Highlights. In the present study, we conducted a retrospective analysis redo mitral valve replacement due to prosthesis failure. Predictors of adverse events and death have been analyzed. The quality of obtained predictors was verified using ROC analysis.Aim. To analyze predictors of adverse events and death in redo mitral procedure for treating mitral bioprostheses failure.Methods. The retrospective registry study of catamnesis of patients undergoing treatment for mitral valve prosthesis dysfunction was carried out. The age and gender of patients, medical history, underlying and comorbid diseases, features of primary replacement and reoperation, durability of the primary prosthesis, adverse events and death after redo mitral valve replacement and others (more than 30 factors) were taken into account. Stepwise logistic regression was used for statistical processing. The quality of obtained predictors was verified using ROC analysis.Results. The results of the study have revealed that there are modifiable risk factors such as cardiopulmonary bypass time, ischemia time, and nonmodifiable risk factors like reoperation, baseline severity of the patient's condition, complications of median resternotomy, concomitant interventions on aortic valve during reoperation. Moreover, treating atrial fibrillation during primary replacement proved to have an overall protective effect.Conclusion. Тhe only modifiable factors that can be influenced during redo mitral valve replacement due to biological prosthesis failure are cardiopulmonary bypass duration and aortic cross-clamping time.