Ann Thorac Surg 2002;73:778
© 2002 The Society of Thoracic Surgeons
Invited commentary
Dieter Horstkotte, MDa
a Department of Cardiology Heart Center North Rhine-Westphalia Ruhr University of Bochum Georgstrasse 11 32545 Bad Oeynhausen, Germany
e-mail: akohlstaedt{at}hdz-nrw.de
Presented at the Thirty-seventh Annual Meeting of The Society of Thoracic Surgeons, New Orleans, LA, Jan 29–31, 2001.
An outer diameter given, stentless valves are purported to have hemodynamic benefits due to larger effective orifice areas. While past observational studies demonstrated low pressure gradients, Cohen and colleagues sought to examine the influence of valve design on myocardial re-remodeling and the patient’s postoperative functional status in a randomized trial. The authors are to be commended for undertaking such a rigorous study design on a topic of great interest to clinicians. In doing so, they reached unexpected conclusions. First, left ventricular mass regression, as their primary endpoint, does not differ after 12 months of follow-up in a statistically significant manner between the two patient groups with stentless versus stented prostheses. Second, functional capacity as measured by the Duke activity status index (DASI) also did not show differences between the two groups. Two characteristics of the study sample may account for these findings:
- Fifty percent of the patients had arterial hypertension. Persistent arterial hypertension has a significant influence on left ventricular re-remodeling after relief of valve-related pressure overload. The arguments of the authors that there was an equal distribution of hypertensive subjects in both randomized arms is not to the point: potential differences in left ventricular mass index reduction between the two groups may be obscured by the residual pressure load. In other words, if hypertension is considered a major and independent factor influencing postoperative myocardial re-remodeling, the power calculation itself may be invalid.
- Eighty three percent of the patients enrolled in the study had pure aortic stenosis and 76% of patients were in New York Heart Association (NYHA) class III/IV (but only 14% had signs of congestive heart failure, which seems to be a discrepancy). Patients with such advanced aortic stenosis, far beyond a time point that is considered "optimal" for surgical intervention, always present with a specifically altered collagen matrix. After relief of LV pressure overload by valve replacement, the collagen matrix does not show a re-remodeling within the first postoperative years, while regression of the contractile elements (regression of hypertrophy) is usually completed after 6–12 months. The functional capacity of the patients (measured in this study by the DASI) is, however, more significantly influenced by the diastolic than the systolic function of the left ventricle. Correlation between diastolic left ventricular function and myocardial/collagen remodeling is close. Again, the lack of differences between these two study groups may be entirely due to the fact that the disturbed left ventricular diastolic function could not recover during the short follow-up time irrespective of any replacement device.
It must thus be argued that the lack of expected differences between the two study groups is the result of major confounders, which highlights the extreme influence of an adequate study design on the study results. This makes assessment of potential differences between patients receiving the one or the other valve substitute extremely difficult, if not impossible. You will miss (expected) speed differences between two race cars if both participants are out of fuel.
Related Article
-
Are stentless valves hemodynamically superior to stented valves? A prospective randomized trial
- Gideon Cohen, George T. Christakis, Campbell D. Joyner, Christopher D. Morgan, Miguel Tamariz, Naoji Hanayama, Hari Mallidi, J.P. Szalai, Marko Katic, Vivek Rao, Stephen E. Fremes, and Bernard S. Goldman
Ann. Thorac. Surg. 2002 73: 767-778.
[Abstract]
[Full Text]
[PDF]
