Post-TAVR Thrombus May Be More Common Than Thought, Often Subclinical

Valve thrombosis following TAVR may be more common than previously thought, according to 2 studies published online in the April 2015 issue of Circulation: Cardiovascular Interventions. Many patients are asymptomatic at diagnosis, but regardless of symptom status, prompt treatment with anticoagulants appears to be effective in the majority of cases. 

5.5.15 one study, Nicolaj C. Hansson, MD, of Aarhus University Hospital Skejby (Aarhus, Denmark), and colleagues used multidetector CT to identify valve thrombosis in 140 patients (mean age 81.4 years; 57% women) 1 to 3 months after TAVR with the Sapien XT prosthesis (Edwards Lifesciences).

Overall, 5 patients (4%) were identified as having valve thrombosis, which appeared as low-attenuation masses attached to the aortic side of the valve cusps. Four of these patients were asymptomatic at the time of CT and had a normal TEE exam without signs of thrombus formation or flow obstruction. Gradients remained unchanged or slightly elevated. 

Three of the 5 thrombosis patients had LVEF of < 35%, 1 had A-fib, and 3 had moderate/severe mitral regurgitation. Although 1 patient had a history of idiopathic thrombocytopenic purpura, none had histories of venous thrombosis or coagulation disorders. No embolic events occurred, and none of the patients had clinical or biochemical evidence of endocarditis. 

Anticoagulants Used to Treat Thrombosis

At time of diagnosis, 3 of the 5 patients with thrombosis were on standard dual antiplatelet therapy and 2 were on aspirin or clopidogrel alone. Anticoagulants were initiated in 3 patients (target INR 2.0-3.0) in combination with either aspirin or clopidogrel. The patient with idiopathic thrombocytopenic purpura received subcutaneous low-molecular weight heparin, and the patient with obstructive valve thrombosis and severe congestive heart failure received unfractionated heparin infusion followed by low-molecular weight heparin and warfarin but died on day 137 post-TAVR.  

The surviving 4 patients remained asymptomatic and demonstrated full thrombus resolution and normalized cusp mobility after 3 months on warfarin or low-molecular weight heparin therapy.  

“The frequency of incidental [valve] thrombosis calls into question the adequacy of current antithrombotic/anticoagulation strategies,” Dr. Hansson and colleagues write. “In this study, we observed [valve] thrombosis in 2 patients with severely depressed LVEF despite sufficient dual antiplatelet therapy. As such,” they advise, “anticoagulation may be indicated [for 3 to 6] months post-TAVI in certain high-risk patients.”  

They caution, however, that the findings may not be generalizable to other types of valves and patient cohorts. Furthermore, the researchers suggest further study on the usefulness of postprocedural multidetector CT when added to routine transesophageal and transthoracic echocardiography for detecting valve thrombosis in high-risk patients. 

One-Third of Cases Subclinical

Another study, conducted by Azeem Latib, MB BCh, of San Raffaele Scientific Institute (Milan, Italy), and colleagues, looked at 4,266 consecutive patients (mean age 79.7 years; 53.8% men) who received a Sapien/Sapien XT or CoreValve (Medtronic) device at 12 centers between January 2008 and September 2013.  

Valve thrombosis was diagnosed at a mean of 181 days after TAVR in 26 patients—20 Sapien/Sapien XT patients and 6 CoreValve—for an overall incidence of 0.61%. Only 1 patient had discontinued antiplatelet therapy by the time of diagnosis.  

While 65% of patients with valve thrombosis had worsening dyspnea as the presenting symptom, 31% had no worsening symptoms and were diagnosed on routine follow-up echocardiography. TEE, performed in 20 of the 26 patients, showed a markedly elevated mean gradient (40.5 ± 14.0 mm Hg). Thickened leaflets or thrombotic apposition of leaflets were seen in 76.9% and a thrombotic mass on the leaflets in 23.1%. Patients with thrombotic mass tended to present earlier than those with other causes of valve thrombosis. 

Importantly, neither echocardiography nor CT showed geometric deformation or dislocation of the implanted valve or evidence of thrombus on the frame.  

Medical therapy was initiated and recommended indefinitely in 23 patients with valve thrombosis and included oral vitamin K antagonists with or without bridging heparin or heparin without subsequent oral vitamin K antagonists in patients at high bleeding risk. Anticoagulation resulted in substantial decrease in gradient (P < .001) or disappearance of the thrombotic mass in all. Of the remaining patients, 2 underwent percutaneous valve-in-valve procedures and 1 had surgical valve replacement. 

Of the 3 patients in the cohort who died, only 1—a patient with recurrent thrombosis at 3 months following a valve-in-valve procedure—did so due to valve thrombosis. 

According to Dr. Latib and colleagues, the existence of subclinical valve thrombosis may be explained by the fact that these patients “have been preconditioned to the severe accompanying symptoms of severe aortic stenosis; hence, slight changes in breathlessness may go unnoticed.”  

They suggest that regardless of symptom status, all TAVR patients should have routine follow-up and receive anticoagulation therapy “as soon as valve thrombosis is suspected, not only in cases with visible thrombotic mass on valve leaflets but also in cases with [valve] dysfunction associated with thickened leaflets or apposition of leaflets with increased [mean gradient] on echocardiography.” 

Data Suggest Underdiagnosis

In an editorial accompanying the studies, Darren Mylotte, MB BCh, MD, of University Hospital Galway (Galway, Ireland), and Nicolo Piazza, MD, PhD, of McGill University Health Centre (Montreal, Canada), note that thrombosis after TAVR has been considered relatively rare, with just 10 published reports comprising 15 cases. There were no cases in the PARTNER trials and just 1 in the PARTNER EU registry, they add. 

Drs. Mylotte and Piazza say the 4% rate found on planned CT makes for “uncomfortable reading and lend[s] further credence to the belief that bioprosthetic thrombosis is underdiagnosed.” Additionally, they say, the study raises a number of questions about bioprosthetic thrombosis, such as:  

  • Is echocardiography insensitive to early valve thrombosis?
  • In what circumstances should CT be considered? 
  • What is the clinical course of asymptomatic nonobstructive valve thrombosis? 
  • What is the real contribution of valve thrombosis to post-TAVR stroke?  
  • At what time points should patients undergo routine valve imaging follow-up? 

“Given the low rates of clinically apparent [valve] thrombosis, routine postimplantation [CT] is hardly justifiable in the asymptomatic patient,” they write, “but should probably be considered in patients with echocardiographic evidence of [valve] dysfunction or where cardiac embolism is suspected, irrespective of the echocardiographic findings.”

1. Leetmaa T, Hansson NC, Leipsic J, et al. Early aortic transcatheter heart valve thrombosis: diagnostic value of contrast-enhanced multidetector computed tomography. Circ Cardiovasc Interv. 2015;8:e001596. 

2. Latib A, Naganuma T, Abdel-Wahab M, et al. Treatment and clinical outcomes of transcatheter heart valve thrombosis. Circ Cardiovasc Interv. 2015;8:e001779.

3. Mylotte D, Piazza N. Transcatheter aortic valve replacement failure: déjà vu ou jamais vu [editorial]? Circ Cardiovasc Interv. 2015;8:e002531.

Related Stories:

  • Dr. Hansson reports receiving unrestricted grant support from the Danish Heart Foundation and Edwards Lifesciences.
  • Dr. Latib reports serving as a consultant for Direct Flow Medical, Medtronic, and Valtech.
  • Dr. Piazza reports serving as a consultant for Medtronic.
  • Dr. Mylotte reports no relevant conflicts of interest.