MODERATOR PENNINGTON: Our first panel for this morning is to deal with the important question of patient selection. In my own mind patient selection in applying these devices is extremely important and perhaps the most important issue. I don‘t know that we know the answers, but at least we have some guidelines that we will be able to offer you. The panelists are Dr John Burkholder from Allegheny Hospital in Pittsburgh, who has had extensive experience with the Bio-Medicus pump; Dr Walter Pae of HersheyPenn State, who has a large experience with the PierceDonachy and the Penn State Artificial Heart; and Dr Lyle Joyce from Minneapolis, who has had experience with the Jarvik heart and the centrifugal pumps. So, we have most of the various disciplines represented. I would submit this rather worn-out information (Table 1) to you as being still the criteria by which patients must be selected, that is, the definition of cardiogenic shock. After all, that is the group of patients we are concerned about in the application of virtually all these devices. The exact numbers, of course, are not so critical, but it is critical that we actually make the measurements. Cardiac output measurements in the operating room, for example, are absolutely essential to the use of these systems. One must measure the atrial pressures, and we believe it is absolutely essential that the left atrial pressure line be used, not a wedge pressure from a Swan catheter. All these measurements must be carefully made and repeated to be certain that they are accurate. The definition of maximal inotropic support is one that might be debated long and hard, but most of us know that when we are in the polypharmacy and our anesthesiologist is at his wit’s end with the patient still not making it, we have to offer more support. Almost all of our patients receive a balloon pump before the insertion of one of these devices. Whether these exact criteria would be agreeable to you or not is not important. The important thing is that the measurements be made and that some sort of systematic method be used to choose patients. Probably one of the most important things not stated in this table concerning postcardiotomy patients is that we must be absolutely certain that the operation that we set out to do has been done appropriately and properly. These devices are definitely not to be used to cover up inadequate or inefficient operations. So we want to be sure that the patient has had an optimal operation. Now, who should be excluded? The criteria we have adopted were somewhat difficult to come by, but these