Objectives:PORTAS-3 was designed to compare the frequency of pneumothorax or haemothorax in a primary open versus closed strategy for port implantation.Background Data:The implantation strategy for totally implantable venous access ports with the optimal benefit/risk ratio remains unclear.Methods:PORTAS-3 was a multicentre, randomized, controlled, parallel-group superiority trial. Adult patients with oncological disease scheduled for elective port implantation were randomized to a primary open or closed strategy. Primary endpoint was the rate of pneumothorax or haemothorax. Assuming a difference of 2.5% between the 2 groups, a sample size of 1154 patients was needed to prove superiority of the open group. A logistic regression model after the intention-to-treat principle was applied for analysis of the primary endpoint.Results:Between November 9, 2014 and September 5, 2016, 1205 patients were randomized. Of these, 1159 (open n = 583; closed n = 576) were finally analyzed. The rate of pneumothorax or haemothorax was significantly reduced with the open strategy [odds ratio 0.27, 95% confidence interval (CI) 0.09-0.88; P = 0.029]. Operation time was shorter for the closed strategy. Primary success rates, tolerability, morbidity, dose rate of radiation, and 30-day mortality did not differ significantly between the groups.Conclusion:A primary open strategy by cut-down of the cephalic vein, if necessary enhanced by a modified Seldinger technique, reduces the frequency of pneumothorax or haemothorax after central venous port implantation significantly compared with a closed strategy by primary puncture of the subclavian vein without routine sonographic guidance. Therefore, open surgical cut-down should be the reference standard for port implantation in comparable cohorts.Trial Registration:German Clinical Trials Register DRKS 00004900.