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

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

INTRODUCTION

Percutaneous treatment of coronary bifurcation lesions can represent up to 20% of all coronary lesions treated.1 The definition of bifurcation lesion given by the European Bifurcation Club2 includes those that effect a relevant side branch (SB) whether by its angiographic diameter or the myocardium at risk associated with such SB.

The pseudo-fractal anatomy of coronary bifurcation lesions3 involves a significant caliber difference between proximal and distal segments. The study of the structural limits of tubular stents has favored the development of treatment techniques4,5 designed to optimize implantation in an anatomy that is not completely cylindrical with good angiographic and clinical results.6-10 Also, a progressive escalated strategy has been agreed based on parameters of damage to the SB from the provisional stenting technique to the complex double stenting one.

Although specific platforms have been designed to treat coronary bifurcation lesions, these have been limited for expert operator use only. On the one hand, dedicated stents can be categorized into stents designed to treat the main vessel by securing proper access to the SB, (Nile Croco & Pax, Minvasys, France, the Multi-Link Frontier, Abbott Vascular Devices, United States or the TAXUS Petal, Boston Scientific, United States). On the other hand, stents designed to treat the SB first to later complete the main branch (MB) with a tubular stent (Tryton Side Branch stent, Tryton Medical, United States; Sideguard, Cappella Inc, United States). However, because of the discrete results reported or their complexity, they have not become entirely popular.

The BIOSS LIM C stent11 (Balton, Poland) is a 70 µm ultra-thin-strut chrome-cobalt platform with a sirolimus-eluting biodegradable polymer of polylactic acid. It is a dedicated stent for bifurcations that consists of 2 segments of different size linked by 2 long connective struts (figure 1). Goal is to keep the pseudo-fractal correlation between the proximal and distal portions of the SB and facilitate the technique to access the SB or the provisional stenting technique. Former studies have given good results with successful implantation rates of 100%, and target lesion revascularization rates from 6.8% to 9.8%.12-14

Figure 1. Image of the BIOSS stent design. Note the structure in 2 bodies with central space for the side branch.

The objective of this study is to describe immediate angiographic results assessed through quantitative coronary angiography (QCA) in terms of deformation of the lesion native angulation and expansion, especially at the level of the polygon of confluence and at the origin of the side branch.

METHODS

Patients

This is a prospective, multicenter registry started by independent investigators that included patients with ischemic heart disease referred for percutaneous coronary revascularization of a bifurcated lesion considered as a lesion with distal branches with a minimum diameter of 2 mm.

Patients with lesions damaging the bifurcation on the SB only were excluded. Also, patients with restenosis, complete total coronary occlusions, a contraindication for dual antiplatelet therapy, cardiogenic shock, minors or patients who gave their express rejection to be included.

Study was conducted according to the Declaration of Helsinki and the study protocol was approved by the different ethics committees of participant centers. The specific written informed consent was obtained from all the patients included in the study.

Procedure

Stenting was performed following the implantation recommendations of every device (figure 2) by implanting and performing the final control in the angiographic view with better deployment of bifurcation. Anticoagulation with sodium heparin or low-molecular weight heparin was administered according to the usual standards of every cath lab. Specific treatment of each coronary bifurcation lesion was left to the operator’s criterion. Predilatation of both branches, the provisional stenting technique or the early double stenting technique were allowed whenever, at least, 1 dedicated stent from the study was used.

Figure 2. Scheme of the positioning of the BIOSS stent for correct implantation. Central marker needs to be adjusted to the carina of bifurcation, and the ostium of the distal main branch.

Procedural success was defined as the implantation of a BIOSS LIM C stent into the bifurcation lesion with residual stenosis on visual estimate < 30% in the MB and 50% in the SB.

Clinical follow-up

Telephone or on-site follow-up was conducted at 30 days and 12 months. Patients were surveyed on adverse cardiovascular events of death, myocardial infarction, stroke, stent thrombosis, need for new revascularization or bleeding.

Angiographic analysis

Angiographic analysis was conducted independently by an imaging lab (BARCICORE-Lab, Spain) using a specific dedicated software for bifurcations (QAngio XA 7.3, The Netherlands) with which all angiographic measurements were acquired including the measurements of bifurcation angulation. Two analysts selected the images before and after implantation without the intracoronary guidewire and in the same view (< 10º of difference). Angiographic analysis was conducted in end-diastole following the lab internal protocols.

Software used allows us to measure the 3 segments of a bifurcation simultaneously (proximal MB, distal MB, and SB) and obtain individual results from all the segments including the polygon of confluence of bifurcation. All analyses were conducted taking the proximal and distal borders of the stent deployed as the reference framework. When the double stenting technique was used, the distal border of the stent implanted into the SB was used as the analysis distal limit. In case of single-stent implantation, only the proximal 5 mm of the SB were used. Figure 3 shows an example of angiographic analysis.

Figure 3. Quantitative coronary angiography showing the percentage diameter stenosis (PDS) at the proximal main vessel (1), distal main vessel (2), and 5 mm proximal to the side branch (3) before (A) and after the procedure (B). Also, it measures changes to the bifurcation angle between the distal branches. Image C shows the limits of each segment (1, 2, and 3), and the borders of the polygon of confluence.

Statistical analysis

All quantitative data are expressed as mean ± standard deviation (SD) while qualitative data are expressed as number (percentage). For the quantitative angiographic analysis (QCA) between angiographic values before and after implantation, Student t test was used for paired data (quantitative data) while the McNemar test (qualitative data) was used when appropriate. For comparison purposes between the cohorts treated with provisional stenting and double stenting, Student t test or the Mann-Whitney U test (quantitative data) were used. Also, the chi-square test or Fisher’s exact test (qualitative data) were used, when appropriate. P values ≤ .05 were considered statistically significant. Statistical analyses were conducted using the statistical software package SPSS version 20.

RESULTS

Baseline clinical data

From August 2018 through February 2021 a total of 124 patients were included in the study (figure 4). The demographic data of patients are shown on table 1. We should mention the rates of patients with diabetes [26.9% (32/124)], and acute coronary syndrome [52,8% (66/124)], 12,8% (16/124) of whom had ongoing ST-segment elevation. No significant differences were reported in the baseline clinical characteristics between the single-stent and the double stenting cohorts.

Table 1. Description of population

Figure 4. Flowchart of the EPIC03-BIOSS study. QCA, quantitative coronary angiography.

Angiographic and procedural data

Angiographic and procedural data are shown on table 2 and table 3. Most procedures were performed via radial access (120, 96.8%) and the angiography revealed the presence of 3-vessel coronary artery disease in 19 patients (15.3%). In 10 cases (8%) the target lesion was found in the left main coronary artery. Regarding complexity, in 55 cases (44.3%) the lesions treated fell into the B2/C categories of the American Heart Association/American College of Cardiology while 32 lesions (25.8%) were categorized as moderate or severe.

Table 2. Angiographic data on visual estimate

Table 3. Procedural description and follow-up

When we analyzed the differences between the cohorts treated with the provisional stenting technique and the double stenting one (table 2 and table 3) we saw that in the latter the rate of true bifurcations, proximal tortuosity, and use of imaging modalities was significantly higher.

During the procedure (table 3) the BIOSS LIM C stent was successfully implanted in 121 patients (97.6%). In the 3 cases when the stent was not implanted, the lesions showed moderate or severe calcification and proximal tortuosity. One case out of the 121 treated with stenting became complicated with SB occlusion following a dissection that could not be revascularized. In 30 (26.5%) out of the 113 cases (90.4%) where the early provisional stenting strategy was used, the POT (proximal optimization technique) was used. In 43 (38.1%) it was necessary to dilate the SB through kissing-balloon or simple dilatation (table 3). Finally, in 9 cases (7.2%) initially treated with the single-stent strategy, a second stent was needed in the SB. The double stenting strategy was initially adopted in 11 patients (9.6%). However, after SB dilatation and stent implantation into the MB the implantation of a second stent was deemed as unnecessary in 2 of them (18.2%). Therefore, 18 patients (14.5%) were eventually treated with the double stenting technique. The rates of predilatation, rotational atherectomy, use of POT, and successful implantation were similar in both cohorts.

Quantitative angiography of bifurcation

The angiographic images of 92 patients were available (table 4). Mean residual stenosis of 18% was seen in the proximal segment and nearly 0% in the MB distal segment. In the SB, the postoperative mean residual stenosis was 21% with significant residual stenosis in 5% of all patients treated with the provisional stenting technique.

Table 4. Quantitative coronary angiography of bifurcation in the entire cohort. Comparison between simple and double stent

Comparing patients treated with the single-stent technique or the double stenting technique (table 4) revealed that, at proximal segment level, expansion results are better with the double stenting technique with residual stenosis of 11% (vs 19% with the single-stent) although starting from greater baseline reference diameters (3.65 ± 0.98 vs 4.75 ± 1.42). The double stenting technique showed excellent results on the SB with a significantly greater minimum lumen diameter (2.31 ± 0.50 vs 1.62 ± 0.43, P = .01), and minimum residual stenosis of 5.9% (vs 24.3% with the provisional stenting technique) with special attention to the lack of stenosis > 50%. Results were similar in the distal segment.

Quantitative angiography of the polygon of confluence

At the polygon of confluence (table 5) results show residual stenoses of 17.15% ± 10.96% at the polygon core, and 19.21% ± 20.56% for the ostium of the SB. When the provisional stenting and double stenting techniques were compared (table 5), data from the QCA show better minimum lumen diameters for the double stenting technique in the bifurcation core and the ostium of the SB, and almost identical for the ostium of the distal segment.

Table 5. Quantitative coronary angiography. Polygon of confluence. Comparative between simple and double stent

Bifurcation angle (table 4) showed a slight change after stenting with a statistically significant reduction from 52.8 ± 18.4 º down to 47.5 ± 17.2º (P = .001). In the double stenting cohort, angulation modification was similar—in absolute terms—with a reduction of some 4º. However, this difference was not statistically significant (table 4). No significant correlation between the degree of angulation modification and the SB residual stenosis was reported (P = .86).

Clinical outcomes

Only 1 procedural complication was reported consisting of the occlusion of the SB in a patient treated with the provisional stenting technique that could not be solved. No major clinical events, cases of stent thrombosis or target lesion revascularization were reported at 30 days.

One year after implantation, 110 patients (88.7%) were contacted. A case of definitive device thrombosis (0.91%) was reported at 1-year follow-up especially in a case of double stenting treated with primary angioplasty. This case added to other 5 cases of new target lesion revascularization due to restenosis (4.54%) reveal a 12-month rate of target lesion failure of 5.45%. Two of these restenoses were found in the ostium of the SB and the remaining 3 were in the main vessel. No deaths and 3 infarctions (2.72%) were reported all of them associated with the device in relation to stent thrombosis and, in the other 2 cases, due to restenosis with minimum mobilization of troponin.

DISCUSSION

The study main findings are: a) the BIOSS LIM C dedicated stent has a high rate of success at 30 days in patients with complex coronary bifurcation lesions; b) such device is basically used with the provisional stenting strategy and is associated with a very reduced need for stenting in the SB; c) immediate angiographic outcomes show the proper behavior from the stent in the 3 bifurcation segments, as well as in the polygon of confluence where the contact surface between the stent and the artery is minimum.

Demographic data confirm that this is a non-selected population with a prevalence of risk factors, comorbidities, heart disease, and anatomical characteristics of the lesions we see in the routine clinical practice of any cath lab these days.

The device had a high rate of implantation success that was consistent with the easiness of its design being successfully implanted in > 97% of the cases. Success rate is similar to that reported in most studies with tubular stents used in bifurcations, something unreported in previous series of dedicated stents (Axxess, Frontier, and Nile studies). For example, the Frontier stent15 had a rate of restenosis of nearly 29.9%. The Nile stent16,17 was successfully used in tortuous arteries and distal segments with acceptable results with a rate of target lesion revascularization of 8.4%. However, it required distribution of angiographic stenosis focused on the carina. The self-expandable Axxess stent18 showed favorable results with a 1-year rare of cardiovascular events and restenosis of 7.7% and 6.4%, respectively. Nonetheless, it only treated the polygon of confluence and the segment immediately proximal to the ostia of the branches. Also, it was limited to certain angles and lengths needing, on many occasions, the use of additional stents.

In 14.5% of the cases, the double stenting technique was used. This is a dedicated stent in such a way that, when crossing the SB, the lack of struts in the polygon of confluence facilitates its advance without requiring previous opening. Results from the study support just how easy it is to use it with the double stenting technique. In all the cases where it was used, a second stent was successfully implanted into the SB. In 6 (66%) out of the 9 cases where the double stenting strategy was planned, the SB stent was directly implanted without dilatation. This data, though very limited, could signal a possible advantage of the BIOSS LIM C dedicated stent to facilitate access of a second stent to the SB when necessary. On the other hand, angiographic results after implantation are particularly remarkable for the double stenting technique: both the minimum lumen diameters and the residual stenosis of the proximal segments and the SB are better compared to those seen in the cohort where the provisional stenting technique was used.

If procedural data are analyzed, something that calls our attention is the SB predilatation in 47 cases where the SB was not damaged significantly. The protocol did not define the obligation to predilate the SB. According to the investigator’s criterion in each case, the observation of a lesion that did not reach a 50% stenosis was still considered to pose risk of carina displacement.

Another aspect we should mention is the strikingly low use of the POT reported in 33 cases (26%), and similarly in both cohorts. On this regard, we should mention that the device is designed with a proximal segment of a greater caliber in such a way that with simple inflation this proximal postdilatation is already incorporated. The POLBOS I19 and II14 clinical trials showed that the use of POT improved the rate of target lesion revascularization. Also, a tendency towards less late lumen loss was reported. However, POT was used at a rate of 37%, which is similar to that of our cohort.

One of the main study endpoints was to assess the potential disadvantage of design in 2 stent segments. This design provides a space for the polygon of confluence—of 0.9 mm to 1.5 mm—between both segments linked by 2 connectors. In this space, the metal-to-artery ratio is significantly lower, which may contribute to a relatively systematic underexpansion.

Results of the provisional cohort on the polygon of confluence show that mean residual stenosis is 19% at the core of the polygon and 23% in the ostium of the SB. These data suggest a certain impact in the angiographic results of this relative lack of scaffold between both segments that we believe could be the cause for a certain degree of underexpansion at the polygon of confluence.

Another study primary endpoint was to see the degree of damage in the bifurcation native angulation. Godino et al.20 analyzed changes to the bifurcation angle after angioplasty using the 1 or 2-stent technique in a cohort of 215 patients. They described a mean reduction of around 10º of the angle in the left main coronary artery, and 7º in the remaining bifurcations with the double stenting technique. However, with the provisional stenting technique no significant differences were reported. In our study, the results seen in the overall population show a statistically significant change—although not very relevant numerically—of the bifurcation angle that went from 53º to 47º. Contrary to what was reported by Godino et al.20, in our study, in the 2-stent cohort, changes to the bifurcation angle were not significant with mean reductions of 3º. However, in the provisional stenting cohort, a significant reduction of the angle was seen of 6º. In any case, we consider that this just has simple statistical significance; it seems very unlikely that a 5º variation of the bifurcation angle can be seen through visual estimate, and much less that any clinical disadvantages can occur.

Overall, the follow-up results were good and consistent with what was described in former studies.14,19 There was a significant loss of cases for the QCA since 92 cases were available only. Another study limitation is that regarding the analysis of the double stenting technique. Although the QCA data suggest good results for the double stenting technique, the study was not designed to draw comparisons between the provisional and the double stenting techniques. Also, the information collected is limited, and the number of patients treated with the double stenting technique is not enough to reach any conclusions.

An additional limitation we would like to mention is the low use of imaging modalities. Probably in coronary bifurcation lesions its systematic use can be beneficial.

In conclusion, we believe that this study conducted at several centers with different operators reveals how relatively easy it is to use the BIOSS LIM C stent to treat coronary bifurcation lesions. We’re pretty sure this is significant enough to simplify the double stenting approach where the scarce distortion overlapping the bifurcation native angulation called our attention. On the other hand, the weak spot would be the feeble metal scaffold that remains in the polygon of confluence probably due to a certain degree of underexpansion. However, the good clinical outcomes reported at 1-year follow-up are indicative that such design is not really a problem.

Limitations

This was a prospective observational study, which means that any comparisons among the routine techniques used to treat coronary bifurcation lesions can be limited. The informed consent from all the patients was not obtained. Although information from 88% of the cohort was obtained, losses to follow-up were slightly higher than they should have been.

CONCLUSIONS

The BIOSS LIM C dedicated stent works well to treat coronary bifurcation lesions. Angiographically, the stent has a space at the polygon of confluence to facilitate access to the SB. This is associated with a lower metal-to-artery ratio conditioning residual stenosis of around 20%. However, such residual stenosis does not necessarily trigger more events at 1 year and, at the end, this carinal design allows easy access to the SB in case a second stent would be needed and with excellent results. Finally, the stent-induced distortion on the angle of the carina is limited, around 5º.

FUNDING

The study was funded with a graft from Fundación EPIC, which was unconditionally funded by the LOGSA group.

AUTHORS’ CONTRIBUTIONS

As the study lead co-investigators, B. García del Blanco, and A. Pérez de Prado drafted the protocol, managed funds, directed the project, recruited the patients, wrote part of the article, and made their contributions to the overall drafting of the article. J. Gómez-Lara conducted the angiographic analysis at the core lab, the statistical analysis, drafted part of the article, and made his contributions to the article overall draft. I. In his capacity of sub-investigator, Otaegui Irurueta recruited patients by performing procedures and protocol follow-ups, entered the data required in the data curation notebook, depurated and finalized the data entered in the database, was involved in statistical analysis, drafted part of the article, participated in the article overall draft by compiling all the sections drafted from the remaining authors, and responded to the corrections requested by reviewers. M.A. Carmona Ramírez dealt with all regulatory actions needed to start the study and include the different participant centers both with the Spanish Agency of Medicines and Medical Devices and the different centers and ethics committees. As study sub-investigators, the remaining authors recruited patients, performed the procedures and follow-ups according to protocol, filled out the data curation notebook, and responded to all the questions asked.

CONFLICTS OF INTEREST

A. Pérez de Prado 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 research grants from the following research sponsors (Fundación EPIC): Abbott, Biosensors, Biotronik, Bristol-Myers-Squibb, Boston Scientific, Cardiva, iVascular, Shockwave Ltd, Terumo, Volcano Philips; also, fees for his theoretical or practical proctoring for Braun, Boston Scientific, and Terumo. The remaining authors declared no conflicts of interest whatsoever.

REFERENCES