Author + information
In Dr. Bloomfield’s letter discussing the Viewpoint (1) and Commentary (2) recently published in the Journal, Dr. Bloomfield notes that, in long-term follow-up of the Edinburgh Heart Valve Trial (3), 1) mortality at 20 years did not differ between mechanical and tissue prostheses; 2) owing to differences in rates of re-operation, freedom from all valve-related complications was superior after mechanical mitral (but not aortic) valve replacement; and 3) over time, an increasing use of anticoagulant therapy was seen after tissue valve replacement. Dr. Bloomfield concludes that long-term follow-up of patients in randomized trials is the only means by which differences between prostheses can be elucidated.
Equivalence of mortality at 20 years supports that, using this end point, existing data (including those from the Edinburgh Heart Valve Trial) do not favor one prosthesis type over another. Freedom from all valve-related complications favors neither prosthesis type following aortic valve replacement, which comprises a preponderance of valve replacement surgeries in the United States; for both mitral and aortic valves, higher rates of re-operation after tissue compared with mechanical valve replacement exemplifies an inherent difference between prostheses. However, re-operation may be an end point more universally feared by cardiologists than by patients. Operative mortality for elective “re-do” valve replacement is not substantially different from that for initial surgery. Whereas some patients may wish to avoid re-operation at all costs, others are clearly willing, if given the option, to face the near-certainty of additional surgery in order to avoid daily anticoagulation. This argument is not obviated by an early experience showing increasing requirement over time for anticoagulation due to atrial fibrillation. Atrial fibrillation carries a lower thromboembolic risk (and therefore requires less aggressive anticoagulation) compared with mechanical mitral valve replacement. Moreover, earlier intervention as well as newer medical therapies and surgical and percutaneous procedures presently available for the treatment of atrial arrhythmias can be expected to lower the need for additional long-term anticoagulation compared with the cohort studied from the 1970s.
The tenet of the previously published Viewpoint (1) is not that long-term data are not desirable, but that for prosthetic valve choices that are made today, pertinent long-term, randomized data do not (and will never) exist. Although Dr. Bloomfield and his co-investigators (3) are to be commended for their study, the data as they relate to current decisions in prosthetic valve surgery are moot. In 2004, neither the Bjork-Shiley nor the original Hancock valve is available for implantation. Future randomized studies are unrealistic, and even if performed, valve technology would again have evolved by the time long-term data became available. Rather than attempting to extrapolate to current practice “tissue versus mechanical” data from valves that are obsolete and in some cases known to be poorly representative of currently available prostheses, this author asks the clinician to understand the limitations of these data, and to recognize that the gradual and continuing advances made in heart valve prostheses make a demand for only long-term, randomized data unrealistic and therefore unwise.
- American College of Cardiology Foundation
- Bach D.S.
- Rahimtoola S.H.
- Oxenham H.,
- Bloomfield P.,
- Wheatley D.J.,
- et al.