Balloon angioplasty offered a valid alternative to surgery for coronary revascularization,1 but the risk of acute coronary occlusion and high restenosis rates associated with recoil, negative remodeling and cell proliferation have fueled an ongoing search for new therapeutic techniques that carry less risk of restenosis.2

The development of drug-eluting stents (DES) considerably reduced the risk of restenosis by eliminating recoil and negative remodeling and by significantly suppressing neointimal hyperplasia.3 Nevertheless, concerns about the risk of stent thrombosis, dependence on prolonged dual antiplatelet therapy and the risk of restenosis in patients with complex lesions prompted the search for alternative devices with low restenosis risk but without the disadvantages associated with DES,4 which led to the development of drug-eluting balloons (DEB).2

Several randomized studies demonstrated the safety and efficacy of this new technology for the treatment of in-stent restenosis, which was the first indication approved for their use in coronary artery disease (CAD).5,6 Other studies subsequently widened indications to other areas of endovascular intervention, notably de novo small vessel disease and peripheral arterial disease.4

It is thus important to compare the good results obtained in clinical trials with those in real-world patients through prospective multicenter registries of all-comers populations. The objective of this study was to evaluate the clinical efficacy of percutaneous coronary intervention (PCI) with DEBs in a multicenter national registry.

This study clearly demonstrated that DEB angioplasty in real-world patients with complex lesions is effective, with high rates of immediate angiographic success and low rates of MACE, including TLR at one-year follow-up. The results were equally good for in-stent restenosis and de novo CAD.

In recent years, DEBs have proved to be a valid alternative for in-stent restenosis following the good results achieved in clinical trials.5,6 Despite the fact that theoretically DEBs have the same potential disadvantages of acute recoil and late negative remodeling as conventional balloons, late luminal loss during follow-up has been consistently low in studies on DEBs (approximately 0.2 mm).9 PACCOCATH ISR I5 was the first randomized trial to compare DEBs with conventional balloons for in-stent restenosis and showed that DEB angioplasty was associated with a low incidence of restenosis and in-segment late luminal loss of only 0.03±0.48 mm at six months, significantly less than with conventional balloon angioplasty (p=0.002). PEPCAD II compared the safety and efficacy of treatment of BMS restenosis using DEBs vs. angioplasty with Taxus™ paclitaxel-coated stents. Here again, the primary endpoint, late luminal loss at six months, was lower in the group treated with DEBs (0.17±0.42 vs. 0.38±0.61 mm, p=0.03). One limitation of these studies is that the primary endpoint was late luminal loss, which is only a surrogate endpoint of angioplasty outcomes. In the case of PEPCAD II, use of this measure to compare two different strategies (balloon vs. stent) puts stent angioplasty at a disadvantage since calculation of late luminal loss depends directly on acute gain, which is always greater with stenting, and the greater the acute luminal gain, the greater the late luminal loss. It is thus to be expected that balloon angioplasty is associated with less luminal loss during follow-up.10 Nevertheless, both of these studies showed favorable clinical outcomes, with a lower rate of MACE for DEB angioplasty,5,6 a difference that was maintained at two-year11 (46% conventional balloon vs. 11% DEB, p=0.001) and five-year follow-up12 (59.3% conventional balloon vs. 27.8% DEB, p=0.009), mainly due to reduced need for TLR (38.9% conventional balloon vs. 9.3% DEB at five years, p=0.004).

The present study analyzed MACE in a high-risk cohort with complex lesions. The combined incidence of events was 14% at 12 months, higher than in randomized trials (9% in PACCOCATH ISR I and II and in PEPCAD II).6,11 On the other hand, the rate of TLR was similar – 5% in our study compared to 4% in PACCOCATH ISR I and II and 6% in PEPCAD II.6,11 These results may reflect the fact that ours was an all-comers population that included patients with MI (30%), left ventricular dysfunction (21%) and chronic renal failure (9%), conditions that were generally criteria for exclusion in previous studies. The higher rates of death and MI during follow-up thus reflect the greater complexity of our patients. Table 7 summarizes clinical outcomes of DEB angioplasty in the main randomized trials and registries to date.

In our population, the results of DEB angioplasty were equally good in terms of efficacy for in-stent restenosis and de novo small-vessel disease. In the randomized PICCOLETO study,13 DEB angioplasty of small vessels was associated with higher rates of angiographic restenosis and MACE at nine months than Taxus™ stent angioplasty, which prompted early termination of the trial. The DEB used in that study was the first-generation DIOR™, which has different pharmacokinetics and bioavailability from the SeQuent® Please, with an inferior anti-restenotic effect.14 By contrast, the more recent randomized BELLO study15 on de novo lesions in small vessels (reference diameter <2.8 mm) showed DEB angioplasty to be superior to Taxus™ stenting in terms of the primary endpoint at six months (late luminal loss 0.08±0.38 mm vs. 0.29±0.44 mm, p=0.001), with low event rates of 7.8% MACE and 4.4% TLR in the DEB group. Percutaneous intervention in small vessels is associated with a high restenosis rate, and the use of DES to suppress neointimal hyperplasia is recommended whenever possible. The E-SIRIUS16 and C-SIRIUS17 trials showed that use of the Cypher™ stent for the treatment of long de novo coronary lesions in small vessels reduced restenosis at eight months compared to BMS. However, stents implanted in small arteries have a higher metal-to-artery ratio, which can increase the risk of restenosis.18 The fact that DEBs have no metal structure is thus a clear advantage. They also have the advantage of homogeneous release of the drug to the vessel wall, unlike DES, in which the drug is concentrated on the struts and thus cannot suppress neointimal hyperplasia between the struts or at the stent edges. Paclitaxel is the drug of choice for DEBs, as its strongly lipophilic properties mean that it remains on the vessel wall for up to a week, without the need for a controlled-release mechanism based on a metal platform and a polymer. This reduces the risk of chronic inflammation and thrombosis and the need for prolonged antiplatelet therapy.2,4

Among the limitations reported for DEB angioplasty are the fact that occasionally the target lesion cannot be crossed and that it can be associated with dissection or recoil, leading to a suboptimal angiographic result and need for bail-out stenting.2 In our study, the lesion could not be crossed in only three patients and the angiographic success rate was very high, with bail-out stenting required in only 3% of lesions, a much lower percentage than reported in previous studies (28% in PEPCAD I19 and 20% in BELLO15).

Assessment of predictors of MACE and TLR at one year demonstrated once again that prognosis was determined by patient characteristics and lesion complexity and that variables such as DEB length or diameter did not correlate with restenosis during follow-up. However, careful analysis of MACE-free survival curves shows that the curves for diabetes and DEB length (known predictors of TLR in previous studies20) diverged during follow-up (Figure 1), reflecting differences that could be significant with longer follow-up.

With regard to treatment of restenosis, the PEPCAD-DES study21 showed that the results of DEB angioplasty in treating DES restenosis were superior to plain balloon angioplasty, with lower rates of TLR (15.3% vs. 36.8%, p=0.005) and MACE (16.7% vs. 50.0%, p<0.001) at six months. In our registry, the event rate was no higher in interventions for DES restenosis, which confirms the efficacy of DEB angioplasty in these patients.

The study has certain limitations which are common in registries, of which the most important are the small sample size and short follow-up to date. It will be interesting to assess the long-term outcomes (beyond one year) of DEB angioplasty to confirm its safety and the persistence of the good results obtained.

This registry of two interventional cardiology centers shows that PCI with DEBs is effective in high-risk patients with complex lesions, with a high rate of angiographic success and a low rate of MACE, including TLR, at one-year follow-up. The results are equally good whether the intervention is for in-stent stenosis or native coronary disease.

The authors have no conflicts of interest to declare.