Mechanical Aortic Valve Prostheses in the Small Aortic Root: Top Hat Versus Standard CarboMedics Aortic Valve

Background The purpose of this study was to evaluate outcome in patients with a small aortic root receiving either a standard CarboMedics mechanical aortic valve or a Top Hat CarboMedics valve (CarboMedics, Austin, TX), specifically designed for the small aortic root. Methods Between 1986 and 2006, 316 consecutive patients underwent 19- or 21-mm mechanical aortic valve replacement, receiving either a CarboMedics Top Hat bileaflet valve (n = 56; mean age, 66 ± 14 years) or a standard CarboMedics aortic valve replacement (n = 260; mean age, 60 ± 13 years) at our institution based on institutional indications for the choice of type of valve prostheses. Median follow-up time was 83.5 months. We studied survival, valve-related and non–valve-related events, and hemodynamic performance by serial echocardiographic follow-up studies. Results In-hospital mortality was 8.9% in the Top Hat group and 10.0% in the standard group (p = 0.354). Five- and ten-year survival in patients in the Top Hat group was 83% and 67%, respectively. Five- and ten-year survival in the standard group was 73% and 59%, respectively (log-rank = 0.331). There were no differences in regard to valve-related and non–valve-related events. Cox regression analysis revealed age (hazard ratio, 1.045; 95% confidence interval, 1.026 to 1.066), previous cardiac surgery (hazard ratio, 1.812; 95% confidence interval, 1.101 to 2.982), additional procedures at the time of valve replacement (hazard ratio, 2.604; 95% confidence interval, 1.651 to 4.108), New York Heart Association class IV (hazard ratio, 3.645; 95% confidence interval, 1.214 to 10.945), and severely impaired left ventricular ejection fraction (hazard ratio, 2.253; 95% confidence interval, 1.289 to 3.941) to be independent predictors of survival. Conclusions Mechanical aortic valve replacement in the small aortic root is associated with substantial perioperative mortality, in particular in the subset of patients requiring additional cardiac surgical procedures. Nevertheless, long-term outcome is satisfying. Because the type of prosthesis does not predict outcome in the multivariate Cox model, we conclude that use of the smaller Top Hat prosthesis can be recommended for the challenging cohort of patients with a small aortic root.