If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Cardiac Surgery Department, University Hospital of Zurich, Zurich, SwitzerlandPostgraduate Program in Health Sciences: Cardiology and Cardiovascular Sciences, Universidade Federal do Rio Grande do Sul, Porto Alegre, Brazil
Cardiac Surgery Department, University Hospital of Zurich, Zurich, SwitzerlandFundación Interhospitalaria para la Investigación Cardiovascular, Madrid, Spain
Cardiology Department, University Hospital of Zurich, University of Zurich, Zurich, Switzerland. Cardiology Department, Cardiance Clinic, Pfaeffikon, Switzerland
Considering that there is a lack of evidence and guideline-based recommendations on the best preoperative oral anticoagulation management (OAC) for transcatheter aortic valve implantation (TAVI), this cohort study aimed to evaluate bleeding, access site complications, and early safety in patients undergoing TAVI on continued OAC therapy vs no-OAC therapy. Three-hundred forty-four patients submitted to a TAVI procedure (66.3% no-OAC vs 33.7% OAC) were consecutively enrolled. Primary endpoint was defined as in-hospital VARC-2 life-threatening or disabling bleeding. Secondary endpoints were in-hospital VARC-2 major vascular complications and VARC-2 early safety at 30 days. Propensity score matching analysis was performed to reduce potential distribution bias, resulting in 2 well-balanced groups (92 patients in each arm). In the overall cohort, mean age, median EuroScore II, and STS-score were 78.7±7.6 years, 2.9% (1.7-5.9), and 2.3% (1.6-3.6), respectively. Despite being older (78 ± 8 vs 80 ± 6, p = 0.004) and having higher STS score (2.1 vs 2.6, p = 0.001), patients on OAC had similar incidence of in-hospital VARC-2 life-threatening or disabling bleeding (1.3% vs. 0.9%, p = 0.711), major vascular complications (4.8% vs 5.2%, p = 0.888), and VARC-2 early safety at 30 days (10.1% vs 12.1%, p = 0.575). No significant differences in the main outcomes were observed when propensity score matching was applied. In conclusion, the management of patients on OAC submitted to a TAVI procedure is challenging and requires balancing the risk of bleeding with the risk of thromboembolic events. The present study suggests that continued OAC was not associated with increased in-hospital VARC-2 life-threatening or disabling bleeding, major vascular complications, and VARC-2 early safety at 30 days.
Antithrombotic or anticoagulation therapy following transcatheter aortic valve implantation (TAVI) has been recently investigated in the POPular TAVI trial cohorts A and B, respectively.
Results have suggested the superiority of single antiplatelet therapy (aspirin) in patients without indication for oral anticoagulation (OAC), and OAC alone in those with an OAC indication. These trials, however, focused on post-TAVI management and did not investigate the optimal perioperative regimen. Current European guidelines recommend vitamin K antagonists (VKA) discontinuation 5 days prior to elective cardiac surgery, aiming an INR (international normalized ratio) value <1.5 on the procedure day. Preoperative direct OAC (DOAC) discontinuation is recommended at least 48-96 hours before the procedure. However, no specific recommendation on how to manage these medications on the TAVI context has been provided.
Trying to clarify this issue, a retrospective study showed that, in patients with atrial fibrillation submitted to transfemoral TAVI, the lowest rates of early safety (with lower values indicating superior safety) and 1-year mortality were observed in the continued DOAC group compared to continued or interrupted VKA. Continued VKA had similar incidence of bleeding and access site complication than interrupted VKA.
Continued versus interrupted oral anticoagulation during transfemoral transcatheter aortic valve implantation and impact of postoperative anticoagulant management on outcome in patients with atrial fibrillation.
] were compared to warfarin in patients with atrial fibrillation. Considering the lack of evidence and guideline-based recommendations on the best pre-TAVI OAC management, this study aims to evaluate the impact of continued OAC. For that, we compared patients receiving continued OAC with those not receiving any OAC therapy in terms of bleeding, vascular complications, and early safety.
Methods
From January 2019 to July 2020, 344 patients submitted to TAVI in a single center were consecutively enrolled. Patients were analyzed according to the perioperative OAC regimen. Group 1 (no-OAC): patients who were not on OAC at the time of the index procedure. Group 2 (OAC): patients who were on OAC (either VKA or DOAC) at the time of the index procedure. This comparison was possible since, as a local institutional routine, TAVI procedures are systematically performed without interruption of the ongoing antithrombotic or OAC regimen. Patients on VKA therapy were kept on VKA therapy during the whole periprocedural phase with an INR target range between 2 and 3, while patients on DOAC remained receiving the therapy during the whole procedural period, without skipping doses. TAVI indication was driven by the institutional heart team and patients provided written informed consent before the procedure. All patients undergoing TAVI in our institution are included in the nationwide Swiss TAVI Registry approved by local ethic committees.
During TAVI, intravenous heparin was administered, adjusted to baseline activated clotting time, aiming to reach an activated clotting time above 250 seconds in both groups. Decision about anesthetic approach and the access route were taken based on patient and devices characteristics, local expertise, and operator discretion. Cerebral embolic protection, using the Sentinel system (Boston Scientific, Marlborough, MA, USA), was routinely used. First choice access site was percutaneous transfemoral, performed through ultrasound-guided puncture. Second choice access was left subclavian artery and the third left carotid artery or transapical, performed by open surgical dissection. In the case of transfemoral route, closure devices, Perclose/ProGlide system (Abbott Vascular, CA, USA) or Manta (Essential Medical Inc, Exton, PA), were used. TAVIs were performed using CE-approved devices. The Medtronic platform (Medtronic Inc., Minneapolis, MN, USA) has an integrated InLine Sheath (14F for the Evolut R 23, 26, 29 mm, and 16F for Evolut Pro and Evolut R 34 mm). The Edwards Sapien platform (Edwards Lifesciences, Irvine, CA, USA) was introduced through a 14F or16F eSheath, which has a dynamic expansion mechanism (14F for 20, 23, 26 mm, and 16F for 29 mm). The Portico (Abbott Vascular, Santa Clara, CCA, USA) previous generation was introduced sheathless, while the new FlexNav system has an integrated 14 or15F sheath (14F for 23, 25 mm, and 15F for 27, 29 mm). The Acurate-Neo (Boston Scientific Corporation, Natick, MA, USA) was introduced through a 14F iSLEEVE expandable introducer. The Lotus Edge (Boston Scientific Corporation, Natick, MA, USA) and the Allegra (New Valve Technology, Hechingen, Germany) platforms were used with an 18F sheath.
Baseline characteristics, procedural data, and outcomes were prospectively collected and adjudicated according to the Valve Academic Research Consortium updated criteria (VARC-2).
Primary endpoint was defined as in-hospital VARC-2 life-threatening or disabling bleeding. Secondary endpoints were in-hospital VARC-2 major vascular complications, and VARC-2 early safety at 30 days (a composite of all-cause mortality, all-stroke, life-threatening bleeding, acute kidney injury stage 2 or 3, coronary artery obstruction requiring intervention, major vascular complication, and valve-related dysfunction requiring repeat procedure). Preoperative CHA2DS2-VASc and HAS-BLED scores were calculated as recommended by current guidelines.
Sudden cardiac death was defined as unexpected death from a presumptively cardiac cause that occurs in a short time period, generally within 1 hour of symptom onset or without prior symptoms.
Descriptive data were expressed as mean ± standard deviation (SD) or median and interquartile range (IQR). Statistical analyses were performed using the statistical package SPSS 25.0 software (IBM Corporation, Armonk, New York). Categorical variables were analyzed using Chi-square test, and continuous variables using Student's t-test or Mann-Whitney U test, according to the distribution pattern. A two-sided p-value lower than 0.05 was considered significant for all analyses. In order to reduce bias, a propensity score matching analysis was performed. Variables related to the primary endpoint or with an unequal distribution between groups (p<0.2) were included in a 1:1 nearest neighbor model. The caliper width was equal to 0.15 of the standard deviation of the logit of the propensity score. The effectiveness of balancing was analyzed with propensity score histograms, estimating the reduction in the standardized percentage bias and performing Chi-square and t-test between groups.
Results
Among the overall 344 evaluated patients, 116 (33.7%) underwent TAVI receiving continued OAC therapy, while 228 (66.3%) did not receive any OAC therapy. Applying propensity score matching, two well-balanced groups, with 92 patients in each one, were generated. Baseline characteristics and echocardiographic parameters for the overall cohort are presented in Table 1, and for the propensity score-matched cohort in Table 2. Before propensity score matching, patients in the OAC group were older, had higher EuroScore II and STS-score, and also higher previous permanent pacemaker prevalence, but lower previous acute myocardial infarction prevalence. As expected, patients on OAC had significantly higher CHA2DS2Vasc score (3.8 ± 1.3 vs 4.4 ± 1.2, p < 0.001) with similar HAS-BLED score (2.4 ± 1 vs 2.6 ± 0.98, p = 0.303).
Table 1Baseline characteristics and echocardiographic parameters before propensity score matching
Variable
All (n = 344)
OAC
p value
No (n = 228)
Yes (n = 116)
Age (years)
78.7 ± 7.6
78 ± 8.1
80.4 ± 6.1
0.004
Men
214 (62.2%)
149 (65.4%)
65 (56%)
0.092
Body mass index (Kg/m2)
26.4 ± 4
26.4 ± 4.7
26.5 ± 4.7
0.921
EuroScore II (%)
2.9 (1.7-5.9)
2.7 (1.55-5)
3.4 (2.4-7.2)
0.001
STS-score (%)
2.3 (1.6-3.6)
2.1 (1.5-3.3)
2.6 (2.0-3.8)
0.001
CHA2DS2-Vasc score ≥3
311 (90.4%)
199 (87.3%)
112 (96.6%)
0.006
HAS-BLED score ≥3
157 (45.6%)
100 (43.9%)
57 (49.1%)
0.353
NYHA 3/4
185 (53.8%)
114 (50%)
71 (61.2%)
0.214
Coronary artery disease
170 (49.4%)
117 (51.3%)
53 (45.7%)
0.324
Previous myocardial infarction
51 (14.8%)
41 (18%)
10 (8.6%)
0.021
Previous coronary artery bypass grafting
35 (10.2%)
22 (9.6%)
13 (11.2%)
0.651
Previous percutaneous coronary intervention
103 (29.9%)
76 (33.3%)
27 (23.3%)
0.054
Previous pacemaker
30 (8.7%)
10 (4.4%)
20 (17.2%)
<0.001
Arterial hypertension
248 (72.1%)
164 (71.9%)
84 (72.4%)
0.925
Diabetes Mellitus
87 (25.3%)
57 (25%)
30 (25.9%)
0.827
Previous stroke
30 (8.7%)
18 (7.9%)
12 (10.3%)
0.446
Peripheral artery disease
40 (11.6%)
25 (11%)
15 (12.9%)
0.591
Chronic obstructive lung disease
47 (13.7%)
29 (12.7%)
18 (15.5%)
0.475
Active smoker
121 (35.2%)
89 (39%)
32 (27.6%)
0.036
Chronic kidney disease
125 (36.3%)
77 (33.8%)
48 (41.4%)
0.165
Indication
0.058
Native valve
330 (95.9%)
221 (96.9%)
109 (94%)
Valve-in-valve
14 (4.1%)
7 (3.1%)
7 (6%)
Echocardiographic variables
Aortic valve main disease
0.107
Regurgitation
10 (2.9%)
9 (3.9%)
1 (0.9%)
Stenosis
334 (97.1%)
219 (96%)
115 (99.1%)
Bicuspid aortic valve
25 (7.3%)
19 (8.3%)
6 (5.2%)
0.286
Left ventricular ejection fraction (%)
54.4 ± 13
54.8 ± 12
53.5 ± 13
0.378
Mean aortic valve gradient (mm Hg)
44.5 ± 16
45 ± 15
44 ± 17
0.06
Peak aortic valve gradient (mm Hg)
69.9 ± 24
70 ± 23
69.7 ± 25
0.928
Variables are expressed as numbers (%), mean (±SD) or median (IQR). NYHA = new york heart association.
Preoperative OAC or antithrombotic regimes are presented in supplementary Table 1. The main reasons for preoperative OAC indication were: atrial fibrillation (n = 107/116, 92.2%), left ventricle thrombus (n = 2/116, 1.7%), prior thromboembolism (n = 5/116, 4.3%), mechanical mitral valve (n = 1/116, 0.8%), and prior thrombus in the aortic valve (n = 1/116, 0.8%). DOAC were taken by 87 patients (87/116, 75%) and warfarin by 29 patients (29/116, 25%).
Type of transcatheter heart valve used and other periprocedural features are presented in Table 3. Baseline and post-procedural laboratory values are displayed in supplementary Table 2. In patients on VKA, INR value was on target on the day before (2.3 ± 0.5) and after the intervention (2.4 ± 0.67).
Table 3Procedural characteristics
Variable
Overall
p value
Propensity Score Matching
p value
OAC
OAC
All (n = 344)
No (n = 228)
Yes (n = 116)
No (n = 92)
Yes (n = 92)
Conscious sedation
318 (92.4%)
211 (92.5%)
107 (92.2%)
0.480
85 (92.4%)
86 (93.5%)
0.563
Vascular access
0.531
0.361
Right Transfemoral
283 (82.2%)
186 (81.5%)
97 (83.6%)
78 (84.8%)
81 (88.1%)
Left Transfemoral
46 (13.3%)
33 (14.5%)
13 (11.2%)
12 (13.1%)
7 (7.6%)
Left subclavian artery
12 (3.5%)
6 (2.6%)
6 (5.2%)
2 (2.2%)
4 (4.3%)
Transapical
2 (0.5%)
2 (0.9%)
-
-
-
Left carotid artery
1 (0.3%)
1 (0.4%)
-
-
-
TAVI device
0.826
0.503
Portico
116 (33.7%)
79 (34.6%)
37 (31.9%)
30 (32.6%)
30 (32.6%)
Sapien 3/Ultra
112 (32.5%)
75 (32.9%)
37 (31.9%)
23 (25%)
30 (32.6%)
Evolut R/PRO
80 (23.2%)
51 (22.3%)
29 (25%)
30 (32.6%)
20 (21.7%)
Acurate-neo
25 (7.3%)
16 (7%)
9 (7.8%)
5 (5.4%)
8 (8.7%)
Other
11 (3.2%)
7 (3.1%)
4 (3.5%)
4 (4.3%)
4 (4.3%)
Pre-dilatation
141 (40.9%)
96 (42.1%)
45 (38.8%)
0.509
36 (39.1%)
37 (40.2%)
0.880
Post-dilatation
45 (13%)
27 (11.8%)
18 (15.5%)
0.147
6 (6.5%)
14 (15.2%)
0.058
Procedure time (min)
54 (44-68)
53 (44-67)
55 (43-69)
0.974
52 (43-67)
53 (43-67)
0.713
Contrast dye (mL)
88 (70-122)
88 (73-120)
87 (67-125)
0.648
94 (71- 121)
86 (65- 124)
0.489
Variables are expressed as numbers (%), mean (± SD) or median (IQR).
In-hospital clinical and echocardiographic outcomes are summarized in Table 4. In the overall cohort, the primary outcome in-hospital VARC-2 life-threatening or disabling bleeding was not increased in the OAC group (1.3% no-OAC vs 0.9% OAC, p = 0.711) (Figure 1), as well as the secondary outcomes in-hospital VARC-2 major vascular complications (4.8% no-OAC vs 5.2% OAC, p = 0.888) and VARC-2 early safety at 30-days (10.1% no-OAC vs 12.1% OAC, p = 0.575). The overall rates of VARC-2 all-stroke (1.3% no-OAC vs 2.5% OAC, p = 0.398) and any red blood cell transfusion were also not statistically different between groups (1.8% no-OAC vs 2.6% OAC, p = 0.605). Similar results were observed in the propensity score-matched cohort, with a VARC-2 early safety at 30-days of 10.9% in the no-OAC and 14.1% in the OAC group (p = 0.504) (Table 4). Since only 4 events were reported for the primary endpoint, a multivariate analysis could not be performed.
Table 4In-hospital outcomes
Variable
Overall
p value
Propensity Score Matching
p value
OAC
OAC
All (n = 344)
No (n = 228)
Yes (n = 116)
No (n = 92)
Yes (n = 92)
VARC-2 all-bleeding
46 (13.3%)
27 (11.8%)
19 (16.3%)
0.338
13 (14.1%)
17 (18.5%)
0.425
Minor
19 (5.5%)
9 (3.9%)
10 (8.6%)
0.073
5 (5.4%)
8 (8.7%)
0.388
Major
23 (6.7%)
15 (6.6%)
8 (6.9%)
0.911
7 (7.6%)
8 (8.7%)
0.788
Life-threatening or disabling
4 (1.2%)
3 (1.3%)
1 (0.9%)
0.711
1 (1.1%)
1 (1.1%)
1.0
VARC-2 vascular complications
75 (21.8%)
48 (21%)
27 (23.2%)
0.892
17 (18.5%)
20 (21.7%)
0.581
Minor
58 (16.9%)
37 (16.2%)
21 (18.1%)
0.661
13 (14.1%)
15 (16.3%)
0.681
Major
17 (4.9%)
11 (4.8%)
6 (5.2%)
0.888
4 (4.4%)
5 (5.4%)
0.733
VARC-2 all-stroke
6 (1.7%)
3 (1.3%)
3 (2.5%)
0.398
3 (3.2%)
3 (3.2%)
0.549
Disabling
-
-
-
-
-
Non-disabling stroke
5 (1.4%)
3 (1.3%)
2 (1.7%)
3 (3.2%)
2 (2.2%)
TIA
1 (0.3%)
-
1 (0.9%)
-
1 (1.1%)
VARC-2 myocardial infarction
2 (0.6%)
2 (0.9%)
-
0.312
2 (2.2%)
-
0.155
Second valve implant
5 (1.4%)
2 (0.9%)
3 (2.6%)
0.211
-
3 (3.2%)
0.081
Surgical conversion
1 (0.3%)
1 (0.4%)
-
0.475
1 (1.1%)
-
0.316
Any red packed blood cells
7 (2%)
4 (1.8%)
3 (2.6%)
0.605
2 (2.2%)
3 (3.2%)
0.650
New permanent pacemaker
51 (14.8%)
37 (16.2%)
14 (12.1%)
0.305
16 (17.4%)
11 (11.9%)
0.298
New atrial fibrillation
11 (3.2%)
10 (4.4%)
1 (0.9%)
0.079
1 (1.1%)
1 (1.1%)
1.0
Delirium
14 (4%)
11 (4.8%)
3 (2.6%)
0.321
6 (6.5%)
3 (3.2%)
0.305
All-cause mortality
6 (1.7%)
1 (0.4%)
5 (4.3%)
0.01
1 (1.1%)
4 (4.3%)
0.174
Hospital length of stay (days)
5 (4-7)
5 (4-7)
6 (4-7)
0.741
6 (4-8)
6 (4-7)
0.243
Echocardiographic variables
Left ventricular ejection fraction (%)
55.5 ± 11.5
55.3 ± 11
55.8 ± 11
0.683
55.1 ± 11.5
56.1 ± 10.9
0.565
Residual mean gradient (mmHg)
8 ± 3.9
8.2 ± 3.9
7.7 ± 3.8
0.330
7.9 ± 3.7
7.8 ± 3.9
0.905
Residual peak gradient (mmHg)
15 ± 6.9
15.9 ± 7
14.3 ± 6.5
0.076
15.5 ± 7.1
14.5 ± 6.7
0.407
Residual aortic regurgitation ≤Mild
320 (93%)
216 (94.7%)
105 (90.5%)
0.165
86 (93.4%)
83 (90.2%)
0.117
Variables are expressed as numbers (%), mean (SD) or median (IQR). TIA = transient ischemic attack.
In the overall cohort, 30-day all-cause mortality was numerically higher in the OAC group (0.9% no-OAC vs 4.3% OAC, p = 0.033). However, when propensity score matching was applied, the difference in 30-day all-cause mortality lost statistical significance (1.1% no-OAC vs 4.3% OAC, p = 0.174). Death's specific reasons are presented in supplementary Table 3. On the last contact [median follow-up 80 days (36-262 days)], survival rate was similar between the groups [96.9% no-OAC vs 94% OAC, p = 0.188]. Kaplan Meier survival curves are presented in Supplementary Figures 1 and 2.
Discussion
In the last decade, TAVI has evolved from a procedure of exception, reserved for inoperable and high-risk patients, to a well-established intervention, even in low-risk populations. Herein, we reported early outcomes after TAVI according to the perioperative OAC regimen. The main differential of the present article lies in the discussion about perioperative TAVI OAC management, a controversial topic that lacks guideline-based recommendations. This issue is especially relevant once the 2 main trials recently published on TAVI antithrombotic or anticoagulant therapy focused only on post-procedural management.
One of the first evidences supporting the safety of continued OAC in patients with atrial fibrillation submitted to transfemoral TAVI was published in 2019 by Mangner et al. In this study, the lowest rates of VARC-2 early safety (13.2% DOAC vs 19.7% continued VKA vs 23.1% interrupted VKA, p = 0.029) and 1-year all-cause mortality (8.8% DOAC vs 13.7% continued VKA vs. 20.1% interrupted VKA, p = 0.015) were observed in the continued DOAC group. Continued VKA regimen had similar outcomes compared to interrupted VKA in terms of all procedural bleeding (38.8% continued VKA vs. 34.8% interrupted VKA, p = 0.097) and access site complications (40.5% continued VKA vs 32.1% interrupted VKA, p = 0.661).
Continued versus interrupted oral anticoagulation during transfemoral transcatheter aortic valve implantation and impact of postoperative anticoagulant management on outcome in patients with atrial fibrillation.
Similarly to Mangner's study, our results supported the safety of the continued OAC strategy. Our VARC-2 early safety in patients receiving continued OAC (12.1%) was at least comparable to that presented by Mangner, whereas our in-hospital VARC-2 all-bleeding rate (16.3%) was even lower (27.5% in continued DOAC and 38.8% in continued VKA). On the other hand, while in the present study the majority of patients on continued OAC were receiving DOAC (75%), in Mangner's study the DOAC therapy had a lower prevalence (60%). Furthermore, we included also, even in a minority, non-transfemoral routes and patients receiving OAC due to reasons other than atrial fibrillation. Besides these points, the main difference between Mangner's and our study is that, while the former compared continued versus interrupted OAC strategies only in patients previously receiving OAC, we compared continued OAC in patients with indication and previously receiving OAC to those not previously receiving any OAC therapy. Thus, if the continued OAC strategy had been associated with worse outcomes, our study would be more likely to detect a difference in bleeding rates since it compared ongoing OAC with a lower bleeding risk control group. Therefore, this study helps to answer the question: in a patient receiving continued OAC, should we expect worse outcomes after TAVI compared to those not receiving any OAC therapy?
In this same line, a letter from Brinker et al. reported that in patients on continued (186 patients) or interrupted (185 patients) OAC regimen submitted to transfemoral TAVI, the rates of periprocedural major or life-threatening bleeding (10.2% vs 10.8%, p = 0.85), major vascular complications (8.6% vs 10.3, p = 0.58), periprocedural stroke (0.6% vs 3.2%, p = 0.06), and 1-year mortality (9.38% vs 9.83%, p = 0.897) did not have statistical difference.
These same authors recently published an update on the previous analysis with 584 patients receiving continued and 733 interrupted OAC. At 30 days, major or life-threatening bleedings (11.3% vs 14.3%, p = 0.39) and major vascular complications rates (11.0% vs 12.3%, p = 0.52) were similar, but packed red blood cell transfusion was less frequent in the continued group (13.7% vs 17.7%, p = 0.001).
Two meta-analyses comparing continuous OAC versus heparin bridging in patients undergoing cardiac implantable electronic devices had suggested that OAC maintenance did not increase procedural adverse events. In the first, continued VKA was associated with significant lower postoperative bleeding risk [odds ratio (OR) 0.25, 95% Confidence Interval (CI) 0.17-0.36, p<0.001] and no difference in thromboembolic events (OR 1.86, 95% CI 0.29-12.17, p = 0.57).
In the second, uninterrupted OAC was associated with significantly lower bleeding risk (OR 0.31, 95% CI 0.18-0.53, p < 0.0001) and no difference in thromboembolic risk (OR 0.82, 95% CI 0.32-2.09, p = 0.65).
Continued versus interrupted oral anticoagulation during transfemoral transcatheter aortic valve implantation and impact of postoperative anticoagulant management on outcome in patients with atrial fibrillation.
Although the present study had not observed differences in both, primary and secondary outcomes, patients on OAC presented a higher 30-day mortality rate (0.9% no-OAC vs 4.3% OAC, p = 0.033). Nonetheless, only 1 of the 5 deaths observed in the OAC group could be associated or worsened by anticoagulation (intraprocedural cardiac tamponade). It is relevant to highlight that when propensity score matching was performed, OAC was not associated with higher 30-day mortality rate (p = 0.174).
Since DOAC has largely replaced VKA to prevent thrombotic events in atrial fibrillation, which is present in a significant number of patients undergoing TAVI (16%–59%),
we performed also an analysis comparing DOAC to VKA. In this analysis, there's no significant difference in terms of main outcomes between the 2 regimens (Supplementary Table 5).
We would like to highlight that, the present analysis reflects a single-center, non-randomized, but prospectively acquired experience. Hence, all the inherent limitations of such design need to be taken into account. Besides, this study focused on short-term results. Properly designed trials with long-term follow-up are required to confirm the best pre- and post-TAVI anticoagulant management. The low number of observed events limits the statistical power of the logistic regression model. Therefore, even though continued OAC had not been associated with increased outcomes, it should be noted that we had a very low number of events, which may difficult a more generalized or definitive conclusion. Lastly, however DOAC comprised different active ingredients, with a predominance of rivaroxaban, clinical trials evaluating the safety and efficacy of DOACs have indicated a class rather than a specific drug effect.
In conclusion, the management of patients on OAC submitted to a TAVI procedure is challenging and requires balancing the risk of bleeding with the risk of thromboembolic events. The present study suggests that continued OAC was not associated with increased in-hospital VARC-2 life-threatening or disabling bleeding, major vascular complications, and VARC-2 early safety at 30 days.
Authors Contribution
Ana Paula Tagliari: Conceptualization; Data curation; Formal analysis; Writing − Original Draft; Daniel Perez-Camargo: Conceptualization; Data curation; Formal analysis; Writing − Original Draft; Enrico Ferrari: Formal analysis; Supervision; Validation; Writing - Review & Editing; Philipp K. Haager: Formal analysis; Supervision; Validation; Writing - Review & Editing; Lucas Jörg: Supervision; Validation; Marco Gennari: Supervision; Validation; Mi Chen: Supervision; Validation; Mara Gavazzoni: Supervision; Validation; Ahmed Aziz Khattab: Supervision; Validation; Stefan Blöchlinger: Supervision; Validation; Francesco Maisano: Supervision; Validation; Maurizio Taramasso: Conceptualization; Formal analysis; Supervision; Writing - Review & Editing.
Acknowledgment
The authors thank Malik Riva and Leonora Kodzadziku for supporting TAVI procedures.
Disclosures
Ana Paula Tagliari reports a relationship with CAPES - Brazil that includes: funding grants. Enrico Ferrari reports a relationship with Consultant for Edwards Lifesciences and received Grants and Research Support from Edwards Lifesciences, Medtronic and Somahlution that includes: consulting or advisory and funding grants. Marco Gennari reports a relationship with Consultant for Medtronic. that includes: consulting or advisory. Mara Gavazzoni reports a relationship with Consultant for Biotronik and Abbott. that includes: consulting or advisory. Francesco Maisano reports a relationship with Abbott, Medtronic, Edwards Lifesciences, Biotronik, Boston Scientific Corporation, NVT, Terumo, SwissVortex, Perifect, Xeltis, Transseptal solutions, Cardiovalve, Magenta, CardioGard, SwissVortex, 4 Tech that includes: consulting or advisory, funding grants, and speaking and lecture fees. Maurizio Taramasso reports a relationship with Abbott Vascular, Boston Scientific, 4 Tech, Edwards Lifesciences, CoreMedic, SwissVortex, Mitraltech and Somahlution that includes: consulting or advisory, funding grants, and speaking and lecture fees. APT has received a Research Grant from Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brasil (Capes) - Finance Code 001. EF is consultant for Edwards Lifesciences and received Grants and Research Support from Edwards Lifesciences, Medtronic and Somahlution. M Gennari is consultant for Medtronic. M Gavazzoni is consultant for Biotronik and Abbott. FM received Research Support and/or Grant from Abbott, Medtronic, Edwards Lifesciences, Biotronik, Boston Scientific Corporation, NVT, Terumo; Consulting fees and/or Honoraria from Abbott, Medtronic, Edwards Lifesciences, SwissVortex, Perifect, Xeltis, Transseptal solutions, Cardiovalve, Magenta; has Royalty Income/IP Rights Edwards Lifesciences and is Shareholder of CardioGard, Magenta, SwissVortex, Transseptal Solutions, 4 Tech, Perifect. MT is consultant for Abbott Vascular, Boston Scientific, and 4 Tech; and has received Consulting fees from Edwards Lifesciences, CoreMedic, SwissVortex, and Mitraltech.
Continued versus interrupted oral anticoagulation during transfemoral transcatheter aortic valve implantation and impact of postoperative anticoagulant management on outcome in patients with atrial fibrillation.
00265-4&id=gr1.jpg){kind=link}