Thrombosis Seen as a Common, Early Cause of Bioprosthetic Valve Failure

Veronica Hackethal, MD

November 30, 2015

ROCHESTER, MN — Bioprosthetic valve thrombosis (BPVT) is more common than often perceived and develops long before the valve fails structurally in patients with symptoms referred for surgery, suggests an analysis published in the December 1, 2015 issue of the Journal of the American College of Cardiology, with Dr Alexander C Egbe (Mayo Clinic, Rochester, MN) as lead author[1]. It also identified independent clinical and echocardiographic predictors of thrombosis in such valves, predominantly aortic valves but also mitral and tricuspid valves, that included paroxysmal atrial fibrillation (AF), subtherapeutic anticoagulation, and abnormal cusp motion.

The study's novel insights, coauthor Dr Sorin V Pislaru (Mayo Clinic) said to heartwire from Medscape, include a rate of BPVT as high as 11% among patients with bioprosthetic valves who are referred for surgical intervention, suggesting that the perceived rate of BPVT in such patients is "significantly underestimated due to lack of awareness." In addition, "bioprosthetic thrombosis appears significantly earlier than degeneration," he said, as early as 2 years after valve implantation in many cases.

"Aggressive surveillance could help identify such patients and allow them to benefit from anticoagulant therapy, which we and others have shown to be successful," Pislaru said. That suggests the current guidelines should be updated to recommend surveillance of bioprosthetic valves within the first 2 to 3 years after implantation. Currently, he said, an annual evaluation is recommended only for transcatheter aortic-valve replacement (TAVR).

Current European Society of Cardiology (ESC) guidelines do not recommend vitamin-K antagonists (VKAs) like warfarin longer than 3 months after implantation for aortic bioprosthetic valves. Guidelines from the American College of Cardiology (ACC) and American Heart Association (AHA) recommend VKAs for 3 to 6 months in patients with aortic-valve replacements, the group notes. Neither the ESC nor the ACC/AHA guidelines recommend VKAs past 3 months for patients who receive mitral, tricuspid, or pulmonary valves.

"More robust, prospective data on the role of anticoagulation in prevention of leaflet thrombosis should be gained before routinely changing medical regimens for these patients," stressed Prof Lars Søndergaard (Rigshospitalet, Copenhagen, Denmark) for heartwire . "Particularly in elderly patients, anticoagulation can have a high bleeding risk, which may outweigh the benefit of reduced leaflet thrombosis." Søndergaard isn't connected with the analysis from Egbe and colleagues.

"There is no clear indication for anticoagulation in all patients with bioprosthetic valves," added Dr Samir Kapadia (Cleveland Clinic, OH). "Routine echocardiogram is a good idea every year. There is no need for alarm for patients with bioprosthetic valves."


Last month, heartwire covered a report based on patients in the PORTICO-IDE study that described computed tomographic (CT) imaging evidence of subclinical thrombosis in 40% of patients who underwent TAVR using valves from different manufacturers, which resolved with anticoagulation. The same report described CT evidence of reduced leaflet motion in patients with newly implanted aortic valves in cohorts from the RESOLVE and SAVORY registries. In response, the FDA issued a notification through MedWatch recommending that patients who have bioprosthetic aortic valves may need additional diagnostic imaging, such as 3D or 4D CT or transesophageal echocardiography (TEE).

Although the current study and the PORTICO-IDE study both found evidence that therapeutic anticoagulation may protect against both clinical and subclinical valve-leaflet thrombosis, comparisons of the two studies can only go so far, as there are important differences between them, according to Søndergaard. For example, the current study from Egbe and associates included patients in whom symptoms let to referral for surgical intervention and used pathological findings to diagnose BPVT, while the PORTICO-IDE study included asymptomatic patients and used CT for identifying thrombosis.

"The major limitation of CT technology is related to the inability to view 'live' images of the valve," Pislaru pointed out, "In echocardiography, we have learned the hard way that continuous and meticulous imaging is key when assessing rapidly moving thin structures such as valve cusps."

Upon review of original echocardiography reports, only 5% of transthoracic echocardiograms (TTE) and 13% of TEEs were associated with possible thrombosis. When the researchers reviewed digitally stored images, however, they found hallmarks of thrombosis on the TEEs of all tricuspid and mitral bioprostheses and the majority of those with aortic bioprostheses, Pislaru said. Given the current analysis and those from PORTICO-IDE, "we believe bioprosthetic thrombosis after TAVR is underestimated."

Thrombosis Preceded Structural Failure

In the study, researchers identified cases of BPVT diagnosed pathologically between 1993 and 2013 using the Mayo Clinic pathology database. Then researchers matched patients with BPVT in a 1:2 ratio for age, sex, and prosthesis position with patients who had valves removed due to structural failure. They obtained clinical, echocardiographic, and surgical information from medical records. Experienced echocardiographers reviewed randomly selected studies from one-half of the cohort.

Of 397 consecutive patients with explanted bioprosthetic valves, 11.6% (n=46) were identified as cases of BPVT (of which 63% were aortic, 20% mitral, 15% tricuspid, and 2% pulmonary).

Bioprosthetic valve thrombosis occurred a median of 24 months after surgery, whereas structural valve failure occurred a median of 108 months after surgery (P<0.001). In multivariate analysis, independent predictors of valve thrombosis included:

  • Paroxysmal AF (hazard ratio [HR 5.19, 95% CI 1.42–21.77; P=0.01).

  • Subtherapeutic INR for patients on VKAs (HR 7.37, 95% CI 1.60–42.13; P=0.001).

  • Increase in mean transvalvular gradient of >50% above baseline within 5 years, in the absence of a high cardiac output state (HR 12.7, 95% CI 2.83–75.91; P=0.001).

  • Increased cusp thickness (HR 12.2, 95% CI 2.71–75.08; P=0.001).

  • Abnormal cusp mobility (HR 6.94, 95% CI 1.74–33.45; P=0.005).

The presence of three echocardiographic features predicted BPVT with 72% sensitivity and 90% specificity.

Prospective trials are needed to address several questions raised by this study, writes Dr William Stewart (Cleveland Clinic) in an accompanying commentary[1]. Issues include whether the findings can be generalized to a larger population of diverse patients and clinicians, the efficacy and risks of anticoagulation using new agents, the benefits and costs of annual echo screenings for all patients with bioprosthetic valves, and the benefits vs risks of thrombolysis compared with surgery.

"Assuming that the patient groups manifest the important differences between thrombosis and degenerative changes of BPV, this paper . . . has important implications," Dr Stewart concluded, "It incites us to search more diligently for BPVT, obtain baseline 'finger-print' echocardiographic studies, perform yearly echocardiographic studies even in the early postoperative period, and lower the threshold for prophylactic and therapeutic anticoagulation in patients with bioprosthetic valves."

The authors report no relevant financial relationships, as do Stewart, Kapadia, and Søndergaard.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.