Drug-coated balloons vs. drug-eluting stents for coronary artery disease: an updated systematic review and meta-analysis of randomized controlled trials with lesion-specific insights

Objective To compare the efficacy and safety of drug-coated balloons (DCB) vs. drug-eluting stents (DES) in patients with coronary artery disease (CAD) undergoing percutaneous coronary intervention, with a focus on lesion-specific and presentation-specific outcomes. Methods We systematically searched PubMed, Embase, CENTRAL, and Web of Science for randomized controlled trials (RCTs) comparing DCB with DES from inception to March 2026. Outcomes included major adverse cardiac events (MACEs), target lesion revascularization (TLR), device-oriented composite endpoint (DoCE), patient-oriented composite endpoint (PoCE), cardiac death, myocardial infarction (MI), all-cause death, thrombosis (definite/probable), and angiographic endpoints. Pooled odds ratios (ORs) or mean differences (MDs) were calculated using random-effects models. Subgroup analyses were performed by lesion type de novo vs. in-stent restenosis (ISR), vessel size, clinical presentation (STEMI vs. NSTEMI vs. unstable angina), and DCB type (paclitaxel vs. sirolimus). Results Twenty-three RCTs comprising 8,123 patients were included. DCB was associated with significantly higher risks of TLR (OR 2.22, 95% CI 1.49–3.33), DoCE (OR 1.86, 95% CI 1.49–2.31), PoCE (OR 1.43, 95% CI 1.20–1.72), and cardiac death (OR 1.53, 95% CI 1.11–2.10) compared with DES. No significant differences were observed for MACEs, MI, all-cause death, or thrombosis. In the critical subgroup analysis, for ISR, DES was superior to DCB (OR for TLR with DCB vs. DES: 3.54, 95% CI 2.05–6.09), whereas for de novo lesions, DCB was associated with a higher risk of TLR compared to DES (OR 1.76, 95% CI 1.03–3.02). In small vessel disease, TLR did not differ significantly between the two strategies (OR 1.17, 95% CI 0.64–2.14). The increased risk of cardiac death with DCB was observed only in trials using paclitaxel-coated balloons (OR 1.63, 95% CI 1.17–2.28), while no signal was seen with sirolimus-coated balloons (OR 0.96, 95% CI 0.19–4.81). However, this finding is exploratory, derived from post-hoc subgroup analyses with limited events and shorter follow-up in sirolimus studies, and should be interpreted with caution. Exploratory analysis by clinical presentation showed no significant interaction between treatment effect and STEMI, NSTEMI, or unstable angina (P-interactio n = 0.34 for MACEs). Conclusions The comparative effectiveness of DCB vs. DES is lesion-specific. For ISR, DES remains the superior treatment. DCB represents a viable alternative to DES in de novo small vessel disease. However, in de novo lesions of non-small vessels, DES remains superior. The increased cardiac death signal appears to be driven by paclitaxel-coated balloons and warrants further investigation. Clinical presentation (STEMI/NSTEMI/unstable angina) did not modify the relative treatment effect, but these analyses were exploratory and limited by sample size. Systematic Review Registration https://www.crd.york.ac.uk/PROSPERO/view/CRD420261355942 , PROSPERO CRD420261355942.