Despite the advent of antibiotics, cardiac valve replacement for native valve endocarditis is still required in subgroups of non-responders. While experience with single valve replacement has been widely described, reports on mitro-aortic valve surgery in this setting has been rarely addressed. The majority of patients with multivalvular endocarditis have had prior valve disease, mostly rheumatic heart disease. Because of the destructive nature of the infectious process, most valves exhibit regurgitation. In multi-valvular involvement, there is usually a direct continuity of the infective process of the aortic valve with the mitral valve. As for single valve surgery, the indication and timing of multivalvular surgery are highly interdependent: broadly, antibiotics are preferred initially to control infection, however, any delay of the surgery may result in further morbidity and mortality, which is likely to be increased with multivalvular involvement. The most important technical aspect of this surgery is thorough debridement of the infectious process, while neither the type of valve (biological vs. mechanical) nor the type of suture (continuous vs. interrupted) seem to play a role. Mitral valve reconstruction may be used with good long-term results when the nature of the infective process is limited. After cardiac outcome, neurological complication is the other major issue following surgery, underlying the necessity for a complete preoperative work-up. The risk of recurrent endocarditis can be limited, provided that the surgical debridement is radical and that adequate antibiotic therapy is performed. Inherent complications of prosthetic valve surgery, thromboembolism and bleeding, are not increased in comparison with single-valve surgery. In conclusion, with timely surgery, good valve debridement and adequate antibiotic therapy, mitroaortic valve surgery for native valve endocarditis can be performed with satisfactory short- and longterm results.