Introduction: There is no consensus regarding preferred site for central vein cannulation. [1] The jugular vein is the most “popular site” for central vein cannulation and is the “preferred site” for temporary hemodialysis. [2] The subclavian vein is less “infection-prone” but more likely to cause pneumothorax. [3] In case of chemo-port implantation, anatomical landmark puncture (ALP) of subclavian vein or cephalic vein exploration (CVE) at delto-pectoral groove have been the more “convenient” and “preferred” methods to avoid unaesthetic two incisions and visible subcutaneously-tunneled catheter. Recent evidence suggests that real-time ultrasound guided puncture (UGP) of subclavian vein could lower pneumothorax rate from 1-6% to 0%. [4] The safety, successful cannulation rate and pitfalls of this new procedure are our concerns. Methods: From Oct.13rd, 2018, all patients scheduled for chemo-port implantation were enrolled for clinical trial of UGP for subclavian vein cannulation. The exclusion criteria included surgical site skin infection or tumor invasion, uncorrectable coagulopathy and known occluded central veins. The preoperative sonography reveals diameter of target vein (DTV), skin-to-anterior wall of target vein distance (SAVD) and diameter change along with inspiration. The cancer type, laterality of chemo-port, positioning and body mass index (BMI) of patients, attempts of puncture, pain score during puncture, conversion rate and procedure-related complications were recorded. Results: Total 77 patients, 88.3% female, aged 31-81(mean 54.6) underwent UGP of deep vein at subclavian region for chemo-port implantation. 59.7% of target veins were at left side and they included 68 subclavian veins, 5 axillary veins and 4 cephalic veins at clavipectoral fascia. 98.7% of operations were performed under local anesthesia. The diversity of cancer types included 47 breast cancer, 10 colon cancer etc. BMI ranged from 18.1 to 33.4 ( 24.2± 3.6). DTV and SAVD were 8.5±1.7 mm and 19.4±4.7 mm respectively. Severe subclavian vein collapse during inspiration was found in 25% patients. 54.4% patients were put on Trendelenburg position, 7.4% with shoulder cushioned and none with shoulder shrugging. The success rate of UGP was 90.9% with 6.5% converted to ALP and 2.6% to CVE. The mean attempts of UGP was 2.0±1.9. The pain score during UGP was 3.5±2.0. The total procedure time was 27.7±8.0 minutes. There is no statistically difference between failed group and success group in age, BMI, SAVD, but there is difference in DTV (p=0.0247). [Table 1] There was one pneumothorax (1.3%) found after operation and remitted spontaneously without tube drainage. The Kaplan-Meier curve of patient survival and function of chemo-port has been 100% without re-intervention due to infection or obstruction. Conclusion: The UGP for subclavian vein cannulation is a new technique for a surgeon familiar with ALP. It has opened the 3rd eye of the surgeon to see the advancement of the needle tip and the dramatic change of vein diameter along with the inspiration in our 25% patients. The success rate and complications are fair in our study. The DTV is the main factor for cannulation failure.Table 1comparison of failure group and success groupTable 1comparison of failure group and success group Disclosure: Nothing to disclose