Imaging plays a major role in the clinical management of patients with or at risk for hepatocellular carcinoma (HCC). It is used for screening and surveillance of at-risk patients, diagnosis and staging of HCC, informing prognosis and treatment selection, monitoring treatment response, and determining eligibility and priority for liver transplantation. Unlike with other cancers, imaging is frequently used to establish the definitive diagnosis of HCC without histologic confirmation,1,2 and pathology confirmation is often not mandated by clinical practice guidelines before instituting treatment, including major treatment such as liver transplantation. Because of the important role of imaging, numerous scientific organizations and societies have proposed imaging-based diagnostic systems for HCC.3 Although these diagnostic systems represent important advances, they are narrow in scope and have inconsistencies and ambiguities.
To address these limitations, in 2008, the American College of Radiology (ACR) supported the development of the Liver Imaging Reporting and Data System (LI-RADS) for comprehensive and standardized terminology, interpretation, and reporting of imaging examinations for the diagnosis of HCC.4,5
The imaging modalities addressed by this system in 2014 are computed tomography (CT) and magnetic resonance imaging (MRI). This article reviews the basic concepts of LI-RADS, emphasizing those that are most relevant to pathologists, including the categories, diagnostic algorithms, major features, and ancillary features for the diagnosis of HCC. LI-RADS and other system are not designed to detect early HCC and so have limited sensitivity for such lesions. Moreover, despite continuous advances in imaging technology, imaging detection and characterization of small (<1 cm) nodules remains limited; in addition, colocalization of small nodules and pathology is difficult which complicates the assessment of diagnostic performance of imaging for such lesions. For these reasons LI-RADS and most other systems do not permit noninvasive diagnosis of HCC for <1 cm lesions. This article will touch upon the presumed pathophysiological basis of key imaging findings used in the diagnosis of HCC, and briefly discuss future directions.
LIVER IMAGING REPORTING AND DATA SYSTEM OVERVIEW
Role of Imaging
Contrast-enhanced CT and MRI play critical roles in diagnosis and management of HCC. These modalities contribute to screening and surveillance of HCC in at-risk patients,6,7 either for evaluation of positive findings on ultrasound (US) or for detection of HCC missed by US.8,9 In addition, CT and MRI are widely used for detection, diagnosis, and staging of HCC without biopsy confirmation.7,10–14 Unlike other cancers, pathology confirmation is often not mandated for the diagnosis of HCC according to diagnostic systems proposed by numerous international scientific organizations.3 Imaging modalities also guide management of HCC,12–14 determine prognosis,14 assess treatment response,13,15–17 and determine eligibility and priority for liver transplantation.15,16,18–21
Role of the Pathologist
The pathologist plays a critical role in working with other physicians to optimize patient care. In particular, anatomic pathologists are experts in tissue microarchitecture; therefore, they can provide guidance on appropriate terminology for pathologic lesions, can give input on mechanistic or pathophysiological underpinnings of radiologic features, can suggest what types of observations may benefit from tissue sampling, or can discuss the diagnostic limitations of such biopsies for a certain circumstance. The pathologist may help clarify communications and help ensure that clinical-radiologic-pathologic correlations are being performed on a regular basis, although each physician involved in a given patient’s care must take on some of this responsibility. Such correlations become especially important in synthesizing difficult or complex cases, or in cases with discordant findings. This is probably best achieved through tumor boards or other types of multidisciplinary clinical conferences. Overall, the process of standardizing and enhancing diagnostic workups both in general, and for LI-RADS in particular, may hugely benefit from the pathologist’s contribution.
Need for Standardization
Despite the important roles of imaging, there was until recently little or no standardization in the interpretation and reporting of radiologic examinations for diagnosis of HCC. This lack of standardization has unfavorable consequences at several levels: the lack of technique standardization leads to suboptimal technique and impaired diagnostic performance. The lack of terminology and interpretation standardization leads to inconsistent terminology in clinical practice, research, and literature. The lack of reporting structure and content standardization leads to omission of relevant information from reports. Finally, the lack of data collection standardization leads to limited ability to create registries, perform data mining, and monitor outcomes.
Development of Diagnostic Systems
To address the important roles played by imaging, numerous scientific organizations and societies have proposed diagnostic systems for the interpretation of liver imaging examinations for the diagnosis of HCC.3 Several systems have been developed in parallel in Asia, Europe, USA, and developing countries. Six of the most current imaging-based diagnostic systems, organized by geographic region, include those endorsed by the Asian Pacific Association,20 Japan Society of Hepatology,16 European Association for the Study of the Liver (EASL), and European Organization for Research and Treatment of Cancer (EORTC),10 American Association for the Study of Liver Diseases (AASLD),7 United Network for Organ Sharing and Organ Procurement and Transplantation Network (UNOS-OPTN),22and LI-RADS.23Over a 13-year period, many of these systems have been updated and refined to reflect progress in radiologic knowledge and technological advances.
Limitations of Current Radiologic Systems for Hepatocellular Carcinoma Diagnosis
Despite the many important contributions of these diagnostic systems, they have a number of limitations. These systems categorize observations in a binary fashion, either as definite HCC or not definite HCC, thereby not reflecting the hepatocarcinogenesis continuum.24,25 Current systems focus only on HCC and do not address the full spectrum of lesions and pseudolesions commonly encountered in patients at risk for HCC. These commonly are areas of arterialized flow due to arterioportal shunting in non-neoplastic parenchyma. Other lesions can include not only nodules in the hepatocarcinogenesis spectrum (precursor lesions such as dysplastic nodules), but also benign lesions not associated with hepatocarcinogenesis (such as hemangiomas).
Prior systemslack a lexicon and atlas, with supporting diagrams and examples for thorough definition of imaging observations. These systems omit imaging criteria for the diagnosis of vascular invasion, which has major implications in tumor staging, treatment, and transplant eligibility.26–28 They do not address non-HCC malignancies such as intrahepatic cholangiocarcinoma, hepatobiliary biphenotypic tumor, metastasis, lymphoma, and posttransplant lymphoproliferative disorder. Except for UNOS-OPTN,22 most systems do not provide guidance on optimal imaging technique. Except for the Japan Society of Hepatology,16 most systems do not take into account hepatobiliary contrast agents which permit detection and characterization of observations based not only on vascularity but also on hepatocellular uptake due to expression of transmembrane organic anion-transporting polypeptide transporters.29
Organization of Liver Imaging Reporting and Data System
In 2008, the ACR convened a committee to develop a standardized interpretation, reporting, and data collection system for CT and MRI examinations in at-risk patients. The resulting system was named LI-RADS.
The organizational chart is shown in Figure 1. A Steering Committee oversees 12 Working Groups, each with its own charges and responsibilities. These Working Groups develop their own content. The material is vetted by the Atlas and Lexicon Working Group and submitted to the Steering Committee. All together, the Steering Committee and Working Groups consist of 150 members, including radiologists, interventional radiologists, hepatologists, surgeons, biostatisticians, informatics and lexicon experts, ACR representatives, and hepatopathologists. One of the newest Working Groups is the LI-RADS Rad-Path Working Group which seeks to standardize terminology relevant to the cirrhotic liver and cirrhosis-associated nodules across the disciplines of radiology and pathology.
LI-RADS development is regularly updated based on expert opinion, feedback from users, and scientific evidence. LI-RADS also incorporates feedback from the AASLD and UNOS-OPTN to achieve congruency with these systems.30
Liver Imaging Reporting and Data System Content
The ACR-supported LI-RADS includes a standardized terminology, interpretation, and reporting system for CT and MRI examinations of liver in patients at risk for HCC. LI-RADS includes an atlas and lexicon of controlled terminology to standardize the content and structure of reports. The system revolves around a diagnostic algorithm and a table for assigning category codes to imaging observations based on the relative probability of HCC. This comprehensive system addresses the full spectrum of lesions and pseudolesions encountered in cirrhotic patients. It also addresses macrovasculoinvasive HCC and non-HCC malignancies. It includes material on minimum acceptable technical parameters, guidance on reporting, and diagnostic options for different categories of findings.30 LI-RADS is fully indexed and the most recent version is available online on the ACR Web site as a hyperlinked document.23
Evolution of Liver Imaging Reporting and Data System
In 2011, the ACR released version 1.0 of LI-RADS,4 which included basic terminology and definitions. An update was released in 2013, which added a lexicon and an imaging atlas, new content, technical requirements, and reporting standards.5
The newest version of LI-RADS (version 2014) adds several more enhancements,23 including a simplified diagnostic algorithm, expanded background material comparing LI-RADS with AASLD and UNOS-OPTN criteria, and key modifications to achieve congruency between LR-5 and OPTN class 5 and AASLD.
In LI-RADS, ordinal category codes are assigned to individual observations according to their likelihood of benignity or malignancy (HCC or non-HCC malignancy) in patients at risk for HCC. The LI-RADS category codes and corresponding concepts for each category are summarized in Table 1.
The LI-RADS 1 (LR-1) category is assigned to observations with imaging features diagnostic of a benign entity or which demonstrate spontaneous disappearance at follow-up. This category is used when the radiologist is certain that an observation is benign. Examples of definitely benign entities include cysts, hemangiomas, vascular anomalies, perfusion alterations, hepatic fat deposition or sparing, hypertrophic pseudomasses, confluent fibrosis, or focal scars. Notice that regenerative nodules are not categorized LR-1. The reason is that noninvasive imaging is unable to reliably exclude the presence neoplastic changes within hepatocellular nodules. Thus, LI-RADS does not permit radiologists to categorize any cirrhosis-associate nodules as LR-1 (definitely benign). Instead, cirrhosis-associated nodules are either ignored in radiology reports (if they are indistinguishable from other background nodules) or, as described below, they are categorized as LR-2 or higher.
The LI-RADS 2 (LR-2) category is assigned to observations with imaging features suggestive but not diagnostic of a benign entity. This category is used when there is high probability, but not absolute certainty, that an observation is benign. Most probably, benign entities have one or more atypical features that prevent categorization as LR-1.
One of the probably benign entities merits further discussion, the LR-2 cirrhosis-associated nodules. The majority of cirrhosis-associated nodules are regenerative nodules which at imaging appear indistinguishable from any other nodules. Although background cirrhotic nodules are ignored, nodules that are distinctive in their appearance are categorized LR-2 or higher, based upon the estimated probability of HCC. Nodules that differ from background liver in size, signal intensity, or attenuation but that isoenhance to background liver are probably benign and are categorized LR-2. Nodules that enhance differently from background liver are categorized as LR-3, LR-4, or LR-5 depending on their imaging features and exact enhancement pattern.
The LI-RADS 3 (LR-3) category is assigned to observations that do not meet unequivocal criteria for other LI-RADS categories. This category is used when the radiologist considers that both HCC and a benign entity have moderate probability.
The LI-RADS 4 (LR-4) category is assigned to observations with imaging features suggestive but not diagnostic of HCC. This category is used when there is high probability, but not absolute certainty, that an observation is HCC.
The LI-RADS 5 (LR-5) category is assigned to observations with imaging features definitely diagnostic of HCC. This category is used when the radiologist has absolute certainty that an observation is HCC. The imaging criteria for HCC were designed to provide high specificity to prevent false-positive diagnosis of HCC in patients eligible for liver transplantation. LR-5 category code is equivalent to OPTN 5, as described in the OPTN policies.22 Hence, small or early HCCs that do not meet the criteria for LR-5 may be categorized as LR-4 or LR-3. The corollary is that a LI-RADS category lower than LR-5 does not exclude a diagnosis of HCC. Thus, pathologists should not be surprised if they provide a histologic diagnosis of HCC for an observation that was categorized as LR-4 or lower. In contrast, the authors would encourage pathologists to alert their clinical and radiology colleagues if LR-5 observations are found not to be HCC at histologic sampling. It is advisable to include this form of feedback as a part of the pathologic assessment of lesions within the liver, which should generally involve radiologic correlation. This is especially important in cases of multiple lesions or discordant pathologic and radiologic findings.
The LI-RADS 5V (LR-5V) category is assigned to observations associated with definite macrovascular venous invasion. This category is used when the radiologist has absolute certainty than an observation is a tumor invading a portal or hepatic vein. This category was created because vascular invasion usually constitutes a contraindication to curative treatments such as liver transplantation and hepatectomy.
The LI-RADS M (LR-M) category is assigned to observations that are thought to be probably or definitely malignant but that lack imaging features specific for HCC. This category exists because the management is different for these malignancies than for HCC. Examples include intrahepatic cholangiocarcinoma, metastasis, biphenotypic hepatobiliary tumor, lymphoma, and posttransplant lymphoproliferative disorder.2 Because HCC is by far the most common malignancy in patients with cirrhosis, most observations categorized LR-M represent atypical HCC (eg, HCC with extensive necrosis, schirrous HCC). However, the possibility of a non-HCC malignancy is thought to be sufficiently high that biopsy or other diagnostic work-up is warranted to inform prognosis and clinical decision making. Therefore, pathologists play an important role in the characterization and management of such observations.
The LI-RADS Treated (LR-Treated) category is assigned to observations that have undergone locoregional treatment, such as hepatectomy, ablative therapy, or chemoembolization. Formal imaging criteria for assessment of treatment response or detection of residual or recurrent tumor are in development and will be implemented in future versions of LI-RADS.
Correlation Between Liver Imaging Reporting and Data System Categories and Histologic Grade
It is important to emphasize that there is not a one-to-one correspondence between LI-RADS categories and histologic progression or grade for cirrhosis-associated nodules. As illustrated in Figure 2, each nodule type in the hepatocarcinogenesis spectrum may span a range of LI-RADS categories. Also illustrated in the figure and mentioned earlier in the text are the following points. The LR-1 category is not assigned to any cirrhosis-associated nodule. LR-5 categorization is assigned to HCC. Many HCCs are likely to be categorized LR-4 or lower.
Further, LI-RADS makes no attempt to differentiate precursor nodules (cirrhotic nodules, low-grade dysplastic nodules, high-grade dysplastic nodules). This is due in part to the imaging overlap between nodules and in part due to the lack of standardized imaging terminology in the previous radiology literature. With the standardization of radiology terminology provided by LI-RADS it may be possible in the future to perform research studies to inform the development of more precise radiology classification system. Instead, LI-RADS assigns category codes representing the probability of benignity or malignancy.
The LI-RADS diagnostic algorithm was designed to mirror the thought process of a radiologist evaluating imaging observations in patients who are at risk for HCC. A formal structure with combinations of diagnostic criteria was established to achieve reproducibility in lesion categorization among radiologists. The 2014 version of the LI-RADS algorithm is shown in Figure 3.
An observation refers to an area with imaging features that differ from those of the adjacent liver parenchyma. LI-RADS endorses the use of “observation” over “lesion” because many imaging abnormalities are not true lesions but may be pseudolesions or imaging artifacts.
The notion of “high-risk patients” refers to those in whom the incidence of HCC is sufficiently high to justify screening or surveillance according to the AASLD guidelines.7 According to the current AASLD guidelines, surveillance is recommended for patients with cirrhosis of any cause or in subgroups of patients with chronic hepatitis B.
LI-RADS has been constructed for evaluation of observations in high-risk patients. This was done intentionally to preserve high positive predictive value for the diagnosis of HCC. Hence, LI-RADS should not be used to interpret CT or MRI in non–high-risk individuals. It must also be noted that the target population, the intended users, the categorization of observations, and the imaging methods addressed by LI-RADS may differ from other American diagnostic systems for HCC (Table 2). The AASLD target population includes patients at risk for HCC enrolled in a surveillance program.7 OPTN applies to patients with HCC considered for liver transplantation.22 A comparison between diagnostic systems has been discussed in detail elsewhere.3
Once an observation is identified, the radiologist must first determine whether the observation has previously been treated. If this is the case based on patient history and imaging findings, the LR-Treated category code is assigned.
For untreated observations, the radiologist must then determine whether the imaging features are diagnostic of a definitely benign entity (LR-1), suggestive but not typical of a benign entity (LR-2), or neither definitely nor probably benign.
In the latter scenario, where the observations are neither definitely nor probably benign, the radiologist must determine whether there are one or more imaging features that favor non-HCC malignancy (LR-M).
If a malignancy other than HCC is thought to be unlikely, the radiologist must seek the presence of enhancing tumor in vein which would indicate the presence of macrovascular invasive tumor (LR-5V).
If there is no definite macrovascular venous invasion, the radiologist must stratify the LI-RADS category according to the table at the bottom of the algorithm. This table permits assignment of LI-RADS categories (LR-3, LR-4, LR-5) according to the number and type of major features identified on individual observations.
The major features include arterial phase enhancement, maximum observation diameter, washout appearance, capsule appearance, and threshold growth.
Arterial phase enhancement is defined as enhancement in the arterial phase that is unequivocally greater than that of the surrounding liver parenchyma. This enhancement pattern is thought to reflect the gradual change in blood supply from portal to arterial due to capillarization and neoarterialization that accompanies the development of progressed HCC.24 In contrast to progressed HCCs, early HCCs tend to be hypoenhancing or isoenhancing to the background liver in the arterial phase.31 The diagnostic performance of arterial phase hyperenhancement has been evaluated on CT and MRI.7,32–35 Although arterial phase hyperenhancement is a feature of progressed HCC, it is not specific to HCC and may be observed in benign entities such as hemangiomas, arterioportal shunts, and perfusion anomalies. Therefore, to maintain high specificity for HCC, this imaging feature must be observed in combination with other major features discussed below. Arterial phase hyperenhancement is required but not sufficient for LR-5 categorization. Thus, many early HCCs, which lack neovascularization and hence lack arterial phase hyperenhancement, cannot be categorized LR-5.
Diameter is defined as the largest dimension of an observation, measured from outer edge to outer edge in the imaging sequence, phase, and plane in which the margins are most sharply demarcated. In LI-RADS, observation diameter is stratified according to size (<20 vs. ≥20 mm) because larger observations are associated with a higher likelihood of HCC and progression.36–38 In LI-RADS, if an observation is <10 mm, it cannot be categorized as LR-5 (definitely HCC). The reason is that despite continuous advances in imaging technology, imaging detection and characterization of small (<1 cm) nodules remains limited; in addition, colocalization of small nodules and pathology is difficult which complicates the assessment of diagnostic performance of imaging for such lesions.39
Washout appearance is defined as temporal reduction in enhancement relative to liver from an earlier to a later phase resulting in portal venous phase hypoenhancement or delayed phase hypoenhancement. This enhancement pattern may reflect multiple concomitant phenomena: rapid venous drainage, reduced portal venous supply to progressed HCC,24 and later enhancement of the background liver. The diagnostic performance of washout appearance has been assessed on CT and MRI.33–35,40 Like arterial phase hyperenhancement, washout appearance by itself is not specific for HCC, as this feature may be observed in cirrhotic nodules and dysplastic nodules. Furthermore, pseudolesions, such as focal areas of parenchymal distortion and enhancing fibrosis, may also create the perception of “washout.”
Threshold growth is defined as a diameter increase of a mass by a minimum of 5 mm and by the following amounts: at least 50% diameter increase if time interval between examinations is ≤6 months or at least 100% diameter increase if time interval between examinations is >6 months. Furthermore, a new ≥10 mm mass also represents threshold growth, regardless of the time interval. A new <10 mm mass does not represent threshold growth. Tumor growth is an important feature used in radiology for establishing the diagnosis of malignancy. This concept is also applicable to the diagnosis of HCC when interval imaging is available.42,43
The decision to assign LR-3, LR-4, or LR-5 categories to observations is based on the selection of appropriate columns and rows in the LI-RADS table. The column is selected according to the enhancement pattern (arterial phase hypoenhancement or isoenhancement vs. hyperenhancement) and diameter. The row is selected according to the number of major features present among the following: washout appearance, capsule, and threshold growth. The appropriate category is found at the intersection of column and row selected. As illustrated, observations with few features concerning for HCC tend to be situated in the upper left corner of the table and be assigned a lower category, whereas observations with several features concerning for HCC tend to be situated in the lower right corner of the table and be assigned a higher category code.
LI-RADS 2014 introduced 2 LR-5 subcategories: LR-5g (as in “growth”) and LR-5us (as in “ultrasound”), to improve congruency with OPTN and AASLD for 10- to 19-mm observations with arterial phase hyperenhancement and only 1 other major feature (“washout,” “capsule,” or threshold growth). LR-5g refers to observations that meet criteria for OPTN Class 5A-g, that is, those that present ≥50% diameter increase in ≤6 months.22 LR-5us refers to observations that meet the AASLD HCC criteria, that is, those that measure 10 to 19 mm, with arterial phase hyperenhancement, “washout” appearance, and were visible as discrete nodule at antecedent surveillance US.7
Radiologists may, at their discretion, use ancillary imaging features that modulate the probability of HCC or benignity (Table 3). Taken in isolation, these features do not permit reliable categorization of observations. These ancillary features may favor malignancy to upgrade category by one or more categories (up to but not beyond LR-4). Features that may favor benignity may be used to downgrade category by one or more categories. A full discussion of ancillary features is beyond the scope of this review. Two important points must be emphasized: ancillary features cannot be used to upgrade a category to LR-5 and that is based on a need to maintain congruency with OPTN which does not recognize ancillary features. Also, radiologists may apply ancillary features at their discretion; there is not yet sufficient scientific evidence to mandate their use.
A representative case is shown in Figure 4 to illustrate that HCC diagnosis may be established with imaging without mandatory requirement for pathology confirmation. This situation applies to observations that meet the major imaging criteria for categorization as LI-RADS 5.
Future versions of LI-RADS will refine diagnostic imaging criteria in response to technological advances and understanding of HCC pathogenesis. The modular nature of LI-RADS facilitates integration of suggestions and improvements from different Working Groups. A radiology/pathology working group has been formed to harmonize the terminology and exchange information about HCC pathogenesis.
We envision future unification of a comprehensive diagnostic system across geographical regions. Such a system will be shared by relevant stakeholders (pathologists, hepatologists, hepatobiliary surgeons, radiologists) working in collaboration. Future versions will also expand the scope of LI-RADS beyond HCC, to include entire spectrum of observations encountered in liver imaging.
LI-RADS has been developed for the standardization of terminology, interpretation, and reporting of imaging examinations for the diagnosis of HCC in at-risk patients (Table 4). The current version of LI-RADS addresses 2 imaging modalities: CT and MRI. The system incorporates major and ancillary imaging findings to categorize observations. It does so through a diagnostic algorithm, table, list of ancillary features, and tie-breaking rules. The major imaging features that permit the categorization of LR-5 (definite HCC) are features of progressed HCC. Future versions will incorporate technological advances and understanding of HCC pathogenesis. We also envision a future unification of the diagnostic criteria into a single comprehensive system codeveloped by the relevant stakeholders.
1. Tang A, Cruite I, Sirlin CB. Toward a standardized system for hepatocellular carcinoma diagnosis
using computed tomography
and MRI. Expert Rev Gastroenterol Hepatol. 2013;7:269–279.
2. Santillan CS, Tang A, Cruite I, et al.. Understanding LI-RADS: a primer for practical use. Magn Reson Imaging Clin N Am. 2014;22:337–352.
3. Cruite I, Tang A, Sirlin CB. Imaging-based diagnostic systems for hepatocellular carcinoma
. AJR Am J Roentgenol. 2013;201:41–55.
6. Baron RL, Peterson MS. From the RSNA refresher courses. Radiographics. 2001;21(suppl 1):S117–S132.
7. Bruix J, Sherman M. American Association for the Study of Liver Diseases. Management of hepatocellular carcinoma
: an update. Hepatology. 2011;53:1020–1022.
8. Taouli B, Krinsky GA. Diagnostic imaging of hepatocellular carcinoma
in patients with cirrhosis before liver transplantation. Liver Transpl. 2006;12(suppl 2):S1–S7.
9. Krinsky G. Imaging of dysplastic nodules and small hepatocellular carcinomas: experience with explanted livers. Intervirology. 2004;47:191–198.
10. European Association for the Study of the Liver, European Organisation for Research and Treatment of Cancer. EASL-EORTC clinical practice guidelines: management of hepatocellular carcinoma
. J Hepatol. 2012;56:908–943.
11. Song P, Tobe RG, Inagaki Y, et al.. The management of hepatocellular carcinoma
around the world: a comparison of guidelines from 2001 to 2011. Liver Int. 2012;32:1053–1063.
12. Hanna RF, Kased N, Kwan SW, et al.. Double-contrast MRI for accurate staging of hepatocellular carcinoma
in patients with cirrhosis. AJR Am J Roentgenol. 2008;190:47–57.
13. Bruix J, Sherman M, Llovet JM, et al.. Clinical management of hepatocellular carcinoma
. Conclusions of the Barcelona-2000 EASL conference. European Association for the Study of the Liver. J Hepatol. 2001;35:421–430.
14. Ito K, Mitchell DG, Siegelman ES. Cirrhosis: MR imaging features. Magn Reson Imaging Clin N Am. 2002;10:75–92. vi.
15. Bruix J, Sherman M. Management of hepatocellular carcinoma
. Hepatology. 2005;42:1208–1236.
16. Kudo M. Diagnostic imaging of hepatocellular carcinoma
: recent progress. Oncology. 2011;81(suppl 1):73–85.
17. Memon K, Kulik L, Lewandowski RJ, et al.. Radiographic response to locoregional therapy in hepatocellular carcinoma
predicts patient survival times. Gastroenterology. 2011;141:526–535. 535 e1-2.
18. Freeman RB, Mithoefer A, Ruthazer R, et al.. Optimizing staging for hepatocellular carcinoma
before liver transplantation: a retrospective analysis of the UNOS/OPTN database. Liver Transpl. 2006;12:1504–1511.
19. Jelic S, Sotiropoulos GC. Hepatocellular carcinoma
: ESMO Clinical Practice Guidelines for diagnosis
, treatment and follow-up. Ann Oncol. 2010;21(suppl 5):v59–v64.
20. Omata M, Lesmana LA, Tateishi R, et al.. Asian Pacific Association for the Study of the Liver consensus recommendations on hepatocellular carcinoma
. Hepatol Int. 2010;4:439–474.
21. Pomfret EA, Washburn K, Wald C, et al.. Report of a national conference on liver allocation in patients with hepatocellular carcinoma
in the United States. Liver Transpl. 2010;16:262–278.
24. Efremidis SC, Hytiroglou P. The multistep process of hepatocarcinogenesis in cirrhosis with imaging correlation. Eur Radiol. 2002;12:753–764.
25. Hayashi M, Matsui O, Ueda K, et al.. Correlation between the blood supply and grade of malignancy of hepatocellular nodules associated with liver cirrhosis: evaluation by CT during intraarterial injection of contrast medium. AJR Am J Roentgenol. 1999;172:969–976.
26. Catalano OA, Choy G, Zhu A, et al.. Differentiation of malignant thrombus from bland thrombus of the portal vein in patients with hepatocellular carcinoma
: application of diffusion-weighted MR imaging. Radiology. 2010;254:154–162.
27. Sakata J, Shirai Y, Wakai T, et al.. Preoperative predictors of vascular invasion in hepatocellular carcinoma
. Eur J Surg Oncol. 2008;34:900–905.
28. Takizawa D, Kakizaki S, Sohara N, et al.. Hepatocellular carcinoma
with portal vein tumor thrombosis: clinical characteristics, prognosis, and patient survival analysis. Dig Dis Sci. 2007;52:3290–3295.
29. Schneider G, Reimer P, Mamann A, et al.. Contrast agents in abdominal imaging: current and future directions. Top Magn Reson Imaging. 2005;16:107–124.
30. Mitchell DG, Bruix J, Sherman M, et al.. LI-RADS (Liver Imaging Reporting and Data System): summary, discussion, consensus of the LI-RADS Management Working Group and future directions. Hepatology. 2015;61:1056–1065.
31. Sano K, Ichikawa T, Motosugi U, et al.. Imaging study of early hepatocellular carcinoma
: usefulness of gadoxetic acid-enhanced MR imaging. Radiology. 2011;261:834–844.
32. Forner A, Vilana R, Ayuso C, et al.. Diagnosis
of hepatic nodules 20 mm or smaller in cirrhosis: prospective validation of the noninvasive diagnostic criteria for hepatocellular carcinoma
. Hepatology. 2008;47:97–104.
33. Kim TK, Lee KH, Jang HJ, et al.. Analysis of gadobenate dimeglumine-enhanced MR findings for characterizing small (1-2-cm) hepatic nodules in patients at high risk for hepatocellular carcinoma
. Radiology. 2011;259:730–738.
34. Rimola J, Forner A, Tremosini S, et al.. Non-invasive diagnosis
of hepatocellular carcinoma
</=2 cm in cirrhosis. Diagnostic accuracy assessing fat, capsule and signal intensity at dynamic MRI. J Hepatol. 2012;56:1317–1323.
35. Sangiovanni A, Manini MA, Iavarone M, et al.. The diagnostic and economic impact of contrast imaging techniques in the diagnosis
of small hepatocellular carcinoma
in cirrhosis. Gut. 2010;59:638–644.
36. Takayama Y, Nishie A, Nakayama T, et al.. Hypovascular hepatic nodule showing hypointensity in the hepatobiliary phase of gadoxetic acid-enhanced MRI in patients with chronic liver disease: prediction of malignant transformation. Eur J Radiol. 2012;81:3072–3078.
37. Kumada T, Toyoda H, Tada T, et al.. Evolution of hypointense hepatocellular nodules observed only in the hepatobiliary phase of gadoxetate disodium-enhanced MRI. AJR Am J Roentgenol. 2011;197:58–63.
38. Takechi M, Tsuda T, Yoshioka S, et al.. Risk of hypervascularization in small hypovascular hepatic nodules showing hypointense in the hepatobiliary phase of gadoxetic acid-enhanced MRI in patients with chronic liver disease. Jpn J Radiol. 2012;30:743–751.
39. Lim JH, Kim CK, Lee WJ, et al.. Detection of hepatocellular carcinomas and dysplastic nodules in cirrhotic livers: accuracy of helical CT in transplant patients. AJR Am J Roentgenol. 2000;175:693–698.
40. Burrel M, Llovet JM, Ayuso C, et al.. MRI angiography is superior to helical CT for detection of HCC prior to liver transplantation: an explant correlation. Hepatology. 2003;38:1034–1042.
41. Khan AS, Hussain HK, Johnson TD, et al.. Value of delayed hypointensity and delayed enhancing rim in magnetic resonance imaging diagnosis
of small hepatocellular carcinoma
in the cirrhotic liver. J Magn Reson Imaging. 2010;32:360–366.
42. Saito Y, Matsuzaki Y, Doi M, et al.. Multiple regression analysis for assessing the growth of small hepatocellular carcinoma
: the MIB-1 labeling index is the most effective parameter. J Gastroenterol. 1998;33:229–235.
43. Nakajima T, Moriguchi M, Mitsumoto Y, et al.. Simple tumor profile chart based on cell kinetic parameters and histologic grade is useful for estimating the natural growth rate of hepatocellular carcinoma
. Hum Pathol. 2002;33:92–99.