Drug-eluting stents (DES) represent the standard of care in chronic total occlusion (CTO) percutaneous coronary intervention (PCI), as they release antiproliferative drugs that reduce restenosis compared with bare-metal stents.1,2 However, DES implantation is associated with potential drawbacks, including the need for prolonged dual antiplatelet therapy, incomplete apposition and expansion, and the development of neoatherosclerosis, which may cause adverse events such as in-stent restenosis and stent thrombosis.3,4 Drug-coated balloons (DCB) have emerged as an alternative PCI strategy, allowing the delivery of antiproliferative drugs without the need for permanent scaffolding.5

The use of DCBs has been explored in various coronary artery disease settings, including in-stent restenosis and de novo small vessel disease, showing favorable outcomes in terms of late luminal gain and restenosis reduction.6–9 The potential of DCBs has recently gained interest in CTO-PCI, since CTO lesions may benefit from avoiding vessel scaffolding. However, the use of DCBs in CTO-PCI is not without challenges, including the risk of recoil due to the high plaque burden of CTO lesions, flow-limiting dissections after preparation, and lack of standardized sizing criteria for DCBs. Moreover, despite the theoretical benefit of DCBs in CTO-PCI, current evidence is limited, with no large-scale randomized controlled trials directly comparing DCBs with DES in this setting.10–12 Therefore, the aim of this systematic review and meta-analysis was to synthesize the available data and provide a comprehensive comparison of DCBs and DES in CTO-PCI.

This systematic review and meta-analysis was performed according to the Cochrane Collaboration and PRISMA guidelines.13,14 The analysis was registered in PROSPERO, the international prospective register of systematic reviews (PROSPERO record ID=CRD42025642790).

In this meta-analysis, we provide a comprehensive synthesis of available data comparing DCB- and DES-based revascularization strategies in CTO-PCI. The main finding of our study is that, in lesions that achieve favorable angiographic results after predilation, the DCB and DES approaches were associated, with similar rates of TLR, MACE, cardiovascular mortality, TVR and TVMI during a median follow-up of 2.8 years. DCB have been extensively studied in ISR and de novo small vessel disease, showing superior late luminal gain and restenosis reduction compared with conventional balloon angioplasty.22,23 The rationale for DCB use in CTOs lies in the potential benefits of preserving vessel physiology and avoiding long-segment stent implantation, often required in CTOs and associated with an increased risk of stent restenosis and thrombosis.24,25 Moreover, recent evidence suggests that CTO-PCI, which requires extraplaque techniques, is associated with higher required stent length, enhancing the rationale for scaffold-sparing solutions.26 Our findings show that, with appropriate lesion preparation, TLR rates following DCB use were similar to those observed with DES (HR, 0.81; 95%CI, 0-57-1.13; P=.22), suggesting that a stentless strategy may be feasible without increasing restenosis risk. Subgroup analysis showed no significant differences in outcomes between de novo and in-stent CTOs. While DES have outperformed bare-metal stents and plain old balloon angioplasty in reducing TLR,27 our findings suggest that DCBs, when used under optimized conditions, may provide similar long-term effectiveness without the disadvantages of permanent metal scaffolding. Lesion preparation emerged as a pivotal determinant of success in DCB-based CTO-PCI. Our meta-regression analysis identified a significant association between the use of advanced plaque modification techniques—such as cutting balloons, intravascular lithotripsy, or rotational atherectomy—and reduced TLR rates (P=.02).

These findings agree with those of prior studies emphasizing the importance of optimizing vessel compliance and luminal gain before DCB application to prevent early recoil and restenosis. Lee et al.28 highlighted that adequate lesion preparation enhances drug penetration and uniformity of drug delivery, which may explain why well-prepared lesions show similar long-term patency rates to DES-treated lesions. Similarly, Mashayekhi et al.29, reported that CTOs with heavy calcification require meticulous preparation to achieve optimal results, irrespective of the revascularization strategy used. Our results support this concept, suggesting that lesion preparation is not merely a procedural step but a potential key determinant of long-term success in DCB-based CTO-PCI.30 However, given the small number of included studies and the exploratory nature of this meta-regression analysis, these findings should be considered as hypothesis-generating. Finally, it is important to emphasize that the selection of DCBs was often limited to cases with favorable angiographic outcomes after lesion preparation, such as the achievement of Thrombolysis in Myocardial Infarction (TIMI) 3 flow and the absence of flow-limiting dissections. This intrinsic selection bias may have contributed to the comparable outcomes observed and underscores that DCB use in CTO-PCI may be reserved for anatomically and procedurally favorable scenarios.

We observed a trend toward lower MACE and cardiovascular mortality in the DCB group (HR, 0.70 and 0.60, respectively), although this trend did not reach statistical significance. While avoidance of late stent-related complications, such as neoatherosclerosis and very late stent thrombosis, has been proposed as a theoretical advantage of DCBs, these events typically occur beyond 3 years. Given that the median follow-up in our analysis was 2.8 years, our data may not fully capture these long-term differences. Furthermore, no studies to date have specifically assessed the incidence of neoatherosclerosis following DCB-only treatment, highlighting the need for further long-term, imaging-based studies.31–33

While last-generation DES have significantly reduced the rates of stent thrombosis compared with first-generation devices, stent failure remains a long-term issue, particularly in complex lesions.34 Notably, despite a numerical trend in favor of DCBs for MACE rates, moderate heterogeneity was observed for this outcome, and the included studies were underpowered to detect differences in hard endpoints; therefore, these findings should be interpreted with caution and warrant further investigation in larger randomized controlled trials.

No significant differences were observed in TVR (HR, 0.89; P=.65) or TVMI (HR, 1.0; P=.99), suggesting that a DCB-based approach does not increase the risk of major ischemic complications compared with DES. The overall low rates of MI in both groups support the safety of CTO-PCI when adequate lesion preparation is achieved. Moreover, while TVR is recommended by the Chronic Total Occlusion-Academic Research Consortium definition (CTO-ARC) as a standardized endpoint for CTO outcomes, we selected TLR as the primary outcome to provide a focused evaluation of the site of the treated lesion.35

Our findings expand upon prior studies on the role of DCBs in CTO-PCI. Particularly, a recent meta-analysis by Natarajan et al.36 evaluated the role of DCBs in CTO, including 1695 patients from 10 observational studies. However, a substantial proportion of the included studies were single-arm cohorts. In contrast, our meta-analysis focuses on studies that directly compare DCB and DES, encompassing a larger population (n=2221) and enabling a more rigorous assessment of comparative outcomes. Moreover, we performed subgroup analyses according to CTO subtype (de novo vs in-stent) and revascularization strategy (DCB-only vs hybrid), providing a broader overview of current DCB applications in CTO-PCI. Among the included studies, that by Shin et al.21 contributed the largest patient cohort to our meta-analysis. Interestingly, it reported a low number of events. Several factors may explain this finding, including short lesion length, lower rates of diabetes, and local expertise.

Finally, we conducted an exploratory meta-regression analysis which showed a significant interaction between advanced lesion preparation with debulking techniques and reduced TLR rates. However, this finding should be interpreted with caution and as hypothesis-generating. First, the definition of intense lesion preparation encompassed heterogeneous techniques, including cutting balloons, rotational atherectomy, and intravascular lithotripsy, which differ in their mechanisms and clinical applications. Because of their heterogeneity, grouping these approaches together may limit the interpretability of the observed association. Second, the analysis was based on a limited number of studies. Despite these limitations, these findings agree with evidence indicating that optimal lesion preparation is a critical determinant of procedural success with DCB therapy,37,38 suggesting a topic that warrants further investigation in larger, prospective randomized trials.

Zhao et al.39 included a smaller patient cohort and did not provide a comparison of DCB with DES, whereas our analysis integrates a significantly larger population, thereby enhancing statistical power and generalizability. Moreover, their study reported a MACE rate of 13.0% in the DCB group, which appears similar to the MACE rates observed in our study, reinforcing the hypothesis that DCBs may represent a viable alternative to stent implantation in selected CTO scenarios. However, considering the anatomical and procedural complexity inherent to CTOs, a hybrid revascularization approach may represent the most balanced and pragmatic strategy. In fact, combining DCB with DES might be helpful to optimize long-term outcomes while minimizing the drawbacks of full-metal stenting. This technique involves selective stenting of segments with suboptimal angiographic results after lesion preparation while using DCBs in adequately prepared vessel sections. While we did not include hybrid strategies as a formally separate arm due to the lack of consistently reported data, we addressed this issue through stratified analyses. Specifically, we compared outcomes between studies using DCB-only strategies and those with bailout DES implantation, which approximates a hybrid approach. The results were largely consistent between these subgroups (figure S5). However, prior studies suggest that a hybrid approach may reduce stent length, lowering the risk of restenosis, neoatherosclerosis, and very late stent thrombosis.40 Additionally, hybrid revascularization may help preserve future treatment options in patients requiring repeat interventions. While our pooled estimates suggest comparable outcomes between DCB-based and DES strategies in CTO-PCI, since the available evidence is derived from a small number of observational studies with heterogeneous designs, our findings should be interpreted with caution and serve to highlight the urgent need for well-designed randomized trials to validate the role of DCB in this complex setting.

Ongoing studies, such as the CTO-DENOVO (NCT05977842) and Co-CTO (NL-OMON51074) will provide more information on the role of DCBs in CTO PCI. Notably, although both studies are designed to evaluate DCB-based strategies, their protocols mandate drug eluting stent implantation of the main CTO body, with DCB reserved for proximal or distal segments after adequate lesion preparation. This highlights that the hybrid approach combining DES and DCBs could be the potential optimal revascularization strategy in complex CTO anatomy. While promising, further research is needed to define optimal patient selection and procedural techniques for this strategy in CTO-PCI.

This meta-analysis provides a comprehensive comparison between DCBs and DES in CTO-PCI and found no clear differences between DCBs and DES in terms of TLR, MACE, cardiovascular mortality and ischemic endpoints, suggesting that DCB-based revascularization may be a reasonable alternative in appropriately prepared CTO lesions, particularly in patients at high risk of stent-related complications. However, given the anatomical complexity and heterogeneity of CTOs, a hybrid revascularization approach combining DES implantation and DCB use in adequately prepared segments may represent the most balanced and pragmatic strategy. While DES remain the standard of care, DCBs emerge as a promising tool that deserve further validation in large-scale randomized trials.

None.

We confirm that SAGER (Sex and Gender Equity in Research) guidelines were followed where applicable.

No artificial intelligence was used in the preparation of this article.

G. Panuccio, and Y.S. Abdelwahed contributed equally to this work. G. Panuccio, Y.S. Abdelwahed, S. De Rosa, D. Torella: conceptualization, data management, writing, validation. All the other coauthors: validation, writing.

Nothing to declare.