CAD); however, permanent metallic implantation is associated with late adverse events. Drug-coated balloon (DCB) angioplasty has emerged as a “leave-nothing-behind” alternative, but its comparative efficacy and safety in de novo CAD remain uncertain. Objectives: To systematically compare clinical outcomes of DCB angioplasty versus DES implantation in patients with de novo coronary artery disease.Methods: A systematic review and meta-analysis of randomized controlled trials was performed according to PRISMA guidelines. PubMed/MEDLINE, EMBASE, Scopus, and Web of Science were searched from January 2006 to November 2024. Trials comparing DCB angioplasty with DES implantation in de novo coronary lesions with at least 12 months of follow-up were included. The primary outcome was major adverse cardiac events (MACE). Secondary outcomes included cardiac death, all-cause mortality, myocardial infarction, and target lesion revascularization (TLR). Random-effects models were used to pool hazard or incidence rate ratios. Risk of bias was assessed using the Cochrane RoB 2.0 tool. Results: Nine randomized trials comprising 4,284 patients were included. DCB angioplasty demonstrated no significant difference compared with DES for MACE (pooled effect size 0.96, 95% CI 0.72–1.20). Risks of myocardial infarction, TLR, and all-cause mortality were also comparable between strategies. However, DCB angioplasty was associated with a significantly higher risk of cardiac death (pooled effect size 1.36, 95% CI 1.07–1.66), with low heterogeneity across studies. Moderate-to-substantial heterogeneity was observed for MACE and TLR outcomes. Conclusions: In patients with de novo coronary artery disease, DCB angioplasty provides similar efficacy to DES implantation for most clinical outcomes, including MACE and all-cause mortality. The observed increase in cardiac death with DCBs warrants cautious interpretation and highlights the need for further large-scale randomized trials with long-term follow-up to better define patient selection and safety.
Percutaneous coronary intervention (PCI) remains a cornerstone in the management of patients with obstructive coronary artery disease (CAD). Over the past two decades, the introduction and refinement of drug-eluting stents (DES) have substantially improved clinical outcomes compared with bare-metal stents, primarily by reducing neointimal hyperplasia and the need for repeat revascularization [1,2]. Contemporary DES platforms, incorporating thinner struts and advanced antiproliferative agents, have further lowered rates of in-stent restenosis and stent thrombosis, thereby establishing DES-based PCI as the standard of care for de novo coronary lesions across a broad spectrum of clinical presentations [3,4].
Despite these advances, permanent metallic stent implantation is not without limitations. The presence of a lifelong intravascular scaffold may predispose patients to delayed endothelial healing, chronic vascular inflammation, and late adverse events, including very late stent thrombosis and neoatherosclerosis [5–7]. In addition, stent implantation can compromise future treatment options, particularly in small vessels, bifurcation lesions, or patients who may require subsequent surgical revascularization. These concerns have driven interest in alternative “leave-nothing-behind” revascularization strategies.
Drug-coated balloons (DCBs) have emerged as a promising non-stent-based approach to coronary revascularization. DCBs deliver an antiproliferative drug—most commonly paclitaxel—directly to the vessel wall during transient balloon inflation, without leaving a permanent implant behind [8,9]. This strategy allows for homogeneous drug transfer to the intima while preserving native vessel anatomy and vasomotion. Initially, DCB technology demonstrated clear efficacy in the treatment of coronary in-stent restenosis, where multiple randomized trials showed superiority or non-inferiority compared with repeat stenting [10–12]. On the basis of this evidence, DCB angioplasty is now an established treatment option for in-stent restenosis in international guidelines.
Encouraged by these results, investigators have explored the role of DCB angioplasty in de novo coronary artery disease. The theoretical advantages of DCBs in this setting include avoidance of permanent metal implantation, reduced risk of late device-related complications, and the potential for shorter durations of dual antiplatelet therapy—an especially relevant consideration in patients at high bleeding risk [13,14]. These potential benefits are particularly attractive in small-vessel disease, where DES implantation is associated with higher restenosis rates and suboptimal long-term outcomes [15].
Over the past decade, several randomized controlled trials have compared DCB angioplasty with DES implantation for de novo coronary lesions. Many of these trials were designed as non-inferiority studies and focused on specific patient subsets, such as small-vessel disease or acute coronary syndromes [16–20]. While individual trials generally reported comparable angiographic and clinical outcomes between the two strategies, most were limited by modest sample sizes and insufficient statistical power to detect differences in hard clinical endpoints, such as mortality or myocardial infarction. Furthermore, heterogeneity in trial design, lesion characteristics, clinical presentation, and endpoint definitions has complicated the interpretation of individual study findings.
Recent years have seen the publication of larger, multicenter randomized trials evaluating DCBs in broader de novo CAD populations, providing more robust clinical data [21]. Nevertheless, uncertainty persists regarding the comparative safety and efficacy of DCB angioplasty versus DES-based PCI, particularly with respect to major adverse cardiac events, mortality, and repeat revascularization. Importantly, concerns have also been raised about potential signals of increased cardiac mortality associated with paclitaxel-coated devices in peripheral vascular interventions, underscoring the need for careful evaluation of long-term outcomes in the coronary setting [22].
Given the expanding evidence base and ongoing debate, a comprehensive synthesis of randomized data is warranted. Systematic review and meta-analysis offer an opportunity to integrate results across trials, improve statistical power, and provide more precise estimates of treatment effects for clinically relevant outcomes. In particular, pooling trial-level incidence rate or hazard data allows for standardized comparison across studies with varying follow-up durations.
Therefore, the objective of the present systematic review and meta-analysis was to compare the efficacy and safety of drug-coated balloon angioplasty versus drug-eluting stent implantation for the treatment of de novo coronary artery disease. We focused on major adverse cardiac events as the primary outcome, with secondary analyses of cardiac death, all-cause mortality, myocardial infarction, and target lesion revascularization. By synthesizing data from randomized controlled trials with at least one year of follow-up, this study aims to provide contemporary, clinically meaningful evidence to inform decision-making regarding the optimal revascularization strategy for patients with de novo coronary lesions.
Literature Search A comprehensive and systematic literature search was conducted to identify randomized controlled trials comparing drug-coated balloon angioplasty with drug-eluting stent implantation in patients with de novo coronary artery disease. The databases PubMed/MEDLINE, EMBASE, Scopus, and Web of Science were searched from January 2006 to November 2024. Search terms included combinations of “drug-coated balloon,” “drug-eluting stent,” “de novo coronary lesions,” “percutaneous coronary intervention,” and “randomized controlled trial.” Reference lists of relevant reviews and eligible studies were manually screened to identify additional trials. Only peer-reviewed studies with a minimum follow-up of 12 months were considered. Registered with Prospero with number CRD420251276334. Study Selection and Data Extraction Study selection was performed in accordance with PRISMA guidelines [23]. After removal of duplicate records, titles and abstracts were independently screened by two reviewers to identify potentially eligible studies. Full-text articles were subsequently assessed for inclusion based on predefined criteria: randomized controlled design, comparison of drug-coated balloons versus drug-eluting stents, treatment of de novo coronary artery disease, and a minimum follow-up of 12 months. Discrepancies were resolved by consensus or consultation with a third reviewer. Data extraction was conducted independently by two investigators using a standardized form, capturing study design, patient characteristics, procedural details, follow-up duration, and clinical outcomes. Statistical Analysis and Risk of Bias Assessment Statistical analyses were performed using a random-effects meta-analysis framework to account for anticipated clinical and methodological heterogeneity across trials. Effect estimates were expressed as hazard ratios or incidence rate ratios with corresponding 95% confidence intervals. Standard errors were derived from reported confidence intervals when not directly provided. Heterogeneity was quantified using the Cochran Q test and the I² statistic, with values >50% indicating substantial heterogeneity. All analyses were conducted using validated statistical software. Risk of bias for individual studies was independently assessed by two reviewers using the Cochrane Risk of Bias tool, version 2.0 (RoB 2.0), evaluating bias arising from randomization, deviations from intended interventions, missing outcome data, outcome measurement, and selective reporting [24].
Study Population and Baseline Demographics
A total of nine randomized controlled trials were included in the final analysis, encompassing 4,284 patients [24 - 32] with de novo coronary artery disease who were randomly assigned to treatment with drug-coated balloon angioplasty or drug-eluting stent implantation. Across the included studies, sample sizes ranged from small single-center trials to large multicenter investigations, with one contemporary trial contributing more than half of the overall population. The majority of trials employed a non-inferiority design and reported clinical follow-up of at least 12 months. The pooled study population was predominantly male, with men accounting for approximately two-thirds to three-quarters of enrolled participants across trials. The mean age of patients was generally in the early to mid-60s, reflecting a typical population undergoing percutaneous coronary intervention. Cardiovascular risk factors were common, with diabetes mellitus present in roughly one-quarter to one-third of patients and a substantial proportion having a prior history of myocardial infarction. Acute coronary syndrome at presentation was variably represented, ranging from trials exclusively enrolling ACS patients to those including stable and unstable clinical presentations. Most studies focused on small-vessel coronary disease, with mean reference vessel diameters typically below 3.0 mm, although selected trials enrolled all-comer populations without vessel size restrictions. Baseline characteristics were broadly balanced between treatment arms within individual trials, supporting the validity of subsequent comparative outcome analyses.
Table 1. Basic Demographics
|
Study |
Sample Size (DCB / DES) |
Participating Centers |
Geographic Location |
Target Vessel Diameter |
Recruitment Timeline |
Study Design |
Primary Outcome Definition |
Longest Follow-up (months) |
Registration ID |
|
BELLO |
90 / 92 |
15 |
Italy |
<2.8 mm |
2010–2012 |
Non-inferiority RCT |
Composite of mortality, myocardial infarction, or target vessel revascularization |
36 |
NCT01086579 |
|
BASKET-SMALL 2 |
382 / 376 |
14 |
Central Europe |
2.0–<3.0 mm |
2012–2017 |
Non-inferiority RCT |
Cardiac death, non-fatal myocardial infarction, or target vessel revascularization |
36 |
NCT01574534 |
|
RESTORE SVD |
116 / 114 |
12 |
China |
2.25–2.75 mm |
2016–2017 |
Non-inferiority RCT |
Cardiac death, target-vessel myocardial infarction, or ischemia-driven TLR |
24 |
NCT02946307 |
|
REVELATION |
60 / 60 |
1 |
Netherlands |
No restriction |
2014–2017 |
Non-inferiority RCT |
Composite of death, myocardial infarction, TLR, stent thrombosis, or major bleeding |
60 |
NCT02219802 |
|
PICCOLETO II |
114 / 118 |
5 |
Southern Europe |
2.25–2.75 mm |
2015–2018 |
Non-inferiority RCT |
Cardiac death, myocardial infarction, or TLR |
36 |
NCT03899818 |
|
Hao et al. |
38 / 42 |
1 |
China |
2.5–4.0 mm |
2018–2019 |
Non-inferiority RCT |
Cardiac death, target-vessel myocardial infarction, or TLR |
12 |
Not reported |
|
Yu et al. |
84 / 79 |
1 |
China |
2.25–4.0 mm |
2017–2018 |
Non-inferiority RCT |
Angiographic late lumen loss (primary); MACE as secondary endpoint |
12 |
Not reported |
|
Dissolve SVD |
129 / 118 |
10 |
China |
2.25–2.75 mm |
2018–2019 |
Non-inferiority RCT |
Cardiac death, target-vessel myocardial infarction, or ischemia-driven TLR |
12 |
NCT03376646 |
|
REC-CAGEFREE-I |
1133 / 1139 |
43 |
China |
Any diameter |
2021–2022 |
Non-inferiority RCT |
Cardiac death, target-vessel myocardial infarction, or clinically/physiologically indicated TLR |
24 |
NCT04561739 |
Table 2. Baseline Patient Profiles and Device Characteristics Across Included Randomized Trials
|
Trial |
Treatment Arm (N) |
Drug-Coated Balloon Used |
Comparator Stent |
Mean Age (years) |
Male Sex (%) |
Diabetes Mellitus (%) |
Prior MI (%) |
ACS at Presentation (%) |
Reference Vessel Diameter (mm) |
Bail-out Stenting (%) |
|
BELLO |
DCB (90) |
IN.PACT Falcon |
Taxus Liberté (paclitaxel) |
64.8 ± 8.5 |
80.0 |
43.3 |
51.1 |
24.4 |
2.15 ± 0.27 |
20.0 |
|
DES (92) |
— |
Taxus Liberté |
66.4 ± 9.0 |
77.2 |
38.0 |
35.9 |
21.7 |
2.26 ± 0.24 |
— |
|
|
BASKET-SMALL 2 |
DCB (382) |
SeQuent Please |
Taxus Element → Xience |
67.2 ± 10.3 |
77.0 |
32.0 |
42.0 |
30.0 |
Not reported |
5.1 |
|
DES (376) |
— |
Xience |
68.4 ± 10.3 |
70.0 |
35.0 |
35.0 |
27.0 |
Not reported |
— |
|
|
RESTORE SVD |
DCB (116) |
Restore |
Resolute Integrity |
60.1 ± 10.5 |
77.0 |
46.0 |
26.0 |
80.0 |
2.42 ± 0.15 |
5.2 |
|
DES (114) |
— |
Resolute Integrity |
60.5 ± 10.8 |
88.0 |
48.0 |
28.0 |
81.0 |
2.42 ± 0.18 |
— |
|
|
REVELATION |
DCB (60) |
Pantera Lux |
Orsiro / Xience |
57.4 ± 9.2 |
87.0 |
13.0 |
NR |
100 |
NR |
18.0 |
|
DES (60) |
— |
Orsiro / Xience |
57.3 ± 8.3 |
87.0 |
7.0 |
NR |
100 |
NR |
— |
|
|
PICCOLETO II |
DCB (118) |
Elutax SV / Emperor |
Xience |
64 (IQR 48–80) |
70.3 |
38.0 |
38.0 |
31.4 |
2.23 ± 0.40 |
6.8 |
|
DES (114) |
— |
Xience |
66 (IQR 50–82) |
76.9 |
35.4 |
30.0 |
21.1 |
2.18 ± 0.40 |
— |
|
|
Hao et al. |
DCB (38) |
Bingo (Yinyi Biotech) |
Not specified |
59.0 ± 11.0 |
78.0 |
28.0 |
NR |
100 |
2.5–4.0 |
9.5 |
|
DES (42) |
— |
Not specified |
56.0 ± 11.0 |
82.0 |
35.0 |
NR |
100 |
2.5–4.0 |
— |
|
|
Yu et al. |
DCB (84) |
SeQuent Please |
Resolute / Xience / Firehawk |
62.6 ± 8.8 |
73.8 |
19.0 |
NR |
91.5 |
2.77 (2.50–3.25) |
2.4 |
|
DES (79) |
— |
Resolute / Xience / Firehawk |
64.0 ± 10.5 |
70.9 |
29.1 |
NR |
87.3 |
3.01 (2.65–3.39) |
— |
|
|
Dissolve SVD |
DCB (129) |
Dissolve DCB |
Resolute Integrity |
60.2 ± 9.5 |
72.9 |
35.7 |
25.6 |
63.6 |
2.20 ± 0.26 |
3.9 |
|
DES (118) |
— |
Resolute Integrity |
60.1 ± 9.3 |
69.5 |
38.1 |
22.9 |
65.3 |
2.21 ± 0.24 |
— |
|
|
REC-CAGEFREE-I |
DCB (1133) |
Swide DCB |
Firehawk (sirolimus) |
61.5 ± 10.3 |
67.9 |
24.9 |
7.1 |
55.6 |
2.74 ± 0.51 |
9.4 |
|
DES (1139) |
— |
Firehawk |
61.2 ± 10.5 |
70.7 |
29.7 |
9.2 |
55.0 |
2.82 ± 0.51 |
— |
Major Adverse Cardiac Events
Nine randomized trials were included in the analysis of major adverse cardiac events (MACE). The pooled random-effects model demonstrated no significant difference between drug-coated balloon angioplasty and drug-eluting stent implantation (effect size 0.96, 95% CI 0.72–1.20). Moderate heterogeneity was observed across studies (I² = 61.3%), reflecting variability in trial design and patient populations. Individual study estimates showed mixed effects, with no consistent advantage for either strategy. Overall, the forest plot indicates comparable efficacy of drug-coated balloons and drug-eluting stents in reducing MACE among patients with de novo coronary artery disease.
Cardiac Death
Eight randomized trials contributed to the analysis of cardiac death. The pooled random-effects REML model demonstrated a significantly higher risk of cardiac death associated with drug-coated balloon angioplasty compared with drug-eluting stent implantation (pooled effect size 1.36, 95% CI 1.07–1.66). Notably, no between-study heterogeneity was observed (τ² = 0.00; I² = 0%), indicating consistent findings across trials. Larger multicenter studies contributed the majority of statistical weight. Although individual trials showed wide confidence intervals due to low event rates, the overall forest plot suggests a statistically significant increase in cardiac mortality with drug-coated balloons during medium-term follow-up.
Myocardial Infarction
Seven randomized trials were included in the analysis of myocardial infarction. The pooled random-effects REML model showed no statistically significant difference between drug-coated balloon angioplasty and drug-eluting stent implantation (pooled effect size 0.88, 95% CI 0.30–1.46). Moderate heterogeneity was observed across studies (I² = 48.4%), suggesting variability in effect estimates likely related to differences in clinical presentation and trial design. Individual studies demonstrated wide confidence intervals, reflecting low event rates. Overall, the forest plot indicates comparable risks of myocardial infarction between the two treatment strategies.
Target Lesion Revascularization
Eight randomized trials contributed to the analysis of target lesion revascularization (TLR). The pooled random-effects REML model showed no statistically significant difference between drug-coated balloon angioplasty and drug-eluting stent implantation (pooled effect size 1.46, 95% CI 0.69–2.23). However, substantial heterogeneity was observed across studies (I² = 91.6%), indicating marked variability in treatment effects. Several trials, particularly larger all-comer studies, demonstrated a higher rate of TLR with drug-coated balloons, while others showed comparable outcomes. Overall, the forest plot suggests similar TLR risk between strategies, but with considerable between-study heterogeneity warranting cautious interpretation.
All-Cause MortalityNine randomized trials were included in the analysis of all-cause death. The pooled random-effects REML model demonstrated no significant difference in all-cause mortality between drug-coated balloon angioplasty and drug-eluting stent implantation (pooled effect size 0.94, 95% CI 0.79–1.09). There was no observed heterogeneity across studies (I² = 0%), indicating highly consistent findings among trials. Individual study estimates were closely clustered around unity, with overlapping confidence intervals. Overall, the forest plot suggests that both revascularization strategies are associated with comparable all-cause mortality during medium-term follow-up in patients with de novo coronary artery disease.
This systematic review and meta-analysis demonstrate that drug-coated balloon angioplasty offers comparable efficacy to drug-eluting stent implantation for the treatment of de novo coronary artery disease, with similar rates of major adverse cardiac events, myocardial infarction, target lesion revascularization, and all-cause mortality. However, a higher risk of cardiac death was observed with drug-coated balloons, despite low heterogeneity across studies. These findings support the selective use of drug-coated balloons in appropriately chosen patients, particularly in small-vessel disease, while emphasizing the need for further large, well-powered randomized trials to clarify long-term safety and optimize patient selection. Conflict of Interest The authors certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript. Funding The authors report no involvement in the research by the sponsor that could have influenced the outcome of this work. Authors’ contributions. All authors contributed equally to the manuscript and read and approved the final version of the manuscript. Acknowledgement This paper is the collaborative work of all authors under the mentorship for the research work from BIR (Biomedical and International Research). We all authors acknowledge this mentorship for this meta-analysis.
[56] Siontis GCM, Räber L, Jueni P, Pilgrim T, Stefanini GG, Moschovitis A, et al. Long-term outcomes of drug-eluting stents versus bare-metal stents in large coronary arteries. Circulation. 2014;129(23):2383-92. doi:10.1161/CIRCULATIONAHA.113.006890.