JACC: CARDIOVASCULAR INTERVENTIONS VOL. 12, NO. 1, 2019 ª 2019 THE AUTHORS. PUBLISHED BY ELSEVIER ON BEHALF OF THE AMERICAN COLLEGE OF CARDIOLOGY FOUNDATION. THIS IS AN OPEN ACCESS ARTICLE UNDER THE CC BY-NC-ND LICENSE (http://creativecommons.org/licenses/by-nc-nd/4.0/). Assessment of Platelet REACtivity After Transcatheter Aortic Valve Replacement The REAC-TAVI Trial Victor Alfonso Jimenez Diaz, MD, MPH,a,b Antonio Tello-Montoliu, MD, PHD,c,d Raul Moreno, MD,e Ignacio Cruz Gonzalez, MD, PHD,d,f Jose Antonio Baz Alonso, MD,a Rafael Romaguera, MD,g Eduardo Molina Navarro, MD,h Pablo Juan Salvadores, PHARMA, MPH,b,i Emilio Paredes Galan, MD, PHD,a Antonio De Miguel Castro, MD,a Guillermo Bastos Fernandez, MD,a Alberto Ortiz Saez, MD,a Saleta Fernandez Barbeira, MD,a Sergio Raposeiras Roubin, MD, PHD,a Juan Ocampo Miguez, MD,a Antonio Serra Peñaranda, MD, PHD,j Mariano Valdes Chavarri, MD, PHD,c,d Angel Cequier Fillat, MD, PHD,d,g Francisco Calvo Iglesias, MD, PHD,a Andres Iñiguez Romo, MD, PHDa,d ABSTRACT OBJECTIVES The REAC-TAVI (Assessment of platelet REACtivity after Transcatheter Aortic Valve Implantation) trial enrolled patients with aortic stenosis (AS) undergoing transcatheter aortic valve replacement (TAVR) pre-treated with aspirin þ clopidogrel, aimed to compare the efficacy of clopidogrel and ticagrelor in suppressing high platelet reactivity (HPR) after TAVI. BACKGROUND Current recommendations support short-term use of aspirin þ clopidogrel for patients with severe AS undergoing TAVR despite the lack of compelling evidence. METHODS This was a prospective, randomized, multicenter investigation. Platelet reactivity was measured at 6 different time points with the VerifyNow assay (Accriva Diagnostics, San Diego, California). HPR was defined as (P2Y12 reaction units (PRU) $208. Patients with HPR before TAVR were randomized to either aspirin þ ticagrelor or aspirin þ clopidogrel for 3 months. Patients without HPR continued with aspirin þ clopidogrel (registry cohort). The primary endpoint was non-HPR status (PRU <208) in $70% of patients treated with ticagrelor at 90 days post-TAVR. RESULTS A total of 68 patients were included. Of these, 48 (71%) had HPR (PRU 273  09) and were randomized to aspirin þ ticagrelor (n ¼ 24, PRU 277  08) or continued with aspirin þ clopidogrel (n ¼ 24, PRU 269  49). The remaining 20 patients (29%) without HPR (PRU 133  12) were included in the registry. Overall, platelet reactivity across all the study time points after TAVR was lower in patients randomized to ticagrelor compared with those treated with clopidogrel, including those enrolled in the registry (p < 0.001). The primary endpoint was achieved in 100% of patients with ticagrelor compared with 21% with clopidogrel (p < 0.001). Interestingly, 33% of clopidogrel responder patients at baseline developed HPR status during the first month after TAVR. CONCLUSIONS HPR to clopidogrel is present in a considerable number of patients with AS undergoing TAVR. Ticagrelor achieves a better and faster effect, providing sustained suppression of HPR to these patients. (Platelet Reactivity After TAVI: A Multicenter Pilot Study [REAC-TAVI]; NCT02224066) (J Am Coll Cardiol Intv 2019;12:22–32) © 2019 The Authors. Published by Elsevier on behalf of the American College of Cardiology Foundation. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). From the aCardiology Department, Hospital Alvaro Cunqueiro, University Hospital of Vigo, Vigo, Spain; bCardiovascular Research Unit, Cardiology Department, Hospital Alvaro Cunqueiro, University Hospital of Vigo, Vigo, Spain; cCardiology Department, Hospital Universitario Virgen de la Arrixaca, IMIB-Arrixaca, Murcia, Spain; dCentro de Investigación en Red de Enfermedades Cardiovasculares (Network Research Center for Cardiovascular Diseases), CIBER-CV, Madrid, Spain; eCardiology Department, Hospital Universitario La Paz, Madrid, Spain; fCardiology Department, Hospital Universitario de Salamanca, Salamanca, Spain; g Cardiology Department, Hospital Universitario de Bellvitge, Barcelona, Spain; hCardiology Department, Hospital Universitario Virgen de las Nieves, Granada, Spain; iCardiovascular Research Group, Galicia Sur Health Research Institute (IIS Galicia Sur), SERGAS-UVIGO, Vigo, Spain; and the jCardiology Department, Hospital Universitario San Pau, Barcelona, Spain. This is an investigator-initiated trial, sponsored by the Spanish Society of Cardiology, and supported by AstraZeneca. Data monitoring, data ISSN 1936-8798 https://doi.org/10.1016/j.jcin.2018.10.005 JACC: CARDIOVASCULAR INTERVENTIONS VOL. 12, NO. 1, 2019 Jimenez Diaz et al. 23 JANUARY 14, 2019:22–32 The REAC-TAVI Trial T ranscatheter aortic valve replacement stenting, and there are limited data on pro- ABBREVIATIONS (TAVR) for degenerative aortic stenosis files of platelet reactivity in AS patients un- AND ACRONYMS (AS), the most common valvular heart dis- dergoing TAVR treated with clopidogrel. ARU = aspirin reaction unit(s) ease in the elderly (1), has become an established Moreover, the impact of newer generation AS = aortic stenosis treatment for patients of high or intermediate surgi- P2Y 12 inhibitor such as ticagrelor character- DAPT = dual antiplatelet cal risk because it offers superior quality of life and ized by more potent and less heterogeneous therapy similar mortality rates at 2 years follow-up, with less antiplatelet effects has never been prospec- HPR = high platelet reactivity invasiveness than surgical aortic valve replacement tively tested in AS patients undergoing TAVR. PRU = P2Y12 reaction unit(s) (2,3). Application of this technique in younger pa- Therefore, we aimed to analyze profiles of TAVR = transcatheter aortic tients and those of lower surgical risk is currently be- platelet reactivity among patients with AS valve replacement ing explored. undergoing TAVR treated with clopidogrel, on a background of aspirin therapy, and to assess the ef- SEE PAGE 33 fects of ticagrelor among patients presenting with HPR. In patients undergoing TAVR, thrombotic and hemorrhagic complications remain a concern, and are METHODS associated with increased mortality and morbidity (4–6). Although the immediate procedural risk relates STUDY DESIGN AND PATIENT POPULATION. The to vascular access and valvular debris embolization, REAC-TAVI trial (Assessment of platelet REACti- an significant percentage of thrombotic complications vity after Transcatheter Aortic Valve Implantation, takes place in the first days after the procedure and NCT02224066) was a prospective, open-label, inves- during follow-up, remaining raised for up to 6 months tigator-initiated, randomized, multicenter, parallel- (4–7). These are non–procedural-related events and group, phase IV pharmacodynamic clinical trial, implies the presence of a prothrombotic environment designed to investigate the incidence of HPR with in these patients. Therefore, antiplatelet treatment aspirin and clopidogrel in patients with severe plays an important role in trying to maintain the symptomatic AS undergoing TAVR and to test the balance between the suppression of thrombotic hypothesis that ticagrelor is superior to clopidogrel in complications without increasing the risk of bleeding. suppression of HPR after TAVR. An outline of the Presently, antithrombotic therapy in patients with study design is provided in the Online Appendix. AS undergoing TAVR is not standardized. Various In brief, patients with severely symptomatic combinations of antithrombotic regimens (single-an- degenerative AS undergoing TAVR by transfemoral tiplatelet, dual antiplatelet therapy [DAPT], or oral access were included at 7 centers in Spain (see the anticoagulants) have been used, but evidence-based Online Appendix for the participating centers). Pa- guidance is limited. Current guidelines (8,9) for the tients were excluded if they had any of the following: management of patients with valvular heart disease stroke during the 14 days before TAVR, bleeding recommend DAPT combining low-dose aspirin and a diathesis, need for long-term oral anticoagulation thienopyridine (clopidogrel) for up to 6 months after treatment, contraindications to DAPT for 3 months, TAVR. However, the evidence for this recommenda- platelet count <50,000/m l, severe hepatic dysfunc- tion is weak and has no support from dedicated ran- tion, use of potent CYP3A inhibitors or inducers. All domized clinical trials. Therefore, there is patients were on a background of aspirin (100 mg/ heterogeneity in antithrombotic treatment regimens day) in adjunct to clopidogrel. Patients need to be on following TAVR in real-world practice (10,11). clopidogrel maintenance dosing (75 mg/day) for $4 Studies have demonstrated a broad variability in days before TAVR. Clopidogrel-naive patients were individual response profiles to clopidogrel therapy pretreated with a 300-mg loading dose. resulting in adverse clinical outcomes, particularly Patients with HPR at baseline were randomized in a among those who persist with high platelet reactivity 1:1 ratio using computerized random-number (HPR) (12–14). However, most of these studies have generation in an interactive web response system by been conducted in patients undergoing coronary an independent investigator to either continue entry, database maintenance, statistical analysis, and drafting and submission of the final manuscript was exclusively performed by the investigators. AstraZeneca played no role in the design of the study, data analysis, data interpretation, writing of the report, or in the decision to submit for publication. The authors have reported that they have no relationships relevant to the contents of this paper to disclose. Manuscript received July 18, 2018; revised manuscript received September 17, 2018, accepted October 2, 2018. 24 Jimenez Diaz et al. JACC: CARDIOVASCULAR INTERVENTIONS VOL. 12, NO. 1, 2019 The REAC-TAVI Trial JANUARY 14, 2019:22–32 Valve Academic Research Consortium-2 classification F I G U R E 1 Study Flow Chart (16). The study flow chart is showed in Figure 1. The study complied with the Declaration of Helsinki, the International Conference on Harmonization/Good Clinical Practice guidelines, and applicable regulatory requirements. Informed consent was obtained and documented for all patients before conducting any study-related procedures. PLATELET REACTIVITY ASSESSMENT. Platelet func- tion was assessed using the VerifyNow point-of-care assay (Accriva Diagnostics, San Diego, California). Measurements were conducted in all patients at baseline (before TAVR) and at 6 different time points following TAVR: 6 h  30 min, 24  2 h, 5  1 day, 30  3 days, and 90  5 days. HPR was defined as a P2Y12 re- action unit (PRU) value $208, in line with consensus recommendations (17,18). Response to aspirin was also assessed; an aspirin reaction unit (ARU) value $550 defined patients as nonresponders to aspirin. All pa- tients were included in the per protocol safety anal- ysis, whereas the pharmacodynamic analysis at 30 days consisted of 61 patients: 1 patient in the registry was excluded from the analyses due to missing platelet reactivity data results; 2 patients (1 in the ticagrelor group and 1 in the clopidogrel group) were excluded due to development of new-onset atrial fibrillation during hospitalization and were switched to oral anticoagulation; and 4 patients died during hospitalization (2 in the clopidogrel group and 2 in the registry group). At 90 days, 2 patients died in the clo- ASA ¼ acetylsalicylic acid; PRU ¼ P2Y12 reactivity units; TAVR ¼ transcatheter aortic valve pidogrel group. The final pharmacodynamic popula- replacement. tion at 90 days, therefore, comprised 59 patients (ticagrelor n ¼ 23; clopidogrel n ¼ 19; registry n ¼ 17). STUDY ENDPOINTS. The primary endpoint was to clopidogrel (75 mg/day) or switch to ticagrelor (90 obtain non-HPR status (PRU <208) in $70% of pa- mg/twice daily). Patients randomized to the ticagrelor tients treated with ticagrelor, or a net difference group received at least 3 maintenance doses before of $40% in the number of patients with PRU <208 undergoing TAVR. Patients without HPR were between treatment groups at 90  5 days of anti- included in a registry cohort and maintained on clo- platelet treatment following TAVR. A key secondary pidogrel therapy. efficacy endpoint was the net difference in the inci- The TAVR procedure was performed according to dence of HPR 6 h post-TAVR of $30% between groups. the standards of each participating center. Patients The secondary safety endpoints are constituted by received intravenous heparin (100 U/kg), with addi- the incidence of cerebrovascular accident or hemor- tional doses if the activated clotting time was <250 s. rhagic complications (according to Valve Academic Compliance to antiplatelet therapy was assessed by Research Consortium-2 criteria) at 4-month follow-up the 4-item Morisky Green Levine Medication Adher- following TAVR. Exploratory endpoints included the ence Scale (15) at 30 and 90 days after TAVR and by analysis of patients with HPR after 24 h, 5 days, and accountability of the study drugs in both treatment 30 days of antiplatelet treatment, and pre-specified arms. Onsite monitoring and source data verification subanalyses by sex, age, comorbidities (diabetes were conducted by an independent contract research mellitus, renal failure), type of bioprosthetic valve organization in 100% of patients and in all study implanted, and HPR to both antiplatelet agents. procedures, with a final follow-up visit at 4 months after TAVR. Adverse events were adjudicated by an SAMPLE-SIZE CALCULATION AND STATISTICAL METHODS. independent clinical events committee, according the Statistical inference was based on comparison of the JACC: CARDIOVASCULAR INTERVENTIONS VOL. 12, NO. 1, 2019 Jimenez Diaz et al. 25 JANUARY 14, 2019:22–32 The REAC-TAVI Trial incidence of the primary endpoint between treatment prevalence of chronic kidney disease and higher he- groups. To achieve 80% power to detect differences moglobin levels. Baseline characteristics were similar in the contrast of the null hypothesis H ₀ :p1 ¼ p2 using between randomized groups, except for lower body a chi-square test for 2 independent samples with a ¼ mass index and mean aortic valve area, and higher 0.05, a treatment arm allocation ratio of 1:1, under the STS-PROM score in the clopidogrel group compared assumption that the proportion of patients achieving with the ticagrelor group. The majority of patients adequate antiplatelet therapy at 3 months post-TAVR had pre-existing cardiovascular risk factors, nearly (PRU <208) in the aspirin þ clopidogrel group is 28% one-half of the population were women, had dia- and the proportion expected in the aspirin þ tica- betes, or had prior percutaneous coronary interven- grelor group is 70%, it was necessary to enroll 21 pa- tion, and more than one-quarter of patients had a tients in each treatment group. Given an expected history of myocardial infarction or chronic kidney dropout rate of 10%, 24 patients in each treatment disease. group (total n ¼ 48) would be required. All patients were treated by transfemoral access, Conformity to the normal distribution was evalu- with a predominance of a conscious sedation strategy ated for continuous variables with the Shapiro-Wilk in the randomized group and general anesthesia in test. Categorical variables are expressed as fre- the registry. Procedural success was achieved in quencies and percentages. For baseline characteris- >90%, and a balloon-expandable valve was used in tics, continuous variables are expressed as mean  48% of cases (Table 2). SD; PRU and ARU as mean  SE. Chi-square tests or PHARMACODYNAMIC RESULTS. Overall, platelet reac- Fisher exact tests were used, where appropriate, to tivity across all the study time points after TAVR was compare categorical variables between 2 groups. Pri- lower in patients receiving ticagrelor compared with mary analysis of the difference between ticagrelor those receiving clopidogrel, including the registry and clopidogrel in PRU at 90 days was analyzed using group (p < 0.001). The primary endpoint of patients a 2-sample Student’s t-test. Two-sample Student’s with PRU <208 was achieved in 100% of patients with t-tests were also used to evaluate other intergroup ticagrelor compared with 21% with clopidogrel (PRU comparisons and to evaluate the impact of the 2 240  15; p < 0.001), with a net difference of re- different treatments on platelet reactivity across time sponders between groups of 79% (p < 0.001) and an points. overall mean difference of 170 PRU in the ticagrelor group compared to clopidogrel group at the end of the RESULTS treatment. Analysis of PRU levels at the different time-points also showed a significant reduction of PATIENT POPULATION. Between November 2015 and platelet reactivity at 6 h, 24 h, 5 days, and 30 days May 2017, of 73 patients screened, 68 patients met with ticagrelor, compared with clopidogrel (p < study entry criteria (3 patients had difficulties 0.001) (Figure 2). The percentage of responders to completing the follow-up visit, 1 patient was switched ticagrelor was significantly higher throughout the to non-transfemoral TAVR, and 1 patient was entire treatment period compared with responders to excluded due to physician decision). Most patients clopidogrel (Figure 3). Interestingly, important varia- were on long-term use ($4 days) of clopidogrel (84%), tions in PRU levels throughout the treatment period whereas the remaining patients received a loading was observed in the registry group, with an increase dose. A total of 48 patients (71%) had HPR at baseline of 53  26 PRU (Figure 2) and a decrease of 33% re- (PRU 273  09) and were randomized to continue sponders during the first month after TAVR (Figure 3). treatment with clopidogrel (n ¼ 24, PRU 269  49) or A total of 23 patients had at least 1 PRU value #70 to switch to ticagrelor (n ¼ 24, PRU 277  08). The during the treatment period: 1 in the clopidogrel remaining 20 patients (29%) without HPR (PRU 133  group, 18 in the ticagrelor group, and 4 in the registry 12) were included in the registry (Figure 1). Baseline group. Mean ARU values were maintained in ranges ARU was 456  15 in the randomized group and 451  of adequate response during the treatment period in 18 in the registry group; an ARU >550 was present in 9 the 3 groups, but significant variations in the degree patients (13%) (ARU 587  10). Of these, 5 patients of response to aspirin between groups were noticed at were in the registry group. Thus, 4 patients had HPR 5 and 30 days post-TAVR (Figure 4). The proportion of to both aspirin and clopidogrel, and were distributed patients responding to aspirin did not vary signifi- equally in both randomized groups. cantly during the treatment phase. Baseline characteristics of the study population are displayed in Table 1. Compared with the randomized SAFETY RESULTS. There was a total of 55 adverse group, patients in the registry cohort had a lower events adjudicated by the clinical events committee 26 Jimenez Diaz et al. JACC: CARDIOVASCULAR INTERVENTIONS VOL. 12, NO. 1, 2019 The REAC-TAVI Trial JANUARY 14, 2019:22–32 T A B L E 1 Baseline Characteristics of the Study Population High On-Treatment Platelet Reactivity Normal Platelet *p Value †p Value ‡p Value Reactivity Registry vs. Registry vs. Registry vs. Clopidogrel þ Aspirin Ticagrelor þ Aspirin Registry (n ¼ 20) Randomized Clopidogrel þ Aspirin Ticagrelor þ Aspirin (n ¼ 24) (n ¼ 24) §p Value Demographic characteristics Age, yrs 75.8  19.6 0.31 0.25 0.31 81.2  5.5 80.3  8.6 0.67 Women 9 (45.0) 0.58 0.95 0.54 11 (45.8) 13 (54.2) 0.56 BMI, kg/m2 26.3  4.5 0.10 0.48 0.004 27.3  4.77 30.6  4.6 0.02 Cardiovascular risk factors Current smokers 5 (25.0) 0.56 0.92 0.43 6 (25.0) 3 (12.5) 0.46 Hypertension 16 (80.0) 0.96 0.67 0.64 19 (79.2) 22 (91.7) 0.41 Diabetes mellitus 8 (40.0) 0.12 0.10 0.43 16 (66.7) 13 (54.2) 0.37 Insulin-treated 2 (10.0) 0.19 0.08 1.00 8 (33.3) 3 (12.5) 0.08 Dyslipidemia 11 (55.0) 0.29 0.55 0.13 16 (66.7) 19 (79.2) 0.33 History of cardiovascular disease and procedures PVD 0 (0.0) 0.18 0.24 1.00 3 (12.5) 1 (4.2) 0.29 CKD 2 (10.0) 0.03 0.14 0.07 8 (33.3) 9 (37.5) 0.76 COPD 2 (10.0) 0.71 0.36 0.56 5 (20.8) 4 (16.7) 0.71 Chronic liver disease 1 (5.0) 0.84 1.00 1.00 2 (8.3) 1 (4.2) 1.00 Stroke 1 (5.0) 0.63 0.61 1.00 3 (12.5) 1 (4.2) 0.29 TIA 1 (5.0) 0.89 1.00 0.45 2 (8.3) 0 (0) 0.48 Myocardial infarction 5 (25.0) 0.56 0.47 0.72 4 (16.7) 5 (20.8) 0.71 PCI 10 (50.0) 0.52 0.47 0.54 10 (41.7) 10 (41.7) 1.00 CABG 0 (0.0) 0.16 0.24 0.24 3 (12.5) 3 (12.5) 1.00 Clinical data Dyspnea 16 (80.0) 0.96 1.00 0.64 19 (79.2) 22 (91.7) 0.41 Angina 4 (20.0) 0.53 1.00 0.50 5 (20.8) 8 (33.3) 0.33 Syncope 3 (15.0) 0.41 0.64 0.64 2 (8.3) 2 (8.3) 1.00 NYHA II to III 14 (70.0) 0.07 0.29 0.29 22 (95.7) 23 (95.8) 1.00 Echocardiographic data Maximum aortic gradient, mm Hg 80.6  24.5 0.42 0.94 0.33 81.1  17.5 70.7  23.0 0.10 Mean aortic gradient, mm Hg 51.8  17.8 0.06 0.45 0.06 47.9  11.6 41.1  13.3 0.08 Mean aortic valve area, cm2 0.61  0.17 0.29 0.93 0.04 0.62  0.20 0.73  0.13 0.05 Aortic regurgitation $moderate 3 (15.0) 0.11 0.29 0.28 1 (5.3) 1 (4.8) 1.00 LVEF 60.5  12.7 0.52 0.30 0.96 55.9  12.9 60.8  15.3 0.28 Systolic PAP 52.3  15.6 0.08 0.29 0.04 41.7  13.2 36.2  8.1 0.34 Blood analysis Hemoglobin, mg/dl 13.4  1.6 0.01 0.009 0.02 12.0  1.8 12.2  1.8 0.62 Platelets, 109/l 203.1  71 0.95 0.78 0.67 197.3  62 211.7  60 0.42 NT-proBNP 1,901  1,496 0.41 0.27 0.83 4,226  5,178 2,104  2,103 0.33 Risk scores Logistic EuroScore 8.9  6.11 0.21 0.11 0.71 13.37  10.37 9.52  5.31 0.12 STS-PROM, % 6.36  7.61 0.73 0.67 0.23 7.22  5.18 4.08  2.53 0.014 HAS-BLED 2.75  1.30 0.63 0.87 0.34 2.69  0.90 2.40  0.81 0.25 CHA2DS2-VASC 4.00  1.41 0.15 0.11 0.39 4.70  1.36 4.38  1.41 0.43 Values are mean  SD or n (%). *p values of comparison between registry group and randomized group. †p values of comparison between registry group and clopidogrel þ aspirin group. ‡p values of comparison between registry group and ticagrelor þ aspirin group. §p values of comparison between both randomized groups. BMI ¼ body mass index; CABG ¼ coronary artery bypass graft; CKD ¼ chronic kidney disease; COPD ¼ chronic obstructive pulmonary disease; LVEF ¼ left ventricular ejection fraction; NT-proBNP ¼ N-terminal pro–B-type natriuretic peptide; NYHA ¼ New York Heart Association functional class; PAP ¼ pulmonary arterial pressure; PCI ¼ percutaneous coronary intervention; PVD ¼ peripheral vascular disease; STS ¼ Society of Thoracic Surgeons; TIA ¼ transient ischemic attack. in the overall population during the entire study. as well as other adjudicated adverse events according There were no statistically significant differences in VARC-2 definitions (Table 3). Only 1 patient had an in- the incidence of major bleeding or cardiovascular hospital major bleeding related to vascular access death among the study groups at 4 months follow-up, while on a PRU value of 70 at 6 h after TAVR. JACC: CARDIOVASCULAR INTERVENTIONS VOL. 12, NO. 1, 2019 Jimenez Diaz et al. 27 JANUARY 14, 2019:22–32 The REAC-TAVI Trial T A B L E 2 Procedural Characteristics and In-Hospital Outcome High On-Treatment Platelet Reactivity Normal Platelet *p Value †p Value ‡p Value Reactivity Registry vs. Registry vs. Registry vs. Clopidogrel þ Aspirin Ticagrelor þ Aspirin Registry (n ¼ 20) Randomized Clopidogrel þ Aspirin Ticagrelor þ Aspirin (n ¼ 24) (n ¼ 24) §p Value Procedural data Local anesthesia only 3 (15.0) 0.003 0.10 <0.001 3 (12.5) 1 (4.2) 0.07 Conscious sedation 5 (25.0) 14 (58.3) 21 (87.5) General anesthesia 12 (60.0) 7 (29.2) 2 (8.3) Transfemoral access 20 (100.0) 1.00 1.00 1.00 24 (100) 24 (100.0) 1.00 Balloon pre-dilation 4 (20.0) 1.00 1.00 1.00 5 (20.8) 5 (20.8) 1.00 Balloon size, mm 23.3  1.5 0.01 0.17 0.008 20.7  2.3 19.8  1.1 0.48 Valve type, size, and post-dilation Balloon-expandable valve 6 (30.0) 9 (37.5) 14 (58.3) Self-expandable valve 9 (45.0) 11 (45.8) 9 (37.5) Other type of valve 5 (25.0) 0.37 0.86 0.11 4 (16.7) 1 (4.2) 0.21 #23-mm valve 0 (0.0) 0.10 0.10 0.11 7 (29.2) 7 (29.2) 0.83 25- to 29-mm valve 18 (90.0) 15 (62.5) 16 (66.7) >29-mm valve 2 (10.0) 2 (8.3) 1 (4.1) Balloon post-dilation 5 (25.0) 0.70 1.00 0.64 6 (25.0) 3 (12.5) 0.46 Balloon size, mm 24.3  0.96 0.38 0.38 0.49 22.7  2.3 23.3  2.3 0.76 Vascular closure 2 ProGlides 13 (65.0) 0.66 0.63 0.71 19 (79.2) 16 (66.7) 0.58 Prostar 5 (25.0) 3 (12.5) 4 (16.7) Surgical closure 2 (10.0) 2 (8.3) 4 (16.7) Procedural outcome Procedural success 18 (90.0) 1.00 NA 1.00 23 (96) 23 (87.5) 1.00 Final maximum gradient, mm Hg 17.9  15.8 0.37 0.26 0.29 7.0  4.6 11.4  8.1 0.17 Final mean gradient, mm Hg 8.7  8.1 0.75 0.35 0.78 4.8  3.7 9.8  5.5 0.11 Final AR $moderate 2 (10.0) 0.13 0.08 0.35 1 (4.1) 3 (12.5) 0.35 Total dose of UFH, UI 5,090  2,634 0.02 0.06 0.03 6,904  2,462 7,000  2,161 0.89 Need of valve recapture and reposition 2 (10.0) 0.59 0.60 0.60 2 (8.3) 2 (8.3) 1.00 Need for a second valve 0 (0.0) NA NA NA 0 (0) 0 (0.0) NA Conversion to open-heart surgery 0 (0.0) NA NA NA 0 (0) 0 (0.0) NA Need of permanent pacemaker implantation 3 (15.0) 0.71 1.00 1.00 5 (20.8) 5 (20.8) 1.00 Hospital stay, days 7.9  3.6 0.86 0.88 0.85 7.7  4.0 7.6  5.6 0.93 Values are n (%), or mean  SD. *p values of comparison between the registry group and the randomized group. †p values of comparison between the registry group and the clopidogrel þ aspirin group. ‡p values of comparison between the registry group and the ticagrelor þ aspirin group. §p values of comparison between both randomized groups. Other type of valve includes Lotus valve and Direct Flow valve. AR ¼ aortic regurgitation; UFH ¼ unfractionated heparin. DISCUSSION highly effective for the suppression of HPR to clopi- dogrel after TAVR with reduced rates of HPR To date, there are limited studies evaluating profiles observed already a few hours after valve implantation of platelet reactivity in patients with severe AS un- and with consistent effects throughout the treatment dergoing TAVR (19–21). The present study is the first period. to our knowledge to prospectively evaluate profiles of Previous studies have demonstrated that platelet platelet reactivity of patients with AS undergoing reactivity is abnormally increased in patients with TAVR treated with aspirin plus clopidogrel therapy valvular heart disease, generating deposition of and to evaluate the effects of ticagrelor among pa- platelets and formation of thrombi on the surface of tients with HPR. The main results of our study are as altered or diseased, natural heart valves (22). This follows: 1) HPR to clopidogrel is present in more than may act as a nidus for microthrombi formation two-thirds of patients with AS undergoing TAVR; 2) in vivo. Increased platelet activation and the throm- more than one-third of patients who were identified bogenic environment in patients with severe AS un- to be responders to clopidogrel developed HPR status dergoing aortic bioprosthesis implantation might be during the first month after TAVR; and 3) ticagrelor is related to: 1) high transvalvular gradient leading to 28 Jimenez Diaz et al. JACC: CARDIOVASCULAR INTERVENTIONS VOL. 12, NO. 1, 2019 The REAC-TAVI Trial JANUARY 14, 2019:22–32 F I G U R E 2 Mean PRU Values During 3 Months of Treatment Following TAVR HPR ¼ high platelet reactivity; other abbreviations as in Figure 1. F I G U R E 3 Responders to P2Y 12 Inhibitors TAVR ¼ transcatheter aortic valve replacement. JACC: CARDIOVASCULAR INTERVENTIONS VOL. 12, NO. 1, 2019 Jimenez Diaz et al. 29 JANUARY 14, 2019:22–32 The REAC-TAVI Trial F I G U R E 4 Mean ARU Values During 3 Months of Treatment Following TAVR ARU ¼ aspirin reactivity units; NS, nonsignificant; other abbreviations as in Figures 1 and 2. increased shear stress and endothelial injury, thereby hyperactive platelet response found in these patients promoting platelet adhesion and activation; 2) altered sustaining a thrombogenic environment. aortic blood flow and activation of various pro- Platelets are well known to play a major role thrombotic factors (23–26) (von Willebrand factor, in coronary stent thrombosis and adequate anti- factor VIII); 3) exposure of subendothelial thrombus- platelet therapy protects against this phenomenon producing materials and release of activated throm- (17,18,30–34). However, the role of platelets and thus botic factors, such as tissue factor and thrombin, from that of antiplatelet therapy in the development of degenerated native aortic valve leaflets (27) into the clinical (transient ischemic attack, stroke, valve circulation generating microthromboemboli (28); and thrombosis) or subclinical (hypoattenuated leaflet 4) poor antiplatelet effects leading to accumulation of thickening and reduced leaflet motion) thrombotic fibrin in the stent valve and microthrombi formation events after a percutaneous or surgical implant of an on the nonendothelialized surface of the bovine aortic bioprosthesis (35–37) is less established. Theo- pericardial tissue leaflets, due to microfissures retically, poor antiplatelet response (i.e., presence of (potentially produced during crimping) prone to HPR) after bioprosthesis implantation could promote platelet adhesion (Figure 5). fibrin deposition and platelet aggregation, favoring Also, abnormal functional characteristics of plate- the thickening of valvular leaflets and bioprosthetic lets in patients with valvular heart disease have been structures, a phenomenon that would end with the documented, compared with normal subjects, formation of thrombotic material. Symptomatic consisting of morphological alterations and with increased pressure gradients with thickened leaflet subendothelial components, such as collagen fibers, tips, leaflet adhesion, and impairing proper opening exposed to the circulating blood with platelet aggre- but with no evidence of thrombi have been described gates adhering to collagen fibers. Moreover, abnor- after early discontinuation of DAPT, with complete mally increased thromboplastic activity and resolution after DAPT resume (38). Also, acute cere- decreased fibrinolytic capacity in histopathology bral occlusion by a fresh thrombus consisting mainly analysis of rheumatic mitral valves have been of accumulations of fibrin and platelets during TAVR, described (29). These features may explain the have been reported (39). Ongoing clinical trials are 30 Jimenez Diaz et al. JACC: CARDIOVASCULAR INTERVENTIONS VOL. 12, NO. 1, 2019 The REAC-TAVI Trial JANUARY 14, 2019:22–32 T A B L E 3 Clinical Events at 4-Month Follow-Up High On-Treatment Platelet Reactivity Normal Platelet *p Value †p Value ‡p Value Reactivity Registry vs. Registry vs. Registry vs. Clopidogrel þ Aspirin Ticagrelor þ Aspirin Registry (n ¼ 20) Randomized Clopidogrel þ Aspirin Ticagrelor þ Aspirin (n ¼ 24) (n ¼ 24) §p Value Death 2 (10.0) 1.00 0.67 0.20 4 (16.7) 0 (0.0) 0.10 Stroke 1 (5.0) 0.50 1.00 0.45 1 (4.2) 0 (0.0) 1.00 Myocardial infarction 1 (5.0) 0.29 0.45 0.45 0 (0.0) 0 (0.0) NA Major bleeding 1 (5.0) 0.15 0.54 0.61 7 (29.2) 3 (12.5) 0.28 MACE 3 (15.0) 0.68 0.71 0.08 5 (20.8) 0 (0.0) 0.05 NACE 4 (20.0) 0.43 0.07 0.68 11 (45.8) 3 (12.5) 0.01 All bleeding events 1 (5.0) 0.09 0.05 0.35 7 (29.2) 4 (16.7) 0.49 Acute renal failure 0 (0.0) 1.00 NA 1.00 0 (0.0) 1 (4.2) 1.00 Vascular complication 0 (0.0) 0.55 0.49 1.00 2 (8.3) 1 (4.2) 1.00 Vascular closure device failure 0 (0.0) 0.55 1.00 0.49 1 (4.2) 2 (8.3) 1.00 Arrhythmias and conduction disturbances 2 (10.0) 0.19 0.15 0.25 7 (29.2) 6 (25.0) 1.00 Values are n (%). *p values of comparison between the registry group and the randomized group. †p values of comparison between the registry group and the clopidogrel þ aspirin group. ‡p values of comparison between the registry group and the ticagrelor þ aspirin group. §p values of comparison between both randomized groups. All events are according Valve Academic Research Consortium-2 criteria. MACE includes death, stroke, and myocardial infarction. NACE includes death, stroke, myocardial infarction, and major bleeding. MACE ¼ major adverse cardiovascular event(s); NACE ¼ net adverse clinical event(s). evaluating oral anticoagulation and DAPT in TAVR well as the poor efficacy of clopidogrel to maintain patients without atrial fibrillation and will help to adequate platelet inhibition during the following 3 define the role of platelets and thrombin on throm- months after TAVR. botic events after TAVR (Global Study Comparing a Therefore, ticagrelor represent an attractive alter- rivAroxaban-based Antithrombotic Strategy to an native to clopidogrel, not only to treat HPR, but also antipLatelet-based Strategy After Transcatheter to prevent this from occurring in patients who may pre- aortIc vaLve rEplacement to Optimize Clinical Out- sent with optimal response to clopidogrel before TAVR. comes [GALILEO], NCT02556203; Dual Antiplatelet Platelet reactivity seems to be a dynamic phenom- Therapy Versus Oral Anticoagulation for a Short Time enon, giving relevance to the effect of the timing of to Prevent Cerebral Embolism After TAVI [AUREA], measurement on the level of reactivity. Also, interin- NCT01642134). dividual variability in the platelet inhibitory response The TAVR population is continuously increasing and poses unique challenges on implementing the F I G U R E 5 Potential Factors Leading to Increased Platelet optimal antithrombotic regimen. Overall, our results Activation and Prothrombotic Environment in Patients confirm the superior potency of ticagrelor over clopi- With AS Undergoing TAVR dogrel to achieve prompt and potent platelet inhibi- tory effects in a new challenging scenario, comparable to what has been described in prior investigations conducted in other clinical settings (40,41). Notably, by the 6-h time point after TAVR, <10% of ticagrelor- treated patients still had HPR. The suppression of HPR with ticagrelor was reached in 100% at 5 days post-TAVR, remaining unchanged during follow-up in all patients. Furthermore, the fact that over 30% of patients initially responding to clopidogrel became nonresponders during the first month after valve im- plantation, highlights the significant increase in platelet reactivity induced by the bioprosthetic valve implantation procedures and its components at an early stage (use of large-bore catheters and stiff guidewire in the left ventricle, pre-/post-dilation with AS ¼ aortic stenosis; TAVR ¼ transcatheter aortic valve disruption of native aortic valve leaflets, and calcifi- replacement. cations of the aortic valve annulus and aortic arch), as JACC: CARDIOVASCULAR INTERVENTIONS VOL. 12, NO. 1, 2019 Jimenez Diaz et al. 31 JANUARY 14, 2019:22–32 The REAC-TAVI Trial to clopidogrel has been demonstrated in patients un- rates of residual platelet reactivity during the peri- dergoing elective coronary stenting. Gurbel et al. (42), procedural period, with a significant increase in found that the maximum inhibitory response to a 300- platelet reactivity during the first month after TAVR mg loading dose of clopidogrel occurs within 24 h. In and that may last for up to 3 months thereafter, GRAVITAS (Gauging Responsiveness With A Ver- casting doubts of its efficacy in this setting. Ticagrelor ifyNow Assay-Impact On Thrombosis And Safety) trial achieves a better and faster effect, providing sus- (43), HPR measured 12 to 24 h after percutaneous cor- tained benefit to these patients over the course of the onary intervention resolved at the 30-day follow-up in treatment period without safety concerns. Larger 38% of the patients randomly assigned to standard- studies are urgently needed to define its clinical dose clopidogrel. In our study, 79% of patients were benefit in this setting. on long-term (>30 days) use of clopidogrel, and the remaining 29% clopidogrel-naive patients received a 300-mg loading dose, with evaluation of platelet ADDRESS FOR CORRESPONDENCE: Dr. Victor A. reactivity 24 h after the loading dose. This heteroge- Jimenez Diaz, Cardiology Department, Hospital neity to timing of baseline platelet function assess- Alvaro Cunqueiro, University Hospital of Vigo, ment with relationship to timing of clopidogrel intake Estrada Clara Campoamor 341, Vigo 36312, Spain. may explain why 21% of patients initially identified to E-mail: victor.alfonso.jimenez.diaz@sergas.es. be nonresponders became responders after TAVR. Furthermore, the Hawthorne effect could also have a role in this change. Hence, the clinical, procedural, and PERSPECTIVES genetic predictors of the early resolution of HPR after TAVR deserve further evaluation. WHAT IS KNOWN? Lack of compelling evidence exists on the efficacy of aspirin þ clopidogrel as antithrombotic treatment in STUDY LIMITATIONS. The HPR cutoff point used in patients with severe AS undergoing TAVR. our study was the cutoff recommended for the eval- uation of thrombotic events in patients with coronary WHAT IS NEW? HPR to clopidogrel is present in a considerable artery disease. But, the HPR cutoff for assessing number of patients with AS undergoing TAVR and more than one thrombotic events in the TAVR population is still third of patients initially responders to clopidogrel become unclear. Given the pharmacodynamic design of the nonresponder during the treatment period. Ticagrelor achieves a study, it was underpowered for clinical endpoints war- better and faster effect providing sustained suppression of HPR ranted by adequately powered studies for this purpose. to these patients. CONCLUSIONS WHAT IS NEXT? Larger clinical trials are needed to assess the clinical implications of these findings. Our data suggest that patients undergoing TAVR for severe AS and treated with clopidogrel have high REFERENCES 1. Nkomo VT, Gardin JM, Skelton TN, et al. Burden valve implantation: insights for cerebrovascular valvular heart disease: a report of the American of valvular heart diseases: a population-based and bleeding events. J Am Coll Cardiol 2013;62: College of Cardiology/American Heart Association study. Lancet 2006;368:1005–11. 2349–59. Task Force on Clinical Practice Guidelines. J Am Coll Cardiol 2017;70:252–89. 2. Smith CR, Leon MB, Mack MJ, et al. Trans- 6. Piccolo R, Pilgrim T, Franzone A, et al. Fre- catheter versus surgical aortic-valve replacement quency, timing, and impact of access-site and 10. Cerrato E, Nombela-Franco L, Nazif TM, et al. in high-risk patients. N Engl J Med 2011;364: non-access-site bleeding on mortality among Evaluation of current practices in transcatheter 2187–98. patients undergoing transcatheter aortic valve aortic valve implantation: the WRITTEN (WoRld- replacement. J Am Coll Cardiol Intv 2017;10: wIde TAVI ExperieNce) survey. Int J Cardiol 2017; 3. Leon MB, Smith CR, Mack MJ, et al. Trans- 1436–46. 228:640–7. catheter or surgical aortic-valve replacement in intermediate-risk patients. N Engl J Med 2016; 7. Tay EL, Gurvitch R, Wijesinghe N, et al. A high- 11. Ahmad Y, Demir O, Rajkumar C, et al. Optimal 374:1609–20. risk period for cerebrovascular events exists after antiplatelet strategy after transcatheter aortic transcatheter aortic valve implantation. J Am Coll valve implantation: a meta-analysis. Open Heart 4. Kapadia S, Agarwal S, Miller DC, et al. Insights Cardiol Intv 2011;4:1290–7. 2017;5:e000748. into timing, risk factors, and outcomes of stroke and transient ischemic attack after transcatheter 8. Baumgartner H, Falk V, Bax JJ, et al. 2017 ESC/ 12. Angiolillo DJ. Variability in responsiveness to aortic valve replacement in the PARTNER trial EACTS guidelines for the management of valvular oral antiplatelet therapy. Am J Cardiol 2009;103: (Placement of Aortic Transcatheter Valves). Circ heart disease. Eur Heart J 2017;38:2739–91. 27A–34A. Cardiovasc Interv 2016;9:e002981. 9. Nishimura RA, Otto CM, Bonow RO, et al. 2017 13. Brandt JT, Close SL, Iturria SJ, et al. Common 5. Rodes-Cabau J, Dauerman HL, Cohen MG, et al. AHA/ACC focused update of the 2014 AHA/ACC polymorphisms of CYP2C19 and CYP2C9 affect the Antithrombotic treatment in transcatheter aortic guideline for the management of patients with pharmacokinetic and pharmacodynamic response 32 Jimenez Diaz et al. JACC: CARDIOVASCULAR INTERVENTIONS VOL. 12, NO. 1, 2019 The REAC-TAVI Trial JANUARY 14, 2019:22–32 to clopidogrel but not prasugrel. J Thromb Hae- 25. Spiel AO, Gilbert JC, Jilma B. von Willebrand 37. Chakravarty T, Sondergaard L, Friedman J, most 2007;5:2429–36. factor in cardiovascular disease: focus on acute et al. Subclinical leaflet thrombosis in surgical coronary syndromes. Circulation 2008;117: and transcatheter bioprosthetic aortic valves: 14. Neubauer H, Kaiser A, Busse B, et al. Identifi- 1449–59. an observational study. Lancet 2017;389: cation, evaluation and treatment of prasugrel low- 2383–92. response after coronary stent implantation—a 26. Van Belle E, Rauch A, Vincent F, et al. Von preliminary study. Thromb Res 2010;126:e389–91. Willebrand factor multimers during transcatheter 38. Al-Rashid F, Konorza TFM, Plicht B, et al. aortic-valve replacement. N Engl J Med 2016;375: Transient increase in pressure gradients after 15. Morisky DE, Green LW, Levine DM. Concurrent 335–44. termination of dual antiplatelet therapy in a and predictive validity of a self-reported measure patient after transfemoral aortic valve im- of medication adherence. Med Care 1986;24: 27. Otto CM, Prendergast B. Aortic-valve steno- plantation. Circ Cardiovasc Interv 2012;5: 67–74. sis—from patients at risk to severe valve obstruc- 318–20. tion. N Engl J Med 2014;371:744–56. 16. Kappetein AP, Head SJ, Généreux P, et al. 39. Jimenez Diaz VA, Baz Alonso JA, Estevez Updated standardized endpoint definitions for 28. Van Mieghem NM, Schipper ME, Ladich E, Ojea O, Serantes Combo A, Rodriguez Paz CM, transcatheter aortic valve implantation: the Valve et al. Histopathology of embolic debris captured Iñiguez Romo A. Real-time detection of an acute Academic Research Consortium-2 consensus during transcatheter aortic valve replacement. cerebral thrombotic occlusion during a trans- document. J Am Coll Cardiol 2012;60:1438–54. Circulation 2013;127:2194–201. catheter valve intervention. J Am Coll Cardiol Intv 17. Tantry US, Bonello L, Aradi D, et al., for the 29. Homma T, Okudaira S, Iida Y. Studies on 2018;11:e103–4. Working Group on On-Treatment Platelet Reac- thromboplastic and fibrinolytic activities of 40. Bliden KP, Tantry US, Storey RF, et al. The tivity. Consensus and update on the definition of valvular tissue in rheumatic valvular disease. Res effect of ticagrelor versus clopidogrel on high on- on-treatment platelet reactivity to adenosine Exp Med (Berl) 1980;176:193–200. treatment platelet reactivity: combined analysis of diphosphate associated with ischemia and 30. Stone GW, Witzenbichler B, Weisz G, et al. the ONSET/OFFSET and RESPOND studies. Am bleeding. J Am Coll Cardiol 2013;62:2261–73. Platelet reactivity and clinical outcomes after Heart J 2011;162:160–5. 18. Aradi D, Kirtane A, Bonello L, et al. Bleeding coronary artery implantation of drug-eluting 41. Alexopoulos D, Galati A, Xanthopoulou I, and stent thrombosis on P2Y12 inhibitors: collab- stents (ADAPT-DES): a prospective multicentre et al. Ticagrelor versus prasugrel in acute cor- orative analysis on the role of platelet reactivity registry study. Lancet 2013;382:614–23. onary syndrome patients with high for risk stratification after percutaneous coronary 31. Aradi D, Storey RF, Komocsi A, et al. Working on-clopidogrel platelet reactivity following intervention. Eur Heart J 2015;36:1762–71. Group on Thrombosis of the European Society of percutaneous coronary intervention: a pharma- 19. Tousek P, Kocka V, Sulzenko J, et al. Pharma- Cardiology. Expert position paper on the role of codynamic study. J Am Coll Cardiol 2012;60: codynamic effect of clopidogrel in patients un- platelet function testing in patients undergoing 193–9. dergoing transcatheter aortic valve implantation. percutaneous coronary intervention. Eur Heart J 42. Gurbel PA, Bliden KP, Hiatt BL, O’Connor CM. Biomed Res Int 2013;2013:386074. 2014;35:209–15. Clopidogrel for coronary stenting response vari- 20. Orvin K, Eisen A, Perl L, et al. Platelet reac- 32. Gurbel PA, Bliden KP, Guyer K, et al. Platelet ability, drug resistance, and the effect of pre- tivity in patients undergoing transcatheter aortic reactivity in patients and recurrent events post- treatment platelet reactivity. Circulation 2003; valve implantation. J Thromb Thrombolysis 2016; stenting: results of the PREPARE POST-STENTING 107:2908–13. 42:11–8. study. J Am Coll Cardiol 2005;46:1820–6. 43. Price MJ, Berger PB, Teirstein PS, et al., for the 21. Polzin A, Schleicher M, Seidel H, et al. High on- 33. Gurbel PA, Bliden KP, Samara W, et al. Clopi- GRAVITAS Investigators. Standard- vs high-dose treatment platelet reactivity in transcatheter dogrel effect on platelet reactivity in patients with clopidogrel based on platelet function testing af- aortic valve implantation patients. Eur J Pharma- stent thrombosis: results of the CREST study. J Am ter percutaneous coronary intervention. The col 2015;751:24–7. Coll Cardiol 2005;46:1827–32. GRAVITAS randomized trial. JAMA 2011;305: 22. Riddle JM, Stein PD, Magilligan DJ Jr., 34. Wenaweser P, Dorffler-Melly J, Imboden K, 1097–105. McElroy HH. Evaluation of platelet reactivity in et al. Stent thrombosis is associated with an patients with valvular heart disease. J Am Coll impaired response to antiplatelet therapy. J Am Cardiol 1983;1:1381–4. Coll Cardiol 2005;45:1748–52. KEY WORDS antithrombotic therapy, 23. Jaffer IH, Fredenburgh JC, Hirsh J, et al. 35. Hansson NC, Grove EL, Andersen HR, et al. aortic stenosis, high platelet reactivity, Medical device-induced thrombosis: what causes it Transcatheter aortic valve thrombosis: incidence, transcatheter aortic valve replacement, and how can we prevent it? J Thromb Haemost predisposing factors, and clinical implications. VerifyNow 2015;13 Suppl 1:S72–81. J Am Coll Cardiol 2016;68:2059–69. 24. Kopanidis A, Pantos I, Alexopoulos N, et al. 36. Makkar RR, Fontana G, Jilaihawi H, et al. A PPE NDI X For supplemental methods and a Aortic flow patterns after simulated implantation Possible subclinical leaflet thrombosis in bio- list of the trial investigators, coordinators, of transcatheter aortic valves. Hellenic J Cardiol prosthetic aortic valves. N Engl J Med 2015;373: committees, and study sites, please see the 2015;56:418–28. 2015–24. online version of this paper.