MR Imaging of hepatocellular carcinoma in the cirrhotic liver: challenges and controversies.

The incidence of hepatocellular carcinoma (HCC) is expected to increase in the next 2 decades, largely due to hepatitis C infection and secondary cirrhosis. HCC is being detected at an earlier stage owing to the implementation of screening programs. Biopsy is no longer required prior to treatment, and diagnosis of HCC is heavily dependent on imaging characteristics. The most recent recommendations by the American Association for the Study of Liver Diseases (AASLD) state that a diagnosis of HCC can be made if a mass larger than 2 cm shows typical features of HCC (hypervascularity in the arterial phase and washout in the venous phase) at contrast material-enhanced computed tomography or magnetic resonance (MR) imaging or if a mass measuring 1-2 cm shows these features at both modalities. There is an ever-increasing demand on radiologists to detect smaller tumors, when curative therapies are most effective. However, the major difficulty in imaging cirrhosis is the characterization of hypervascular nodules smaller than 2 cm, which often have nonspecific imaging characteristics. The authors present a review of the MR imaging and pathologic features of regenerative nodules and dysplastic nodules and focus on HCC in the cirrhotic liver, with particular reference to small tumors and lesions that may mimic HCC. The authors also review the sensitivity of MR imaging for the detection of these tumors and discuss the staging of HCC and the treatment options in the context of the guidelines of the AASLD and the imaging criteria required by the United Network for Organ Sharing for transplantation. MR findings following ablation and chemoembolization are also reviewed.

Full-dose gemcitabine and concurrent radiotherapy for unresectable pancreatic cancer.

PURPOSE: Full-dose gemcitabine and concurrent radiotherapy is a promising treatment approach in unresectable pancreatic cancer. This study was conducted to assess the pattern of failure and toxicity associated with the use of conformal treatment volumes, omitting prophylactic lymph node irradiation. METHODS AND MATERIALS: Seventy-four patients with locally advanced pancreatic cancer were treated between 1997 and 2005 with full-dose (1000 mg/m(2), Days 1, 8, and 15) gemcitabine and concurrent radiotherapy (36 Gy [median] in 15 daily fractions). The planning target volume (PTV) was limited to the gross tumor volume (GTV) plus 1-cm margin. Patient computed tomography (CT) scans were systematically reviewed to determine the pattern of failure. Kaplan-Meier and Cox-regression models were used to analyze freedom from local progression (FFLP), distant failure, overall survival (OS), and toxicity. RESULTS: With a median follow-up of 10.6 months (20.6 months in living patients), the 1-year and 2-year FFLP rates were 64% and 38%, respectively. Four patients (5%) failed in the peripancreatic lymph nodes (3 in-field and 1 marginal failure). Median OS was 11.2 months. Analyzed as a time-dependent covariate, local failure was a significant predictor of OS (p = 0.0074). Sixteen patients (22%) had significant gastrointestinal (GI) toxicity (> or = Grade 3). PTV correlated with significant GI toxicity (p = 0.007). CONCLUSIONS: Freedom from local progression in unresectable pancreatic cancer is suboptimal. In conjunction with full-dose gemcitabine, the use of conformal fields encompassing only the GTV helps reduce toxicity and does not result in marginal failures. Our findings provide rationale for intensification of local therapy in conjunction with more effective systemic therapy.

MRI of sonographically indeterminate adnexal masses.

OBJECTIVE: The purpose of this study was to assess the ability of MRI to characterize sonographically indeterminate adnexal masses and to define the sonographic features contributing to indeterminate diagnoses. MATERIALS AND METHODS: Two blinded radiologists retrospectively reviewed the MRI examinations of 87 patients with 95 sonographically indeterminate adnexal masses. Reviewers determined the origin of a mass, its tissue content (cystic, solid, complex cystic, or cystic and solid), tissue characteristics (fat, blood, fibrous, or leiomyomatous), and benignity versus malignancy. Sonograms were reviewed by three reviewers to determine the origin of a mass, its tissue content, and reasons for an indeterminate diagnosis. Sensitivity and specificity of MRI were calculated, and agreement of sonography and MRI with the final diagnosis was determined using kappa statistics. The final diagnosis was determined by histopathology, surgical findings, or imaging or clinical follow-up. RESULTS: The sensitivity of MRI for identifying malignancy (n = 5) was 100% and its specificity for benignity (n = 90) was 94%. Excellent agreement was seen between MRI and the final diagnosis for determining the origin (kappa = 0.93), tissue content (kappa = 0.98), and tissue characteristics (kappa = 0.91) of a mass. Sonography had poor agreement with the final diagnosis for the origin (kappa = 0.19) and tissue content (kappa = 0.33) of a mass. The main reasons for indeterminate sonographic diagnoses were the inability to determine origin because of location and large mass size and the appearances of purely solid or complex cystic masses. CONCLUSION: Sonographically indeterminate adnexal masses of uncertain origin and solid or complex cystic content benefit from further evaluation with MRI, which is highly accurate for identifying the origin of a mass and characterizing its tissue content, obviating surgery.

MR imaging of the gallbladder.

T2-weighted imaging and MRCP, which have high sensitivity to edema and fluid, are paramount in the evaluation of certain gallbladder diseases, such as cholelithiasis, cholecystitis, adenomyomatosis, and cystic duct abnormalities. Dynamic gadolinium-enhanced MR imaging has the potential to differentiate among the many nonspecific-appearing lesions involving the gallbladder. MR imaging may not yet replace ultrasound as the workhorse of acute gallbladder imaging. Currently, MRCP is an ideal complementary study to inconclusive sonographic studies and can help plan surgical intervention in the setting of acute cholecystitis. Further investigation of hepatobiliary contrast agents, however, may reveal that MR imaging may be considered as first-line imaging in the acute setting.

Characterization of pancreatic tumor motion using cine MRI: surrogates for tumor position should be used with caution.

PURPOSE: Our current understanding of intrafraction pancreatic tumor motion due to respiration is limited. In this study, we characterized pancreatic tumor motion and evaluated the application of several radiotherapy motion management strategies. METHODS AND MATERIALS: Seventeen patients with unresectable pancreatic cancer were enrolled in a prospective internal review board-approved study and imaged during shallow free-breathing using cine MRI on a 3T scanner. Tumor borders were agreed on by a radiation oncologist and an abdominal MRI radiologist. Tumor motion and correlation with the potential surrogates of the diaphragm and abdominal wall were assessed. These data were also used to evaluate planning target volume margin construction, respiratory gating, and four-dimensional treatment planning for pancreatic tumors. RESULTS: Tumor borders moved much more than expected. To provide 99% geometric coverage, margins of 20 mm inferiorly, 10 mm anteriorly, 7 mm superiorly, and 4 mm posteriorly are required. Tumor position correlated poorly with diaphragm and abdominal wall position, with patient-level Pearson correlation coefficients of -0.18-0.43. Sensitivity and specificity of gating with these surrogates was also poor, at 53%-68%, with overall error of 35%-38%, suggesting that the tumor may be underdosed and normal tissues overdosed. CONCLUSIONS: Motion of pancreatic tumor borders is highly variable between patients and larger than expected. There is substantial deformation with breathing, and tumor border position does not correlate well with abdominal wall or diaphragmatic position. Current motion management strategies may not account fully for tumor motion and should be used with caution.

Sonography of uterine abnormalities in postpartum and postabortion patients: a potential pitfall of interpretation.

OBJECTIVE: The interpretation of postpartum and postabortion uterine abnormalities on sonography can be challenging. The purpose of this study was to identify misleading imaging features that lead to inclusion of a uterine arteriovenous malformation (AVM) in the differential diagnosis of a uterine abnormality because consideration of this diagnosis can potentially alter patient treatment. METHODS: The sonographic examinations of 29 pathologically proven cases of uterine abnormalities in postpartum and postabortion patients were retrospectively reviewed. Two radiologists independently evaluated several features: the presence of a uterine mass, myometrial involvement, and the presence of an associated vascular abnormality. Percent agreement and the relationship between imaging features and inclusion of a uterine AVM in the differential diagnosis were determined. RESULTS: Interobserver agreement was as follows: the presence of a uterine mass, 90%; myometrial involvement, 83%; the presence of an associated vascular abnormality, 72%; and inclusion of a uterine AVM in the differential diagnosis, 86%. Myometrial involvement showed a statistically significant relationship to inclusion of a uterine AVM in the differential diagnosis (P < .05). Final pathologic diagnoses included retained products of conception (RPOC) (n = 26), an endometrial polyp (n = 1), chronic endometritis (n = 1), and an exogenous progestational effect (n = 1). No uterine AVMs were found. CONCLUSIONS: Despite high interobserver agreement in characterizing uterine abnormalities on sonography, readers still include uterine AVMs in the differential diagnosis of uterine masses that are ultimately proven to be RPOC. A myometrial location of a uterine mass is a particularly misleading imaging feature of RPOC.

Apparent wall thickening of cystic renal lesions on MRI.

PURPOSE: To show that cystic renal lesions that would otherwise meet criteria for simple cysts can demonstrate perceptible walls or increased wall thickness on MRI, sometimes causing these lesions to be "upgraded." It was hypothesized that thickening of cyst walls on MRI can be artifactual, due to data truncation, applied filtering, and low signal-to-noise ratio (SNR). MATERIALS AND METHODS: k-Space data for a 4-cm cyst were created in a 40-cm field of view (FOV) (512 x 512 matrix). Additional data sets were created using the central 512 x 256 and 512 x 128 points. Noise was simulated so that the cyst SNR was approximately 7, 14, and 20, respectively. Actual wall thickness was set at 0.25 mm, and cyst:wall signal at 1:4. An inverse two-dimensional (2D) fast Fourier transform (FFT) yielded simulated images. A Fermi filter was applied to reduce ringing. Images/projections were examined for wall thickening. Seven patients with initially thick-walled cysts on fat-saturated spoiled gradient-echo (FS-SPGR) images were scanned with increasing resolution (256 x 128 and 256 x 256; four patients were also scanned with 512 x 512). Average wall thickness at each resolution was compared using a two-tailed paired Student's t-test. RESULTS: Simulations showed apparent wall thickening at low resolution, improving with higher resolutions. Low SNR and application of the Fermi filter made it difficult to identify ringing as the cause of this thickening. The simulation results were confirmed on seven patients, whose cyst walls proved to be artifactually thickened (P < 0.01). CONCLUSION: Thickening of cyst walls on MRI can be artifactual. Upon encountering thick-walled cystic renal lesions, high-resolution images can be acquired to exclude apparent thickening.

Frequency, management, and outcome of extravasation of nonionic iodinated contrast medium in 69,657 intravenous injections.

PURPOSE: To determine retrospectively the frequency, management, and outcomes of extravasations of intravenously injected nonionic iodinated contrast medium. MATERIALS AND METHODS: Institutional review board approval was obtained, and the requirement for informed consent was waived. The study was HIPAA compliant. For 69,657 intravenous injections of nonionic iodinated contrast medium for computed tomographic examinations between January 1, 2000, and March 31, 2005, the incident reports, radiology reports, and medical records of patients in whom contrast medium extravasations occurred were reviewed. Data collected included patient age, catheter gauge, injection site, volume extravasated, patient symptoms, severity of injury, treatment, whether or not the plastic surgery service was consulted and any additional treatment that service instituted, and patient outcome. RESULTS: Extravasations occurred in 475 (0.7%) of 69 657 patients, and follow-up information was available for 442 adults (280 women, 162 men; mean age, 57 years) and 17 children (nine girls, eight boys; mean age, 6 years). Extravasated volumes ranged from 3 to 150 mL. Symptoms usually consisted of swelling and/or pain. The plastic surgery service was consulted for 38 adults and six children and provided additional treatment in eight patients (seven adults, one child). Among adults, 432 had minimal or no adverse effects, nine had moderate adverse effects, and one had a severe complication (75 mL of contrast material extravasated into the hand, causing compartment syndrome). Only one moderate or severe complication in an adult resulted from an extravasation of less than 50 mL. Fifteen children had minimal or no adverse effects, one had moderate adverse effects, and one had a severe complication (18 mL of contrast material extravasated into the arm, causing brachial plexopathy). CONCLUSION: Extravasation of nonionic iodinated contrast medium results only rarely in moderate or severe adverse effects, and these usually occur only when large volumes of contrast medium are involved.

Contrast-enhanced CT accurately detects hemorrhage in torso trauma: direct comparison with angiography.

OBJECTIVE: To assess the ability of contrast enhanced computed tomography (CECT) to detect active hemorrhage and other vascular injuries in chest, abdominal, and pelvic trauma patients, using angiographic findings and need for intervention as paired gold standards. METHODS: We obtained approval from the Institutional review board for a retrospective search of the radiology information system: seeking trauma patients undergoing angiography within 24 hours of CECT for chest, abdominal, or pelvic injuries. CECT protocol was standard trauma CT, not specialized for CT angiography. Angiographic techniques varied with indication. Clinical and imaging reports and selected radiologic studies were reviewed, but the original dictated report was the interpretive standard. We used Fisher's exact test for statistical analysis. RESULTS: During the 30-month study period, 466 patients underwent emergent interventional radiologic procedures. Of those, 418 were excluded for nontrauma indications or neuroangiographic procedures. Fourty-eight patients (33 male, 15 female, average age 43.4) thus constituted the study population in whom we evaluated 63 traumatic injuries. CT findings had statistically significant associations (p < 0.0001) with both angiographic evidence of active hemorrhage and the need for intervention, which were tabulated separately. CT had 94.1% sensitivity and 97.6% negative predictive value (NPV) for detection of active hemorrhage, and 92.6% sensitivity and 91.2% NPV for predicting need for surgical or endovascular intervention. CONCLUSIONS: CECT findings correlate strongly with angiographic findings, though sensitivity remains imperfect. However, when CT is used in the context of other clinical features, particularly hemodynamic instability, it may enable clinicians to reserve emergent angiography for those patients in whom emergent intervention is planned.

3D excretory MR urography: improved image quality with intravenous saline and diuretic administration.

PURPOSE: To assess the effect of diuretic administration on the image quality of excretory magnetic resonance urography (MRU) obtained following intravenous hydration, and to determine whether intravenous hydration alone is sufficient to produce diagnostic quality studies of nondilated upper tracts. MATERIALS AND METHODS: A total of 22 patients with nondilated upper tracts were evaluated with contrast-enhanced MRU. All patients received 250 mL of saline intravenously immediately prior to the examination. A total of 11 patients received 10-20 mg furosemide in addition to saline. Imaging was performed with a three-dimensional (3D) and two-dimensional (2D) breathhold spoiled gradient-echo sequences. Excretory MRU images were acquired five minutes after the administration of 0.1 mmol/kg gadolinium and were independently reviewed by two radiologists, who were blinded to the MRU technique. Readers evaluated the calyces, renal pelvis, and ureters qualitatively for degree of opacification, distention, and artifacts on a four-point scale. Statistical analysis was performed using a permutation test. RESULTS: There was no significant disagreement between the two readers (P=0.14). Furosemide resulted in significant improvement in calyceal and renal pelvis distention (P<0.005), and significant artifact reduction in all upper tract segments (P<0.001) compared to the effect of saline alone. CONCLUSION: Intravenous furosemide significantly improves the image quality of excretory MRU studies obtained following intravenous hydration. Intravenous saline alone is insufficient to produce diagnostic quality studies of the non-dilated upper tracts.

Hepatic fat fraction: MR imaging for quantitative measurement and display--early experience.

UNLABELLED: The institutional review board approved this HIPAA-compliant study. After all five patients with nonalcoholic fatty liver disease signed a consent, they underwent magnetic resonance (MR) imaging for hepatic fat quantification. The purpose of this study was to develop a fast and accurate method to acquire and display quantitative maps of the percentage of hepatic fat. In-phase and out-of-phase gradient-echo MR imaging was performed with dual flip angles (70 degrees, 20 degrees) to resolve ambiguity of the dominant constituent. T2* corrections were also estimated and applied to generate color-coded maps of the estimated percentage of hepatic fat. MR imaging results were compared with biopsy results in two of five patients, and the technique was validated qualitatively and quantitatively with a water-oil phantom. Results of the phantom study confirmed that the dual-flip angle algorithm can be used to correctly identify the dominant constituent, allowing depiction of 0%-100% of fat content. The estimated liver fat fraction was comparable to quantitative fat measurements at biopsy in both patients (MR imaging, 18.3% +/- 2.8 [standard deviation] and 28.6% +/- 2.4, vs quantitative histopathologic analysis, 11.2% and 28.5%, respectively). SUPPLEMENTAL MATERIAL: radiology.rsnajnls.org/cgi/content/full/2373041639/DC1

Magnetic resonance imaging of prostate cancer.

Management of carcinoma of the prostate has traditionally been guided by digital rectal examination, and by laboratory data such as serum prostate-specific antigen (PSA) level and histopathologic tumor grade. The introduction of the endorectal coil has improved the ability of magnetic resonance imaging (MRI) to contribute to staging and treatment planning of prostate cancer, especially in cases of confined or locally invasive disease. Exciting research in the fields of magnetic resonance spectroscopy (MRS) and MR-guided intervention of the prostate may soon expand the role of MRI in the diagnosis and treatment of prostate cancer. This article reviews current MRI techniques, the MRI features of prostate cancer, the role and efficacy of MRI in prostate carcinoma staging, and the current and future uses of MR spectroscopy and MR-guided prostate brachytherapy.