Fifteen years' clinical experience with the CarboMedics prosthetic heart valve

Abstract

Background and aim of the study: The CarboMedics bileaflet prosthetic heart valve was first implanted as part of a prospective clinical study at the authors' institution in November 1987. The patient cohort included was part of a multicenter trial set up by the manufacturer for an FDA application. The present report details findings over a 15-year period, with a continuous follow up on this patient cohort.

Methods: Between November 1987 and August 1990, 132 patients (68 males, 64 females; median age 56 years; range: 12-74 years) received a CarboMedics heart valve prosthesis. All patients were included in the study, whether surgery was elective or emergency, first time or reoperation. There were 69 aortic, 49 mitral and 12 double (aortic and mitral) valve replacements. Two patients had isolated tricuspid valve replacement. Concomitant surgery was performed in 15 patients. Anticoagulation with warfarin was started on postoperative day 1. After discharge, patients were examined regularly as outpatients for up to five years. Subsequent follow up was obtained prospectively by questionnaires to the patients' general practitioners and with telephone calls to the patients. Actuarial estimates of survival and freedom from morbid events were calculated using the Kaplan-Meier method; 95% confidence limits for the distribution function were calculated according to the Greenwood formula.

Results: Complete follow up information was available on 94% of the patients; total follow up was 1,270.3 patient-years (pt-yr). Actuarial survival at 15 years was 51+/-4.9% overall; 56+/-6.2% for single aortic, 51+/-8.0% for single mitral, and 30+/-15.9% for double valve replacements. Actuarial rates of freedom from complications were: valve thrombosis 100%, embolism 89+/-3.3%, and all anticoagulant-related bleeding 76+/-4.5%. The linearized rates per 100 pt-yr were: embolism 0.94 (aortic 0.74, mitral 1.25); anticoagulant-related bleeding 2.28; paravalvular leakage overall 0.24 (aortic 0.29); prosthetic valve endocarditis overall 0.24 (aortic 0.29, mitral 0.21). There was no hemolysis, prosthetic valve dysfunction, or structural deterioration.

Conclusion: Over a 15-year time frame, the CarboMedics prosthetic heart valve has proven to be a highly reliable device with no structural failures and a low incidence of valve-related complications.