Background and aim of the study: Although the incidence of paravalular leaks, the most common cause of non-structural dysfunction after valve replacement, is well defined, the results of their surgical correction are not. Given the growing enthusiasm for interventional catheter-based correction of paravalvular leaks, a current surgical baseline against which to compare these results is important.
Methods: All patients who had surgical correction of an aortic or mitral paravalvular leak unrelated to acute bacterial endocarditis between 1986 and 2001 were identified from a computerized registry. Hospital records were reviewed, and follow up data obtained.
Results: A total of 136 consecutive patients (73 males, 63 females; mean age 64 years) underwent surgical correction of a paravalvular leak. Of the valves, 44 (32%) were aortic and 92 (68%) mitral. More than one previous cardiac operation had been performed in 68 patients (50%). In 107 patients (79%; 32 aortic (73%), 75 mitral (82%)), the leak was the primary indication for reoperation, while for 29 patients (21%; 12 aortic (27%), 17 mitral (18%)) the correction was secondary to another cardiac procedure. In 65 patients (48%; 12 aortic (27%), 53 (58%) mitral)) the leak was repaired primarily, while in 71 patients (52%; 32 aortic (73%), 39 (42%) mitral)) the prosthesis was replaced. Operative mortality was 6.6% (n = 9). There were no significant multivariable predictors of hospital death. Perioperative stroke occurred in seven cases (5.1%), and hospital stay was >14 days in 40 patients (29%). The 10-year Kaplan-Meier survival was 30 (CI 20-39)%. Ten-year actual versus actuarial freedom from repeat paravalvular leak was 84 (CI 68-92)% versus 63 (CI 49-76)%.
Conclusion: Surgical correction of cardiac paravalvular leaks can be performed with acceptable mortality and morbidity. Patients with surgically corrected paravalvular leaks have a limited life expectancy, but reasonable freedom from recurrent paravalvular leak.