Evolution of mechanical heart valves.

The need for prosthetic heart valves was long recognized but seemed an impossible dream before 1952 when Dr Charles Hufnagel clinically introduced a ball valve that he placed into the descending thoracic aorta for treatment of aortic valvular insufficiency. Fulfillment of that dream became a reality with the advent of extracorporeal circulation in the early 1950s. Development of prosthetic heart valves involved the search for biologically compatible materials and hemologically tolerant designs. Success could not be achieved without the union of these two factors. As there was no satisfactory mechanism to scientifically achieve this goal, trial and error was the method of choice. The development of prosthetic heart valves became the purview of the cardiovascular surgeon who often collaborated with engineers. To distinguish one valve from another each prosthesis often became identified with the surgeon developer. The development of bioprostheses occurred later in the development of artificial heart valves and constitutes a separate subject not covered in this presentation.

The Omni design: evolution of a valve.

The Lillehei-Kaster prosthesis has been the subject of an engineering evolution to close the gap between engineering expectations and clinical performance. Advanced engineering development in response to users has narrowed this gap with the Omni design (Omniscience titanium and omnicarbon Pyrolite cages). Our studies on mitral Omniscience valves demonstrated that because anatomic and surgical variations, the anterior orientation was more forgiving than the posterior orientation, resulting in lower thrombotic complications (0.5% versus 3.3% patient-year). A subsequent thicker cuff was used to reduce the probability of anatomic interference and incomplete opening after cardiac recovery. Such findings were also incorporated in the Omnicarbon design. Five European centers implanted 354 patients (198 atrial valve replacement, 115 mitral valve replacement, and 41 double valve replacement with Omniscience valves between August 1984 and January 1986. No restrictive patient selection criteria were used. As of June 1987, 96% of the patients at risk were accounted for. Average follow-up was 1.7 +/- 0.4 years (range, 0.3 to 2.8 years), with a total follow-up of 555 years. There were no cases of structural failure or clinically significant hemolysis. Ninety-one percent of the patients improved one or more functional classes. Actuarial analysis of survival probability at 3 years is 92.5% for atrial valve replacement, 97.9% for mitral valve replacement, and 93.6% overall. There were no late deaths from valve thrombosis or thromboembolism. The actuarial freedom from all thromboembolic events (valve thrombosis, thromboembolism, and transient ischemia) is 97.8% at 2 1/2 years. Through close rapport between user surgeons and designers, the evolution of a valve with improved performance becomes a reality.

Implantation techniques: a primary consideration in valve surgery.

Clean site preparation for a prosthesis is mandatory to avoid tissue interference with the prosthesis. Oversizing of a prosthesis is fraught with peril. In all types of prostheses, orientation has importance. A prosthesis is designed with consideration for axial flow. Tilting the valve in implantation will likely result in a dysfunctional valve. The sewing ring serves not as a gasket but as an additional factor that increases the possibilities of success of a properly implanted valve by taking advantage of its naturally compliant features.

Seventy-six month experience with the Omniscience cardiac valve.

Between August 1978 and September 1984, 440 patients were implanted with the Omniscience cardiac valve at three North American medical centers (210 aortic, AVR; 165 mitral, MVR; and 65 double valve replacements). Total follow-up was 1056 patient years; survivors were followed up to 76 months (mean 32), with 99% accountability. One hundred and eighty-six (42%) also underwent prior or concomitant major cardiac surgical procedures. Mean age was 56 +/- 12 years. Preoperatively, 3% were in New York Heart Association (NYHA) Class I, 18% II, 54% III, 26% IV. The 5 1/2 year actuarial survival rate for death from valve-related causes (thromboembolism, valve thrombosis, perivalvular leak) was 97% overall. The 5 1/2 year actuarial survival rate from all death causes was 78%. The complication rates, expressed as %/patient-year (endocarditis 0.7; anticoagulant-related hemorrhage 1.0; perivalvular leak 1.5; pannus/tissue overgrowth 0.6; and transient ischemic episode 1.2) demonstrate that advanced age at implant and additional major cardiac surgical procedures do not affect occurrence. Actuarial freedom from thrombotic complications (thromboembolism with residual deficit and valve thrombosis) at 5 1/2 years was 95% overall. Postoperatively, of 348 patients with known NYHA, 65% were in NYHA I, 28% II, 5% III, and 2% IV.

The first open heart corrections of tetralogy of Fallot. A 26-31 year follow-up of 106 patients.

Tetralogy of Fallot became a correctable malformation on August 31, 1954, and from that data through 1960, 106 patients (ages 4 months-45 years) who underwent open repairs at the University of Minnesota and were discharged, have been followed (99% complete) until death or for 26-31 years (mean: 23.7 years, 2424 patient years). The purposes of this study were to determine survival, morbidity, hemodynamics, educational/employment attainments, and relation of these to surgical technics. Operations were done by cross circulation (6 patients) and bubble oxygenator (100 patients). This group had the first uses of patch ventricular septal defect closure, outflow root, infundibuloplasty, atresia correction, ischemic arrests, and pacemakers among other innovations. Twenty-one (of 105 patients) have died during the followup: eight deaths in the first 10 years, 12 between 10 and 20 years, and 1 greater than 20 years. The causes of death were sudden (5), accidental (4), congestive failure (2), reoperation (2), suicide (2), and other (2). Actuarial survival at 30 years was 77%. Late complications were ten reoperations, five arrhythmias, and one endocarditis. Actuarial freedom from reoperations at 30 years was 91%. Cardiac recatheterizations in 62 patients disclosed only 10 with residual shunts. Peak right ventricular systolic pressures were less than 40 mmHg (34 patients), 41-60 mm (2 patients), 61-70 mm (4 patients), greater than 71 mm (4 patients). Thirty-four patients (32%) completed college, ten of these completed graduate school (5 masters degrees, 2 M.D.'s, 2 Ph.D.'s, 1 lawyer). Fifteen others attended college, and nine received technical school diplomas. Forty patients (18 men, 22 women) had progeny, with 82 (93%) live births and six major cardiac defects (7.3%). In summary, complete repair gave excellent late results in this group cared for very early in the open heart era. Survivors led productive lives without restrictions in education and employment. Many of the deaths/complications that occurred are now easily preventable, which augurs extremely well for this generation.

Fluorocarbon cardioplegia and myocardial protection.

This study compares myocardial protection using nonoxygenated clear cardioplegia with oxygen-carrying solutions of blood (PO2, 100 mm Hg) and fluorocarbon (FC-47 perfluorotributylamine, PO2, 500 mm Hg), all containing 25 mEq/L of potassium chloride. Three groups of dogs, each consisting of 5 animals, were placed on cardiopulmonary bypass, and the aorta of each dog was cross-clamped for 45 minutes. Hemodynamic and biochemical variables were measured at baseline and during recovery. Levels of the myocardial isoenzyme of creatine phosphokinase (CPK-MB) in the coronary sinus were significantly lower in the fluorocarbon cardioplegia group at 15 minutes of aortic cross-clamping (p less than 0.01), while both the fluorocarbon and blood cardioplegia groups demonstrated lower CPK-MB levels at 45 minutes (p less than 0.001 and p less than 0.05, respectively), compared with the clear cardioplegia group. The blood and fluorocarbon groups had improved mean aortic blood pressure (p less than 0.02 and p less than 0.05, respectively) and left ventricular pressure. At 45 minutes of reperfusion and recovery, all hemodynamic and enzymatic variables were similar in each group. We conclude that oxygenated solutions better protect the myocardial cell, but that further work is needed to determine the most effective oxygen level for maximum protection with cardioplegia.

Membrane vs bubble oxygenators: a prospective study of 52 patients.

Controversy continues about the oxygenator preferable for cardiopulmonary bypass (CPB). This prospective study was undertaken in 52 patients undergoing coronary bypass surgery. Oxygenators were alternated each case between model Q-100, Bentley Laboratories Inc, Irvine, Calif, and Travenol Membrane Oxygenator (TMO), Travenol Laboratories Inc, Deerfield, Ill. The Q-100 group required higher CPB O2 flows, but PO2 levels during CPB were similar for both groups. Heparin sodium dosage and activated clotting, bleeding, prothrombin, and partial thromboplastin times were identical in both groups. Blood loss and platelet reduction after CPB were also similar. Postoperative complications in the Q-100 group included one myocardial infarction, and one neurological problem. The TMO group had no myocardial infarction and one neurological problem. The membrane oxygenator took nine minutes longer to set up and was $63 more expensive to purchase. Blood trauma during CPB was less with the membrane oxygenator (lower plasma hemoglobin level), but we conclude that both oxygenators performing adequately during clinical use in open heart surgery.

Clinical experience with the Lillehei-Kaster valve prosthesis.

The Lillehei-Kaster valve was used in 215 patients over a 7-year period. The aortic valve was replaced in 81 of them. Hospital mortality was 21% (17 patients). Ten patients were lost to follow-up and long-term mortality was 5% (3 patients). The incidence of thromboembolism was 2.6 per 100 patient-years, and actuarial survival was 96% at 5 and 87% at 7 years. Mitral valve replacement was performed in 108 patients. Hospital mortality was 20% (22 patients), 24 patients were lost to follow-up, and long-term mortality was 13% (11 patients). The combined incidence of thromboembolism was 5.0 per 100 patient-years, and actuarial survival was 81% at 5 years and 75% at 7 years. In both groups, the majority of patients improved clinically. Fifteen patients had double-valve replacement, and 11 had coronary revascularization in addition to valve replacement.

Ventricular fibrillation without left ventricular venting. Observations in humans.

Pressures were measured in the heart and great vessels of 52 patients who underwent coronary revascularization. In 25 patients the left ventricle was vented during fibrillation; in the others a vent was not used. Samples for blood gas analysis were obtained twice during fibrillation from the aorta, left atrium, pulmonary artery, and right atrium. Left ventricular venting was found to be effective in keeping mean left heart pressure below 10 mm Hg, although temporary incompetence of the aortic valve or malfunction of the vent occasionally caused higher pressures. Vent use led to air embolism in the aorta in 16% of the patients. In the nonvented patients mean left heart pressures remained between 10 and 20 mm Hg. However, higher values were frequently observed. Blood gas analysis demonstrated that without venting, retrograde pulmonary flow occurred during fibrillation. No abnormality was encountered that could be related to nonventing.