Access to side branches with a sharply angulated origin: usefulness of a specific wire for chronic occlusions

INTRODUCTION

Despite increasingly effective primary and secondary preventive treatments, coronary artery-related events continue to be the leading cause of morbidity and mortality worldwide.1,2 Lifestyle changes (eg, weight loss, low-salt diet, smoking cessation), medical therapy (eg, anti-hypertensive, lipid-lowering, glucose-lowering, and antithrombotic regimens) in addition to coronary revascularization via percutaneous coronary intervention (PCI) and coronary artery bypass graft (CABG) constitute the multifaceted approach of this disease. Yet despite the advances made in this multimodality approach, cardiovascular morbidity and mortality remain high.

More recently, the central role played by inflammation in the pathogenesis of coronary artery disease from atherosclerotic plaque formation to acute coronary syndrome (ACS), and PCI itself have gained important recognition. Colchicine, an anti-inflammatory agent indicated for multiple inflammatory conditions including pericarditis, gout, and familial Mediterranean fever, has gained attention as a potential attenuator of atherosclerotic inflammation. Acting via the inhibition of tubulin polymerization and eventually blunting immune cell activation and inflammatory response,3,4 recent evidence suggests a benefit of colchicine in the management of the cardiovascular events of patients with clinical signs of coronary artery disease.5 However, its impact among patients in the peri-PCI period remain controversial.

Recent trials have begun exploring the effects of colchicine in the PCI setting, albeit with mixed results. In the Colchicine-PCI trial of patients with non-ST-segment elevation acute coronary syndrome, the administration of colchicine immediately before and after PCI resulted in lower interleukin-6 and high-sensitivity C-reactive protein (hsCRP) levels at 24 hours, but did not show fewer PCI-related myocardial injuries.6 This trial was followed by COPE-PCI that found that when administered 6-to-24 hours before the PCI, colchicine did in fact reduce PCI-related myocardial injuries in a population of patients with stable angina and non-ST-elevation acute myocardial infarction (NSTEMI).7 Nevertheless, the more recent COVERT-MI trial8 found no difference in infarct size or left ventricular remodeling on the cardiac magnetic resonance imaging in patients treated with colchicine compared to those untreated with this agent.

These individual studies may not provide properly powered analyses, particularly in low-rate events such as strokes, on the impact of colchicine regarding secondary prevention in patients in the peri-PCI period, thus prompting the need for a systematic appraisal and meta-analysis of the quality of evidence and treatment effects on major adverse cardiovascular events.

METHODS

Protocol

The search process of this meta-analysis was reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines and is registered with PROSPERO (CRD42021247704). The meta-analysis did not require specific institutional review board approval since it utilized results published in former studies. All relevant information can be found in the trials included. The corresponding author had full access to all the data and final responsibility on the decision to submit the manuscript for publication. The data supporting the findings of this study are available from the corresponding author upon reasonable request.

Search strategy

We performed a comprehensive literature search of all published studies—retrospective, observational, and randomized controlled trials—available on Web of Science, Embase, PubMed, Ovid MEDLINE, Cochrane Central Register of Controlled Trials (CENTRAL), and ClinicalTrials.gov (inception through August 23, 2021, without language restrictions. Case reports, letters to the editor, reviews, and book chapters were not included in this meta-analysis. The keywords used in the search were ‘colchicine,’ ‘coronary artery disease,’ ‘coronary heart disease,’ ‘angina,’ ‘myocardial infarction,’ ‘acute myocardial infarction,’ ‘myocardial ischemia,’ ‘acute coronary syndrome,’ ‘ischemic heart disease,’ ‘percutaneous coronary intervention,’ ‘percutaneous transluminal angioplasty,’ ‘percutaneous coronary revascularization,’ and ‘myocardial revascularization’ including their subheadings, MeSH terms, and all synonyms. References for each of the studies se lected were also screened (the detailed search strategy can be found on the supplementary data). The search process was reported according to the Preferred Reported Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.

Selection criteria

Studies were eligible if they included any the following criteria: a) compared the efficacy of colchicine treatment, at any dose and for any duration, to standard medical treatment with or without placebo; b) included populations of patients treated with PCI regardless of the indication; and c) reported, at least, 1 of the following cardiovascular outcomes: all-cause mortality, cardiovascular mortality, myocardial infarction (MI), stroke or urgent coronary revascularization. Study selection was conducted by 2 independent reviewers (C.E. Soria Jiménez, and J. Chang) first by screening titles and abstracts and then by reviewing full texts and their corresponding references. In case of disagreement over eligibility, a third reviewer (H.M. García-García) assessed discrepancy, and decisions were reached by consensus.

Data collection and study endpoints

Data on study characteristics, patient characteristics, and endpoint event rates were independently drawn and organized into a structured dataset by 2 reviewers (C.E. Soria Jiménez, and F. Hayat), and then compared. All discrepancies resulted in the re-evaluation of primary data and involvement of a third reviewer (H.M. García García). Disagreements were resolved by consensus.

Endpoints

The prespecified primary endpoint was all-cause mortality. Secondary clinical endpoints were cardiovascular mortality, MI, stroke, and any revascularization. Each endpoint was assessed according to the definitions reported in the original study protocols (summarized on table 1 of the supplementary data).

Risk of bias

The risk of bias in each study was assessed using the revised Cochrane Risk of Bias tool (RoB 2.0) for randomized controlled trials (RCTs), and the Risk of Bias in Non-randomized Studies of Interventions assessment Tool from the Cochrane handbook (ROBINS-I) for observational studies. Two investigators (C.E. Soria Jiménez, and J. Sanz Sánchez) independently assessed 5 domains of bias in RCTs: (1) randomization process, (2) deviations from intended procedures, (3) missing outcome data, (4) outcome measurement, and (5) selection of results reported. The same investigators independently assessed 7 domains of bias in observational studies: (1) confounding, (2) selection of participants, (3) classification of procedures, (4) deviations from intended interventions, (5) missing outcome data, (6) outcome measurement, and (7) selection of results reported (table 2 and 3 of the supplementary data).

Statistical analysis

Odds ratios (OR) and 95% confidence intervals (95%CI) were estimated using the DerSimonian and Laird random-effects model with the estimate of heterogeneity taken from the Mantel-Haenszel method. The presence of heterogeneity among the studies was evaluated using the Cochran Q test referred to chi-square distribution (P ≤ .10 was considered statistically significant) plus the I² test to assess inconsistencies. Values of 0% indicated no observed heterogeneity, and values ≤ 25%, ≤ 50%, and > 50% indicated low, moderate, and high heterogeneity, respectively. The presence of publication bias was investigated using Harbord test and visual estimation with funnel plots. We conducted a leave-one-out sensitivity analysis for all outcomes by iteratively removing 1 study at a time to confirm that our findings were not driven by any single study. To account for the different follow-up durations across the studies, another sensitivity analysis was conducted using a Poisson regression model with random intervention effects to calculate the means of inverse-variance weighting of trial-specific log stratified incidence rate ratios. Results were shown as incidence rate ratios, which are exponential coefficients of the regression model.

A meta-regression analysis was conducted using the empirical Bayesian method to estimate the between-study variance tau-squared to assess the effect of colchicine dosage, follow-up duration, percentage of patients with ACS, and percentage of those with diabetes mellitus on treatment effects on the primary endpoint.

Two-tailed P values < .05 were considered statistically significant. Statistical analyses were conducted using the Stata software version 13.1 (StataCorp LP, College Station, United States).

RESULTS

Search results

Figure 1 shows the PRISMA study search and selection process. Out of a total of 1239 unique reports, 12 RCTs5-16 and 1 observational study17 were identified and included in this analysis. The corresponding author of the COOL trial15 was contacted regarding data from a number of patients treated with PCI; 58 out of a total of 80 patients evaluated (72.5%) underwent PCI. The study ultimately met the inclusion criteria and was included in our analysis. The main features of the studies included are shown on table 1. Data on the outcomes, mortality, MI, stroke, and urgent revascularization were reported in 12, 9, 5, and 6 trials, respectively. A total of 3741 and 3673 patients treated with and without colchicine were included (for a total of 7414 patients). Time elapsed from the PCI to the start of colchicine went from immediately before PCI to 13.5 days later as shown on table 1.

Figure 1. Preferred Reported Items for Systematic Reviews and Meta-Analyses (PRISMA) flowchart of database search results and study selection.

Table 1. Characteristics of trials selected

Baseline characteristics

Main baseline characteristics of the patients included are shown on table 2. Most patients were men with a mean age of 60 years, had ACS, and underwent revascularization with drug-eluting stents.

Table 2. Baseline characteristics of patients from each trial

Publication bias and asymmetry

Funnel-plot distributions of pre-specified outcomes indicate absence of publication bias for all the outcomes (figures 1 to 5 of the supplementary data).

Risk of bias assessment

Table 2 and table 3 of the supplementary data summarize the results of the risk of bias assessment. A total of 11 trials were ranked as trials with a low overall risk of bias, 1 presented some concerns while another one was ranked as a trial with a high overall risk of bias.

Outcomes

No differences were seen between patients treated with colchicine and those treated without it or placebo regarding all-cause mortality (OR, 1.06; 95%CI, 0.72-1.55; I² = 0%), cardiovascular mortality (OR, 0.98; 95%CI, 0.42-2.28; I² = 14.2%) or coronary revascularization (OR, 0.64; 95%CI, 0.29-1.42; I² = 49.3%). However, patients treated with colchicine had a lower risk of stroke (OR, 0.38; 95%CI, 0.18-0.81; I² = 0%), and a trend towards a lower risk of MI (OR, 0.84; 95%CI, 0.66-1.07; I² = 0%) (figure 2).

Figure 2. Forrest plot analyses for the main outcomes of death (A), myocardial infarction (B), stroke (C), and revascularization (D). 95%CI, confidence interval; OR, odds ratio.

Sensitivity analyses

In the leave-one-out sensitivity analysis, results were consistent with the primary analysis (tables 4 to 8 of the supplementary data). Similarly, in a sensitivity analysis on the use of estimated incidence rate ratios to account for different lengths of follow-up, findings remained unchanged (table 9 of the supplementary data).

When the risk ratios with random-effects models were estimated, findings remained consistent with the main analysis for all endpoints (table 10 of the supplementary data). Random effect meta-regression analyses found no significant impact of colchicine dosage (P = .33), follow-up duration (P = .88), percentage of patients with ACS (P = .37) or percentage of patients with diabetes mellitus (P = .96) on treatment effect regarding the primary endpoint (table 11 of the supplementary data).

DISCUSSION

This meta-analysis included 7414 patients across 12 RCTs and 1 observational study. It showed some clinical benefits on cardiovascular events with the addition of colchicine to standard medical therapy in patients undergoing PCI. Specifically, we found that the addition of colchicine compared to no colchicine or placebo reduced the risk of stroke showing a trend towards a lower risk of MI both with no observed heterogeneity. Additionally, we observed no differences in all-cause mortality, cardiovascular mortality or coronary revascularization. Significantly, colchicine dosage, follow-up duration, percentage of patients with ACS or diabetes mellitus showed no impact on treatment effect (see PRISMA checklist on table 12 of the supplementary data).

Our outcomes regarding all-cause and cardiovascular mortality are consistent with a prior meta-analysis of 5 RCTs conducted by Fu et al.,18 that also found no significant reduction of mortality, MI, serious adverse events, and restenosis. One explanation for the lack of mortality benefit of both trials may be that although mortality rate was generally low and differences were largely not statistically significant in many of these trials, follow-up duration was generally short (< 30 days) in most studies, and it is possible that higher event rates may be seen with longer follow-up data. We should mention that the meta-analysis conducted by Fu et al.18 included 1 RCT of patients treated with CABG, not PCI. It is possible that the inflammatory profiles of this cohort of patients differ from those treated with PCI (eg, multivessel coronary artery disease, longer postoperative recovery, and higher risk of postoperative complications). As a matter of fact, this mixed population may have led to the lack of reduction seen in the overall rate of MI, serious adverse events, and restenosis. Similarly, a prior meta-analysis conducted by Fiolet et al.19 demonstrated that the addition of colchicine to standard medical therapy in patients with acute and chronic coronary syndromes reduced the risk of the primary endpoint significantly (a composite of MI, stroke, and cardiovascular mortality), and the individual endpoint of MI, stroke, and coronary revascularization with no differences whatsoever on all-cause or cardiovascular mortality. Our results demonstrating a lower risk of stroke and a trend towards a lower risk of MI are more consistent with this meta-analysis. A key difference among the different meta-analyses is the population of patients. Fiolet et al.19 included the LoDoCo20 and LoDoCo221 trials whose inclusion criteria were patients with chronic coronary disease and clinical stability for over 6 months. This amounted to > 50% of patients analyzed who were not in the peri-PCI period and likely had a different inflammatory profile at the time of colchicine administration. These 2 trials also had longer follow-ups (36 and 29 months, respectively) potentially allowing for more time to capture outcome differences like MI and urgent revascularization between the different treatment groups. In contrast, our meta-analysis only focused on patients in the peri-PCI as conducted by Fu et al.18 and expanded the total number of studies analyzed to 12 RCTs and 1 observational study. As far as we know, our study is the largest meta-analysis ever conducted to this date to assess the effects of colchicine on the clinical outcomes of patients in the peri-PCI period.

Alkouli et al.22 reported that the adjusted rate of ischemic stroke increased for patients treated with PCI due to ST-segment elevation myocardial infarction (STEMI) (0.6% to 0.96%), NSTEMI (0.5% to 0.6%), and unstable angina or stable ischemic heart disease (UA/SIHD, 0.3% to 0.72%). In turn, in-hospital mortality was higher (23.5% vs 11.0%, 9.5% vs 2.8%, and 11.5% vs 2.4% for STEMI, NSTEMI, and UA/SIHD cohorts, respectively), and post-PCI stroke was associated with a > 2-fold increase in LoS, a > 3-fold increase in non-home discharges, and a > 60% increase in cost. Given the increasing complexity of patients treated as well as the PCI techniques utilized over the past decade, effective preventive strategies and treatments are needed, and herein lies the opportunity for other anti-inflammatory drugs such as colchicine to further mitigate the morbidity and mortality of patients with post-PCI stroke. In the acute phase of MI, activated inflammasomes mount an intense inflammatory response.23 There is also endothelial damage after PCI, which may result in atherosclerotic plaque destabilization with subsequent thromboembolism causing cerebrovascular events.24 Colchicine may play a role preventing stroke by helping stabilize atherosclerotic plaques in patients undergoing PCI, though this effect may not be robust enough to overcome the direct endothelial injury present at the time of PCI.

Colchicine is a widely available drug with known anti-inflammatory properties. Its mechanism of action is yet to be fully elucidated but has been shown to work partly via the inhibition of NLRP3 (nucleotide-binding oligomerization domain-, leucine-rich repeat- and pyrin domain-containing protein 3) inflammasome, which ultimately downregulates interleukin-1B and interleukin-6, 2 known inflammatory mediators.23-27 It also causes microtubule disruption and decreased neutrophil activation and extravasation. Since elevated levels of inflammatory biomarkers are an independent predictor of major adverse cardiovascular events28-31 our results show that colchicine joining the current medical therapy is a potential addition to further attenuate inflammation regarding the secondary prevention of cardiovascular disease in patients undergoing PCI.

Some limitations of our study include the use of aggregate study-level data as opposed to patient-level data. While this limits subgroup analyses, the overall conclusions would remain the same. There was also a small percentage of patients in each of the studies analyzed who did not undergo PCI, which poses some limitations on the overall effects on a PCI population. However, in all studies, the vast majority of patients eventually underwent this procedure. Similarly, the LoDoCo221 trial enrolled patients who underwent PCI but was ultimately excluded from this analysis as patients required a period of clinical stability 6 months after PCI before starting colchicine therapy. A 6-month gap from PCI to colchicine initiation did not fit in with our period of interest (the peri-PCI period). The study conducted by O’keefe16 was completed in an era of balloon angioplasty, and colchicine treatment in this setting may not be comparable to patients who underwent PCI in the era of statins, modern stents, and antiplatelet agents. Additionally, most patients from our study underwent PCI due to the presentation of ACS, yet there were other clinical presentations including stable ischemic heart disease and unstable angina, and yet others that specifically excluded patients with acute MI. Given the different clinical status at presentation for PCI, it’s likely that the inflammatory profile of these different populations of patients also varied resulting in different clinical outcomes. Nevertheless, despite variation in the inclusion and exclusion criteria, outcome definitions, and colchicine dose and duration, this did not introduce heterogeneity into our results.

CONCLUSIONS

In patients undergoing PCI, the addition of colchicine to optimal medical therapy resulted in a significant reduction of strokes, and a trend towards a lower risk of MI. However, this did not result in lower all-cause and cardiovascular mortality rates, and urgent revascularization.

FUNDING

This research did not receive any specific grants from public, private or non-profit sectors.

AUTHORS’ CONTRIBUTIONS

M.B. Levine was involved in data curation and research. F. Hayat, and J. Chang were involved in data curation and research, as well as in drafting, editing, and reviewing the early draft of the manuscript. C.E. Soria Jiménez, J. Sanz Sánchez, and H. García-García were involved in project conceptualization, data curation, formal data analysis and investigation, methodology, project administration, resources, validation, visualization, as well as drafting, editing, and reviewing all manuscript drafts and its final version.

CONFLICTS OF INTEREST

H.M. García-García declared institutional grant support from Biotronik, Boston Scientific, Medtronic, Abbott, Neovasc, Shockwave, Phillips, and Corflow. The remaining authors declared no conflicts of interest whatsoever.

SUPPLEMENTARY DATA

REFERENCES

INTRODUCTION

Uncovered stent struts is one of the key predictors of stent thrombosis,1,2 and dual antiplatelet therapy (DAPT) has shown to reduce its risk.3 However, DAPT increases the risk of hemorrhage, and nearly one third of the patients treated with percutaneous coronary intervention (PCI) are considered at high bleeding risk. Given the desire for earlier discontinuation of DAPT to reduce the risk of bleeding complications, healing of the stents at earlier time points is desirable.

Drug-eluting stents (DES) significantly reduce neointimal hyperplasia and restenosis compared to bare-metal stents (BMS). However, the main concern regarding first-generation DES was late stent thrombosis due to lack of endothelization of stent struts.1 Therefore, a new generation of DES was developed based on improved thinner metal platforms, new drugs (alternative antiproliferative -limus analogues),4-6 and more biocompatible polymers.7 The evolution of DES moved towards DES with biodegradable polymers.8-10 The comparative studies between DES with biodegradable polymers and BMS showed a lower rate of cardiac death, target vessel-related myocardial reinfarction, and revascularization at 1 year.10 Compared to biodegradable polymer DES, those with durable polymer were noninferior regarding acute coronary syndromes with respect to all-cause mortality, nonfatal myocardial infarction, and revascularization.11 Furthermore, the most recent advance to overcome stent thrombosis has been the polymer-free DES. This type of DES was initially designed to reduce the risk of stent thrombosis in patients with high bleeding risk who could only take short courses of DAPT. It was compared to BMS showing a better efficacy and safety profile.13 It has recently been compared to DES in large clinical trials, especially in patients with high bleeding risk and need for shorter DAPT courses. In these studies, the use of polymer-based zotarolimus-eluting stent was noninferior to polymer-free DES14, and non-measurable differences in device-oriented composite primary endpoints were found.15

However, despite these large clinical trials, there is scarce information on the difference between the characteristics of arterial healing among different types of latest generation DES. Optical coherence tomography (OCT) is a widely used high-resolution intracoronary imaging modality to assess vascular response after stent implantation, thus detecting stent strut coverage and its apposition to the vessel wall.16,17 Stent strut coverage as studied by the OCT is considered a valuable surrogate marker for vessel healing after DES implantation.

The objective of this study was to compare durable polymer-coated everolimus-eluting stents (durable-polymer EES) vs biodegradable polymer-coated everolimus-eluting stents (biodegradable-polymer EES) vs polymer-free BES using stent strut coverage as assessed by the optical coherence tomography (OCT) as a surrogate marker to evaluate short-term arterial healing.

METHODS

Patient population and data collection

This was a prospective, multicenter, non-randomized study that compared 3 different types of DES: a) the durable polymer-coated everolimus-eluting stent Xience DES (Abbott, United States); b) the biodegradable polymer-coated everolimus-eluting stent Synergy DES (Boston Scientific, United States), and c) the polymer-free BES Biofreedom DES (Biosensors International Ltd, Singapore). The study was conducted at 4 Spanish teaching hospitals.

A total of 144 patients were consecutively recruited from January 2018 through December 2019. Patients were eligible if they had been admitted due to stable coronary artery disease or acute coronary syndrome without cardiogenic shock. The medical team selected the type of stent that should be implanted. The detailed study flowchart is shown on figure 1. Inclusion criteria were a) de novo lesions; b) ≥ 1 target lesions in the same or different coronary artery; c) no need for stent overlapping, and a 10 mm minimal distance between the stents; d) stent length between 8 mm and 30 mm; e) use of stents with diameters ≥ 2.5 mm.

Figure 1. Flowchart of patient inclusion.

Exclusion criteria were a) complex lesions including ostial lesions, chronic total coronary occlusions, calcified lesions requiring calcium modification techniques, and bifurcations requiring the kissing balloon technique; b) target lesions in small vessels (< 2.5 mm) and long lesions (> 30 mm) requiring small diameter stents (2.25 mm) or overlapping stents; c) diabetic patients; d) very tortuous arteries that anticipated the impossibility of access with the OCT catheter for follow-up purposes; and e) complications during index procedure. Patients with diabetes mellitus were excluded from the study because of their pro-inflammatory status that facilitates both stent thrombosis and restenosis.18

Once recruited, patients were consecutively assigned to a 1- or 6-month OCT follow-up group. The baseline characteristics, angiographic and procedural data, follow-up data, and outcome data were prospectively collected by the study coordinators. Clinical data at follow-up were obtained from the clinical records. This study was performed following the principles established by the Declaration of Helsinki, ISO14155, and the clinical practice guidelines. The study protocol was approved by the Institutional Ethics Committee (IEC) and the hospital research committee. Informed consent was obtained from all the patients.

Percutaneous coronary intervention, angiographic analysis, and optical coherence tomography

In the index procedure stents were implanted according to the standard approach. Patients were medically treated following the European guidelines on the management of chronic ischemic heart disease or acute coronary syndrome.19

Regarding the initial angiographic analysis, 2 orthogonal projections without coronary guidewire were obtained after finishing the index procedure. These same projections were acquired at follow-up. The off-line analysis of the angiographic images (quantitative coronary angiography [QCA]) was performed in an independent core lab (Barcelona Cardiac Imaging Core-Lab [BARCICORElab]), following their standard protocol. They used a dedicated software (CAAS, version 5.9; Pie Medical BV, The Netherlands). Methods used in this core lab have been previously reported20.

The follow-up angiography was performed at 1-or-6-month follow-up. Angiographic and OCT images were obtained from each patient. The Dragonfly frequency domain OCT C7-XR system (St. Jude Medical, United States) was used. This analysis was performed at the same independent core lab with a dedicated software (St. Jude Medical). Further information can be found on the supplementary data. The struts were classified as non-covered if their surface was totally or partially exposed to the lumen, and without any tissue coverage above its high-density scaffold. Stent strut apposition was defined as the perpendicular distance between the luminal edge of the strut and the vascular wall. Incomplete apposition was considered when distance was higher compared to the total strut thickness considering the addition of strut plus polymer. Intimal hyperplasia was measured as the perpendicular distance between the luminal surface of the stent strut and the luminal surface of the neointima.

Endpoints

The study primary endpoint was the percentage of uncovered struts among durable-polymer EES vs biodegradable-polymer EES vs polymer-free BES as seen on the OCT at 1 month.

The study secondary endpoint was to compare the coverage and apposition of these 3 different types of DES on the OCT 1 vs 6 months after implantation. In addition, we evaluated the intimal hyperplasia in the 3 stent groups over time.

Statistical analysis

Continuous variables were expressed as mean and standard deviation except when they did not follow normal distribution, in which case they were expressed as median and 25^th-75^th percentile. Categorical variables were expressed as frequency and percentage. The analysis of the clinical differences was performed using the chi-square test or Fisher’s exact test for qualitative variables. Comparison among quantitative variables was performed using the 1-way ANOVA test. Generalized estimating equations, considering the clustering nature of the OCT data, were used to conduct analyses at strut level. All probability values were two-sided, P values < .05 were considered statistically significant. Statistical analysis was performed using the SPSS software package, version 22.0 (SPSS, United States). The sample size estimate is shown on the supplementary data.

RESULTS

Baseline clinical characteristics

A total of 104 patients from 4 different hospitals were included in the study; 44 patients were treated with a polymer-free BES, 35 with a biodegradable-polymer EES, and 25 patients with a durable-polymer EES. Of these, 37 patients underwent follow-up angiography and OCT 1 month after DES implantation, and 67 patients after 6 months. Mean age was 57 years; most patients were man (11% women). The inter-group baseline clinical characteristics are shown on table 1 according to the type of stent implanted. We observed a statistically significant difference in the left ventricular ejection fraction that was slightly lower in patients who received a polymer-free BES (54% vs 60%). The number of patients who needed postdilatation was higher in the durable-polymer EES group (68%) especially compared to the polymer-free BES group (38%).

Table 1. Baseline clinical, lesion, and procedural characteristics

Procedural and lesion characteristics

Procedural characteristics based on the type of stent implanted are shown on table 1. We found no significant differences in stent diameter or length in the 3 stent groups. The left anterior descending coronary artery was the most treated of all whereas secondary arteries were scarcely included in this study.

Angiographic analysis

The lesion angiographic characteristics are shown on table 2 and table 1 of the supplementary data. There were 2 patients with angiographic images with insufficient quality for analysis, both from the biodegradable-polymer EES group. There were no significant differences in the lesions before or after the PCI. After 1 month, no significant differences were reported regarding lumen loss or percent diameter stenosis. No differences were reported among the 3 stent groups at 6 months.

Table 2. Angiographic analysis

OCT outcomes

OCT results are shown on table 3, and table 2 in the supplementary data. There were 6 patients with OCT images with insufficient quality for analysis, 2 from the polymer-free BES, 3 from the biodegradable-polymer EES group, and 1 from the durable-polymer EES group. Stent strut coverage and apposition were analyzed, as well as neointimal hyperplasia. Overall, 15 906 struts were examined. Of these, 4380 struts were from durable-polymer EES; 5122 from biodegradable-polymer EES; and 6404 from polymer-free BES; 6184 and 9722 struts were analyzed at 1 and 6 months, respectively.

Table 3. Optical coherence tomography analysis

A high rate of uncovered struts was found among stents 1 month after implantation, with no significant differences (P = .209). We observed ≥ 5% of uncovered struts in > 80% of the patients. There was better coverage in the 3 stent groups at 6 months compared to 1 month (P < .001 polymer-free BES; P = .007 biodegradable-polymer EES; P = .001 durable-polymer EES). No statistically significant differences were reported in strut coverage at 6 months among the different stents (P = .172) (figure 2, figure 3A).

Figure 2. Endpoint comparison at 1 and 6 months. A: percentage of struts covered at 1 and 6 months; B: percentage of struts apposed at 1 and 6 months; C: neointimal hyperplasia at 1 and 6 months. BES, biolimus-eluting stent; EES, everolimus-eluting stent.

Figure 3. Optical coherence tomography. A: optical coherence tomography showing covered and uncovered stent struts at 1 and 6 months; B: optical coherence tomography showing malapposed stent struts at 1 and 6 months. BES, biolimus-eluting stent; EES, everolimus-eluting stent; MO, months.

Regarding strut apposition to the artery walls, no significant differences were reported among the 3 stents after 1 month (P = .497). We observed ≥ 5% of malapposed struts in 29% to 30% of the patients with no differences among stents. No significant differences were reported among the stents after 6 months either. The rate of apposition was higher at 6 months compared to 1 month in all stent groups (P < .001 polymer-free BES; P = .001 biodegradable-polymer EES; P = .029 durable-polymer EES) (figure 2, figure 3B).

When we analyzed neointimal hyperplasia, no significant differences were found at 1 month among the 3 stent groups (P = .083). At 6 months, we found higher hyperplasia in the polymer-free BES compared to the durable-polymer EES (P < .001). We found more hyperplasia at 6 months compared to 1 month in all groups (P < .001 polymer-free BES; P < .001 biodegradable-polymer EES; P = .005 durable-polymer EES; figure 2).

DISCUSSION

The main findings of this prospective multicenter registry are: a) in non-diabetic patients with de novo non-complex coronary lesions treated with durable vs biodegradable vs polymer-free DES, strut coverage was similar and low (≥ 5% of uncovered struts in > 80% of patients) at 1 month; b) there was a similar high rate of malapposed stent struts (4% to 6%) at 1 month; c) intimal coverage and apposition improved significantly at 6 months; d) polymer-free BES had higher intimal hyperplasia at 6 months.

OCT findings suggest that, in non-diabetic patients with non-complex coronary lesions treated with 3 latest generation DES, there is a similar high rate of uncovered and malapposed struts at 1 month. It is after 6 months when we could see better coverage and apposition.

Several small-scale OCT studies have been performed to compare the coverage and apposition of stents with permanent and absorbable polymers.21-24 Conclusions of these studies differ with most of them stating that the absorbable polymer stent has better coverage than the permanent polymer, and 1 of them concluding the opposite.23 One of the studies found coverage to be sufficient after 3 months,22 whereas another stated that coverage improved at 12 months.24 In this study, we found no significant differences in stent strut coverage or apposition between permanent and absorbable polymer at 1 or 6 months on the OCT analysis (figure 2). Differences in stent strut coverage at follow-up may in part be explained by the stent platform, the polymers used to control drug release, and the antiproliferative drug itself. The stents of the study had a similar drug (-limus analogue) but different polymeric features (durable vs biodegradable vs polymer-free), and different platform thickness (the polymer-free BES had a thicker platform). Probably within the first month the drug effect is most important, and it was similar among the study stents (antiproliferative -limus analogues). However, over time (between 1 and 6 months), other stent features such as platform thickness or polymeric features might play a role that could explain the differences seen regarding coverage at 3 months in other studies.22,23

Accordingly, other studies have analyzed other types of polymer-free stents different to the one from our study.25,26 They found that coverage was achieved in a high percentage at 3 to 9 months, reaching conclusions that were similar to our study. One of the studies26 performed an OCT at 1, 3, and 9 months, demonstrating higher strut coverage over time, which is consistent with our results. Only 1 study analyzed the Biolimus A9 polymer-free stent with OCT without comparing it to other stents.27 It was a prospective single-center single-armed study that examined strut coverage of the Biolimus A9 polymer-free stent at 1, 2, 3, 4, 5, and 9 months. Researchers found that coverage was fast and improved over time with the stent remaining safe and effective. These results are similar to ours in the sense that coverage was significantly better at 6 months. However, we also compared polymer-free stents to other polymer-based stents, something that, to the best of our knowledge, has not been tested before.

One of the limitations of extrapolating the clinical safety of the stents and the degree of intimal coverage as seen on the OCT is that there is no consensus on the cut-off value of coverage that would allow safe discontinuation of DAPT. Few studies1,28 have tried to decide on a percentage of coverage, with the only in-vivo study estimating that > 5.9% of uncovered struts was an independent risk factor for stent thrombosis.28 However, these studies were limited in the number of patients included, and only some stents were tested. Larger studies are needed to decide on a security threshold for strut coverage that would make it safe to stop DAPT without increasing the risk of stent thrombosis. Therefore, the rate of coverage at 1 month (80% to 88%) in our study seems insufficient, reaching a very high percentage after 6 months (94% to 97%) in the 3 types of stents.

Finally, our study shows that intimal hyperplasia was significantly higher in polymer-free stents at 6 months. Polymer-free Biolimus A9 has a stainless steel and thicker platform, which has been associated with more intimal hyperplasia and in-stent restenosis in previous studies.²⁷ The other 2 types of DES have a cobalt-chromium platform which has largely substituted stainless steel to provide sufficient strength and visibility with thinner struts of around 70-90 μm, resulting in lower rates of angiographic and clinical restenosis.29 Thus, inflammatory response to this thicker stent platform (130-140 μm) could be in part responsible for this finding.

Study limitations

This was an OCT-based study; unfortunately, it was not powered to assess clinical outcomes. Our study was non-randomized. However, we minimized the confounding factors through selected inclusion/exclusion criteria for patients and lesions (non-diabetic patients with non-complex coronary lesions). We analyzed the differences among the groups and no significant differences were found, except in the left ventricular ejection fraction, which was significantly lower in the group of polymer-free stents. It has not been described in the medical literature whether a lower left ventricular ejection fraction has any correlation with stent thrombosis. This study included selected non-diabetic patients with simple coronary artery lesions. Therefore, the conclusions cannot be extrapolated to other groups with different characteristics.

The distribution of patients who underwent the follow-up angiography in the polymer-free DES group at 1 or 6 months was uneven. More patients rejected the follow-up angiography at 1 month in the polymer-free DES group, which may account for this difference. Finally, the complexity of the analysis of 3 different groups in 2 different moments of time caused a disgregation of cases. This led to a small N in each group, with the potential biases associated.

CONCLUSIONS

In non-diabetic patients, a significantly high percentage of uncovered struts was detected at 1 month with OCT in latest generation DES regardless of the polymeric features of the stent (durable vs biodegradable vs polymer-free stent) in the non-complex coronary artery lesion setting. Our OCT findings do not support improved short-term healing characteristics of stents with biodegradable polymer or polymer-free based -limus elution compared to current generation of durable polymer DES.

FUNDING

This study received funding from Abbott Vascular and Biosensors International. Funders were not involved in the study design, collection, analysis, interpretation of data, drafting of this article or decision to submit it for publication.

AUTHORS’ CONTRIBUTIONS

A. Calvo-Fernández, R. Elosua, and B. Vaquerizo designed the study. J. Gómez-Lara, Héctor Cubero-Gallego, Helena Tizón-Marcos, N. Salvatella, A. Negrete, R. Millán, J.L. Díez, J.M. de la Torre Hernández, and B. Vaquerizo participated in the recruitment of patients and database contribution. A. Calvo-Fernández, C. Ivern, A. Sánchez-Carpintero, and B. Vaquerizo managed the database. A. Calvo-Fernández, X. Durán, N. Farré, and B. Vaquerizo conducted the data analysis and statistics. A. Calvo-Fernández, R. Elosua, N. Farré, H. Cubero-Gallego, and B. Vaquerizo drafted the paper with feedback from all the authors.

CONFLICTS OF INTEREST

J.M. de la Torre Hernandez is an associate editor of REC: Interventional Cardiology; the journal’s editorial procedure to ensure impartial handling of the manuscript has been followed; he has received grants and research support from Abbott Medical, Biosensors, Bristol Myers Squibb, Amgen; honoraria or consultation fees from Boston Scientific, Medtronic, Biotronik, Astra Zeneca, Daiichi-Sankyo. N. Salvatella has received teaching honoraria from Abbott Vascular, and consultation fees from Boston Scientific. The remaining authors declared no other competing interests.

SUPPLEMENTARY DATA

REFERENCES

INTRODUCTION

Ischemic heart disease is the leading cause of mortality in developed countries.1 The rate of acute coronary syndrome (ACS), specially non-ST-segment elevation ACS (NSTEACS), has increased over the last few years, in part, due to the ageing of the population.2-3 Given the underlying pathophysiology4 patients receive specific antithrombotic treatment, and invasive approach is used in most of the cases.1-3 The new guidelines published by the European Society of Cardiology (ESC) on the management of NSTEACS1 include changes compared to the guidelines published back in 2016. The most significant ones include antithrombotic treatment, the revascularization strategy, and several controversial innovations.

In the guidelines published in 2020, early cardiac catheterization within the first 24 hours after admission was advised (level of evidence IA) in patients diagnosed with acute myocardial infarction with GRACE scores (Global Registry of Acute Coronary Events) > 140 or dynamic electrocardiographic changes suggestive of ischemia.1 Also, the previous window of recommendation of 0 to 72 hours for moderate risk patients is now gone.4 On the other hand, the systematic use of pretreatment at admission with an P2Y₁₂ inhibitor antiplatelet drug (ticagrelor, prasugrel or clopidogrel) in patients to be treated with an early invasive strategy is now ill-advised.1

The objective of the IMPACT registry (Time of intervention in patients with myocardial infarction with non-ST segment elevation, management and outcomes [IMPACT-TIMING-GO]) is to get the big picture on the current treatment of NSTEACS, in Spain, in association with catheterization times, use of pretreatment in these patients, and describe the possible prognostic implications of the different strategies used in real life.

METHODS

Study design and population

This is an observational, prospective, multicenter, and nationwide registry that will include all consecutive patients admitted with a diagnosis of NSTEACS to the different participant centers, treated with diagnostic coronary angiography, and with unstable or causal atherosclerotic disease regardless of further treatment administered by the heart team. The baseline characteristics of the patients included, and their clinical progression regarding in-hospital events will be studied. Patients will undergo a 1-and-3-year clinical follow-up period.

This registry has been promoted by the Spanish Society of Cardiology Young Cardiologists Working Group with scientific support from the Spanish Society of Cardiology Research Agency. Also, it has been approved by different Research Ethics Committees with drugs from all the participant hospitals. Finally, it has been designed according to the STROBE checklist for observational studies.

The list of centers that will eventually participate in the registry is shown on figure 1. Inclusion and exclusion criteria are shown on table 1. The presence of elevated markers of myocardial damage or electrocardiographic changes is not mandatory. Patients with a clinical diagnosis of unstable angina can be included as long as coronary angiography confirms the clinical diagnosis.

Figure 1. Map with the Spanish participant centers in the IMPACT-TIMING-GO registry.

Table 1. Inclusion and exclusion criteria of the IMPACT-TIMING-GO registry

Endpoints

The study primary endpoint is to know the degree of compliance of the recommendations included in the clinical practice guide-lines in patients admitted due to NSTEACS treated with coronary angiography, in Spain, describe the use of antithrombotic treatment before cardiac catheterization, and the time elapsed until it is performed in the real-world clinical practice.

The secondary endpoints are:

• – To describe the baseline, clinical, and epidemiological characteristics of the study population.
• – To study the rates of cardiovascular mortality, new revascularization, stent thrombosis, and hospitalizations due to heart failure during admission and at the 1-and-3-year follow-up.
• – To describe major cardiovascular adverse events of all-cause mortality, non-fatal stroke, non-fatal infarction, and the rate of major bleeding grades 3, 4, and 5 according to the BARC scale (Bleeding Academic Research consortium.5) Data will be analyzed during admission and at the 1-and-3-year follow-up.
• – To know the medical treatment at discharge and at follow-up of patients discharged in Spain after NSTEACS.
• – To know the degree of control of the different cardiovascular risk factors associated with the endpoints defined in the ESC guidelines 2021 on prevention of cardiovascular disease in the routine clinical practice.6

Data curation and definitions

Data will be collected prospectively by trained medical investigators from each participant center in a specific standard form. Demographic data, the baseline clinical characteristics, and all analytical, electrocardiographic, and echocardiographic data will be included as well.

Similarly, data on disease progression and the in-hospital stay, indication for coronary angiography and when it is be performed, type of treatment received (conservative, stent implantation or revascularization surgery), and the in-hospital complications occurred (hemorrhages and severity, heart failure or shock, reinfarction, stroke, confusional state, mechanical and arrhythmic complications, infectious complications requiring antibiotic therapy, and mortality causes) will be collected. Finally, the medical treatment at hospital discharge and level of compliance of the current recommendations based on the clinical practice guidelines will be studied too.

The definitions of the variables are shown on table 2.7-8

Table 2. Definitions of target variables

Follow-up

Clinical follow-up to detect events will be conducted by medical investigators through on-site visits, health record reviews or phone calls with the patient, family members or treating physician at 1 and 3 years. Clinical variables, functional class, and additional variables (analytical, electrocardiographic, and echocardiographic, and treatment received) will be included. The overall mortality rate and its causes, need of emergency hospitalization (duration > 24 hours) and its causes, and the rates of non-fatal infarction and stroke will be collected as well. All deaths due to myocardial infarction, sudden death or heart failure will be considered cardiovascular deaths.

Sample size estimate

Taking the events seen in previous studies with a population of similar characteristics as the reference,9-14 a sample size of 800 patients will be enough to know the baseline characteristics of the study population, and the therapeutic approach currently used in Spain in our routine clinical practice. Patients lost to follow-up will be handled by multiple imputation.

Statistical analysis

Categorical variables will be expressed as number and percentage. Quantitative variables will be expressed as mean ± standard deviation. Quantitative variables with normal distribution will be expressed as median and interquartile range [25%-75%]. The normal distribution of quantitative variables will be assessed using the Kolmogorov-Smirnoff test. Regarding the reference variables, Student t test will be used to compare quantitative variables, and the chi-square test or Fisher’s exact test, if applicable, to compare categorical variables. Statistical analysis will be performed using the SPSS statistical software version 22.0 (IBM Corp., Armonk, United States).

Specific studies on subgroups of special interest will be conducted: feminine sex, patients ≥ 75 years, those with GRACE scores > 140, diabetic patients, those with a past medical history of renal failure, with an indication for chronic oral anticoagulation, with multivessel disease, acute myocardial infarction, ventricular dysfunction according to the current clinical practice guidelines and based on the day of admission (holiday vs working day), and patients who require transfer to tertiary centers to receive a coronary angiography.

Ethical principles

Inclusion in the study will not imply changes to the patients’ treatment. Instead, it will follow the routine clinical practice and the recommendations set forth by the current clinical practice guidelines. Therefore, antithrombotic treatment and additional examinations including the need for a coronary angiography and the time it is performed will all be decided by the heart team based on the routine clinical practice. Coronary angiography, vascular access, antithrombotic treatment during the procedure, and the material and devices used will all be decided by the treating operator in charge of the case. All patients will sign a written informed consent form before being included in the study that will be conducted in full compliance with the Declaration of Helsinki. This study will also observe all legal regulations applicable to this type of studies and follow the good clinical practice rules while being conducted.

DISCUSSION

The IMPACT-TIMING-GO registry will give us information on the current real-world management of patients with NSTEACS with invasive treatment and causal coronary artery disease, which will allow us to assess the degree of implementation of the current recommendations of ESC guidelines 2020 on cardiac catheterizations performed within the first 24 hours and no pretreatment with P2Y₁₂ inhibitors. Similarly, different prognostic differences that early invasive treatment and no pretreatment could have in the real life of patients diagnosed with NSTEACS could be suggested.

Despite the clinical practice guidelines recommendations on the invasive treatment of patients with NSTEACS, the main clinical trials published to this date have been unable to demonstrate any clear benefits from systematic early invasive treatment.9-14 The VERDICT trial,9 published in 2018, randomized 2147 patients with NSTEACS on a 1:1 ratio to receive early (< 12 hours) or delayed (48 to 72 hours) cardiac catheterization. No significant differences were found in the composite endpoint of major cardiovascular events at 4-year follow-up. However, in the subgroup of patients with GRACE scores > 140 statistically significant differences were seen favorable to the early strategy regarding major adverse cardiovascular events (hazard ratio, 0.81; 95% confidence interval, 0.67-1.01; P = .023). Consistent with this, the TIMACS clinical trial10 published in 2008 of 3031 patients with NSTEACS found no differences in the primary endpoint when early invasive strategy (< 24 hours) and delayed approach (> 36 hours) were compared, except for, once again, in patients with GRACE scores > 140. Other randomized clinical trials with fewer patients show contradictory results11-14 some without significant differences.15 Also, in many cases, the results favorable to the early strategy are associated with refractory ischemia, not with hard endpoints like cardiovascular mortality or non-fatal myocardial infarction. In Spain, evidence on the management of NSTEACS is prior to the current clinical practice guidelines,16-17 and the most recent registry is retrospective, which is suggestive of a possible mortality benefit in patients with GRACE scores > 140.18 Over the last 2 decades, in our country, the use of an invasive strategy in patients with NSTEACS has increased significantly from 20% in the MASCARA registry in 200516 up to 80% in the DIOCLES study from 2012.17 However, evidence is scarce on catheterization times, our capacity to adapt to current recommendations (the median time of the DIOCLES trial was 3 days), the possible impact this time reduction can have, and on the consequencies from not starting antiplatelet pretreatment in patients who don’t meet the times recommended.

On the other hand, the current formal recommendation from the clinical practice guidelines of not pretreating systematically with a P2Y₁₂ inhibitor (level of recommendation IIIA1) patients on early invasive treatment is mainly based on 3 clinical trials and their meta-analysis.19 In the ACCOAST trial, pretreatment with prasugrel did not reduce thrombotic events in patients with NSTEMI. However, cardiac surgery-related and potentially fatal hemorrhages increased.20 We should mention that the median time elapsed since the prasugrel loading dose until the coronary angiography was performed was 4 hours. In the ISAR-REACT 5 trial published in 2019, a non-pretreatment strategy with prasugrel in patients with ACS vs pretreatment with ticagrelor proved superior regarding the primary endpoint of thrombotic events with a tendency towards fewer hemorrhagic events.21 We should mention that the intrinsic effect of the drug used should not be obviated or else the fact that the median time elapsed since randomization until the prasugrel loading dose was received in the non-pretreatment group was 60 minutes. Finally, the first study that compared 2 different pretreatment strategies vs the intraoperative administration of ticagrelor did not show any clear benefits regarding thrombosis or a deleterious effect of pretreatment regarding bleeding.22 Once again, the median time elapsed until the cardiac catheterization was performed was < 24 hours since hospital admission (23 hours). Surprisingly, clinical practice guidelines leave the door opened to a weak level of recommendation (IIbC) regarding pretreatment of patients in whom early catheterization < 24 hours is not possible.1

In conclusion, current recommendations on early invasive treatment and no antiplatelet pretreatment in patients with NSTEACS are controversial and can also be difficult to implement in the routine clinical practice in our setting. The ultimate objective of the IMPACT-TIMING-GO registry is to shed light on the current management of NSTEACS in Spain. After the impact that the COVID-19 pandemic has had on the general structure of the healthcare system and the drop in the number of interventional procedures performed in 2020,23 we should expect to see pre-pandemic numbers in 2022 and cath labs and cardiac surgery intensive care units going back to normal. Therefore, moment seems ripe to conduct a real-world registry.

CONCLUSIONS

The IMPACT-TIMING-GO registry is the first prospective study ever conducted in Spain that will be giving us information on the early therapeutic strategies—both pharmacological and interventional—performed in our country in patients with NSTEACS after the publication of the ESC guidelines 2020, and the impact of these and other measures indicated in these patients at follow-up.

FUNDING

This unfunded study has been promoted by the Spanish Society of Cardiology Young Cardiologists Working Group with scientific endorsement from the Spanish Society of Cardiology.

AUTHORS’ CONTRIBUTIONS

Study design, data curation and review, statistical analysis, and manuscript drafting: P. Díez-Villanueva, F. Díez-Delhoyo, and M.T. López-LLuva. All the authors participated in the manuscript review and approval process.

CONFLICTS OF INTEREST

None reported.

ACKNOWLEDGEMENTS

We wish to thank the Spanish Society of Cardiology Young Cardiologists Working Group for their drive to engage the youth in medical research.

REFERENCES

INTRODUCTION

Coronary bifurcation lesions are still challenging for interventional cardiologists. The complexity surrounding such lesions regarding their anatomical, functional, and even clinical aspects truly complicates the management of this entity despite its high incidence rate that can be up to 20% of all the lesions that are treated at a cath lab on a routine basis. The relentless publication of articles on such lesions over the last few decades, the creation of specific study groups like the European Bifurcation Club, and the periodic publication of consensus documents for the management of this entity shows, without a doubt, that this scenario is in constant change and has not been solved today yet. One of the most controversial aspects is the importance of the side branch (SB) regarding the long-term prognosis of such lesions. Drug-coated balloon (DCB) is part of the therapeutic armamentarium of interventional cardiologists to treat coronary bifurcation lesions. Its utility for the management of certain anatomical settings like in-stent restenosis (ISR) type of lesions has already been demonstrated. However, its effectiveness to treat the SB is much less evident with scarce studies available in the medical literature. The theoretical advances posed by this device to treat the SB would be the administration of antiproliferative drugs into the ostium mainly, the lack of distortion of its original anatomy, and the minimization of strut deformation at carina level.1

This article presents a registry with the results obtained in our unit with the management of SB with DCB with a longer than usual clinical follow-up in this type of studies.

METHODS

This was a single-center, prospective-retrospective registry started back in 2019 of all coronary bifurcation lesions where the SB was treated with paclitaxel-coated DCB from October 2018 through March 2022. The device used was the SeQuent Please NEO (Braun, Germany), a paclitaxel-iopromide coated polymer-free balloon using Paccocath technology. Inclusion criteria were the presence of coronary bifurcation lesions with 1 compromised SB of, at least, 2 mm in diameter through visual angiographic estimate regardless of the aprioristic presence of a diseased SB or the appearance of carina displacement or slow flow after treating the main branch (MB). Also, the operator should consider the DCB approach of clinical and prognostic interest. Patient recruitment in the registry was on the rise: 4 patients in 2018, another 4 in 2019, 9 patients in 2020, 31 in 2021, and finally 7 within the first 3 months of 2022. No exclusion criteria were established. Approach strategy consisted of an early provisional stenting or DCB technique to treat the MB when damaged. Further management of SB with DCB was left to the operator’s criterion if, after treating the MB, significant damage done to the SB would require stenting in such branch. In that case, the patient would not be included, and the SB would not be eligible for treatment with a DCB. If, after preparing the lesion, the operator would actually consider using the DCB option, that would be the time to include the patient in the study. The rate of procedural failure—defined as the impossibility to cross the lesion with the DCB once it was used or unsatisfactory angiographic outcomes after balloon inflation involving SB stenting. The protocol for using the DCB—based on the recommendations established on the use of such devices—consisted of SB predilatation with non-compliant or scoring balloons in a 0.8-1 vessel/balloon diameter ratio, use of the device if an acceptable angiographic result with TIMI grade-3 flow was achieved, lack of significant dissection, and residual stenosis < 30%. If other lesions different from the one that triggered the inclusion in the registry needed revascularization, this was scheduled for a second surgical act. The study design followed a per protocol analysis to estimate the benefits of the technique described compared to the routine clinical practice including cases with successful DCB treatment at the follow-up and excluding those with acute device failure or impossibility to use the device once opened for being unable to cross the lesion. The lack of dissection after DCB that required stenting with residual stenosis < 50%, and final TIMI grade-3 flow was considered as procedural success. Device failure, on the other hand, was considered as an impossible DCB inflation once used or the need for stenting the SB with unsatisfactory DCB results. Different clinical variables from the patient were analyzed, as well as the lesion anatomy, and the procedural intervention per se. Retrospective clinical follow-up of patients successfully treated with the DCB was conducted. Follow-up went on for a maximum of 2 years after the procedure, and prospectively since the registry started back in 2019 until present time. This follow-up was conducted through phone calls or by checking the patients’ electronic health records. The ARC-2 definitions2 were used to collect the adverse clinical events including a composite endpoint of all-cause mortality, cardiac death, myocardial infarction, device thrombosis, clinically driven target lesion failure and revascularization, target vessel failure outside the target lesion, and revascularization of other lesions occurred at follow-up. All patients signed their written informed consent forms, and the study was approved by our center research ethics committee.

Statistical analysis

Continuous variables are expressed as mean ± standard deviation. Categorical variables are expressed as frequency and percentage. Also, actuarial curves of adverse event-free survival using the Kaplan-Meier method were built, specifically target lesion failure-free and adverse event-free curves (all-cause mortality, target lesion revascularization, target vessel failure, and revascularization of other lesions).

RESULTS

A total of 55 patients and 56 lesions were included since 2 different bifurcations found in 1 of the patients were treated in the same procedure. The patient/lesion flowchart included in the study is shown on figure 1. The patients’ clinical characteristics are shown on table 1. Vascular access was radial in 100% of the cases using a 6-Fr introducer sheath also in all of them. Table 2 shows the anatomical characteristics of target lesions. Figure 2 shows a schematic representation of the type of lesion according to the Medina classification. Table 3 shows the variables associated with the procedure. We should mention that all the clinical and anatomical data shown here, the patients’ high-risk profile with high prevalence of cardiovascular risk factors, and the large number of ISR-type of lesions reached 32.1% of the sample. The rate of lesions included with damage to 2 or 3 different bifurcation segments was 71.4% (40 out of 56). Regarding procedural factors the high rate of procedural success was significant (low rate of acute device failure with only 1 case of a type A dissection image after DCB inflation without damage to the distal flow and > 30% residual stenosis). Therefore, because of lesion location at ostium level, and possible damage to the MB (the left anterior descending coronary artery in this case), the operator decided to perform drug-eluting stent implantation for sealing purposes (figure 3). In all the remaining procedures, the acute result of the DCB was successful. In our series, the scarce use of intracoronary imaging modalities (only 7.1%) was also remarkable.

Figure 1. Flowchart of patients/lesions included in the study.

Figure 2. Number of lesions based on the type of bifurcation damage according to the Medina classification.

Figure 3. Only case of acute device failure. A: diagonal branch ostial lesion prior to the intervention (arrow); B: suboptimal outcome after drug-coated balloon (arrow); C: final outcome after stenting the side branch.

Table 1. Patients’ clinical characteristics

Table 2. Anatomical characteristics of the lesions

Table 3. Procedural characteristics

The rate of adverse events at follow-up is shown on table 4. After a median follow-up of 12 months (377 ± 244 days; range, 64-734 days) only 2 clinically driven target lesion revascularizations (3.6%) were reported. Both were performed due to in-stent lesions that did not reach the target lesion proximal or distal borders. The first one was performed in a case of ISR of the SB in a very small vessel without acute ischemia whose new revascularization was performed late, more specifically, 23 months after the index procedure (figure 4). The second one was performed 6 months after the procedure—also without acute ischemic signs—but with ISR in the main vessel while the SB remained patent without significant restenosis (figure 5). Both cases were treated with drug-eluting stent implantation. Two deaths were reported: 1 cardiac death due to advanced left ventricular dysfunction in an 80-year-old woman who, after percutaneous coronary intervention, was implanted with a transfemoral aortic valve and a definitive pacemaker, with poor disease progression that eventually led to her death. The other death was septic shock related. No admissions due to acute myocardial infarction or episodes of target lesion thrombosis (both probable and definitive) were reported. No cases of target vessel failure outside the target lesion were reported either. A total of 5 revascularizations of other lesions (9.3%) were performed—all of them scheduled—but none due to acute coronary syndrome. The Kaplan-Meier curves showing target lesion revascularization-free and adverse event-free survival are shown on figure 6.

Table 4. Rate of adverse cardiovascular events at follow-up

Figure 1. First case of target lesion failure due to late restenosis. A: early in-stent restenosis type of lesion in obtuse marginal artery (arrow); B: final outcomes after drug-coated balloon; C: new in-stent restenosis in the side branch at 23 months (arrow).

Figure 5. Second case of target lesion failure. A: early obtuse marginal artery bifurcation lesion with distal left circumflex artery (arrow); B: outcomes after stenting the main branch, and drug-coated balloon implantation into the side branch; C: 6-month follow-up with restenosis at main branch level (arrow).

Figure 6. Kaplan-Meier curve of actuarial target lesion revascularization (TLR)-free survival and composite adverse events-free survival (all-cause mortality, TLR, revascularization of other lesions).

DISCUSSION

The latest document of the European Bifurcation Club on the utility of DCBs to treat SBs in coronary bifurcation lesions pays little attention to it due to the lack of large enough clinical trials to be conclusive.3 Despite the huge amount of medical literature available on the management of coronary bifurcation lesions, the actual significance of the SB and its involvement in target lesion failure has not been properly explained to this date. A study conducted by Oh et al.4 conclude that treating the SB in 1089 patients with bifurcation lesions at left anterior descending coronary artery-diagonal branch level was associated with a lower—yet not statistically significant—rate of target vessel failure. However, this difference was statistically significant when the subgroup studied included low-risk patients. On the other hand, a different clinical trial that studied factors associated with failed revascularizations of the left main coronary artery bifurcation found that the presence of MB stent struts inside the SB ostium was one of them5 suggestive that the use of intracoronary imaging modalities like intracoronary ultrasound or optical coherence tomography could improve results, at least, on such location, by telling us what patients would benefit from specifically treating the SB.

The strongest evidence available to this date leans towards the utility of DCB to treat ISR-type of lesions without a word dedicated to the SB. Very few studies have focused on the effectiveness of DCB to treat the SB. Such document advocates for treating coronary bifurcation lesions with the provisional stenting strategy according to the latest clinical practice guidelines drafted by the European Cardiology Society followed by treating the SB with a DCB. The first clinical trials on this regard were published back in 2011 like the DEBIUT,6 BABILON,7 DEBSIDE,8 the study conducted by Herrador et al.,9 the PEPCAD V,10 and the PEPCAD-BIF11 clinical trials. These studies showed contradictory—yet overall satisfactory—data regarding the effectiveness of DCB. These studies presented better quantitative angiographic parameters regarding restenosis or late lumen loss. However, not in every one of them this was associated with a lower rate of revascularization. As a matter of fact, there were doubts around the possibility of a higher rate of late thrombosis suggested by some of these trials. The recently published BEYOND clinical trial conducted by Jing et al.12 compared the use of a conventional balloon vs DCB to treat the SB with a 9-month angiographic follow-up. This trial found that the DCB was associated with better results regarding less late lumen loss. However, such an improvement did not translate into a lower rate of clinical adverse events since surprisingly enough no new revascularizations were reported in any of the 2 groups. A recent meta-analysis13 that included 10 studies on the effect of DCB on the SB concluded that such technique improved the angiographic outcomes significantly. However, this did not translate either into statistically significant clinical outcomes (target lesion failure mainly) basically due, according to the authors, to the low rate of this adverse event reported, and the fact that the study was statistically underpowered due to its small sample size. In a different study published in 2022,14 the management of coronary bifurcation lesions of left main coronary artery using 2 strategies was compared: double stenting for the MB and the SB vs 1 stent into the MB, and 1 DCB into the SB. They found controversial results at follow-up between both groups in different angiographic parameters with similar rates of restenosis and adverse events. However, the group treated with DCB significantly improved all the parameters associated with the SB (left circumflex artery, in this study)—as opposed to those associated with the MB (left main coronary artery-left anterior descending coronary artery)—with less late lumen loss (0.43 vs -0.17; P < .001), less luminal narrowing (16.7 vs 32.1; P = .002), and greater minimal lumen diameter (2.4 vs 1.8; P = .0031). Still, the rate of restenosis in the left circumflex artery (SB in this study) was 4 times higher in the double stenting group compared to the DCB group (30.4% vs 7.7%) although this difference was not statistically significant (P = .09). This could be indicative of greater superiority of the DCB if studies with larger samples would be conducted. Another recent study published in 202115 randomized 219 true de novo coronary bifurcation lesions where the SB was treated with conventional balloon vs DCB. At 12-month clinical and angiographic follow-up, significant improvements were reported in both the angiographic (less late lumen loss and greater minimal lumen diameter) and clinical parameters with a lower rate of major adverse cardiovascular events being reported. This improvement, however, did not translate into significant reductions regarding new revascularizations or target vessel failure.

The rate of target lesion failure requiring new revascularization was 3.6%, a figure that is consistent with most former studies. However, the range found goes from a surprising 0% up to a whopping 22%. However, we should mention the truly unfavorable clinical and anatomical profile of our sample since in most clinical trials, ISR-type of lesions, left main coronary artery disease or ST-segment elevation acute coronary syndrome—all allowed in our registry—were considered exclusion criteria regarding.

Out of the only 2 cases reported of target lesion failure requiring new revascularization, 1 occurred in a patient with an ISR-type of lesion. This occurred precisely in the SB while in former studies7—as already explained—the main incidence rate of failure occurred in the MB, not in the SB. The exclusion of patients with ISR would account for this difference. In our sample this type of lesions were 32.7% of all the lesions included. This added to the high rate (30.6%) of Medina 1,1,1 coronary bifurcation lesions (the one with the greatest complexity of all bifurcations) demonstrates the truly unfavorable profile of our sample. As a matter of fact, the rate of lesions included with damage to, at least, 2 segments of 1 bifurcation according to the Medina classification reached 71.4%. Very few studies have been conducted on this subgroup of patients. One of the most significant ones is the one conducted by Harada et al.16 that included 177 patients with ISR-type of lesions both in the MB and the SB treated with DCB. The latter was used in 80.6% of the SBs. The rate of binary restenosis was 24% at 6-to-8-month angiographic follow-up while the 1-year rate of new target lesion revascularization was 22%.

Limitations

Our study main limitation is the lack of a control group with lesions of similar characteristics, which would have allowed us to compare both groups and determine exactly the impact DCB has on the prognosis of patients. Similarly, the lack of angiographic follow-up does not discard the possibility of device failure. However, this would probably occur in the SB, not the MB, since it is in the latter where target lesion failure occurs according to the BABILON clinical trial.7 Another study limitation we should take into consideration is the elevated presence of small SB with a rate of use, in our sample, of DCB sizes < 2.25 mm of 43.7%. This would make target lesion failure go clinically inadvertently in some of these cases. Finally, we should mention that this study is limited by the relatively small number of patients included. Also, because due to its observational nature, no selection biases can be excluded.

CONCLUSIONS

The findings presented here show the experience of a single center with a very low rate of acute procedural complications, and a low rate of long-term adverse events despite dealing very high-risk profile lesions and patients with a 3.6% rate of target lesion failure reported. It is crucial to select the right type of lesions that can benefit from such therapy (basically the lack of a large plaque burden in the SB), a very refined technique of lesion preparation, and a greater use of tools to guide the angioplasty like intracoronary ultrasound or optical coherence tomography, preferably in ISR-type of lesions whose clinical progression is more unfavorable compared to that of de novo lesions. Despite the low rate of adverse events reported since no comparison with a control group was made, no definitive conclusions can be drawn on the advantages of DCBs in this clinical setting. We can only say that both in the «real-world» and the routine clinical practice described here, such strategy yields good long-term results without prejudice to other strategies may have given better or worse results regarding effectiveness. Randomized clinical trials are needed with enough statistical power and large enough samples to corroborate the promising data obtained from former studies to confirm or discard the superiority of DCB in the management of the SB in coronary bifurcation lesions. Until that time, the DCB can be considered a therapeutic tool that can be tremendously useful to improve the long-term results obtained in this type of complex lesions.

FUNDING

None whatsoever.

AUTHORS’ CONTRIBUTIONS

J. Valencia drafted the manuscript. J. Valencia, F. Torrez-Mezcua, and M. Herrero-Brocal participated in data curation, and in the clinical follow-up of the patients. J. Valencia, J. Pineda, P. Bordes, F. Torres-Saura, and J.M. Ruiz-Nodar participated in patient recruitment and in the manuscript critical review process. All the authors gave their final approval to the manuscript.

CONFLICTS OF INTEREST

None reported.

REFERENCES