| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Ann Thorac Surg 2003;76:657-658
© 2003 The Society of Thoracic Surgeons
a Department of Cardiac Surgery, Sunnybrook and Women’s College Health Sciences Centre, 2075 Bayview Ave, Room H-429, Toronto, ON M4N 3MJ, Canada
e-mail: g.cohen{at}sympatico.ca
To the Editor:
I thank Drs Pibarot and Dumesnil for their comments. My coauthors and I [1] realize that our findings are not in keeping with previously held misconceptions, and therefore, Drs Pibarot and Dumesnil are correct in approaching any such data with caution. However, in formulating an opinion, one must take into account a complete review of the literature. The statement that "these results are much better than those previously reported for the same prosthesis ... as well as for other stented bioprostheses" is erroneous. In fact, various authors have previously published findings with respect to valvular gradients and effective orifice areas that are similar, if not superior, to ours.
Takakura and colleagues [1] performed a hemodynamic evaluation of 19-mm Carpentier-Edwards (CE) pericardial bioprostheses and found mean gradients of 12.3 ± 4.8 mm Hg and effective orifice areas of 1.39 ± 0.26 cm2, values markedly superior to those reported by Banbury and associates [3], Dellgren and coauthors [4], or ourselves. Cosgrove and colleagues [5] similarly evaluated the CE pericardial valve in the aortic position. Mean valve orifice area was 2.0 ± 0.6 cm2 in 23-mm valves and 2.2 ± 0.7 cm2 in 25-mm valves, once again values exceeding those reported by Banbury and coauthors [3] or Dellgren and associates [4] and superior to those published by our group. A nearly identical trend was noted by Thomson and colleagues [6], who compared the hemodynamics of stentless, stented, and mechanical valves and reported mean effective orifice areas of 2.0 ± 0.6 cm2 in the stented group, a value larger than that reported by us. With respect to other stented bioprostheses, Williams and coworkers [7] carried out a randomized, controlled trial comparing the CE SAV (supra-annular valve) with the Toronto stentless porcine valve and found mean gradients of 8.9 ± 2.5 mm Hg and 6.3 ± 0.6 mm Hg in the stented group at 1 week and 32 months, respectively. In one of the earliest studies, Levine and co-workers [8] undertook a hemodynamic evaluation of Hancock and CE porcine bioprostheses and reported a mean gradient of 7.6 ± 6.2 mm Hg in the 23-mm CE subgroup, again a value similar to ours and much less than that of either Banbury and colleagues [3] or Dellgren and coauthors [4]. Such data are especially relevant in light of previous studies by Cosgrove and colleagues [5] that demonstrated superiority (less obstructive nature) of the CE pericardial bioprosthesis in comparison to conventional porcine bioprostheses.
Is this to say that one institution performs echocardiographic evaluations that are superior to those done by another? This is unlikely. However, it does speak to the wide variability inherent in echocardiographic measurements, especially when performed by multiple observers at varying sites. A 2001 study by Marquez and associates [9] demonstrated that mean pressure gradients and effective orifice areas reported by several institutions differed for similar valve types; for a given size, differences of 200% to 400% were observed.
Why do such wide discrepancies exist? They are likely due to both interobserver variability and confounding factors, which are often ignored (the most common being cardiac output). For this reason, our group adopted primary (left ventricular mass regression) and secondary (Duke Activity Status Index) outcomes that are less prone to contamination and conducted our study in the most sound fashion possible, using a randomized technique and only 2 blinded echocardiographers previously assessed for less than 5% interobserver variability. As all comparisons were performed by the same echocardiographer, any measurement biases would naturally have been applied to both valve types equally.
It is true that no study should be accepted at face value. However, one must make use of all available data to reach an informed decision as to the most optimal prosthetic heart valve for his or her patients. Ultimately, we must choose what is best for our patients and avoid the temptation to rely on selective referencing and industry-generated biases. ([2])
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
