Description: Objective To evaluate dual-energy spectral CT imaging in evaluating the degree of differentiation in colon cancer. Methods Forty-seven colon cancer patients underwent spectral CT during arterial phase (AP) and portal venous phase (PP), and were characterized pathologically differentiated to well-differentiated (A, n  = 18) and poorly differentiated or undifferentiated carcinoma group (B, n  = 29). Lesion iodine concentration (IC) was measured and normalized to that of aorta (NIC). CT numbers were measured and the slope ( λ HU ) of the spectral HU curve was calculated. These parameters were statistically compared between the two groups. ROC curves were used to evaluate their diagnostic efficacies. Results There were significant differences in IC (1.01 ± 0.20 vs. 1.59 ± 0.57 mg/ml), NIC (0.12 ± 0.03 vs. 0.19 ± 0.09), λ HU (1.41 ± 0.29 vs. 2.03 ± 0.85), and CT number at 70 keV (48.61 ± 9.03HU vs. 63.97 ± 15.86HU) between groups A and B in AP ( p  〈 0.05), but no difference in PP. Using IC = 1.13 mg/ml in AP as the threshold, one obtained a sensitivity of 81.8% and a specificity of 71.4% for differentiating well-differentiated from poorly differentiated or undifferentiated carcinoma. These values were statistically higher than those (64.7% and 62.3%) using CT number at 70 keV. Conclusion Spectral CT imaging parameters (IC, NIC, and λ HU ) in AP provide improved accuracy for evaluating the degrees of differentiation in colon cancer than CT number at 70 keV.

Description: Purpose To investigate clinical features and results of follow-up in patients with arterial injury during transcatheter arterial chemoembolization (TACE) for hepatocellular carcinoma. Methods From 2005 to 2015, 2219 TACE procedures were performed in 906 patients in our hospital. Iatrogenic arterial injury occurred during 38 TACE procedures (sessions) in 35 patients (24 men, 11 women; mean age 71.8 years). The incidence of arterial injury was 1.7%. We evaluated the characteristics of arterial injury, and evaluated the risk factors for incomplete recanalization of the injured artery at follow-up angiography conducted after 1–11 months (mean 102.5 days). Results Iatrogenic arterial injury was caused by the microcatheter in 34 of 38 cases (86.8%). There were 15 cases (39.5%) in which the replaced hepatic artery or the extrahepatic artery was the parasitic supply. Extravasation occurred in five cases. In 36 of 38 cases, follow-up angiography was performed. We divided 36 cases into two groups: complete reopening ( n  = 24) and non-complete reopening ( n  = 12). The two groups were compared regarding the factors associated with incomplete recanalization of the injured artery at follow-up. Injury length 〉3 cm ( p  = 0.0002) and proximal arterial injury (proximal to the segmental artery; p  = 0.03) were significant risk factors for non-complete reopening of the injured artery. Conclusion Iatrogenic arterial injury frequently occurred in the extrahepatic artery or replaced hepatic artery. Recanalization rate of arterial injury was high; however, injury length 〉3 cm and proximal arterial injury were risk factors for non-complete reopening of the injured artery.

Description: Purpose To evaluate the image quality of routine diagnostic images generated from a novel detector-based spectral detector CT (SDCT) and compare it with CT images obtained from a conventional scanner with an energy-integrating detector (Brilliance iCT), Routine diagnostic (conventional/polyenergetic) images are non-material-specific images that resemble single-energy images obtained at the same radiation, Methods ACR guideline-based phantom evaluations were performed on both SDCT and iCT for CT adult body protocol. Retrospective analysis was performed on 50 abdominal CT scans from each scanner. Identical ROIs were placed at multiple locations in the abdomen and attenuation, noise, SNR, and CNR were measured. Subjective image quality analysis on a 5-point Likert scale was performed by 2 readers for enhancement, noise, and image quality. Results On phantom studies, SDCT images met the ACR requirements for CT number and deviation, CNR and effective radiation dose. In patients, the qualitative scores were significantly higher for the SDCT than the iCT, including enhancement (4.79 ± 0.38 vs. 4.60 ± 0.51, p  = 0.005), noise (4.63 ± 0.42 vs. 4.29 ± 0.50, p  = 0.000), and quality (4.85 ± 0.32, vs. 4.57 ± 0.50, p  = 0.000). The SNR was higher in SDCT than iCT for liver (7.4 ± 4.2 vs. 7.2 ± 5.3, p  = 0.662), spleen (8.6 ± 4.1 vs. 7.4 ± 3.5, p  = 0.152), kidney (11.1 ± 6.3 vs. 8.7 ± 5.0, p  = 0.033), pancreas (6.90 ± 3.45 vs 6.11 ± 2.64, p  = 0.303), aorta (14.2 ± 6.2 vs. 11.0 ± 4.9, p  = 0.007), but was slightly lower in lumbar-vertebra (7.7 ± 4.2 vs. 7.8 ± 4.5, p  = 0.937). The CNR of the SDCT was also higher than iCT for all abdominal organs. Conclusion Image quality of routine diagnostic images from the SDCT is comparable to images of a conventional CT scanner with energy-integrating detectors, making it suitable for diagnostic purposes.

Description: This review article focuses on gastroduodenal and pancreatic surgeries with the goal of identifying radiologic findings that translate to important surgical considerations. The topics covered include partial gastrectomy with reconstruction techniques, total gastrectomy, pancreaticoduodenectomy, and pancreaticojejunostomy. Indications, contraindications, surgical techniques, and postoperative imaging are described within each of these topics. Knowledge of these surgical techniques is extremely helpful for the interpreting radiologists to identify expected postoperative anatomy and related complications that would remain clinically relevant to our surgical colleagues and direct timely patient management.

Description: Large bowel perforation is an abdominal emergency that results from a wide range of etiologies. Computed tomography is the most reliable modality in detecting the site of large bowel perforation. The diagnosis is made by identifying direct CT findings such as extraluminal gas or contrast and discontinuity along the bowel wall. Indirect CT findings can help support the diagnosis, and include bowel wall thickening, pericolic fat stranding, abnormal bowel wall enhancement, abscess, and a feculent collection adjacent to the bowel. Common etiologies that cause large bowel perforation are colon cancer, foreign body aspiration, stercoral colitis, diverticulitis, ischemia, inflammatory and infectious colitides, and various iatrogenic causes. Recognizing a large bowel perforation on CT can be difficult at times, and there are various entities that may be misinterpreted as a colonic perforation. The purpose of this article is to outline the MDCT technique used for evaluation of suspected colorectal perforation, discuss relevant imaging findings, review common etiologies, and point out potential pitfalls in making the diagnosis of large bowel perforation.

Description: Purpose To evaluate the usefulness of a novel contrast-injection protocol for high-resolution abdominal computed tomography angiography (CTA) using nitroglycerin (NTG). Methods Abdominal CTA was performed in 80 patients using two 64-detector-row CT scanners. Forty patients were examined after administration of sublingual NTG (NTG group), while 40 were examined without NTG administration (non-NTG group). Arterial phase images were acquired with maximum intensity projection and volume rendering. Reduction rates: vessel cross-sectional areas ratio of 10 cm distal to origin at the superior mesenteric artery, contrast enhancements, signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR) were assessed. Three reviewers evaluated degree of depiction of the peripancreatic vasculature using a four-point scale (1 = poor, 4 = excellent). Results Reduction rates were significantly lower in the NTG group ( P  〈 0.001), while there were no significant differences in contrast enhancements, SNR, or CNR between groups. Visual evaluation results of the NTG group were significantly better than those of the non-NTG group ( P  〈 0.01). Conclusion Abdominal CTA using NTG improved visualization of the abdominal peripheral vessels. This improved arterial view may be beneficial for preoperative evaluation of the arterial anatomy.

Description: Purpose To determine the clinical efficiency and long-term outcomes between unilateral and bilateral stent insertion in patients with malignant hilar biliary obstruction. Methods From August 2012 to February 2016, 63 consecutive patients with malignant hilar biliary obstruction were treated with unilateral or bilateral stent insertion at our center. The bilateral stents were inserted using the side-by-side technique. The clinical efficiency and long-term outcomes were compared between the two groups. Results Unilateral and bilateral stent insertions were successfully performed in 31 of 33 and 27 of 30 patients, respectively ( P  = 0.912). No procedure-related complication occurred. Clinical success was achieved in 29 of 31 patients in the unilateral stent group and in 26 of 27 patients in the bilateral stent group ( P  = 0.637). During the follow-up, re-obstruction of stent occurred in five patients in the unilateral stent group and in three patients in the bilateral stent group ( P  = 0.58). The significant differences were not observed in the stent patency time (368 vs. 387 days, P  = 0.685) and survival (200 vs. 198 days, P  = 0.751) between two groups. Based on the univariate and multivariate analyses, the independent risk factors for decreasing the survival time included higher Eastern Cooperative Oncology Group performance status ( P  = 0.018), higher alanine aminotransferase level ( P  = 0.009), and absence of anticancer treatment after stent insertion ( P  = 0.002). Conclusion Compared to bilateral stent insertion for malignant hilar biliary obstruction, unilateral stent insertion can provide comparable clinical efficiency and long-term outcomes.

Description: Purpose To determine if hepatic venous pressure gradient (HVPG) correlates with advanced hepatic fibrosis, as a complement to transjugular (transvenous) core needle liver biopsy. Materials and methods After institutional review board approval, a retrospective review was conducted on 340 patients who underwent transjugular (transvenous) core needle liver biopsy with concurrent pressure measurements between 6/1/2007 and 6/1/2013. Spearman correlation and linear regression were performed. A receiver operating characteristic (ROC) curve was created and sensitivity, specificity, predictive values and likelihood ratios were calculated. Results Indications included hepatitis C, abnormal liver function tests, non-alcoholic steatohepatitis, autoimmune hepatitis, and cirrhosis, among others. Biopsies showed stage 1 or 2 fibrosis in 15.6% each, stage 3 fibrosis in 21.6%, stage 4 fibrosis in 40.7%, and no fibrosis in 6.5%. Mean HVPG was 6.5 mm Hg (SD 5.0) with a range of 0–26 mm Hg. Spearman correlation coefficient for association between HVPG and fibrosis stage was 0.561 ( p  〈 0.001). R2 on linear regression was 0.247 ( p  〈 0.001). ROC curve for the prediction of stage 4 fibrosis had an area under the curve of 0.79 (95% CI 0.73–0.85). HVPG of ≥6 mm Hg had a sensitivity of 71.3%, specificity of 79.6%, positive predictive value of 70.5%, negative predictive value of 80.2%, positive likelihood ratio of 3.49 (95% CI 2.45–4.97) and negative likelihood ratio of 0.36 (95% CI 0.26–0.50) for diagnosis of stage 4 fibrosis. Conclusions HVPG correlates with stage 4 (advanced) hepatic fibrosis.

Description: Purpose To compare magnetic resonance cholangiopancreatography (MRCP) and Gd-EOB-DTPA-enhanced MRI in the evaluation of the biliary anatomy in potential living liver donors (LLDs). Methods A retrospective study was conducted in a tertiary care liver transplant center after obtaining ethics and institutional approvals. A total of 42 potential LLD MRI examinations were performed between November 2013 and March 2016. All patients underwent a standard MRI protocol which included MRCP and Gd-EOB-DTPA-enhanced MRI sequences in a single session. Three abdominal MR radiologists independently reviewed the studies and completed a customized data collection sheet for each MR sequence. The readers subjectively scored the bile duct visualization on each MR sequence on a Likert scale and classified the biliary anatomic configuration. Statistical analysis was performed using intraclass correlation coefficient and the McNemar Chi-square ( χ 2 ) test. Results The 42 potential LLDs included 22 males and 20 females with an age range of 18–60 years. There was ‘good’ or ‘excellent’ inter-reader agreement on either MRI examination for the visualization of the first- and second-order ducts and the majority of third-order ducts. ‘Good’ inter-reader agreement on Gd-EOB-DTPA-enhanced MRI and ‘fair’ inter-reader agreement on MRCP was noted for the left third-order medial duct. There was significantly better visualization of the cystic duct, left hepatic duct, and right second-order ducts on Gd-EOB-DTPA-enhanced MRI compared with MRCP. A 12.6% improvement in classifying the biliary branch pattern was also observed on Gd-EOB-DTPA-enhanced MRI compared with MRCP ( P  = 0.03). Conclusion Gd-EOB-DTPA-enhanced MRI provides additional diagnostic confidence over MRCP in the evaluation of the biliary ductal anatomy in potential LLDs.