Background Although thrombolytic therapy with rescue percutaneous coronary intervention (PCI) is a common treatment strategy for ST-segment elevation acute myocardial infarction (STEMI), scant data are available on its efficacy relative to primary PCI, and comparison was therefore the aim of this study. Methods This multicenter, open-label, randomized, parallel trial was conducted in 12 hospitals on patients (age 570 years) with STEMI who presented within 12 hours of symptom onset (mean interval >3 hours). Patients were randomized to three groups: primary PCI group (n=101); recombinant staphylokinase (r-Sak) group (n=104); and recombinant tissue-type plasminogen activator (rt-PA) group (n=106). For all patients allocated to the thrombolytic therapy arm, coronary angiography was performed at 90 minutes after drug therapy to confirm infarct-related artery (IRA) patency; rescue PCI was performed in cases with TIMI flow grade <= 2. Bare-metal stent implantation was planned for all patients. Results After randomization it required an average of 113.4 minutes to start thrombolytic therapy (door-to-needle time) and 141.2 minutes to perform first balloon inflation in the IRA (door to balloon time). Rates of IRA patency (TIMI flow grade 2 or 3) and TIMI flow grade 3 were significantly lower in the thrombolysis group at 90 minutes after drug therapy than in the primary PCI group at the end of the procedure (70.5% vs. 98.0%, P <0.0001, and 53.0% vs. 85.9%, P <0.0001, respectively). Rescue PCI with stenting was performed in 117 patients (55.7%) in the thrombolytic therapy arm. Rates of patency and TIMI flow grade 3 were still significantly lower in the rescue PCI than in the primary PCI group (88.9% vs. 97.9%, P=0.0222, and 68.4% vs. 85.0%, P=0.0190, respectively). At 30 days post-therapy, mortality rate was significantly higher in the thrombolysis combined with rescue PCI group than in primary PCI group (7.1% vs. 0, P=0.0034). Rates of death/MI and bleeding complications were significantly higher in the thrombolysis with rescue PCI group than in the primary PCI group (10.0% vs. 1.0%, P=0.0380, and 28.10% vs. 8.91%, P=0.0001, respectively). Conclusions Thrombolytic therapy with rescue PCI was associated with significantly lower rates of coronary patency and TIMI flow grade 3, but with significantly higher rates of mortality, death/MI and hemorrhagic complications at 30 days, as compared with primary PCI in this group of Chinese STEMI patients with late presentation and delayed treatments. Chin Med J 2010;123(11):1365-1372