Introduction

Drug-eluting stents (DESs) have been shown to reduce long-term restenosis rates as compared to bare metal stents (BMSs) [1] and have become the preferred therapy in patients undergoing percutaneous coronary intervention (PCI). The presence of an anti-proliferative agent on the DES impairs the endothelialization process. In fact, angioscopic studies have shown incomplete endothelialization 6–12 months post DES implantation [2–4], and post-mortem studies conducted 40 months after implantation have confirmed poor endothelialization in 45 % of cases [5]. The delayed endothelialization is believed to be a risk factor for stent thrombosis (ST) which, though rare, can be a fatal event. Several large trials have demonstrated the effectiveness of dual antiplatelet therapy (DAPT) to reduce stent thrombosis and reduce future cardiovascular events in patients with acute coronary syndromes (ACS) as well as post elective PCI [6–9]. DAPT also prevents thrombosis from plaque rupture in native coronary lesions that were not stented [7, 9]. Hence, the initiation of DAPT (the combination of aspirin and a P2Y12 inhibitor) is recommended for all patients with ACS, regardless of stenting, and post PCI, regardless of their presentation, i.e., ACS or stable angina (SA).

Historically, the initial recommendations for first-generation DES (sirolimus-eluting and paclitaxel-eluting stents) were for 3 and 6 months of DAPT, respectively. Several subsequent observational studies however suggested a marked increase in major adverse cardiovascular events (MACE) upon the discontinuation of DAPT after 6 months from implantation of a first-generation DES [10–13]. The results of these studies led to the current recommendation of the American College of Cardiology/American Heart Association [14] for patients to receive at least 12 months of DAPT following implantation of a DES. The European Society of Cardiology, in contrast [15], recommends 6 to 12 months of DAPT post DES implantation. This difference highlights the challenge in determining the optimal duration of DAPT as we know that prolonged DAPT is also associated with an increased bleeding risk, and these bleeding complications have been reported to increase the mortality of patients receiving longer DAPT by two to fourfold [16]. Prolonged DAPT can also impact a patient’s quality of life by delaying elective surgical, dental, or endoscopic procedures [16]. This review will summarize the most recent data to help guide treating physicians make informed, patient-specific recommendations for optimal DAPT duration after a DES is placed.

Basis of current recommendations of 12 months of DAPT

The current recommendations by the ACC/AHA are based on several large observational studies demonstrating that less than 12 months of DAPT post DES is associated with worse outcomes. In an observational study of 4666 patients (all comers) from the Duke registry [10], adjusted rates of death or MI at 24 months were significantly lower when patients received 12 months of ASA plus clopidogrel compared to 6 months (3.1 vs. 7.2 %, p = 0.02). In contrast, patients who received BMSs had a similar long-term mortality and rates of death/MI irrespective of the duration of DAPT at 6, 12, and 24 months [11]. In the BASKET-LATE trial [12], designed to define the incidence of cardiac death or myocardial infarction and late ST (defined as occurring between 30 days and 12 months post-stent placement) in patients (42.2 % stable angina, 57.7 % ACS) treated with a DES (first generation) (n = 545) vs. BMS (n = 201), patients who stopped DAPT between 7 and 18 months had more thrombotic events with a DES (4.9 % event rate) than those who received a BMS (1.3 %). Furthermore, the discontinuation of DAPT prior to 12 months in the Dutch stent thrombosis registry [13] was a significant predictor of ST (hazard ratio 5.9, 95 % CI 1.7–19.8).

Observational data from the PREMIER registry [17] demonstrated a marked increase in cardiovascular mortality (7.5 %) in patients who stopped their DAPT within 1 month of DES placement. Mortality was also shown to be greater in patients who received 6 vs. 12 months of DAPT (5.3 vs. 2.8 %, p = 0.012) in the Melbourne registry (n = 2980 patients; 38.5 % SA and 61.5 % ACS) [10]. Finally, in patients presenting with ACS, which is a risk factor for ST, data from Sweden showed that less than 6 months of DAPT vs. greater than 6 months of DAPT significantly increased the risk of death, stroke, or re-infarction (hazard ratio (HR) 0.75, 95 % CI 0.59–0.95) [18]. Based on these studies, the current recommendations were made. Importantly, these registry studies involved a generation of DES (first) that are no longer clinically available and have been shown to be inferior to the current second-generation DES currently used [19]. These studies also were not randomized clinical trials and, therefore, subject to biases and confounding. Finally, many of the registries combined patients presenting with both ACS and stable angina and, thus, have both high- and low-risk patients which can influence the data.

Data for greater than 12 months of DAPT post DES

Secondary to heightened and intense public scrutiny of DESs and the risk of ST and related morbidity and mortality, some cardiologists began prescribing lifelong DAPT in the late 2000s for their patients post DES implantation even in the absence of supporting data. In an effort to better assess whether longer DAPT is indeed beneficial, the ZEST-LATE/REAL-LATE [20] studies were developed. Data were combined from these two studies to create an open-label, non-placebo-controlled study which randomized event-free patients at 12 months post DES to receive DAPT for another 12 months or aspirin alone. At 24 months, there was no difference in the composite primary endpoint of cardiac death, MI, or the risk of ST. This negative study involved different types of DESs and had an event rate of cardiac death/MI that was lower than expected (1.8 vs. 1.2 %, p = 0.17) undermining its power and applicability.

Additional data negating the benefit of greater than 12 months of DAPT in patients with DESs was provided in the PROlonging Dual-antiplatelet treatment after Grading stent-induced Intimal hyperplasia study (PRODIGY) trial [19] (Table 1), which compared four different types of DESs (first and second generation) in 2013 patients randomized to 6 vs. 24 months of DAPT. There was no difference in the primary endpoint (death from any cause, MI, or CVA) with 24 months of DAPT compared to 6 months (10.0 vs. 10.1 %, p = 0.91). Furthermore, the shorter DAPT group had a significantly lower bleeding rate (HR 0.67, 95 % CI 0.47–0.95, p = 0.026) [21]. Next, the DES-LATE trial [22] was designed to address whether an additional 24 months of DAPT in patients who completed 12 months of DAPT post DES without complications would reduce a composite endpoint of death from cardiac causes, myocardial infarction, or stroke compared to aspirin monotherapy. This prospective study was an open label and randomized 5045 patients whose initial presentation included both stable angina (38 %) as well as ACS (39 % UA, 11 % NSTEMI, 12 % STEMI). At the end of the study period, the primary endpoint was no different between the 12- and 36-month groups (2.4 vs. 2.6 %) with a non-significant decrease in bleeding in the 12 month DAPT group (hazard ratio 0.71, 95 % confidence interval 0.42–1.20, p = 0.20). With no difference between the two groups in terms of death from any cause, myocardial infarction, stent thrombosis, or stroke, this study would suggest that extending DAPT to 36 months provides no additional benefit and may increase the risk of bleeding.

Table 1 Major Trial highlights
Full size table

The limitations of the DES-LATE trial included the open-label nature, the low event rate, and the fact that it included only patients who were event free at 1 year and therefore were likely more stable. In order to overcome these limitations, the DAPT trial was designed. This large multi-center international double-blinded RCT was designed to examine the impact of long-term DAPT (30 months) vs. 12 months on the prevention of adverse cardiovascular events as well as define the magnitude of bleeding risk associated with prolonged DAPT. This study was powered for the primary endpoints of stent thrombosis and major adverse cardiovascular and cerebrovascular events (MACCE) and a composite of death, non-fatal myocardial infarction, or stroke [23••]. Treatment with either clopidogrel or prasugrel was permitted. In November 2014, the DAPT trial results were released at the American Heart Association Scientific Sessions and concomitantly published [24••]. In summary, 9961 patients (initial presentation 42 % ACS, 38 % stable angina, and ∼20 % other indication) were randomly assigned after 12 months of DAPT post DES to aspirin plus a thienopyridine or aspirin plus a placebo for an 18 additional months. Of the initially screened population, 23 % were excluded for a number of reasons including having had ischemic events within the first 12 months (n = 2638 patients [death, MI, CVA, ST, revascularization, or bleeding]). By excluding these patients, the study excluded many patients at the highest risk of the primary endpoint, i.e., patients wherein prolonged DAPT might have provided the maximal benefit.

Over 4700 patients in both arms completed follow-up to 30 months. The DAPT study showed that compared with subjects treated with DAPT for 12 months, subjects randomized to 30 months of treatment had a significantly lower cumulative incidence of stent thrombosis (0.4 vs. 1.4 %, hazard ratio 0.29, p < 0.001) and MACCE (4.3 vs. 5.9 %, hazard ratio 0.71, p < 0.001). This result was primarily driven by a reduction in myocardial infarction (2.1 vs. 4.1 %, hazard ratio 0.47, p < 0.001). Highlighting that DAPT reduces both ST and non-stent-related thrombosis; non-stent thrombosis-related myocardial infarction comprised 55 % of the treatment benefit (1.8 vs. 2.9 %, hazard ratio 0.59, p < 0.001). There was no difference in the incidence of stroke between the two treatment arms (0.8 vs. 0.9 %, p = 0.32).

The negative impact of prolonged DAPT was highlighted in the primary safety endpoint of this trial, namely bleeding. Prolonged DAPT was associated with increased moderate to severe bleeding (2.5 vs. 1.6 %, HR 1.61, 95 % CI 1.2–2.6, p = 0.001), though there was no difference in GUSTO severe bleeding or BARC defined fatal bleeding (type 5 bleeding) between the two groups. More concerning was an increased risk of death from any cause in the prolonged DAPT group compared to placebo (2.0 vs. 1.5 %, HR = 1.36, 95 % CI 1.00–1.85, p = 0.05). The authors note that this may be related to an imbalance in cancer prevalence and cancer-related mortality. When the 22 patients with malignancy-associated death were removed from the analysis, there was no difference in overall mortality between the two groups. How this large and influential study will impact the guidelines is unknown. The editorial accompanying this publication suggested dividing DAPT post DES into two distinct periods, i.e., a period of “mandatory” DAPT and a period wherein DAPT may be “possibly beneficial” [25•].

Approximately 20 % of the DAPT DES cohort came from the Taxus Liberté Post Approval Study (TL-PAS) which was initially designed as a post market surveillance study (n = 4199) [26]. Patients who were free of events after 12 months of open-label prasugrel (n = 3494) were then eligible to enter into the DAPT trial and received either placebo or prasugrel for another 18 months. Of this original cohort, 2191 patients were randomized and 97 % completed the 30-month study. This is the single largest study involving the use of prolonged prasugrel therapy and demonstrated that 30 months of prasugrel and aspirin reduced the composite endpoint (death, MI, or stroke) by a significant amount (3.7 vs. 8.8 %, HR 0.407, p < 0.001) with most of the reduction being recurrent MI. With the enhanced efficacy of prasugrel, a greater than 90 % reduction in ARC-defined definite or probable ST (0.2 vs. 2.9 %, HR 0.063, p < 0.001) was also shown with a longer DAPT duration. There was a trend toward GUSTO moderate to severe bleeding with the prolonged therapy (2.4 vs. 1.7 %, HR 1.438, p = 0.234) with no difference in severe bleeds.

Both in the DAPT trial and within the TL-PAS sub-study, there was noted to be an acute rise in ischemic events upon cessation of DAPT (both in the 12- and 30-month groups) independent of the generation of thienopyridine and that most of these events were not stent thrombosis. In an effort to decrease this rise in events, the concept of a tapered withdrawal of DAPT has been tested. ISAR-CAUTION (NCT00640679) [27••] attempted to answer whether abrupt cessation or a progressive downgraded dosing would be better, i.e., have less ST or ischemic events, in patients in whom DAPT cessation was planned after 12 months. These patients were randomized in a double-blinded fashion to gradual discontinuation over 4 weeks or abrupt cessation and were followed for 3 months for a composite of cardiac death, MI, stroke, ST, major bleeding, or re-hospitalization. Though target enrollment was 3000 patients, the study was terminated due to slow enrollment, and only 782 patients were randomized. The tapering protocol involved a 4-week period wherein patients went from daily thienopyridine dosing to every other day for 1 week followed by every 2 days for a week followed by every 3 days for a week and finally every 4 days for a week. This tapering approach was not superior to abrupt cessation, though this study was clearly underpowered due to the low event rate and the number of subjects. Given the fact that there is a clear increase in the number of ischemic events with DAPT cessation, future studies examining how to mitigate this risk are required.

Data for a shorter duration of DAPT (<12 months)

The current European Heart Society guidelines recommend 6–12 months of DAPT following DES placement. This shorter period is based on several randomized trials that have shown the safety and efficacy of a shorter period of DAPT in select patients. The PRODIGY trial [21] included bare metal, first- and second-generation DESs, and randomized 2000 patients to 6 or 24 months of DAPT. These patients would be considered high risk for ST, and future events given the cohort comprised 74.4 % of ACS patients. Despite this, there was no difference in the primary endpoint of death from any cause, MI, or CVA between the two groups (10.0 vs. 10.1 %, p = 0.91), and there was a higher incidence of bleeding in the 24-month cohort.

The EXCELLENT trial (Efficacy of Xience/Promus Versus Cypher to Reduce Late Loss After Stenting) was an open-label trial comparing efficacy of Xience/Promus to Cypher stents, but also randomized patients to 6 vs. 12 months of DAPT [28]. For the 1443 patients (51.1 % ACS, 48.4 % SA) enrolled, there was no difference between the 6- and 12-month groups for the endpoint of primary target vessel failure (composite of cardiac death, MI, or ischemia-driven target vessel revascularization). Stent thrombosis tended to occur more frequently in the 6-month DAPT group than in the 12-month group (0.9 vs. 0.1 %, hazard ratio 6.02, 95 % CI 0.72–49.96, p = 0.10). Despite this trend, the risk of death or myocardial infarction did not differ in the two groups (2.4 vs. 1.9 %, hazard ratio 1.21, 95 % CI 0.60–2.47, p = 0.58). Additionally, there was a higher rate of target vessel failure (TVF) among diabetic patients in the 6-month group (HR 3.16, 95 % CI 1.42–7.03, p = 0.005) highlighting the higher risk of these patients.

The issue of stent specificity and duration of DAPT was highlighted in the RESET (REal Safety and Efficacy of 3-month DAPT following Endeavor zotarolimus-eluting stent implantation) trial [29]. This unique study enrolled 2117 patients (53.3 % ACS, 44.7 % SA) and randomized patients to receive either Endeavor zotarolimus-eluting stents (E-ZESs) plus 3 months of DAPT or any other drug-eluting stent plus 12 months of DAPT. At 12 months, the primary composite of cardiac death, MI, ST, target vessel revascularization (TVR), or bleeding was no different between the two groups (4.7 vs. 4.7 %, p < 0.001 for non-inferiority). Interestingly, the ST rate was no different between a shorter DAPT group with E-ZES stents vs. any other DES with prolonged DAPT (0.2 vs. 0.3 %, p = 0.65). These results are consistent with data from the PROTECT study (NCT00476957) [30] which also showed a low ST rate with E-ZESs. These studies have raised the question of whether the safety of a shorter DAPT might be stent specific.

Stent specificity and its association with low ST rates were further raised in the Optimized Duration of Clopidogrel Therapy Following Treatment with the Zotarolimus-Eluting Stent in Real-World Clinical Practice (OPTIMIZE, NCT01113372) trial [31]. This non-inferiority study compared 3 vs. 12 months of DAPT in 3119 patients undergoing PCI with a zotarolimus-eluting stent. These patients had either stable angina or a history of “very low-risk ACS” which was defined as unstable angina or recent (but not acute) myocardial infarction (<30 days). There was a formal recommendation not to enroll patients with elevated biomarkers. The primary endpoint (all-cause death, MI, CVA, or major bleeding) occurred in only 6.0 % of patients receiving 3 months of therapy vs. 5.8 % of patients receiving 12 months (risk difference 0.17, 95 % CI −1.52 to 1.86). MACE rates at 1 year did not differ between the groups (8.3 % in the 3 month vs. 7.4 % in the 12 month cohort (HR 1.12, 95 % CI 0.87–1.45, p = .002 for non-inferiority). Of note, between 91 and 360 days, no statistically significant association was observed for the primary endpoint between the short- and long-term DAPT groups (2.6 vs. 2.6 %, HR 1.03, 95 % CI 0.66–1.60), MACE (5.3 vs. 4.3 %, HR 1.22, 95 % CI 0.88–1.70) or stent thrombosis (4 [0.3 %] vs. 1 [0.1 %], HR 3.97, 95 % CI, 0.44–35.49). Though not adequately powered, the low event rate in this study raises the question of whether a 3-month course of DAPT is safe in low-risk patients receiving a zotarolimus-eluting stent.

The ITALIC trial (NCT01476020) [32] was a multi-centered, prospective, open-label randomized trial that also hypothesized that 6 months of DAPT was non-inferior to 24 months of DAPT in aspirin-sensitive patients. Of the 2031 patients (1822 actually analyzed) (44 % ACS) who had received at least 1 Xience V DES, there was no difference in the primary end point (composite of all-cause mortality, MI, urgent TVR, CVA, or thrombolysis in myocardial infarction (TIMI) major bleeding) between both groups (1.5 vs. 1.6 %, p = 0.85). The authors concluded that in aspirin-sensitive patients (all patients were tested), 6 months of DAPT was non-inferior to 24 months when a Xience V stent was placed. However, this study had low event rates and was not powered adequately nor was it powered for stent thrombosis. Additionally, the trial was terminated early due to slow enrollment.

Unfortunately, none of these studies were powered to compare stent thrombosis. Critics have stated that these studies were even underpowered to identify actual differences in the composite endpoints given the relatively low event rates. There were also methodological issues within the studies that have impaired the applicability of these trials [23••]. For instance, PRODIGY pooled analysis of DES and BMS which clearly differ with regard to necessary DAPT duration. Several meta-analyses have been completed using some of the aforementioned trials. The most recent meta-analysis [33•] examined the REAL/ZEST-LATE, PRODIGY, EXCELLENT, and RESET trials. These four RCTs provide a median short duration DAPT of 6.2 months (n = 4081) vs. 16.8 months in the prolonged DAPT group (n = 4076). There was no difference in the rate of the composite endpoint of cardiac death or myocardial infarction between the short (3.3 %) and prolonged (3.0 %) DAPT groups (OR 1.11, 95 % CI 0.81–1.43, p = 0.41). In this study, an analysis [34] performed at the time of DAPT discontinuation clearly demonstrated a non-significant higher rate of ST in the shorter duration of DAPT vs. the longer period (0.35 vs. 0.20 %, p = 0.22). Major bleeding was higher in the prolonged DAPT group (0.29 vs. 0.71 %, p = 0.01). Another recent meta-analysis of the same trials confirmed that longer DAPT therapy did not reduce the risk of all-cause mortality (OR 0.89, 95 % CI 0.67–1.17, p = 0.4, I 2 = 0 %), myocardial infarction (OR 1.16, 95 % CI 0.85–1.57, p = 0.35, I 2 = 0 %), cardiac death (OR 0.88, 95 % CI 0.61–1.25, p = 0.47, I 2 = 0 %), stent thrombosis (OR 1.29, 95 % CI 0.76–2.21, p = 0.35, I 2 = 0 %), or cerebrovascular accidents (OR 0.73, 95 % CI 0.41–1.27, p = 0.26, I 2 = 0 %). They also found an increase in major TIMI bleeding (OR 51, 95 % CI 0.29–0.89, p = 0.02, I 2 = 0 %) and concluded that there is no difference in the efficacy outcomes of ≤6 and ≥12 months of DAPT in patients receiving DES stents.

Recently, a study assessing the safety of 6- vs. 12-month DAPT in second-generation DES, i.e., the SECURITY RCT trial (NCT00944333), was reported [35]. Initially designed to enroll 4000 patients, this study was prematurely terminated because of “enrollment futility because of minimal differences in the rate of the primary endpoint between the two groups.” Overall, a total of 1399 patients with stable or unstable angina or documented silent ischemia who received at least one second-generation DES were randomized. The primary composite endpoint (cardiac death, MI, stroke, definite or probable stent thrombosis, or BARC type 3 or 5 bleeding at 12 months) occurred in 4.5 vs. 3.7 % (risk difference 0.8 %, 95 % CI −2.4–1.7 %, p = 0.469) of patients in the 6- vs. 12-month groups, respectively. There was also no difference in the secondary endpoint (the primary endpoint + BARC type 2 bleeding) or stent thrombosis at 12 or 24 months. There was no difference in bleeding, but overall, the event rates were far less than expected, and this study was grossly underpowered. Further confounding its findings is that DAPT was still being used in 33.8 % of the 6-month group at the 12-month follow-up. This fact coupled with the low event rates and the underpowered nature of the study limit the applicability of this study. Interestingly, the authors used multivariable analysis to assess for factors that influenced the primary endpoint. They determined that several procedure-related factors (mean stent length, size, and number) and patient factors (age) were strong independent predictors highlighting that each patient’s risk for adverse events likely differs [36].

The results of the ISAR-SAFE trial (NCT00661206) [37], a multi-center international study comparing 6 vs. 12 months of DAPT, were released in November 2014. This study initially planned to enroll 6000 patients to achieve a power sufficient for the primary endpoint of death, myocardial infarction, stent thrombosis, or major bleeding at 1 year. This number was determined to detect a non-inferiority difference of 2 % from an expected rate of 10 % and was not powered to detect a difference in ST alone. The trial randomized 4005 patients (6 months n = 1997, 12 months n = 2007) prior to early termination due to a lower than anticipated event rate. Patients enrolled in this study (24 % diabetics) were real world with various presentations (SA 48 %, ACS 52 %) and stent type (everolimus-eluting stents 49 %, zotarolimus-eluting stents 15 %, newer generation sirolimus-eluting stents 16 %, biolimus 8 %, and BMS 0.4 %). There was no difference in MACE between the 6- vs. 12-month group (1.5 vs. 1.6 %, p for non-inferiority <0.001) and the composite of death, MI, CVA, and stent thrombosis was similar (1.3 vs. 1.5 %, p = 0.59). Individual endpoints included mortality (0.4 vs. 0.6 %, p = 0.37), MI (0.7 vs. 0.7 %, p = 0.85), stent thrombosis (0.3 vs. 0.2 %, p = 0.74), and stroke (0.4 vs. 0.3 %, p = 0.57). Major and minor bleeding were numerically lower with the 6-month group (0.3 vs. 0.7 %, p = 0.12). BARC >2 class bleeding was significantly reduced with the abbreviated DAPT therapy (1 vs. 2 %, p = 0.01). Readers should interpret these data, understanding that the trial was stopped early due to a significantly lower than expected event rate of 1.6 % vs. an anticipated 10 % rate and therefore was underpowered.

Additional considerations that impact risk

Currently, the guidelines suggest a one size fits all approach for all patients after they receive a DES. These recommendations do not take into consideration patient or procedural variables, both of which are known to impact individual patient risks for both ST and future events such as bleeding. In fact, inconsistency in the current data investigating longer vs. shorter DAPT may derive from inclusion of patients presenting with very distinct clinical presentations (stable angina vs. acute coronary syndrome) and treatments (stent type). Future studies involving specific sub-groups are needed.

The most feared complication of an abbreviated DAPT duration is stent thrombosis (ST) and for those physicians contemplating this approach, it is critical to understand the patient, procedural, and anatomic risk factors for this potentially fatal complication. Patient-specific variables that increase the risk of ST include ACS, smoking, diabetes mellitus, renal failure, and low ejection fraction [38–41]. A pooled analysis from the SPIRIT and COMPARE trials [19] showed that younger patients (<65 years of age) have a higher risk of ST compared to their elderly colleagues (>65 years of age). Procedural variables that increase the risk of ST with DES include direct stenting during ACS, longer stent length, overlapping stents, incomplete lesion coverage, persistent slow flow, residual stenosis, and dissection [42–45]. Though intravascular ultrasound (IVUS)-guided stenting to ensure appropriate expansion has not been proven in RCTs, a propensity-score matched analysis showed that patients undergoing IVUS-guided DES placement had a lower definite ST rate at 30 days and 12 months than those wherein IVUS was not used [46]. Additionally, bifurcation stenting, treatment of in-stent restenosis, chronic total occlusion, as well as stenting in lesions with necrotic cores (as assessed by IVUS with virtual histology) appear to increase the risk of ST [40]. These procedural and patient presentation factors need to be considered when determining the optimal duration of DAPT.

There are stent-related factors that also increase the risk of ST including the anti-proliferative agent used and its dose, strut/polymer thickness, and coating technologies which are all factors that differ between stent types and the two different generations. Second-generation differ from first-generation DESs with respect to both the anti-proliferative agents (second-generation everolimus and zotarolimus vs. first-generation paclitaxel and sirolimus), the coating technologies employed toward the polymer layer, and the stent frame [47]. With advancements in technology, the second-generation stents may result in better stent apposition leading to improved endothelialization which may reduce the risk of ST [48]. With thinner struts, more endothelium may be covered. This likely contributes as much as the type of anti-proliferative drug present as lower ST rates have been seen with all of the second-generation stents [48]. Therefore, providers considering a shorter DAPT for their patients should also consider the type of stent implanted. In a pre-specified analysis from the PRODIGY trial [49], paclitaxel stents showed a significantly higher rate of definite, probable, or possible ST compared to everolimus-eluting stent, zotarolimus-eluting stents, or bare metal stents. Additionally, the OPTIMIZE trial showed non-inferiority of a shorter DAPT (3 months) with zotarolimus-eluting stents over prolonged DAPT [31]. The PROTECT trial [30] showed no difference in ST beyond 1 year with zotarolimus-eluting stents and sirolimus-eluting stents. In contrast, the TL-PAS study [26] suggests that for the TAXUS Liberté paclitaxel-eluting stent, >30 months of DAPT is beneficial.

Future directions/areas of investigation

The current ACC/AHA guidelines [14] recommend that patients receiving DES can receive prasugrel or ticagrelor along with aspirin post ACS or stent implantation (in place of clopidogrel) based on large trials showing the enhanced efficacy of the newer agents in the reduction of ischemic events [50, 51]. The optimal duration of DAPT with these newer agents is unknown, though the TL-PAS study [26] would suggest that in patients who receive paclitaxel stent, prolonged DAPT out to 30 months should be considered. The EDUCATE trial (NCT01069003) is evaluating in a double-blind, placebo-controlled fashion the impact of 12 vs. 30 months of DAPT (clopidogrel or prasugrel + aspirin) in 2500 patients receiving the Endeavor zotarolimus-eluting stent on a composite outcome (cardiac death, MI, ST, bleeding, and DAPT compliance). The OPTIDUAL trial (NCT00822536) [52•] is ongoing and will add additional information about the long-term impact of prolonged DAPT therapy out to 48 vs. 12 months. Various permutations of DAPT are also being evaluated. The GLOBAL LEADERS study (NCT01813435) (n = 16,000) will assess the safety and efficacy of 1 month of DAPT with ticagrelor and aspirin followed by 23 months of ticagrelor monotherapy vs. 12 months of DAPT with aspirin and ticagrelor upon the outcome of all-cause death or MI.

The optimal duration of DAPT with the newer bioabsorbable polymers or scaffolds is also the subject of scrutiny and debate. Though late stent thrombosis may not be an issue with these bioabsorbable stents, the delayed absorption of these scaffolds may actually require longer therapy than our current DES platforms.

Conclusion

Overall, the optimal duration of DAPT following implantation of a DES remains debatable and is dependent upon patient, procedural, and anatomical characteristics that most physicians do not consider. Current ACC/AHA and European Heart Society guidelines suggest a one size fits all approach; however, benefit vs. risk of longer DAPT or shorter DAPT is patient specific. Additionally, whether to abruptly discontinue or taper DAPT remains an important unanswered question given increased event rates following DAPT cessation. The further identification of risk factors for early and late events including trials involving the newer agents, prasugrel, and ticagrelor are needed.