Percutaneous Coronary Intervention for Very Small Vessels With the Use of a Newer-Generation 2.0 mm Drug-Eluting Stent

Objectives. The outcomes of treating coronary artery disease (CAD) in very small vessels <2.25 mm are sparse. The present study aimed to compare the safety and efficacy of the Resolute Onyx 2.0 mm drug-eluting stent (DES) (Medtronic) with the Onyx 2.25 mm DES for the treatment of CAD. Methods. We retrospectively evaluated patients who underwent percutaneous coronary intervention (PCI) for CAD involving Onyx 2.0 mm DES (Onyx 2.0 group) and Onyx 2.25 mm DES (Onyx 2.25 group) in the 2 consecutive years from November 2016 to November 2018. Major adverse cardiac and cerebral event (MACCE) rate, defined as all-cause mortality, non-fatal myocardial infarction, stroke, and repeat revascularization for target-lesion failure, was reported. Results. A total of 152 subjects with 160 lesions were enrolled. The baseline demographics, lesion characteristics, and procedural results between the two groups were similar. The lesions were significantly shorter (P<.01), fewer stents were consequently deployed (P=.04), and the total stent length was shorter (P<.01) in the Onyx 2.0 group vs the Onyx 2.25 group. At a median follow-up of 673 days, MACCE rate did not differ significantly between the two groups. Multivariate analysis identified the presence of atrial fibrillation, chronic kidney disease, and the use of statins to be independently associated with MACCE. Conclusions. Our data suggest that the use of the Onyx 2.0 mm DES to treat CAD in very small vessels (<2.25 mm) is feasible and safe, and the clinical outcomes were similar to those of the Onyx 2.25 mm group.

J INVASIVE CARDIOL 2021;33(7):E565-E574.

Key words: drug-eluting stent, outcomes, percutaneous coronary intervention, very small vessel

Percutaneous coronary interventions (PCIs) for lesions in small coronary vessels in the era of bare-metal stents and first-generation drug-eluting stents (DESs) are challenging and often associated with an increased risk of adverse clinical events — especially the occurrence of restenosis.^1-5 Strut thickness, lesion length, and the minimum stent lumen diameter were previously identified as independent predictors of restenosis after DES implantation.^6-8 Treatment for small-vessel lesions with second-generation durable-polymer DES options may produce somewhat more favorable results, but data regarding coronary lesions with even smaller reference vessel diameters (RVDs) <2.25 mm are scarce, since the smallest commercially available DES has been 2.25 mm in diameter until recently.^4,5,9-13

However, treating diseases in very small vessels <2.25 mm is becoming more critically important in women with heart disease and in diabetic patients with long, diffuse disease. Furthermore, after angioplasty or stenting in a proximal vessel, it is still imperative to maximize blood flow to the muscle itself by ensuring that smaller run-off vessels are open; this is especially true with regard to side-branch occlusion of small coronary vessels after the stenting of main stems, coronary artery bypass grafting performed on vessels with RVDs between 1.5 mm and 2.25 mm, and so forth.^14,15

Recently, the first report of a DES with a dedicated size to treat lesions with RVD <2.25 mm, ie, the 2.0 mm Resolute Onyx zotarolimus-eluting stent (Medtronic) was published, and it showed that the DES was associated with a low rate of target-lesion failure (TLF) and late lumen loss, without a signal for stent thrombosis.¹⁶ However, it is unclear whether the excellent DES outcomes observed in vessels with a RVD ≥2.25 mm^5,17 and the first report of Resolute Onyx 2.0 mm DES¹⁶ will persist in an all-comers population. Hence, the aim of the present study was to compare the safety and efficacy of the Onyx 2.0 mm DES with those of the Resolute Onyx 2.25 mm DES when treating coronary artery disease (CAD) in very small vessels.

Patient population. The study population was composed of 152 patients who underwent PCI with the use of Onyx 2.0 mm or Onyx 2.25 mm DES at Cheng Hsin General Hospital during two consecutive years from November 2016 to November 2018. The institutional review board at Cheng Hsin General Hospital approved the study protocol, and written informed consent was waived because there were minimal risks to all subjects.

PCI procedures and medications. Before PCI, vessel size was reassessed after the administration of 200 µg of intracoronary nitroglycerin. Then, computer-assisted quantitative coronary angiography (QCA) was performed to assess vessel diameter. Interventional procedures and application of concomitant medication were performed in accordance with medical guidelines, clinical standards, and the physician’s judgment. Lesion predilation, direct stenting, and stent postdilation were left to the treating physician’s discretion. A loading dose of 600 mg clopidogrel, 180 mg ticagrelor, or 20 mg prasugrel was administered to all patients at least 2 hours prior to the coronary angiography. After the intervention, the protocol mandated dual-antiplatelet therapy with acetylsalicylic acid (100 mg per day) and either a thienopyridine (clopidogrel 75 mg per day or prasugrel 3.75 mg per day) or ticagrelor (90 mg twice per day) was prescribed. After PCI, dual-antiplatelet therapy was continued for 6 months, or extended to 12 months in cases with acute coronary syndrome. All endpoints were adjudicated by an independent physician blinded to the clinical outcomes.

Follow-up protocol. After the stenting procedure, all patients remained in the hospital for at least 48 hours. Electrocardiograms were recorded and blood samples were collected for determination of creatine kinase, its MB isoenzyme, and cardiac troponin I within the first 24 hours post procedure, and daily afterward if the data were persistently elevated. At 30 days post procedure, a telephone interview was conducted to assess each patient’s clinical status. Phone interviews were repeated at 6-month intervals thereafter, with structured clinical questionnaires to assess clinical events, medication, and quality of life. All patients with symptoms considered possibly cardiac in origin underwent a complete clinical, electrocardiographic, and laboratory evaluation at the outpatient clinic. When necessary, an angiographic study was performed. Research nurses who were blinded to the study purposes obtained relevant data retrospectively from our catheterization laboratory and Hospital Information System and then entered the data into a computer database.

QCA evaluation. Baseline, postprocedural, and follow-up coronary angiograms were digitally recorded and assessed offline by experienced analysts. The complexity of the lesions was defined according to the modified American College of Cardiology/American Heart Association grading system¹⁸ and SYNTAX score, which was calculated using the SYNTAX score calculator at https://www.syntaxscore.com/calculator/start.htm. For all patients, offline QCA analysis was performed according to current standards by angiographic analysts at the catheterization laboratory; analysts were blinded to the stent size and clinical outcome. The parameters of the RVD, the minimal lumen diameter (MLD), and percent diameter stenosis (%DS, defined as difference between the RVD and MLD divided by the reference diameter and multiplied by 100) were duly measured.

Definitions of angiographic success, device success, procedural success, and outcomes. Device success was defined as the achievement of < 30% residual stenosis by QCA (or <20% by visual assessment) and Thrombolysis in Myocardial Infarction (TIMI) flow grade 3 after successful deployment of the study device. Angiographic success was defined as the achievement of <30% residual stenosis by QCA (or <20% by visual assessment) and TIMI flow grade 3 after the PCI procedure using any percutaneous method. Procedural success was defined as the achievement of device and angiographic successes without the occurrence of MACCE during the hospital stay.

The primary objective of this trial was to show non-inferiority of Onyx 2.0 mm DES vs Onyx 2.25 mm DES regarding MACCE, ie, all-cause mortality, non-fatal myocardial infarction (MI), stroke, and repeat revascularization for TLF. Cardiac death was defined as any death not clearly of extracardiac origin and MI was defined according to the guidelines.¹⁹ The diagnosis of stroke during follow-up required the presence of new lesions in the brain imaging studies. TLF was defined as the composite of cardiac death, target-vessel MI (ST-segment elevation or non–ST segment elevation MI) or target-lesion revascularization (TLR). TLR was defined as any revascularization procedure, percutaneous or surgical, involving the target lesion and performed in the presence of symptoms or objective signs of ischemia during the follow-up interval. Secondary endpoints were the single components of the primary endpoint, as well as probable or definite stent thrombosis according to the Academic Research Consortium definition.²⁰

Statistical analysis. Data were transferred from the database to the Statistical Program for Social Sciences program, version 18.0 for Windows (SPSS). Statistical analysis was performed using a per-patient approach. In patients with a multilesion intervention, only 1 lesion was selected randomly for analysis. The random selection was performed before the analysis of the data by assigning a random number to each lesion and selecting for analysis the lesion with the smallest random number among patients with multilesion intervention. The adequacy of this method was checked by evaluating the reproducibility of the results after selecting the lesion with the greatest random number.

Univariate comparisons of demographic, procedural, and outcome parameters between the two groups (Onyx 2.0 mm and Onyx 2.25 mm) were made. Continuous variables are expressed as mean ± standard deviation and were compared using the Student’s t-test or the Wilcoxon rank sum test. Categorical variables are presented as frequency (percentage) and were compared by the Pearson’s Chi-square test or the Fisher’s exact test. As for the survival analysis, the patients were divided into two groups, dependent upon whether or not MACCE occurred during follow-up. Univariate comparisons of baseline clinical and angiographic characteristics as well as procedural variables between the two groups were made with appropriate tests. In the multivariate Cox proportional hazards analyses, the independent predictors of MACCE in the study patients were determined. Additionally, MACCE-free survival analysis was performed with the Kaplan-Meier method.

A two-sided P<.05 was considered statistically significant for all analyses. Statistical analysis was performed using SPSS, version 18.0 statistical software (IBM SPSS).

Baseline demographics and clinical presentations. A total of 152 subjects with 160 lesions were enrolled. Baseline demographic and clinical characteristics of both groups implanted with Onyx 2.0 mm or 2.25 mm DES are summarized in Table 1. In general, the groups were well matched. There were no significant differences in clinical presentation, incidence of New York Heart Association (NYHA) functional class II-IV at presentation, or complexity of CAD.

Lesion characteristics and procedural results. Lesion characteristics and procedural results are presented in Table 2. The study population consisted of 160 lesions. In the Onyx 2.0 group, target lesions occurred in the left anterior descending (LAD) coronary artery in 44%, in the left circumflex (LCX) coronary artery in 22%, in the right coronary artery (RCA) in 30%, and in saphenous vein grafts in 4%. In the Onyx 2.25 group, the target lesions occurred in the LAD coronary artery in 34%, in the LCX coronary artery in 29%, and in the RCA in 37%. There was no significant difference between the two groups. The majority of lesions (93% of the Onyx 2.0 group and 98% of the Onyx 2.25 group) were anatomically complex (type B2/C). Although the Onyx 2.0 group had more lesions of ≥ moderate calcification than the Onyx 2.25 group, the statistical difference was non-significant. On the other hand, the Onyx 2.25 group had significantly more chronic total occlusion lesions (13% in the Onyx 2.0 group vs 29% in the Onyx 2.25 group; P=.02) and involved numerically more bifurcation lesions (5% in the Onyx 2.0 group vs 14% in the Onyx 2.25 group; P=.07).

The mean lesion length was significantly shorter (21.5 ± 5.9 mm vs 26.1 ± 9.7 mm; P<.01), the mean RVD was significantly smaller (2.0 ± 0.1 mm vs 2.3 ± 0.2 mm; P<.001), and the mean MLD was significantly larger (0.8 ± 0.3 mm vs 0.6 ± 0.4 mm; P<.01) in the Onyx 2.0 group vs the Onyx 2.25 group, respectively, as measured by QCA. The mean %DS was similar in both groups. During PCI, nearly all lesions were similarly predilated in both groups. Postdilation was performed in 39% of the implanted stents in the Onyx 2.0 group vs 43% of the implanted stents in the Onyx 2.25 group (P=.71). Debulking devices, thrombo-suction, and extension catheters were used with similar frequencies in both groups. Significantly fewer mean total stents were deployed in the target lesions of the Onyx 2.0 group vs the Onyx 2.25 group (1.3 ± 0.5 stents vs 1.5 ± 0.7 stents, respectively; P=.04); moreover, the mean total stent length of the Onyx 2.0 group was significantly shorter than in the Onyx 2.25 group (25.0 ± 17.6 mm vs 34.5 ± 23.2 mm, respectively; P<.01). Device success, angiographic success, and procedural success rates were 100% in both groups.

Clinical outcomes. Clinical outcomes are summarized in Table 3. There were no new events during hospitalization. At discharge, the rates of acute or subacute stent thrombosis were 0% in both groups. Dual-antiplatelet therapy was prescribed to 87% of the Onyx 2.0 group and 95% of the Onyx 2.25 group (P=.19). Statins were prescribed to 92% of the Onyx 2.0 group and 91% of the Onyx 2.25 group.

During the mean follow-up periods of 581 ± 253 days in the Onyx 2.0 group and 729 ± 223 days in the Onyx 2.25 group (P<.001), the rates of MACCE were similar in both groups (6% in the Onyx 2.0 group vs 13% in the Onyx 2.25 group; P=.30). There were no significant differences in cardiac mortality (2% in the Onyx 2.0 group vs 2% in the Onyx 2.25 group) or non-cardiac mortality (1% in the Onyx 2.0 group vs 4% in the Onyx 2.25 group; P=.63). The rates of TLF and TLR were also similar in both groups (TLF occurred in 2.0% of the Onyx 2.0 group vs 7.0% of the Onyx 2.25 group [P=.26]; TLR occurred in 2.0% of the Onyx 2.0 group vs 5.0% of the Onyx 2.25 group [P=.54]). However, target-vessel MI occurred in 1/96 Onyx 2.0 patients (1%), and was probably due to late stent thrombosis; no episodes of stent thrombosis occurred in the Onyx 2.25 group.

The study patients were also divided into two groups depending upon whether or not MACCE occurred during follow-up (Table 4A and 4B). In the multivariate Cox proportional hazard analyses, the independent predictors of MACCE in the study patients were determined using variables with a P-value <.10 in the univariate analysis. Multivariate analysis identified the presence of atrial fibrillation (hazard ratio [HR], 14.491; 95% confidence interval [CI], 2.501-83.976; P<.01), chronic kidney disease ≥ stage 3 (HR, 5.121; 95% CI, 1.121-23.399; P=.04), and the use of statins (HR, 0.109; 95% CI, 0.014-0.823; P=.03) to be independently associated with MACCE. Additionally, there was no statistical significance in MACCE rate between stent types (6.0% of Onyx 2.0 patients vs 13% of Onyx 2.25 patients; P=.39). The Kaplan-Meier survival curves are shown in Figure 1.

The main findings of our study are: (1) in this “real world” patient cohort, the Onyx DES of either size yielded favorable results in terms of safety and efficacy and showed no differences in terms of major clinical events at a median follow-up of 673 days; (2) target-vessel MI occurred in 1/96 (1%) of the Onyx 2.0 group, probably due to late stent thrombosis, but no episodes of stent thrombosis occurred the Onyx 2.25 group; and (3) the presence of atrial fibrillation, chronic kidney disease ≥ stage 3, and no use of statins are the three independent predictors of MACCE.

Traditionally, the term “small vessel” denoted a coronary artery with RVD <2.8-3.0 mm.²¹ However, 10 years ago, small vessels were redefined as those with RVD <2.5-2.75 mm. In the past, small vessels caused many technical difficulties during PCI with the use of bare-metal stents and first-generation DESs, and small coronary vessel PCI was an independent predictor of repeat revascularization and adverse cardiac events.^1-8 Because the contemporary new-generation DES options use substantially thinner struts than first-generation DESs, they can be particularly advantageous when used in small-vessel PCI and their smaller strut size may achieve greater lumen gain in small vessels.²² With the availability of 2.25 mm or even 2.0 mm stents, “very small vessel disease” has become fairly treatable with PCI; in patients with de novo lesions in small coronary arteries <2.5 mm, thin-strut or very thin-strut DESs may yield excellent results, as revealed by previous trials.^22-24 Recently, the first report on the DES with a dedicated size to treat lesions with RVD <2.25 mm was published, in which the Resolute Onyx 2.0 mm stent was associated with a low rate of TLF and late lumen loss, without a signal for stent thrombosis.¹⁶ However, a study that compared the outcomes of DES or drug-coated balloons (DCB) for the treatment of very small-caliber CAD tells a somewhat different story. The study involved 111 patients treated with 131 2.0 mm DESs (95 Xience Xpedition SV + 36 Resolute Onyx) and another 83 patients treated with 97 DCBs; the patients treated with DES implantation had a higher incidence of MI than those who received DCBs (7.2% vs 1.2%, respectively; P=.049).²⁵ However, in the present study, the overall incidences of TLR and target-vessel MI were 2.0% and 2.0%, respectively, and were comparable with rates in the Onyx 2.25 group as well as with rates reported in previous Onyx 2.0 very small vessel studies,¹⁶ and much better than rates reported in another study (5.4% and 7.2%, respectively).²⁵ Our data suggest that with thorough preparation prior to stenting (predilation was performed in ≥95% of our procedures), and the availability of <2.5 mm DES with very thin struts, PCI with DES implantation will be the therapy of choice for patients with lesions in small and very small coronary vessels.

Some may argue that DCBs can also be recommended as an optimal treatment strategy for patients with de novo small or very small CAD, since DCB has been established as a safe and effective option for treating in-stent restenosis in BMS or DES. However, it remains unclear whether DCBs have a role to play in de novo coronary lesions, for DES remains their standard treatment. According to recently published clinical trials and meta-analyses, the DCB strategy is non-inferior to the DES and can be considered as a feasible treatment strategy for very small vessel caliber CAD with a favorable outcome.^25-27 There was a significant reduction in rates of MI and death in the patients treated with DCB as compared with those treated with DES, and the difference was statistically significant.^25,27 Moreover, the potential benefit of leaving behind an intact vessel without a stent may reduce the thrombotic events, and no prolonged dual-antiplatelet therapy is necessary. However, there is a steep learning curve for DCB implantation, and operators cannot always expect a stent-like result, as a certain degree of recoil or minor dissection may occur. The theoretical benefit of a stent-free treatment for CAD with DCB in terms of very late adverse events will require a long-term follow-up to prove its efficacy. It is fair to say that in vessels with diameters >2.5 mm, it is probably best to utilize current-generation DES options; however, for vessels with diameters of 2.0-2.5 mm or even smaller, DCBs may be considered a useful backup plan. We therefore prefer DES in the majority of cases where small or very small vessels require revascularization.^14,15 However, DCBs may be used to treat lesions in truly small vessels where DES options will cause concern.

Although the first report of the Resolute Onyx 2.0 mm zotarolimus-eluting stent showed that the DES was associated with a low rate of TLF and late lumen loss, without a signal for stent thrombosis,¹⁶ differences in TLR (and TLF) between the Onyx 2.0 and the Onyx 2.25 groups in our study were not evident after 1 year. However, target-vessel MI occurred in 1 out of 96 patients (1%) in the Onyx 2.0 group, probably due to late stent thrombosis, and there was no episode of stent thrombosis in the Onyx 2.25 group. Whether the probable late stent thrombosis in our study patient has anything to do with the strut thickness (thicker than those of ultra-thin strut thickness, in the ranges of <80 µm) or by insufficient radial force of the Onyx 2 mm DES remains uncertain. Moreover, early cessation of dual-antiplatelet therapy at 6 months is common practice in Taiwan due to the reimbursement policy of the National Insurance, and perhaps somewhat contributed to the late stent thrombosis event.

There was no statistical significance in terms of the incidences of MACCE in the Onyx 2.0 and Onyx 2.25 groups. Multivariate analysis identified the presence of atrial fibrillation, chronic kidney disease ≥ stage 3, and the use of statins to be independently associated with MACCE. Although the very small vessels perfuse a relatively small territory and carry a high risk of dissection, perforation, and restenosis after stenting, we nevertheless consider that PCI for very small vessels <2.25 mm is becoming more critically important. This is because an increasing number of patients suffer from diseased, very small coronary arteries, with diameters between 1.5 and 2.25 mm, and require our help and care. The clinical characteristics of patients with lesions in very small vessels usually include women, elderly people, and patients with concomitant health issues such as diabetes mellitus, heart failure, and peripheral vascular disease.^16,17,24,25 Lesions tend to be more complex and multivessel disease is more common (ACC/AHA type C lesions).^16,17,24,25 The clinical, lesion, and procedural characteristics of the patient population in our study were similar to those reported in previous studies.^16,25 Intervention of these vessels is a challenge because lesions are frequently located distally, and negotiation of the stent is hampered by the distal location of the lesion, vessel tortuosity, and vessel calcification. Besides, the small vessel lumen size leaves little space for error in sizing and stent expansion, and limits the stent option availability for these vessels. Fortunately, due to technological advances, PCI has progressed tremendously over the past decades, and such very small vessels with diffuse CAD can now be treated with excellent results.

Study limitations. Our study was limited by its relatively small sample size, but the overall population was similar in size to other very small vessel studies. Although there is another commercially available 2.0 mm DES (Xience Xpedition; Abbott Cardiovascular), it was not available at our institution during the study period, so we used Onyx 2.25 mm DES as the comparator arm. Second, while most small-vessel or very small-vessel studies reported outcome data within the period of 2 years, longer-term follow-up data are needed. Third, while the TLR and TLF rates were similar in both groups, the lack of a statistical significance between the Onyx 2.0 and Onyx 2.25 groups may be attributed to the relatively small sample size and a shorter follow-up period in the Onyx 2.0 group. Finally, although guidance by intracoronary imaging (eg, intravascular ultrasonography or optical coherence tomography) can facilitate stent sizing in small vessels, it was used infrequently during the present study.

Our analysis suggests that the use of Onyx 2.0 mm DES to treat very small CAD with lesions of RVD <2.25 mm is feasible and safe, and the clinical outcomes were similar to those of the Onyx 2.25 mm DES. Further studies are required to ascertain the long-term benefits of the Onyx 2.0 mm DES and other commercially available DES options of its kind in this setting before unrestricted use of them can be recommended.

Acknowledgment. We thank Ms Chin-Feng Cheng for her assistance in preparing this manuscript.