineural invasion as selection criteria for adjuvant therapy in intrahepatic cholangiocarcinoma: a multi-institution analysis. HPB 2012; 14:514–522.
6. Mouli S, Memon K, Baker T, et al. Yttrium-90 radioembolization for intrahepatic cholangiocarcinoma: safety, response, and survival analysis. J Vasc Interv Radiol. 2013; 24:1227–1234.
Additional educational materials:
EURORAD Case No:1347 – Cholangiocarcinoma
Bizimi V, Kailidou E, Katsiva V, Douridas G, Tibishrani M.
ESGAR 2013: EE-007 – Multimodality assessment of Cholangiocarcinoma. A challenge for the radiologists.
Sanchez GA, Parrilla JS, Radosevic A, Pedrosa NR, Lopez JCA, Barrera MB.
Dr. Ruben Vandenbulcke is a second year resident working at the Department of Radiology at AZ Delta Hospital Roeselare, following the radiology training scheme at the University Hospitals Leuven. He completed his undergraduate medical degree at the Catholic University of Leuven in 2014. He has a main interest in abdominal radiology, with particular regard to interventional abdominal radiology.
Comments may be sent directly to Dr. Ruben Vandenbulcke.
McInnes MD, Hibbert RM, Inácio JR, Schieda N. Focal Nodular Hyperplasia and Hepatocellular Adenoma: Accuracy of Gadoxetic Acid-enhanced MR Imaging-A Systematic Review. Radiology. 2015 Nov; 277(2):413-423.
Dr. Annalisa Mantarro (Radiology Resident, Pisa University Hospital, Pisa/IT)
Prof. Emanuele Neri (Associate Professor of Radiology, Pisa University Hospital, Pisa/IT)
The purposes of this systematic review, published in Radiology by McInnes in November 2015, were: 1) to assess the diagnostic accuracy of hepatobiliary phase (HBP) in gadoxetic acid–enhanced magnetic resonance imaging (MRI) of the liver for the diagnosis of focal nodular hyperplasia (FNH) versus hepatocellular adenoma (HCA); 2) to identify the rate of reported iso- or hyperintense HCAs and hypointense FNHs in HBP.
Both FNHs and HCAs are benign hepatic neoplasms, which affect similar adult population with overlapped features (such as hypervascularity on enhanced dynamic study) at MR examination. On this ground, accurate imaging differentiation of these lesions is crucial, since the therapeutic management is very different: frequently FNHs are treated conservatively, while HCAs require resection due to its possible complications, such as haemorrhage or malignant degeneration. Based on current knowledge, the HBP of gadoxetic acid-enhanced MRI guides the differential diagnosis between these two hepatic entities with high diagnostic accuracy: FNHs are typically iso/hyperintense owing to abnormal biliary ductules; while HCAs are commonly hypointense, for the lack of biliary ductules1,2,3.
However, recent studies reported that inflammatory HCAs represent more than half of all HCAs, showing higher rate of iso- or hyperintensity at HBP4.
This systematic review was written according to PRISMA guidelines (Preferred Reporting Items for Systematic Reviews and Meta-Analyses), and the Cochrane Handbook of Diagnostic Test Accuracy Reviews. The inclusion criteria are summarized in the Table 1.
The literature search yielded 1286 literature citations. After the review of all records, 61 articles were retrieved in full-text for further evaluation and a total of 6 studies (309 patients; 164 with HCA, 233 with FNH) fulfilled the inclusion criteria for the diagnostic accuracy assessment (primary outcome). Moreover, a total of 12 case series and case reports were included, since they reported either the rate of iso- or hyperintense HCAs or hypointense FNHs in the HBP (secondary outcomes, Figure 1).
The sensitivity in FNH diagnosis based on HBP was high, ranging from 0.91 to 1.00 (lower margin of the 95% confidence interval, CI 0.77). The specificity also was generally high, ranging from 0.87 to 1.00 (lower margin of the 95% CI 0.54, Figure 2). Of note, specificity was lowest among studies in which molecular subtyping of HCA was performed. The rate of iso- or hyperintensity of HCA in HBP ranged considerably from 0% to 70%, occurring more frequently in the inflammatory subtype; while the percentage of hypointense FNHs in HBP were 0%-23% (Figure 3).
This study pointed out the high diagnostic accuracy of gadoxetic acid–enhanced MRI in differentiating HCAs from FNHs. Among the included studies for the diagnostic accuracy evaluation, it is noteworthy that only one study classified HCA subtypes with molecular imaging, reporting them in a minority of patients (29/71)5, and thus producing the lowest specificity rate (87%; 95% CI 77-94). On the basis of these findings, among the studies in which molecular subtyping was not performed, the detected higher specificity could be the result of inflammatory HCA misclassification (iso- or hyperintense in HBP) as telangectatic FNH. The lack of HCA subtyping potentially produced fewer false-positive results, resulting in overestimation of specificity. This hypothesis was supported by data of the included case series, that provided molecular subtyping and a much higher rate of iso- or hyperintense HCAs in HBP. In this respect, among the included studies for diagnostic accuracy assessment the percentage of iso- or hyperintense HCAs in HBP was much lower (range 0%-13%) than the rate reported in case series (range 9%-70%).
This study further strengthens the importance of gadoxetic acid–enhanced MRI with liver-specific contrast agents for HCA/FNH differentiation. Nevertheless, additional large studies using molecular imaging are needed to both correctly identify inflammatory HCA and determine the role of HBP in predicting HCA subtype6.
1. Grazioli L, Bondioni MP, Haradome H et al (2012). Hepatocellular adenoma and focal nodular hyperplasia: value of gadoxetic acid-enhanced MR imaging in differential diagnosis. Radiology 262:520–529.
2. Neri E, Bali MA, Ba-Ssalamah A et al (2015). ESGAR consensus statement on liver MR imaging and clinical use of liver-specific contrast agents. Eur Radiol [Epub ahead of print].
3. Bieze M, van den Esschert JW, Nio CY et al (2012) Diagnostic accuracy of MRI in differentiating hepatocellular adenoma from focal nodular hyperplasia: prospective study of the additional value of gadoxetate disodium. AJR Am J Roentgenol 199:26–34.
4. Shanbhogue AK, Prasad SR, Takahashi N et al (2011). Recent advances in cytogenetics and molecular biology of adult hepatocellular tumors: implications for imaging and management. Radiology 258:673–693.
5. Grieser C, Steffen IG, Kramme IB et al (2014). Gadoxetic acid enhanced MRI for differenti- ation of FNH and HCA: a single centre expe- rience. Eur Radiol 24:1339–1348.
6. Merkle EM1, Zech CJ, Bartolozzi C et al (2015). Consensus report from the 7th International Forum for Liver Magnetic Resonance Imaging. Eur Radiol [Epub ahead of print].
Additional educational material:
EURORAD Case No: 11885 - Inflammatory hepatocellular adenoma - CT and MRI findings
Ramos Andrade D, Costa Y, Curvo Semedo L et al.
ESGAR 2015: SS 14.09 - Focal nodular hyperplasia versus inflammatory adenoma: clues for a confident differential diagnosis during MR imaging with a hepatobiliary (Gd-EOB-DTPA) contrast agent.
Battaglia V, Turturici L, Mantarro A et al.
ESGAR 2015: EE-163 - Hepatic adenoma: pictorial review of MRI findings and correlation with new molecular-pathological classification
Ciresa M, De Gaetano AM, Guerra A et al.
Dr. Annalisa Mantarro is a final year Resident at University of Pisa, in Italy. Her major field of interest is abdominal radiology, with particular regard to MR liver imaging and CT colonography. She has published in European Journal of Radiology, British Journal of Radiology, Abdominal Imaging, World Journal of Radiology, and Medical Radiology. Has been presenter of scientific papers at ESGAR, ECR, RSNA and SIRM meetings.
Comments may be sent directly to Dr. Mantarro.
M.H. Martens, M.M. van Heeswijk, J.J. van den Broek, et al. Prospective, multicenter validation study of magnetic resonance volumetry for response assessment after preoperative chemoradiation in rectal cancer: can the results in the literature be reproduced? Int J Radiat Oncol Biol Phys 2015;93:1005-1014.
Monique Maas – Radiology Registrar, Maastricht University Medical Centre
Doenja Lambregts – Radiology Registrar, Maastricht University Medical Centre / The Netherlands Cancer Institute
The aim of this paper published in the Red Journal (International Journal of Radiation Oncology Biology Physics) in December 2015 by Martens et al. is twofold: (1) to review existing literature on tumour size and volume measurements on MRI for assessment of response to chemoradiotherapy in rectal cancer and (2) prospectively validate different size and volume cut-offs derived from literature in an independent cohort of patients.
Response assessment in rectal cancer is an emerging topic. As pointed out in a meta-analysis from 2013 by van der Paardt et al. published reports show heterogeneous results for the restaging of rectal cancer after chemoradiotherapy (CRT) on morphological MRI.1 The main issue is that in irradiated patients it is difficult to differentiate between post-treatment fibrosis and areas of viable residual tumour. As such, there is a need for additional tools to improve the diagnostic performance of imaging in assessing response, which is stressed by the increasing interest for organ preserving treatment. With organ-preservation, patients with a good or complete tumour response are deferred from surgery and undergo a local excision or even a ‘watch-and-wait’ policy. Selection of the right patients is one of the most important challenges, which requires an accurate response assessment after CRT.2,3
An approach that has often been studied is to measure changes in the size and/or volume of rectal tumours on MRI as a predictor of response. However, methodology, study outcomes and results vary considerably amongst the published reports, making it difficult to draw conclusions regarding its potential clinical diagnostic value. The paper by Martens et al. aimed to shed some light on this matter. They included 14 papers in their retrospective review and subdivided these into papers that used one-dimensional size measurements (tumour length), 3-dimensional size measurements (length x anteroposterior diameter x left-right diameter) and whole volume measurements. The outcome measures of response were the histopathological tumour regression grade, complete pathologic response vs. residual tumour and T-downstaging (downstaging of the tumour at histopathology compared to the initial cT-stage). Optimal size and volume thresholds were derived from these studies and prospectively tested in a new cohort of 146 patients from 4 different centres. The study findings for the literature review and corresponding prospective validation are presented in Table 3. The main conclusions from the authors were that one-dimensional and 3-dimensional size measurements are not of any clinical benefit in assessing response. Accuracies for one-dimensional measurement ranged between 53% and 68% in the literature review and dropped to 48-53% in the prospective validation. Similarly, results for the 3-dimensional measurements ranged between 53-79% in literature and 52-56% in the prospective validation. Best results were found for whole-volume tumour measurements. Whole-volume tumour measurements were particularly valuable in identifying patients with a complete tumour response with accuracies of up to 88% reported in literature, which could be reproduced with accuracies up to 80% in the prospective validation cohort.
Despite these relatively high overall accuracies, the main issue for clinical practice is the rather disappointing sensitivities to select complete responders, which – as illustrated in Table 3 – often not exceed 50%. Such a low sensitivity will result in missing a substantial part of the complete responders. On the other hand, the high specificity of around 90% shows that the risk for missing residual tumour is very low. All together the value of T2W-volumetry to identify patients eligible for a watch-and-wait policy appears to be limited, apart from the fact that whole volume tumour measurements are quite time consuming and therefore unpractical for daily routine.
Additionally, this paper only focuses on MR-volumetry using T2-weighted MRI. Several studies have demonstrated that diffusion-weighted MRI is more accurate than T2-weighted MRI in identifying residual tumour after CRT.1,4 Moreover, recent reports have shown promising results for DWI-volumetry, with improved sensitivities and similar specificities compared to T2W-volumetry for selecting complete responders.5,6 As such, DWI-volumetry may be of more value in clinical practice than T2W-volumetry.
As the authors acknowledge, nodal staging was not a part of this study, but of course is very relevant for assessment of a complete response.
1. van der Paardt MP, Zagers MB, Beets-Tan RG, Stoker J, Bipat S. Patients Who Undergo Preoperative Chemoradiotherapy for Locally Advanced Rectal Cancer Restaged by Using Diagnostic MR Imaging: A Systematic Review and Meta-Analysis. Radiology 2013;269:101-112.
2. Habr-Gama A, Perez RO, Nadalin W, et al. Operative versus nonoperative treatment for stage 0 distal rectal cancer following chemoradiation therapy: long-term results. Ann Surg 2004;240:711-717; discussion 717-718.
3. Lezoche G, Baldarelli M, Guerrieri M, et al. A prospective randomized study with a 5-year minimum follow-up evaluation of transanal endoscopic microsurgery versus laparoscopic total mesorectal excision after neoadjuvant therapy. Surg Endosc 2008;22:352-358
4. Kim SH, Lee JM, Hong SH, et al. Locally advanced rectal cancer: added value of diffusion-weighted MR imaging in the evaluation of tumor response to neoadjuvat chemo- and radiation therapy. Radiology 2009;253:116-125
5. Curvo-Semedo L, Lambregts DM, Maas M, et al. Rectal cancer: assessment of complete response to preoperative combined radiation therapy with chemotherapy--conventional MR volumetry versus diffusion-weighted MR imaging. Radiology 2011;260:734-743.
6. Ha HI, Kim AY, Yu CS, Park SH, Ha HK. Locally advanced rectal cancer: diffusion-weighted MR tumour volumetry and the apparent diffusion coefficient for evaluating complete remission after preoperative chemoradiation therapy. Eur Radiol 2013;23:3345-3353.
Dr. Monique Maas is a final year registrar at Maastricht University Medical Centre (The Netherlands). She completed her PhD on organ preserving treatment in rectal cancer in 2013 and is still involved in research projects on imaging and treatment of rectal cancer, in which she mainly focuses on the value of imaging to select patients for organ preservation. She has presented scientific talks and lectures at many conferences and has published in high impact journals such as The Lancet Oncology and the Journal of Clinical Oncology. Currently she is supervising PhD students in the field of rectal cancer imaging.
Furuhashi N, Suzuki K, Sakurai Y et al. Differentiation of focal-type autoimmune pancreatitis from pancreatic carcinoma: assessment by multiphase contrast-enhanced CT. Eur Radiol. 2015; 25: 1366-1374.
Dr. Gemma Miles (Radiology Registrar, Derriford Hospital, Plymouth/UK)
Dr. Simon Jackson (Consultant GI Radiologist, Derriford Hospital, Plymouth/UK)
This paper published in the European Journal of Radiology by Furhashi et al in May 2015 aims to determine the use of increasingly recognised multiphase contrast enhanced CT (CECT) features (both alone and in combination) to differentiate focal-type autoimmune pancreatitis (f-AIP) from pancreatic cancer. This retrospective, single centre study adds to emerging characteristic CECT data, useful in the interrogation of the diagnostically challenging f-AIP, where CT is the commonest first-line investigation for characterisation.
Previous studies have demonstrated various highly specific CECT imaging findings seen in focal-type AIP, which however remain of low sensitivity when considered alone. The authors demonstrate considerable increase in diagnostic sensitivity when used in combination.
A retrospective review of the standard quadruple-phase (pre-, pancreatic, portal venous and delayed phase) CECT imaging of the pancreas of two cohorts of patients; with focal AIP and pancreatic cancer, was carried out by two radiologists, blinded to the final diagnosis, demographic data and other examination findings (e.g. levels of serum IgG4 and CA19-9). The patients were initially identified from their institutions database of pancreatic and biliary system CT examinations.
Following a small number of exclusions, CECT examinations of the pancreas of twenty-two f-AIP lesions diagnosed between January 2007 and August 2012 were evaluated (the AIP diagnosis was made on the basis of International Consensus Diagnostic Criteria and Revised Japanese Pancreas Society criteria)1, whilst sixty-one histologically confirmed cases of pancreatic cancer surgically resected between November 2010 and August 2012 were compared. The authors defined a ‘focal’ lesion as “a mass-forming, focal enlargement or abnormal enhancement of the pancreas within a third of the pancreas”.
The lesion size (maximum diameter), location and enhancement patterns were assessed as seen in the table below…(TABLE 2), with previously described specific parenchymal enhancement features also commented upon within the affected pancreas including dotted enhancement in the pancreatic phase, duct penetrating sign (the main pancreatic duct visible within the lesion) – previously described by Ichikawa in 20012 on MRCP and enhanced duct sign (wall enhancement of the main pancreatic duct within the lesion).
Further assessment was made on the presence of a peri-pancreatic capsule-like rim, calcification, cyst formation and vascular involvement (as usually assessed with pancreatic neoplasms).
Peri-pancreatic strands (>10mm in length) – as described by Matsumoto et al in 2012 with respect to pancreatic body and tail carcinoma 3 were also identified.
The presence/absence of main pancreatic duct upstream dilatation and distal atrophy were commented upon within the unaffected gland. The absence of upstream main pancreatic duct dilatation in the presence of stenoses having been previously demonstrated on MR/MRCP 4.
Finally various characteristics with regard to the bile duct were assessed.
In both f-AIP and pancreatic carcinoma the majority of lesions were present within the head (82% and 74% respectively). The f-AIP lesions were overall significantly larger (35+/- 11mm in f-AIP compared with 25+/- 9mm in pancreatic cancer) (p<0.001).
In addition, significant differences in enhancement were demonstrated between the two groups; with f-AIP more frequently showing homogenous enhancement in both portal venous and delayed phases compared with pancreatic carcinoma, which showed heterogeneous decreased enhancement in both phases. This conforms the findings of previous studies. There was also a significant increase in frequency of dotted enhancement in the pancreatic phase, duct-penetrating sign, enhancing duct sign and capsule-like rim in f-AIP compared with pancreatic carcinoma. All of these findings demonstrated high specificities in the region of 0.93-0.98 but relatively low sensitivities (0.36-0.59)
Furthermore, ring-like enhancement during delayed phase and peri-pancreatic strands were significantly more common in the pancreatic cancer cohort. These features demonstrated considerable sensitivity (0.95) however limited specificities of 0.39-0.46. Whilst individual findings were not able to show both high sensitivity and specificity, the authors demonstrated by combining three-four of these significant CT features, up to 0.91 sensitivity and 0.98 specificity, with accuracies in the region of 0.94. In general there was reasonable reproducibility of the significant CT findings (moderate-good inter-observer agreement) with the exception of enhancement patterns during the pancreatic phase (K=0.24).
Interestingly, there was no significant difference identified between the two in the presence of calcification, cyst formation, vascular involvement, upstream dilatation of the main pancreatic duct and distal atrophy. No significant differences were identified within CBD features in either stented or non-stented patients.
This study adds to the growing radiological experience of differentiating focal AIP from pancreatic cancer, essential to avoid unnecessary surgical resections and their potential considerable morbidity. The study confirms similar CECT features of f-AIP, previously described on MR/MRCP such as duct-penetrating sign and dotted enhancement during the pancreatic phase 5 which the authors postulate represents residual normal or mildly involved pancreatic lobules within a lesion. In addition this study further reinforces the significance of previously described CECT signs such as enhanced-duct sign, showing high specificity (0.98) compared with pancreatic carcinoma.
As one might expect, and confirmed by this study, when three or more of the significant CT features are considered together there is a considerable increase in sensitivity.
Furthermore, as the authors suggest, in reality, reviewing the entire scan, to identify extra-pancreatic manifestations of IgG4 disease and considering both the biochemical and clinical findings, would clearly enhance diagnostic accuracy of f-AIP. In equivocal cases, the importance of multi-modality imaging and specialist hepato-pancreaticobilliary MDT discussion should not be underestimated.
1. Shimosegawa T, Chiari ST, Frulloni L et al (2011). International consensus diagnostic criteria for autoimmune pancreatitis: guidelines of the International Association of Pancreatology. Pancreas 40:352-358.
2. Ichikawa T, Sou H, Araki T et al (2001). Duct-penetrating sign at MRCP: usefulness for differentiating inflammatory pancreatic mass from pancreatic carcinomas. Radiology 221:107-116.
3. Matsumoto S, Mori H, Kiyonaga M et al (2012). “Peripancreatic strands appearance” in pancreatic body and tail carcinoma: evaluation by multi-detector CT with pathological correlation, Abdom Imaging 37: 602-608.
4. Negrelli R, Manfredi R, Pedrinolla B et al (2015). Pancreatic duct abnormalities in focal AIP: MR/MRCP imaging findings. Eur Radiol 25(2):359-67.
5. Sugiyama Y, Fujinaga Y, Kadoya M et al (2012). Characteristic magnetic resonance features of focal autoimmune pancreatitis useful for differentiation from pancreatic cancer. Jpn Radiol 30:296-309.
Additional educational material:
EURORAD Case No: 5373 – A case of focal autoimmune pancreatitis.
Corriero A, Pratali A, Cappelli C, Caproni G, Battaglia V, Forasassi F, Mazzeo S, Bartolozzi C.et al.
ESGAR 2015: LS 2.3 – Autoimmune pancreatitis.
SA Jackson (Plymouth/UK)
ESGAR 2014, Cum Laude e-poster: EE-138 Autoimmune Pancreatitis – the Clinicoradiological Story So Far.
G. Miles, T. Adlan, F. Wotton, S.A. Jackson.
Dr. Gemma Miles is a final year registrar on the Peninsula radiology training scheme in the UK, having completed her undergraduate medical degree and intercalated BSc at Imperial College, London. She is currently undertaking an MSc in Clinical Leadership, Management and Innovation through the University of Plymouth. She has previously undertaken research into the correlation of CT with histopathological outcome in the assessment of tumour origin in pancreatic cancer, with work presented at a number of international meetings. She achieved a Cum Laude award at ESGAR 2014 for her e-Poster on autoimmune pancreatitis and in addition has published articles in Emergency Radiology, Journal of Clinical Ultrasound and Imaging.
Comments may be sent directly to Dr. Gemma Miles.
J.J. Atema, S.L. Gans, A. Van Randen et al. Comparison of Imaging Strategies with Conditional versus Immediate Contrast-Enhanced Computed Tomography in Patients with Clinical Suspicion of Acute Appendicitis. Eur Radiol. 2015; 25(8): 2445–2452.
Oliver Hulson (Radiology Registrar, St. James’s University Hospital, Leeds/UK)
Damian Tolan (Consultant Radiologist, St. James’s University Hospital, Leeds/UK)
Atema and colleagues from the Netherlands published this paper in European Radiology in April 2015 with a primary aim to, ‘compare the diagnostic accuracy of a conditional CT strategy… versus an immediate CT strategy in adult patients with a clinical suspicion of acute appendicitis’. Whilst the results are from a historic cohort of patients that were collected prospectively as part a multi-centred study published prior to this between 2005 and 2006, the outcomes are still valid and serve to highlight a number of prescient points. The conditional CT strategy is one in which a CT examination is only performed when a prior ultrasound investigation is either negative or inconclusive, whereas an immediate CT strategy is one where, as implied, a CT examination is requested as the first line imaging investigation.
The ultrasound study was performed using both curved 3.5-5.0 MHz and linear 10MHz array probes, using widely accepted features for the diagnosis of acute appendicitis (including a thickened appendix, the presence of an appendicolith and free fluid amongst others). The CT examination was performed with intravenous contrast in the portal venous phase, with no oral or rectal contrast administered. Again, widely accepted imaging features were utilized in the diagnosis of acute appendicitis, but the final diagnosis was at the discretion of the reporting radiologist.
Over the 19-month study period, 422 patients were randomised on a 1:1 ratio to the conditional or immediate CT imaging strategy cohorts. Figure 1 demonstrates flowcharts for both study populations, including the ‘imaging diagnosis’ and ‘final diagnosis’ in both groups. This final diagnosis was assigned to each patient by an expert panel forming a composite assessment, based on histopathology where available, imaging findings, surgical findings and at least 6 months of follow up.
As flowchart A demonstrates, 199/422 (47.2%) patients in the ‘conditional’ group had an ultrasound that was either negative or inconclusive for appendicitis and thus proceeded to CT. Of these 57/199 (28.6%) subsequently were found to be positive for acute appendicitis on CT. A further 10 were given a final diagnosis of appendicitis, despite a negative or inconclusive CT study.
In the immediate CT cohort in Flowchart B, 260/422 (61.6%) patients were diagnosed with appendicitis on CT, with 238 (91.5%) of these cases given a final diagnosis concurring with this.
Table 1 illustrates the diagnostic accuracy of each imaging strategy. Interestingly the conditional CT group had a higher false positive rate of acute appendicitis compared to the immediate CT group, of 14% and 8% respectively, but no p-value was calculated for this. The authors hypothesize that this is a statistical quirk, due to the stepwise nature of the flowchart accumulating both more true and false positives. Otherwise there were no another statistically significant differences between the cohorts. Importantly, the sensitivity and specificity for both cohorts were broadly similar (with the specificity a little higher for the immediate CT strategy), as were the number of missed cases and the negative and positive predictive values.
The study justifies the use of ultrasound as the first line imaging investigation in adult patients being assessed for acute appendicitis. Numerous studies have highlighted the increasing dependence on CT in the assessment of the acute abdomen(1)(2), and acute appendicitis is generally regarded as a pathology predominantly, but not exclusively, affecting young people. This is particularly important as a number of papers have been published recently highlighting the long-term effects of clinical radiation exposure, in particular the potential increased lifetime cancer risk(3),(4). If the diagnosis can be achieved, or an alternative cause for the patient’s symptoms established without the use of CT, then these potentially harmful effects of ionizing radiation are obviously avoided.
Having said that, the sonographic diagnosis of appendicitis can be challenging and other studies have demonstrated significant operator-dependence and interobserver variability(5). However this trial took place across a wide number of institutions and a large number of radiologists which strengthens its wider applicability. Another potential issue arises in the surgeons’ mistrust of ultrasound in this context(6), and it can envisaged that a number of CT requests will arise in real life clinical practice in order to prove or disprove the sonographic findings.
Ultimately, each patient’s assessment and the imaging strategy employed should take into account the clinical presentation, examination findings and laboratory investigations to achieve the diagnosis. By working closely with surgical colleagues, and offering the most appropriate imaging modality for that particular case, it is hoped that the most rational imaging strategy is employed, and the use of ionizing radiation is kept as low as reasonably applicable. As Atema and colleagues suggest, ultrasound as the first line imaging investigation seems a rational and cost effective approach for the assessment of uncomplicated acute appendicitis. It will be interesting to see whether health care service commissioners see this research and mandate radiologists to improve their ultrasound skills to reduce cost and radiation for this patient group.
Dr. Oliver Hulson is a final year registrar on the West Yorkshire radiology training scheme in the UK. He completed his undergraduate medical degree at the University of Leeds. He has a keen interest in training and education, completing his MSc in Clinical Education through the University of Edinburgh alongside his clinical studies. He has presented at a number of international radiology meetings, has published in Paediatric Radiology and provided a number of radiology teaching cases for the British Medical Journal.
Additional educational materials:
EURORAD Case Nr.: 12077 - Femoral hernial sac containing inflamed appendix: De Garengeot's hernia
ESGAR 2013: SE-080 Free-hand sonoelastographic evaluation of appendicitis
G. Pekindil, T. Coskun; Manisa/TR
ESGAR 2014: WS 22.01 - Acute lower abdominal pain
M. D'Onofrio (Verona, Italy)
1. Brown J, Shesser R. Computed tomography scan use in the emergency department evaluation of patients with nonspecific abdominal pain. Ann Emerg Med. 2004 Oct;44(4):S32.
2. Larson DB, Johnson LW, Schnell BM, Salisbury SR, Forman HP. National Trends in CT Use in the Emergency Department: 1995–2007 1. Radiology. 2011;258(1):164–73.
3. Brenner D, Hall E. Computed tomography--an increasing source of radiation exposure. N Engl J Med. 2007 Nov;29(357):2277–84.
4. Pearce M, Salotti J, Little M, McHugh K. Radiation exposure from CT scans in childhood and subsequent risk of leukaemia and brain tumours: a retrospective cohort study. The Lancet. 2012 Aug;380(9840):499–505.
5. Pinto F, Pinto A, Russo A, Coppolino F, Bracale R, Fonio P, et al. Accuracy of ultrasonography in the diagnosis of acute appendicitis in adult patients: review of the literature. Crit Ultrasound J. 2013;5(Suppl 1):S2.
6. Exadaktylos AK, Sadowski-Cron C, Mader P, Weissmann M, Dinkel HP, Negri M, et al. Decision making in patients with acute abdominal pain at a university and at a rural hospital: Does the value of abdominal sonography differ? World J Emerg Surg. 2008;3(1):29.
You can send your comments directly to Dr. Oliver Hulsen.