Author + information
- Raj Makkar, MD∗ ( and )
- Tarun Chakravarty, MD
- ↵∗Address for correspondence:
Dr. Raj Makkar, Smidt Heart Institute, Cedars-Sinai Medical Center, 8700 Beverly Boulevard, Los Angeles, California 90048.
Subclinical leaflet thrombosis occurs in up to 10% to 15% of patients undergoing bioprosthetic aortic valve replacement (1–3). This finding has been described in both transcatheter (1–3) and surgical (1,2) bioprosthetic aortic valves as well as in multiple valve types (1–3). Subclinical leaflet thrombosis is associated with a small but significant increase in aortic valve gradients. A greater proportion of patients with subclinical leaflet thrombosis develop clinically significant valve thrombosis, associated with a rise in aortic valve gradients (2). The published studies, individually (1–3) as well as in a meta-analysis (4), have shown no impact on mortality or myocardial infarction or stroke but a trend toward an increase in neurologic events, especially transient ischemic attacks (TIAs) (1,2,4). With increasing adoption of transcatheter aortic valve replacement (TAVR), there is tremendous interest in understanding subclinical leaflet thrombosis to further optimize TAVR outcomes.
In this issue of JACC: Cardiovascular Interventions, Ruile et al. (5) report the intermediate-term clinical outcomes associated with early leaflet thrombosis. The study included 754 patients undergoing TAVR at a single medical center. Although multiple transcatheter heart valves were studied, the predominant valve types included balloon-expandable (80.1%) and self-expanding (14.5%) valves. All patients underwent computed tomographic angiography (CTA) at a median of 5 days after TAVR. The prevalence of atrial fibrillation was lower in patients with leaflet thrombosis. The investigators report clinical endpoints during a median follow-up period of 406 days. The presence of leaflet thrombosis did not significantly affect mortality (freedom from mortality 86.6% vs. 85.4%; p = 0.912) or strokes or TIAs (stroke- and TIA-free survival 98.5% vs. 96.8%; p = 0.331). Early leaflet thrombosis was not a predictor of mortality or strokes or TIAs in univariate or multivariate analysis.
The strength of this study that it is a large dataset from a group of investigators who are among the forerunners in studying this finding. A previously reported registry, the largest dataset (in 931 patients) on subclinical leaflet thrombosis, reported increased rates of TIAs in patients with subclinical leaflet thrombosis (2). The investigators, in the present study, did not notice an association between stroke or TIA and leaflet thrombosis. This finding should be viewed in the context of the limitations of this study and methodological differences from the previously published studies. First and foremost, the definition of leaflet thrombosis used in this study is not the same as the definition used for subclinical leaflet thrombosis in previously reported studies (1,2). The investigators define leaflet thrombosis as hypoattenuated leaflet thickening, with or without rigidity or reduced leaflet motion of 1 or more leaflet segments. Prior studies have defined subclinical leaflet thrombosis as hypoattenuation affecting leaflet motion with at least 50% reduction in leaflet motion, in addition to the presence of hypoattenuated leaflet thickening on the aortic valve leaflets (Figure 1). It seems reasonable to assume that the studies on subclinical leaflet thrombosis that reported an association between this finding and TIAs included a more advanced stage of leaflet thrombosis.
Second, the investigators do not differentiate between periprocedural neurological events and events occurring during follow-up after performing CTA. Periprocedural strokes and TIAs are related primarily to the procedure and can be multifactorial, including advanced age, history of stroke, atrial fibrillation, aortic wall and aortic valve calcification, aortic wall atheroma, left ventricular dysfunction, and transient cerebral hypoperfusion during rapid pacing. Strokes and TIAs occurring during follow-up after performing CTA could, at least in theory, be related to leaflet thrombosis.
Third, the data on neurological events were not independently adjudicated by a stroke neurologist. Although strokes are less likely to be over- or underdiagnosed, the evaluation and diagnosis of TIA is challenging and subject to interobserver variability. The incidence of strokes and TIAs after TAVR depends on the rigor of follow-up and whether mandated assessment by a neurologist was included in the protocol. For instance, the 30-day stroke rate after TAVR was 9.1% in the SENTINEL trial (which included mandated neurological follow-up) (6), 5.5% in the PARTNER 2 trial (which lacked mandated neurological follow-up) (7), and 2.1% in the TVT registry (with no protocol-driven clinical or neurological follow up) (8). A 96.8% rate of stroke- and TIA-free survival observed in this study at 18 months in a patient cohort with a mean age of 82 years is concerning for significant underestimation of neurological events.
Fourth, only 53% of patients undergoing TAVR underwent post-TAVR CTA; no information is provided on the remaining patients who did not undergo CTA.
Fifth, the investigators performed CTA early after TAVR, the median time from TAVR to CTA being 5 days; in contrast, other studies of subclinical leaflet thrombosis have performed CTA 1 to 3 months after TAVR. Fifty-one of 120 patients (42.5%) with leaflet thrombosis were discharged on anticoagulation (either for clinical indications or for the treatment of leaflet thrombosis). Early detection of leaflet thrombosis (unlike other studies) and its treatment in almost one-half of the patients may have affected the incidence of neurological events during follow-up. Anticoagulation alters the natural history of this finding, and in this series, it occurred very early after TAVR.
Unlike in other studies, the investigators did not notice a correlation between anticoagulation at discharge and the prevalence of leaflet thrombosis. This does not imply that anticoagulation is not protective for leaflet thrombosis. This is due to CTA being performed 5 days after TAVR, before therapeutic anticoagulation has taken effect. Previous studies have reported a significant protective effect of anticoagulation, with both warfarin and direct oral anticoagulant agents, when CTA is performed 1 to 3 months after TAVR (1–3). This study does confirm the impact of leaflet thrombosis on transcatheter valve hemodynamic status, consistent with the previously reported experience.
Data on the impact of leaflet thrombosis on valve durability, or the role of routine CTA or routine anticoagulation after TAVR, are lacking. The imaging substudies of low-risk TAVR trials, mandated by the U.S. Food and Drug Administration, will shed light on the natural history and clinical impact of subclinical leaflet thrombosis. The GALILEO (Global Study Comparing a rivAroxaban-based Antithrombotic Strategy to an antipLatelet-based Strategy After Transcatheter aortIc vaLve rEplacement to Optimize Clinical Outcomes) and ATLANTIS (Anti-Thrombotic Strategy After Trans-Aortic Valve Implantation for Aortic Stenosis) randomized trials will provide evidence on the safety and efficacy of routine anticoagulation after TAVR. This study does not provide supportive evidence for or against the initiation of anticoagulation for the treatment of leaflet thrombosis. Lack of an association with clinical outcomes at 1 year does not rule out the potential adverse outcomes associated with this finding at longer term follow-up, especially as this finding is more likely to persist or progress than spontaneously resolve in the absence of anticoagulation (1,2). A small proportion of patients with leaflet thrombosis do progress to clinically significant valve thrombosis. Given the lack of an association between clinical outcomes and leaflet thrombosis in this and other studies and the clinical benefits of TAVR, it is reasonable not to be alarmed by this finding. Nonetheless, it is equally important to not neglect or prematurely consider this a “benign” finding.
↵∗ Editorials published in JACC: Cardiovascular Interventions reflect the views of the authors and do not necessarily represent the views of JACC: Cardiovascular Interventions or the American College of Cardiology.
Dr. Chakravarty is a consultant and proctor for Edwards Lifesciences and Medtronic Inc. Dr. Makkar has reported that he has no relationships relevant to the contents of this paper to disclose.
- 2018 American College of Cardiology Foundation
- Chakravarty T.,
- Sondergaard L.,
- Friedman J.,
- et al.
- Hansson N.C.,
- Grove E.L.,
- Andersen H.R.,
- et al.
- Rashid H.N.,
- Gooley R.P.,
- Nerlekar N.,
- et al.
- Ruile P.,
- Minners J.,
- Breitbart P.,
- et al.
- Kapadia S.R.,
- Kodali S.,
- Makkar R.,
- et al.
- Grover F.L.,
- Vemulapalli S.,
- Carroll J.D.,
- et al.