Antithrombotic Therapy and Freedom From Bridging Stent Occlusion After Elective Branched Endovascular Repair: A Multicenter International Cohort Study.

CLINICAL IMPACT: Based on our study, no antithrombotic therapy is significantly associated with bridging stent occlusion, and no evidence of the superiority of other antithrombotic therapy exists. Nevertehless, due to the low number of bridging stent occlusions, this study can neither support nor reject the PRINCE2SS recommendations. Further studies with larger cohorts are needed to determine clear guideliness of the best antithrombotic treatment regimen after complex enfovascular aortic repair.

Interventional radiology for prevention and management of postpartum haemorrhage: a single centre retrospective cohort study.

PURPOSE: Postpartum haemorrhage (PPH) remains a leading cause of maternal death despite current medical management. Surgical interventions are still needed for refractory bleeding. Interventional radiology (IR) can be a successful intermediary that avoids the need for hysterectomy. Nevertheless, IR outcome data in a peripartum setting are limited. The objective of this study is to document the efficacy and safety of IR. METHODS: Retrospective study reviewed the records of consecutive patients who underwent peripartum IR from 01/01/2010 until 31/12/2020 in a tertiary academic centre. Patients were divided in a prophylactic and a therapeutic group. Information about interventions before and after IR, and IR specific complications was retrieved. Efficacy was defined by the number of transfusions and additional surgical interventions needed after IR, and safety was assessed by the incidence of IR related complications. RESULTS: Fifty-four patients, prophylactic group (n = 24) and therapeutic group (n = 30), were identified. In both groups, IR was successful with 1.5 ± 2.9 packed cells transfused post-IR (1.0 ± 2.1 prophylactic vs 1.9 ± 3.3 therapeutic; p = 0.261). Additional surgical interventions were required in n = 5 patients (9.2%), n = 1 (4.2%) in the prophylactic vs. n = 4 (13.3%) in the therapeutic group. Complications were reported in n = 12 patients (22.2%), n = 2 (8.3%) prophylactic vs. n = 10 (33.3%) in therapeutic group. Mostly minor complications, as puncture site hematoma or bleeding, were reported in n = 4 (7.4%). Severe complications as necrosis and metabolic complications were reported in n = 2 patients (3.9%). CONCLUSION: IR for prevention and treatment of PPH was highly successful and associated with minor complications.

Prospective, randomized, multicenter clinical study comparing a self-expanding covered stent to percutaneous transluminal angioplasty for treatment of upper extremity hemodialysis arteriovenous fistula stenosis.

Use of a covered stent after percutaneous transluminal angioplasty (PTA) was compared to PTA alone for treatment of upper extremity hemodialysis patients with arteriovenous fistula (AVF) stenoses. Patients with AVF stenosis of 50% or more and evidence of AVF dysfunction underwent treatment with PTA followed by randomization of 142 patients to include a covered stent or 138 patients with PTA alone. Primary outcomes were 30-day safety, powered for noninferiority, and six-month target lesion primary patency (TLPP), powered to test whether TLPP after covered-stent placement was superior to PTA alone. Twelve-month TLPP and six-month access circuit primary patency (ACPP) were also hypothesis tested while additional clinical outcomes were observed through two years. Safety was significantly non-inferior while six- and 12-month TLPP were each superior for the covered stent group compared to PTA alone (six months: 78.7% versus 55.8%; 12 months: 47.9% versus 21.2%, respectively). ACPP was not statistically different between groups at six-months. Observed differences at 24 months favored the covered-stent group: 28.4% better TLPP, fewer target-lesion reinterventions (1.6 ± 1.6 versus 2.8 ± 2.0), and a longer mean time between target-lesion reinterventions (380.4 ± 249.5 versus 217.6 ± 158.4 days). Thus, our multicenter, prospective, randomized study of a covered stent used to treat AVF stenosis demonstrated noninferior safety with better TLPP and fewer target-lesion reinterventions than PTA alone through 24 months.

Multicenter Randomized Controlled Trial of APERTO-Paclitaxel Drug-Eluting Balloon Angioplasty Versus Standard Percutaneous Transluminal Angioplasty in Dysfunctional Hemodialysis Grafts and Native Fistulae.

PURPOSE: to assess the safety and efficacy of APERTO-Paclitaxel-coated balloon angioplasty versus standard angioplasty for the treatment of dysfunctional hemodialysis shunts and native arteriovenous fistulae. METHODS: consecutive patients with dysfunctional dialysis related to underlying efferent vein stenosis were included and randomized 1:1 to either APERTO-paclitaxel drug-coated balloon (study arm) or standard percutaneous transluminal angioplasty (control arm). Primary endpoint is time from treatment until dialysis access dysfunction according to standardized Kidney Disease Outcomes Quality Initiative (KDOQI)-guidelines and assessed by Kaplan-Meier survival curves and tested for significance with log-rank analysis. Secondary endpoints include device, technical, and clinical success of the index angioplasty procedure. RESULTS: The study included 103 patients (n=51 study-group) with a de novo (n=33) dysfunctional native arteriovenous fistula (n=79) in the forearm (n=60). The majority of included patients were male with a mean age of 69.8 years, presenting with a dysfunctioning autologous arteriovenous fistula in the forearm. Device-related complications did not occur in any of the included patients. Functional hemodialysis access without need for re-intervention at 1 year after index procedure was found in n=10 (19.6%) and n=5 (9.6%) of patients treated with, respectively, paclitaxel drug-coated balloon and percutaneous transluminal angioplasty (p=0.612). A nonsignificant benefit of paclitaxel drug-coated balloon (n=5; 25%) over percutaneous transluminal angioplasty (n=1; 11%) was found (p=0.953) in de novo lesions in autologous fistulas. CONCLUSION: APERTO-paclitaxel drug-coated balloon is a safe balloon catheter to manage dysfunctional hemodialysis access; however, longer period of adequate hemodialysis circuit functioning after endovascular index stenosis treatment, using APERTO-paclitaxel drug-coated balloon versus percutaneous transluminal angioplasty could not be demonstrated. CLINICAL IMPACT: APERTO-paclitaxel drug-coated balloon catheter is a safe device to manage dysfunctional hemodialysis access. Compared to conventional angioplasty balloon, the APERTO drug-coated balloon will not result in longer period of adequate hemodialysis circuit functioning. A non-significant benefit of APERTO drug-coated balloon was found in de novo lesions in autologous fistulas.

Monographic Issue on Pulmonary Hypertension: Medical and Interventional Treatment for Chronic Thromboembolic Pulmonary Hypertension.

Chronic thromboembolic pulmonary hypertension (CTEPH) is a rare complication of acute pulmonary embolism. The reasons why clots do not resorb are incompletely understood, but the result is partial or complete fibrothrombotic obstruction of pulmonary arteries. A secondary microvasculopathy aggravates the pulmonary hypertension (PH) as a consequence of high flow and shear stress in the nonoccluded arteries. The treatment of CTEPH has long been purely surgical, but many patients were inoperable because of inaccessible lesions or severe comorbidities. Alternatives were developed, including medical therapy and more recently balloon pulmonary angioplasty (BPA). Depending on the generation of the obstructed vessels, the treatment will be surgical, up to the (sub)segmental level, or by BPA for more distal vessels. PH drugs are used to treat the microvasculopathy. The current paper describes the therapeutic management of inoperable patients: the medical approach with PH drugs used in mono- or combination therapy; the proper use of anticoagulants in CTEPH; the technique, indications, and results at short- and long-term of BPA; the multimodal approach for inoperable patients combining PH drugs and BPA; and the effects of rehabilitation. It shows the importance of a multidisciplinary approach to the disease.

Closing spontaneous portosystemic shunts in cirrhosis: Does it make sense? Does it work? What does it take?

Spontaneous portosystemic shunts (SPSS) are an often neglected cause of hepatic encephalopathy associated with cirrhosis. Nowadays, SPSS are considered as radiological biomarkers of clinically significant portal hypertension rather than the previous dogmatic perceived decompressive vessels. SPSS are not rare as they can be diagnosed in over 60% of the patients with cirrhosis by mere contrast-enhanced CT. Moreover, they are clinically relevant since they impact on all portal hypertensive related complications, in particular medically refractory HE, and represent an independent predictor of decompensation and mortality in cirrhosis, irrespective of the type of SPSS. Taken together, these elements warrant strategies to target these shunts directly which is currently is achieved via interventional radiology embolization. In this review, we discuss why it makes sense to tackle SPSS, how to do it and what it takes to do it right based on aggregated literature.

Posttraumatic hemorrhagic bladder rupture managed with transurethral foley catheter placement and bilateral transcatheter vesical artery embolization.

BACKGROUND: Outcome of minimally invasive treatment of posttraumatic, hemorrhagic bladder rupture is unknown. METHODS: A 41-year-old male presented at the emergency department with pelvic and scrotal pain and macroscopic hematuria after a motor vehicle accident. Contrast-enhanced computed tomography revealed an open book fracture and an arterial phase contrast media extravasation posterior to the symphysis pubis and anterior to the urethra-vesical junction. RESULTS: The open book fracture was treated with an external fixation and the persistent bleeding was managed with insertion of a Foley catheter and bilateral embolization of the vesical arteries. CONCLUSION: Minimally invasive treatment, including vesical artery embolization and placement of a Foley catheter can be effective in the treatment of posttraumatic, hemorrhagic bladder rupture.

Oxygen Pathway Limitations in Patients With Chronic Thromboembolic Pulmonary Hypertension.

BACKGROUND: Exertional intolerance is a limiting and often crippling symptom in patients with chronic thromboembolic pulmonary hypertension (CTEPH). Traditionally the pathogenesis has been attributed to central factors, including ventilation/perfusion mismatch, increased pulmonary vascular resistance, and right heart dysfunction and uncoupling. Pulmonary endarterectomy and balloon pulmonary angioplasty provide substantial improvement of functional status and hemodynamics. However, despite normalization of pulmonary hemodynamics, exercise capacity often does not return to age-predicted levels. By systematically evaluating the oxygen pathway, we aimed to elucidate the causes of functional limitations in patients with CTEPH before and after pulmonary vascular intervention. METHODS: Using exercise cardiac magnetic resonance imaging with simultaneous invasive hemodynamic monitoring, we sought to quantify the steps of the O2 transport cascade from the mouth to the mitochondria in patients with CTEPH (n=20) as compared with healthy participants (n=10). Furthermore, we evaluated the effect of pulmonary vascular intervention (pulmonary endarterectomy or balloon angioplasty) on the individual components of the cascade (n=10). RESULTS: Peak Vo2 (oxygen uptake) was significantly reduced in patients with CTEPH relative to controls (56±17 versus 112±20% of predicted; P<0.0001). The difference was attributable to impairments in multiple steps of the O2 cascade, including O2 delivery (product of cardiac output and arterial O2 content), skeletal muscle diffusion capacity, and pulmonary diffusion. The total O2 extracted in the periphery (ie, ΔAVo2 [arteriovenous O2 content difference]) was not different. After pulmonary vascular intervention, peak Vo2 increased significantly (from 12.5±4.0 to 17.8±7.5 mL/[kg·min]; P=0.036) but remained below age-predicted levels (70±11%). The O2 delivery was improved owing to an increase in peak cardiac output and lung diffusion capacity. However, peak exercise ΔAVo2 was unchanged, as was skeletal muscle diffusion capacity. CONCLUSIONS: We demonstrated that patients with CTEPH have significant impairment of all steps in the O2 use cascade, resulting in markedly impaired exercise capacity. Pulmonary vascular intervention increased peak Vo2 by partly correcting O2 delivery but had no effect on abnormalities in peripheral O2 extraction. This suggests that current interventions only partially address patients' limitations and that additional therapies may improve functional capacity.

Long-term outcome results after endovascular aortoiliac aneurysm repair with the bifurcated EXCLUDER Endoprosthesis.

OBJECTIVE: To report the long-term outcome of patients presenting with an aortic, aortoiliac, or isolated common iliac aneurysm treated with the bifurcated EXCLUDER Endoprosthesis. Furthermore, potential differences in late outcome results between the original- and low-permeability endoprosthesis were analyzed. METHODS: A retrospective analysis of prospectively collected data of 182 patients who underwent endovascular aneurysm repair with the EXCLUDER Endoprosthesis between June 1998 and October 2015 in an academic, tertiary care center for aortic disease was performed. Patient follow-up was from 3 to 20 years (mean follow-up of 6.9 years). Primary end points were overall survival and reintervention-free survival. Secondary end points were device-related complications, endoleaks, and reinterventions. RESULTS: Overall survival at 5, 10, and 15 years was 72.8%, 42.1%, and 12.2%, respectively, with no aneurysm-related mortality and no difference in overall survival between the original- vs low-permeability endoprosthesis group (P = .617). Freedom from type I endoleak at 5 years was 94.8%. No new type I endoleak was detected beyond the 5-year follow-up mark. No type III endoleak was identified. Reintervention-free survival was 83.6%, 66.7%, and 66.7% at 5-, 10-, and 15-year follow-up, respectively. There was a significant difference in intervention-free survival between the original- vs low-permeability endoprosthesis group (P = .029) and after the 5-year follow-up mark. In addition, patients with the low-permeability endoprosthesis showed significantly fewer device-related complications (P = .002) and endoleaks (P = .005). CONCLUSIONS: Endovascular aneurysm repair using the EXCLUDER Endoprosthesis is effective and durable on long-term follow-up, with acceptably low device-related complications and reinterventions. The low-permeability endoprosthesis was associated with significantly fewer new device-related complications and endoleaks after 5 years of follow-up.

Glue embolization of a bleeding lingual artery pseudoaneurysm related to tongue surgery.

Postoperative lingual artery pseudoaneurysm associated with hemorrhage is a rare entity mainly found after tonsillectomy or laryngectomy. In this report, a symptomatic, lingual artery pseudoaneurysm after repeated base of tongue surgery was identified on computed tomography and successfully managed with catheter-directed glue-embolization.

Direct EBUS-guided transtracheal lymphosclerosis for plastic bronchitis after Fontan.

We report on a new puncture technique with direct transtracheal mediastinal lymphatic access to treat plastic bronchitis after Fontan repair. High resolution contrast-enhanced spiral CT identified enlarged lymph nodes in the paratracheal region. Inguinal intranodal Gadolinium Dynamic Contrast-enhanced Magnetic Resonance lymphangiography (DCMRL) confirmed the pathologic centrifugal lymph flow passing through these lymph nodes before leaking into the bronchial tree. The abnormal hypertrophic paratracheal, subcarinal, and hilar lymph nodes were punctured with a 22G needle through an endobronchial ultrasound bronchoscope. Occlusion of the lymph vessels was obtained by injecting a mixture of lipiodol/NBCA N-butyl cyanoacrylate (Histoacryl) 5/1 under fluoroscopic control. There was a total remission of PB with now 10 months of follow-up.

Hemoptysis after Lung Transplantation Caused by Bronchial Arterial Neovascularization: Angiographic Analysis and Successful Embolization.

This report discusses 3 bilateral lung transplant recipients (2 female, 1 male) who presented with late hemoptysis (10 y, 18 y, and 19 y after transplantation). All patients had a history of pulmonary infections, bronchiectasis, and/or Aspergillus infection. Arteriography, through catherization of the common femoral artery, demonstrated spontaneous bronchial and systemic neovascularization arising from the thyrocervical trunk, internal thoracic artery, intercostal arteries, and dorsal scapular artery. Embolization was performed with microspheres, polyvinyl alcohol microparticles, and/or glue and effectively terminated hemoptysis. One patient died 10 d later as a result of fungal infection, and the 2 others remained in stable condition (18- and 26-mo postembolization follow-up available).

Intranodal Lymphangiography With High-Dose Ethiodized Oil Shows Efficient Results in Patients With Refractory, High-Output Postsurgical Chylothorax: A Retrospective Study.

OBJECTIVE. The purpose of this study is to evaluate the safety and efficacy of intranodal lymphangiography (INL) with high-dose ethiodized oil in patients with postoperative refractory chylothorax. MATERIALS AND METHODS. A retrospective review of a cohort of 18 patients with refractory postoperative chylothorax seen between May 2015 and March 2019 was conducted. All patients underwent intranodal lymphangiography with high doses of ethiodized oil (mean, 75 mL; range, 40-140 mL). The following information was gathered from the institutional database: patient demographics, type of surgery, output volumes, interval between surgery and lymphangiography, imaging results, amount of ethiodized oil injected, clinical success, and time to resolution. RESULTS. Of the 18 patients, 11 (61%) had previously undergone thoracic duct ligation, and seven (39%) had not. A lymphatic leak was confirmed by lymphangiography in 12 of 18 patients (67%). A total of five patients underwent a second session of INL, which was successful in three of the patients (60%). Removal of all chest tubes was possible in 15 of 18 patients (83%) after a mean of 12 days (range, 1-25 days). Two patients had an anastomotic leak develop after esophagectomy and died with their chest tubes in situ. One patient underwent thoracic duct ligation after two failed INL procedures. No complications were recorded. CONCLUSION. INL with a high dose of injected ethiodized oil is a safe and effective procedure for the management of postsurgical refractory chylothorax, with chest tube removal in 83% of patients.

Recanalization of portal axis after cavoportal hemitransposition in a liver transplant recipient with complete splanchnic thrombosis.

BACKGROUND: Diffuse splanchnic thrombosis may render standard LTx difficult or even technically impossible. A 19-year-old woman with acute-on-chronic Budd-Chiari syndrome and complete splanchnic thrombosis underwent conventional LTx. Only limited anatomical portal inflow could be restored, and urgent re-transplantation for recurrent splanchnic vein thrombosis became necessary. METHODS: At re-transplant, and in addition to the reestablishment of some portal inflow through the preserved original porto (native)-portal (graft) connection, a cavoportal shunt was created (first partial via 30% tapering of the vena cava, but eventually complete by total occlusion of the vena cava). RESULTS: The postoperative course was then uneventful, and interestingly, the native portomesenteric axis gradually reopened. Two years post-transplant, the liver graft is perfused via both physiological and non-physiological sources. Liver function is normal. There is no IVC syndrome and no residual PHT. She is leading a normal life. CONCLUSION: Creation of CPHT, in addition to the preservation of portal inflow from the native splanchnic system, should be considered in patients with diffuse splanchnic thrombosis, when sufficient physiological portal inflow cannot be restored at the time of LTx, but in whom the splanchnic circulation may reopen up later.

Predictive Factors of Operative Need in Medically Managed Type B Aortic Dissections.

BACKGROUND: Aneurysmal degeneration of medically managed type B aortic dissection (TBAD) can be a life-threatening condition. Preventive thoracic endovascular aorta repair (TEVAR) in patients at risk could potentially be beneficial. The aim of this study was to examine the predictors for late aneurysmal dilatation after TBAD. METHODS: A retrospective study was conducted on 82 patients with medically managed acute TBAD for a minimum of 14 days. Relevant demographic, biochemical, and radiographic variables at presentation were studied. The aortic dissection risk calculator tool developed by Sailer et al., predicting the risk of adverse events after aortic dissection based on demographic and radiographic variables at presentation, was tested retrospectively. RESULTS: With a median follow-up of 36 months (range 13-68), 25 (30.5%) patients underwent surgery (92% TEVAR). A larger initial aortic and false lumen diameter as well as a greater distal extension of the dissection was associated with higher need for surgery (respectively, P = 0.003, P = 0.004, and P = 0.001). We observed higher growth rates of maximum aortic diameter in patients with a greater distal extension of the dissection, larger false lumen diameters and false lumen outflow, and entry tears located at the inner aortic arch (respectively, P = 0.001, P = 0.005, P = 0.001 and P = 0.014). No significant correlations could be found for the risks provided by the calculator tool. CONCLUSIONS: The initial maximum aortic diameter of TBAD is a key predictor for aortic growth. Furthermore, the distal extension of the dissection also seems to play an important role in late aneurysmal degeneration. However, we were not able to confirm the added value of the risk calculator tool in our study group.

Factors predicting long-term outcomes of percutaneous angioplasty and stenting of the superior mesenteric artery for chronic mesenteric ischemia.

Background: To analyse the long-term outcomes of percutaneous angioplasty and stenting of the superior mesenteric artery (SMA) in the treatment of chronic mesenteric ischemia (CMI), and to assess predictive factors for a better clinical outcome. Patients and methods: Retrospective analysis of 76 consecutive patients, treated percutaneously for CMI between January 1999 and January 2018 and followed up until the end of 2018. Patients' pre-, peri- and post-interventional clinical and radiological data were gathered from the institutional electronic medical records. The Kaplan Meier method with log rank test or the Cox model were used to analyse overall survival; the cumulative incidence function with Pepe and Mori test or the Fine and Grey model were used to analyse relapse-free survival, considering death as a competing event. Results: Seventy-six consecutive patients with a mean age of 72 years were included in the study. Catheter-angiography revealed an ostial or non-ostial >90% stenosis in n=23 (29.7%) and n=53 (69.7%) of included patients, respectively. Immediate clinical success was achieved in n=68 (89.5%), and procedural complications were observed in n=13 (17.1%) patients. Long-term follow-up revealed relapse of symptoms in n=21 (28.8%) patients, and overall survival estimates are 81.8%, 57.0% and 28.2% after two, five and ten years of follow-up, respectively. A trend towards longer relapse-free survival was found in the circumferential stenosis group (78.2% at five years) compared with the non-circumferential stenosis group (55.5%) (P=0.063). Conclusions: Angioplasty and stenting of the SMA for CMI is relatively safe and effective despite a substantial number of patients experiencing clinical relapse over time. Patients with focal, circumferential stenosis might have longer relapse-free survival than patients with non-circumferential stenosis.

The impact of type 1a endoleak on the long-term outcome after EVAR.

OBJECTIVE: Endoleaks remains a major determinant of outcome after endovascular aortic repair (EVAR) for infrarenal abdominal aortic aneurysms. The objective of this study is to evaluate the long-term impact of proximal type 1 endoleak encountered at the case end of the EVAR procedure or during follow-up. METHODS: All patients treated with EVAR between February 2001 and August 2017 in our institution were included. Data were collected retrospectively. A descriptive analysis of subgroups with type 1a endoleak at case end or with type 1a endoleak first encountered during follow-up was performed. RESULTS: In total, 468 patients were included for analysis. At the case end of the EVAR procedure, in 5.3% (25/468) of patients, a type 1a endoleak was seen. In 72% of cases (18/25) the type 1a endoleak spontaneously disappeared on the first follow-up computed tomography angiography and never recurred. At the end of the follow-up period, no patient in this subgroup died with a type 1a endoleak on follow-up. In 3.4% (16/468) of all cases, a type 1a endoleak was encountered for the first time during follow-up. In 87.5% (14/16) of these cases, reintervention was performed. Two patients died with a persisting type 1a endoleak during follow-up. CONCLUSIONS: Clinicians should differentiate between the type 1a endoleak at the case end and the type 1a endoleak first occurring during follow-up. For type 1a endoleak at the case end, expectant management can be adopted. Delayed type 1a endoleak seen during follow-up rarely seals spontaneously and needs reintervention in the majority of cases.

Use of hyperglycemic clamp to assess pancreatectomy and islet cell autotransplant in patient with heterotaxy syndrome and dorsal pancreas agenesis leading to chronic pancreatitis.

Patients with heterotaxy syndrome (HS) can present with an associated complete dorsal pancreas agenesis (DPA). They are considered to be at increased risk for developing diabetes due to a reduced functional beta cell mass (FBM) as well as for chronic pancreatitis leading to unmanageable pain. We report the case of a young woman with chronic pancreatitis due to HS and associated DPA. She presented with a severe persisting upper abdominal pain refractory to nonsurgical treatment. Unlike in previously reported cases, she had a high FBM (ie, 150% of normoglycemic controls) as determined by hyperglycemic clamp. She underwent a total pancreatectomy followed within 24 hours by an intraportal autologous islet cell transplant containing 4 × 106 beta cells (4700 islet equivalent)/kg body weight. After surgery, the pain resolved, eliminating the need for analgesics. The intraportal implant established an adequate FBM (72% of controls at posttransplant month 2), achieving glycemic control without need for insulin administration. A hyperglycemic clamp can assess the utility and efficacy of an intraportal islet cell autotransplant following total pancreatectomy in patients with HS and complete DPA.

Total area of spontaneous portosystemic shunts independently predicts hepatic encephalopathy and mortality in liver cirrhosis.

BACKGROUND & AIMS: Spontaneous portosystemic shunts (SPSS) frequently develop in liver cirrhosis. Recent data suggested that the presence of a single large SPSS is associated with complications, especially overt hepatic encephalopathy (oHE). However, the presence of >1 SPSS is common. This study evaluates the impact of total cross-sectional SPSS area (TSA) on outcomes in patients with liver cirrhosis. METHODS: In this retrospective international multicentric study, CT scans of 908 cirrhotic patients with SPSS were evaluated for TSA. Clinical and laboratory data were recorded. Each detected SPSS radius was measured and TSA calculated. One-year survival was the primary endpoint and acute decompensation (oHE, variceal bleeding, ascites) was the secondary endpoint. RESULTS: A total of 301 patients (169 male) were included in the training cohort. Thirty percent of all patients presented with >1 SPSS. A TSA cut-off of 83 mm2 was used to classify patients with small or large TSA (S-/L-TSA). Patients with L-TSA presented with higher model for end-stage liver disease score (11 vs. 14) and more commonly had a history of oHE (12% vs. 21%, p <0.05). During follow-up, patients with L-TSA experienced more oHE episodes (33% vs. 47%, p <0.05) and had lower 1-year survival than those with S-TSA (84% vs. 69%, p <0.001). Multivariate analysis identified L-TSA (hazard ratio 1.66; 95% CI 1.02-2.70, p <0.05) as an independent predictor of mortality. An independent multicentric validation cohort of 607 patients confirmed that patients with L-TSA had lower 1-year survival (77% vs. 64%, p <0.001) and more oHE development (35% vs. 49%, p <0.001) than those with S-TSA. CONCLUSION: This study suggests that TSA >83 mm2 increases the risk for oHE and mortality in patients with cirrhosis. Our results support the clinical use of TSA/SPSS for risk stratification and decision-making in the management of patients with cirrhosis. LAY SUMMARY: The prevalence of spontaneous portosystemic shunts (SPSS) is higher in patients with more advanced chronic liver disease. The presence of more than 1 SPSS is common in advanced chronic liver disease and is associated with the development of hepatic encephalopathy. This study shows that total cross-sectional SPSS area (rather than diameter of the single largest SPSS) predicts survival in patients with advanced chronic liver disease. Our results support the clinical use of total cross-sectional SPSS area for risk stratification and decision-making in the management of SPSS.

Whole liver segmentation based on deep learning and manual adjustment for clinical use in SIRT.

PURPOSE: In selective internal radiation therapy (SIRT), an accurate total liver segmentation is required for activity prescription and absorbed dose calculation. Our goal was to investigate the feasibility of using automatic liver segmentation based on a convolutional neural network (CNN) for CT imaging in SIRT, and the ability of CNN to reduce inter-observer variability of the segmentation. METHODS: A multi-scale CNN was modified for liver segmentation for SIRT patients. The CNN model was trained with 139 datasets from three liver segmentation challenges and 12 SIRT patient datasets from our hospital. Validation was performed on 13 SIRT datasets and 12 challenge datasets. The model was tested on 40 SIRT datasets. One expert manually delineated the livers and adjusted the liver segmentations from CNN for 40 test SIRT datasets. Another expert performed the same tasks for 20 datasets randomly selected from the 40 SIRT datasets. The CNN segmentations were compared with the manual and adjusted segmentations from the experts. The difference between the manual segmentations was compared with the difference between the adjusted segmentations to investigate the inter-observer variability. Segmentation difference was evaluated through dice similarity coefficient (DSC), volume ratio (RV), mean surface distance (MSD), and Hausdorff distance (HD). RESULTS: The CNN segmentation achieved a median DSC of 0.94 with the manual segmentation and of 0.98 with the manually corrected CNN segmentation, respectively. The DSC between the adjusted segmentations is 0.98, which is 0.04 higher than the DSC between the manual segmentations. CONCLUSION: The CNN model achieved good liver segmentations on CT images of good image quality, with relatively normal liver shapes and low tumor burden. 87.5% of the 40 CNN segmentations only needed slight adjustments for clinical use. However, the trained model failed on SIRT data with low dose or contrast, lesions with large density difference from their surroundings, and abnormal liver position and shape. The abovementioned scenarios were not adequately represented in the training data. Despite this limitation, the current CNN is already a useful clinical tool which improves inter-observer agreement and therefore contributes to the standardization of the dosimetry. A further improvement is expected when the CNN will be trained with more data from SIRT patients.