Differentiation of hepatic malignancies from hemangiomas and cysts by T2 relaxation times: early experience with multiply refocused four-echo imaging at 1.5 T

SSAT/AHPBA joint symposium on evaluation and treatment of benign liver lesions

BACKGROUND

Benign liver lesions are common incidental radiologic findings.

METHODS

Experts convened in 2011 at a Society for Surgery of the Alimentary Tract/ Americas Hepato-Pancreato-Biliary Association joint symposium to discuss the evaluation and treatment of benign liver lesions.

RESULTS

Most benign liver lesions can be accurately diagnosed with high-quality imaging, including ultrasonography, multiphase computed tomography, and magnetic resonance imaging, particularly with hepatocyte-specific contrast agents. Percutaneous biopsy is reserved for lesions that cannot be characterized radiographically, and its accuracy is improved with immunophenotypic markers. Hepatic cysts are the most commonly diagnosed benign liver lesions; these must be distinguished from malignant cystic lesions, which are rare. Among the solid benign liver lesions, hemangiomas and focal nodular hyperplasia seldom require treatment. In contrast, hepatocellular adenomas are associated with a risk for complications. A new classification system for hepatocellular adenomas based on genetic and phenotypic features can help guide patient care. In patients who are symptomatic or at risk for complications, multidisciplinary evaluation and treatment based on clinicopathologic, radiographic, and molecular analysis is needed.

CONCLUSIONS

Most benign liver lesions can be accurately diagnosed radiographically and do not require treatment. Treatment is necessary for patients with symptoms or at risk for complications.

MR T1 and T2 relaxations in cysts and abscesses measured by 1.5 T MRI

OBJECTIVES

The main objective of this study was to make a comparison between the relaxation rates in jaw cysts and abscesses. Such a comparison should provide quantitative information for MR image analysis.

METHODS

A phantom containing 20 odontogenic jaw cysts and 11 jaw abscesses was imaged with 1.5 T MR. T(1) measurements were performed by using a mixed sequence of inversion recovery and spin echo, while T(2) measurements were carried out by the Carr-Purcell Meiboom-Gill (CPMG) sequence. Cystic fluids and abscesses were compared statistically.

RESULTS

In cysts and abscesses, respectively, the mean 1/T(1) was 0.9355 s(-1) and 0.8245 s(-1) and the mean 1/T(2) was 2.4575 s(-1) and 4.7073 s(-1). The 1/T(2) in cysts was very highly significantly different from that in abscesses (p = 0.0001). Both T(1) and T(2) were linearly proportional to material contents. T(2) relaxivities [26.458 ml (g s)(-1) for abscesses and 21.455 ml (g s)(-1) for cysts] were higher than T(1) relaxivities [5.4766 ml (g s)(-1) for abscesses and 10.075 ml (g s)(-1) for cysts].

DISCUSSION

Present T(2) measurements differentiate cysts from abscesses with a confidence interval of 95%. Because in vivo and in vitro image contrasts are changed by the same parameters, the T(2) findings should present valuable information for in vivo MRI. Hence the significant difference and the relaxivities may provide quantitative information for clinicians and researchers making image analyses.

CONCLUSION

T(2) may differentiate cysts from abscesses. The difference in T(2) is related to the material content of samples.

T2 mapping from highly undersampled data by reconstruction of principal component coefficient maps using compressed sensing

Recently, there has been an increased interest in quantitative MR parameters to improve diagnosis and treatment. Parameter mapping requires multiple images acquired with different timings usually resulting in long acquisition times. While acquisition time can be reduced by acquiring undersampled data, obtaining accurate estimates of parameters from undersampled data is a challenging problem, in particular for structures with high spatial frequency content. In this work, principal component analysis is combined with a model-based algorithm to reconstruct maps of selected principal component coefficients from highly undersampled radial MRI data. This novel approach linearizes the cost function of the optimization problem yielding a more accurate and reliable estimation of MR parameter maps. The proposed algorithm--reconstruction of principal component coefficient maps using compressed sensing--is demonstrated in phantoms and in vivo and compared with two other algorithms previously developed for undersampled data.

Imaging of the liver for hepatocellular cancer

BACKGROUND

Imaging of the liver is a key component in the detection, diagnosis, management, and follow-up of patients with hepatocellular carcinoma.

METHODS

The author uses his own experience as well as a review of pertinent literature to describe the capabilities and the limitations of the principal currently available imaging techniques for the liver.

RESULTS

Ultrasound is widely available, but sensitivity and specificity for small nodules are limited. Computed tomography effectively demonstrates extrahepatic lesions and can differentiate between cysts or hemangiomas and hepatocellular carcinomas. Magnetic resonance imaging better characterizes hepatic lesions, but positron emission tomography is of limited value.

CONCLUSIONS

Cross-sectional imaging with ultrasound, CT, or MRI is critical for nodule characterization in the cirrhotic liver, surgical planning of HCC, and treatment response evaluation.

Management of an incidental liver mass

The increased use of sensitive imaging modalities has led to increased identification of the incidental liver mass (ILM). A combination of careful consideration of patient factors and imaging characteristics of the ILM enables clinicians to recommend a safe and efficient course of action. Using an algorithmic approach, this article includes pertinent clinical factors and the specific radiologic criteria of ILMs and discusses the indications for potential procedures. It is the aim of this article to assist with the development of an individualized strategy for each patient with an ILM.

Differentiation between hepatic haemangiomas and cysts with an inversion recovery single-shot turbo spin-echo (SSTSE) sequence using the TI nulling value of hepatic haemangioma with sensitivity encoding

OBJECTIVE

To evaluate the additional value of inversion recovery (IR) single-shot turbo spin-echo (SSTSE) imaging with sensitivity encoding (SENSE) using the inversion time (TI) value of hepatic haemangioma as a supplement to conventional T2-weighted turbo spin-echo (TSE) imaging for the discrimination of hepatic haemangiomas and cysts.

METHODS

A total of 134 lesions (77 hepatic haemangiomas, 57 hepatic cysts) in 59 patients were evaluated. Three readers evaluated these images and used a five-point scale to evaluate the lesion status. A receiver operating characteristic (ROC) analysis and 2 x 2 table analysis were used.

RESULTS

The ROC analysis for all the readers and all the cases revealed a significantly higher area under the curve (AUC) for the combination of moderately and heavily T2-weighted TSE with IR-SSTSE images (0.945) than for moderately and heavily T2-weighted TSE images alone (0.894) (P < 0.001). For the combination of T2-weighted TSE with IR-SSTSE versus T2-weighted TSE alone, the 2 x 2 table analysis revealed a higher true-positive rate; this difference was statistically significant (P < 0.0001).

CONCLUSION

The introduction of IR-SSTSE with SENSE sequences significantly improves the diagnostic accuracy of the differentiation of hepatic haemangioma and cysts while increasing the time required for routine abdominal imaging by only 20 s.

The diagnosis and management of benign hepatic tumors

Benign hepatic tumors include a broad spectrum of regenerative and true neoplastic processes. Because of advances in imaging studies such as computed tomography (CT) and magnetic resonance imaging (MRI) as well as progress in immunohistochemistry, accurate diagnosis can now be made in a large percentage of patients without surgical laparotomy or resection. This article will focus on the pathogenesis, diagnosis, and management of focal benign lesions of the liver. Many of these tumors present with typical features in various imaging studies. On occasions, biopsies are required and/or surgical removal is needed. The most common benign hepatic tumors include cavernous hemangioma, focal nodular hyperplasia, hepatic adenoma, and nodular regenerative hyperplasia. In the majority of cases of benign hepatic tumors, patients are asymptomatic, and no treatment is indicated. The main indication for treatment is the presence of significant clinical symptoms or suspicion of malignancy or fear of malignant transformation.

Processing of radial fast spin-echo data for obtaining T2 estimates from a single k-space data set

Radially acquired fast spin-echo data can be processed to obtain T2-weighted images and a T2 map from a single k-space data set. The general approach is to use data at a specific TE (or narrow TE range) in the center of k-space and data at other TE values in the outer part of k-space. With this method high-resolution T2-weighted images and T2 maps are obtained in a time efficient manner. The mixing of TE data, however, introduces errors in the T2-weighted images and T2 maps that affect the accuracy of the T2 estimates. In this work, various k-space data processing methods for reconstructing T2-weighted images and T2 maps from a single radial fast spin-echo k-space data set are analyzed in terms of the accuracy of T2 estimates. The analysis is focused on the effect of image artifacts, object dependency, and noise on the T2 estimates. Results are presented in computer-generated phantoms and in vivo.

Determinants of MR relaxation rates in jaw cysts: implications for diagnostic values of the relaxation times

OBJECTIVES

The MRI signal intensities and the relaxation times (T1 and T2) have already been used for comparisons among various types of cysts. The signal intensities (or T1 and T2) were suggested to be related to cyst content. However, such a relationship between the relaxation rates (1/T1 and 1/T2) and the aspirated content of jaw cysts was not given. Therefore, an investigation on the determinants of the relaxation rates in the jaw cysts may be helpful in elucidating the biological basis for the differences in relaxation rates among the categories of jaw cysts.

METHODS

The aspirated contents of 23 cysts (16 radicular and 7 haemorrhagic) were scanned using an MR imager operating at 1 T magnetic field strength. The viscosities and the dry-weight-to-water-weight ratios (Ms/Mw) were determined. Also, the mean values of cyst categories were compared using statistical analysis.

RESULTS

The 1/T1 had a moderate correlation with viscosity (0.66) and Ms/Mw (0.56), while the 1/T2 had a good correlation with viscosity (0.87) and Ms/Mw (0.82). The mean values of viscosity, Ms/Mw and relaxation rates in radicular cysts were significantly lower (P < 0.05) than those of haemorrhagic cysts.

CONCLUSIONS

The present data suggest that similarities and differences between relaxation rates of radicular and haemorrhagic cysts can be explained by the solid content and viscosity of cysts.

Magnetic resonance imaging of the liver: true fast imaging with steady state free precession sequence facilitates rapid and reliable distinction between hepatic hemangiomas and liver malignancies

OBJECTIVE

To assess the capability of the true fast imaging with steady state free precession (true FISP) sequence in the distinction between hemangiomas and malignant liver lesions.

METHODS

Sixty-eight patients with 45 hemangiomas and 51 liver malignancies were included in this study. A 1.5-T magnetic resonance system and a phased-array body coil were used. In addition to true FISP, breath-hold and fat-suppressed, T2-weighted, half-Fourier single-shot turbo spin echo (HASTE) and both unenhanced and gadolinium (Gd)-enhanced T1-weighted sequences were acquired. Two radiologists evaluated the magnetic resonance images independently in a blinded fashion. Interobserver variations with true FISP and HASTE were determined. Lesion contrast-to-noise ratios were calculated from true FISP images.

RESULTS

With true FISP, readers 1 and 2 made a correct distinction between hemangiomas and liver malignancies in 43 of 45 (96%) cases and 40 of 45 (89%) cases, respectively. The kappa value was 0.65. With HASTE, the success rates were 40 of 45 cases (89%) and 36 of 45 cases (80%), respectively, and the kappa value was 0.33. With a Gd-enhanced T1-weighted sequence, the correct classifications were 35 of 45 cases (78%) and 37 of 45 cases (82%), respectively. All hemangiomas appeared as bright and well-demarcated lesions on true FISP images. Malignant liver foci were heterogeneous with unsharp margins and nearly isointense relative to liver. The specificities of true FISP in lesion differentiation were 100% and 98% for readers 1 and 2, respectively. The mean contrast to noise ratio value of hemangiomas was 21.2 (standard deviation [SD] = 9.2), and that of malignant lesions was 4.9 (SD = 3.9). This difference was highly significant (P < 0.0001).

CONCLUSION

Noninvasive, rapid, and reliable differentiation between hemangiomas and malignant liver lesions is possible by using the true FISP sequence.

Evaluation of liver diffusion isotropy and characterization of focal hepatic lesions with two single-shot echo-planar MR imaging sequences: prospective study in 66 patients

PURPOSE

To (a) evaluate liver diffusion isotropy, (b) compare two diffusion-weighted magnetic resonance (MR) imaging sequences for the characterization of focal hepatic lesions by using two or four b values, and (c) determine an apparent diffusion coefficient (ADC) threshold value to differentiate benign from malignant lesions.

MATERIALS AND METHODS

Sixty-six patients were examined with two single-shot echo-planar diffusion-weighted MR sequences. In the first sequence, liver diffusion isotropy was evaluated by using diffusion gradients in three directions with two b values. In the second sequence, a unidirectional diffusion gradient was used with four b values. ADCs were measured in 43 patients with 52 focal hepatic lesions more than 1 cm in diameter and in 23 patients with 14 normal and nine cirrhotic livers and were compared by using nonparametric tests.

RESULTS

Diffusion in the liver parenchyma was isotropic. ADCs of focal hepatic lesions were significantly different between sequences (P <.01). The mean (+/- SD) ADCs in the first sequence were 0.94 x 10(-3) mm(2)/sec +/- 0.60 for metastases, 1.33 x 10(-3) mm(2)/sec +/- 0.13 for HCCs, 1.75 x 10(-3) mm(2)/sec +/- 0.46 for benign hepatocellular lesions, 2.95 x 10(-3) mm(2)/sec +/- 0.67 for hemangiomas, and 3.63 x 10(-3) mm(2)/sec +/- 0.56 for cysts. There was a significant difference between benign (2.45 x 10(-3) mm(2)/sec +/- 0.96, isotropic value) and malignant (1.08 x 10(-3) mm(2)/sec +/- 0.50) lesions (P <.01 for both sequences).

CONCLUSION

Diffusion-weighted MR imaging can help differentiate benign from malignant hepatic lesions. The use of two b values in one direction could be sufficient for the design of MR sequences in the liver.

Differentiation between hemangiomas and cysts of the liver with single-shot fast-spin echo image using short and long TE

PURPOSE

This study was designed to examine whether hepatic hemangioma can be differentiated from cyst by use of single-shot fast spin echo (SSFSE) sequence without contrast material.

METHODS

SSFSE images with a short TE (90 msec) and a long TE (600-700 msec) were obtained in 30 patients with cyst (n= 19) or hemangioma (n= 26) on a 1.5-T unit. In short TE, both cysts and hemangiomas showed well-defined hyperintense masses.

RESULTS

In long TE, however, cysts showed well-defined hyperintense masses, whereas hemangiomas were poorly defined faint masses with intermediate signal intensity. The contrast-to-noise ratio of hemangiomas and cysts were 3.23 and 52.9 (p < 0.001) in long TE; there was no overlap for the ranges, and the result did not depend on lesion size. All 45 lesions were differentiated correctly with these qualitative criteria.

CONCLUSIONS

SSFSE with short and long TE is useful in the evaluation of cyst and hemangioma of the liver.

Radial fast spin-echo method for T2-weighted imaging and T2 mapping of the liver

PURPOSE

To evaluate a multishot radial fast-spin echo (RAD-FSE) method developed to improve the quality of abdominal T2-weighted imaging as well as the characterization of focal liver lesions.

MATERIALS AND METHODS

The RAD-FSE sequence used in this work consisted of a preparatory period followed by a short echo train (ETL = 16). A novel radial k-space trajectory was used to minimize streaking artifacts due to T2 variations and motion. Small diffusion gradients (b = 1.2 mm/s(2)) were used to improve flow suppression. The quality of images obtained with RAD-FSE was compared to multishot 2DFT fast spin-echo (2DFT-FSE) and half-Fourier acquisition single-shot turbo-spin-echo (HASTE) images using data from 16 patients. A postprocessing algorithm was used to generate multiple high-resolution images (at different effective TE values) as well as a T2 map from a single RAD-FSE data set. The T2 maps were used to differentiate malignant from benign lesions for a set of 33 lesions ranging from 0.8-194 cm(3).

RESULTS

RAD-FSE produces high-resolution images of the liver in a breath-hold without the motion artifacts of 2DFT-FSE methods, and without the blurriness and loss of small lesion detectability of HASTE. The inclusion of diffusion weighting in RAD-FSE decreases the signal from blood in hepatic vessels, which improves lesion visualization. The T2 values obtained by postprocessing a single RAD-FSE data set can differentiate malignant from benign lesions. The mean T2 values obtained for malignancies, hemangiomas, and cysts are 108 +/- 30 msec, 240 +/- 14 msec, and 572 +/- 334 msec, respectively.

CONCLUSION

These results indicate that RAD-FSE produces abdominal images of higher quality than 2DFT-FSE and HASTE. In addition, lesions can be characterized using T2 maps generated from a single RAD-FSE data set.

Benign liver neoplasms

A variety of benign focal liver lesions are easily characterized with currently available imaging techniques and contrast agents. The most common benign liver lesions, such as hemangioma, bile duct cyst, and FNH, reveal characteristic cross-sectional imaging features that allow an accurate diagnosis. For atypical variants and more uncommon lesions, including HCA, angiomyelioma, infantile hemagioendothelioma, and mesenchymal hamartoma, integration of clinical data can often help in the interpretation of imaging studies. Finally, for the remaining lesions, such as hepatic adenomatosis, the imaging findings may not be specific enough to negate the need for a tissue biopsy.