Dr. Schaff, Dr. Jamison, members and guests:
I would first like to state that Medtronic provides some financial support for the clinical and echocardiographic follow-up. Aortic valve dysfunction is the most common cause of valve replacement in our cardiac surgical population. The growing popularity of homograft and autograft valves in the late '80s combined with their outstanding hemodynamics encouraged the valve companies to produce stentless porcine aortic valves.
The Freestyle aortic root bioprosthesis was approved by the FDA in July 1992 for human investigation and use. It is a full aortic root with porcine coronary ostia intact, preserved in a buffered 0.2% glutaraldehyde solution and fixed with zero pressure gradient across the leaflets. This method of fixation preserves the natural collagen structure of the leaflets and maintains the native valve anatomy. The Freestyle valve is treated with an anti-calcification process, alpha amino oleic acid, synthesized from oleic acid, a naturally occurring fatty acid. A thin polyester cloth covering has been added to strengthen the proximal suture line and cover any exposed porcine myocardium.
This is a prospective, non-randomized clinical trial of the Freestyle aortic root bioprosthesis. Patients aged 20 years or older requiring aortic valve or aortic root replacement were eligible to participate. Those undergoing concomitant procedures such as coronary artery bypass or valve reconstruction were also considered, but patients requiring concomitant valve replacement or having a pre-existing prosthetic valve in another position were excluded from the study. From January 1993 to July 1997, 281 valves were implanted. Five valves were explanted during the same surgery, three for unacceptable hemodynamics, one valve because of occlusion of the right coronary ostium, and one because of a tear between the prosthesis and the anterior leaflet of the mitral valve. All of these explantations were early in the study, and the remaining 276 patients were used to complete the study.
There are three implant techniques commonly used for either homograft, autograft, or stentless bioprosthesis. The valve replacement, which was done in 238 patients, the root replacement in 36 patients, and the intra-aortic cylinder or inclusion in 2 patients.
Gender distribution was almost equal with 54% male and 46% female. The mean age was 67.7 years; 88.4 % were in normal sinus rhythm preoperatively and 86.2% were either NYHA functional class III or IV.
The most common etiology was senile degeneration in 52%, followed by congenital malformation in 29%. Risk factors and concomitant disease are listed on this slide. The most common was left ventricular dysfunction, which included left ventricular hypertrophy, wall motion abnormality, or LV dilatation. Hypertension, angina, and coronary artery disease were also frequent. It is important to note that 11.2% of the patient population had a history of previous CVA or TIA.
The distribution of implant sizes is summarized on this slide. Almost 60% of the implants were 25 or 27 mm. Total cross clamp time averaged 100 minutes for single valve replacement and 130 minutes when associated with another procedure. Cardioplegia was given according to the surgeon's preference, either cold, warm, or both. The most common associated surgical procedure was coronary artery bypass in 90 patients. Moderate to extensive annular debridement was also needed in 84% of the cases.
Data were available for the 259 survivors for follow-up of 100%. Two hundred were followed for more than a year, 144 for more than 2 years, 83 for more than 3 years, and 28 for more than 4 years. This is a short follow-up, however, the NYHA class remained stable with 95% of the group being in either class I or II. Sinus rhythm was present at follow-up in almost 90% of patients. The antithromboembolic therapy of choice was aspirin. Warfarin was reserved for patients in atrial fibrillation or with high risk of thromboembolism.
Fifteen patients died within 30 days of surgery, 8 in the aortic valve replacement group and 7 in the aortic root replacement group. The total early mortality was 5.4%, 3.3% in the valve replacement group and 19.4% for the root replacement group. Root replacement in this series was done for aortic root pathologies.
Freedom from death from any causes at 4 years is 86%. Eighteen patients died during follow-up. There were 2 valve-related deaths, 2 cardiac deaths, 1 unexplained death, and 13 non-cardiac. The valve-related deaths were caused by renal failure secondary to multi-organ failure 7 months postop in 1 patient and CVA 3 years postop in another one. Freedom from valve-related death at 4 years was 98.7%.
There have been 4 related reoperations. Two were for prosthesis endocarditis almost a year postoperatively. One patient was reoperated 33 days postoperatively for a perivalvular leak and the last one 4 years postoperatively for progressive central regurgitation due to aortic root enlargement. She underwent a successful root replacement with a homograft. Freedom from reoperation at 4 years is 98.8%.
There were 4 cases of endocarditis. Two were treated surgically and 2 were treated medically. Freedom from endocarditis at 4 years was 98.3%. The incidence of TIA and CVA was 7.6%, which is similar to series reported in population older than 70 years of age. There were 21 early events, of which 15 were permanent and 6 transient. Of the 12 late events, 6 were permanent and 6 were transient. The freedom from any neurologic event at 4 years was 87.3%.
Hemodynamic performance was assessed by echocardiography. Echo Doppler assessments were made of all survivors prior to discharge, at 3 to 6 months, and annually thereafter. Measurements were made of cardiac output, mean pressure gradient, effective orifice area, and regurgitation. The most interesting data encountered was regarding the mean gradient and the effective orifice area. Although there was no statistical difference between cardiac output at any time during follow-up, there was a statistically significant decrease in mean systolic gradient from discharge to 3-6 months, for a p of less than 0.001. Similarly, there was a statistically significant increase in the average effective orifice area from discharge to 3-6 months, for a p of less than 0.05.
Regurgitation is either trivial or mild in the majority of patients, and it is not hemodynamically significant. There is no evidence that regurgitation progresses; actually, it seems to get better with time.
In summary, the Freestyle aortic root bioprosthesis is versatile and can be used either as a valve or as a root replacement. Hemodynamic performances are comparable to those of homografts. There was a significant decrease in the mean systolic gradient from discharge to 3-6 months and a significant increase in the effective orifice area from discharge to 3-6 months. At 4 years postoperatively, 86% of the patients had either no or trivial regurgitation, and 11% of the patients had mild transvalvular regurgitation. The absence of a stent allows the implantation of a larger size prosthesis. Technically, the implantation is more demanding than with a conventional prosthesis and the cross clamp time is longer. There was no structural failure, and AOA treatment of the prosthesis is promising. Only a longer follow-up will give us the answer about the durability.
I would like to thank the Society for the privilege of presenting this paper.