Ninety-Day Readmission and Long-Term Mortality in

Medicare Patients (≥65 Years) Treated With Ticagrelor

Versus Prasugrel After Percutaneous Coronary
Intervention (from the Blue Cross Blue Shield of
Michigan Cardiovascular Consortium)
Chris Song, BAa, Devraj Sukul, MDb,*, Milan Seth, MSb, James M. Dupree, MD, MPHc,
Akshay Khandelwal, MDd, Simon R. Dixon, MBChBe, David Wohns, MDf, Thomas LaLonde, MDg,
and Hitinder S. Gurm, MBBSb,h

Ticagrelor and prasugrel were found to be superior to clopidogrel for the treatment of acute
coronary syndrome (ACS) after percutaneous coronary intervention (PCI); however, the
comparative effectiveness of these 2 drugs remains unknown. We compared postdischarge
outcomes among older patients treated with ticagrelor versus prasugrel after PCI for ACS.
We linked clinical data from PCIs performed in older patients (age ≥65) for ACS at 47 Michi-
gan hospitals to Medicare fee-for-service claims from January 1, 2013, to December 31,
2014, to ascertain rates of 90-day readmission and long-term mortality. We used propen-
sity score matching to adjust for the nonrandom use of ticagrelor and prasugrel at discharge.
Logistic regression and Cox proportional hazards models were used to compare rates of
90-day readmission and long-term mortality, respectively. Patients discharged on ticagrelor
(n = 1,243) were more frequently older, female, had a history of cerebrovascular disease,
and presented with ST- or non-ST-elevation myocardial infarction compared with prasugrel
(n = 1,014). After matching (n = 756 per group), there were no significant differences in
the rates of 90-day readmission (16.7% ticagrelor vs 14.6% prasugrel; adjusted odds ratio
1.15, 95% confidence interval 0.86 to 1.55, p = 0.35) or 1-year mortality (5.4% ticagrelor
vs 3.7% prasugrel; hazard ratio 1.3, 95% confidence interval 0.8 to 2.2, p = 0.31). In con-
clusion, we found no significant differences in the rates of 90-day readmission or long-
term mortality between older patients treated with ticagrelor and patients treated with
prasugrel after PCI for ACS. In the absence of randomized data to the contrary, these 2
treatments appear similarly effective. © 2017 Elsevier Inc. All rights reserved. (Am J Cardiol
2017;120:1926–1932)

Over the past decade, the armamentarium of P2Y12 in- coronary syndrome (ACS).1,2 Due to the dearth of research
hibitors has expanded with the approval of ticagrelor and directly comparing the effectiveness of ticagrelor and prasugrel,
prasugrel. In randomized controlled trials (RCTs), these medi- it remains difficult to make evidence-based decisions regarding
cations were found to be superior to clopidogrel among patients the choice of treatment between them. A recent head-to-
treated with percutaneous coronary intervention (PCI) for acute head RCT comparing clinical end points between ticagrelor

a
Division of Cardiovascular Medicine, Department of Internal Medi- all of the data in the study and take responsibility for the integrity of the
cine, University of Michigan Medical School, Ann Arbor, Michigan; data and the accuracy of the data analysis.
b
Division of Cardiovascular Medicine, Department of Internal Medicine, Uni- Dr. Sukul is supported by the National Institutes of Health T32 post-
versity of Michigan, Ann Arbor, Michigan; cDepartment of Urology, University doctoral research training grant (T32-HL007853) Bethesda, Maryland. This
of Michigan, Ann Arbor, Michigan; dDivision of Cardiology, Henry Ford work was supported by the Blue Cross Blue Shield of Michigan and Blue
Health System, Detroit, Michigan; e Department of Cardiovascular Care Network as part of the Blue Cross Blue Shield of Michigan Value Part-
Medicine, Beaumont Hospital, Royal Oak, Michigan; fDivision of Cardiology, nerships program. The funding source supported data collection at each site
Spectrum Health, Grand Rapids, Michigan; gDepartment of Cardiovascular and funded the data-coordinating center but had no role in study concept,
Medicine, St. John Hospital and Medical Center, Detroit, Michigan; interpretation of findings, or in the preparation, final approval, or decision
and hDivision of Cardiology, Department of Internal Medicine, VA Ann to submit the manuscript.
Arbor Healthcare System, Ann Arbor, Michigan. Manuscript received June Disclaimer: Although Blue Cross Blue Shield of Michigan (BCBSM)
16, 2017; revised manuscript received and accepted August 1, 2017. and BMC2 work collaboratively, the opinions, beliefs, and viewpoints ex-
Authorship declaration: All authors listed meet the authorship criteria pressed by the authors do not necessarily reflect the opinions, beliefs, and
according to the latest guidelines of the International Committee of Medical viewpoints of BCBSM or any of its employees.
Journal Editors, and all authors agree with the manuscript. See page 1931 for disclosure information
Chris Song and Devraj Sukul contributed equally to the drafting and prepa- *Corresponding author: Tel: (734) 936-8214; fax: (734) 615-3326.
ration of the manuscript. Hitinder Gurm and Milan Seth had full access to E-mail address: [email protected] (D. Sukul).

0002-9149/© 2017 Elsevier Inc. All rights reserved. www.ajconline.org

https://doi.org/10.1016/j.amjcard.2017.08.009
 Coronary Artery Disease/Ticagrelor and Prasugrel After PCI 1927

and prasugrel has demonstrated no significant differences in matching algorithm. We required exact matching on the type
primary or secondary end points; however, the trial was of PCI including balloon only, bare metal stent, drug-
terminated early due to futility.3 There is also a paucity of eluting stent, or mixed.
observational research comparing these 2 treatments.4,5 Because Next, we reported adjusted odds ratios (aORs) for the
older patients are at an increased risk of ACS and post-PCI outcome of 90-day readmission using logistic regression
complications,6–8 we sought to assess the comparative effec- models adjusting for potential confounding variables listed
tiveness of these 2 drugs after PCI among Medicare patients in Supplementary Table S1. We used Kaplan-Meier curves
aged ≥65 years presenting with ACS. in the matched cohort to depict differences in cumulative mor-
tality between the 2 treatment groups. Cox proportional hazards
models adjusting for confounding (variables listed in
Methods
Supplementary Table S1) were used to estimate the effect of
We performed a retrospective analysis on data collected treatment choice on long-term mortality.
by the Blue Cross Blue Shield of Michigan Cardiovascular In a sensitivity analysis, we excluded patients with rela-
Consortium (BMC2), a registry of all patients who under- tive or absolute contraindications to prasugrel from the
went PCI in the state of Michigan. This is a prospective, derivation cohort. These contraindications included a history
multicenter, statewide registry of patients who underwent of cerebrovascular disease, age >75 years, or a weight of
PCI at any nonfederal hospital in Michigan. A more de- <60 kg. All primary analyses were repeated in this cohort.
tailed description of the registry, including data collection Per the NCDR CathPCI registry, cerebrovascular disease was
and auditing practices, has been described previously.9,10 defined as a history of stroke, transient ischemic attack, non-
For the present study, we evaluated consecutive patients invasive or invasive carotid testing with >79% occlusion, or
who underwent PCI and were discharged alive between a previous carotid artery surgery or intervention for carotid
January 1, 2013, and December 31, 2014, in the 47 hospi- artery stenosis.12
tals involved in the registry.
To evaluate postdischarge outcomes including 90-day
Results
readmission and long-term mortality, we linked clinical data
from the BMC2 PCI registry with administrative Medicare From January 1, 2013, to December 31, 2014, a total of
claims through indirect matching on multiple variables in- 13,702 Medicare patients who underwent PCI were dis-
cluding admission, discharge, and procedure dates for the charged alive on clopidogrel, ticagrelor, or prasugrel at 47
index hospitalization, patient gender, date of birth, and hos- hospitals in Michigan. Of these patients, 10,261 were dis-
pital and operator National Provider Identifier numbers. charged on clopidogrel and were excluded. After further
Medicare data were made available through a collaboration exclusions including those aged <65 years old, there were
with the Michigan Value Collaborative.11 1,243 patients in the ticagrelor group and 1,014 patients in
Medicare beneficiaries discharged alive after PCI from the prasugrel group (Figure 1). Baseline characteristics of
January 1, 2013, to December 31, 2014, and assigned a the unmatched and matched cohorts are presented in Table 1.
P2Y12 inhibitor (clopidogrel, ticagrelor, or prasugrel) at Finally, after propensity score matching, 756 patients were
discharge served as the initial study cohort to which inclu- included in each group. The absolute standardized differ-
sions and exclusions were applied. First, we included all ences were <10% on all matched variables, suggesting
patients discharged alive on ticagrelor or prasugrel as docu- good covariate balance between the 2 groups (Table 1 and
mented in the BMC2 PCI clinical registry. Next, we excluded Figure 2).
patients aged <65 years, on dialysis before PCI, discharged In the matched cohort, 16.7% (n = 125 of 756) and 14.6%
to a nursing home, or patients who underwent PCI for a (n = 109 of 756) of patients treated with ticagrelor and
non-ACS presentation. We also excluded those patients who prasugrel, respectively, had at least 1 readmission within 90
experienced postprocedural complications including stroke, days of discharge. There was no significant difference in the
major bleeding, or major surgery, as these clinical compli- risk of 90-day readmission between ticagrelor and prasugrel
cations may significantly influence the choice of P2Y12 (aOR 1.15, 95% confidence interval [CI] 0.86 to 1.55,
inhibitor use. p = 0.35). The 1-year mortality rate for patients discharged
The primary outcomes included the rate of readmission on ticagrelor was 5.4% (95% CI 3.3% to 7.4%), whereas that
within 90 days of discharge and long-term mortality. Read- for prasugrel was 3.7% (95% CI 2.2% to 5.2%) (Figure 3).
mission was defined as readmission to the hospital for any There was no significant difference in the hazard rates of mor-
reason within 90 days after discharge from the hospitaliza- tality between patients treated with ticagrelor and patients
tion where PCI occurred. Only the first readmission was treated with prasugrel (hazard ratio 1.3, 95% CI 0.8 to 2.2,
counted per PCI episode. Finally, the patient’s vital status and p = 0.31).
date of death were determined from each patient’s Medi- In a sensitivity analysis, we excluded 653 (52.5%)
care beneficiary file. patients treated with ticagrelor and 257 (25.3%) patients
We used propensity score matching to account for the treated with prasugrel who had absolute or relative
nonrandom use of ticagrelor and prasugrel. A logistic regres- contraindications to prasugrel, which included a history of
sion model was used to estimate propensity scores among cerebrovascular disease, age >75 years, or a weight of <60 kg
patients receiving ticagrelor or prasugrel with demographic (Supplementary Table S2). This left 590 patients treated with
and clinical variables included as predictors (Supplementary ticagrelor and 757 patients treated with prasugrel available
Table S1). Patients receiving ticagrelor were matched in a 1:1 for matching. After propensity score matching, 486 patients
fashion to similar patients receiving prasugrel using a greedy remained in each group. Within this subgroup, there was no
1928 The American Journal of Cardiology (www.ajconline.org)

Figure 1. Study flow diagram.

significant difference in the rate of 90-day readmission between prasugrel is a prodrug that requires CYP450 enzymes to
the 2 treatments (14.4% ticagrelor vs 13.0% prasugrel; aOR become active.13,14 This may factor into important differences
1.14; 95% CI 0.78 to 1.68, p = 0.5). The 1-year mortality rates in platelet inhibition seen in several in vivo pharmacologic
for ticagrelor and prasugrel were 4.8% (95% CI 2.4% to 7.2%) studies.15–17 In addition, given that prasugrel is metabolized
and 2.3% (95% CI 0.9% to 3.7%), respectively, and the hazard by the CYP450 system,14 active metabolite levels could be
ratio of mortality was 1.6 (95% CI 0.8 to 3.5, p = 0.19) affected by common medications or dietary interactions. Also,
(Figure 4). as stent thrombosis is becoming less common due to ad-
vances in stent technologies, pathologic bleeding from
excessive platelet inhibition may contribute more toward mor-
Discussion
bidity and mortality among patients taking these drugs.
In this retrospective observational study examining pa- Indeed, compared with patients treated with clopidogrel,
tients discharged on ticagrelor or prasugrel after undergoing patients with a history of cerebrovascular disease, aged >75
PCI for ACS, we found no significant differences in the rates years, or with a weight of <60 kg treated with prasugrel
of 90-day readmission and 1-year mortality among matched experienced inferior outcomes in TRITON-TIMI 38.2 In
patients. After excluding patients with absolute or relative our unmatched cohort, 257 (25.3%) patients discharged on
contraindications to prasugrel in a sensitivity analysis, we found prasugrel had these co-morbidities. This relatively high rate
no significant differences in the rates of readmission or of patients having a contraindication to prasugrel may par-
mortality. tially be explained by the broader definition of cerebrovascular
There is reason to believe that there may be important dif- disease used in this study, which included patients with
ferences in the effectiveness of these 2 treatments. Ticagrelor carotid artery stenosis (>79% occlusion) or a previous carotid
is active in both its parent and metabolite forms, whereas artery surgery or intervention.12 Thus, some patients may
 Coronary Artery Disease/Ticagrelor and Prasugrel After PCI 1929

Table 1
Baseline characteristics of the unmatched and matched cohorts
Before Matching After Matching
Variable Prasugrel Ticagrelor p-Value Prasugrel Ticagrelor p-Value
(n = 1,014) (n = 1,243) (n = 756) (n = 756)
Age (Years) 70.96 ± 4.64 74.50 ± 6.54 <0.001 71.73 ± 4.83 71.80 ± 5.07 0.783
Women 328/1,014 (32.3%) 517/1,243 (41.6%) <0.001 262/756 (34.7%) 268/756 (35.4%) 0.746
Body Mass Index (kg/m2) 30.39 ± 6.29 29.49 ± 7.14 0.001 30.19 ± 6.27 30.20 ± 6.12 0.991
White 936/1,014 (92.3%) 1,143/1,243 (92.0%) 0.757 695/756 (91.9%) 693/756 (91.7%) 0.851
Black 59/1,014 (5.8%) 74/1,243 (6.0%) 0.892 45/756 (6.0%) 45/756 (6.0%) 1.000
Current/Recent Smoker 180/1,014 (17.8%) 170/1,243 (13.7%) 0.008 125/756 (16.5%) 125/756 (16.5%) 1.000
Hypertension 877/1,014 (86.5%) 1,086/1,242 (87.4%) 0.504 658/756 (87.0%) 655/756 (86.6%) 0.820
Dyslipidemia* 840/1,014 (82.8%) 993/1,242 (80.0%) 0.080 618/756 (81.7%) 621/756 (82.1%) 0.841
Family History of Premature Coronary Artery Disease 138/1,014 (13.6%) 126/1,241 (10.2%) 0.011 93/756 (12.3%) 95/756 (12.6%) 0.876
Prior Myocardial Infarction 298/1,014 (29.4%) 314/1,243 (25.3%) 0.028 206/756 (27.2%) 213/756 (28.2%) 0.688
Prior Heart Failure 114/1,014 (11.2%) 163/1,241 (13.1%) 0.173 85/756 (11.2%) 90/756 (11.9%) 0.688
Prior Valve Surgery/Procedure 11/1,014 (1.1%) 16/1,242 (1.3%) 0.658 <11 11/756 (1.5%) 0.489
Prior Percutaneous Coronary Intervention 458/1,014 (45.2%) 442/1,242 (35.6%) <0.001 309/756 (40.9%) 306/756 (40.5%) 0.875
Cerebrovascular Disease 99/1,014 (9.8%) 202/1,243 (16.3%) <0.001 88/756 (11.6%) 85/756 (11.2%) 0.808
Peripheral Arterial Disease 143/1,014 (14.1%) 166/1,243 (13.4%) 0.607 112/756 (14.8%) 116/756 (15.3%) 0.774
Chronic Lung Disease 203/1,013 (20.0%) 237/1,243 (19.1%) 0.562 157/756 (20.8%) 156/756 (20.6%) 0.949
Diabetes Mellitus 402/1,014 (39.6%) 446/1,243 (35.9%) 0.066 290/756 (38.4%) 295/756 (39.0%) 0.792
Heart Failure within 2 Weeks 88/1,014 (8.7%) 101/1,242 (8.1%) 0.641 54/756 (7.1%) 67/756 (8.9%) 0.218
Left Ventricular Systolic Dysfunction 80/1,014 (7.9%) 77/1,243 (6.2%) 0.115 56/756 (7.4%) 52/756 (6.9%) 0.690
Cardiogenic Shock within 24 Hours <11 17/1,243 (1.4%) 0.288 <11 <11 0.795
Cardiac Arrest within 24 Hours <11 19/1,243 (1.5%) 0.171 <11 <11 1.000
Pre-Procedure Creatinine (mg/dL) 1.04 ± 0.33 1.06 ± 0.36 0.352 1.04 ± 0.33 1.04 ± 0.38 0.633
Pre-Procedure Hemoglobin (g/dL) 13.62 ± 1.64 13.45 ± 1.73 0.015 13.65 ± 1.65 13.64 ± 1.69 0.889
CAD presentation
Unstable angina pectoris 619/1,014 (61.0%) 614/1,243 (49.4%) <0.001 436/756 (57.7%) 435/756 (57.5%) 0.959
Non-ST-Elevation Myocardial Infarction 233/1,014 (23.0%) 358/1,243 (28.8%) 0.002 191/756 (25.3%) 201/756 (26.6%) 0.557
ST-Elevation Myocardial Infarction 162/1,014 (16.0%) 271/1,243 (21.8%) <0.001 129/756 (17.1%) 120/756 (15.9%) 0.533
Procedural characteristics
Intra-Aortic Balloon Pump 11/1,014 (1.1%) 21/1,242 (1.7%) 0.226 <11 11/755 (1.5%) 0.650
Other Mechanical Ventricular Support <11 <11 0.256 <11 <11 0.562
Arterial Access Site: Femoral 778/1,014 (76.7%) 876/1,243 (70.5%) <0.001 560/756 (74.1%) 551/756 (72.9%) 0.600
Arterial Access Site: Radial 233/1,014 (23.0%) 364/1,243 (29.3%) <0.001 193/756 (25.5%) 204/756 (27.0%) 0.520
Balloon only 16/1,014 (1.7%) 28/1,243 (2.8%) 0.286 11/756 (1.5%) 11/755 (1.5%) >0.99
Bare metal stent 88/1,014 (8.7%) 134/1,243 (10.8%) 0.102 64/756 (8.5%) 64/756 (8.5%) >0.99
Drug eluting stent 902/1,014 (89.0%) 1,073/1,243 (86.3%) 0.064 678/756 (90.0%) 678/756 (90.0%) >0.99
Both bare metal and drug eluting stent used <11 <11 0.802 <11 <11 >0.99

* A history of dyslipidemia is defined per the NCDR CathPCI registry as documentation of the following: Total cholesterol > 200 mg/dL; or low-density
lipoprotein > = 130 mg/dL; or high-density lipoprotein <40 mg/dL. For patients with known coronary artery disease, treatment is frequently initiated for low-
density lipoprotein >100 mg/dL, thus qualifying for hypercholesterolemia.12

have been classified as having a history of cerebrovascular patients initially treated with ticagrelor compared with pa-
disease without having a history of stroke or transient isch- tients treated with prasugrel and attributed this to the inferior
emic attack. These patients would not be considered to side-effect profile of ticagrelor.5 In contrast, the increased
have true contraindications to prasugrel. Regardless, this risk of bleeding with prasugrel, as demonstrated in the
finding emphasizes the need to focus quality improvement TRITON-TIMI 38 RCT, may be associated with decreased
efforts at using these antiplatelet agents in the appropriate compliance.2
patients. Finally, costs are an important consideration as studies have
Beyond pharmacologic differences, other factors may play shown decreasing medication compliance with increasing out-
a role in the use of these agents. For instance, ticagrelor’s of-pocket costs.19,20 In upcoming years, these drugs will become
twice daily dosing may impact patient compliance com- generic and thus significantly cheaper. The specific time of
pared with prasugrel’s once daily dosing. A meta-analysis transition to generic medications remains unclear as both
explored this concept and found a significantly lower com- manufacturers have applied for 30-month stays on their
pliance rate in twice daily versus once daily dosing of patents.21,22 However, given the earlier patent expiration date
antihypertensive medications.18 Beigel et al observed a trend for prasugrel, it will likely become generic sooner than
toward lower compliance rate among patients treated with ticagrelor.23,24 As such, if there remains a clinical equipoise
ticagrelor compared with prasugrel.5 They also noted cross- between the 2 drugs, prasugrel may be the more cost-
over rates to a different P2Y12 inhibitor to be higher among effective treatment option.
1930 The American Journal of Cardiology (www.ajconline.org)

Figure 2. Plot of absolute standardized differences before and after matching. Absolute standardized differences before and after matching in PCI patients
discharged on ticagrelor and prasugrel. CAD = coronary artery disease; CVA = cerebrovascular accident; MI = myocardial infarction.

Although we demonstrated no significant differences in

the rates of 90-day readmission and long-term mortality
between ticagrelor and prasugrel, RCTs are essential in guiding
clinical decision making. An ongoing RCT, ISAR-REACT
5, is currently enrolling patients to evaluate the safety and
efficacy of ticagrelor and prasugrel in patients who underwent
PCI for ST-elevation myocardial infarction, non-ST-elevation
myocardial infarction, or unstable angina.25 This study should
provide insights into the comparative efficacy of these 2 drugs.
Until RCTs can produce definitive evidence for which drug
to use, we need to rely on observational studies such as ours
to guide practice.
The present study should be interpreted in the context of
several limitations. First, although we attempted to account
for the nonrandom administration of ticagrelor and prasugrel
through propensity-matching techniques, we were unable
to account for all potential confounders due to the observa-
tional nature of the present study. Second, all hospitals
participating in this registry are actively engaged in state-
wide collaborative quality improvement initiatives. Therefore,
these findings may not be generalizable to hospitals that do Figure 3. Kaplan-Meier curve for cumulative long-term mortality. Kaplan-
not participate in such initiatives.26 Third, we studied older Meier curve demonstrating long-term mortality in propensity-matched patients.
Medicare patients in the state of Michigan, thus potentially The hazard ratio is presented in the upper left corner.
 Coronary Artery Disease/Ticagrelor and Prasugrel After PCI 1931

Skene A, Steg PG, Storey RF, Harrington RA. Ticagrelor versus

clopidogrel in patients with acute coronary syndromes. N Engl J Med
2009;361:1045–1057.
2. Wiviott SD, Braunwald E, McCabe CH, Montalescot G, Ruzyllo W,
Gottlieb S, Neumann F-J, Ardissino D, De Servi S, Murphy SA,
Riesmeyer J, Weerakkody G, Gibson CM, Antman EM. Prasugrel versus
clopidogrel in patients with acute coronary syndromes. N Engl J Med
2007;357:2001–2015.
3. Motovska Z, Hlinomaz O, Miklik R, Hromadka M, Varvarovsky I, Dusek
J, Knot J, Jarkovsky J, Kala P, Rokyta R, Tousek F, Kramarikova P,
Majtan B, Simek S, Branny M, Mrozek J, Cervinka P, Ostransky J,
Widimsky P. Prasugrel versus ticagrelor in patients with acute
myocardial infarction treated with primary percutaneous coronary
intervention: clinical perspective. Circulation 2016;134:1603–
1612.
4. Larmore C, Effron MB, Molife C, DeKoven M, Zhu Y, Lu J, Karkare
S, Lieu HD, Lee WC, Vetrovec GW. “Real-world” comparison of
prasugrel with ticagrelor in patients with acute coronary syndrome treated
with percutaneous coronary intervention in the United States. Cath-
eter Cardiovasc Interv 2016;88:535–544.
5. Beigel R, Iakobishvili Z, Shlomo N, Segev A, Witberg G, Zahger D,
Atar S, Alcalai R, Kapeliovich M, Gottlieb S, Goldenberg I, Asher E,
Matetzky S. Real-world use of novel P2Y12 inhibitors in patients with
acute myocardial infarction: a treatment paradox. Cardiology 2016;
Figure 4. Kaplan-Meier curve for cumulative long-term mortality after ex- 136:21–28.
6. Wu C, Camacho FT, King SB, Walford G, Holmes DR, Stamato NJ,
clusion of patients with contraindications to prasugrel. The Kaplan-Meier
Berger PB, Sharma S, Curtis JP, Venditti FJ, Jacobs AK, Hannan EL.
curve demonstrates long-term mortality in a sensitivity analysis excluding Risk stratification for long-term mortality after percutaneous coronary
patients with a history of cerebrovascular disease, age >75 years, or weight intervention. Circ Cardiovasc Interv 2014;7:80–87.
of <60 kg. The hazard ratio is presented in the upper left corner. 7. Yost GW, Puher SL, Graham J, Scott TD, Skelding KA, Berger PB, Blan-
kenship JC. Readmission in the 30 days after percutaneous coronary
limiting generalizability. However, older patients are at es- intervention. JACC Cardiovasc Interv 2013;6:237–244.
pecially high risk of postdischarge complications including 8. Benjamin EJ, Blaha MJ, Chiuve SE, Cushman M, Das SR,
Deo R, Ferranti SD, de Floyd J, Fornage M, Gillespie C, Isasi CR,
readmission and death6,7; therefore, this is an important popu- Jiménez MC, Jordan LC, Judd SE, Lackland D, Lichtman JH,
lation to study and understand the potential adverse outcomes Lisabeth L, Liu S, Longenecker CT, Mackey RH, Matsushita K,
of these treatments. Finally, as previously noted, we could Mozaffarian D, Mussolino ME, Nasir K, Neumar RW, Palaniappan L,
not account for medication compliance, treatment discon- Pandey DK, Thiagarajan RR, Reeves MJ, Ritchey M, Rodriguez CJ,
Roth GA, Rosamond WD, Sasson C, Towfighi A, Tsao CW,
tinuation, or treatment switching, all of which may have Turner MB, Virani SS, Voeks JH, Willey JZ, Wilkins JT, Wu JH,
important effects on outcomes and warrant further research. Alger HM, Wong SS, Muntner P, Subcommittee on behalf of the
In this retrospective analysis of ticagrelor and prasugrel AHASC and SS. Heart Disease and Stroke Statistics—2017 Update:
treatments in older Medicare patients who underwent PCI for a report from the American Heart Association. Circulation 2017;135:e146–
ACS, there were no significant differences in the rates of 90- e603.
9. Kline-Rogers E, Share D, Bondie D, Rogers B, Karavite D, Kanten S,
day hospital readmission or long-term mortality. With Wren P, Bodurka C, Fisk C, McGinnity J, Wright S, Fox S, Eagle KA,
prasugrel’s potential earlier patent expiration and generic drug Moscucci M. Development of a multicenter interventional cardiology
approval, it may emerge as a more cost-effective agent com- database: the Blue Cross Blue Shield of Michigan Cardiovascular
pared with ticagrelor in the near future. Consortium (BMC2) experience. J Intervent Cardiol 2002;15:387–
392.
10. Moscucci M, Rogers EK, Montoye C, Smith DE, Share D, O’Donnell
Acknowledgment: The authors are indebted to all the study M, Maxwell-Eward A, Meengs WL, Franco ACD, Patel K, McNa-
coordinators, investigators, and patients who participated in mara R, McGinnity JG, Jani SM, Khanal S, Eagle KA. Association of
the Blue Cross Blue Shield of Michigan Cardiovascular Con- a continuous quality improvement initiative with practice and outcome
sortium registry. variations of contemporary percutaneous coronary interventions.
Circulation 2006;113:814–822.
11. Herrel LA, Syrjamaki JD, Linsell SM, Miller DC, Dupree JM. Identi-
Disclosures fying drivers of episode cost variation with radical prostatectomy. Urology
2016;97:105–110.
Hitinder S. Gurm receives research funding from Blue 12. NCDR® CathPCI Registry® v4.4 coder’s data dictionary. Available
Cross Blue Shield of Michigan and the National Institutes of at: https://www.ncdr.com/WebNCDR/docs/public-data-collection
Health and is a consultant for Osprey Medical. None of the -documents/cathpci_v4_codersdictionary_4-4.pdf. Accessed on May 25,
authors have any conflicts directly relevant to this study. 2017.
13. Butler K, Teng R. Pharmacokinetics, pharmacodynamics, safety and tol-
erability of multiple ascending doses of ticagrelor in healthy volunteers.
Supplementary Data Br J Clin Pharmacol 2010;70:65–77.
14. Dobesh PP. Pharmacokinetics and pharmacodynamics of prasugrel, a
Supplementary data associated with this article can be thienopyridine P2Y12 inhibitor. Pharmacother J Hum Pharmacol Drug
found, in the online version, https://doi.org/10.1016/ Ther 2009;29:1089–1102.
j.amjcard.2017.08.009. 15. Franchi F, Rollini F, Aggarwal N, Hu J, Kureti M, Durairaj A, Duarte
VE, Cho JR, Been L, Zenni MM, Bass TA, Angiolillo DJ. Pharmaco-
dynamic comparison of prasugrel versus ticagrelor in patients with type
1. Wallentin L, Becker RC, Budaj A, Cannon CP, Emanuelsson H, Held 2 diabetes mellitus and coronary artery disease: clinical perspective.
C, Horrow J, Husted S, James S, Katus H, Mahaffey KW, Scirica BM, Circulation 2016;134:780–792.
1932 The American Journal of Cardiology (www.ajconline.org)

16. Rollini F, Franchi F, Cho JR, DeGroat C, Bhatti M, Muniz-Lozano A, 22. Kass D. AstraZeneca Hits Mylan With Patent Suit Over Blood Thinner—
Singh K, Ferrante E, Wilson RE, Dunn EC, Zenni MM, Guzman LA, Law360. Available at: https://www.law360.com/articles/722595/
Bass TA, Angiolillo DJ. A head-to-head pharmacodynamic compari- astrazeneca-hits-mylan-with-patent-suit-over-blood-thinner. Accessed
son of prasugrel vs. ticagrelor after switching from clopidogrel in patients on April 6, 2017.
with coronary artery disease: results of a prospective randomized study. 23. United States Food and Drug Administration. Approved Drug Prod-
Eur Heart J 2016;37:2722–2730. ucts with Therapeutic Equivalence Evaluations: Orange Book. Available
17. Laine M, Frère C, Toesca R, Berbis J, Barnay P, Pansieri M, at: https://www.accessdata.fda.gov/scripts/cder/ob/patent_info
Michelet P, Bessereau J, Camilleri E, Ronsin O, Helal O, Paganelli F, .cfm?Product_No=001&Appl_No=022307&Appl_type=N. Accessed on
Dignat-George F, Bonello L. Ticagrelor versus prasugrel in diabetic April 5, 2017.
patients with an acute coronary syndrome. Thromb Haemost 24. United States Food and Drug Administration. Approved Drug Prod-
2014;111:273–278. ucts with Therapeutic Equivalence Evaluations : Orange Book. Available
18. Iskedjian M, Einarson TR, MacKeigan LD, Shear N, Addis A, Mittmann at: https://www.accessdata.fda.gov/scripts/cder/ob/patent_info.cfm?Product
N, Ilersich AL. Relationship between daily dose frequency and adherence _No=002&Appl_No=022433&Appl_type=N. Accessed on April 5, 2017.
to antihypertensive pharmacotherapy: evidence from a meta-analysis. 25. Schulz S, Angiolillo DJ, Antoniucci D, Bernlochner I, Hamm C, Jaitner
Clin Ther 2002;24:302–316. J, Laugwitz K-L, Mayer K, Merzljak B, von Morath T, Neumann F-J,
19. Ye X, Gross CR, Schommer J, Cline R, St Peter WL. Association between Richardt G, Ruf J, Schömig G, Schühlen H, Schunkert H, Kastrati A,
copayment and adherence to statin treatment initiated after coronary heart Investigators IS and ARREA for CT (ISAR-R 5 T). Randomized com-
disease hospitalization: a longitudinal, retrospective, cohort study. parison of ticagrelor versus prasugrel in patients with acute coronary
Clin Ther 2007;29:2748–2757. syndrome and planned invasive strategy—design and rationale of the
20. Curkendall S, Patel V, Gleeson M, Campbell RS, Zagari M, Dubois R. Intracoronary Stenting and Antithrombotic Regimen: Rapid Early Action
Compliance with biologic therapies for rheumatoid arthritis: do patient for Coronary Treatment (ISAR-REACT) 5 trial. J Cardiovasc Transl
out-of-pocket payments matter? Arthritis Care Res 2008;59:1519– Res 2014;7:91–100.
1526. 26. Share DA, Campbell DA, Birkmeyer N, Prager RL, Gurm HS, Moscucci
21. Lerner K. Eli Lilly Sues Par Pharma to Block Generic Effient—Law360. M, Udow-Phillips M, Birkmeyer JD, How A. Regional collaborative of
Available at: https://www.law360.com/articles/505154/eli-lilly-sues hospitals and physicians in Michigan cut costs and improved the quality
-par-pharma-to-block-generic-effient. Accessed on April 6, 2017. of care. Health Aff (Millwood) 2011;30:636–645.