Modifying LI-RADS on Gadoxetate Disodium-Enhanced MRI: A Secondary Analysis of a Prospective Observational Study.

BACKGROUND: The Liver Imaging Reporting and Data System (LI-RADS) is widely used for diagnosing hepatocellular carcinoma (HCC), however, with unsatisfactory sensitivity, complex ancillary features, and inadequate integration with gadoxetate disodium (EOB)-enhanced MRI. PURPOSE: To modify LI-RADS (mLI-RADS) on EOB-MRI. STUDY TYPE: Secondary analysis of a prospective observational study. POPULATION: Between July 2015 and September 2018, 224 consecutive high-risk patients (median age, 51 years; range, 26-83; 180 men; training/testing sets: 169/55 patients) with 742 (median size, 13 mm; interquartile range, 7-27; 498 HCCs) LR-3/4/5 observations. FIELD STRENGTH/SEQUENCE: 3.0 T T2 -weighted fast spin-echo, diffusion-weighted spin-echo based echo-planar, and 3D T1 -weighted gradient echo sequences. ASSESSMENT: Three radiologists (with 5, 5, and 10 years of experience in liver MR imaging, respectively) blinded to the reference standard (histopathology or imaging follow-up) reviewed all MR images independently. In the training set, the optimal LI-RADS version 2018 (v2018) features selected by Random Forest analysis were used to develop mLI-RADS via decision tree analysis. STATISTICAL TESTS: In an independent testing set, diagnostic performances of mLI-RADS, LI-RADS v2018, and the Korean Liver Cancer Association (KLCA) guidelines were computed using a generalized estimating equation model and compared with McNemar's test. A two-tailed P < 0.05 was statistically significant. RESULTS: Five features (nonperipheral "washout," restricted diffusion, nonrim arterial phase hyperenhancement [APHE], mild-moderate T2 hyperintensity, and transitional phase hypointensity) constituted mLI-RADS, and mLR-5 was nonperipheral washout coupled with either nonrim APHE or restricted diffusion. In the testing set, mLI-RADS was significantly more sensitive (72%) and accurate (80%) than LI-RADS v2018 (sensitivity, 61%; accuracy 74%; both P < 0.001) and the KLCA guidelines (sensitivity, 64%; accuracy 74%; both P < 0.001), without sacrificing positive predictive value (mLI-RADS, 94%; LI-RADS v2018, 94%; KLCA guidelines, 92%). DATA CONCLUSION: In high-risk patients, the EOB-MRI-based mLI-RADS was simpler and more sensitive for HCC than LI-RADS v2018 while maintaining high positive predictive value. LEVEL OF EVIDENCE: 2 TECHNICAL EFFICACY: Stage 2.

Data-Driven Modification of the LI-RADS Major Feature System on Gadoxetate Disodium-Enhanced MRI: Toward Better Sensitivity and Simplicity.

BACKGROUND: The Liver Imaging Reporting and Data System (LI-RADS) is widely accepted as a reliable diagnostic scheme for hepatocellular carcinoma (HCC) in at-risk patients. However, its application is hampered by substantial complexity and suboptimal diagnostic sensitivity. PURPOSE: To propose data-driven modifications to the LI-RADS version 2018 (v2018) major feature system (rLI-RADS) on gadoxetate disodium (EOB)-enhanced magnetic resonance imaging (MRI) to improve sensitivity and simplicity while maintaining high positive predictive value (PPV) for detecting HCC. STUDY TYPE: Retrospective. POPULATION: Two hundred and twenty-four consecutive at-risk patients (training dataset: 169, independent testing dataset: 55) with 742 LR-3 to LR-5 liver observations (HCC: N = 498 [67%]) were analyzed from a prospective observational registry collected between July 2015 and September 2018. FIELD STRENGTH/SEQUENCE: 3.0 T/T2-weighted fast spin-echo, diffusion-weighted spin-echo based echo-planar and three-dimensional (3D) T1-weighted gradient echo sequences. ASSESSMENT: All images were evaluated by three independent abdominal radiologists who were blinded to all clinical, pathological, and follow-up information. Composite reference standards of either histopathology or imaging follow-up were used. STATISTICAL TESTS: In the training dataset, LI-RADS v2018 major features were used to develop rLI-RADS based on their associated PPV for HCC. In an independent testing set, diagnostic performances of LI-RADS v2018 and rLI-RADS were computed using a generalized estimating equation model and compared with McNemar's test. A P value <0.05 was considered statistically significant. RESULTS: The median (interquartile range) size of liver observations was 13 mm (7-27 mm). The diagnostic table for rLI-RADS encompassed 9 cells, as opposed to 16 cells for LI-RADS v2018. In the testing set, compared to LI-RADS v2018, rLI-RADS category 5 demonstrated a significantly superior sensitivity (76% vs. 61%) while maintaining comparably high PPV (92.5% vs. 94.1%, P = 0.126). DATA CONCLUSION: Compared with LI-RADS v2018, rLI-RADS demonstrated improved simplicity and significantly superior diagnostic sensitivity for HCC in at-risk patients. LEVEL OF EVIDENCE: 3 TECHNICAL EFFICACY STAGE: 2.

Coaxial Stent Graft inside a Constraining Stent for Transjugular Intrahepatic Portosystemic Shunt Reduction.

A technique to create a coaxial, self-expanding stent graft inside a constraining, bare-metal, balloon-expandable stent for transjugular intrahepatic portosystemic shunt (TIPS) reduction is described. The key steps are performed on a back table rather than inside the patient, and the resulting construct is deployed using standard unsheathing maneuvers. The construct was used in 4 patients to make 6 TIPS diameter reductions (mean postreduction diameter, 6 mm; range, 0-8 mm), all resulting in increases in the portosystemic pressure gradient (mean increase, 6 mm Hg; range, 1-19 mm Hg). On average, hepatic encephalopathy improved 1 point on the West Haven scale (range, 0-2).

Proton Pump Inhibitor Use Is Associated with an Increased Frequency of New or Worsening Hepatic Encephalopathy after Transjugular Intrahepatic Portosystemic Shunt Creation.

PURPOSE: To determine whether proton pump inhibitor (PPI) use increases the rate of new or worsening hepatic encephalopathy (HE) after transjugular intrahepatic portosystemic shunt (TIPS) creation. MATERIALS AND METHODS: In this retrospective study, 284 of 365 patients who underwent TIPS creation from January 1, 2005, to December 31, 2016, were analyzed (186 male, mean age 56 y, range 19-84 y). Dates of PPI use and dates of new or worsening HE, defined as hospitalization or escalation in outpatient medical management, were extracted from medical records. Mixed-effects negative binomial regression was used to test for an association between PPI usage and HE. RESULTS: After TIPS creation, among 168 patients on PPIs chronically, there were 235 episodes of new or worsening HE in 106,101 person-days (0.81/person-year). Among 55 patients never on PPIs, there were 37 episodes in 31,066 person-days (0.43/person-year). Among 61 patients intermittently taking PPIs, there were 78 episodes in 37,710 person-days while on PPIs (0.75/person-year) and 25 episodes in 35,678 person-days while off PPIs (0.26/person-year). In univariate regression, PPI usage was associated with a 3.34-fold increased rate of new or worsening HE (incidence rate ratio [IRR] 3.34; P < .001). In multivariate regression, older age (IRR 1.05; P < .001), male sex (IRR 1.58; P = .023), higher Model for End-Stage Liver Disease score (IRR 1.06; P = .015), previous HE or HE-preventive medication use (IRR 1.51; P = .029), and PPI use (IRR 3.19; P < .001) were significant risk factors. Higher PPI doses were associated with higher rates of HE (IRR 1.16 per 10 mg omeprazole equivalent; P = .046). CONCLUSIONS: PPI usage is associated with increased rates of new or worsening HE after TIPS creation.

Risk factors for ocular complications in periocular infantile hemangiomas.

BACKGROUND/OBJECTIVES: Infantile hemangiomas are the most common benign tumors of childhood. Although some children with periocular infantile hemangiomas do not require treatment, these lesions may result in amblyopia and visual impairment if not properly treated. We have attempted to characterize clinical features of periocular infantile hemangiomas that predict negative ocular outcomes and thus require prompt referral to an ophthalmologist and initiation of therapy. METHODS: This study included children with periocular infantile hemangiomas consecutively seen at Ann & Robert H. Lurie Children's Hospital of Chicago from January 1994 through December 2014. Only patients evaluated by both a dermatologist and an ophthalmologist were included. Medical records of patients who met inclusion criteria were reviewed. Ocular findings were reviewed for the presence of ptosis, refractive errors, strabismus, proptosis, and amblyopia. RESULTS: Ninety-six patients (74% female, median age of onset 0.48 months) were included. Periocular infantile hemangiomas larger than 1 cm in diameter, with a deep component, and with involvement of the upper eyelid were significantly associated with astigmatism (P = .002, P = .02, and P = .003, respectively) and amblyopia (P = .002, P = .02, and P = .04, respectively). Using logistic regression, diameter greater than 1 cm (odds ratio = 14.13, P = .01) and amblyopia (odds ratio = 21.00, P = .04) were the strongest predictors of astigmatism. Lower lid and medial and lateral canthal involvement were not predictive of ocular complications. CONCLUSION: Predictive factors for ocular complications in patients with periocular infantile hemangiomas are diameter greater than 1 cm, a deep component, and upper eyelid involvement, with size being the most consistent predictor. These patients should be promptly referred to an ophthalmologist, and treatment should be strongly considered.

Towards optimal deep fusion of imaging and clinical data via a model-based description of fusion quality.

BACKGROUND: Due to intrinsic differences in data formatting, data structure, and underlying semantic information, the integration of imaging data with clinical data can be non-trivial. Optimal integration requires robust data fusion, that is, the process of integrating multiple data sources to produce more useful information than captured by individual data sources. Here, we introduce the concept of fusion quality for deep learning problems involving imaging and clinical data. We first provide a general theoretical framework and numerical validation of our technique. To demonstrate real-world applicability, we then apply our technique to optimize the fusion of CT imaging and hepatic blood markers to estimate portal venous hypertension, which is linked to prognosis in patients with cirrhosis of the liver. PURPOSE: To develop a measurement method of optimal data fusion quality deep learning problems utilizing both imaging data and clinical data. METHODS: Our approach is based on modeling the fully connected layer (FCL) of a convolutional neural network (CNN) as a potential function, whose distribution takes the form of the classical Gibbs measure. The features of the FCL are then modeled as random variables governed by state functions, which are interpreted as the different data sources to be fused. The probability density of each source, relative to the probability density of the FCL, represents a quantitative measure of source-bias. To minimize this source-bias and optimize CNN performance, we implement a vector-growing encoding scheme called positional encoding, where low-dimensional clinical data are transcribed into a rich feature space that complements high-dimensional imaging features. We first provide a numerical validation of our approach based on simulated Gaussian processes. We then applied our approach to patient data, where we optimized the fusion of CT images with blood markers to predict portal venous hypertension in patients with cirrhosis of the liver. This patient study was based on a modified ResNet-152 model that incorporates both images and blood markers as input. These two data sources were processed in parallel, fused into a single FCL, and optimized based on our fusion quality framework. RESULTS: Numerical validation of our approach confirmed that the probability density function of a fused feature space converges to a source-specific probability density function when source data are improperly fused. Our numerical results demonstrate that this phenomenon can be quantified as a measure of fusion quality. On patient data, the fused model consisting of both imaging data and positionally encoded blood markers at the theoretically optimal fusion quality metric achieved an AUC of 0.74 and an accuracy of 0.71. This model was statistically better than the imaging-only model (AUC = 0.60; accuracy = 0.62), the blood marker-only model (AUC = 0.58; accuracy = 0.60), and a variety of purposely sub-optimized fusion models (AUC = 0.61-0.70; accuracy = 0.58-0.69). CONCLUSIONS: We introduced the concept of data fusion quality for multi-source deep learning problems involving both imaging and clinical data. We provided a theoretical framework, numerical validation, and real-world application in abdominal radiology. Our data suggests that CT imaging and hepatic blood markers provide complementary diagnostic information when appropriately fused.

Clinical and imaging predictors of the natural course of bland portal vein thrombus in cirrhotic patients.

BACKGROUND: Portal vein thrombus (PVT) in cirrhotic patients is associated with worsening portal hypertension, leads to increased complexity of necessary interventions such as transjugular liver portosystemic shunt or liver transplantation, and is associated with worse outcomes after liver transplantation. Additionally, there are no established consensus guidelines for management of bland PVT in cirrhotic patients, which currently exists on a spectrum and is patient and provider dependent. PURPOSE: The aim of this study was to determine whether there are associations between key clinical and imaging variables and bland PVT burden over time. MATERIAL AND METHODS: This exploratory, retrospective, single-center study included patients who underwent two or more multiphase CT or MRI examinations between 1/1/2013 and 12/31/2019 and had a diagnosis of PVT. Three readers independently evaluated all index and follow-up examinations for PVT burden using a proposed 8-point scale and the established Yerdel score. Key clinical factors were collected from the electronic medical record. The PVT burden over time and the association of this burden with key clinical and imaging variables were assessed using logistic regression models. RESULTS: 138 patients with cirrhosis and bland PVT were included in the analyses. Median age was 60 years (interquartile range 55-67; 90 men) and median follow-up time was 19.6 months. At baseline, the mean score was 2.31 (± 1.44) and at the final follow-up examination, the mean score was 2.52 (± 1.99). Baseline occlusion level was the only statistically significant association with worsening PVT burden in all four vascular territories (p-value < 0.0001). Anticoagulation status did not have a statistically significant association with change in thrombus burden in any vascular territory or cumulative thrombus burden across all territories (p-value 0.11-0.43). CONCLUSION: In conclusion, our study shows that in cirrhotic patients with bland PVT, the thrombus burden did not change significantly over time and baseline thrombus burden is the only clinical factor significantly associated with increasing burden over time.

An Endovascular Approach to the Entrapped Central Venous Catheter After Cardiac Surgery.

PURPOSE: Entrapment of central venous catheters (CVC) at the superior vena cava (SVC) cardiopulmonary bypass cannulation site by closing purse-string sutures is a rare complication of cardiac surgery. Historically, resternotomy has been required for suture release. An endovascular catheter release approach was developed. MATERIALS AND METHODS: Four cases of CVC tethering against the SVC wall and associated resistance to removal, suggestive of entrapment, were encountered. In each case, catheter removal was achieved using a reverse catheter fluoroscopically guided over the suture fixation point between catheter and SVC wall, followed by the placement of a guidewire through the catheter. The guidewire was snared and externalized to create a through-and-through access with the apex of the loop around the suture. A snare placed from the femoral venous access provided concurrent downward traction on the distal CVC during suture release maneuvers. RESULTS: In the initial attempt, gentle traction freed the CVC, which fractured and was removed in two sections. In the subsequent three cases, traction alone did not release the CVC. Therefore, a cutting balloon was introduced over the guidewire and inflated. Gentle back-and-forth motion of the cutting balloon atherotomes successfully incised the suture in all three attempts. No significant postprocedural complications were encountered. During all cases, a cardiovascular surgeon was present in the interventional suite and prepared for emergent resternotomy, if necessary. CONCLUSION: An endovascular algorithm to the "entrapped CVC" is proposed, which likely reduces risks posed by resternotomy to cardiac surgery patients in the post-operative period.

Membrane glycoproteins associated with breast tumor cell progression identified by a lectin affinity approach.

The membrane glycoprotein component of the cellular proteome represents a promising source for potential disease biomarkers and therapeutic targets. Here we describe the development of a method that facilitates the analysis of membrane glycoproteins and apply it to the differential analysis of breast tumor cells with distinct malignant phenotypes. The approach combines two membrane extraction procedures, and enrichment using ConA and WGA lectin affinity columns, prior to digestion and analysis by LC-MS/MS. The glycoproteins are identified and quantified by spectral counting. Although the distribution of glycoprotein expression as a function of MW and p I was very similar between the two related cell lines tested, the approach enabled the identification of several distinct membrane glycoproteins with an expression index correlated with either a precancerous (MCF10AT1), or a malignant, metastatic cellular phenotype (MCF10CA1a). Among the proteins associated with the malignant phenotype, Gamma-glutamyl hydrolase, CD44, Galectin-3-binding protein, and Syndecan-1 protein have been reported as potential biomarkers of breast cancer.