Spontaneous perforation of the common bile duct: an uncommon complication of primary sclerosing cholangitis.

OBJECTIVES: To describe the MR features and prognosis of patients with an uncommon complication of primary sclerosing cholangitis (PSC) characterized by a spontaneous perforation of the common bile duct (CBD) resulting in a peri-biliary collection and a pseudo-cystic appearance of the CBD. METHODS: A single-center cohort of 263 patients with PSC who had at least two MRIs between 2003 and 2022 and a minimum follow-up of 1 year was retrospectively analyzed. MRI data (characteristics of CBD perforation and MR features of PSC) and clinical data were assessed. Analysis of survival without liver transplantation according to type of PSC (classical or CBD spontaneous perforation) was performed according to the Kaplan-Meier method and the curves were compared using the Log-Rank test. RESULTS: A total of nine (3.4%) PSC patients (5 males) had perforation of the CBD with a median age at diagnosis of 18 years compared to 33 years for the control group (p = 0.019). The peri-biliary collections were variable in appearance (fusiform or pedunculated), with a diameter ranging from 5 to 54 mm. All nine patients showed intra- and extra-hepatic bile duct involvement, dysmorphia, and high ANALI scores. The clinical course was characterized by numerous complications in most patients, and five patients (56%) underwent liver transplantation at a median time of 5 years from diagnosis, compared to 40 patients (16%) in the control group (p = 0.02). CONCLUSION: The spontaneous perforation of the common bile duct is an uncommon complication of primary sclerosing cholangitis that affects young patients and is associated with a poor prognosis. CLINICAL RELEVANCE STATEMENT: This uncommon complication of primary sclerosing cholangitis with perforation of the common bile duct resulting in a peri-biliary collection and a pseudo-cystic appearance of the common bile duct is characterized by a poor prognosis in younger patients. KEY POINTS: • Among 263 patients with primary sclerosing cholangitis (PSC), nine patients (3.6%) had an uncommon complication characterized on MRI by perforation of the common bile duct (CBD). • This perforation of the CBD was responsible in all nine cases for the formation of a peri-biliary collection, giving a pseudo-cystic appearance to the CBD. • The spontaneous perforation of the common bile duct is an uncommon complication of primary sclerosing cholangitis that affects young patients with a poor prognosis.

Deep learning for automatic bowel-obstruction identification on abdominal CT.

RATIONALE AND OBJECTIVES: Automated evaluation of abdominal computed tomography (CT) scans should help radiologists manage their massive workloads, thereby leading to earlier diagnoses and better patient outcomes. Our objective was to develop a machine-learning model capable of reliably identifying suspected bowel obstruction (BO) on abdominal CT. MATERIALS AND METHODS: The internal dataset comprised 1345 abdominal CTs obtained in 2015-2022 from 1273 patients with suspected BO; among them, 670 were annotated as BO yes/no by an experienced abdominal radiologist. The external dataset consisted of 88 radiologist-annotated CTs. We developed a full preprocessing pipeline for abdominal CT comprising a model to locate the abdominal-pelvic region and another model to crop the 3D scan around the body. We built, trained, and tested several neural-network architectures for the binary classification (BO, yes/no) of each CT. F1 and balanced accuracy scores were computed to assess model performance. RESULTS: The mixed convolutional network pretrained on a Kinetics 400 dataset achieved the best results: with the internal dataset, the F1 score was 0.92, balanced accuracy 0.86, and sensitivity 0.93; with the external dataset, the corresponding values were 0.89, 0.89, and 0.89. When calibrated on sensitivity, this model produced 1.00 sensitivity, 0.84 specificity, and an F1 score of 0.88 with the internal dataset; corresponding values were 0.98, 0.76, and 0.87 with the external dataset. CONCLUSION: The 3D mixed convolutional neural network developed here shows great potential for the automated binary classification (BO yes/no) of abdominal CT scans from patients with suspected BO. CLINICAL RELEVANCE STATEMENT: The 3D mixed CNN automates bowel obstruction classification, potentially automating patient selection and CT prioritization, leading to an enhanced radiologist workflow. KEY POINTS: • Bowel obstruction's rising incidence strains radiologists. AI can aid urgent CT readings. • Employed 1345 CT scans, neural networks for bowel obstruction detection, achieving high accuracy and sensitivity on external testing. • 3D mixed CNN automates CT reading prioritization effectively and speeds up bowel obstruction diagnosis.

Chyluria: non-enhanced MR lymphography.

Chyluria is an uncommon medical condition resulting from an abnormal communication between the abdominal lymphatic system and the urinary tract, which results in the presence of chyle in the urine, making it appear milky white. Proper diagnosis is demonstrated by the concentration of urinary lipids. Worldwide, chyluria is most commonly associated with the parasite Wuchereria bancrofti. However, in Europe and North America, where the condition is rare, non-parasitic etiologies predominate. Identifying the cause and location of the uro-lymphatic communication is essential in guiding therapeutic management, but imaging the lymphatic channels remains a challenge. Magnetic resonance (MR) lymphography, a non-invasive free-breathing 3D high-resolution fast-recovery fast spin-echo sequence similar to that used for 3D MR cholangiopancreatography, may demonstrate the cause and location of an abnormal communication between the lymphatic system and urinary tract. In parasitic causes of chyluria, dilated lymphatics vessels communicating with the lymphatic system are demonstrated. In non-parasitic causes of chyluria channel type lymphatic malformations are the most common. Markedly dilated and dysplastic lymphatic vessels communicating with the urinary tract are demonstrated. In addition, other cystic or channel type lymphatic malformations such as thoracic, soft tissue or bone abnormalities may be observed. This review describes the abdominal lymphatic diseases leading to chyluria and presents the technique and images obtained with non-enhanced MR lymphography to enable radiologists in identifying and classifying uro-lymphatic fistulae.Critical relevance statement: Non-enhanced MR lymphography enables the identification and categorization of uro-lymphatic fistulae.

Imaging of the biliary tract.

PURPOSE OF REVIEW: Magnetic resonance cholangiopancreatography (MRCP) has become the reference examination for the exploration of the biliary tract and has replaced endoscopic cholangiography for the analysis of the biliary tract because of its equivalent performance and its noninvasive character. RECENT FINDINGS: Based on the International Primary Sclerosing Cholangitis (PSC) Study Group recommendations for MR imaging in PSC, two protocols can be distinguished for the imaging of biliary tract: a basic protocol and a more complete protocol. It is essential to know the main pitfalls in order not to wrongly describe biliary anomalies. In addition to the excellent performance of MR imaging with MRCP in analyzing the anatomy and the anomalies of the biliary tree, complementary techniques have recently been developed. Several MR prognostic factors have been described. New hepato-specific contrast agents are now available for assessment of the general and segmental liver function. MR Elastography and Diffusion-weighted MR sequences are accurate to evaluate the degree of hepatic fibrosis. Finally, images obtained in MRCP can be postprocessed by a software that will analyze and model the biliary tree in order to quantitatively evaluate the biliary system. SUMMARY: Magnetic resonance imaging with its recent developments becomes by now an essential tool for the evaluation of biliary diseases.

Improving pain control during transarterial chemoembolization for hepatocellular carcinoma performed under local anesthesia with multimodal analgesia.

PURPOSE: The purpose of this study was to assess the performance of a reinforced analgesic protocol (RAP) on pain control in patients undergoing conventional trans-arterial chemoembolization (cTACE) for hepatocellular carcinoma (HCC). MATERIALS AND METHODS: Eighty-one consecutive patients (57 men, 24 women) with a mean age of 69 ± 10 (standard deviation) years (age range: 49-92 years) underwent 103 cTACEs. Standard antalgic protocol (50 mg hydroxyzine, 10 mg oxycodone, 8 mg ondansetron, and lidocaine for local anesthesia) was prospectively compared to a RAP (standard + 40 mg 2-h infusion nefopam and 50 mg tramadol). The individual pain risk was stratified based on age, the presence of cirrhosis and alcoholic liver disease, and patients were assigned to a low-risk group (standard protocol) or high-risk group (RAP). The primary endpoint was severe periprocedural abdominal pain (SAP), defined as a visual analog scale score ≥30/100. A predefined intermediate analysis was performed to monitor the benefit-risk of the RAP. Based on the intermediate analysis, all patients were treated with the RAP. RESULTS: The intermediate analysis performed after 52 cTACE showed that 2/17 (12%) high-risk patients (i.e., those receiving the RAP) experienced SAP compared to 15/35 (43%) low-risk patients (odds ratio [OR] = 0.18; 95% confidence interval [CI]: 0.02-0.98; P = 0.03). Analysis of all procedures showed that 12/67 (18%) patients in cTACE receiving the RAP experienced SAP compared to 15/36 (42%) patients who did not receive it (OR = 3.27; 95% CI: 1.32-8.14; P = 0.01). There were no statistical differences in adverse events, particularly for nausea, between groups. CONCLUSION: Reinforcing the analgesic protocol by combining non-opioid and opioid molecules reduces perioperative pain in patients undergoing cTACE for HCC.

A BCB4 variant is associated with hepatobiliary MR abnormalities in people with low-phospholipid-associated cholelithiasis syndrome.

Background & Aims: The low-phospholipid-associated cholelithiasis (LPAC) syndrome is a recently described peculiar form of cholelithiasis associated with the ATP-binding-cassette subfamily B, member 4 (ABCB4) gene deficiency. The purpose of our study was to analyse the relationship between magnetic resonance (MR) features and the genetic status of ABCB4 in people with LPAC syndrome. Methods: A total of 233 individuals with proven LPAC syndrome were enrolled between January 2003 and June 2018 in a retrospective single-centre study. Inclusion criteria included availability of clinical files, MR images, and genetic data. MR images were analysed by consensus among 3 senior radiologists blinded to the status of ABCB4 gene mutation. Results: A total of 125 individuals (mean age at first MR imaging 40.8 years; 66% females; 48% ABCB4 variant) were included. MR abnormalities were found in 61 (49%) of the 125 individuals. Forty (67%) of the 60 individuals with an ABCB4 gene variant had MR abnormalities as compared with 21 (33%) of the 65 individuals without an ABCB4 gene variant (odds ratio [OR] 4.1, 95% CI 1.9-9.5, p = 0.0001). Compared to individuals with no variant, individuals with an ABCB4 variant were more likely to show intrahepatic macrolithiasis (56 vs. 17%; OR 6.3, 95% CI 2.6-16.2, p <0.0001), bile duct dilatation (60 vs. 18%; OR 6.5, 95% CI 2.7-16.3, p <0.0001), and at least 1 MR feature of complication (35 vs. 15%; OR 2.9, 95% CI 1.1-7.8, p <0.05). Conclusions: ABCB4-related LPAC syndrome is associated with more frequent and severe hepatobiliary MR abnormalities. This finding strongly supports the major role of the ABCB4 gene in the pathogenesis of LPAC syndrome and highlights a genotype-phenotype association in this inherited disease with genetic heterogeneity. Lay summary: ABCB4-related LPAC syndrome associated with an ABCB4 gene variant demonstrates more frequent and severe hepatobiliary MR abnormalities. This finding supports the major role of the ABCB4 gene in the pathogenesis of LPAC syndrome.

Quantitative magnetic resonance cholangiopancreatography metrics are associated with disease severity and outcomes in people with primary sclerosing cholangitis.

Background & Aims: People with primary sclerosing cholangitis (PSC) have a variable and often progressive disease course that is associated with biliary and parenchymal changes. These changes are typically assessed by magnetic resonance imaging (MRI), including qualitative assessment of magnetic resonance cholangiopancreatography (MRCP). Our aim was to study the association of novel objective quantitative MRCP metrics with prognostic scores and patient outcomes. Methods: We performed a retrospective study including 77 individuals with large-duct PSC with baseline MRCP images, which were postprocessed to obtain quantitative measures of bile ducts using MRCP+™. The participants' ANALI scores, liver stiffness by vibration-controlled transient elastography, and biochemical indices were collected at baseline. Adverse outcome-free survival was measured as the absence of decompensated cirrhosis, liver transplantation (LT), or liver-related death over a 12-year period. The prognostic value of MRCP+-derived metrics was assessed by Cox regression modelling. Results: During a total of 386 patients-years, 16 cases of decompensation, 2 LTs, and 5 liver-related deaths were recorded. At baseline, around 50% of the patients were classified as being at risk of developing disease complications. MRCP+ metrics, particularly those describing the severity of bile duct dilatations, were correlated with all prognostic factors. Univariate analysis showed that MRCP+ metrics representing duct diameter, dilatations, and the percentage of ducts with strictures and/or dilatations were associated with survival. In a multivariable-adjusted analysis, the median duct diameter was significantly associated with survival (hazard ratio 10.9, 95% CI 1.3-90.3). Conclusions: MRCP+ metrics in people with PSC correlate with biochemical, elastographic, and radiological prognostic scores and are predictive of adverse outcome-free survival. Lay summary: In this study, we assessed in people with primary sclerosing cholangitis (PSC) the association of novel objective quantitative MRCP metrics automatically provided by a software tool (MRCP+) with prognostic scores and patient outcomes. We observed that MRCP+ metrics in people with PSC correlate with biochemical, elastographic, and radiological prognostic scores and are predictive of adverse outcome-free survival.

Adhesion-related small bowel obstruction: deep learning for automatic transition-zone detection by CT.

BACKGROUND: To train a machine-learning model to locate the transition zone (TZ) of adhesion-related small bowel obstruction (SBO) on CT scans. MATERIALS AND METHODS: We used 562 CTs performed in 2005-2018 in 404 patients with adhesion-related SBO. Annotation of the TZs was performed by experienced radiologists and trained residents using bounding boxes. Preprocessing involved using a pretrained model to extract the abdominopelvic region. We modeled TZ localization as a binary classification problem by splitting the abdominopelvic region into 125 patches. We then trained a neural network model to classify each patch as containing or not containing a TZ. We coupled this with a trained probabilistic estimation of presence of a TZ in each patch. The models were first evaluated by computing the area under the receiver operating characteristics curve (AUROC). Then, to assess the clinical benefit, we measured the proportion of total abdominopelvic volume classified as containing a TZ for several different false-negative rates. RESULTS: The probability of containing a TZ was highest for the hypogastric region (56.9%). The coupled classification network and probability mapping produced an AUROC of 0.93. For a 15% proportion of volume classified as containing TZs, the probability of highlighted patches containing a TZ was 92%. CONCLUSION: Modeling TZ localization by coupling convolutional neural network classification and probabilistic localization estimation shows the way to a possible automatic TZ detection, a complex radiological task with a major clinical impact.

Comparison and validation of seven white matter hyperintensities segmentation software in elderly patients.

BACKGROUND: Manual segmentation is currently the gold standard to assess white matter hyperintensities (WMH), but it is time consuming and subject to intra and inter-operator variability. PURPOSE: To compare automatic methods to segment white matter hyperintensities (WMH) in the elderly in order to assist radiologist and researchers in selecting the most relevant method for application on clinical or research data. MATERIAL AND METHODS: We studied a research dataset composed of 147 patients, including 97 patients from the Alzheimer's Disease Neuroimaging Initiative (ADNI) 2 database and 50 patients from ADNI 3 and a clinical routine dataset comprising 60 patients referred for cognitive impairment at the Pitié-Salpêtrière hospital (imaged using four different MRI machines). We used manual segmentation as the gold standard reference. Both manual and automatic segmentations were performed using FLAIR MRI. We compared seven freely available methods that produce segmentation mask and are usable by a radiologist without a strong knowledge of computer programming: LGA (Schmidt et al., 2012), LPA (Schmidt, 2017), BIANCA (Griffanti et al., 2016), UBO detector (Jiang et al., 2018), W2MHS (Ithapu et al., 2014), nicMSlesion (with and without retraining) (Valverde et al., 2019, 2017). The primary outcome for assessing segmentation accuracy was the Dice similarity coefficient (DSC) between the manual and the automatic segmentation software. Secondary outcomes included five other metrics. RESULTS: A deep learning approach, NicMSlesion, retrained on data from the research dataset ADNI, performed best on this research dataset (DSC: 0.595) and its DSC was significantly higher than that of all others. However, it ranked fifth on the clinical routine dataset and its performance severely dropped on data with artifacts. On the clinical routine dataset, the three top-ranked methods were LPA, SLS and BIANCA. Their performance did not differ significantly but was significantly higher than that of other methods. CONCLUSION: This work provides an objective comparison of methods for WMH segmentation. Results can be used by radiologists to select a tool.