A comparison of the main outcomes from BP-BES and DP-DES at five years of follow-up: A systematic review and meta-analysis

Biodegradable polymer biolimus-eluting stents (BP-BES) are third-generation drug-eluting stents (DES) composed of biodegradable polymers that may improve prognosis after percutaneous coronary intervention (PCI). After five years of follow-up, BP-BES showed conflicting results compared to durable polymer drug-eluting stents (DP-DES). We performed a meta-analysis of the outcomes of studies on BP-BES and DP-DES after percutaneous coronary intervention (PCI) at five years of follow-up. Eligible studies were retrieved from PubMed, Embase and the Cochrane Library and reported the results of all-cause mortality, myocardial infarction (MI), target lesion revascularization (TLR), target vessel revascularization (TVR) and stent thrombosis (ST) at five years of follow-up. Five studies of a total of 4687 patients were included in the meta-analysis. At five years of follow-up, BP-BES was associated with lower rates of major adverse cardiac events (MACE) (OR = 0.83, 95%CI = [0.71, 0.97]), TLR (OR = 0.77, 95%CI = [0.62, 0.96]) and ST (OR = 0.60, 95%CI = [0.43 to 0.84]), whereas no significant differences in mortality, MI, or TVR rates were detected. Our results demonstrated that at five years of follow-up, BP-BES can significantly reduce the risk of MACE, TLR and ST, which indicate that safety and efficacy were increased after PCI.


Results
Characteristics of the Included Studies. One-thousand-six hundred-fifty-two articles were obtained by online and manual searches. After removing duplicates and screening titles and abstracts, six independent trials that contained data for BP-BES versus DP-DES were included. The LEADERS (Limus Eluted From A Durable Versus ERodable Stent Coating) trial had four and five years of follow-up, and we obtained the latest data 13,18 . Finally, five studies 13,16,17,19,20 were selected that included 4687 patients (2002 randomized to BP-BES and 2685 to DP-DES). (Fig. 1) (as seen in the flow chart).
All of the trials included five years of clinical follow-up data. The incidences of clinical outcomes at five years of follow-up are shown in Table 2. Dual antiplatelet therapy was administered to all patients for at least six or twelve months after discharge. Aspirin and clopidogrel were used in the majority of patients 13,16,17,19 ; however, prasugrel or ticagrelor were also used 20 . Diabetes prevalence among the included trials ranged from 15.7-65.4%. The number of PCI patients with acute coronary syndrome ranged from 6.7%-100.0%.
The pooled ORs and 95%CIs of the comparisons of the main outcomes between DP-BES and BP-DES are shown in Table 3. Three outcomes were significantly different: MACE, TLR and ST; the results of death, MI and TVR were not significantly different. Figure 2 As seen in Figure 3, 366 patients (7.8%) underwent repeat revascularization of the target lesion. In the 1st DP-DES group, the use of BP-BES significantly reduced the risk of TLR compared to 1st DP-DES ( Definite/probable ST was observed in a total of 155 patients (3.3%), as presented in Fig. 4. In the 1st DP-DES group, the use of BP-BES significantly reduced the risk of ST compared to 1st DP-DES (2.9% versus 5.3%; OR       Table 2. Clinical outcomes at five years of follow-up. *The data from five years of follow-up were unpublished, so we included the data from four years of follow-up BP-BES = biodegradable polymer biolimus-eluting stents, DP-DES = durable polymer drug-eluting stents, MACE = major adverse cardiac events, MI = myocardial infarction, PCI = percutaneous coronary intervention, TLR = target lesion revascularization, TVR = target vessel revascularization, ST = stent thrombosis, N/A = not applicable.  Table 3. Pooled ORs and 95% CIs of the main outcomes between DP-BES and BP-DES. OR = odd ratio; CI = confidence interval, BP-BES = biodegradable polymer biolimus-eluting stents, DP-DES = durable polymer drug-eluting stents, MACE = major adverse cardiac events, MI = myocardial infarction, TLR = target lesion revascularization, TVR = target vessel revascularization, ST = stent thrombosis. a P value for between-study heterogeneity based on the Q test. Significant results are presented in bold.

Sensitivity analysis.
A sensitivity analysis was performed by sequentially omitting 1 individual at a time to reflect the influence of each study on the overall meta-analysis. No heterogeneity was observed in the polymorphism (Fig. 8); thus, the results of our meta-analysis were stable.  The difference between outcomes may attributed to the different time point of follow-up. The potential clinical advantage of the BP-BES might be expected to emerge once the biodegradable polymer has dissolved, and this may have occurred 9 months after implantation 14,15 . Thus, it was not surprising that BP-BES did not reduce the risk of MACE at one or even three years of follow-up. Moreover, only three trials were included at three  years of follow-up; more trials with larger populations of patients are needed to support the conclusion. Five trials were included in our study; four 13,16,19,20 of them demonstrated that BP-BES were associated with a lower risk of MACE than DP-DES. However, of these four trials, the results of three of the trials were not statistically significant (Serruys et 20 ), but all of them showed that BP-BES were associated with a lower risk of ST than DP-DES, so we archived an inspiring result. Incomplete endothelialization and the inflammatory response caused by the persistence of a  durable polymer play important roles in very late ST 6,10 . Drug-eluting stents gradually release drugs from polymer coatings that are applied to the stent surface, which prolongs the time of completely endothelialization. Three to four months are required for complete endothelialization with BMS 28,29 , whereas with DES, more time is required 6 . At the same time, the level of endothelial coverage in BP-BES was comparable to that of BMS at four weeks, with no significant increase in inflammatory reactions up to 15 months 30 . Moreover, compared with DP-DES, BP-BES contains a biodegradable polymer that gradually dissolves into water and carbon dioxide, which are associated with a lower risk of inflammatory responses in animal studies 31 . Thus, these observations may explain why the BP-BES may be associated with a lower risk of very late ST and a better long-term outcome in our meta-analysis.
In However, a previous meta-analyses showed no difference between BP-BES and DP-DES at one and three years of follow-up 22,25,27,32 . The major reasons for this difference were as follows: first, the data in our included studies was collected at five years, which was longer than the above studies. Second, PES represented a weak competitor in comparison with SES and EES [33][34][35] . The number of PES in our meta-analysis was 2.7% (125/4687), which is higher than in the above studies ( 25 ), which may contribute to the better result.
Based on our study we found no differences in mortality, MI, or TVR between BP-BES and DP-DES, but the insufficient sample size may contribute to this discrepancy. However, based on the current studies, we found that BP-BES could reduce the risk of MACE, ST and TLR compared to DP-DES. These results had statistical significance and no heterogeneity was found. Many elements may have resulted in this discrepancy, including the lack of an adequate sample size, differences between stents, experience of operators, presence of complications after PCI, seriousness of lesion, and so on.
BP-BES also had some disadvantages. In this study, after implantation of the BP-BES, 18.8% of patients presented with MACE, 7.9% of patients underwent TLR and 2.6% of patients had ST, which compromised the prognosis after PCI. Furthermore, after implantation, the biodegradable polymer gradually dissolved into water and carbon dioxide, alleviating self-perpetuating inflammation and late stent thrombosis, which in turn necessitates prolonged dual antiplatelet therapy that increases the risk of long-term bleeding events after PCI 36 .
There were several limitations in this meta-analysis. First, only English language articles were included in our study, which may bias the results. Second, patient heterogeneity and confounding factors might have affected the analysis. Third, significant heterogeneity was detected in some pooled analyses, which may have affected the meta-analysis results, even though we adopted the random effects model or introduced sensitivity analysis. Fourth, the number of included studies was relatively small, and the results should be interpreted with caution; further studies are needed to confirm these results. In conclusion, BP-BES can significantly reduce the risk of MACE, TLR and ST compared with DP-DES at five years of follow-up, which indicates that BP-BES are associated with a better safety and efficiency after PCI in the long term.

Materials and Methods
Identification of Studies. PubMed, the Cochrane Library and Embase databases were thoroughly searched in September 2016 by the first two investigators to identify potential studies of BP-BES and DP-DES. The terms "Biolimus", "Nobori", "Biomatrix", "BioFreedom", and "stent" were used. Missing data (the data that we failed to identify during the electronic search) were obtained by reviewing the citations of review articles and all eligible studies.
Inclusion and Exclusion Ceriteria. Citations were screened at the title and abstract level and retrieved as full reports. The inclusion criteria were: (1) comparison of BP-BES vs DP-DES; (2) studies reporting at least one of the following outcomes: MACE, all/cardiac death, MI, TVR, TLR, ST; and (3) clinical follow-up at five years. When more than one report of the same study was retrieved, the one with the longest follow-up was included. The exclusion criteria: (1) a duplication of previous publications; (2) a comment, review or editorial; and (3) a study without data. The studies were independently selected by two investigators, according to the inclusion and exclusion criteria by screening the title, abstract and full-text. Any dispute was resolved by discussion.
Data Extraction. From each study, the following data were independently extracted by the first two investigators using a standardized form: first author's last name, year of publication, journal, BP-BES, DP-DES, sample size, age, gender, patients with diabetes, patients with ACS, left ventricular ejection fraction (LVEF), multi-vessel disease, SYNTAS scores, numbers of stents used per patient, total stent length, and mm per patient. For data from multiple treatment groups, the approach recommended in the Cochrane handbook was adopted to avoid a unitof -error analysis that may result from entering several comparisons into one meta-analysis, which could lead to "double-counts" of patients based on the same study. Disagreements were resolved through discussion.

Statisticals analysis. RevMan version 5.3 (The Nordic Cochrane Centre, The Cochrane Collaboration,
Copenhagen, Denmark) was used for statistical analyses. We calculated the odds risk (OR) and its 95%CI (confidence interval) for the five outcomes as binary data. Heterogeneity was evaluated by the magnitude of the Chi 2 , corresponding P value and I 2 statistic. When the I 2 value was above 50%, a random effects model based on the Mantel-Haenszel (MH) or inverse variance (IV) statistical approach was selected to combine the data. If the I 2 value was below 50%, a fixed effects model based on the MH or IV statistical approach was selected, and a sensitivity analysis was conducted to detect the robustness of the result.