PURPOSE: Total knee replacement (TKR) failure represents a hard challenge for knee surgeons. TKR failure can be managed in revision with different constraint, related with soft and bone knee damages. The choice of the right constraint for every failure cause represents a not summarized entity. The purpose of this study is identifying distribution of different constraints in revision TKR (rTKR) for failure cause and the overall survival. METHODS: A registry study based on the Emilia Romagna Register of the Orthopaedic Prosthetic Implants (called RIPO) was performed with a selection of 1432 implants, in the period between 2000 and 2019. Selection implants including primary surgery constraint, failure cause and constraint revision for every patient, and divided for constraint degrees used during procedures (Cruciate Retaining-CR, Posterior Stabilized-PS, Condylar Constrained Knee-CCK, Hinged). RESULTS: The most common cause of primary TKR failure was aseptic loosening (51,45%), followed by septic loosening (29,12%). Each type of failure was managed with different constraint, the most used was CCK in the most of failure causes, such as to manage aseptic and septic loosening in CR and PS failure. Overall survival of TKA revisions has been calculated at 5 and 10 years for each constraint, with a range of 75.1-90.0% at 5 years and 75.1-87.5% at 10 years. CONCLUSION: Constraint degree in rTKR is typically higher than primary, CCK is the most used constraint in revision surgery with an overall survival of 87.5% at 10 years.