Percutaneous valve implantation: back to the future?

In the July issue of the European Heart Journal an experimental study was reported on the percutaneous implantation of heart valves in the systemic circulation. Since the report by Bonhoeffer et al. in 2000 on the first percutaneous pulmonary valve replacement in man, the two authors felt challenged to move to the percutaneous implantation of the aortic valve. Despite many commercial activities in this area, the authors are, to our knowledge, the first to report on this. After the creation of severe or moderate aortic valve regurgitation in eight lambs, a bovine jugular vein was mounted on an expandable, metallic stent and placed through the femoral artery into the descending aorta immediately after the take-off of the brachiocephalic trunk. The device was delivered successfully and the immediate function was normal in all animals. Four lambs with massive aortic regurgitation died within 24 h despite proper graft function. Four lambs with ‘mild’ aortic valve regurgitation had valve insufficiency for 2 months but good graft function. However, the native valve insufficiency was cured after 3 months and the recovered valve graft was dysfunctional. The behaviour of all the animals was similar. These experiments mirrored the approach of Hufnagel et al. from Washington, 50 years ago. Their animal experiments were mainly committed to the development of a prosthetic ball valve and fast and secure fixation of the device in the descending aorta. With their publication in 1954, describing 23 prosthetic valve implants in the proximal descending aorta of patients with severe aortic valve regurgitation, the modern era of surgical valve replacement therapy was started. By controlling an estimated 75% of the regurgitant volume, the 17 survivors improved in functional status and there were no early catastrophic device failures. What can be learned from the experience of Hufnagel et al. concerning descending aortic prostheses? Unfortunately the documentation of this procedure is extremely limited and over-optimistic. Twenty years after the initial report, Fishbein and Roberts conclude, in a review of the literature, that despite the distribution of 4000 prostheses for human implantation, the data of only 55 patients could be retrieved. The remarkable performance of the device was based on the fact that in 26 deceased patients