I think it is really important and essential that when you are looking at the results of a clinical trial to see how it was designed, the hypotheses that were generated, the assumptions that were made, and the statistical analysis plan as well as the conduct of the trial itself. We need to look at whether it was multi-centered, randomized......
<International Circulation>:In the session this morning about pre-clinical studies you were discussing whether animal studies translate into human results. What has been your personal experience in this area? Do results of animal studies generally translate into humans?
Roxana Mehran: It is more often that animal studies do translate into what happens clinically, otherwise they wouldn’t be done. It is our only tool to assess the safety profile of something before it is placed into human beings. Of course, you must know that there are always limitations, and nothing is a perfect science. There are obviously good examples of when animal studies did not translate; they look good in the animal but they look terrible in the human being. However, I think for the most part, if you see a signal in the animal, one needs to be certainly worried about that particular signal and look for it in the clinical trials. That does not mean if you see a signal that you can not put that device in a human being, but you have to be careful about it. Think about what Cypher and Taxus stents have done in protecting so many patients against restenosis. There certainly was a signal, but had we not been given the opportunity to do human studies, we would not have had opportunity to have drug-eluting stents in our arena. Obviously the signals did mean something, especially with late stent thrombosis, and that was something we were able to address.