Transcatheter aortic valve implantation (TAVI) is the standard of care for patients
with severe aortic stenosis who are inoperable, and a valid alternative to surgery
for patients at high or intermediate surgical risk.[1] Current candidates to TAVI are typically octogenarians, which makes their propensity
to thrombus formation or bleeding higher at baseline compared with younger individuals.[2] It is of no surprise that TAVI itself further contributes to increase the risk of
ischaemic and haemorrhagic complications in this population.[3]
Despite numerous technical innovations over the years, cerebrovascular events are
still detected in 4 to 6% of intermediate-risk patients at 30 days, typically as the
consequence of procedural factors.[4] Subsequently, the risk of cerebrovascular events peaks at 8 to 10% at 1 year and
10 to 12% at 2 years, with late events mostly explained by concurrent patient-related
(e.g. atrial fibrillation) and valve-related (e.g. stent surface exposure, leaflet
thrombosis) conditions.[4] Leaflet thrombosis is an emerging concern of uncertain clinical significance, more
frequently observed with transcatheter valves rather than surgical bioprostheses,
which is more likely to resolve with anti-coagulants rather than anti-platelets.[5]
[6]
Bleeding in TAVI patients is approximately 31% at 5 years, with similar proportions
of access-site and non-access site-related events, and a well-known detrimental impact
on prognosis.[7]
[8] Of non-access-site bleeding, approximately 40% of the episodes (mostly neurological
and gastrointestinal) accrue beyond 30 days.[4]
[8] The incidence and timing of bleeding and ischaemic complications after TAVI call
into question the net benefit of using adjuvant anti-thrombotic therapies in this
setting. Current guidelines from the European Society of Cardiology recommend clopidogrel
in addition to aspirin for 3 to 6 months after TAVI in patients who are not candidates
to oral anti-coagulation, followed by single anti-platelet therapy lifelong (class
IIa).[9] Patients at high risk of bleeding may be considered eligible upfront for anti-platelet
monotherapy (class IIb). In the United States, joint guidelines from the American
College of Cardiology and American Heart Association recommend dual-anti-platelet
therapy (DAPT) for 6 months (class IIb).[10] Therefore, anti-platelet therapy with aspirin and/or clopidogrel is the current
anti-thrombotic standard for non-anti-coagulated patients undergoing TAVI.
Investigations from the field of percutaneous coronary intervention (PCI) have consistently
demonstrated some degree of inter-individual variability in the platelet response
to clopidogrel,[11] and high on-treatment platelet reactivity may have prognostic implications.[12]
[13]
[14]
[15] Several platelet function assays (e.g., light transmission aggregometry, VerifyNow,
multiple electrode aggregometry) allow distinguishing patients with on-clopidogrel
high platelet reactivity (HPR) or low platelet reactivity (LPL) based on standardized
cut-off values.[16]
[17] PCI studies suggest that HPR and LPR denote a status of impaired or undue response
to clopidogrel, and carry a higher risk of thrombosis and bleeding, respectively.[11]
Nonetheless, the impact of tailoring anti-platelet therapy based on platelet function
profiles is controversial. In elderly patients from the ANTARCTIC study—a population
that resembles but does not necessarily match the population of patients undergoing
TAVI—platelet function monitoring with treatment adjustment did not improve the clinical
outcomes of PCI.[18] In the TAVI field, the association of platelet reactivity and clinical outcomes
has been investigated in small studies correlating on-treatment LPR and bleeding[19]
[20]
[21]
[22]
[23] ([Table 1]).
Table 1
Relation between platelet function measurement and bleedings in TAVI population
|
Trial
|
Polzin et al
|
Orvin et al
|
Watanabe et al
|
Czerwińska-Jelonkiewicz et al
|
Kibler et al
|
Gross et al
|
|
Year of publication
|
2015
|
2016
|
2016
|
2018
|
2018
|
2018
|
|
Journal
|
European Journal of Pharmacology
|
Journal of Thrombosis and Thrombolysis
|
International Heart Journal
|
Archives of Cardiovascular Diseases
|
Thrombosis and Haemostasis
|
Thrombosis and Haemostasis
|
|
Patients, N°
|
140
|
39
|
32
|
100
|
219
|
146
|
|
Age, median ± SD
|
83 ± 6
|
81.7 ± 6.5
|
84.2 ± 5.0
|
78.3 ± 9.9[a] and 79.5 ± 7.1[a]
|
82.8 ± 6.5[b] and 83.4 ± 6.9[b]
|
81 ± 8.4
|
|
Coronary artery disease, n (%)
|
98 (70)
|
NA
|
13 (40.6)
|
96 (96)
|
NA
|
133 (91.1)
|
|
Anti-coagulant intake, n (%)
|
25 (18)
|
12 (30.8)
|
6 (18.8)
|
34 (34)
|
85 (38.8)
|
NA
|
|
Platelet function assessment method
|
VASP, LTA
|
VerifyNow, Multiplate analyser
|
LTA, VASP, VerifyNow
|
VerifyNow
|
PFA-100
|
Multiplate analyser
|
|
Platelet function assessment timing
|
During hospital stay
|
1–3 days before TAVI, at 3–5 days and at 30 days
|
At 7 and 30 days
|
24 hours before TAVI and at 6 days
|
Post-procedural
|
Along with the TAVI procedure
|
|
Primary outcome
|
Ischaemic and bleeding complications
|
Platelet reactivity
|
Platelet function parameters
|
In-hospital bleeding and vascular complications
|
PVAR
|
Association between platelet reactivity and bleedings
|
|
Secondary outcomes
|
None
|
None
|
In-hospital thrombotic or bleedings events
|
NA
|
Bleeding or its composite with PVAR
|
None
|
|
Anti-thrombotic therapy
|
Aspirin + Clopidogrel
|
Clopidogrel + Aspirin or OAC
|
Aspirin ± Clopidogrel
|
Aspirin + Clopidogrel or OAC + Aspirin or Clopidogrel
|
Aspirin + Clopidogrel
|
Aspirin + P2Y12 inhibitor (mostly Clopidogrel)
|
|
Bleeding definition
|
VARC-2
|
NA
|
VARC-2
|
VARC-2
|
VARC-2
|
VARC-2
|
|
Conclusion
|
No correlations between platelet reactivity and ischaemic or bleeding events
|
TAVI patients treated with DAPT have high rates of residual platelet reactivity
|
A hyper-response to clopidogrel is associated with bleeding
|
Platelet reactivity to clopidogrel seems to predict TAVI-related bleeding
|
CT-ADP (vWf-dependent platelet function) predict PVAR and bleedings
|
LPR is associated with bleeding in TAVI patients
|
Abbreviations: CT-ADP, closure time adenosine diphosphate; DAPT, dual-anti-platelet
therapy; LPR, low platelet reactivity; LTA, light transmittance aggregometry; NA,
not available; OAC, oral anti-coagulant; PFA, platelet function analyser; PVAR, paravalvular
aortic regurgitation; TAVI, transcatheter aortic valve implantation; VARC-2, valve
academic research consortium-2 consensus document; VASP, vasodilator stimulated phosphoprotein
phosphorylation; vWf, von Willebrand factor.
a In the bleedings and no bleedings group, respectively. Comprehensive data are not
reported.
b In the PVAR and no PVAR group, respectively. Comprehensive data are not reported.
In this issue of Thrombosis and Haemostasis, Gross et al add to this evidence with another instructive investigation.[24] The authors studied platelet reactivity by using the Multiplate analyzer[25] in 136 consecutive TAVI patients on DAPT with aspirin plus a P2Y12 inhibitor (mostly clopidogrel). Bleedings were assessed according to the Valve Academic
Research Consortium-2 (VARC-2) definition.[26] At 30 days post-TAVI, there was a significant association between LPR and VARC-2
bleeding both in unadjusted (hazard ratio [HR], 2.10, 95% confidence interval [CI],
1.17–3.79; p = 0.01) and age-adjusted (HRadj, 2.06, 95% CI, 1.14–3.71; p = 0.02) analyses. In contrast, HPR was not significantly associated with the risk
of ischaemic events. Although these results were obtained in a relatively small sample
of patients, the study shows some admirable methodological aspects. Bleeding assessment
was based on the broadly accepted VARC-2 definition,[26] which is context-specific, and HPR and LPR were defined according to the cut-off
values that were previously shown to link with adverse events in the PCI setting.[27] In addition, the study population reflects contemporary TAVI practice in intermediate-
to high-risk patients treated with either balloon-expandable or self-expandable bioprostheses,
and the follow-up rate was complete.
Unfortunately, some caveats of this study should also be underlined. First, the conclusion
that HPR was not associated with thrombotic events is undermined by the small number
of ischaemic events at 30 days and resulting low power. Bleeding events were more
frequent, which makes the inference about the association between LPR and haemorrhagic
complications more statistically robust. Of interest, as eloquently shown by the Kaplan–Meier
analysis, the higher propensity of LPR patients to bleed was mostly confined to the
first 5 days from the procedure. Yet, the comparison between LPR and non-LPR patients
was adjusted only by age and suffers from residual confounding, possibly including
identifiable risk factors, because the result of univariate analyses for many variables
of interest (e.g. demographic, clinical, procedural) was not reported. It would have
been also valuable to know how many of the bleeding events occurring in LPR patients
were life-threatening, major, or minor. The cut-off value of LPR identified for bleeding
suffered from relatively low positive (53.6%) and negative (75.6%) predictive values,
which is consistent with the multifactorial nature of bleeding. Other limitations
include the availability of a single platelet function measurement, the lack of confirmation
from other point-of-care assays and the inclusion of patients with different backgrounds
of loading and maintenance P2Y12 inhibitor doses.
The net benefit of DAPT after TAVI has been recently questioned by a patient-level
meta-analysis of three small randomized trials comparing DAPT with aspirin.[28] Over a total of 421 patients, life-threatening VARC-2 bleeding at 30 days were significantly
increased with DAPT (6.8% vs. 2.4%, p = 0.036), corresponding to a 2.68-fold relative increase compared with aspirin. No
differences in stroke and all-cause death were reported, but numbers were small. Further
evidence on the merits of DAPT in the TAVI setting is expected from the on-going POPular-TAVI
and CLOE trials.[29] If DAPT will not be ultimately shown to be more beneficial than aspirin alone, speculating
on HPR and LPR will become a sterile exercise in the TAVI scenario.
Another field of intense investigation is the role of anti-coagulation in TAVI patients
with no baseline indication for vitamin K antagonists or non-vitamin K antagonist
oral anticoagulants. The joint American College of Cardiology and American Heart Association
guideline for valvular heart disease affirms that a short term of anti-coagulation
may be considered in these patients (class IIb), although it remains unclear if this
regimen should be stacked on top of anti-platelet therapy.[10] As expected, dual-pathway inhibition increases the risk of bleeding in the elderly
population currently referred to TAVI.[30] Also notably, the seminal GALILEO study, comparing a rivaroxaban-based with an antiplatelet-based
strategy in TAVI patients, has been recently stopped due to an increase in safety
events in the rivaroxaban arm (NCT02556203). Further insight on the comparison between
anti-coagulants and anti-platelets in TAVI will come from the ATLANTIS trial of apixaban.[31]
Concomitant coronary artery disease is frequently discovered during the clinical work-up
prior to TAVI and requires treatment in 20 to 40% of patients.[32] Mostly based on empiric considerations, guidance on the use of anti-thrombotic therapy
after TAVI for these patients is given by two recent joint European consensus documents.[33]
[34] TAVI patients who are already taking oral anti-coagulants and have recent PCI or
an acute coronary syndrome should be treated similarly to patients receiving a stent
without TAVI. Conversely, waiting for the results of on-going studies, patients undergoing
TAVI without concomitant need for oral anti-coagulation should receive an anti-platelet
regimen consisting of lifelong aspirin monotherapy or aspirin and clopidogrel for
3 to 6 months followed by aspirin monotherapy, depending on the bleeding risk, and
concomitant treated or untreated coronary artery disease.
In conclusion, accruing evidences suggest that LPR may be one of the predisposing
factors for bleeding in the TAVI setting, particularly early after the procedure.
While this argument is not a call for guiding anti-platelet therapy by means of platelet
function testing, it voices about the risk of anti-thrombotic over-treatment if one-size-fits-all
strategies are implemented. Because the boundary between tailored therapy and under-treatment
is equally thin, further evidence is needed to find the sweet spot of optimal pharmacotherapy
in the TAVI setting ([Fig. 1]). These investigations must take into account the ever-changing nature of TAVI in
the context of improved safety and extension towards individuals at lower risk.
Fig. 1 Spectrum of potential platelet reactivity profiles in patients undergoing transcatheter
aortic valve implantation (TAVI), ranging from high platelet reactivity (HPR) to low
platelet reactivity (LPR). An area of “sweet spot” for the optimal use of anti-thrombotic
medications is envisaged. Risk factors for HPR and LPR are proposed based on lessons
from the field of percutaneous coronary intervention.[11]
