Objective: We aimed to appraise the risk benefit-balance of the Avantgarde Carbofilm-coated stent in patients undergoing percutaneous coronary intervention (PCI) at high risk of premature dual antiplatelet therapy (DAPT) discontinuation.
Background: Carbofilm coating may increase the biocompatibility of coronary stents, but their risk-benefit balance according to diabetic status remains unclear.
Methods: Patients underwent PCI with the Avantgarde Carbofilm-coated stent and judged by the caring physician at high risk of premature DAPT interruption were retrospectively identified. Subjects were distinguished in 3 groups: non-diabetic (ND), non-insulin-dependent diabetic (NIDDM), and insulin-dependent diabetic (IDDM). Outcomes of interest were major adverse cardiac events (MACE), and their individual components.
Results: A total of 619 patients were included: 490 (79.2%) in the ND group, 95 (15.3%) in the NIDDM group, and 33 (5.3%) IDDM group. After 15±7 months, MACE occurred in 6.9% in the ND group, 17.9% in the NIDDM group, and 21.2% in the IDDM group (p<0.001), with similarly trends for death and cardiac death (all p<0.05). Multivariable analysis confirmed that those in the ND group were at lower risk than those in the NIDDM group for MACE and death (respectively p=0.020 and p=0.005).
Conclusion: The Avantgarde Carbofilm-coated stent appears associated with favorable results in patients without diabetes mellitus undergoing PCI and judged at high risk of premature DAPT withdrawal.
Keywords: Acute Coronary Syndrome; Coronary Artery Disease; PTCA; Stent
There has been momentous improvements in the medical management of patients with coronary artery disease over the last few decades, but revascularization is still required in unstable patients or those without a satisfactory response to maximal medical therapy [1-2]. Whereas surgical revascularization by means of coronary artery bypass grafting (CABG) appears clearly superior to percutaneous coronary intervention (PCI) in patients with multivessel disease, [3-4] PCI is still beneficial in those with more focal disease or at high risk for surgery.
The ideal device for PCI is still a matter of debate, as baremetal stents (BMS) are fraught with a predictable risk of restenosis and repeat intervention,  and drug-eluting stents (DES) are still considered a potential long-term hazard, especially when DES with permanent polymers with a proven risk of late thrombosis are concerned . Biocompatible coatings aimed at reducing thrombogenicity of metallic endoprosthesis have been proposed, and devices exploiting proprietary Carbofilm coatings have been recently introduced, with favorable preliminary data [7-8]. These devices could prove uniquely useful in those patients at high risk of premature discontinuation of dual antiplatelet therapy (DAPT), which may have ominous implications [9-10]. However, there is uncertainty on their risk-benefit balance, especially in unselected patients, and even more so in patients with diabetes mellitus, which are typically at higher risk of both restenosis and thrombosis [11-13].
We aimed thus to retrospectively appraise the risk-benefit balance of PCI with the Carbofilm-coated Avantgarde stent in real-world patients at high risk of premature DAPT interruption, focusing on diabetic status.
This was a multicenter retrospective observational study exploiting a prospectively-maintained administrative database for data collection and extraction (Cardioplanet, Esaote, Genoa, Italy). All patients provided written informed consent.
Patients receiving one or more Avantgarde stent during PCI,or in which implantation of this device was attempted, and who were judged at physician’s discretion as being at high risk of premature DAPT discontinuation, were retrospectively retrieved from an explicit query of our administrative database. Indications for PCI with such device, beyond the fact that patients were not deemed suitable for prolonged (>1 month) DAPT, were at operator’ discretion. Indeed, all participating centers had previously purchased the Avantgarde stent with the explicit willingness to use it selectively in patients at high risk of premature DAPT discontinuation and this device is typically reserved for patients without high risk of restenosis or those in whom compliance to dual antiplatelet therapy is uncertain. Accordingly, the prototypical subject in whom PCI with Avantgarde is considered is a subject with current or recent unstable coronary artery disease, who is a high risk of bleeding because of his or her specific baseline features, or who is likely to undergo non-cardiac surgery within a few weeks or months (e.g. because of malignancy).
Percutaneous coronary intervention was performed as per standard of care. Thus, intravascular ultrasound (IVUS), optical coherence tomography (OCT) or fractional flow reserve (FFR) were used only when clinically indicated. Similarly, no formal provision was present for pre-dilation or post-dilation, which were also at operators’ discretion.
Ancillary medical therapy, including type, intensity, and duration of dual antiplatelet therapy, were also left at the preference of caring physicians. Patients were followed as per routine practice with office visits or phone interviews at 2 weeks, 6 months, 1 year, and 2 years after the index procedure. Specifically, the 2 week time frame was chosen as this was likely the best one to capture in-hospital events or events occurring shortly after discharge, whether 30-day events may have included also events occurring several days or weeks after discharge.
Endpoints and Definitions
The main endpoint of interest for this work was the risk of major adverse cardiac events (MACE), i.e. the composite of death, myocardial infarction, or repeat target vessel revascularization (TVR). Details on individual components of MACE, plus cardiac death, target lesion revascularization (TLR), CABG, and stent thrombosis were also collected.
For the purpose of this work, we distinguished patients without diabetes (non-diabetic [ND]) from thsoe with diabetes, defined as those on oral anti-diabetic medications (non-insulin-dependent diabetic [NIDDM]) or requiring insulin for glycemic control (insulin-dependent diabetic [IDDM]). Indeed, we chose to analyze separately patients with IDDM and NIDDM as they are typically very different in terms of risk of restenosis to bare-metal stents such as the Avantagarde device, as well as in terms of risk of non-target-vessel events.
Continuous variables are reported as mean±standard deviation, and were compared with ANOVA for bivariate analysis.Categorical variables are reported as n (%), and were compared with chi-squared test for bivariate analysis. Multivariable adjusted analyses were then performed with binary logistic regression, including as covariates all those significantly associated with diabetic status at bivariate analysis, plus all those with an established epidemiologic role on clinical outcomes . Results of multivariable analysis are reported as odds ratio (OR) with 95% confidence interval. Statistical significance was set at the 2-tailed 0.05 level, and p values unadjusted for multiplicity are reported throughout. Computations were performed with SPSS 20 (IBM, Armonk, NY, USA).
A total of 619 patients were included, 490 (79.2%) in the ND group, 95 (15.3%) in the NIDDM group, and 33 (5.3%) IDDM group. Significant differences between these groups were apparent for several baseline variables, including age, prevalence of hypertension, renal failure, prior myocardial infarction, and procedure timing (all p<0.05) (Table 1). Significant differences were found also for several procedural features, including target vessel, stent diameter, and stent length (all p<0.05) (Table 2). In general, patients with diabetes were older, sicker, and had more complex coronary artery lesions.
Table 1. Baseline characteristics.
AMI=acute myocardial infarction; PCI=percutaneous coronary intervention
Table 2. Procedural features.
Significant differences in clinical outcomes were apparent already at 2 weeks at bivariate unadjusted analysis (Table 3). Specifically, patients without diabetes were at lower short-term risk of MACE, death, and cardiac death (all p<0.05). Similarly significant differences in favor of the ND group were evident at long-term (14.6±6.9 months) follow-up, with a lower risk of MACE, death, and cardiac death. Conversely, the rate of myocardial infarction, TLR, TVR, CABG or stent thrombosis was homogeneously low despite the ubiquitous high risk for premature DAPT discontinuation, and not significantly different in the groups. Notably, all cardiac deaths ischemic in etiology (either sudden or due to myocardial infarction).
Table 3. Clinical results at short- and long-term.
AMI=acute myocardial infarction; CABG=coronary artery bypass grafting;MACE=major adverse cardiac events; TLR=target lesion revascularization; TVR=target vessel revascularization
Multivariable analysis confirmed that patients in the ND group were still at lower long-term risk of MACE and death even when adjusting for potential confounders (Table 4). Conversely, the differences between the IDDM and ND groups were no longer significant after adjustment, suggesting that most of the risk associated with diabetes might lie in the NIDDM group.
Table 4. Unadjusted and multivariable-adjusted clinical results at longterm.
*reported as odds ratio (OR) with 95% confidence intervals, and simultaneously adjusting for age, gender, hypertension, dyslipidemia, smoking status, renal failure, prior AMI, prior PCI, elective procedure, left main revascularization, baseline diameter stenosis, lesion length, calcified lesion, Ellis C lesion, bifurcation lesion, stent diameter, stent length, and procedural success; MACE=major adverse cardiac events; PCI=percutaneous coronary intervention; TVR=target vessel revascularization.
This study, the largest reporting to date on the use of the Carbofilm-coated Avantgarde stent in unselected patients undergoing PCI, suggests that this device appears associated with favorable results in patients without diabetes mellitus undergoing PCI. These findings appear even more insightful given that the Avantgarde was used selectively in patients at high risk of premature DAPT discontinuation.
The management of coronary artery disease has improved substantially in recent years, thanks to developments in medical therapy and revascularization strategy. More poignantly, devices for percutaneous coronary intervention have changed substantially, and the interventionist’s armamentarium now includes drug-eluting stents with bioresorbable polymers, bioresorbable vascular scaffolds, and drug-coated balloons [15-17]. Nonetheless, bare metal stents still represent a useful tool especially in patients unsuitable for long-term DAPT and/ or at low risk of restenosis (e.g. because of focal lesions in large vessels). In addition, specific coatings can be added to bare metal stents to improve their biocompatibility, foster endothelialization, and reduce thrombogenicity. Carbofilm coating is a typical example of such coatings, as it aims at improving both short-term safety and long-term efficacy of bare metal stents.
Preliminary data on the Carbofilm-coated Avantgarde stent have already been reported, and include a pilot clinical experience , as well as a mechanistic study exploiting early optical coherence tomography to appraise strut coverage .
Briguori et al used the Avantgarde in 42 patients with undeferrable non-cardiac surgery (actually performed an average of 27 days after PCI)  Only one patient had a MACE,precisely 12 days after PCI and 3 days after vascular surgery for abdominal aortic aneurysm. Prati et al showed in 20 patients with STEMI and multivessel disease, that strut coverage 4-7 days after Avantgarde implantation in the culprit lesion was satisfactory, with only 3.9% uncovered struts, 8.0% malapposed struts, and 2.6% thrombus-laden struts . These data suggest that the Avantgarde provides satisfactory results even in patients requiring early DAPT discontinuation or with high-risk lesions with substantial thrombus burden.
Our present findings confirm these preliminary observations in much smaller studies. Specifically, we found that Avantgarde was associated with favorable results in non-diabetic patients.Results in diabetics were inferior to those without this condition, but this in keeping with prior evidence on the unfavorable short- and long-term prognosis faced by such patients when undergoing PCI, given their more diffuse atherosclerotic burden and higher risk for thrombotic complications [18-20]. The overall evidence suggests that insulin-dependent diabetics are even at higher risk of adverse events than noninsulin-dependent diabetics, as they may represent a subgroup of diabetics with poorly controlled glycemia and persisting insulin resistant despite optimal oral medication regimens [21,22]. Reports calling into question this paradigm are however available, and testify that improvements in management strategies may finally equalize the prognosis of diabetics and non-diabetics [23-24].
An important consideration to bear in mind when analyzing our results is that all patients included in our retrospective registry were deemed, at the physician’s discretion, at high risk of premature DAPT discontinuation. Accordingly, all subjects were at heightened risk of thrombotic complications. In light of these specific clinical features, the fact that only 1 stent thrombosis was adjudicated supports the hypothesis that the Carbofilm coated Avantgarde has low thrombogenicity, and may thus be safely used in such patients. The fact that diabetics had a more unfavorable prognosis than non-diabetics suggests conversely that additional efforts to improve the outcome of these patient should be pursued. These may include more potent antithrombotic therapy , optimization of stenting techniques with intravascular imaging  or reliance on other revascularization strategies (i.e. coronary artery bypass surgery) .
This work has several drawbacks, including those of retrospective observational studies including unselected real-world patients and the lack of an explicit set of inclusion and exclusion criteria such as those typical of prospective studies. Specifically, the risk of confounding due to selection, performance, and attrition and adjudication bias cannot be ruled out. Indeed, stent choice, as well as all other management decisions, was at physicians’ discretion. Accordingly, we cannot rule out that the Avantgarde stent was used selectively in very high risk diabetics in comparison to very low risk non-diabetics, with such high risk features only partially adjustable for. Thus, these results should best be viewed as hypothesis-generating, and require formal prospective testing in a pragmatic randomized clinical trial comparing patients undergoing PCI with Avantgarde versus subjects undergoing PCI with new-generation DES,(15) or bioresorbable devices .
In addition, details on type and duration of DAPT were not collected systematically and thus no formal analysis of their impact on clinical outcomes (such as MACE) could be performed in the present work. Notwithstanding these limitations, the prevalence and burden of coronary artery disease in patients with or without diabetes, and the common need for minimally invasive coronary revascularization despite the frequent occurrence of high risk features for premature dual antiplatelet therapy discontinuation mean that work may be informative to the large audience of cardiovascular specialists.
In conclusion, the Avantgarde Carbofilm-coated stent appears associated with favorable results in patients without diabetes mellitus undergoing PCI despite being deemed at high risk of premature DAPT.