Dilatation of the intrahepatic bile ducts associated with benign liver lesions: an unusual finding

In three patients presenting different types of liver lesions, including isolated cyst, focal nodular hyperplasia (FNH), and hemangioma, intrahepatic bile duct dilatation was observed on US and CT. Final diagnosis was obtained by surgery in two cases (cyst and FNH) and by 1-year follow-up in one patient presenting an isolated hemangioma. The only common characteristic in our three cases was that lesions were present in segment four according to Couinaud's classification, at the level of the transverse fissure, suggesting that a space-occupying lesion at this site may cause compression of the common hepatic duct and right or left intrahepatic bile ducts. Our report indicates that compression may occur even with lesion of moderate size (35–40 mm in diameter). A benign liver lesion may cause a bile duct dilatation, particularly if located in segment 4, close to the hilum. Awareness of this possibility is important to avoid unnecessary invasive diagnostic procedures, particularly when all imaging criteria are consistent with a benign lesion. Electronic Publication     

Benign focal liver lesions: discrimination from malignant mimickers

Focal lesions of the liver often have various imaging characteristics which may be interpreted as either benign or malignant. Understanding the underlying pathophysiology of these liver lesions may lead to characteristic imaging manifestations, which direct the radiologist to the diagnosis. Benign lesions include congenital hepatic cyst, autosomal dominant polycystic disease, hemangioma, focal nodular hyperplasia (FNH), hepatic adenoma, inflammatory pseudotumor, peliosis hepatis, focal fatty infiltration, hamartoma, and infectious processes such as hepatic abscess, echinococcal cyst, and candidiasis. Characteristic imaging features, clinical symptoms, and treatment/prognosis will be discussed. Emphasis will be placed on key reliable features of each disease to develop a method of discriminating these lesions from other benign and malignant disorders.     

Prevalence of hepatic hemangioma in patients with focal nodular hyperplasia: MR imaging analysis

PURPOSE: To evaluate the prevalence of hepatic hemangioma in a large population of patients with focal nodular hyperplasia (FNH) and to determine if the prevalence is higher than that in patients with other hepatic masses. MATERIALS AND METHODS: Over a period of 40 months, 247 patients with one or more hepatic masses underwent magnetic resonance (MR) imaging at our institution and met the inclusion criteria for this study. One hundred forty-eight patients received a diagnosis of FNH (study group). Ninety-nine patients had a lesion other than FNH and had no history of chronic liver disease (control group). Imaging findings of the main lesion and presence of associated hemangioma were investigated. chi2 analysis was used to determine if there was a statistically significant difference in the two groups regarding the number of patients with associated masses. RESULTS: Twenty-nine of 148 patients (20%) in the study group had FNH with one or more associated hepatic hemangiomas. Among the 99 patients in the control group, nine (9%) had an associated hemangioma. The prevalence of hemangioma was significantly higher (P <.02) in the study group than in the control group. The prevalence of hemangioma in patients with a solitary FNH lesion compared with that in patients with multiple FNH lesions was not significantly different. CONCLUSION: Patients with FNH are more likely to have an associated hepatic hemangioma than are those with another type of focal hepatic mass.     

Focal nodular hyperplasia (FNH): a potential cause of false-positive positron emission tomography

Positron emission tomography (PET) using F-18 fluorodeoxyglucose (FDG) has been proven to be a useful tool in the differential diagnosis of liver tumors. Focal nodular hyperplasia (FNH) is an uncommon benign liver lesion, which can be difficult to differentiate from other benign and malignant liver pathologies. FDG PET imaging usually shows uptake similar or even decreased compared to that of the normal liver. We describe a hypermetabolic FNH lesion in a patient with a history of breast cancer. Computed tomography scan, ultrasonography (US), and magnetic resonance imaging were negative. The lesion was resected, and histologic findings were consistent with FNH.     

探讨双b值对比法在肝脏局灶性病变诊断中的作用

目的：探讨双b值对比法在肝脏局灶性病变诊断中的价值.方法：回顾性分析108例行肝脏MRI平扫及动态增强扫描患者的病例（肝癌44例.转移瘤4例,血管瘤40例,囊肿18例,局灶性结节增生2例）,分为良恶性两组（恶性组48例,A组;良性组60例,B组）.均行b值为500和800 s/mm2的DWI扫描,测量肝脏病变的信号强度值,计算相对信号比R500、R800（R=病灶信号/背景信号）,观察不同b值时病变的信号变化情况△R（△R=R800-R500）.采用X2检验分析采用△R值诊断良恶性病变与病理结果的一致性.结果：△R≥0.25有42例（PA组）,△R〈0.25（PB组）有66例,其中PA中有4个为良性病变（均为血管瘤）,PB中有10个为恶性病变（2个转移瘤,8个肝癌）,差异无统计学意义（X2=1.786,P〉0.05）.结论：双b值对比法在肝脏局灶性病变的定性诊断中具有可行性.     

Magnetization transfer contrast of hepatic lesions in breath-hold gradient-echo images of different T1 weighting

Seventeen patients with hepatic lesions [six metastases from colon, breast, and gallbladder carcinoma; one gallbladder carcinoma; five hepatocellular carcinoma; three focal nodular hyperplasia (FNH); one adenoma; and one cyst] were examined by MR breath-hold two-dimensional gradient-echo imaging to assess the potential of magnetization transfer contrast (MTC) for improved conspicuity and classification. Imaging sequences were applied with and without irradiation of off-resonant radiofrequency (RF) prepulses, but other parameters were unchanged. Therefore, quantitative assessment of MTC could be performed. In contrast to former examinations of other researchers, no significant difference of MTC was found between malignant liver lesions and benign lesions as FNH or adenoma. MTC might provide differentiation between hemangioma and cysts versus solid tumors, but MTC is not capable of distinguishing benign and malignant types of solid liver tumors. Effects of unchanged MTC prepulses on signal intensity of normal liver tissue and most lesions were more pronounced for nearly proton density-weighted fast low-angle shot (FLASH) images than for T1-weighted FLASH images, obtained by using higher excitation flip angles. Liver-to-lesion contrast could not be improved clearly by MTC prepulses. The contrast between liver and lesions in the gradient-echo breath-hold images was compared with standard T1- and T2-weighted spin-echo images. Liver-to-lesion contrast in the breath-hold images was found to be inferior to T2-weighted spin-echo images in 14 of 17 cases. Lesion conspicuity in regions near the diaphragm was better in breath-hold images, because problems with marked breathing motion (as in standard imaging) could be avoided.     

Hepatic neoplasia: Selected clinical aspects

Benign liver tumors are relatively uncommon and, even when large enough to be symptomatic, they usually remain undiagnosed prior to exploratory laparotomy. Hemangiomas constitute the majority of benign hepatic neoplasms and are 9 times as frequent in females as in males. Most are asymptomatic but abdominal swelling, a mass, or symptoms due to compression of adjacent organs may occur and abdominal hemorrhage is reported in 4.5% of patients. Hepatic hemangioma may produce a large arteriovenous communication serious enough to cause heart failure. Recently an increased frequency of liver tumors, mostly adenomas, has been noted in women taking oral contraceptives (OCs); the cause has been attributed to estrogens. The exact incidence is unknown but believed to be low. It is most common in women in their late 20s who have been on OCs for 7 years or more. The tumor occasionally completely regresses on withdrawal of the OCs. The tumor may be discovered incidentally at laparotomy or may manifest inself by pain, a palpable mass, or catastrophic hemoperitoneum. Hepatic adenoma is usually a solitary lesion and infrequently degenerates into malignancy. Differential diagnosis includes chronic gall bladder disease and peptic ulcer. Focal nodular hyperplasia (FNH) is apparently much less frequently related to OC use and is less likely to bleed seriously than adenoma. Hepatic chemistry is usually normal in adenoma and FNH, but slight increases in serum bilirubin, serum alkaline phosphatase, and serum transaminase may occur. Primary liver cancer (hepatocellular carcinoma or hepatoma) is mostly a disease of males and in the US and Western Europe seldom develops before age 40. Fibrolamellar carcinoma, which characteristically develops in adolescents and young adults, occurs with equal sex incidence. Doubt has been expressed about its relationship to OCs. In the US about 75% of primary hepatocellular carcinomas are associated with cirrhosis, and about 5% of cirrhosis cases develop primary liver cancer. Clinical manifestations of hepatoma have been divided into 5 groups: frank cancer (62.7%), acute abdominal cancer (8%), febrile cancer (8%), occult cancer (16%), and metastatic cancer (5%). Detection of large amounts of alpha fetoprotein has proven useful in diagnosis of hepatocellular carcinoma, but values may be negative in OC users. It has been estimated that 1/3 to 1/2 of all malignant tumors eventually metastasize to the liver.     

Discrimination of benign from malignant hepatic lesions based on their T2-relaxation times calculated from moderately T2-weighted turbo SE sequence

The differentiation of hemangioma from other hepatic neoplasms using MRI usually relies on the evaluation of heavily T2-weighted images. The aim of this study was to assess the value of T2-relaxation times calculated from moderately T2-weighted turbo spin-echo (TSE) sequence in characterization of focal hepatic lesions, including hepatic malignancies, focal nodular hyperplasia (FNH), hemangioma, and cyst. Fifty-two patients with 114 proven lesions (61 malignant masses, 6 focal nodular hyperplasias, 28 hemangiomas, 19 cystic lesions) were examined on 1.5-T system using a double-echo TSE sequence (TR=1800 ms; TE(eff) 1=40 ms; TE(eff) 2=120 ms). Signal intensities (SI) of the liver as well as SI of all lesions were measured, and then the T2-relaxation times were calculated. The mean T2 time for the liver was 54 ms (+/-8 ms), for FNH 66 ms (+/-7 ms), for malignant hepatic lesions 85 ms (+/-17 ms), for hemangiomas 155 ms (+/-35 ms), and for cystic lesions 583 ms (+/-369) ms. Most malignant hepatic lesions were best differentiated between the thresholds of 67 and 116 ms, generating a sensitivity of 90% and a specificity of 94%. There were six false-negative diagnoses of malignant tumor and three false-positive cases (two hemangiomas and one FNH). Calculation of the T2-relaxation times obtained from the double-echo TSE sequence with moderate T2-weighting allowed differentiation between malignant and benign hepatic lesions with high sensitivity and specificity.     

Characterization of unifocal liver lesions with pulse inversion harmonic imaging after Levovist injection: preliminary results

The aim of this study was to evaluate capabilities of pulse inversion harmonic imaging (PIHI) in characterization of unifocal liver lesions. We evaluated with PIHI (HDI5000, ATL, Bothell, Wash.) and spiral CT 46 consecutive patients with a single liver lesion identified by fundamental US [7 hepatocellular carcinomas (HCC), 2 cholangiocarcinomas, 7 focal nodular hyperplasias (FNH), 17 hemangiomas and 13 metastases]. The PIHI was performed before and 30 s, 2 and 4 min after bolus administration of Levovist (2.5 g, 300 mg/ml). Scans were digitally stored and reviewed using a dedicated software. Hepatocellular carcinoma was hyperechoic on 30-s scan, and hypoechoic (n = 5) or isoechoic (n = 2) on 2-min scan. Cholangiocarcinoma had inhomogeneous persistent enhancement. Focal nodular hyperplasia was hyperechoic (n = 5) or isoechoic (n = 2) on 30-s scan, hyperechoic (n = 4), isoechoic (n = 2) or slightly hypoechoic (n = 1) on 2-min scan. Large hemangioma revealed peripheral enhancement on 30-s scan which extended centripetally on 2-min scan. Small hemangioma appeared isoechoic on 2-min scan in all but two cases in which they were hypoechoic on 2-min scans and hyperechoic on 4-min scan. Metastasis was hypoechoic on all scans, 70% with rim enhancement. Similar changes in enhancement pattern have been observed at spiral CT. The 30-s and the 2-min scans revealed a conclusive importance in characterization of HCC, cholangiocarcinoma, and large hemangioma. The 2-min scan often furnished enough information for characterization of small hemangioma and metastasis. The 4-min scan allowed characterization of two hemangiomas which appeared hypoechoic on 2-min scans. In the other cases it did not provide further information. Diagnosis of FNH is usually reached with Colour Doppler US; PIHI should be used when colour Doppler is biased by artefacts or when colour Doppler findings are not characteristic. Our results seem to show that PIHI could be a valuable alternative diagnostic approach to spiral CT for unifocal liver lesion characterization. This hypothesis needs to be confirmed with an increased number of lesions.     

Telangiectatic focal nodular hyperplasia: US,CT, and MR imaging findings with histopathologic correlation in 13 cases

PURPOSE: To review the ultrasonographic (US), computed tomographic (CT), and magnetic resonance (MR) imaging findings in 13 patients with telangiectatic focal nodular hyperplasia (FNH) and to compare imaging features with histopathologic results from resected specimens. MATERIALS AND METHODS: US, helical multiphasic CT, and MR images in 13 patients with pathologically proven telangiectatic FNH were reviewed retrospectively. Two abdominal radiologists evaluated lesions for number, size, heterogeneity, surface characteristics, presence of a central scar, presence of a pseudocapsule, US appearance, attenuation at CT, signal intensity at MR imaging, and presence of associated lesions. Imaging and pathologic findings were compared. RESULTS: Sixty-one lesions (5-140 mm in diameter) were seen at imaging. Lesions were multiple in eight of 13 (62%) patients. Imaging characteristics were heterogeneity in 26 of 61 lesions (43%), well-defined margins in 43 of 61 (70%), lack of a central scar in 56 of 61 (92%), presence of a pseudocapsule in three of 61 (5%), hyperintensity on T1-weighted MR images in 17 of 32 (53%), strong hyperintensity on T2-weighted MR images in 24 of 54 (44%), and persistent enhancement on delayed contrast material-enhanced CT or T1-weighted MR images in 23 of 38 (61%). No specific US pattern was noted. Two patients had additional lesions: One had classic FNH, and the other had a cavernous hemangioma. Hyperintensity on T1-weighted MR images was due to sinusoidal dilatation. Hyperintensity on T2-weighted MR images correlated well with the presence of inflammation. CONCLUSION: Telangiectatic FNH differs from typical FNH at imaging: Atypical FNH features often observed with telangiectatic FNH are lack of a central scar, lesion heterogeneity, hyperintensity on T1-weighted MR images, strong hyperintensity on T2-weighted MR images, and persistent contrast enhancement on delayed contrast-enhanced CT or T1-weighted MR images.     

Liver metastases on serial contrast-enhanced multidetector computed tomography examinations: was the detection possible on previous examinations?

OBJECTIVE: To verify the earliest detectability of liver metastases in patients who underwent serial multidetector computed tomography (MDCT) examinations. METHODS: We selected 12 patients with known primary cancer who underwent 4 or more contrast-enhanced, 4-detector MDCTs. When metastases had been reported, an evaluation of the preceding MDCT was done to define whether the lesion was detectable, detectable only by minimal signs, undetectable, or detected but misdiagnosed as a benign lesion (MBL). RESULTS: Eighty-eight lesions were analyzed. Evaluating the preceding examination, we defined detectable (n=8), detectable only by minimal signs (n=5), undetectable (n=74), and MBL (n=1). The group with minimal signs was composed of 4 small hypodense foci and 1 calcification. The MBL was a non-Hodgkin lesion first misdiagnosed as a hemangioma. CONCLUSION: Approximately 15% of liver metastases were prospectively missed, 9% of them being retrospectively detectable, 6% being retrospectively visible as minimal signs, whereas only 1% of liver metastases were misdiagnosed as a benign lesion.     

DWI技术在肝脏占位性病变鉴别诊断中的价值

目的:探讨不同b值时表现扩散系数(ADC)值、指数化表现扩散系数(eADC)值、灌注ADC值和eADC值在肝脏常见占位性病变鉴别诊断中的价值。方法:对73例肝脏恶性肿瘤及40例良性占位性病变行磁共振扩散加权成像(DWI),并计算低、中和高b值时病灶的ADC值和eADC值及灌注ADC和eADC值。分析不同b值时各测量参数在不同病变之间的差异。结果:低、中b值时,在血管瘤与其他病变之间,局灶性结节增生(FNH)与转移瘤、肝脓肿之间,肝细胞癌(HCC)与肝脓肿之间,ADC和eADC值差异均具有统计学意义(P<0.05)。当高b值时,在血管瘤与HCC、肝内胆管细胞癌(CCC)及FNH之间,FNH与转移瘤、CCC、肝脓肿之间,肝脓肿与CCC之间,及转移瘤与HCC、CCC之间,ADC值和eADC值差异均具有统计学意义(P<0.05)。在血管瘤与其他病变之间,FNH与转移瘤、肝脓肿、CCC之间,肝脓肿与HCC之间灌注ADC和eADC值差异也具有统计学意义(P<0.05),灌注eADC值在HCC和CCC之间差异也具有统计学意义(P<0.05)。结论:灌注ADC和eADC值,恶性肿瘤采用高b值、良性病变采用低中b值得到ADC和eADC值有助于肝脏占位性病变的鉴别诊断。     

Differential diagnosis of focal nodular hyperplasia with quantitative parametric analysis in contrast-enhanced sonography

OBJECTIVES: We investigated the potential of quantitative parametric analysis in the differential diagnosis of focal nodular hyperplasia (FNH) from other hypervascularized liver focal lesions. MATERIALS AND METHODS: Eighty-five focal liver lesions (in 83 patients) were explored using contrast-enhanced ultrasound (SonoVue and Cadence Contrast Pulse Sequencing) consisting of typical FNH (n=52), hepatocellular carcinoma (n=11), hemangioma with high flow (n=8), hypervascular metastases (n=10), and hepatocellular adenoma (n=4). QontraXt software (AMID, Italy) was used here to estimate the following parameters: maximum peak value, Tr (time corresponding to time for obtaining 63% of the plateau), beta parameter corresponding to the exponential factor, and slope corresponding to the tangent value of the first phase of enhancement. These parameters were obtained from the time-intensity curves derived from the enhancement observed in 2 regions of interest corresponding, respectively, to the whole lesion and the central region only. RESULTS: A significant statistical difference (P<0.05) was found in the values of Tr, beta, and slope between FNH and other hypervascularized lesions on both the whole lesion and central region. Among these parameters, slope appeared as the most valuable whatever the region of interest, ie, central or whole lesion (P<0.01). Central region was more accurate in the differentiation of FNH and concordant with visual characterization. CONCLUSION: Quantitative parametric curve analysis of the different hypervascularized lesions confirms the depiction of the central artery in FNH and thus could help in differentiating this specific focal liver lesion from the others.     

Hepatic focal nodular hyperplasia: specific findings at dynamic contrast-enhanced US with carbon dioxide microbubbles.

Dynamic contrast material-enhanced ultrasonography (US) with intraarterial infusion of carbon dioxide microbubbles was performed for four cases of histologically proved focal nodular hyperplasia (FNH) in four patients and for 167 cases of various hepatic nodules in 144 patients. No complications due to dynamic US were observed in any of the 148 patients. All FNH nodules were less than 3 cm in diameter. Consistent specific findings of FNH were not obtained with US, computed tomography, magnetic resonance imaging, radiocolloid scanning, or angiography in the four cases of FNH. In contrast, the characteristic vascular pattern (ie, early central hypervascular supply with centrifugal filling to the periphery at the arterial phase and a uniform or lobulated dense stain at the capillary phase) was observed in all four cases of FNH with dynamic US. This vascular pattern demonstrated in FNH with dynamic US was not seen in any of the 167 hepatic nodules, including 44 small hepatocellular carcinomas less than 3 cm in diameter. Therefore, the newly developed, dynamic contrast-enhanced US technique seems to be extremely sensitive and specific for diagnosing FNH and is useful in the differentiation of FNH from other hepatic tumors, especially hepatocellular carcinoma.     

Quantitative dynamic contrast-enhanced sonography of hepatic tumors

Liver tumors are defined using quantitative dynamic contrast-enhanced ultrasound compared to histological diagnosis, respectively, long-term follow-ups. Forty-two focal liver lesions in 39 patients were examined by contrast harmonic imaging over a period of 2 min after bolus injection of 10-ml galactose-based contrast agent. Vascular enhancement was quantified by using a dedicated software that allowed us to place representative regions of interest (ROI) in the center of the lesion, in the complete lesion, in regular liver parenchyma and in representative liver vessels (artery, vein and portal vein). Peak enhancement was judged to be either in the arterial, portal venous or in the late phase of liver perfusion. The lesion was described as hypovascular, isovascular and hypervascular compared to liver parenchyma. Contrast uptake was described as centrifugal or centripetal and peripheral or homogenous, respectively. Characterization of the lesions was performed unenhanced and after contrast by four independent specialists unaware of histology. Diagnosis of malignancy was evaluated by using a receiver operating characteristic (ROC) analysis, also overall accuracy, average sensitivity, specificity and negative and positive predictive values were calculated. Interobserver agreement was defined by the Kappa statistics. Histologic examination revealed 29 malignant [hepatocellular carcinoma (HCC), n=11; cholangiocellular carcinoma (CCC), n=1; lymphoma, n=1; metastases, n=16)] and 7 benign [hemangioma, n=1; focal nodular hyperplasia (FNH), n=4, adenoma, n=2)] lesions. Six benign lesions (hemangioma n=1; FNH n=5) were proved by long-term follow-up. ROC analysis regarding the diagnosis of malignancy showed values from 0.43 to 0.62 (mean 0.57) before and from 0.70 to 0.80 (mean 0.75) after contrast agent, respectively. The average values for sensitivity, specificity, accuracy and negative and positive predictive values were 66, 26, 62, 45 and 73% unenhanced and 83, 49, 73, 65 and 82% after contrast, respectively. The interobserver agreement was 0.54 and 0.65 for unenhanced and enhanced examinations, respectively. Quantitative dynamic contrast-enhanced sonography improves the diagnosis of malignancy in liver lesions.     

几种肝脏病变的MRI菲立磁增强表现及诊断价值

目的:探讨菲立磁(SPIO)增强在肝脏病变中的诊断价值。方法:16例患者行SPIO增强的MRI检查,其中6例为海绵状血管瘤,不均匀脂肪肝4例,肝细胞癌(HCC)3例,局灶结节增生(FNH)、肝腺瘤及肝转移瘤各1例。结果:与MRI平扫相比,肝内海绵状血管瘤在SPIO增强后的T1WI上信号明显增高,而在T2WI上信号强度有轻度下降;在不均匀脂肪肝的病例中,增强后兴趣区与周围肝组织的信号强度对比关系延续了增强前的表现;肝腺瘤和FNH在增强后的T2WI上有明显的信号强度下降:HCC在增强后的T2WI上常无明显信号下降,而在T1WI上却有轻度信号升高;转移瘤在T2WI上无信号下降。结论:SPIO强化MRI对肝内病变的良恶性鉴别诊断有一定的帮助。     

131I治疗先天性甲状腺单叶缺如合并Graves病甲巯咪唑治疗后复发一例

笔者报道了1例131I治疗先天性甲状腺单叶缺如合并Graves病甲巯咪唑治疗后复发的病例,从临床症状、血清学化验、甲状腺超声及颈胸部SPECT/CT融合显像等方面分析该病例特点,并通过文献复习加深了对先天性甲状腺单叶缺如合并Graves病的认识。甲状腺单叶缺如合并Graves病患者在规律口服抗甲状腺药物治疗后出现复发,应尽早选择131I治疗。131I治疗仍是先天性甲状腺单叶缺如伴Graves病甲巯咪唑治疗后复发患者较为理想的方法。     

肝脏局灶结节性增生的CT特征

目的探讨肝脏局灶结节性增生(Focalnodularhyperplasia,FNH)这种较少见的肝细胞良性占位性病变的CT特征。材料和方法自1990年至1999年期间我院CT室诊为肝脏占位病变的900余例中,将有手术切除标本并经病理证实的肝脏FNH8例的临床及CT片逐例进行分析和归纳,重点讨论其鉴别要点。结果8例均为单发病灶。发生在肝左叶3例,肝右叶5例。其病灶大小在5cm以下者4例;5cm以上者4例(其中10cm以上者2例)。7例为类圆形,1例不规则形。平扫时7例为低密度,1例为等密度(有脂肪肝影响)。均行增强扫描,2例动脉期有显著增强,5例门静脉期扫描仍有显著增强,8例延迟3分钟后扫描为等密度。7例病灶内均可见裂隙或小片状密度减低影。1例为均匀密度影。结论利用CT扫描,尤其是多时相螺旋CT扫描,可以对肝脏局灶结节性增生(FNH)进行明确的术前诊断。     

Radiologists' performance in the diagnosis of liver tumors with central scars by using specific CT criteria

PURPOSE: To determine the performance of radiologists with differing levels of expertise in the diagnosis of the most common types of liver tumors with central scars (ie, focal nodular hyperplasia [FNH], fibrolamellar hepatocellular carcinoma [HCC], and large hepatic hemangioma) by using specific computed tomographic (CT) findings. MATERIALS AND METHODS: Review of medical records at the University of Pittsburgh Medical Center identified patients with a total of 64 liver tumors that had central scars-including 29 cases of FNH, 20 fibrolamellar HCCs, and 15 large (>3.5 cm in diameter) hemangiomas-and with CT scans available for review. Retrospective review of these scans was performed individually by six radiologists who were blinded to the diagnosis, including two faculty abdominal radiologists, one abdominal imaging fellow, and three radiology residents. Individual performance was evaluated by means of receiver operating characteristic analysis, and interobserver agreement was measured by using the Cronbach alpha. Individual CT findings that may allow differentiation of tumor types were identified with the Kruskal-Wallis test. RESULTS: CT allowed good to excellent interobserver agreement in the diagnosis of tumor type and in recognition of differential findings among the three types. The individual accuracy of diagnosis was very good, with the average area under the receiver operating characteristic curve ranging from 0.81 to 0.90. Although the faculty radiologists performed the best, the differences in performance between the subgroups of readers and the levels of confidence in diagnosis were not statistically significant. The diagnosis of fibrolamellar HCC was the most accurate and had the highest sensitivity, followed by FNH and large hemangioma. Clinical and CT findings that were found to be statistically significant in differentiating tumor types were patient age and sex, tumor size larger than 10 cm, width of tumor scars, invasion of vessels, nodular centripetal enhancement, marked hyperattenuation on arterial phase images, lymphadenopathy, heterogeneity, extrahepatic metastases, surface lobulation, calcification, and isoattenuation with liver tissue on portal venous phase images. CONCLUSION: CT allows accurate differentiation of the most common types of liver tumors with central scars, including FNH, fibrolamellar HCC, and large hemangioma.