Keywords
mitral regurgitation - M-TEER - bleeding - thromboelastography - platelet function
Introduction
Transcatheter mitral valve repair is an established treatment for multimorbid patients
suffering from symptomatic severe mitral valve regurgitation who are at high surgical
risk. Previous studies have shown the safety and efficacy of this procedure.[1]
[2] Nevertheless, the transcatheter procedure is also associated with several complications.
Bleeding complications are one of the most common adverse events occurring in up to
22% of patients[3]
[4] and lead to a prolonged hospital stay.[5] Extensive bleeding can also increase mortality.[4] On the other hand, preventing thromboembolic events is also important. Stroke rates
after mitral transcatheter edge-to-edge repair (M-TEER) vary from 0.7 to 2.6% within
30 days of the intervention.[6] Up to 80% of patients suffer from atrial fibrillation or flutter (AF) and there
is a high AF burden particularly in those patients with functional mitral regurgitation
(MR).[7] Yet, there is no well-established standardized anticoagulation regimen for these
patients.[8] Premedication with antithrombotic drugs due to other indications (e.g., AF, history
of percutaneous coronary intervention [PCI]) is common in this patient cohort. M-TEER
might lead to change in flow conditions in the left atrium and shear stress causing
enhanced thrombogenicity and platelet activation. For transcatheter aortic valve implantation
(TAVI), several studies have demonstrated the effect of this procedure on platelet
function and coagulation.[9]
[10] These studies showed that coagulation capacity increased while platelet function
decreased,[9] and that low strength of fibrin clot was associated with life-threatening bleeding
complications.[10] The impact on hemostatic parameters after implantation of the PASCAL system has
not been examined so far.
The primary objective of this study was to evaluate the occurrence of bleeding complications
after 6 months. Six months is the usual time frame for follow-up examinations after
M-TEER to evaluate the long-term result with echocardiography. Additionally, we investigated
whether any of the TEG or Multiplate parameters could be used as an independent predictor
of bleeding complications.
Methods
Study Design and Population
In this single-center study, we analyzed 100 consecutive patients with MR (primary,
secondary, and mixed etiology) receiving M-TEER with the PASCAL system after Heart
Team decision at the Department of Cardiology of the Medical University Tübingen,
Germany between July 2019 and December 2022. According to the instructions for use
and our procedural protocol, the guide catheter (GC) was fully aspirated before positioning
of the steering catheter in the left atrium and before removal of the GC to avoid
air-/thromboembolism. Blood from the left atrium that would otherwise have been discarded
was used for the analysis of thrombogenicity. After the implantation of the device,
the second blood sample was taken from the left atrium also via aspiration. With these
two blood samples, we performed thromboelastography (TEG) analysis and Multiplate
impedance aggregometry. Other relevant information such as laboratory parameters (hemoglobin,
glomerular filtration rate, etc.), periprocedural anticoagulation treatment, comorbidities,
bleeding events, and follow-up examinations was collected in a specifically designed
electronic database. The follow-up was set as scheduled hospital visit 6 months after
M-TEER with a clinical examination of the patient and a transthoracic echocardiographic
examination. We classified bleeding according to the Bleeding Academic Research Consortium
(BARC) definition.[11] We categorized patients in two groups considering the outcome: no bleeding, which
corresponds to BARC = 0, and bleeding event occurred (BARC ≥ 1). To describe the individual
bleeding risk of the patient, we used the PRECISE-DAPT score that has been previously
developed and validated in patients with coronary artery disease undergoing PCI.[12] This score has not been systemically validated in patients undergoing valvular interventions,
but recent studies show an impact on bleeding risk prediction in valvular cohorts.[13] Although risk scores for mortality after M-TEER have been developed and validated,[14] there is currently no dedicated bleeding risk score for this patient cohort.
Ethical approval was already established as part of a biomaterial approval for a national
research consortium (DFG, German Research Foundation)—Project number 374031971–TRR
240, 141/2018BO2. Written informed consent was previously obtained from all patients.
Multiplate and Thromboelastography
During the procedure, blood was taken directly from the left atrium before and after
placing the device. These blood samples were analyzed using impedance aggregometry
(Multiplate) and TEG6s. Both assays, Multiplate and TEG, were performed by a single
operator who was trained in these methods and blinded regarding the clinical history
and outcome of the patients. The results before and after successful procedure were
compared. The results remained blinded to the interventional cardiologist. Quality
checks of the Multiplate and TEG Analyzer were performed routinely.
The Multiplate Analyzer is used to determine platelet dysfunction and reactivity as
well as the effects of antiplatelet therapy. Aggregation of platelets is measured
in the blood by adding hirudin to prevent clotting. Platelet function was investigated
after stimulation with adenosine diphosphate (ADP) test, arachidonic acid (ASPI test)
or collagen (COL test), thrombin receptor activating peptide (TRAP test), and ristocetin
(RISTO-test). The adherence of activated platelets to the surface of the sensor conductor
leads to an increase in the electric resistance. The measured impedance correlates
with the strength of aggregation. Details of this method and its association with
outcome in other patient populations have been described previously.[15]
TEG is a standardized viscoelastic hemostatic assay. It includes measurement of clot
formation time, velocity of clot formation, clot firmness, and fibrinolysis. The blood
sample is rotated in a cup through 4° 45̀ six times a minute, imitating venous flow
to activate coagulation. Velocity and strength of clot formation are measured by a
computer. The result is displayed as a thromboelastogram.[16] We used the standardized TEG6s with Platelet Mapping.
Procedural Technique
The PASCAL system is a CE-marketed device for edge-to-edge treatment of MR. In brief,
the device is placed after transseptal puncture over a guide under transesophageal
echocardiographic control. In most cases, the procedure was performed under deep sedation.
All patients received central venous access for safe administration of drugs. The
vascular access site was the right femoral vein, if suitable.
Antithrombotic Regimen
Oral anticoagulation was paused the day before the procedure. Oral antiplatelet drugs
were continued throughout the hospitalization. Only two patients had no antithrombotic
premedication. Patients on aspirin monotherapy (n = 11) were either loaded with clopidogrel after the procedure (n = 4) or were administered clopidogrel without loading. During procedure unfractionated
heparin (UFH) was administered and coagulation was controlled by frequent measurements
of activated clotting time (ACT). To prevent thromboembolic events, the target ACT
was ≥300 seconds after transseptal puncture. After the procedure, all patients received
low-dose heparinization for 8 to 12 hours during femoral compression. Thereafter,
the antithrombotic regimen was heterogeneous and individual, dependent on comorbidities
such as the need for permanent oral anticoagulation because of AF or the need for
dual antiplatelet therapy (DAPT) due to recent PCI.
Statistical Methods
For statistical analysis, we used IBM SPSS Statistics (Version: 28.0.0.0 (190)). Continuous
data are presented as mean with standard deviation, dependent on the normality of
the distribution, as assessed by visual inspection of the histograms. Categorical
variables are presented as counts and proportions. We used t-test for differences between the means and Fisher's exact test for differences between
proportions. Differences in mean values with a two-tailed test result of p <0.05 were considered statistically significant. Receiver operating characteristic
(ROC) curves were built using predictions from regression analysis and area under
the curve (AUC) calculations were performed using the programming language Python
(Python Software Foundation). The Youden index was calculated to identify an optimal
cut point for the predictor of interest (i.e., delta ADP test). Cox regression and
logistic regression analyses were performed using IBM SPSS Statistics.
Results
We included 100 patients in our study, of whom 41 developed bleeding events within
6 months. Baseline characteristics of the patient cohort, stratified according to
patients with and without bleeding events, are shown in [Table 1]. Major bleeding (BARC ≥ 3) was observed in 14 patients (14%). The most frequent
bleeding was bleeding at puncture site (n = 23, 56.1%), and only three patients experienced a gastrointestinal bleeding. Preinterventional
oral anticoagulation or oral antiplatelet medication showed no differences at baseline
between patients with and without bleeding events. Importantly, pre- and periprocedural
activated coagulation time (ACT) test showed similar results between those patients
who had a bleeding event and those who did not. Patients who developed bleeding had
a significantly higher PRECISE-DAPT score and more frequently a history of bleeding
than those who did not.
Table 1
Baseline characteristics
|
Total cohort,
n = 100
|
No bleeding,
n = 59
|
Bleeding,
n = 41
|
p-Value
|
|
Age (y), mean
|
78.7 ± 7.2
|
78.6 ± 7.3
|
78.8 ± 7.1
|
0.908
|
|
Gender (f/m)
|
45/55
|
29/30
|
16/25
|
0.317
|
|
Chronic kidney disease, n
|
44 (44%)
|
25
|
19
|
0.694
|
|
Prehistory of bleeding, n
|
12
|
3
|
9
|
0.011
|
|
PRECISE-DAPT score, mean
|
28 ± 13.4
|
24.8 ± 10.8
|
32.6 ± 15.4
|
0.004
|
|
Preinterventional echocardiography
|
|
MR severe (≥III°), n
|
70
|
43
|
27
|
0.431
|
|
Type MR (primary/secondary/mixed), n
|
66/20/13
|
35/15/8
|
31/5/5
|
0.211
|
|
Cardiomyopathy (ischemic/dilated/other), n
|
17/10/12
|
13/7/7
|
4/3/5
|
0.293
|
|
EF (%), mean
|
47.1 ± 13.2
|
45.1 ± 14.1
|
50.1 ± 11.2
|
0.057
|
|
PA pressure mean (mmHg)
|
25.7 ± 9.5
|
25.9 ± 9.4
|
25.4 ± 9.7
|
0.816
|
|
Preinterventional antithrombotics
|
|
SAPT
|
12
|
9
|
3
|
0.230
|
|
DAPT
|
12
|
7
|
5
|
0.960
|
|
(N)OAC
|
51
|
31
|
20
|
0.711
|
|
(N)OAC + APT
|
23
|
11
|
12
|
0.214
|
|
Preinterventional laboratory variables
|
|
Hemoglobin (g/dL), mean
|
12 ± 2
|
12.2 ± 1.8
|
11.7 ± 2.2
|
0.294
|
|
WBC
|
7,455.2 ± 2,182.9
|
7,611.9 ± 2,411.2
|
7,229.8 ± 1,809.9
|
0.368
|
|
Platelet count
|
215.6 ± 73.6
|
218.8 ± 69.8
|
211 ± 79.4
|
0.613
|
|
Creatinine (mg/dL), mean
|
1.4 ± 1
|
1.3 ± 0.6
|
1.6 ± 1.4
|
0.230
|
|
GFR
|
54.2 ± 19.4
|
55.4 ± 18.4
|
52.5 ± 21
|
0.477
|
|
INR, mean
|
1.3 ± 0.3
|
1.3 ± 0.3
|
1.3 ± 0.3
|
0.968
|
|
aPTT (s), mean
|
28.8 ± 11.2
|
27.4 ± 8.7
|
30.7 ± 13.8
|
0.190
|
|
Procedural characteristics
|
|
Highest ACT (s), mean
|
339.3 ± 45
|
335.9 ± 40.6
|
344 ± 51
|
0.452
|
|
Number of implanted devices, n
|
1.3 ± 0.6
|
1.2 ± 0.5
|
1.4 ± 0.8
|
0.155
|
|
P
mean (mmHg), mean
|
2.9 ± 1.3
|
2.8 ± 1.5
|
3 ± 0.9
|
0.454
|
Abbreviations: DAPT, dual antiplatelet therapy; EF, ejection fraction; GFR, glomerular
filtration rate; MR, mitral regurgitation; (N)OAC, (new) oral anticoagulation; PA,
pulmonary artery; PRECISE DAPT, clinical bleeding risk score (PREdicting bleeding
Complications In patients undergoing Stent implantation and subsEquent Dual Anti Platelet
Therapy); SAPT, single antiplatelet therapy; WBC, white blood cells.
Note: p-Values were calculated using t-test for differences between the means and Fisher's exact test for differences between
proportions.
[Table 2] demonstrates the changes in thrombogenicity before and after placing the device.
p-Values were calculated with t-test for paired samples. [Figs. 1] to [3] also illustrate these changes. [Fig. 1] shows the differences pre- and post-M-TEER considering all Multiplate parameters.
For better illustration, [Figs. 2] and [3] only demonstrate statistically significant changes in TEG parameters.
Fig. 1 Changes in Multiplate parameters pre- and post-M-TEER. Boxplots were created using
SPSS, showing median, minimum, and maximum.
Fig. 2 Changes in TEG parameters pre- and post-M-TEER (1). Boxplots were created using SPSS,
showing median, minimum, and maximum. M-TEER, mitral transcatheter edge-to-edge repair;
TEG, thromboelastography.
Fig. 3 Changes in TEG parameters pre- and post-M-TEER (2). Boxplots were created using SPSS,
showing median, minimum, and maximum. M-TEER, mitral transcatheter edge-to-edge repair;
TEG, thromboelastography.
Table 2
Differences pre- and post-TEER in Multiplate and TEG parameters
|
Mean value pre-
|
Mean value post-
|
Mean value delta
|
p-Value
|
|
Multiplate
|
|
ADP test (units)
|
35.38 ± 20.07
|
37.59 ± 21.86
|
2.60 ± 11.48
|
0.042
|
|
ASPI test (units)
|
39.03 ± 21.12
|
44.70 ± 28.68
|
5.29 ± 12.99
|
<0.001
|
|
TRAP test (units)
|
72.94 ± 27.31
|
76.18 ± 26.05
|
3.70 ± 13.77
|
0.012
|
|
COL test (units)
|
59.12 ± 37.03
|
65.42 ± 42.98
|
6.44 ± 26.52
|
0.023
|
|
RISTO test (units)
|
81.99 ± 48.10
|
87.17 ± 52.14
|
5.66 ± 25.98
|
0.043
|
|
TEG
|
|
R (Min.)
|
7.33 ± 2.06
|
7.08 ± 1.88
|
−0.37 ± 1.12
|
0.002
|
|
K (Min.)
|
1.35 ± 0.40
|
1.38 ± 0.44
|
0.02 ± 0.35
|
0.504
|
|
Angle (degree)
|
72.27 ± 4.05
|
71.89 ± 4.5
|
−0.34 ± 3.19
|
0.310
|
|
MAHKH (mm)
|
67.15 ± 2.99
|
66.49 ± 3.33
|
−0.62 ± 1.97
|
0.004
|
|
MAADP (mm)
|
59.49 ± 8.83
|
57.08 ± 10.7
|
−1.93 ± 3.78
|
< 0.001
|
|
MAActF (mm)
|
16.17 ± 7.53
|
16.21 ± 5.54
|
−0.08 ± 4.59
|
0.873
|
|
LY30 (%)
|
0.091 ± 0.71
|
0.045 ± 0.21
|
−0.035 ± 0.77
|
0.672
|
|
ADP aggregation (%)
|
84.53 ± 17.36
|
81.27 ± 19.7
|
−2.39 ± 10.75
|
0.038
|
|
ADP inhibition (%)
|
14.78 ± 15.55
|
18.73 ± 19.7
|
3.15 ± 7.15
|
<0.001
|
Abbreviations: ADP, adenosine diphosphate; COL, collagen; TEER, transcatheter edge-to-edge
repair; TRAP, thrombin receptor activating peptide.
Note: p-Values were calculated using t-test for differences between the means.
[Table 3] shows the differences in Multiplate and TEG assays between bleeding and nonbleeding
groups. We selected exclusively those parameters for further analyses which showed
statistical significance in the t-test (p < 0.05). In the bleeding cohort, ADP aggregation was reduced before and after the
implantation. Furthermore, M-TEER led to a further decrease in ADP-induced platelet
activation. This test shows efficacy of ADP-receptor antagonists such as clopidogrel
and prasugrel. Consistent with this, TEG analysis demonstrated similar significant
decrease in ADP aggregation and increase in ADP inhibition.
Table 3
Differences pre- and post-TEER between patients with versus without bleeding events
|
No bleeding
|
Bleeding
|
p-Value
|
|
TEG pre-TEER
|
|
R (Min.)
|
7.1 ± 1.6
|
7.6 ± 2.6
|
0.230
|
|
K (Min.)
|
1.3 ± 0.3
|
1.4 ± 0.5
|
0.440
|
|
Angle (degree)
|
72.4 ± 3.2
|
72 ± 5.2
|
0.311
|
|
MAHKH (mm)
|
67.1 ± 3.3
|
67.2 ± 2.5
|
0.882
|
|
MAADP (mm)
|
60.8 ± 7.1
|
57.6 ± 10.6
|
0.053
|
|
MAActF (mm)
|
15.4 ± 5.4
|
17.3 ± 9.8
|
0.213
|
|
LY30 (%)
|
0.1 ± 0.9
|
0.05 ± 0.2
|
0.625
|
|
ADP aggregation (%)
|
87.8 ± 11.7
|
79.8 ± 22.5
|
0.041
|
|
ADP inhibition (%)
|
12.2 ± 11.7
|
18.8 ± 19.3
|
0.066
|
|
Multiplate pre-TEER
|
|
ADP
|
38.4 ± 17.6
|
31.1 ± 22.7
|
0.086
|
|
ASPI
|
38.8 ± 26.8
|
37.9 ± 27.9
|
0.736
|
|
TRAP
|
76.7 ± 24.2
|
67.5 ± 30.8
|
0.115
|
|
COL
|
61.4 ± 35.8
|
55.9 ± 38.9
|
0.464
|
|
RISTO
|
86 ± 42.8
|
76.3 ± 54.9
|
0.356
|
|
TEG post-TEER
|
|
R (Min.)
|
6.8 ± 1.7
|
7.4 ± 2.0
|
0.162
|
|
K (Min.)
|
1.4 ± 0.4
|
1.4 ± 0.4
|
0.806
|
|
Angle (degree)
|
72.1 ± 4.3
|
1.62 ± 4.8
|
0.633
|
|
MA HKH (mm)
|
66.3 ± 3.8
|
66.7 ± 2.5
|
0.579
|
|
MAADP (mm)
|
59 ± 7.9
|
54.4 ± 13.4
|
0.069
|
|
MAActF (mm)
|
15.5 ± 5.1
|
17.2 ± 6.1
|
0.132
|
|
LY30 (%)
|
0.1 ± 0.3
|
0.01 ± 0.04
|
0.099
|
|
ADP aggregation (%)
|
85.6 ± 13.1
|
75 ± 25.4
|
0.024
|
|
ADP inhibition (%)
|
14.4 ± 13.1
|
25 ± 25.4
|
0.024
|
|
Multiplate post-TEER
|
|
ADP
|
43.1 ± 20.1
|
29.9 ± 22.1
|
0.004
|
|
ASPI
|
49.3 ± 28.8
|
38.3 ± 27.6
|
0.073
|
|
TRAP
|
80.3 ± 23.4
|
70.4 ± 28.7
|
0.073
|
|
COL
|
73.3 ± 46.5
|
54.5 ± 35.3
|
0.039
|
|
RISTO
|
94.9 ± 53.5
|
76.3 ± 48.9
|
0.096
|
|
TEG delta
|
|
R (Min.)
|
−0.4 ± 1
|
−0.4 ± 1.2
|
1
|
|
K (Min.)
|
0.05 ± 0.4
|
−0.01 ± 0.3
|
0.418
|
|
Angle (degree)
|
−0.4 ± 3.5
|
−0.3 ± 2.8
|
0.977
|
|
MA HKH (mm)
|
−0.8 ± 1.9
|
−0.4 ± 2.1
|
0.430
|
|
MAADP (mm)
|
−1.5 ± 3.1
|
−2.6 ± 4.5
|
0.164
|
|
MAActF (mm)
|
0.1 ± 2
|
−0.3 ± 6.7
|
0.731
|
|
LY30 (%)
|
−0.1 ± 1
|
0.01 ± 0.04
|
0.667
|
|
ADP aggregation (%)
|
−1.6 ± 4.9
|
−3.5 ± 15.8
|
0.423
|
|
ADP inhibition /%)
|
1.6 ± 4.9
|
5.3 ± 9.2
|
0.014
|
|
Multiplate delta
|
|
ADP
|
5.5 ± 11.7
|
−1.5 ± 11.4
|
0.006
|
|
ASPI
|
7.9 ± 15.2
|
1.7 ± 8.1
|
0.014
|
|
TRAP
|
4.5 ± 14
|
2.6 ± 13.6
|
0.522
|
|
COL
|
10.1 ± 27.7
|
0.1 ± 23.6
|
0.053
|
|
RISTO
|
7.9 ± 28.1
|
2.6 ± 22.7
|
0.350
|
Abbreviations: ADP, adenosine diphosphate; COL, collagen; TEER, transcatheter edge-to-edge
repair; TRAP, thrombin receptor activating peptide.
Note: p-Values were calculated using t-test for differences between the means and Fisher's exact test for differences between
proportions.
We also found a significantly reduced platelet activation by collagen in the bleeding
group after M-TEER as shown in [Table 3]. In addition, differences in platelet function between pre- and post-PASCAL implantation
were observed as shown in [Table 3] and COL-dependent platelet aggregation was significantly reduced in the bleeding
group.
For further investigation of possible predictors of bleeding, we conducted ROC curve
analysis for those values which proved to be statistically significant in the univariate
analysis. Results are shown in [Figs. 4] to [6].
Fig. 4 ROC curve for predicting bleeding with pre- and post-implantation-TEER TEG parameters.
(A) ROC curve for ADP aggregation and (B) ROC curve for ADP Inhibition. ROC curves were generated with Python. ROC, receiver
operating characteristic; TEER, transcatheter edge-to-edge repair; TEG, thromboelastography.
Fig. 5 ROC curve for predicting bleeding with pre- and post-implantation-TEER Multiplate
parameters. (A) ROC curve for ADP test and (B) ROC curve for COL test. ROC curves were generated with Python. ADP, adenosine diphosphate;
COL, collagen; ROC, receiver operating characteristic; TEER, transcatheter edge-to-edge
repair.
Fig. 6 (A) ROC curve for predicting bleeding with delta ASPI test; (B) ROC curve for predicting bleeding with PRECISE-DAPT score. ROC curves were generated
with Python. ROC, receiver operating characteristic.
For TEG analysis, ADP aggregation and inhibition showed significant differences between
the bleeding and no bleeding groups at baseline pre-M-TEER. Nevertheless, with ROC
analysis, ADP aggregation and inhibition were not found to be accurate predictors
of bleeding, with AUC values between 0.57 and 0.63. Similarly, the COL test was not
a useful independent predictor of bleeding (AUC: 0.57–0.60). The best predictor of
bleeding was the delta-ADP test (AUC: 0.69).
[Fig. 6B] shows the prediction of bleeding using the established clinical score, the PRECISE-DAPT
score, with an AUC of 0.65, which is inferior to the delta-ADP test. For the delta-ADP
test, we performed the Kaplan–Meier analysis for bleeding events during the follow-up
of 180 days, using the Youden index of our delta-ADP ROC curve as cut-off, and showed
this to be highly significant (p = 0.004; [Fig. 7]). [Fig. 8] shows that a delta-ADP above the Youden index of 2.5 is associated with a reduced
rate of bleeding events. [Fig. 9] illustrates a classification of bleeding risk according to delta-ADP and PRECISE-DAPT
score, showing all patients with bleeding events and the percentage of patients with
bleeding in each risk category.
Fig. 7 Kaplan–Meier curve showing bleeding-free time considering Youden index for delta-ADP.
Delta ADP, postADP − preADP (Multiplate parameters post- and pre-M-TEER). Kaplan–Meier
curve was generated with SPSS. ADP, adenosine diphosphate; M-TEER, mitral transcatheter
edge-to-edge repair.
Fig. 8 Bleeding events categorized by delta-ADP values below or above Youden index 2.5.
Delta ADP, postADP − preADP (Multiplate parameters post- and pre-M-TEER). ADP, adenosine
diphosphate; M-TEER, mitral transcatheter edge-to-edge repair.
Fig. 9 Classification of bleeding risk by delta-ADP and PRECISE-DAPT score. Delta ADP, postADP − preADP
(Multiplate parameters post- and pre-M-TEER); PRECISE DAPT, clinical bleeding risk
score (PREdicting bleeding Complications In patients undergoing Stent implantation
and subsEquent Dual Anti Platelet Therapy). ADP, adenosine diphosphate; M-TEER, mitral
transcatheter edge-to-edge repair.
COX regression analysis confirmed the statistically significant influence of delta-ADP
(p = 0.002) on bleeding events over time, and its ability to predict bleeding compared
with the established PRECISE-DAPT score (p = 0.018), while gender and age alone were not predictive of bleeding ([Table 4]). To prove these results, logistic progression analysis was performed ([Table 5]). A delta-ADP value ≤2.5 is correlated with a 4.712 times chance of bleeding event
(p < 0.001).
Table 4
COX regression analysis with hazard ratios
|
Significance
|
Hazard ratio
|
Confidence interval
|
|
Age (per year increase)
|
0.744
|
1.008
|
0.961
|
1.057
|
|
Gender (if female)
|
0.231
|
0.678
|
0.359
|
1.28
|
|
PRECISE DAPT (per unit increase)
|
0.016
|
1.025
|
1.005
|
1.047
|
|
Delta ADP (per unit increase)
|
0.002
|
0.955
|
0.928
|
0.983
|
Abbreviations: Delta ADP, postADP − preADP (Multiplate parameters post- and pre-M-TEER);
PRECISE DAPT, clinical bleeding risk score (PREdicting bleeding Complications In patients
undergoing Stent implantation and subsEquent Dual Anti Platelet Therapy).
Note: COX regression analysis was performed using SPSS.
Table 5
Logistic regression analysis with odds ratios
|
Significance
|
Odds ratio
|
Confidence interval
|
|
Age (per year increase)
|
0.907
|
1.003
|
0.949
|
1.061
|
|
Gender (if female)
|
0.318
|
0.662
|
0.295
|
1.487
|
|
SAPT (if true)
|
0.239
|
0.439
|
0.111
|
1.731
|
|
DAPT (if true)
|
0.960
|
1.032
|
0.303
|
3.508
|
|
NOAK (if true)
|
0.711
|
0.860
|
0.387
|
1.910
|
|
PRECISE DAPT (per unit increase)
|
0.006
|
1.047
|
1.013
|
1.081
|
|
Delta ADP ≤ 2.5 (if true)
|
<0.001
|
4.712
|
1.990
|
11.159
|
Abbreviations: Delta ADP, postADP − preADP (Multiplate parameters post- and pre-M-TEER);
PRECISE DAPT, clinical bleeding risk score (PREdicting bleeding Complications In patients
undergoing Stent implantation and subsEquent Dual Anti Platelet Therapy).
Note: Logistic regression analysis was performed using SPSS.
Discussion
To the best of our knowledge, this is the first study to assess changes in thrombogenicity
in relation to TEER, as detected by TEG and aggregometry. Both assays—Multiplate and
TEG6s—are established methods to assess platelet function, efficacy of antithrombotic
therapy, and parameters of clot formation. If these parameters are deemed to be predictive
for bleeding risk, antithrombotic management could be personalized in a more restrictive
anticoagulation or antiplatelet regime in high-risk patients, for example, single
antiplatelet therapy instead of DAPT or a single loading dose of clopidogrel instead
of continuous dose.
In patients undergoing TAVI, several studies have evaluated the usefulness of TEG
to predict major cardiovascular and bleeding events.[9]
[10]
[17] One of these, which also used TEG platelet mapping, revealed a decrease in ADP-dependent
platelet aggregation and clot strength measured 3 days after TAVI implantation.[9] Another study showed an increase in platelet aggregation measured by Multiplate.[18] This study revealed similar results: an increase in platelet aggregation in Multiplate
test, and a decrease in platelet function in TEG analysis. The discrepancies between
both tests indicate a complex dysfunction of platelets. One explanation might be that
the tests focus on different aspects of platelet function. Multiplate measures the
platelet aggregation capacity under the influence of different agonists, while TEG
measures the mechanic stability of clot strength and the interactions between platelets,
fibrin, and other coagulation factors. Therefore, the discrepancy might indicate that
platelets are enhanced in aggregation when stimulated with different agonists but
might be impaired in their capacity to contribute to mechanical clot strength when
interacting with fibrin and other coagulation factors in overall hemostasis. This
could be due to qualitative platelet dysfunction, impaired fibrin interaction, or
other systemic factors. Nevertheless, clinical implications include potential impact
on bleeding risk.
Although M-TEER procedures represent a great advance in the treatment of valvular
heart disease,[19] there are no standardized guidelines for the antithrombotic regimen and the assessment
of bleeding risk. There is a great heterogeneity in anticoagulation and antiplatelet
therapy pre- and postintervention among patients undergoing TEER, whereas the peri-interventional
regimen is usually standardized.[20]
[21] Stroke and thrombus formation after TEER are known complications which should impact
the anticoagulatory regimen.[21]
[22]
[23]
[24] A high proportion (80%) of patients undergoing M-TEER have AF due to concomitant
atriopathy, and the optimal antithrombotic regimen and risk for thrombotic and bleeding
risk after TEER is unknown. To date, there are no standardized recommendations regarding
the addition of antiplatelet therapy after TEER in these patients. The risk of bleeding
increases with comorbidities such as kidney failure, age, and frailty.[4]
[25]
[26] In our cohort of patients, we found that periprocedural testing may provide prognostic
value for the prediction of bleeding. The most useful test in our study was the delta-ADP
test. Other parameters, which are usually considered to increase bleeding risk, such
as chronic kidney disease or diabetes mellitus, were not significantly associated
with bleeding events in our cohort. Furthermore, we demonstrate that the PRECISE-DAPT
score was less accurate in the prediction of bleeding events than the delta-ADP test
as demonstrated in [Fig. 5B]. Accordingly, the delta-ADP test might improve bleeding prediction. However, in
this study AUC was moderately high, which does not permit for individual prognosis
estimation.
This study has a few limitations. We are aware that the present findings are hypothesis-generating
and warrant validation in larger cohorts. The sample size is small and there was a
relatively low number of major bleeding events, limiting the statistical power of
the analysis for more severe bleeding events. However, minor (BARC-2) bleeding events
are also clinically relevant, as shown previously in other cardiovascular disease
conditions.[5]
[27]
[28] Nevertheless, the rate of major bleeding is relatively high in comparison to previous
registries investigating bleeding events after M-TEER. A possible explanation is that
most of these studies used the Mitral Valve Academic Research Consortium (MVARC) definition,
which is more restrictive to the classifications of major bleeding events.[5] Thus, transfusion of ≥3 units of whole blood or packed red blood cells qualifies
for major bleeding in the MVARC classification, whereas in the BARC classification,
any transfusion with overt bleeding is sufficient for major bleeding definition.[11]
[29] We could not analyze the relationship between the hematological tests and thrombotic
events, as these were too few events during follow-up in the present cohort. Thus,
we believe that the bleeding risk is potentially more clinically relevant in the early
phase (i.e., 6 months) after M-TEER and identification of bleeding risk predictors
is more important. Whether the trade-off between bleeding and thrombotic risk changes
over time is uncertain. The etiology of MR was heterogenous, representing an all-comer
scenario. We did not systematically analyze whether there are differences in the results
with regard to the etiology of MR and the stage of heart failure. The results of the
tests are prone to processing errors and the individual results have not been reproduced
by intra-individual sequential testing. The timing and sampling of the blood was standardized
according to the procedural protocol, minimizing the potential bias of the effect
of periprocedural anticoagulation. Nevertheless, the periprocedural use of UFH might
suggest dynamic chances in coagulation activity. Thus, the effect of UFH does not
explain the differences in thrombogenicity before and after placing the device since
ACT measurements were comparable among all patients. Furthermore, we did not observe
a significant difference in ACT levels between patients with and without bleeding
events. Further studies should test the hypothesis that use of platelet function and
TEG markers to guide antithrombotic therapy may modify postprocedural bleeding events
in patients undergoing M-TEER. Another limitation are the differences in pre- and
post-TEER platelet measurements between the Multiplate and TEG systems as discussed
before. Although these differences could be explained, large-scale clinical evidence
based on this topic still needs to be updated.
Conclusion
This study using TEG6s platelet mapping and Multiplate showed that thrombogenicity
is immediately affected after the PASCAL implantation, possibly caused by shear stress
on platelets and altered flow conditions. These effects were significantly correlated
with bleeding events. Whether these results could guide periprocedural and long-term
antithrombotic therapy, and thus improve clinical outcome, requires further investigation.
What is known about this topic?
-
Transcatheter mitral valve repair is an established treatment: previous studies have
shown the safety and efficacy of this procedure.
-
Bleeding complications are common adverse events: bleeding is associated with prolonged
hospital stay and also increased mortality but until now there is no well-established
standardized anticoagulation regimen for these patients
What does this paper add?
-
Thrombogenicity in left atrium is immediately affected after the PASCAL implantation.
-
Intraprocedural assessment of TEG and Multiplate parameters may help to predict bleeding
complications and personalize antithrombotic therapy.
