Initial experience with bioresorbable vascular scaffolds for percutaneous revascularisation in patients with acute coronary syndrome

Aim: Bioresorbable vascular scaffolds (BVS) have recently been introduced to minimise the long-term complications of metallic stents in acute coronary syndrome (ACS), but their benefits have not been well analysed. Methods: The authors studied all ACS patients treated with any kind of stent at a single centre between March 2013 (when the first BVS was implanted) and June 2016. Results: The study included 951 subjects, mean age 67.9 ± 13.3 years, mean Global Registry of Acute Coronary Events (GRACE) score 148.5 ± 44.8, 75.2% men and 38.2% with an ST-segment elevation myocardial infarction. The mean number of stents implanted was 1.3 ± 1.0 and 54 subjects (5.7%) received at least 1 BVS. Drug-eluting stents were implanted in 57.3% subjects, followed by bare-metal stents (19.0%). The subjects treated with BVS were younger and had lower GRACE scores compared to the rest. Inhospital mortality was 4.8% and no subject treated with BVS died before discharge. BVStreated patients received dual antiplatelet therapy or new antiplatelet agents more frequently. During a median follow-up of 13 months, all-cause mortality was 7.8%, cardiovascular mortality was 6.1%, and at least 1 major cardiovascular event occurred in 26.4% of the subjects. Stent type did not affect prognosis. Conclusion: Coronary revascularisation using BVS in selected ACS patients is safe and effective.


INTRODUCTION
Cardiovascular disease is the leading cause of mortality in the world, and coronary heart disease is the primary contributor to that cardiovascular mortality. [1]Coronary heart disease is a progressive condition resulting from atherosclerosis that produces unstable coronary plaque, with episodes of erosion and intraluminal thrombosis that become manifest as an acute coronary syndrome (ACS), [2] ventricular fibrillation, asystole, and sudden death.[5] Bare-metal stents (BMS) were the first devices developed to improve the results of balloon angioplasty by the resolving post-balloon dissections, and preventing coronary wall recoil and constrictive remodelling. [6]Percutaneous coronary intervention (PCI) with stents was progressively adopted as the standard of care for interdiction of ACS, myocardial ischemia and infarction. [7]However, as this strategy was being adopted it became clear that stent restenosis was a frequent and potentially fatal complication in patients treated with BMS. [6]Drug-eluting stents (DES) largely overcame the restenosis problem with BMS but they were also linked to many long-term complications such as late stent-thrombosis, neo-atherosclerosis, side-branch jailing and/or preclusion of future surgical revascularisation at the same lesion. [8]oresorbable vascular scaffolds (BVS) are newly adopted device for coronary stenting and were designed to minimise the long-term complications of stents [6] by providing an on-permanent scaffold. [9]The first BVS was designed and tested in humans at the end of last century and now shows promising results in terms of feasibility and complete reabsorption. [10]everal randomised clinical trials and meta-analyses have outlined the usefulness of BVS for coronary revascularisation, [11][12][13] although the clinical benefit for ACS patients has not been well analysed. [14,15]The objective of our study was to report the safety and utility of BVS compared to other stents in a cohort of ACS patients.

Study design
In 2009, we initiated an ongoing prospective registry of all patients admitted for ACS in our institution.Several results have already been published. [5,16]For this study, we included all patients admitted with the diagnosis of ACS between January 2013 and March 2016, resulting in a cohort of 951 consecutive subjects.The patients were classified according to the stent type that was used and the cohort was divided in 4 groups: no stent, BMS, DES, or BVS.The interventional cardiologist made the treatment decisions and stent selection after considering the clinical situation.The DES implanted during the study period were all second-and thirdgeneration devices.Intravascular ultrasound and optical coherence tomography were performed in most cases where a BVS was implanted but the final decisions were made according to the treating physicians' clinical judgements.
ACS was defined by the presence of typical clinical symptoms of unstable angina or impending myocardial infarction.ACS was classified as ST-elevation myocardial infarction (STEMI) and non-STEMI according to the electrocardiographic findings.The mortality risk was assessed by the Global Registry of Acute Coronary Events (GRACE) score; according to individual scores patients with categorised in low risk < 108), intermediate risk (109-130) and high-risk (> 140). [3]All patients with STEMI received primary PCI and patients with non-STEMI were referred for PCI, if appropriate as determined by a comprehensive medical evaluation.According to previous reports, incomplete coronary revascularisation was defined as when at least one of the main coronary arteries, or a secondary branch > 1.5 mm, with significant lesions (> 70%), was treated but not fully revascularised. [17,18]sk factors, clinical antecedents, treatments, complementary tests, and main diagnosis at discharge were tabulated for all subjects by trained medical staff. [5]For the diagnosis of previous coronary artery disease, subjects needed to have a clinical diagnosis of myocardial infarction, stable or unstable angina or known prior coronary revascularisation.Previous heart failure was determined based upon their prior clinical diagnosis of heart failure.Glomerular filtration rate was estimated from serum creatinine values with the Modification of Diet in Renal Disease Study equation.Overall estimation of comorbidities was assessed by the Charlson Index, adapted for patients with coronary heart disease; [19] patients with a Charlson score ≥ 4 were considered as having high-comorbidity risks.Following current recommendations, optimal medical treatment was codified when patients received prescriptions for four medical treatments: an antiplatelet agent, a statin, a beta-adrenergic blocker, and an angiotensinconverter enzyme inhibitor or angiotensin-receptor blocker. [20,21]Statin treatment was classified as low intensity, moderate intensity, or high intensity based on the current guidelines. [22,23]Ticagrelor and prasugrel were analysed together as a group of new antiplatelet agents in place of clopidogrel.
The post-discharge follow-up of all subjects followed a well-established protocol.The end-points analysed were cardiovascular and all-cause mortality as well as time to first major cardiovascular event (MACE) (ACS, heart failure hospitalisation, fatal or non-fatal stroke, or major bleeding).Two staff members reviewed clinical records, and (in absence of hospital contact), the electronic medical history was consulted for outpatient follow-up care.All physicians in the geographic area use a unified electronic medical record (EMR) that documents every contact with the health care system, for either medical or nursing visits.If electronic medical reports were lacking, one nurse who had been trained to acquire the needed data by telephone was directed to call the subject and assess all endpoints through a follow-up conversation.All emergency calls, visits to the emergency room of the hospital, or hospital readmissions also are registered in the single informatics application.Fatalities directly related to cardiac events, such as ACS, heart failure hospitalisation or sudden death were attributed to cardiovascular causes; noncardiovascular mortality was coded when another concurrent process was thought to be the main cause of the fatality, representing mainly infections, cancer deaths or accidents.The ethics committee of the hospital approved the protocol for the study and for obtaining informed consent from the subjects.

Statistical analysis
Data were processed with IBM SPSS 22.0 and STATA 14-0 statistical packages for Mac computers.Quantitative variables are presented as mean ± standard deviation (SD) and differences were assessed by a Students t-test or Chi-squared tests.Qualitative variables are presented as percentages and differences were analysed by the analysis of variance (ANOVA) test.An analysis of interactions and collinearity between main clinical variables was performed and results were taken under consideration for further analysis.Statistical differences were accepted as significant if the P value was < 0.05.

Cox regression models performed survival analyses
once the proportional risk tests were verified.The model was adjusted by all variables that obtained P values < 0.1 in the univariate analysis or could have plausible clinical implications; results are presented as the hazard ratio (HR) and 95% confidence intervals (CI).Stent type was analysed as a categorical variable in dummy models.The model's discriminative accuracy was assessed by the Harrell's C-statistic, and its calibration was verified by the Gronnesby and Borgan test.Analysis of recurrent cardiovascular events was performed by negative binomial regression, and results are presented as incidence rate ratio (IRR) and rates/100 patients/year. [24]

RESULTS
During the study period, a total of 951 patients were enrolled as subjects.They had a mean age of 67.9 (± 13.4) years, 75.2% were male, 38.2% were diagnosed with STEMI, and their mean GRACE score was 148.5 (± 44.8).The revascularisation strategies included DES (57.3%) and BMS (19.0%); 54 (5.7%) received at least 1 BVS [Table 1].In 171 subjects (18.0%), no stents were placed after the angiography and diagnostic studies were done.BVS-treated patients were the younger, and had the highest percentage of current smokers but the lowest prevalence of known heart failure, atrial fibrillation, or stroke.No differences were observed in The details of the revascularisation outcomes are presented in Table 2.The radial artery approach was utilised in most cases and in all BVS-treated subjects.The mean number of coronary lesions was 1.6 (± 0.9) which was significantly higher in BVS-treated subjects; the mean number of stents per subject was 1.3 (± 1.0) and BVS-treated cases also had the highest rate.As a consequence, the rate of complete revascularisation was the highest in patients treated with BVS.No BVS implantations were attempted in the left main coronary artery.
In-hospital mortality was 4.8% (46 patients out of 951) and no patient treated with BVS died prior to discharge.
The highest in-hospital mortality rate was observed in patients who were not (or could not be) stented (8.8%), followed by those treated with BMS (7.2%) and DES (3.3%).As shown in Table 3, BVS-treated patients received the highest rate of dual antiplatelet treatment and 70% received a new antiplatelet drug (prasugrel or ticagrelor).Generally, patients treated with DES or BVS received more intensive medical treatments.

DISCUSSION
This single-centre experience with BVS supports their safety and effectiveness for revascularisation during ACS.BVS-treated patients had a lower risk profile despite the fact that they had more coronary lesions and were treated more aggressively, as they received a higher number of stents, more complete revascularisation, and more intensive antiplatelet and statin regimens.
Decision-making for stent use is influenced by many factors related to patient characteristics, coronary lesions, and other risks.DES were introduced to better control the rate of restenosis that occurs in patients treated with BMS.The superiority of DES has been largely demonstrated, [25] although mortality benefit was only clearly outlined with the later generation of DES. [8]VS were conceived to avoid long-term complications of the metal structures by providing temporary structural integrity before being resorbed completely within the vessel wall. [9,26]The long-term incidence of cardiovascular events related to DES-treated vessels is around 2% to 3% per year for at least 5 years [27,28] and the contribution of permanent metallic devices in lumen target lesions has a relevant contribution. [29]he Absorb ® BVS is a 150 μm thick bioresorbable polyl-lactide scaffold with a conformal bioresorbable coating (with a total thickness of 7 μm) that elutes everolimus. [30]Angiographic follow-up of BVS has clearly demonstrated the complete reabsorption of the scaffolds [9] and significant lumen gain. [31]The initial experience of BVS in our institution supports the clinical safety and efficacy of BVS in ACS patients with a significant post-discharge follow-up.
As with every innovation, concerns arose when data from randomised clinical trials and large numbers of patients treated with BVS became available.The pooled analysis of first studies showed the equivalence at 1-year of BVS compared with everolimus-DES although a non-significant trend (HR: 2.09, 95% CI: 0.92 to 4.75, P = 0.08) to higher late stent-thrombosis was already outlined. [30]In the ABSORB-II trial, BVS had similar rates of repeat revascularisation at 1 year of follow-up, despite inferior mid-term angiographic performance, in comparison with everolimus-eluting metallic stents. [11]onetheless, patients treated with a BVS had a threefold increased risk of subacute stent thrombosis.These results have been verified in subsequent meta-analyses. [12,13]Stent thrombosis is a challenging clinical problem and related to many factors, including the different antiplatelet regimens, discontinuation of dual antiplatelet therapy, procoagulant states, stent malposition, polymer content, and many others. [32,33]ajor cardiovascular events and mortality rates in the BVS-treated patients in our cohort were similar or even lower than in patients treated with other stents.Moreover, patients treated with BVS in our study had a higher number of vessels with significant lesions, and subsequently received more stents.This suggests that in well-selected ACS patients under intensive platelet treatment, the use of BVS can be a reasonable strategy of percutaneous revascularisation as has been proposed by other reports [14,15] and meta-analyses. [12,13]S-treated patients in our cohort received dual antiplatelet therapy with clopidogrel or the newer antiplatelet drugs, prasugrel or ticagrelor, more frequently than the rest of patients.This could be influenced by many factors, such as the percutaneous coronary intervention characteristics, the number of stents, the younger age or absence of concomitant anticoagulation, but reflects that patients that received a BVS were treated more intensively.The effect of BVS for late adverse events could be especially important for young patients with ACS because they have an impaired vascular healing [34] and new antiplatelet agents could have similar effects. [35]The efficacy of BVS has been tested in only two randomised clinical trials [36,37] and most reports come from observational studies. [14,15,38]The primary endpoint in many of these studies was not cardiovascular mortality, and only the feasibility of the BVS implantation was assessed.We conducted an observational prospective study with the aim of providing mortality rates in a real-world cohort of patients.Moreover, we examined the cumulative incidence of recurrent events, which has been proposed as the best approach to monitor the actual course and prognosis of coronary heart disease. [24]ronary heart disease is a chronic inflammatory disease and it develops as a result of a progressive process. [39]Despite optimal medical treatment and revascularisation, recurrent events are common. [21,40]he most frequent statistical analyses for follow-up events are based in time-to-events, and therefore, patients are excluded from further analysis once they experience such events.Nonetheless, analysis of recurrent events has been proposed as a more accurate way to assess the actual life-long course and prognosis. [41]Optimal medical treatment has been demonstrated to provide benefit in patients with stable [42] and unstable [21] coronary heart disease regardless of revascularisation.Nonetheless, revascularisation has a much more critical role in ACS patients, and it has been identified as one of the major factors related to long-term ACS survival.Our analysis identified a negative association between revascularisation and recurrent events that provides additional support to its key role in the treatment of ACS patients.STEMI represents less than 40% of ACS and its emergent treatment requires many resources because acute phase mortality is much higher than in non-STEMI. [1]onetheless, the long-term mortality and medical costs are equivalent for both types of ACS.Non-STEMI patients comprise a heterogeneous group, and these patients are usually older and have more comorbidities.This can yield challenging decision-making with regard to revascularisation and medical treatment.As a result, revascularisation rates are significantly lower in non-STEMI compared to STEMI patients, highlighting the unmet needs for validated and organised systems of care for non-STEMI patients. [43]Our results highlight the role of revascularisation for ACS and the recurrence of ischemic events.Stent type was not associated with any of these events although our follow-up suggested that revascularisation has higher impact than the other techniques used.
Our study has several limitations.The first is that a very limited number of patients were treated with BVS.Second, there may have been unmeasured confounders, details about the physician's decisionmaking, or patient factors that are not captured by the registry but account for the treatment differences observed.Furthermore, our analysis uses observational, non-randomised data, and thus, associations between various treatments and outcomes may be confounded by unmeasured variables.Finally, long-term outcomes could be modified by factors that were not captured by the follow-up protocol that was employed at our outpatient centres.Nonetheless, since clinical features and event rates are similar to previous reports, [4,[36][37][38] we believe that our results should be representative of daily clinical practice and that our conclusions are probably valid.
In conclusion, our study supports the hypothesis that coronary revascularisation using BVS is safe and effective in selected ACS patients.Percutaneous coronary revascularisation with stents is the cornerstone of ACS treatment but there are long-term complications of metallic devices that can be relevant.The use of BVS is a novel strategy with promising results, and there is an expectation that they will prove useful in ACS selected patients.The long term prognosis of BVS-treated patients remains unknown, but given the growing number of patients being treated with these devices, further evidence regarding their net clinical benefit will likely emerge with additional time.

Table 1 : Clinical features of the cohort according to the stent type
For comparisons between BVS and the rest; *for comparisons between BVS or DES and the rest.BMS: bare metal stent; DES: drugeluting stents; BVS: bioresorvable vascular scaffold; HF: heart failure; CHD: coronary heart disease; PCI: percutaneous coronary intervention; CABG: coronary arterial bypass graft; COPD: chronic obstructive pulmonary disease; STEMI: ST-elevation myocardial infarction; GFR: glomerular filtration rate; GRACE: Global Registry of Acute Coronary Events; LVEF: left ventricle ejection fraction the prevalence of diabetes or dyslipidaemia according to the stent type received.BMS-treated patients had the highest prevalence of comorbidities, reflected by the highest mean Charlson index. #

Table 3 : Medical treatments recommended at discharge according to the stent received within hospital revascularization
BMS: bare metal stent; DES: drug-eluting stents; BVS: bioresorvable vascular scaffold; DAPT: double antiplatelet treatment; ARB: angiotensin receptor blocker; ACEI: angiotensin-converter enzyme inhibitors BMS (9.5%), DES (4.0%), and BVS (3.7%) (log-rank test, P < 0.01).No differences were observed between DES and BVS in mortality rates.No differences were observed in the incidence of MACE according to stent type, although a tendency to lower incidence of time to first MACE was noted for BVS: no stent (28.8%),BMS (31.5%),DES (25.5%), and BVS (16.7%) (P = 0.16).The highest rate of MACEs/year/100 patients was recorded in patients that received no stents, followed by BMS, DES, and BVS [Figure2].Multivariate analysis did not find any associations of mortality or cumulative MACEs with stent type [Table4].Diabetes, age > 75 years and GRACE score > 140 were associated with higher cardiovascular and all-cause mortality, whereas predictors of recurrent MACEs included age > 75 years, diabetes, and previous cardiovascular disease; revascularisation was negatively associated with recurrent events.