Leaflet thrombosis refers to the formation of both calcified and non-calcified thrombi on transcatheter heart valve (THV) leaflets, most commonly observed in transcatheter aortic valves (TAVs) made of bioprosthetic material (1). In 2015, Makkar et al. first recognized subclinical leaflet thrombosis (SLT) in transcatheter aortic valve intervention (TAVR). In their landmark study, they utilized multidetector computed tomography (CT) to assess leaflet abnormalities after balloon-expanding TAVR and demonstrated the occurrence of reduced leaflet motion associated with hypoattenuation opacities on CT in a subset of asymptomatic patients (2). This pioneering work prompted further investigation into its mechanisms, implications, and preventive and management strategies.

Pathophysiological mechanisms underlying leaflet thrombosis are still being elucidated. A modified Virchow's triad has been proposed, encompassing (i) hypercoagulability at the bioprosthetic surface, (ii) leaflet surface damage and/or endothelial injury during device deployment, and (iii) stasis and turbulent flow (3, 4). The formation of neosinuses between THV leaflets and the valve cage and between native leaflets and aortic wall creates semi-permeable barriers that lead to flow stasis and reduced washout forces (Figure 1) (5). Factors including larger-sized or overexpanded balloon-expandable valves, low-deployment depth of self-expanding valves (particularly CoreValve Evolut, Medtronic), intra-annular valve position, small annular perimeter, valve-in-valve (ViV) TAVR for stented surgical valves, non-commissural alignment, low coronary washout, small sinus size, and reduced flow can increase thrombosis risk (6–8). Valve maldeployment, which may occur due to factors such as extrinsic calcification or valve-in-valve deployment, can lead to incomplete cusp coaptation and excursion and increase thrombosis risk (9).

Two distinct entities have been identified, namely, hypoattenuated leaflet thickening (HALT) and restricted leaflet motion (RELM). HALT refers to the abnormal thrombotic thickening of valve leaflets. RELM occurs secondary to HALT and impairs valve functionality. The co-occurrence of HALT and RELM defines hypoattenuation affecting motion (HAM) (Figure 2) (10, 11).

Clinical leaflet thrombosis after TAVR is rare, affecting approximately 0.5% of TAVR patients, and can manifest as increased valve gradient and double the risk of cerebrovascular events and embolization to other arterial beds (12–14).

SLT occurs more commonly in up to 15% of TAVR patients and can vary dynamically over time based on the valve type and position. Inherent differences in THV compared with surgical valves, including mounting the valve within a rigid vs. more flexible stent, valve crimping with delivery, and native calcification, may alter thrombotic risk (15). In the PARTNER-3 sub-study involving 435 low-risk patients, HALT was more than twice as common after TAVR as compared with surgical aortic valve replacement (SAVR) at 30 days (13% vs. 5%, respectively), but the difference was no longer seen after 1 year. In the Evolut Low-Risk Leaflet Thickening or Immobility (LTI) sub-study evaluating the frequency of HALT in 503 patients undergoing TAVR using the self-expanding Evolut THV or SAVR, there was no difference in HALT between TAVR and SAVR at 30 days and 1 year (16,17). Notably, a recent study examining ViV TAVR demonstrated no difference in HALT or RELM between ViV and native-valve TAVR, despite historical concerns about the impact of ViV on sinus thrombosis (18).

Four-dimensional CT (4DCT) has emerged as a standard to identify HALT and RELM and is comparable to transesophageal echocardiology. Seen as hypoattenuating leaflet lesions on CT, HALT is assessed in diastole and requires visualization of leaflet coaptation. RELM is assessed in systole with maximal leaflet opening. Each leaflet is classified as mildly (<50%), moderately (50%–69%), or severely (70%–99%) restricted or immobile. HALT and >50% restriction signifies HAM (8).

The clinical significance of SLT is not well understood. Although the evidence is limited and conflicting, HALT may be associated with increased transient ischemic attack or stroke risk (RR: 2.6), with several studies showing no increase in cerebral ischemic events (19–21). If persistent or progressive, HALT minimally increases the valve gradient at 1 year compared with patients with no HALT or with spontaneously resolved HALT (17.8 vs. 12.7 mmHg, p = 0.04). An increase in the valve gradient appears concordant with the degree of HALT (17). Most studies have failed to demonstrate an association between HALT or RELM and heart failure symptomatology (22).

Early prevention strategies centered on antiplatelets, based on expert consensus and extrapolation from coronary stenting and historical bioprosthesis studies. Single antiplatelet therapy (SAPT) is preferred over dual antiplatelet therapy (DAPT) for long-term prevention (23, 24).

Oral anticoagulation (OAC)–based prevention strategies have produced mixed results. The GALILEO trial, which randomized 1,644 TAVR patients to either rivaroxaban (plus aspirin for 3 months) or antiplatelet (aspirin plus clopidogrel for 3 months) treatment, demonstrated a substantial reduction in HALT and RELM with the use of rivaroxaban. However, the trial was terminated early due to higher deaths and bleeding in the rivaroxaban treatment arm (25). The ATLANTIS study, which randomly assigned 1,500 TAVR patients to apixaban or standard care [vitamin K antagonist (VKA) or antiplatelet], demonstrated that apixaban vs. antiplatelet resulted in lower obstructive valve thrombosis but higher non-cardiovascular mortality with apixaban, while apixaban vs. VKA produced no difference in the primary outcome or safety (26). A small study randomizing TAVR patients to VKA plus aspirin or aspirin alone showed a reduction in the primary endpoint (7.0% vs. 26.5%), with VKA having no excess bleeding (27). Using routine OACs post-TAVR remains under debate, and future studies including comparisons of type and strength should be pursued.

The treatment selection for subclinical or clinical thrombosis requires careful assessment of thrombus burden and individual patient factors (Figure 2). Once confirmed on imaging, the preferred treatment involves OAC (apixaban, rivaroxaban, or warfarin) for 3–6 months until resolution. A reasonable approach is to monitor SLT in patients with a high bleeding risk because of its dynamic nature (17). Severe cases may require balloon valvuloplasty or transcatheter valve-in-valve implantation (17, 28).

Leaflet thrombosis is a dynamic phenomenon that relies on local and systemic factors to develop. The best prevention strategies are still under debate, arising from concerns about OAC bleeding risk with or without antiplatelet therapy. Further investigation is needed to improve the prevention strategies and better understand the clinical implications and progression to valve degeneration.