Objective To evaluate the clinical benefit of pre-procedural antiplatelet therapy in patients undergoing transfemoral (TF) transcatheter aortic valve implantation (TAVI).
Methods OCEAN (Optimized transCathEter vAlvular interveNtion)-TAVI is a prospective, multicentre, observational cohort registry, enrolling 749 patients who underwent TAVI from October 2013 to August 2015 in Japan. We identified 540 patients (median age 85 years, 68.1% female) undergoing TF-TAVI; of these, 80 had no pre-procedural antiplatelet therapy and 460 had antiplatelet therapy. The endpoints were any bleeding (life-threatening, major, and minor bleeding) and thrombotic events (stroke, myocardial infarction, and valve thrombosis) during hospitalisation.
Results Patients with dual antiplatelet therapy (DAPT) had a significantly higher incidence of any bleeding than those with single antiplatelet therapy (SAPT) (36.5% vs 27.5%, p=0.049) and no antiplatelet therapy (36.5% vs 21.3%, p=0.010). Patients without pre-procedural antiplatelet therapy did not experience an increased risk of thrombotic events. In multivariable logistic regression analysis, DAPT before TF-TAVI significantly increased any bleeding compared with SAPT (OR 2.05, 95% CI 1.16 to 3.65) and no antiplatelet therapy (OR 2.30, 95% CI 1.08 to 4.90).
Conclusions The current study demonstrated that DAPT before TF-TAVI increased the risk of bleeding compared with single or no antiplatelet therapy. Lower intensity antiplatelet therapy was not associated with thrombotic events. In modern practice, it might be reasonable to perform TAVI using single or no pre-procedural antiplatelet therapy with an expectation of no increase of adverse events.
Trial registration number UMIN-ID; 000020423; Results.
Statistics from Altmetric.com
Transcatheter aortic valve implantation (TAVI) is becoming the mainstream treatment for severe aortic valve stenosis.1 ,2 Before TAVI, dual antiplatelet therapy (DAPT) including aspirin (75–100 mg/day) and clopidogrel (75 mg/day after a loading dose of 300 mg) is initiated empirically, based on the experience of percutaneous coronary intervention (PCI).3 ,4 However, compared to coronary intervention, TAVI is associated with a higher incidence of bleeding and has the risk of emergent open heart surgery during the procedure.5 Because almost all candidates for TAVI have advanced age, frailty, and several comorbidities, DAPT before TAVI has been associated with a high incidence of bleeding or vascular complications.6 Moreover, the benefit of DAPT over single antiplatelet therapy (SAPT) or no pre-procedural antiplatelet therapy has not been well studied. The aim of this study was to evaluate the incidence of adverse events according to pre-procedural antiplatelet therapy in patients from a multicentre cohort who underwent TAVI.
This retrospective study used data for the period from October 2013 to July 2015, obtained from the OCEAN (Optimized transCathEter vAlvular interveNtion) TAVI registry. The OCEAN-TAVI is a prospective, multicentre, observational registry of symptomatic patients with severe aortic valve stenosis (AS), undergoing TAVI using the Edwards Sapien XT prosthesis (Edwards Lifesciences) at nine collaborating hospitals located in Japan. This trial was registered with the University Hospital Medical Information Network Clinical Trials Registry, as accepted by the International Committee of Medical Journal Editors (UMIN-ID; 000020423). Inclusion criteria were: (1) the presence of symptoms; (2) degenerative AS with New York Heart Association class ≥II; (3) a mean gradient >40 mm Hg or a jet velocity >4.0 m/s; or (4) an aortic valve area <1.0 cm2 (or an effective orifice area index <0.6 cm2/m2). Indication for TAVI was determined based on the clinical consensus of a multidisciplinary team including cardiac surgeons, interventional cardiologists, anaesthetists, and imaging specialists. Exclusion criteria were: (1) bicuspid or non-calcified aortic valve; (2) failed surgical bioprosthesis implantation; (3) severe aortic regurgitation; or (4) patients on dialysis. The study protocol was developed in accordance with the Declaration of Helsinki and was approved by the ethics committee of each participating hospital. All patients gave written informed consent before participating in this study.
Of the 749 patients registered in the OCEAN-TAVI registry, we excluded those patients who had undergone transapical, transiliac, and transaortic TAVI (n=162) and those who had taken anticoagulants before TAVI (n=47). Finally, we identified 540 patients who had undergone transfemoral (TF) TAVI with no pre-procedural antiplatelet therapy, SAPT, and DAPT during hospitalisation (figure 1). Pre-procedural antiplatelet therapy, including SAPT (aspirin 100 mg or clopidogrel 75 mg) or DAPT (75 mg of clopidogrel plus 100 mg of aspirin), was usually started 1 week before TAVI. After achieving homeostasis, post-procedural antiplatelet therapy, including SAPT, DAPT, SAPT with oral anticoagulation (OAC), and DAPT with OAC, was started, usually 1 day after TAVI. The regimen and timing of antithrombotic medication were left to the physician's discretion in each hospital. Any TAVI-related adverse events in the patients during hospitalisation were retrospectively tracked.
All TAVIs were performed using the Sapien XT-Novaflex Delivery System (Edwards Lifesciences, Inc). The prosthesis size (20, 23, 26, and 29 mm) was determined using pre-procedural echocardiographic and multidetector CT measurements. In each procedure, unfractionated heparin was administered in a weight-adjusted dose at the beginning of the procedure and additional doses were given during the procedure to achieve an activated clotting time of 250–300 s. The approach methods, including percutaneous or cut-down, pre-dilatation and post-dilatation, were left to the physician's discretion. The surgical risk was estimated by the logistic EuroSCORE-II.
The clinical endpoints of this study were: (1) bleeding, including life-threatening, major, minor, and any bleeding, which was defined as the composite endpoint of life-threatening, major, and minor bleeding; (2) transfusion; (3) all-cause death; (4) myocardial infarction (MI); (5) stroke; and (6) valve thrombosis, during hospitalisation. Procedural complications were based on the definition of Valve Academic Research Consortium (VARC–2) criteria.7
Continuous variables were assessed for normal distribution using the Shapiro-Wilk test and presented as mean±SD or median and interquartile range (IQR 25–75%). Dichotomous variables are presented as numbers and percentages. Differences between the patients with no pre-procedural antiplatelet therapy and patients with SAPT or DAPT were compared using the χ2 test or Fisher's exact test for categorical variables and Student's t-tests or Wilcoxon rank-sum tests, as appropriate, for continuous variables. Multivariable logistic regression analysis was performed to identify independent predictors of any bleeding. Variables showed p value <0.25 in univariate analysis8 and variables clinically associated with all bleeding were entered into the multivariate model. The variance inflation factor (VIF) was used to check for multicollinearity for each variable. Statistical analysis was performed using the Statistical Package for Social Sciences, V.21 (SPSS Inc, Chicago, Illinois, USA) software. A value of p<0.05 was considered statistically significant.
Of 540 patients, 531 (98.3%) were discharged alive. Baseline clinical characteristics of the study participants are shown in table 1.
Those patients with any bleeding during hospitalisation were more likely to be older, female, and to have a lower body mass index (BMI), a lower body surface area (BSA), congestive heart failure (CHF) on admission, and chronic kidney disease (CKD) than patients without any bleeding. The STS (Society of Thoracic Surgeons) score and Logistic EuroSCORE II were significantly higher in patients with any bleeding than in those without any bleeding. According to pre-antiplatelet therapy, the incidence of any bleeding was significantly lower in patients without pre-procedural antiplatelet therapy and significantly higher in patients with pre-procedural DAPT. Pre-procedural SAPT did not affect the incidence of bleeding in this population. Patients who had any bleeding during hospitalisation required longer hospital stay than patients without any bleeding. There were no significant differences concerning the morphologic parameter measured by CT and echocardiography.
Except for the nine cases of in-hospital death, 531 patients were prescribed post-procedural antiplatelet therapy. Of the 244 patients with no pre-procedural antiplatelet therapy and pre-procedural SAPT, 65.9% (n=161) had started post-procedural DAPT; and of the 287 patients with pre-procedural DAPT, 89.2% (n=256) continued DAPT after TAVI. The alteration to post-procedural DAPT from pre-procedural no antiplatelet therapy or SAPT was significantly higher in patients without any bleeding compared to the patients with any bleeding (36.1% vs 17.2%, p<0.001).
Table 2 shows the procedural index in this population. The patients with any bleeding required more procedure and fluoroscopy time. The puncture approach was significantly lower in patients with any bleeding than in patients without any bleeding. The rate of pre-dilatation was lower in patients without any bleeding than in patients with any bleeding. There were no differences regarding the size of the prosthesis valve and the rate of post-dilatation.
During hospitalisation, the cumulative incidences of any bleeding, transfusion, all-cause death, stroke, MI, and valve thrombosis were 31.4% (n=170), 28.9% (n=156), 1.7% (n=9), 2.6% (n=14), 0.6% (n=3), and 0% (n=0), respectively. The causes of in-hospital all-cause death were cardiovascular disease 37.5% (n=3), respiratory disease 25.0% (n=2), gastrointestinal disease 12.5% (n=1), and unknown 25.0% (n=3). The bleeding sources were access site (n=163), intracranial haemorrhage (n=1), cardiac tamponade (n=5), and gastrointestinal (n=1) (table 3). During the TAVI procedure, the incidence of emergency conversion to open heart surgery was 1.7% (n=9), and the incidence of surgical repair for access site and root, including femoral artery and aorto-iliac artery, was 3.1% (n=17).
The incidence of any bleeding was significantly higher in patients with DAPT compared to those with no pre-procedural antiplatelet therapy and SAPT (36.5% vs 25.5%, p=0.006). According to each antiplatelet therapy, patients with DAPT had a significantly higher incidence of any bleeding than those with SAPT (36.5% vs. 27.5%, p=0.049) and no antiplatelet therapy (36.5% vs. 21.3%, p=0.010). The details of complications during hospitalisation are shown in table 3. There were significant differences only in the incidence of minor bleeding between the patients with DAPT and those with no pre-procedural antiplatelet therapy and SAPT (19.8% vs 12.6%, p=0.024). The patients with DAPT had a significantly higher incidence of bleeding from the access site that required transfusion compared to the patients with no pre-procedural antiplatelet therapy and SAPT (34.1% vs 25.5%, p=0.030; 32.4% vs 24.7%, p=0.048, respectively). According to the incidence of all-cause death, MI, stroke, and valve thrombosis, there were no differences between the patients who received DAPT and those with no pre-procedural antiplatelet therapy and SAPT (table 3).
Multivariable logistic regression analysis was performed to identify predictors of any bleeding in the study population. After adjustment for age (per 1 increase), female sex, BSA (per 0.1 increase), history of atrial fibrillation, congestive heart failure on admission, history of liver disease, previous PCI, chronic kidney disease, percutaneous puncture approach, procedure time (per 1 min increase), and Logistic EuroSCORE II (per 0.1 increase), DAPT before TAVI significantly increased the risk of any bleeding (OR 2.30, 95% CI 1.08 to 4.90, p=0.031) compared to no antiplatelet therapy before TAVI (table 4). None of the VIF values was >2 and the Hosmer-Lemeshow goodness-of-fit test for this model had a p value of 0.512. In the same multivariate model, SAPT before TAVI did not increase the risk of any bleeding compared with no antiplatelet therapy before TAVI (OR 1.13, 95% CI 0.50 to 2.57; p=0.768). Compared with pre-procedural SAPT, DAPT before TAVI increased the risk of any bleeding in the same multivariate model (OR 2.05, 95% CI 1.16 to 3.65; p=0.014).
In the present study, patients with DAPT before TAVI had a significantly higher incidence of bleeding and derived no clinical benefit from reduced thrombotic events compared to patients with SAPT or no pre-procedural antiplatelet therapy who underwent TF-TAVI. Our results indicate that the routine administration of DAPT before TAVI should not be recommended, and that risk stratification is warranted to avoid bleeding events resulting from current clinical practice.
Despite its clinical benefits, TAVI has a potential risk of catastrophic bleeding and ischaemic events during the procedure.9 Moreover, regardless of its origin, major and/or life-threatening bleeding is associated with a high incidence of acute and/or midterm mortality after TAVI.10 Until now, the evidence of antiplatelet and anticoagulant therapy before TAVI has been lacking. In a randomised study, Stabile et al11 demonstrated that 30-days safety endpoint, all-cause mortality, and cardiovascular mortality did not differ between those patients treated with aspirin alone and those receiving DAPT. In a 6 month follow-up randomised study, it was demonstrated that DAPT before TAVI did not reduce the incidence of major adverse cardiac and cerebrovascular events compared with aspirin before TAVI.12 Nevertheless, DAPT has been commonly used before TAVI. Recently, in systematic and pooled analysis, Hassell et al13 reported that there was no difference in 30-day net adverse clinical events, including all-cause mortality, acute coronary syndrome, stroke, and life-threatening and major bleeding, between those patients treated with aspirin alone and those receiving DAPT following TAVI.
In this study, we demonstrated that DAPT before TF-TAVI is associated with an increased risk of bleeding during hospitalisation, compared with pre-procedural SAPT and with no pre-procedural antiplatelet therapy. This result was consistent with previous reports on the assessment of antiplatelet therapy in TF-TAVI.12 ,13 We also demonstrated that SAPT and no pre-procedural antiplatelet therapy did not increase the risk of thrombotic events compared to DAPT in patients undergoing TF-TAVI. Moreover, the alteration to post-procedural DAPT from pre-procedural no antiplatelet therapy and SAPT might reduce the risk of bleeding during hospitalisation.
The candidates for TAVI are elderly and have multiple comorbidities, including renal impairment, pulmonary disease, and atherosclerotic disease. In these elderly patients, it was reported that the combination of clopidogrel and aspirin was associated with a high incidence of major bleeding.14 In this cohort, participants were more likely to have high comorbidity from chronic kidney disease, which was reported to be associated with bleeding during the administration of DAPT.15 Watanabe et al16 reported that the administration of DAPT to all TAVI patients was associated with bleeding risk because there is high prevalence of CYP2C19-induced metabolism abnormalities in Asian populations. These are the reasons why our cohort data had a high incidence of bleeding. The rationale for pre-procedural antiplatelet therapy is based on initial TAVI experience using extracorporeal bypass, which resulted in severe post-procedural thrombocytopenia in patients without clopidogrel.17 Platelet activation and consumption secondary to an interaction between the bioprosthesis and blood flow was responsible for the prolonged and severe thrombocytopenia. A previous histopathological study demonstrated that it takes about 3 months to achieve endothelialisation and integration of the valve stent in the porcine pericardial bioprosthesis of the CoreValve.18 Therefore, DAPT for 3 months or longer was thought to reduce the post-procedural thromboembolic events after valve implantation. However, current studies failed to establish that DAPT reduced adverse events and that SAPT or no pre-procedural antiplatelet therapy did not increase the thrombotic events in TF-TAVI. Even in modern practice, there is the possibility of emergency surgical conversion during the TAVI procedure. In our data, 1.7% of cases required surgical conversion and these data were consistent with previous data.19 This report showed that emergency surgical conversion during TAVI is rare but is related to a high incidence of mortality. Until now, there were no available data about the clinical impact of pre-procedural antiplatelet therapy on perioperative adverse events in patients who required emergent surgical conversion during TAVI. In the multicentre retrospective analysis of urgent coronary artery bypass grafting in patients with acute coronary syndrome, patients who undergo surgery within 5 days of receiving clopidogrel are at high risk for reoperation, major bleeding, and prolongation of hospital stay.20 Because the candidates who undergo TAVI have serious comorbidities, the increased risks of reoperation, major bleeding, and long hospitalisation are associated with a high mortality risk in patients who require emergent surgical conversion after TAVI. In multivariate analysis, previous PCI was not associated with a high incidence of bleeding. This might be the reason why almost all of the previous PCIs had been done more than 1 year ago and DAPT for coronary stenting has been changed to SAPT. In fact, the number of previous PCIs was different from the number of pre-procedural DAPT in our cohort. Therefore, the previous PCI had no impact on bleeding events.
Even though the patients with no pre-procedural antiplatelet therapy had high operative risk, the cumulative incidence of adverse events was lower than in patients with pre-procedural antiplatelet therapy. Based on the possibility of emergent surgical conversion, our data indicate that the combination of pre-procedural SAPT or no antiplatelet therapy and post-procedural DAPT could be an appropriate strategy for TF-TAVI.
There were several limitations in this study. First, this is a retrospective analysis of prospective multicentre cohort registry data. Second, the administration of pre- and post-procedural antiplatelet therapy was not randomised; therapeutic choice was left to the discretion of the doctors in each hospital, because there is no consensus guideline of antiplatelet therapy for TAVI in Japan. Third, the number of patients with concomitant atrial fibrillation (AF) in our study population was relatively low (13.3%) compared to previous reports.21–23 However, originally the OCEAN-TAVI registry registered 150 patients (20%) who had pre-existing AF and this was consistent with previous European registry data.24 Although there is a possibility of under- or mis-diagnosis of pre-existing AF due to a lack of routine continuous ECG monitoring before TAVI, our data might be representative of the real-world severe AS population undergoing TAVI. Fourth, there were differences in the baseline characteristics between patients with no pre-procedural antiplatelet therapy, SAPT and DAPT because of selection bias. However, since our study is based on observational registry data, we believe that this represents a real-world unselected population of patients with severe aortic valve stenosis undergoing TAVI. Finally, we did not assess the mid-term or long-term outcomes in patients with DAPT, SAPT, or no pre-procedural antiplatelet therapy. Until now, only the comparison between DAPT and SAPT has been assessed and revealed that TAVI with SAPT could be performed safely through mid-term follow-up.11 A longer follow-up study is required to evaluate the impact of pre-procedural SAPT or no antiplatelet therapy in patients undergoing TF-TAVI.
In the modern TAVI era, it may be reasonable to perform TAVI with pre-procedural single or no antiplatelet therapy without increasing the risk of thrombotic events. Therefore, the combination of post-procedural DAPT with lower intensity pre-procedural antiplatelet therapy might be considered in order to reduce the incidence of bleeding complications in TAVI. Further long-term follow-up study is required to evaluate the clinical benefit of pre-procedural single or no antiplatelet therapy in TAVI.
What is already known on this subject?
No prospective, randomised, large study has been undertaken to investigate the appropriate antiplatelet regimen before transcatheter aortic valve implantation (TAVI). A recent report demonstrated that TAVI with aspirin alone could be performed without an increase in thrombotic events compared with TAVI with dual antiplatelet therapy (DAPT). However, there are no data about the risk or benefit of no pre-procedural antiplatelet therapy in TAVI.
What might this study add?
DAPT before TAVI was associated with an increased risk of bleeding compared to single antiplatelet therapy (SAPT) (OR 2.05, 95% CI 1.16 to 3.65; p=0.014) and no pre-procedural antiplatelet therapy (OR 2.30, 95% CI 1.08 to 4.90; p=0.031). Interestingly, SAPT or no pre-procedural antiplatelet therapy did not increase the risk of thrombotic events compared to pre-procedural DAPT (3.6% vs 1.7%, p=0.158).
How might this impact on clinical practice?
TAVI with pre-procedural single or no antiplatelet therapy could be performed without an increase in thrombotic events compared with conventional pre-procedural DAPT. In modern TAVI practice, pre-procedural DAPT might not be required in all cases.
The authors thank the investigators and institutions that have participated in the OCEAN-TAVI registry.
Contributors Conception and design—YW, MY, KT, MA, NT, SS, FY, and KH. Acquisition of data—HH, YW, YN, HK, AK, MY, KT, MA, NT, SS, FY, and KH. Analysis and interpretation—HH, YW, KK. Drafting of the manuscript—HH. Critical revision of the manuscript—YW, KK, MY, and KH. Corresponding author—HH.
Competing interests YW is a proctor for transfemoral-TAVI for Edwards Lifesciences. MY is a proctor for transfemoral-TAVI for Edwards Lifesciences. MA is a proctor for transfemoral-TAVI for Edwards Lifesciences. NT is a proctor for transfemoral-TAVI for Edwards Lifesciences. SS is a proctor for transfemoral-TAVI for Edwards Lifesciences. KH is a proctor for transfemoral-TAVI for Edwards Lifesciences.
Patient consent Obtained.
Ethics approval Ethics committee of each participating hospital.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.