Purpose: Suture-mediated hemostasis device takes a long time to train and have limitations in finding appropriate compression point depending on the tactile sense of the hands. If the appropriate compression point is determined using ultrasound and the instrument is used under ultrasound guidance, it is expected to reduce the technical failure and additional manual compression. Materials and methods: This retrospective study included 104 patients in whom the ProGlide vascular closure device (Abbott Vascular, Redwood City, CA, USA) was deployed to close common femoral artery access between January 2022 and June 2022. Ultrasound-guided hemostasis was performed in 54 patients and 50 conventional hemostasis was performed. We analyzed the medical records (procedure, patient body mass index, coagulation function, visual imaging, ultrasound imaging, time to achieve hemostasis) and post-treatment medical records (progress records, nursing records) to investigate the technical success rate and complications (hematoma, pseudoaneurysm formation, SFA occlusion, access site infection) of the procedure. Results: Technical success rate was 52/54 (Ultrasound guided group) and 47/50 (conventional group), respectively ( p > 0.05). Mean hemostasis time was 271 s (US guided group) and 317 s (conventional group), respectively ( p > 0.05). Additional manual compression was applied in 7/54 (US guided group) and 15/50 (conventional group) ( p < 0.05). In both groups, complications such as hematoma formation, access site infection and SFA occlusion did not occur. Conclusion: Using ultrasound guidance when using the suture mediated closing device reduces the frequency of additional manual compression without increased risk of complication.

Background: This study aimed to assess the safety and feasibility of a 3-month flushing interval for totally implantable venous access ports (TIVAPs), particularly regarding intraluminal clot formation. Methods: Between May 2017 and September 2018, we established a single-center cohort of 151 patients who were referred for TIVAP removal and categorized them into three flushing-interval groups: A (⩽1 month), B (1–2 months), and C (2–3 months). Odds ratios (OR) with 95% confidence intervals (CI) were computed using logistic regression to determine the risk of intraluminal clot occurrence. To compare clot lengths among the groups, we employed linear regression. Restricted cubic splines were used to model the flushing interval as a continuous variable. Results: Compared to Group A, Groups B and C had adjusted ORs of 0.82 (0.35–1.92) and 0.78 (0.29–2.11) for intraluminal clot risk. Occlusion was rare (1/151; 0.7%) and successfully treated with mechanical recanalization. Adjusted differences in clot length compared to Group A were 0.01 (−0.85 to 0.87) for Group B and −0.23 (−1.21 to 0.76) for Group C. Spline regression analysis showed no significant association between the flushing interval and clot length ( p for trend = 0.84). Conclusion: The incidence and burden of intraluminal clots did not increase significantly within the first 3 months. Occlusion was rare, even in the presence of clots, and was successfully treated. Therefore, a 3-month flushing interval appears to be a safe and practical option, aligning with the common recommendation for surveillance intervals in patients with cancer.

Background/Objectives: Early detection of hepatocellular carcinoma (HCC), particularly small lesions (<2 cm), which is crucial for curative treatment, remains challenging with conventional liver dynamic computed tomography (LDCT). We aimed to develop a deep learning algorithm to generate synthetic CT during hepatic arteriography (CTHA) from non-invasive LDCT and evaluate its lesion detection performance. Methods: A cycle-consistent generative adversarial network with an attention module [Unsupervised Generative Attentional Networks with Adaptive Layer-Instance Normalization (U-GAT-IT)] was trained using paired LDCT and CTHA images from 277 patients. The model was validated using internal (68 patients, 139 lesions) and external sets from two independent centers (87 patients, 117 lesions). Two radiologists assessed detection performance using a 5-point scale and the detection rate. Results: Synthetic CTHA significantly improved the detection of sub-centimeter (<1 cm) HCCs compared with LDCT in the internal set (69.6% vs. 47.8%, p < 0.05). This improvement was robust in the external set; synthetic CTHA detected a greater number of small lesions than LDCT. Quantitative metrics (structural similarity index measure and peak signal-to-noise ratio) indicated high structural fidelity. Conclusions: Deep-learning–based synthetic CTHA significantly enhanced the detection of small HCCs compared with standard LDCT, offering a non-invasive alternative with high detection sensitivity, which was validated across multicentric data.

This study aimed to validate a landmark-based catheter length prediction formula for bedside peripherally inserted central catheter (PICC) placement in a real-world clinical setting, especially in complex environments where advanced navigation technologies may be limited or unsuitable. We retrospectively analyzed 449 consecutive bedside PICC placements performed between March 2022 and September 2025. Catheter insertion length was estimated beforehand using a formula based on anteroposterior chest radiograph (CXR) measurements, including clavicle length and vertebral body units. Final tip positions were assessed by two independent radiologists using post-procedural CXR in relation to the cavoatrial junction (CAJ), with optimal positioning defined as the tip being within ±2.8 cm of the CAJ. Of the 449 cases, technical success was achieved in 436 (97.1%), and optimal tip positioning was seen in 74.3% of successful cases. The interobserver agreement for tip assessment was 98.45% (95% confidence interval: 96.84–99.25%). The average distance from the catheter tip to the CAJ was −0.10 ± 2.59 cm, with a median of 0.0 cm. Rates of optimal positioning did not significantly differ based on operator experience (p = 0.496) or approach side (p = 0.590). Importantly, the technical success rate remained high (97.5%) even in patients with other intravascular devices, such as extracorporeal membrane oxygenation (ECMO) cannulas or hemodialysis catheters. In conclusion, the landmark-based formula offers high predictive accuracy and excellent reproducibility for bedside PICC placement. It serves as a reliable initial safety measure and a practical alternative in complex intensive care settings where real-time navigation is limited or systemically constrained.