Efficacy of drug-eluting balloons (DEB) for treatment of de novo coronary lesions remains controversial. The present systematic review and meta-analysis of randomised controlled trials assessed DEB with bare-metal stents (BMS) and also DEB with provisional bail-out stents ('DEB-only' strategy), to other conventional options: plain-old balloon angioplasty (POBA), BMS and drug-eluting stents (DES). A systematic literature search from January 2000 until May 2017 was conducted. Primary outcome measure, late lumen loss (LLL); and secondary outcomes; binary restenosis, major adverse cardiac events (MACE), target lesion revascularization (TLR), myocardial infarction (MI), cardiovascular death and stent thrombosis were analysed. Seventeen RCTs were included with 2,616 patients. Several comparative groups showed significant differences. DEB with BMS were inferior to DES for LLL [mean difference (MD) =0.12 mm; 95% confidence interval (CI), 0.03 to 0.22; P=0.01]; and binary restenosis [risk ratio (RR) =1.89; (CI, 1.13 to 3.18); P=0.02]. DEB with BMS was superior to BMS for LLL [MD =-0.27 mm; (-0.45 to -0.10); P=0.002]; and MACE [RR =0.64; (0.46 to 0.90); P=0.010]. Finally, DEB alone was superior to POBA for LLL [MD =-0.39 mm; (-0.67 to -0.11); P=0.006] and binary restenosis [RR =0.20; (0.05 to 0.85); P=0.03] in bifurcation lesions. The results of this meta-analysis showed that whilst DEB with BMS is superior to BMS alone, the combination is inferior to DES for treatment of de novo coronary lesions. Thus, DEB + BMS should not be applied in de novo lesions unless in patients who have absolute contraindications to DES. DEB alone, however, should be considered for relative contraindications to DES such as small vessel disease and bifurcation lesions.