- Email: [email protected]
Assessment of P2Y12 inhibitor usage and switching in acute coronary syndrome patients undergoing percutaneous coronary revascularization
International Journal of Cardiology 223 (2016) 854–859
Contents lists available at ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
Assessment of P2Y12 inhibitor usage and switching in acute coronary syndrome patients undergoing percutaneous coronary revascularization☆,☆☆ Mrudula Kudaravalli d, Andrew D. Althouse a, Oscar C. Marroquin a, Sameer J. Khandhar c, Michael S. Sharbaugh a, Catalin Toma a, A.J. Conrad Smith a, John T. Schindler a, Joon S. Lee a, Suresh R. Mulukutla a,b,⁎ a
Heart and Vascular Institute, Department of Medicine, University of Pittsburgh, Pittsburgh, PA, United States Division of Cardiology, Veterans Affairs Hospital, Pittsburgh, PA, United States c Division of Cardiology, University of Pennsylvania, Philadelphia, PA, United States d Department of Medicine, Allegheny General Hospital, Pittsburgh, PA, United States b
a r t i c l e
i n f o
Article history: Received 13 May 2016 Received in revised form 5 August 2016 Accepted 6 August 2016 Available online 08 August 2016 Keywords: Thienopyridine P2Y12 inhibitor Percutaneous coronary intervention Coronary disease
a b s t r a c t Background: Dual antiplatelet therapy is recommended for patients with acute coronary syndrome (ACS) that undergo percutaneous coronary intervention (PCI). However, the effect of switching P2Y12 inhibitors between the loading dose and therapy after discharge is not well described. Methods: This post-hoc analysis of a prospectively collected registry included 3219 consecutive ACS patients who underwent PCI. Patients were categorized into four groups: clopidogrel at load and discharge (C–C), loading dose of clopidogrel and discharged on prasugrel/ticagrelor (C–PT), loading dose of prasugrel/ticagrelor and discharged on clopidogrel (PT–C), and prasugrel/ticagrelor at load and discharge (PT–PT). Results: While 77.6% of patients received the C–C treatment regimen and 13.6% received the PT–PT strategy, the strategy of P2Y12 switching was fairly common with 6.2% in the PT–C group and 2.6% in the C–PT group. While C– C was the most common treatment regimen, PT–C and PT–PT were more commonly used in STEMI patients than in NSTEMI or unstable angina patients. A significantly lower unadjusted incidence of the composite outcome (death, MI, and repeat revascularization) was appreciated in both the PT–C (1.0%) and PT–PT (2.3%) groups than the C–C group (4.0%). Propensity-score matched analysis still showed significantly reduced risk (HR = 0.22, 95% CI 0.05–0.93, p = 0.04) in the PT–C group vs. a matched group of C–C controls. Conclusions: The strategy of utilizing a newer P2Y12 inhibitor and then switching to clopidogrel in ACS patients following PCI is used with some frequency in routine clinical practice and further studies should evaluate the safety and efficacy of such a strategy. Published by Elsevier Ireland Ltd.
1. Introduction With the advent of clopidogrel in 1997, dual anti-platelet therapy (DAPT) with P2Y12 inhibitor and aspirin became the standard of care in patients with acute coronary syndrome (ACS) undergoing percutaneous
☆ Statement of authorship: Each author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation. ☆☆ Funding: This research was funded, in part, by funds from an Abbott Vascular grant and from St. Jude Medical. The remaining funding was from the University of Pittsburgh Medical Center. ⁎ Corresponding author at: University of Pittsburgh, VA Pittsburgh Healthcare System, Division of Cardiology, University Drive C, Cardiology/Cath Lab 111C-U, Pittsburgh, PA 15240, United States. E-mail address: [email protected] (S.R. Mulukutla).
http://dx.doi.org/10.1016/j.ijcard.2016.08.144 0167-5273/Published by Elsevier Ireland Ltd.
coronary intervention (PCI) and stenting [1–3]. However, patients continued to have adverse events, in part due to variability in response to clopidogrel secondary to patient phenotype, genetic polymorphism and drug–drug interactions [4–6]. More recently, novel P2Y12 inhibitors (prasugrel and ticagrelor) were introduced to the market. Subsequently, randomized clinical trials conducting head-to-head comparison between these novel agents and clopidogrel showed improved outcomes with novel agents over DAPT with clopidogrel [7,8]. However, the novel P2Y12 inhibitors are not without limitations. Ticagrelor is known to cause dyspnea, increases in uric acid and creatinine levels and ventricular pauses [9], while prasugrel increases the risk of major bleeding and is contraindicated in numerous patient populations [10]. Current AHA guidelines do not recommend one specific P2Y12 inhibitor, and therefore the choice of P2Y12 agents is at the discretion of the physician [2].
M. Kudaravalli et al. / International Journal of Cardiology 223 (2016) 854–859
With this gray area in recommendations, it is possible that some patients may receive one P2Y12 inhibitor for their loading dose, but be switched to another for treatment after hospital discharge because of the different pharmacologic aspects of the newer agents, the lower cost of clopidogrel, or compliance issues [11,12]. While a few studies have addressed the impact of switching between P2Y12 agents, most focus primarily on the level of platelet inhibition achieved [13–15]. There is little in the literature to date to document the frequency of P2Y12 switching and its potential impact upon clinical outcomes, but there are some potential concerns when switching related to gaps in effective platelet inhibition or excessive platelet inhibition resulting in bleeding [16,17]. Therefore, we sought to (1) define the frequency of switching P2Y12 agents after PCI in routine clinical practice, (2) to describe the differences in populations where different DAPT strategies were used, and (3) to attempt to evaluate the efficacy and safety of this strategy. 2. Methods Patients at four University of Pittsburgh Medical Center (UPMC) hospitals with ACS undergoing PCI and stenting were consecutively enrolled starting in August 2011 as part of an IRB-approved hospital-based registry carried on through May 2014. Of the 7009 patients undergoing PCI in this time, 4617 patients presented with ACS (STEMI, NSTEMI and Unstable angina). Patients that did not receive at least one drug-eluting stent or died in hospital were excluded from the study. Our analytic cohort was comprised of the 3219 patients that received a drug-eluting stent who were discharged alive with a P2Y12 inhibitor (Fig. 1). Information regarding P2Y12 inhibitors given procedurally and at discharge was obtained using the hospital-based registry. Patients are divided into four groups for analysis: those receiving a clopidogrel loading dose (600 mg) and discharged on clopidogrel (C–C), those receiving a loading dose of clopidogrel (600 mg) and discharged on prasugrel/ ticagrelor (C–PT), those receiving a loading dose of prasugrel/ticagrelor and discharged on clopidogrel (PT–C), and those treated with prasugrel/ticagrelor at load and discharge (PT–PT). Baseline data (demographics, clinical characteristics, cardiac presentation, and
855
procedural details) were obtained from the registry, as were patient outcomes in the first 90 days. With a limited number of events in the short follow-up, the primary outcome used to measure the impact of switching between P2Y12 inhibitors is a composite of death, MI, and repeat PCI/revascularization.
2.1. Statistical methods Continuous variables are presented as median (range), while categorical variables are presented as n (%). Analysis of variance (ANOVA) was used to compare the distribution of continuous variables between the four treatment groups; chi-squared tests were used to compare the distribution of categorical variables (or Fisher's Exact Test in the case of small sample sizes). We present overall “global” comparisons of all four treatment groups as well as pairwise comparisons for each of the C–PT, PT–C, and PT–PT groups compared against the C–C group. Outcomes are presented as n (%) occurring within the first 90 days. Kaplan–Meier analysis is used to present freedom from the composite outcome in patients receiving each treatment strategy. However, due to a) the large number of baseline variables that differ significantly between treatment groups and b) the impracticality of multivariable analyses adjusting for large numbers of confounding variables with a low number of events, we present propensity-matched analyses to allow proper comparison between groups. Briefly, all baseline variables were considered candidate variables for a multivariate logistic regression to create propensity scores for receiving the C–PT treatment regimen (versus C–C treatment). The propensity scores were used to match patients that received C–PT to those getting C–C on a set of prespecified demographic and clinical characteristics. The final multivariate logistic regression used to create propensity scores included the following variables: age, gender, race, diabetes, hypertension, dyslipidemia, prior HF, prior PAD, prior PCI, prior CABG, prior MI, CAD Presentation, and number of diseased vessels. After calculating the propensity scores, patients receiving C–PT treatment were randomly sequenced and the C–C patients were matched 4:1 (without replacement) to each C–PT patient with using a caliper distance of 0.10 for matching. Baseline comparisons of between-group differences are repeated in the propensity-matched cohort to ensure adequate balance in risk profile. Cox proportional-hazards models were then used to estimate the hazard ratio (HR) for the composite outcome in the C–PT group vs. the C–C group. This process was repeated for both C–PT group and the PT–C group to evaluate the effects of switching P2Y12 inhibitor against comparable control groups of patients treated with C–C strategy. All statistical analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC).
Fig. 1. Flow chart representing analytic cohort used for analysis.
856
M. Kudaravalli et al. / International Journal of Cardiology 223 (2016) 854–859
3. Results The relative frequency of the four treatment groups is plotted over time in Fig. 2. The large majority of patients through the entire duration of the study received the C–C treatment regimen (2500/ 3219 = 77.6% overall), while very few patients (83/3219 = 2.6% overall) were given the C–PT treatment regimen. Frequency of the PT–C group (199/3219 = 6.2% overall) steadily climbed until early 2014, while the PT–PT group (437/3219 = 13.6% overall) accounted for a relatively steady proportion across most of the study. Demographics, clinical characteristics, and cardiac presentation of the each treatment group is summarized Table 1. As expected, there were many significant differences between groups in baseline risk factors, especially when comparing the PT–C and PT–PT groups against the C–C group. In particular, patients in both the PT–C and PT–PT groups tended to be younger, had fewer comorbid conditions, had less history of cardiovascular intervention, and were more likely to have singlevessel disease than patients in the C–C group. While C–C was the most common treatment regimen for all levels of CAD presentation, it is worth noting that PT–C and PT–PT was more commonly used in STEMI patients than in NSTEMI or unstable angina patients (Fig. 3). Outcome counts for the entire cohort are summarized in Table 2. We see a lower unadjusted incidence of the composite outcome (death, MI, and repeat revascularization) in both the PT–C (1.0%) and PT–PT (2.3%) groups than the C–C group (4.0%) (Fig. 4). However, given the baseline differences between groups, the propensity-score matched analysis is more appropriate for evaluating the strategy of switching P2Y12 inhibitors against an appropriate control group. Characteristics of the respective propensity-score matched cohorts are shown in Table 3. The matching algorithm successfully created cohorts with four C–C controls per case for each of the “switch” groups (both C–PT and PT–C) with good balance between groups on a plurality of baseline characteristics. Slight differences remained in prior CABG and CAD presentation; for the C–PT vs. C–C comparison, this yields higher prevalence of prior CABG, but a lower prevalence of STEMI patients in the C–C control group vs. the C–PT group. For the PT–C vs. C–C comparison, this again returned a higher percentage of prior CABG, but lower prevalence of STEMI patients in the C–C control group vs. the PT–C group. The risk of composite outcome in the C–PT group was not much different than their matched C–C controls (HR = 1.34, 95% CI 0.43–4.16, p = 0.612). However, outcomes in the PT–C group were quite good
Fig. 2. P2Y12 inhibitor use during & after cardiac catheterization, 2011–2014 (by quarter). Abbreviations: C–C = clopidogrel load, clopidogrel discharge; C–PT = clopidogrel load, prasugrel/ticagrelor discharge; PT–C = prasugrel/ticagrelor load, clopidogrel discharge; PT–PT = prasugrel/ticagrelor load, prasugrel/ticagrelor discharge.
relative to their matched C–C controls, with a significantly lower incidence of the composite outcome (HR = 0.22, 95% CI 0.05–0.93, p = 0.04) despite the small sample size. While this hazard ratio is derived from a small number of events and therefore may not be a truly representative estimate of the absolute treatment effect, it is striking enough to warrant attention as a potentially viable treatment strategy that could be explored further. 4. Discussion Using data from a hospital-based registry, we evaluated the prevalence of P2Y12 switching in routine clinical practice at a large healthcare institution. In addition, we evaluated the incidence of clinical outcomes in PCI patients receiving stents, broken down by the P2Y12 inhibitor(s) used for loading dose and at discharge, though these results are only hypothesis generating at this point. Despite the superiority of prasugrel and ticagrelor in randomized clinical trials, we observed that the majority of patients in our clinical practice continued to receive clopidogrel, while a substantial minority received a different P2Y12 inhibitor for their loading dose than at discharge. Over the course of our study, 23% of patients got one of the newer agents at some point in their course of therapy, and this is consistent with other studies [18]. Data from the National Cardiovascular Data Registry (NCDR) shows an increase in prasugrel use from 3% to 18% between 2009 and 2012 [19]. Our study also revealed that physicians were more likely to switch P2Y12 agents in STEMI patients, as they tend to have a higher thrombotic burden compared to the unstable angina population, who primarily received clopidogrel. However, even with the higher prevalence of STEMI patients in the PT–C arms, the clinical outcomes for these patients were still quite good. Despite the established superiority of prasugrel and ticagrelor, ongoing physician preference for clopidogrel could be explained by the availability of generic clopidogrel allowing for lower cost to the patients, as well as increased compliance with once a day dosing of clopidogrel [20]. When comparing the groups that switched P2Y12 agents (C–PT and PT–C) to patients that remained on clopidogrel throughout (C–C), the group loaded with prasugrel/ticagrelor (PT–C) experienced particularly good outcomes. However, our analyses cannot make any firm conclusions regarding these outcomes given that this was not a randomized study, and it is impossible to account for all potential factors that would account for why P2Y12 switching would occur. From the standpoint of hypothesis generation, our data suggest improved outcomes when one of the newer agents is given for the loading dose and highlight the potential importance of loading acute coronary syndrome patients with one of the newer agents, prasugrel or ticagrelor, and that this may be the most critical aspect in determining short-term, 3-month outcomes after ACS. In both TRITON-TIMI 38 and the PLATO studies, the curves for the primary endpoint separate early [7,8]. For instance, in TRITON-TIMI 38, at 90 days, 5.6% of the clopidogrel group met the primary endpoint compared with 4.7% in the prasugrel group. Similarly, in our study, the composite endpoint was met in 4.0% of patients in the C–C group and 2.2% in any of the groups that received any of the newer agents for their loading dose. Mechanistically, our observations of improved outcomes in patients after loading with prasugrel or ticagrelor may be explained by the rapid onset of action of the newer agents, in comparison to clopidogrel, and the fact that the newer agents are able to provide effective platelet inhibition after a very short duration after administration of the loading dose [21–23]. Although both the C–PT group and the PT–C group did experience a slightly higher incidence of bleeding events than the C–C group postdischarge, the absolute number of events was very low (1/83 = 1.2% in the C–PT group and 4/199 = 2.0% in the PT–C group), and these bleeding rates were not statistically significant. Hence, it appears that the decision to switch P2Y12 medication from load to discharge does not increase bleeding risk. However, again, we only evaluated 3month outcomes, and prior studies have suggested that prasugrel and
M. Kudaravalli et al. / International Journal of Cardiology 223 (2016) 854–859
857
Table 1 Patient demographics, clinical characteristics, and cardiac presentation.
# patients Age at admission Male, % Race White Black Other BMI Diabetes Hypertension Dyslipidemia Prior HF Prior PAD Prior PCI Prior CABG Prior MI CAD presentation Unstable angina NSTEMI STEMI Postcardiac arresta (prior 24 h) Postcardiogenic shock (prior 24 h)a Number of diseased vessels 0 1 2 3
p-Value1
C–C
C–PT
2500 67.0 (19.0–96.0) 1658 (66.3%)
83 59.0 (31.0–90.0) 59 (71.1%)
2304 (92.2%) 178 (7.1%) 14 (0.6%) 29.3 (13.9–61.1) 985 (39.4%) 2144 (85.8%) 1983 (79.3%) 337 (13.5%) 347 (13.9%) 1182 (47.3%) 607 (24.3%) 949 (38.0%)
73 (88.0%) 9 (10.8%) 1 (1.2%) 30.5 (21.0–65.0) 36 (43.4%) 61 (73.5%) 58 (69.9%) 7 (8.4%) 7 (8.4%) 35 (42.2%) 10 (12.0%) 30 (36.1%)
1445 (57.8%) 744 (29.8%) 311 (12.4%) 37 (1.5%) 26 (1.0%)
30 (36.1%) 30 (36.1%) 23 (27.7%) 3 (3.6%) 2 (2.4%)
14 (0.6%) 799 (32.0%) 997 (39.9%) 690 (27.6%)
1 (1.2%) 28 (33.7%) 33 (39.8%) 21 (25.3%)
b.001 0.365 0.322
0.042 0.466 0.002 0.037 0.183 0.156 0.359 0.010 0.737 b.001
0.121 0.236
PT–C 199 64.0 (23.0–96.0) 136 (68.3%) 193 (97.0%) 3 (1.5%) 2 (1.0%) 28.9 (14.7–52.7) 49 (24.6%) 142 (71.4%) 131 (65.8%) 15 (7.5%) 9 (4.5%) 54 (27.1%) 13 (6.5%) 48 (24.1%) 44 (22.1%) 55 (27.6%) 100 (50.3%) 6 (3.0%) 2 (1.0%)
0.847
p-Value2
PT–PT 437 59.0 (30.0–92.0) 317 (72.5%)
b.001 0.561 0.008
411 (94.1%) 25 (5.7%) 0 (0.0%) 30.0 (15.1–60.1) 135 (30.9%) 319 (73.0%) 305 (69.8%) 31 (7.1%) 30 (6.9%) 154 (35.2%) 73 (16.7%) 129 (29.5%)
0.244 b.001 b.001 b.001 0.017 b.001 b.001 b.001 b.001 b.001
184 (42.1%) 144 (33.0%) 109 (24.9%) 9 (2.1%) 1 (0.2%)
0.096 0.963 0.005
1 (0.5%) 88 (44.2%) 67 (33.7%) 43 (21.6%)
p-Value3
p-Value4
b.001 0.011 0.161
b.001 0.067 0.013
0.013 b.001 b.001 b.001 b.001 b.001 b.001 b.001 b.001 b.001
0.008 b.001 b.001 b.001 b.001 b.001 b.001 b.001 b.001 b.001
0.368 0.101
0.178 0.217
b.001
b.001
2 (0.5%) 190 (43.5%) 168 (38.4%) 77 (17.6%)
Continuous variables reported as median (range); comparisons performed using Kruskal–Wallis test. Categorical variables reported as n (%); comparisons performed using chi-squared tests. Abbreviations: C–C = clopidogrel load, clopidogrel discharge; C–PT = clopidogrel load, prasugrel/ticagrelor discharge; PT–C = prasugrel/ticagrelor load, clopidogrel discharge; PT–PT = prasugrel/ticagrelor load, prasugrel/ticagrelor discharge; HF = heart failure; PAD = peripheral artery disease; PCI = percutaneous coronary intervention; CABG = coronary artery bypass graft; MI = myocardial infarction; NSTEMI = non-ST elevation myocardial infarction; STEMI = ST-elevation myocardial infarction. a Postcardiac arrest and postcardiogenic shock refer to patients who suffered one of these events within the 24 h prior to their PCI procedure. 1 p-Value comparing C–PT group to C–C group. 2 p-Value comparing PT–C group to C–C group. 3 p-Value comparing PT–PT group to C–C group. 4 p-Value for test across all 4 groups.
ticagrelor may be associated with higher bleeding rates, at least in specific patient populations [24]. While the increased bleeding risk with the newer agents may exist only in specific patient subsets, our findings suggest that there may be an attractive alternative by loading ACS patients with a newer agent and discharging them on clopidogrel, thereby giving patients the benefit of a more potent antiplatelet agent
for the load and an agent that may be associated with lower risk for bleeding on discharge. Mechanistically, our observations of improved outcomes in patients after loading with prasugrel or ticagrelor may be explained by the rapid onset of action of the newer agents, in comparison to clopidogrel, and the fact that the newer agents are able to provide effective platelet inhibition after a very short duration after administration of the loading dose.
C-C
Table 2 90-day patient outcomes, by treatment group.
C-C
C-C
PT-C
PT-PT
PT-PT PT-PT C-PT
PT-C PT-C
C-PT
C-PT
Fig. 3. Bar chart of drug groups by CAD presentation. Abbreviations: C–C = clopidogrel load, clopidogrel discharge; C–PT = clopidogrel load, prasugrel/ticagrelor discharge; PT– C = prasugrel/ticagrelor load, clopidogrel discharge; PT–PT = prasugrel/ticagrelor load, prasugrel/ticagrelor discharge.
C–C
C–PT
PT–C
PT–PT
# patients
2500
83
199
437
Primary outcomes Compositea Death MI Repeat revascularization PCI CABG
99 (4.0%) 53 (2.1%) 18 (0.7%) 55 (2.2%) 51 (2.0%) 4 (0.2%)
4 (4.8%) 0 (0.0%) 0 (0.0%) 4 (4.8%) 3 (3.6%) 1 (1.2%)
2 (1.0%) 2 (1.0%) 0 (0.0%) 2 (1.0%) 1 (0.5%) 1 (0.5%)
10 (2.3%) 3 (0.7%) 0 (0.0%) 8 (1.8%) 7 (1.6%) 1 (0.2%)
Other outcomes Stroke Bleeding Stent thrombosis
4 (0.2%) 28 (1.1%) 0 (0.0%)
0 (0.0%) 1 (1.2%) 0 (0.0%)
0 (0.0%) 4 (2.0%) 2 (1.0%)
1 (0.2%) 1 (0.2%) 0 (0.0%)
All statistics reported as n (%); unadjusted comparisons of event rates performed using chi-squared tests. Abbreviations: C–C = clopidogrel load, clopidogrel discharge; C–PT = clopidogrel load, prasugrel/ticagrelor discharge; PT–C = prasugrel/ticagrelor load, clopidogrel discharge; PT–PT = prasugrel/ticagrelor load, prasugrel/ticagrelor discharge; MI = myocardial infarction; PCI = percutaneous coronary intervention; CABG = coronary artery bypass graft. a Composite endpoint includes first incident of death, MI, or repeat revascularization.
858
M. Kudaravalli et al. / International Journal of Cardiology 223 (2016) 854–859
Fig. 4. Freedom from death, MI, and repeat revascularization in first 90 days after stenting. Abbreviations: C–C = clopidogrel load, clopidogrel discharge; C–PT = clopidogrel load, prasugrel/ticagrelor discharge; PT–C = prasugrel/ticagrelor load, clopidogrel discharge; PT–PT = prasugrel/ticagrelor load, prasugrel/ticagrelor discharge.
If our results suggesting that delivering a load with prasugrel/ ticagrelor and subsequent treatment with clopidogrel upon discharge results in better clinical outcomes than C–C treatment and equivalent outcomes to PT–PT treatment, it is possible that a “combination strategy” of loading with a dose of prasugrel/ticagrelor and discharging with a prescription for clopidogrel could achieve the optimal balance between favorable clinical outcomes and reduced patient costs (relative to giving a full course of treatment with prasugrel/ticagrelor). However, this needs to be tested within a randomized clinical trial setting given the noted limitations of this analysis.
5. Limitations The hospital-based registry only had complete follow-up for all patients through 90 days after discharge, resulting in a relatively low number of events during follow-up. Therefore, further analysis with greater follow-up time is required to confirm the outcomes for patients treated with PT–C strategy relative to all other groups. A randomized controlled trial should be considered confirm these results, especially because the prevalence of P2Y12 switching is not infrequent and is not supported by clinical data as yet. The significant baseline differences
Table 3 Patient demographics, clinical characteristics, and cardiac presentation: propensity-score matched cohorts.
# patients Age at admission Male, % Race White Black Other BMI Diabetes Hypertension Dyslipidemia Prior HF Prior PAD Prior PCI Prior CABG Prior MI CAD presentation Unstable angina NSTEMI STEMI Postcardiac arrest (prior 24 h) Postcardiogenic shock (prior 24 h) Number of diseased vessels 0 1 2 3
C–PT
Matched C–C Cohort A
83 59.0 (31.0–90.0) 59 (71.1%)
332 64.0 (19.0–93.0) 216 (65.1%)
73 (88.0%) 9 (10.8%) 1 (1.2%) 30.5 (21.0–65.0) 36 (43.4%) 61 (73.5%) 58 (69.9%) 7 (8.4%) 7 (8.4%) 35 (42.2%) 10 (12.0%) 30 (36.1%)
302 (91.0%) 27 (8.1%) 3 (0.9%) 29.7 (13.9–61.1) 143 (43.1%) 269 (81.0%) 246 (74.1%) 45 (13.6%) 51 (15.4%) 154 (46.4%) 77 (23.2%) 132 (39.8%)
30 (36.1%) 30 (36.1%) 23 (27.7%) 3 (3.6%) 2 (2.4%)
158 (47.6%) 115 (34.6%) 59 (17.8%) 5 (1.5%) 4 (1.2%)
1 (1.2%) 28 (33.7%) 33 (39.8%) 21 (25.3%)
4 (1.2%) 111 (33.4%) 129 (38.9%) 88 (26.5%)
p-Value1
0.001 0.299 0.707
0.215 0.960 0.128 0.437 0.207 0.103 0.490 0.025 0.546 0.071
0.211 0.410 0.996
PT–C
Matched C–C Cohort B
198 64.0 (23.0–96.0) 135 (68.2%)
792 63.0 (25.0–96.0) 558 (70.5%)
193 (97.5%) 3 (1.5%) 2 (1.0%) 29.0 (14.7–52.7) 49 (24.7%) 142 (71.7%) 130 (65.7%) 15 (7.6%) 9 (4.5%) 54 (27.3%) 13 (6.6%) 48 (24.2%)
751 (94.8%) 39 (4.9%) 2 (0.3%) 29.1 (13.9–61.1) 233 (29.4%) 585 (73.9%) 550 (69.4%) 73 (9.2%) 75 (9.5%) 252 (31.8%) 107 (13.5%) 219 (27.7%)
44 (22.2%) 55 (27.8%) 99 (50.0%) 6 (3.0%) 2 (1.0%)
227 (28.7%) 267 (33.7%) 298 (37.6%) 23 (2.9%) 18 (2.3%)
1 (0.5%) 87 (43.9%) 67 (33.8%) 43 (21.7%)
7 (0.9%) 308 (38.9%) 282 (35.6%) 195 (24.6%)
p-Value2
0.987 0.532 0.035
0.475 0.192 0.540 0.304 0.467 0.026 0.215 0.007 0.333 0.006
0.924 0.258 0.574
Continuous variables reported as median (range); comparisons performed using Kruskal–Wallis test. Categorical variables reported as n (%); comparisons performed using chi-squared tests. Abbreviations: C–C = clopidogrel load, clopidogrel discharge; C–PT = clopidogrel load, prasugrel/ticagrelor discharge; PT–C = prasugrel/ticagrelor load, clopidogrel discharge; PT–PT = prasugrel/ticagrelor load, prasugrel/ticagrelor discharge; HF = heart failure; PAD = peripheral artery disease; PCI = percutaneous coronary intervention; CABG = coronary artery bypass graft; MI = myocardial infarction; NSTEMI = non-ST elevation myocardial infarction; STEMI = ST-elevation myocardial infarction. 1 p-Value comparing C–PT group to C–C group. 2 p-Value comparing PT–C group to C–C group.
M. Kudaravalli et al. / International Journal of Cardiology 223 (2016) 854–859
between groups and limited number of events preclude accurate treatment comparisons in our registry data. Finally, the reasons for switching P2Y12 agents were not recorded. Therefore, we were limited in identifying the reasons encountered in clinical practice by a physician switching P2Y12 agents, which could further explain the observed differences between groups. 6. Conclusion Treatment with a novel P2Y12 inhibitor (prasugrel/ticagrelor) followed by continued DAPT with clopidogrel is a strategy that appears to be commonly employed in routine clinical practice, though, still, the great majority of patients still receive a strategy of exclusively clopidogrel. While our data make comparisons challenging between groups using different strategies, our exploratory analysis suggest that outcomes in the C–PT group were not different compared to their matched C–C controls while outcomes in the PT–C group were better than their matched C–C controls. Further investigation should be conducted to determine the clinical impact of such strategies. Conflict of interest There are no conflicts of interest regarding this manuscript for any author. Acknowledgments The authors gratefully acknowledge the University of Pittsburgh Medical Center (UPMC), Abbott Vascular, and St. Jude Medical. A grant from Abbott Vascular and funds from UPMC helped to support the infrastructure to collect data and to enroll patients into this study. Funds from St. Jude Medical and UPMC helped to support our biostatistics staff to complete the analyses. References [1] S. Yusuf, F. Zhao, S.R. Mehta, S. Chrolavicius, G. Tognoni, K.K. Fox, Effects of clopidogrel in addition to aspirin in patients with acute coronary syndromes without ST-segment elevation, N. Engl. J. Med. 345 (2001) 494–502. (OpenUrl). [2] E.A. Amsterdam, N.K. Wenger, R.G. Brindis, D.E. Casey Jr., T.G. Ganiats, D.R. Holmes Jr., A.S. Jaffe, H. Jneid, R.F. Kelly, M.C. Kontos, G.N. Levine, P.R. Liebson, D. Mukherjee, E.D. Peterson, M.S. Sabatine, R.W. Smalling, S.J. Zieman, ACC/AHA Task Force Members, AHA/ACC guideline for the management of patients with non-STelevation acute coronary syndromes: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines, J. Am. Coll. Cardiol. 64 (24) (2014) e139–e228. [3] Z.M. Chen, L.X. Jiang, Y.P. Chen, J.X. Xie, H.C. Pan, R. Peto, R. Collins, L.S. Liu, COMMIT (ClOpidogrel and Metoprolol in Myocardial Infarction Trial) collaborative group, Addition of clopidogrel to aspirin in 45,852 patients with acute myocardial infarction: randomised placebo-controlled trial, Lancet 366 (2005) 1607–1621. [4] P.A. Gurbel, K.P. Bliden, Durability of platelet inhibition by clopidogrel, Am. J. Cardiol. 91 (2003) 1123–1125. [5] V.L. Serebruany, S.R. Steinhubl, P.B. Berger, A.I. Malinin, D.L. Bhatt, E.J. Topol, Variability in platelet responsiveness to clopidogrel among 544 individuals, J. Am. Coll. Cardiol. 45 (2005) 246–251. [6] G. Campo, G. Parrinello, P. Ferraresi, B. Lunghi, M. Tebaldi, M. Miccoli, J. Marchesini, F. Bernardi, R. Ferrari, M. Valgimigli, Prospective evaluation of on-clopidogrel platelet reactivity over time in patientstreated with percutaneous coronary intervention relationship with gene polymorphisms and clinical outcome, J. Am. Coll. Cardiol. 57 (2011) 2474–2483. [7] S.D. Wiviott, E. Braunwald, C.H. McCabe, G. Montalescot, W. Ruzyllo, S. Gottlieb, F.J. Neumann, D. Ardissino, S. De Servi, S.A. Murphy, J. Riesmeyer, G. Weerakkody, C.M. Gibson, E.M. Antman, TRITON-TIMI 38 Investigators, Prasugrel versus clopidogrel in patients with acute coronary syndromes, N. Engl. J. Med. 357 (2007) 2001–2015.
859
[8] L. Wallentin, R.C. Becker, A. Budaj, C.P. Cannon, H. Emanuelsson, C. Held, J. Horrow, S. Husted, S. James, H. Katus, K.W. Mahaffey, B.M. Scirica, A. Skene, P.G. Steg, R.F. Storey, R.A. Harrington, PLATO Investigators, A. Freij, M. Thorsén, Ticagrelor versus clopidogrel in patients with acute coronary syndromes, N. Engl. J. Med. 361 (2009) 1045–1057. [9] B.M. Scirica, C.P. Cannon, H. Emanuelsson, E.L. Michelson, R.A. Harrington, S. Husted, S. James, H. Katus, P. Pais, D. Raev, J. Spinar, P.G. Steg, R.F. Storey, L. Wallentin, PLATO Investigators, The incidence of bradyarrhythmias and clinical bradyarrhythmic events in patients with acute coronary syndromes treated with ticagrelor or clopidogrel in the PLATO (platelet inhibition and patient outcomes) trial: results of the continuous electrocardiographic assessment substudy, J. Am. Coll. Cardiol. 57 (19) (2011) 1908–1916. [10] S.D. Wiviott, E.M. Antman, E. Braunwald, Prasugrel, Circulation 122 (2010) 394–403. [11] A.D. Michelson, A.L. Frelinger III, E. Braunwald, W.E. Downey, D.J. Angiolillo, N.P. Xenopoulos, J.A. Jakubowski, Y. Li, S.A. Murphy, J. Qin, C.H. McCabe, E.M. Antman, S.D. Wiviott, TRITON-TIMI 38 Investigators, Pharmacodynamic assessment of platelet inhibition by prasugrel vs. clopidogrel in the TRITON-TIMI 38 trial, Eur. Heart J. 30 (2009) 1753–1763. [12] R.F. Storey, D.J. Angiolillo, S.B. Patil, B. Desai, R. Ecob, S. Husted, H. Emanuelsson, C.P. Cannon, R.C. Becker, L. Wallentin, Inhibitory effects of ticagrelor compared with clopidogrel on platelet function in patients with acute coronary syndromes, J. Am. Coll. Cardiol. 56 (2010) 1456–1462. [13] D. Alexopoulos, I. Xanthopoulou, S. Deftereos, G. Sitafidis, I. Kanakakis, M. Hamilos, C. Angelidis, S. Petousis, D. Stakos, H. Parissis, M. Vavouranakis, P. Davlouros, J. Goudevenos, C. Stefanadis, In-hospital switching of oral P2Y12 inhibitor treatment in patients with acute coronary syndrome undergoing percutaneous coronary intervention: prevalence, predictors and short-term outcome, Am. Heart J. 167 (1) (Jan 2014) 68–76. [14] R.F. Storey, D.J. Angiolillo, S.B. Patil, B. Desai, R. Ecob, S. Husted, H. Emanuelsson, C.P. Cannon, R.C. Becker, L. Wallentin, Optimizing P2Y-receptor inhibition in acute coronary syndrome patients based on platelet function testing: impact of prasugrel and high-dose clopidogrel, J. Am. Coll. Cardiol. 63 (11) (2014) 1061–1070. [15] D.J. Angiolillo, N. Curzen, P. Gurbel, P. Vaitkus, F. Lipkin, W. Li, J.A. Jakubowski, M. Zettler, M.B. Effron, D. Trenk, Pharmacodynamic evaluation of switching from ticagrelor to prasugrel in patients with stable coronary artery disease: results of the SWAP-2 Study (Switching Anti Platelet-2), J. Am. Coll. Cardiol. 63 (2014) 1500–1509. [16] A. Bagai, E.D. Peterson, E. Honeycutt, M.B. Effron, D.J. Cohen, S.G. Goodman, K.J. Anstrom, A. Gupa, J.C. Messenger, T.Y. Wang, In-hospital switching between adenosine diphosphate receptor inhibitors in patients with acute myocardial infarction treated with percutaneous coronary intervention: insights into contemporary practice from the TRANSLATE-ACS study, Eur. Heart J. Acute Cardiovasc. Care 4 (2015) 499–508. [17] A. Bagai, Y. Wang, T.Y. Wang, J.P. Curtis, H.S. Gurm, B. Shah, A.N. Cheema, E.D. Peterson, J.F. Saucedo, C.B. Granger, M.T. Roe, D.L. Bhatt, R.L. McNamara, K.P. Alexander, In-hospital switching between clopidogrel and prasugrel among patients with acute myocardial infarction treated with percutaneous coronary intervention: insights into contemporary practice from the national cardiovascular data registry, Circ. Cardiovasc. Interv. 7 (2014) 585–593. [18] A. Sandhu, M. Seth, S. Dixon, D. Share, D. Wohns, T. Lalonde, M. Moscucci, A.L. Riba, M. Grossman, H.S. Gurm, Contemporary use of prasugrel in clinical practice: insights from the Blue Cross Blue Shield of Michigan Cardiovascular Consortium, Circ. Cardiovasc. Qual. Outcomes 6 (3) (2013) 293–298. [19] M.W. Sherwood, S.D. Wiviott, S.A. Peng, M.T. Roe, J. Delemos, E.D. Peterson, T.Y. Wang, Early clopidogrel versus prasugrel use among contemporary STEMI and NSTEMI patients in the US: insights from the National Cardiovascular Data Registry, J. Am. Heart Assoc. 3 (2014). [20] D.S. Kazi, M.A. Hlatky, A delicate balance: the cost effectiveness of new antiplatelet agents, J. Am. Coll. Cardiol. 65 (2015) 477–479. [21] T. Jernberg, C.D. Payne, K.J. Winters, C. Darstein, J.T. Brandt, J.A. Jakubowski, H. Naganuma, A. Siegbahn, L. Wallentin, Prasugrel achieves greater inhibition of platelet aggregation and a lower rate of non-responders compared with clopidogrel in aspirin-treated patients with stable coronary artery disease, Eur. Heart J. 27 (2006) 1166–1173. [22] G.M. Tan, Y.Y. Lam, B.P. Yan, Novel platelet ADP P2Y12 inhibitors in the treatment of acute coronary syndrome, Cardiovasc. Ther. 30 (2012) 167–173. [23] K.P. Bliden, U.S. Tantry, R.F. Storey, Y.H. Jeong, M. Gesheff, C. Wei, P.A. Gurbel, The effect of ticagrelor versus clopidogrel on high on-treatment platelet reactivity: combined analysis of the ONSET/OFFSET and RESPOND studies, Am. Heart J. 162 (2011) 160–165. [24] N.B. Norgard, J.J. Dinicolantonio, Clopidogrel, prasugrel, or ticagrelor? A practical guide to use of antiplatelet agents in patients with acute coronary syndromes, Postgrad. Med. 125 (2013) 91–102.
Contents lists available at ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
Assessment of P2Y12 inhibitor usage and switching in acute coronary syndrome patients undergoing percutaneous coronary revascularization☆,☆☆ Mrudula Kudaravalli d, Andrew D. Althouse a, Oscar C. Marroquin a, Sameer J. Khandhar c, Michael S. Sharbaugh a, Catalin Toma a, A.J. Conrad Smith a, John T. Schindler a, Joon S. Lee a, Suresh R. Mulukutla a,b,⁎ a
Heart and Vascular Institute, Department of Medicine, University of Pittsburgh, Pittsburgh, PA, United States Division of Cardiology, Veterans Affairs Hospital, Pittsburgh, PA, United States c Division of Cardiology, University of Pennsylvania, Philadelphia, PA, United States d Department of Medicine, Allegheny General Hospital, Pittsburgh, PA, United States b
a r t i c l e
i n f o
Article history: Received 13 May 2016 Received in revised form 5 August 2016 Accepted 6 August 2016 Available online 08 August 2016 Keywords: Thienopyridine P2Y12 inhibitor Percutaneous coronary intervention Coronary disease
a b s t r a c t Background: Dual antiplatelet therapy is recommended for patients with acute coronary syndrome (ACS) that undergo percutaneous coronary intervention (PCI). However, the effect of switching P2Y12 inhibitors between the loading dose and therapy after discharge is not well described. Methods: This post-hoc analysis of a prospectively collected registry included 3219 consecutive ACS patients who underwent PCI. Patients were categorized into four groups: clopidogrel at load and discharge (C–C), loading dose of clopidogrel and discharged on prasugrel/ticagrelor (C–PT), loading dose of prasugrel/ticagrelor and discharged on clopidogrel (PT–C), and prasugrel/ticagrelor at load and discharge (PT–PT). Results: While 77.6% of patients received the C–C treatment regimen and 13.6% received the PT–PT strategy, the strategy of P2Y12 switching was fairly common with 6.2% in the PT–C group and 2.6% in the C–PT group. While C– C was the most common treatment regimen, PT–C and PT–PT were more commonly used in STEMI patients than in NSTEMI or unstable angina patients. A significantly lower unadjusted incidence of the composite outcome (death, MI, and repeat revascularization) was appreciated in both the PT–C (1.0%) and PT–PT (2.3%) groups than the C–C group (4.0%). Propensity-score matched analysis still showed significantly reduced risk (HR = 0.22, 95% CI 0.05–0.93, p = 0.04) in the PT–C group vs. a matched group of C–C controls. Conclusions: The strategy of utilizing a newer P2Y12 inhibitor and then switching to clopidogrel in ACS patients following PCI is used with some frequency in routine clinical practice and further studies should evaluate the safety and efficacy of such a strategy. Published by Elsevier Ireland Ltd.
1. Introduction With the advent of clopidogrel in 1997, dual anti-platelet therapy (DAPT) with P2Y12 inhibitor and aspirin became the standard of care in patients with acute coronary syndrome (ACS) undergoing percutaneous
☆ Statement of authorship: Each author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation. ☆☆ Funding: This research was funded, in part, by funds from an Abbott Vascular grant and from St. Jude Medical. The remaining funding was from the University of Pittsburgh Medical Center. ⁎ Corresponding author at: University of Pittsburgh, VA Pittsburgh Healthcare System, Division of Cardiology, University Drive C, Cardiology/Cath Lab 111C-U, Pittsburgh, PA 15240, United States. E-mail address: [email protected] (S.R. Mulukutla).
http://dx.doi.org/10.1016/j.ijcard.2016.08.144 0167-5273/Published by Elsevier Ireland Ltd.
coronary intervention (PCI) and stenting [1–3]. However, patients continued to have adverse events, in part due to variability in response to clopidogrel secondary to patient phenotype, genetic polymorphism and drug–drug interactions [4–6]. More recently, novel P2Y12 inhibitors (prasugrel and ticagrelor) were introduced to the market. Subsequently, randomized clinical trials conducting head-to-head comparison between these novel agents and clopidogrel showed improved outcomes with novel agents over DAPT with clopidogrel [7,8]. However, the novel P2Y12 inhibitors are not without limitations. Ticagrelor is known to cause dyspnea, increases in uric acid and creatinine levels and ventricular pauses [9], while prasugrel increases the risk of major bleeding and is contraindicated in numerous patient populations [10]. Current AHA guidelines do not recommend one specific P2Y12 inhibitor, and therefore the choice of P2Y12 agents is at the discretion of the physician [2].
M. Kudaravalli et al. / International Journal of Cardiology 223 (2016) 854–859
With this gray area in recommendations, it is possible that some patients may receive one P2Y12 inhibitor for their loading dose, but be switched to another for treatment after hospital discharge because of the different pharmacologic aspects of the newer agents, the lower cost of clopidogrel, or compliance issues [11,12]. While a few studies have addressed the impact of switching between P2Y12 agents, most focus primarily on the level of platelet inhibition achieved [13–15]. There is little in the literature to date to document the frequency of P2Y12 switching and its potential impact upon clinical outcomes, but there are some potential concerns when switching related to gaps in effective platelet inhibition or excessive platelet inhibition resulting in bleeding [16,17]. Therefore, we sought to (1) define the frequency of switching P2Y12 agents after PCI in routine clinical practice, (2) to describe the differences in populations where different DAPT strategies were used, and (3) to attempt to evaluate the efficacy and safety of this strategy. 2. Methods Patients at four University of Pittsburgh Medical Center (UPMC) hospitals with ACS undergoing PCI and stenting were consecutively enrolled starting in August 2011 as part of an IRB-approved hospital-based registry carried on through May 2014. Of the 7009 patients undergoing PCI in this time, 4617 patients presented with ACS (STEMI, NSTEMI and Unstable angina). Patients that did not receive at least one drug-eluting stent or died in hospital were excluded from the study. Our analytic cohort was comprised of the 3219 patients that received a drug-eluting stent who were discharged alive with a P2Y12 inhibitor (Fig. 1). Information regarding P2Y12 inhibitors given procedurally and at discharge was obtained using the hospital-based registry. Patients are divided into four groups for analysis: those receiving a clopidogrel loading dose (600 mg) and discharged on clopidogrel (C–C), those receiving a loading dose of clopidogrel (600 mg) and discharged on prasugrel/ ticagrelor (C–PT), those receiving a loading dose of prasugrel/ticagrelor and discharged on clopidogrel (PT–C), and those treated with prasugrel/ticagrelor at load and discharge (PT–PT). Baseline data (demographics, clinical characteristics, cardiac presentation, and
855
procedural details) were obtained from the registry, as were patient outcomes in the first 90 days. With a limited number of events in the short follow-up, the primary outcome used to measure the impact of switching between P2Y12 inhibitors is a composite of death, MI, and repeat PCI/revascularization.
2.1. Statistical methods Continuous variables are presented as median (range), while categorical variables are presented as n (%). Analysis of variance (ANOVA) was used to compare the distribution of continuous variables between the four treatment groups; chi-squared tests were used to compare the distribution of categorical variables (or Fisher's Exact Test in the case of small sample sizes). We present overall “global” comparisons of all four treatment groups as well as pairwise comparisons for each of the C–PT, PT–C, and PT–PT groups compared against the C–C group. Outcomes are presented as n (%) occurring within the first 90 days. Kaplan–Meier analysis is used to present freedom from the composite outcome in patients receiving each treatment strategy. However, due to a) the large number of baseline variables that differ significantly between treatment groups and b) the impracticality of multivariable analyses adjusting for large numbers of confounding variables with a low number of events, we present propensity-matched analyses to allow proper comparison between groups. Briefly, all baseline variables were considered candidate variables for a multivariate logistic regression to create propensity scores for receiving the C–PT treatment regimen (versus C–C treatment). The propensity scores were used to match patients that received C–PT to those getting C–C on a set of prespecified demographic and clinical characteristics. The final multivariate logistic regression used to create propensity scores included the following variables: age, gender, race, diabetes, hypertension, dyslipidemia, prior HF, prior PAD, prior PCI, prior CABG, prior MI, CAD Presentation, and number of diseased vessels. After calculating the propensity scores, patients receiving C–PT treatment were randomly sequenced and the C–C patients were matched 4:1 (without replacement) to each C–PT patient with using a caliper distance of 0.10 for matching. Baseline comparisons of between-group differences are repeated in the propensity-matched cohort to ensure adequate balance in risk profile. Cox proportional-hazards models were then used to estimate the hazard ratio (HR) for the composite outcome in the C–PT group vs. the C–C group. This process was repeated for both C–PT group and the PT–C group to evaluate the effects of switching P2Y12 inhibitor against comparable control groups of patients treated with C–C strategy. All statistical analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC).
Fig. 1. Flow chart representing analytic cohort used for analysis.
856
M. Kudaravalli et al. / International Journal of Cardiology 223 (2016) 854–859
3. Results The relative frequency of the four treatment groups is plotted over time in Fig. 2. The large majority of patients through the entire duration of the study received the C–C treatment regimen (2500/ 3219 = 77.6% overall), while very few patients (83/3219 = 2.6% overall) were given the C–PT treatment regimen. Frequency of the PT–C group (199/3219 = 6.2% overall) steadily climbed until early 2014, while the PT–PT group (437/3219 = 13.6% overall) accounted for a relatively steady proportion across most of the study. Demographics, clinical characteristics, and cardiac presentation of the each treatment group is summarized Table 1. As expected, there were many significant differences between groups in baseline risk factors, especially when comparing the PT–C and PT–PT groups against the C–C group. In particular, patients in both the PT–C and PT–PT groups tended to be younger, had fewer comorbid conditions, had less history of cardiovascular intervention, and were more likely to have singlevessel disease than patients in the C–C group. While C–C was the most common treatment regimen for all levels of CAD presentation, it is worth noting that PT–C and PT–PT was more commonly used in STEMI patients than in NSTEMI or unstable angina patients (Fig. 3). Outcome counts for the entire cohort are summarized in Table 2. We see a lower unadjusted incidence of the composite outcome (death, MI, and repeat revascularization) in both the PT–C (1.0%) and PT–PT (2.3%) groups than the C–C group (4.0%) (Fig. 4). However, given the baseline differences between groups, the propensity-score matched analysis is more appropriate for evaluating the strategy of switching P2Y12 inhibitors against an appropriate control group. Characteristics of the respective propensity-score matched cohorts are shown in Table 3. The matching algorithm successfully created cohorts with four C–C controls per case for each of the “switch” groups (both C–PT and PT–C) with good balance between groups on a plurality of baseline characteristics. Slight differences remained in prior CABG and CAD presentation; for the C–PT vs. C–C comparison, this yields higher prevalence of prior CABG, but a lower prevalence of STEMI patients in the C–C control group vs. the C–PT group. For the PT–C vs. C–C comparison, this again returned a higher percentage of prior CABG, but lower prevalence of STEMI patients in the C–C control group vs. the PT–C group. The risk of composite outcome in the C–PT group was not much different than their matched C–C controls (HR = 1.34, 95% CI 0.43–4.16, p = 0.612). However, outcomes in the PT–C group were quite good
Fig. 2. P2Y12 inhibitor use during & after cardiac catheterization, 2011–2014 (by quarter). Abbreviations: C–C = clopidogrel load, clopidogrel discharge; C–PT = clopidogrel load, prasugrel/ticagrelor discharge; PT–C = prasugrel/ticagrelor load, clopidogrel discharge; PT–PT = prasugrel/ticagrelor load, prasugrel/ticagrelor discharge.
relative to their matched C–C controls, with a significantly lower incidence of the composite outcome (HR = 0.22, 95% CI 0.05–0.93, p = 0.04) despite the small sample size. While this hazard ratio is derived from a small number of events and therefore may not be a truly representative estimate of the absolute treatment effect, it is striking enough to warrant attention as a potentially viable treatment strategy that could be explored further. 4. Discussion Using data from a hospital-based registry, we evaluated the prevalence of P2Y12 switching in routine clinical practice at a large healthcare institution. In addition, we evaluated the incidence of clinical outcomes in PCI patients receiving stents, broken down by the P2Y12 inhibitor(s) used for loading dose and at discharge, though these results are only hypothesis generating at this point. Despite the superiority of prasugrel and ticagrelor in randomized clinical trials, we observed that the majority of patients in our clinical practice continued to receive clopidogrel, while a substantial minority received a different P2Y12 inhibitor for their loading dose than at discharge. Over the course of our study, 23% of patients got one of the newer agents at some point in their course of therapy, and this is consistent with other studies [18]. Data from the National Cardiovascular Data Registry (NCDR) shows an increase in prasugrel use from 3% to 18% between 2009 and 2012 [19]. Our study also revealed that physicians were more likely to switch P2Y12 agents in STEMI patients, as they tend to have a higher thrombotic burden compared to the unstable angina population, who primarily received clopidogrel. However, even with the higher prevalence of STEMI patients in the PT–C arms, the clinical outcomes for these patients were still quite good. Despite the established superiority of prasugrel and ticagrelor, ongoing physician preference for clopidogrel could be explained by the availability of generic clopidogrel allowing for lower cost to the patients, as well as increased compliance with once a day dosing of clopidogrel [20]. When comparing the groups that switched P2Y12 agents (C–PT and PT–C) to patients that remained on clopidogrel throughout (C–C), the group loaded with prasugrel/ticagrelor (PT–C) experienced particularly good outcomes. However, our analyses cannot make any firm conclusions regarding these outcomes given that this was not a randomized study, and it is impossible to account for all potential factors that would account for why P2Y12 switching would occur. From the standpoint of hypothesis generation, our data suggest improved outcomes when one of the newer agents is given for the loading dose and highlight the potential importance of loading acute coronary syndrome patients with one of the newer agents, prasugrel or ticagrelor, and that this may be the most critical aspect in determining short-term, 3-month outcomes after ACS. In both TRITON-TIMI 38 and the PLATO studies, the curves for the primary endpoint separate early [7,8]. For instance, in TRITON-TIMI 38, at 90 days, 5.6% of the clopidogrel group met the primary endpoint compared with 4.7% in the prasugrel group. Similarly, in our study, the composite endpoint was met in 4.0% of patients in the C–C group and 2.2% in any of the groups that received any of the newer agents for their loading dose. Mechanistically, our observations of improved outcomes in patients after loading with prasugrel or ticagrelor may be explained by the rapid onset of action of the newer agents, in comparison to clopidogrel, and the fact that the newer agents are able to provide effective platelet inhibition after a very short duration after administration of the loading dose [21–23]. Although both the C–PT group and the PT–C group did experience a slightly higher incidence of bleeding events than the C–C group postdischarge, the absolute number of events was very low (1/83 = 1.2% in the C–PT group and 4/199 = 2.0% in the PT–C group), and these bleeding rates were not statistically significant. Hence, it appears that the decision to switch P2Y12 medication from load to discharge does not increase bleeding risk. However, again, we only evaluated 3month outcomes, and prior studies have suggested that prasugrel and
M. Kudaravalli et al. / International Journal of Cardiology 223 (2016) 854–859
857
Table 1 Patient demographics, clinical characteristics, and cardiac presentation.
# patients Age at admission Male, % Race White Black Other BMI Diabetes Hypertension Dyslipidemia Prior HF Prior PAD Prior PCI Prior CABG Prior MI CAD presentation Unstable angina NSTEMI STEMI Postcardiac arresta (prior 24 h) Postcardiogenic shock (prior 24 h)a Number of diseased vessels 0 1 2 3
p-Value1
C–C
C–PT
2500 67.0 (19.0–96.0) 1658 (66.3%)
83 59.0 (31.0–90.0) 59 (71.1%)
2304 (92.2%) 178 (7.1%) 14 (0.6%) 29.3 (13.9–61.1) 985 (39.4%) 2144 (85.8%) 1983 (79.3%) 337 (13.5%) 347 (13.9%) 1182 (47.3%) 607 (24.3%) 949 (38.0%)
73 (88.0%) 9 (10.8%) 1 (1.2%) 30.5 (21.0–65.0) 36 (43.4%) 61 (73.5%) 58 (69.9%) 7 (8.4%) 7 (8.4%) 35 (42.2%) 10 (12.0%) 30 (36.1%)
1445 (57.8%) 744 (29.8%) 311 (12.4%) 37 (1.5%) 26 (1.0%)
30 (36.1%) 30 (36.1%) 23 (27.7%) 3 (3.6%) 2 (2.4%)
14 (0.6%) 799 (32.0%) 997 (39.9%) 690 (27.6%)
1 (1.2%) 28 (33.7%) 33 (39.8%) 21 (25.3%)
b.001 0.365 0.322
0.042 0.466 0.002 0.037 0.183 0.156 0.359 0.010 0.737 b.001
0.121 0.236
PT–C 199 64.0 (23.0–96.0) 136 (68.3%) 193 (97.0%) 3 (1.5%) 2 (1.0%) 28.9 (14.7–52.7) 49 (24.6%) 142 (71.4%) 131 (65.8%) 15 (7.5%) 9 (4.5%) 54 (27.1%) 13 (6.5%) 48 (24.1%) 44 (22.1%) 55 (27.6%) 100 (50.3%) 6 (3.0%) 2 (1.0%)
0.847
p-Value2
PT–PT 437 59.0 (30.0–92.0) 317 (72.5%)
b.001 0.561 0.008
411 (94.1%) 25 (5.7%) 0 (0.0%) 30.0 (15.1–60.1) 135 (30.9%) 319 (73.0%) 305 (69.8%) 31 (7.1%) 30 (6.9%) 154 (35.2%) 73 (16.7%) 129 (29.5%)
0.244 b.001 b.001 b.001 0.017 b.001 b.001 b.001 b.001 b.001
184 (42.1%) 144 (33.0%) 109 (24.9%) 9 (2.1%) 1 (0.2%)
0.096 0.963 0.005
1 (0.5%) 88 (44.2%) 67 (33.7%) 43 (21.6%)
p-Value3
p-Value4
b.001 0.011 0.161
b.001 0.067 0.013
0.013 b.001 b.001 b.001 b.001 b.001 b.001 b.001 b.001 b.001
0.008 b.001 b.001 b.001 b.001 b.001 b.001 b.001 b.001 b.001
0.368 0.101
0.178 0.217
b.001
b.001
2 (0.5%) 190 (43.5%) 168 (38.4%) 77 (17.6%)
Continuous variables reported as median (range); comparisons performed using Kruskal–Wallis test. Categorical variables reported as n (%); comparisons performed using chi-squared tests. Abbreviations: C–C = clopidogrel load, clopidogrel discharge; C–PT = clopidogrel load, prasugrel/ticagrelor discharge; PT–C = prasugrel/ticagrelor load, clopidogrel discharge; PT–PT = prasugrel/ticagrelor load, prasugrel/ticagrelor discharge; HF = heart failure; PAD = peripheral artery disease; PCI = percutaneous coronary intervention; CABG = coronary artery bypass graft; MI = myocardial infarction; NSTEMI = non-ST elevation myocardial infarction; STEMI = ST-elevation myocardial infarction. a Postcardiac arrest and postcardiogenic shock refer to patients who suffered one of these events within the 24 h prior to their PCI procedure. 1 p-Value comparing C–PT group to C–C group. 2 p-Value comparing PT–C group to C–C group. 3 p-Value comparing PT–PT group to C–C group. 4 p-Value for test across all 4 groups.
ticagrelor may be associated with higher bleeding rates, at least in specific patient populations [24]. While the increased bleeding risk with the newer agents may exist only in specific patient subsets, our findings suggest that there may be an attractive alternative by loading ACS patients with a newer agent and discharging them on clopidogrel, thereby giving patients the benefit of a more potent antiplatelet agent
for the load and an agent that may be associated with lower risk for bleeding on discharge. Mechanistically, our observations of improved outcomes in patients after loading with prasugrel or ticagrelor may be explained by the rapid onset of action of the newer agents, in comparison to clopidogrel, and the fact that the newer agents are able to provide effective platelet inhibition after a very short duration after administration of the loading dose.
C-C
Table 2 90-day patient outcomes, by treatment group.
C-C
C-C
PT-C
PT-PT
PT-PT PT-PT C-PT
PT-C PT-C
C-PT
C-PT
Fig. 3. Bar chart of drug groups by CAD presentation. Abbreviations: C–C = clopidogrel load, clopidogrel discharge; C–PT = clopidogrel load, prasugrel/ticagrelor discharge; PT– C = prasugrel/ticagrelor load, clopidogrel discharge; PT–PT = prasugrel/ticagrelor load, prasugrel/ticagrelor discharge.
C–C
C–PT
PT–C
PT–PT
# patients
2500
83
199
437
Primary outcomes Compositea Death MI Repeat revascularization PCI CABG
99 (4.0%) 53 (2.1%) 18 (0.7%) 55 (2.2%) 51 (2.0%) 4 (0.2%)
4 (4.8%) 0 (0.0%) 0 (0.0%) 4 (4.8%) 3 (3.6%) 1 (1.2%)
2 (1.0%) 2 (1.0%) 0 (0.0%) 2 (1.0%) 1 (0.5%) 1 (0.5%)
10 (2.3%) 3 (0.7%) 0 (0.0%) 8 (1.8%) 7 (1.6%) 1 (0.2%)
Other outcomes Stroke Bleeding Stent thrombosis
4 (0.2%) 28 (1.1%) 0 (0.0%)
0 (0.0%) 1 (1.2%) 0 (0.0%)
0 (0.0%) 4 (2.0%) 2 (1.0%)
1 (0.2%) 1 (0.2%) 0 (0.0%)
All statistics reported as n (%); unadjusted comparisons of event rates performed using chi-squared tests. Abbreviations: C–C = clopidogrel load, clopidogrel discharge; C–PT = clopidogrel load, prasugrel/ticagrelor discharge; PT–C = prasugrel/ticagrelor load, clopidogrel discharge; PT–PT = prasugrel/ticagrelor load, prasugrel/ticagrelor discharge; MI = myocardial infarction; PCI = percutaneous coronary intervention; CABG = coronary artery bypass graft. a Composite endpoint includes first incident of death, MI, or repeat revascularization.
858
M. Kudaravalli et al. / International Journal of Cardiology 223 (2016) 854–859
Fig. 4. Freedom from death, MI, and repeat revascularization in first 90 days after stenting. Abbreviations: C–C = clopidogrel load, clopidogrel discharge; C–PT = clopidogrel load, prasugrel/ticagrelor discharge; PT–C = prasugrel/ticagrelor load, clopidogrel discharge; PT–PT = prasugrel/ticagrelor load, prasugrel/ticagrelor discharge.
If our results suggesting that delivering a load with prasugrel/ ticagrelor and subsequent treatment with clopidogrel upon discharge results in better clinical outcomes than C–C treatment and equivalent outcomes to PT–PT treatment, it is possible that a “combination strategy” of loading with a dose of prasugrel/ticagrelor and discharging with a prescription for clopidogrel could achieve the optimal balance between favorable clinical outcomes and reduced patient costs (relative to giving a full course of treatment with prasugrel/ticagrelor). However, this needs to be tested within a randomized clinical trial setting given the noted limitations of this analysis.
5. Limitations The hospital-based registry only had complete follow-up for all patients through 90 days after discharge, resulting in a relatively low number of events during follow-up. Therefore, further analysis with greater follow-up time is required to confirm the outcomes for patients treated with PT–C strategy relative to all other groups. A randomized controlled trial should be considered confirm these results, especially because the prevalence of P2Y12 switching is not infrequent and is not supported by clinical data as yet. The significant baseline differences
Table 3 Patient demographics, clinical characteristics, and cardiac presentation: propensity-score matched cohorts.
# patients Age at admission Male, % Race White Black Other BMI Diabetes Hypertension Dyslipidemia Prior HF Prior PAD Prior PCI Prior CABG Prior MI CAD presentation Unstable angina NSTEMI STEMI Postcardiac arrest (prior 24 h) Postcardiogenic shock (prior 24 h) Number of diseased vessels 0 1 2 3
C–PT
Matched C–C Cohort A
83 59.0 (31.0–90.0) 59 (71.1%)
332 64.0 (19.0–93.0) 216 (65.1%)
73 (88.0%) 9 (10.8%) 1 (1.2%) 30.5 (21.0–65.0) 36 (43.4%) 61 (73.5%) 58 (69.9%) 7 (8.4%) 7 (8.4%) 35 (42.2%) 10 (12.0%) 30 (36.1%)
302 (91.0%) 27 (8.1%) 3 (0.9%) 29.7 (13.9–61.1) 143 (43.1%) 269 (81.0%) 246 (74.1%) 45 (13.6%) 51 (15.4%) 154 (46.4%) 77 (23.2%) 132 (39.8%)
30 (36.1%) 30 (36.1%) 23 (27.7%) 3 (3.6%) 2 (2.4%)
158 (47.6%) 115 (34.6%) 59 (17.8%) 5 (1.5%) 4 (1.2%)
1 (1.2%) 28 (33.7%) 33 (39.8%) 21 (25.3%)
4 (1.2%) 111 (33.4%) 129 (38.9%) 88 (26.5%)
p-Value1
0.001 0.299 0.707
0.215 0.960 0.128 0.437 0.207 0.103 0.490 0.025 0.546 0.071
0.211 0.410 0.996
PT–C
Matched C–C Cohort B
198 64.0 (23.0–96.0) 135 (68.2%)
792 63.0 (25.0–96.0) 558 (70.5%)
193 (97.5%) 3 (1.5%) 2 (1.0%) 29.0 (14.7–52.7) 49 (24.7%) 142 (71.7%) 130 (65.7%) 15 (7.6%) 9 (4.5%) 54 (27.3%) 13 (6.6%) 48 (24.2%)
751 (94.8%) 39 (4.9%) 2 (0.3%) 29.1 (13.9–61.1) 233 (29.4%) 585 (73.9%) 550 (69.4%) 73 (9.2%) 75 (9.5%) 252 (31.8%) 107 (13.5%) 219 (27.7%)
44 (22.2%) 55 (27.8%) 99 (50.0%) 6 (3.0%) 2 (1.0%)
227 (28.7%) 267 (33.7%) 298 (37.6%) 23 (2.9%) 18 (2.3%)
1 (0.5%) 87 (43.9%) 67 (33.8%) 43 (21.7%)
7 (0.9%) 308 (38.9%) 282 (35.6%) 195 (24.6%)
p-Value2
0.987 0.532 0.035
0.475 0.192 0.540 0.304 0.467 0.026 0.215 0.007 0.333 0.006
0.924 0.258 0.574
Continuous variables reported as median (range); comparisons performed using Kruskal–Wallis test. Categorical variables reported as n (%); comparisons performed using chi-squared tests. Abbreviations: C–C = clopidogrel load, clopidogrel discharge; C–PT = clopidogrel load, prasugrel/ticagrelor discharge; PT–C = prasugrel/ticagrelor load, clopidogrel discharge; PT–PT = prasugrel/ticagrelor load, prasugrel/ticagrelor discharge; HF = heart failure; PAD = peripheral artery disease; PCI = percutaneous coronary intervention; CABG = coronary artery bypass graft; MI = myocardial infarction; NSTEMI = non-ST elevation myocardial infarction; STEMI = ST-elevation myocardial infarction. 1 p-Value comparing C–PT group to C–C group. 2 p-Value comparing PT–C group to C–C group.
M. Kudaravalli et al. / International Journal of Cardiology 223 (2016) 854–859
between groups and limited number of events preclude accurate treatment comparisons in our registry data. Finally, the reasons for switching P2Y12 agents were not recorded. Therefore, we were limited in identifying the reasons encountered in clinical practice by a physician switching P2Y12 agents, which could further explain the observed differences between groups. 6. Conclusion Treatment with a novel P2Y12 inhibitor (prasugrel/ticagrelor) followed by continued DAPT with clopidogrel is a strategy that appears to be commonly employed in routine clinical practice, though, still, the great majority of patients still receive a strategy of exclusively clopidogrel. While our data make comparisons challenging between groups using different strategies, our exploratory analysis suggest that outcomes in the C–PT group were not different compared to their matched C–C controls while outcomes in the PT–C group were better than their matched C–C controls. Further investigation should be conducted to determine the clinical impact of such strategies. Conflict of interest There are no conflicts of interest regarding this manuscript for any author. Acknowledgments The authors gratefully acknowledge the University of Pittsburgh Medical Center (UPMC), Abbott Vascular, and St. Jude Medical. A grant from Abbott Vascular and funds from UPMC helped to support the infrastructure to collect data and to enroll patients into this study. Funds from St. Jude Medical and UPMC helped to support our biostatistics staff to complete the analyses. References [1] S. Yusuf, F. Zhao, S.R. Mehta, S. Chrolavicius, G. Tognoni, K.K. Fox, Effects of clopidogrel in addition to aspirin in patients with acute coronary syndromes without ST-segment elevation, N. Engl. J. Med. 345 (2001) 494–502. (OpenUrl). [2] E.A. Amsterdam, N.K. Wenger, R.G. Brindis, D.E. Casey Jr., T.G. Ganiats, D.R. Holmes Jr., A.S. Jaffe, H. Jneid, R.F. Kelly, M.C. Kontos, G.N. Levine, P.R. Liebson, D. Mukherjee, E.D. Peterson, M.S. Sabatine, R.W. Smalling, S.J. Zieman, ACC/AHA Task Force Members, AHA/ACC guideline for the management of patients with non-STelevation acute coronary syndromes: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines, J. Am. Coll. Cardiol. 64 (24) (2014) e139–e228. [3] Z.M. Chen, L.X. Jiang, Y.P. Chen, J.X. Xie, H.C. Pan, R. Peto, R. Collins, L.S. Liu, COMMIT (ClOpidogrel and Metoprolol in Myocardial Infarction Trial) collaborative group, Addition of clopidogrel to aspirin in 45,852 patients with acute myocardial infarction: randomised placebo-controlled trial, Lancet 366 (2005) 1607–1621. [4] P.A. Gurbel, K.P. Bliden, Durability of platelet inhibition by clopidogrel, Am. J. Cardiol. 91 (2003) 1123–1125. [5] V.L. Serebruany, S.R. Steinhubl, P.B. Berger, A.I. Malinin, D.L. Bhatt, E.J. Topol, Variability in platelet responsiveness to clopidogrel among 544 individuals, J. Am. Coll. Cardiol. 45 (2005) 246–251. [6] G. Campo, G. Parrinello, P. Ferraresi, B. Lunghi, M. Tebaldi, M. Miccoli, J. Marchesini, F. Bernardi, R. Ferrari, M. Valgimigli, Prospective evaluation of on-clopidogrel platelet reactivity over time in patientstreated with percutaneous coronary intervention relationship with gene polymorphisms and clinical outcome, J. Am. Coll. Cardiol. 57 (2011) 2474–2483. [7] S.D. Wiviott, E. Braunwald, C.H. McCabe, G. Montalescot, W. Ruzyllo, S. Gottlieb, F.J. Neumann, D. Ardissino, S. De Servi, S.A. Murphy, J. Riesmeyer, G. Weerakkody, C.M. Gibson, E.M. Antman, TRITON-TIMI 38 Investigators, Prasugrel versus clopidogrel in patients with acute coronary syndromes, N. Engl. J. Med. 357 (2007) 2001–2015.
859
[8] L. Wallentin, R.C. Becker, A. Budaj, C.P. Cannon, H. Emanuelsson, C. Held, J. Horrow, S. Husted, S. James, H. Katus, K.W. Mahaffey, B.M. Scirica, A. Skene, P.G. Steg, R.F. Storey, R.A. Harrington, PLATO Investigators, A. Freij, M. Thorsén, Ticagrelor versus clopidogrel in patients with acute coronary syndromes, N. Engl. J. Med. 361 (2009) 1045–1057. [9] B.M. Scirica, C.P. Cannon, H. Emanuelsson, E.L. Michelson, R.A. Harrington, S. Husted, S. James, H. Katus, P. Pais, D. Raev, J. Spinar, P.G. Steg, R.F. Storey, L. Wallentin, PLATO Investigators, The incidence of bradyarrhythmias and clinical bradyarrhythmic events in patients with acute coronary syndromes treated with ticagrelor or clopidogrel in the PLATO (platelet inhibition and patient outcomes) trial: results of the continuous electrocardiographic assessment substudy, J. Am. Coll. Cardiol. 57 (19) (2011) 1908–1916. [10] S.D. Wiviott, E.M. Antman, E. Braunwald, Prasugrel, Circulation 122 (2010) 394–403. [11] A.D. Michelson, A.L. Frelinger III, E. Braunwald, W.E. Downey, D.J. Angiolillo, N.P. Xenopoulos, J.A. Jakubowski, Y. Li, S.A. Murphy, J. Qin, C.H. McCabe, E.M. Antman, S.D. Wiviott, TRITON-TIMI 38 Investigators, Pharmacodynamic assessment of platelet inhibition by prasugrel vs. clopidogrel in the TRITON-TIMI 38 trial, Eur. Heart J. 30 (2009) 1753–1763. [12] R.F. Storey, D.J. Angiolillo, S.B. Patil, B. Desai, R. Ecob, S. Husted, H. Emanuelsson, C.P. Cannon, R.C. Becker, L. Wallentin, Inhibitory effects of ticagrelor compared with clopidogrel on platelet function in patients with acute coronary syndromes, J. Am. Coll. Cardiol. 56 (2010) 1456–1462. [13] D. Alexopoulos, I. Xanthopoulou, S. Deftereos, G. Sitafidis, I. Kanakakis, M. Hamilos, C. Angelidis, S. Petousis, D. Stakos, H. Parissis, M. Vavouranakis, P. Davlouros, J. Goudevenos, C. Stefanadis, In-hospital switching of oral P2Y12 inhibitor treatment in patients with acute coronary syndrome undergoing percutaneous coronary intervention: prevalence, predictors and short-term outcome, Am. Heart J. 167 (1) (Jan 2014) 68–76. [14] R.F. Storey, D.J. Angiolillo, S.B. Patil, B. Desai, R. Ecob, S. Husted, H. Emanuelsson, C.P. Cannon, R.C. Becker, L. Wallentin, Optimizing P2Y-receptor inhibition in acute coronary syndrome patients based on platelet function testing: impact of prasugrel and high-dose clopidogrel, J. Am. Coll. Cardiol. 63 (11) (2014) 1061–1070. [15] D.J. Angiolillo, N. Curzen, P. Gurbel, P. Vaitkus, F. Lipkin, W. Li, J.A. Jakubowski, M. Zettler, M.B. Effron, D. Trenk, Pharmacodynamic evaluation of switching from ticagrelor to prasugrel in patients with stable coronary artery disease: results of the SWAP-2 Study (Switching Anti Platelet-2), J. Am. Coll. Cardiol. 63 (2014) 1500–1509. [16] A. Bagai, E.D. Peterson, E. Honeycutt, M.B. Effron, D.J. Cohen, S.G. Goodman, K.J. Anstrom, A. Gupa, J.C. Messenger, T.Y. Wang, In-hospital switching between adenosine diphosphate receptor inhibitors in patients with acute myocardial infarction treated with percutaneous coronary intervention: insights into contemporary practice from the TRANSLATE-ACS study, Eur. Heart J. Acute Cardiovasc. Care 4 (2015) 499–508. [17] A. Bagai, Y. Wang, T.Y. Wang, J.P. Curtis, H.S. Gurm, B. Shah, A.N. Cheema, E.D. Peterson, J.F. Saucedo, C.B. Granger, M.T. Roe, D.L. Bhatt, R.L. McNamara, K.P. Alexander, In-hospital switching between clopidogrel and prasugrel among patients with acute myocardial infarction treated with percutaneous coronary intervention: insights into contemporary practice from the national cardiovascular data registry, Circ. Cardiovasc. Interv. 7 (2014) 585–593. [18] A. Sandhu, M. Seth, S. Dixon, D. Share, D. Wohns, T. Lalonde, M. Moscucci, A.L. Riba, M. Grossman, H.S. Gurm, Contemporary use of prasugrel in clinical practice: insights from the Blue Cross Blue Shield of Michigan Cardiovascular Consortium, Circ. Cardiovasc. Qual. Outcomes 6 (3) (2013) 293–298. [19] M.W. Sherwood, S.D. Wiviott, S.A. Peng, M.T. Roe, J. Delemos, E.D. Peterson, T.Y. Wang, Early clopidogrel versus prasugrel use among contemporary STEMI and NSTEMI patients in the US: insights from the National Cardiovascular Data Registry, J. Am. Heart Assoc. 3 (2014). [20] D.S. Kazi, M.A. Hlatky, A delicate balance: the cost effectiveness of new antiplatelet agents, J. Am. Coll. Cardiol. 65 (2015) 477–479. [21] T. Jernberg, C.D. Payne, K.J. Winters, C. Darstein, J.T. Brandt, J.A. Jakubowski, H. Naganuma, A. Siegbahn, L. Wallentin, Prasugrel achieves greater inhibition of platelet aggregation and a lower rate of non-responders compared with clopidogrel in aspirin-treated patients with stable coronary artery disease, Eur. Heart J. 27 (2006) 1166–1173. [22] G.M. Tan, Y.Y. Lam, B.P. Yan, Novel platelet ADP P2Y12 inhibitors in the treatment of acute coronary syndrome, Cardiovasc. Ther. 30 (2012) 167–173. [23] K.P. Bliden, U.S. Tantry, R.F. Storey, Y.H. Jeong, M. Gesheff, C. Wei, P.A. Gurbel, The effect of ticagrelor versus clopidogrel on high on-treatment platelet reactivity: combined analysis of the ONSET/OFFSET and RESPOND studies, Am. Heart J. 162 (2011) 160–165. [24] N.B. Norgard, J.J. Dinicolantonio, Clopidogrel, prasugrel, or ticagrelor? A practical guide to use of antiplatelet agents in patients with acute coronary syndromes, Postgrad. Med. 125 (2013) 91–102.