Potentials and pitfalls of clinical outcome research studies in cardiac surgery.

Al-Halees et al. [1], Duran et al. [2], Amano et al. [3], and Grinda et al. [4] had mostly rheumatic patients (78—100%), which allowed us to draw the conclusions in the paper. In addition, it is important to understand that one of the main reasons for attempting aortic valve repair is because the patients are young. The twomain aortic valve pathologies found in young patients are rheumatic and congenital. Calcific or degenerative aortic disease will be found in older patients in whom we already have a good therapy: prosthetic replacement. There may not be much justification for attempting repair in these patients, and therefore it is important to describe the results in the specific patient population in whom repair has been tried, namely the young patient with rheumatic or congenital pathology. Thus far, repair does not seem to have durable results in the rheumatic population. We must hope that the congenital population will fare better. Dr Izumoto’s second point is also well taken. Many different types of repair were described in each case series and it was very difficult to ascertain which methods of repair were successful and which were not. It is currently unknown if a triangular resection may perform better than a leaflet placation. This is another area where future research will need to be done. I again thank Dr Izumoto for his insightful comments in an area where he has a great deal of expertise.