Search Results (624)

Background: Deep vein thrombosis (DVT) is a serious thromboinflammatory complication of acute ischemic stroke (AIS). The true incidence, mechanistic risk factors, and optimal prophylactic strategies remain uncertain, particularly in the era of reperfusion therapy. Methods: This systematic review and meta-analysis (IRIS-DVT) searched PubMed, Embase, Cochrane, Scopus, and Web of Science for studies reporting DVT incidence, risk factors, or prophylaxis in AIS (2004–2025). Random-effects models were used to generate pooled prevalence and effect estimates, and the certainty of evidence was graded using the GRADE framework. Results: Forty-two studies (n = 6,051,729 patients) were included. The pooled prevalence of DVT was 7% (95% CI, 6–9%), approximately seventy-fold higher than in the general population, with wide heterogeneity influenced by screening timing and diagnostic modality. Pathophysiological risk factors included higher stroke severity (NIHSS; SMD 0.41; 95% CI, 0.38–0.43), older age (SMD 0.32; 95% CI, 0.18–0.46), elevated D-dimer (SMD 0.55; 95% CI, 0.38–0.72), female sex (OR 1.33; 95% CI, 1.19–1.50), and malignancy (OR 2.69; 95% CI, 1.56–5.22), supported by moderate-certainty evidence. Respiratory infection and admission hyperglycemia showed weaker, low-certainty associations. Traditional vascular risk factors (hypertension, diabetes, atrial fibrillation, dyslipidemia) were not significantly related to DVT risk. Evidence for prophylaxis with low-molecular-weight heparin, direct oral anticoagulants, or intermittent pneumatic compression was limited and graded very low certainty. Conclusions: DVT complicates approximately one in fourteen AIS cases, reflecting a distinct thromboinflammatory process driven more by acute neurological severity, systemic hypercoagulability, and malignancy than by conventional vascular risk factors. Early systematic screening (≤72 h) and consistent use of mechanical prophylaxis are warranted. Dedicated AIS-specific mechanistic and interventional trials are urgently needed to refine prevention strategies and improve post-stroke outcomes. Full article

Background/Objectives: The variability in the structure of the deep cerebral venous system in neonates is increasingly recognised, as are the vascular structural factors involved in the development of the germinal matrix/intraventricular haemorrhage (GM/IVH) in premature infants. We aimed to characterise the ultrasound patterns of these veins in different categories of newborns and to assess if there is a correlation between certain patterns and angles and the presence of GM/IVH. Methods: One hundred neonates (68 at-term and 32 preterm) were included in this research. The pattern of venous drainage and the angle at the confluence between the terminal vein (TV) and internal cerebral vein (ICV) were identified on coronal sections through the anterior fontanel. The normal pattern was considered as that in which the confluence between the TV and the ICV could be identified, and the atypical pattern was considered the situation in which no confluence or terminal vein was identified. Results: There was no statistically significant difference regarding the normal or atypical venous patterns between the groups (p < 0.443), neither regarding the angles between TV and ICV between term and preterm neonates (p < 0.279—left; p < 0.718—right), and singletons and twins (p < 0.745 left; p < 0.418 right), or between the angles on the left and on the right in the whole group (p < 0.121 and the subgroups of term (p < 0.440) and preterm neonates (p < 0.092). The mean value of the angle at the confluence between the TV and the ICV on the left, was significantly lower in the premature infants with GM/IVH (124.90° vs. 137.02°; p = 0.012), being a good predictor for the occurrence of the lesion (AUC = 0.793; IC 95%: 0.580–1.006; p = 0.018), with a sensitivity of 79%, a specificity of 67%, and a cut-off value of 126.90°. In patients with GM/IVH, the angle was significantly lower on the side with the haemorrhage than on the side without haemorrhage (p < 0.043). Conclusions: There is no difference in the central venous pattern or angle at the confluence of the TV and the ICV between different categories of neonates. The angle at the confluence between the TV and ICV could identify the cases at risk for GM/IVH as well as the side of occurrence of the haemorrhage, offering the opportunity of developing personalised prevention strategies. The lack of an MRI comparator of these measurements limits the practical importance of this study. Full article

Immune cells like neutrophils, monocytes/macrophages, and lymphocytes play key roles in the development, progression, and resolution of deep vein thrombosis (DVT) by contributing to inflammation, coagulation, and fibrinolysis. IFN-γ, a cytokine mainly secreted by natural killer (NK) and T cells, is a critical factor in DVT pathogenesis. It links immune responses to coagulation activation by promoting endothelial activation, leukocyte recruitment, cytokine release, and coagulation imbalance. Its strong pro-inflammatory and prothrombotic effects make IFN-γ a promising target for DVT treatment beyond standard anticoagulants. Exploring ways to block IFN-γ signaling or its downstream effects could open doors to novel therapies for DVT, aiding in resolution and preventing post-thrombotic complications. This review delves into DVT pathophysiology, diagnostics, and management, emphasizing the importance of targeting immune cells and IFN-γ to advance treatment options. Full article

Background: End-stage renal disease (ESRD) patients on haemodialysis present a high burden of cardiovascular comorbidities and require anticoagulation, which increases bleeding risk. Methods: We performed a retrospective observational study (2021–2024) in the Haemodialysis Centre of The Clinical Emergency Hospital of Constanta County, Romania, including 50 adults with stage G5 CKD on haemodialysis for ≥3 months and receiving anticoagulant therapy. We collected from electronic medical records detailed demographic data (age, sex, place of residence), comorbidities (hypertension, atrial fibrillation, ischaemic heart disease, diabetes, deep-vein thrombosis, stroke, myocardial infarction, pulmonary embolism, cirrhosis), lifestyle factors (smoking and alcohol consumption), vascular access type (arteriovenous fistula or central venous catheter) and laboratory parameters (haemoglobin, haematocrit, creatinine, albumin, total protein, electrolytes, LDL- and HDL-cholesterol, total cholesterol, INR, APTT, D-dimer, BNP, CK-MB, troponin). All laboratory units were standardised and checked for plausibility. Results: Median age was 71 years; 48% were female. The most common comorbidities were: hypertension (100%), atrial fibrillation (100%) and ischaemic heart disease (62–81%). Patients exhibited severe anaemia (mean Hb ~9.7 g/dL), nephrotic-range proteinuria, hypoalbuminaemia, and impaired coagulation profiles (INR ~1.8–1.9; prolonged APTT in men). Female patients had higher platelet counts and D-dimer levels, suggesting a stronger prothrombotic profile, while males showed longer APTT. Cardiovascular strain was reflected by elevated BNP in men and also troponin/CK-MB. Correlations included smoking with leukocytosis, alcohol with increased urine density, diabetes with higher urea and lower protein, and subtherapeutic INR in cerebrovascular disease. Conclusions: Patients with ESRD on haemodialysis and anticoagulant therapy display a complex interplay of cardiovascular comorbidities, anemia, overlapping thrombotic and bleeding risks, with sex-specific differences. Therefore, systematic monitoring of proteinuria, haemoglobin, D-dimer, and coagulation markers is crucial to balance thrombotic and bleeding risks. Objective: To characterise the clinical and paraclinical profile and comorbidity–laboratory correlations of ESRD patients undergoing haemodialysis and anticoagulant therapy. Full article

Background and Objectives: Performance of robot-assisted total knee arthroplasty (TKA) procedures has continued to increase in popularity. However, tourniquet use is necessary for longer periods of time in robot-assisted TKA than conventional manual TKA because the robot-assisted procedure requires an additional registration process. The use of tourniquets for long periods increases the risk of hidden blood loss and ischemic soft tissue injury in the lower extremity. The purpose of this study was to compare the value of performing robot-assisted TKA without the use of a tourniquet to that of performing this surgery with the use of a tourniquet. Parameters we assessed were blood loss, degree of postoperative thigh and knee pain, and occurrence of early post-operative complications. Materials and Methods: Data from 100 consecutive patients who underwent primary unilateral robot-assisted TKA between July 2024 and July 2025 were included in this study’s analyses. Patients were divided into three groups chronologically. The first 29 patients comprised group 1, the early tourniquet group; the next 30 patients were assigned to group 2, the no tourniquet group; and group 3 was the late tourniquet group and comprised the remaining 41 subjects. However, because allocation was chronological rather than randomized, the outcomes of later groups may partly reflect the surgeon’s accumulated experience (learning curve), which should be considered when interpreting the results. The primary outcome measure was estimated blood loss (EBL). The secondary outcome measures included transfusion rate, visual analog scale (VAS) pain scores for the knee and thigh on the third postoperative day, readmission rate due to surgical complications, superficial and deep infection rate, length of operation, and length of tourniquet use. Results: Group 2 participants, the no tourniquet participants, experienced significantly greater EBL on postoperative days (PODs) 1, 2, and 3 compared to the subjects assigned to groups 1 and 3 (p = 0.003, p < 0.001, and p = 0.005, respectively). However, there were no significant differences in transfusion rates (p = 0.290) among the 3 groups. VAS scores for knee and thigh pain were also not significantly different among the three groups (all p-values > 0.05). Three patients in group 1 (10.3%), one patient in group 2 (3.3%), and one patient in group 3 (2.4%) were readmitted for complications related to wound healing (p = 0.289). Additionally, two patients in group 1 developed superficial wound infections from which the causative bacteria were cultured. No infections were observed in the other groups (p = 0.082), however. Two patients in group 1 and two patients in group 2 experienced symptomatic deep vein thrombosis (DVT) (p = 0.235). No group 3 patients experienced DVT, and only one patient in group 2 was confirmed with DVT using an enhanced CT scan (p = 0.308). Group 3 patients had shorter lengths of surgery (p < 0.001) than group 1 and 2 patients and had shorter periods of tourniquet use (p = 0.034) than group 1 patients. Conclusions: Tourniquet non-use in robot-assisted TKA surgeries was associated with greater EBL in acute postoperative periods, but this finding was not accompanied by any change in transfusion rate. Tourniquet non-use was not clinically beneficial for reducing immediate postoperative thigh and knee pain or reducing the prevalence of early post-operative complications. Tourniquet use in robot-assisted TKA may be beneficial because of the advantages its use provides in maintaining a clear surgical field and in facilitating the cementing process. Full article

Objectives: This study aimed to determine the incidence of postoperative deep vein thrombosis (DVT) and pulmonary embolism (PE) following orthopedic surgeries and to identify independent clinical, laboratory, and procedural factors associated with thromboembolic risk. Materials and Methods: A retrospective cohort analysis was conducted on 300 patients who underwent elective or emergency orthopedic surgeries (hip/knee arthroplasty, fracture fixation, and spinal procedures) between January 2020 and December 2024 at two tertiary centers. Demographic, clinical, and biochemical data were collected. Patients were stratified into two groups: those who developed DVT/PE and those who did not. Univariate analyses were performed to identify significant factors, and a multivariate logistic regression model with stepwise variable selection was applied in accordance with the events-per-variable (EPV) criterion. Receiver operating characteristic (ROC) curve analyses were conducted to evaluate the discriminative performance of significant predictors. Results: Among 300 patients who underwent orthopedic surgery, postoperative deep vein thrombosis (DVT) and/or pulmonary embolism (PE) occurred in 50 cases (16.7%). Patients who developed thromboembolic events were older (72.5 ± 8.7 vs. 65.2 ± 10.1 years, p < 0.001), had higher body mass index (32.1 ± 5.3 vs. 28.3 ± 4.5 kg/m², p < 0.001), and showed a greater prevalence of diabetes mellitus (40% vs. 20%, p < 0.01) and chronic kidney disease (24% vs. 10%, p < 0.001) compared to those without DVT/PE. Laboratory analyses revealed significantly elevated neutrophil count, D-dimer, C-reactive protein (CRP), glucose, and troponin levels in the DVT/PE group. In the stepwise multivariate logistic regression model, age (OR = 1.44, p = 0.003), diabetes mellitus (OR = 2.88, p = 0.046), chronic kidney disease (OR = 2.33, p = 0.014), D-dimer (OR = 2.15, p = 0.019), and immobilization duration (OR = 2.21, p = 0.028) emerged as independent predictors of thromboembolic events. ROC analysis revealed that D-dimer > 0.9 mg/L had the highest discriminative performance (AUC = 0.89, sensitivity 88%, specificity 84%, p = 0.003), followed by troponin > 0.5 U/L (AUC = 0.86, p = 0.005), immobilization > 3 days (AUC = 0.82, p = 0.012), and age > 65 years (AUC = 0.74, p = 0.021). Conclusions: DVT and PE remain significant postoperative complications with a multifactorial etiology in orthopedic surgeries. Advanced age, comorbidities (such as diabetes mellitus and chronic kidney disease), and elevated inflammatory and metabolic markers (including neutrophil count, glucose, CRP, and D-dimer), together with procedural factors like prolonged immobilization, were identified as independent risk factors. Early recognition of these high-risk features and implementation of individualized prophylaxis strategies may improve postoperative outcomes and reduce thromboembolic risk. Full article

Deep vein thrombosis is a condition associated with substantial morbidity and a high risk of pulmonary embolism, underscoring the need for rapid and reliable diagnostic solutions. Although machine learning and deep learning techniques are increasingly being applied for clinical decision support, comprehensive analyses of their contributions to early detection, risk prediction, and monitoring remain limited. Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses 2020 guidelines, we conducted a systematic search in ScienceDirect, IEEE Xplore, Scopus, and Web of Science for studies published between January 2014 and March 2025. Eligible studies applied machine learning or deep learning approaches for the early prediction, monitoring, or risk assessment of deep vein thrombosis, or described reference datasets for algorithm development. Two authors independently extracted data and evaluated methodological quality using the Quality Assessment of Diagnostic Accuracy Studies-2 framework. The included studies were categorized into four domains: Early prediction, monitoring, risk assessment, and reference datasets. In total, 66 studies met the inclusion criteria. Recent advances include deep learning-assisted ultrasound interpretation and real-time implementation of machine learning algorithms. While most studies demonstrated a low overall risk of bias, recurring limitations were identified in terms of patient selection, reporting practices, and validation strategies. Dataset harmonization and external validation were infrequently performed, and documentation of data provenance and class imbalance handling was inconsistent. Machine learning and deep learning approaches demonstrate considerable potential to accelerate accurate diagnoses and facilitate individualized risk stratification; however, their translation into routine practice requires standardized datasets, rigorous external validation, and integration into existing clinical workflows. This review consolidates a decade of research, links methodological quality to clinical applicability, and provides a task-oriented roadmap for advancing machine learning-enabled diagnostics and monitoring in the context of deep vein thrombosis. Full article

Background: Few data are available on the outcomes of patients with venous thromboembolism (VTE) on long-term reduced dose of direct oral anticoagulants (DOACs). We evaluated the effectiveness and safety of reduced dose of DOACs for the extended treatment of VTE. Methods: In this monocenter, ambispective cohort study, 140 patients receiving a reduced dose of DOACs for VTE were included. The primary outcomes were recurrent VTE, major bleeding and clinically relevant non-major bleeding. The secondary outcomes were arterial events and minor bleedings. The incidence of the primary outcomes was calculated. The rate for secondary outcomes was descriptively reported. Results: The mean age of the overall cohort was 72 years. Half of the patients were female, 51.4% had a persistent risk factor, 40.0% an unprovoked VTE, and 8.6% a minor transient risk factor. Most patients had lower extremity deep vein thrombosis with or without pulmonary embolism (55.0%) and received apixaban (73.6%) or rivaroxaban (14.3%) for a mean duration of 2.7 years. Regarding the primary outcomes, there was one recurrent VTE (0.7%), four major bleedings (2.9%) and two clinically relevant non-major bleedings (1.4%). Regarding the secondary outcomes, there were four acute ischemic strokes (2.9%) and two minor bleedings (1.4%). Conclusions: Reduced dose of DOACs was associated with a low rate of recurrent VTE and an acceptably low rate of bleeding complications. The rate of arterial events during follow-up suggests the need for an assessment of cardiovascular risk factors in this study population. Full article

Background: Post-thrombotic syndrome (PTS) is a frequent complication of deep vein thrombosis (DVT), with significant clinical and quality-of-life implications. Endovascular techniques have emerged as potential strategies to reduce PTS severity in selected patients, though evidence remains inconclusive. Methods: We conducted a multicenter, retrospective study including 176 patients with iliac or iliofemoral DVT from four hospitals in Murcia, Spain. Patients were treated either with anticoagulation alone (n = 121) or with endovascular techniques followed by anticoagulation (n = 55). The primary outcome was the presence of PTS at 12 months, defined by the presence of ≥5 signs/symptoms from the Villalta scale. Multivariate analysis was performed to identify independent predictors of PTS. Results: No significant differences were observed in the overall prevalence of PTS between the endovascular and anticoagulation-only groups at 12 months (25.4% vs. 23.2%, p = 0.63). However, edema and skin hyperpigmentation were significantly more frequent in the anticoagulation-only group. Patients treated with endovascular techniques had lower rates of residual thrombosis at follow-up (34.6% vs. 53.9%, p = 0.03). Multivariate analysis identified residual thrombosis (OR 6.3, 95% CI 2.74–14.5), persistently elevated D-dimer (OR 3.26, 95% CI 1.1–9.7), and age (OR 1.02, 95% CI 1.003–1.042) as independent predictors of PTS. Mortality was significantly higher in the anticoagulation-only group, largely driven by a higher cancer prevalence. Conclusions: Endovascular techniques did not significantly reduce the overall incidence of PTS but were associated with lower rates of specific signs such as edema and hyperpigmentation. Residual thrombosis remains a key predictor of PTS and may represent a modifiable therapeutic target. Full article

Background and Clinical Significance: Arterial and venous thromboses are typically distinct clinical entities, each governed by unique pathophysiological mechanisms. The concurrent manifestation of both, particularly in the setting of massive pulmonary embolism (PE), is exceptionally rare and poses significant diagnostic and therapeutic challenges. Case Presentation: This report describes a 61-year-old male with well-controlled hypertension and type 2 diabetes who developed extensive thromboses involving deep vein thrombosis (DVT) of the right popliteal vein, arterial thrombosis of the left iliac artery, and massive PE. The patient was initially managed conservatively, in accordance with the European Society of Cardiology (ESC) 2019 Guidelines for Acute PE, using unfractionated heparin (UFH), low-molecular-weight heparin, a direct oral anticoagulant (DOAC), and adjunctive therapy. This approach was chosen due to the absence of hemodynamic instability. However, given failed percutaneous revascularization and persistent arterial occlusion, surgical thromboendarterectomy (TEA) was ultimately required. Post hoc genetic testing was prompted by the complex presentation in the absence of classical provoking factors—such as trauma, surgery, malignancy, or antiphospholipid syndrome—consistent with recommendations for selective thrombophilia testing in atypical or severe cases. The analysis revealed four thrombophilia-associated polymorphisms: heterozygous Factor V Leiden (FVL; R506Q genotype), Plasminogen Activator Inhibitor-1 (PAI-1; 4G/5G genotype), Methylenetetrahydrofolate reductase (MTHFR; c.677C > T genotype), and homozygous Angiotensin-Converting Enzyme Insertion/Deletion (ACE I/D; DD genotype). Conclusions: While each variant has been individually associated with thrombotic risk, their co-occurrence in a single patient with simultaneous arterial and venous thromboses has not, to our knowledge, been previously documented. This case underscores the potential for gene–gene interactions to amplify thrombotic risk, even in the presence of variants traditionally considered to confer only modest to moderate risk. It highlights the need for a multidisciplinary approach and raises questions regarding pharmacogenetics, anticoagulation, and future research into cumulative genetic risk in complex thrombotic phenotypes. Full article

Introduction: The introduction of radial-firing laser fibers has minimized catheter-to-vein distance during endovenous laser ablation (EVLA) for the great saphenous vein (GSV) and anterior saphenous veins (ASVs). This study investigates flush endovenous laser ablation (fEVLA) effectiveness in addressing chronic superficial venous insufficiency (CVI). Materials and methods: This single-center retrospective study analyzed consecutive fEVLA cases with duplex ultrasound follow-up at 1, 4, and 12 weeks. The primary endpoint was clinically significant endovenous heat-induced thrombosis (classes III–IV). Results: Three hundred and twelve patients were recruited (405 affected limbs, 369 GSV, and 36 ASV). CEAP classifications were stratified as follows: C2 in 6.1%, C3 in 34.2%, C4 in 44%, C5 in 2.7%, and C6 in 12.7% of cases. Perforator ligation, phlebectomy, or foam sclerotherapy were carried out in conjunction with EVLA. fEVLA was feasible in all cases. The success rate was 96.78%, defined as EHIT classes 1 and 2, and assessed by ultrasound one week postoperatively. Clinically significant EHIT (2.5% class 3 and 0.2% class 4) was managed with 15 mg rivaroxaban twice daily for 21 days. Follow-up at 4 weeks revealed complete resolution of all EHIT 3–4 cases. No cases of pulmonary embolism or deep vein thrombosis were observed during the study or follow-up period. Conclusions: fEVLA is a safe treatment for superficial CVI across various CEAP classes, and with prompt detection, the minimal complication rate can be completely resolved. Full article

Citrus diseases severely impact fruit yield and quality. To facilitate in-field, non-destructive, and rapid detection of citrus leaf physiological and pathological conditions, this study proposes a classification method for citrus leaf physiological and pathological statuses that integrates visible/near-infrared multispectral technology with deep learning. First, a handheld spectrometer was employed to acquire spectral images of five sample categories—Healthy, Huanglongbing, Yellow Vein Disease, Magnesium Deficiency and Manganese Deficiency. Mean spectral data were extracted from regions of interest within the 350–2500 nm wavelength range, and various preprocessing techniques were evaluated. The Standard Normal Variate (SNV) transformation, which demonstrated optimal performance, was selected for data preprocessing. Next, we innovatively introduced an adaptive spectral positional encoding mechanism into the Transformer framework. A lightweight, learnable network dynamically optimizes positional biases, yielding the ASTransformer (Adaptive Spectral Transformer) model, which more effectively captures complex dependencies among spectral features and identifies critical wavelength bands, thereby significantly enhancing the model’s adaptive representation of discriminative bands. Finally, the preprocessed spectra were fed into three deep learning architectures (1D-CNN, 1D-ResNet, and ASTransformer) for comparative evaluation. The results indicate that ASTransformer achieves the best classification performance: an overall accuracy of 97.7%, underscoring its excellent global classification capability; a Macro Average of 97.5%, reflecting balanced performance across categories; a Weighted Average of 97.8%, indicating superior performance in classes with larger sample sizes; an average precision of 97.5%, demonstrating high predictive accuracy; an average recall of 97.7%, showing effective detection of most affected samples; and an average F1-score of 97.6%, confirming a well-balanced trade-off between precision and recall. Furthermore, interpretability analysis via Integrated Gradients quantitatively assesses the contribution of each wavelength to the classification decisions. These findings validate the feasibility of combining a handheld spectrometer with the ASTransformer model for effective citrus leaf physiological and pathological detection, enabling efficient classification and feature visualization, and offer a valuable reference for disease detection of physiological and pathological conditions in other fruit crops. Full article

As shallow mineral resources become increasingly depleted, the search for deep-seated orebodies has emerged as a crucial focus in modern gold exploration. This study investigates Vein 171 in the Linglong gold field, Jiaodong Peninsula, using 3D attribute modeling for deep mineralization prediction and precise orebody delineation. The research integrates surface and block models through Vulcan 2021.5 3D mining software to reconstruct the spatial morphology and internal attribute distribution of the orebody. Geostatistical methods were applied to identify and process high-grade anomalies, with grade interpolation conducted using the inverse distance weighting (IDW) method. The results reveal that Vein 171 is predominantly controlled by NE-trending extensional structures, and grade enrichment occurs in zones where fault dips transition from steep to gentle. The grade distribution of the 171₁ and 171_sub-1 orebodies demonstrates heterogeneity, with high-grade clusters exhibiting periodic and discrete distributions along the dip and plunge directions. Key enrichment zones were identified at elevations of –1800 m to –800 m near the bifurcation of the Zhaoping Fault, where stress concentration and rock fracturing have created complex fracture networks conducive to hydrothermal fluid migration and gold precipitation. Nine verification drillholes in key target areas revealed 21 new mineralized bodies, resulting in an estimated additional 2.308 t of gold resources and validating the predictive accuracy of the 3D model. This study not only provides a reliable framework for deep prospecting and mineral resource expansion in the Linglong Goldfield but also serves as a reference for exploration in similar structurally controlled gold deposits globally. Full article

Background: Perihilar cholangiocarcinoma (pCCA), especially the periductal-infiltrating subtype, is notoriously difficult to diagnose due to subtle imaging findings and the absence of a mass. Case Presentation: We describe a 56-year-old man with morbid obesity and deep vein thrombosis (DVT), admitted for severe cholestatic jaundice. Initial ultrasound and two ERCPs were inconclusive, with only mild hilar duct dilation on CT. MRI was not possible due to the severe weight of the patient. Only at the 3rd ERCP with digital cholangioscopy were irregular mucosa and tumor infiltration observed, and a biopsy confirmed moderately to poorly differentiated adenocarcinoma. The patient deteriorated rapidly after discharge, returning in septic shock. Despite laparoscopy excluding cholecystitis and cirrhosis, he died from multiorgan failure. Autopsy revealed diffuse hilar tumor infiltration, nodal metastases, and fatal pulmonary tumor embolism (Bismuth IV). Conclusions: This case highlights the necessity of early escalation to cholangioscopy in unresolved cholestasis, the importance of recognizing paraneoplastic thrombosis, and the value of autopsy in clarifying cause of death. Full article