In rheumatoid arthritis (RA), non constrained or semi-constrained prostheses can be used. The authors used the Kudo III, IV or V or iBP prostheses 54 times from 1994 to 2003. After initial satisfactory results, they had to change one or both implants for several reasons: humeral stem fracture (5 cases), unipolar humeral loosening (1 case), ulnar loosening without laxity (8 cases), polyethylene wear (11 cases), due to progressive ulnar collateral ligament lengthening and progressive valgus deformity, without or with metallosis, due to contact between Cr-Co humeral component and titanium alloy ulnar component, chronic infection (1 case). When the local conditions were satisfactory (bone stock, ligament balance), the fractured or loosened component was changed. When the conditions were bad (poor bone stock, ligament misbalance, metallosis), both implants were removed; posterior humeral and/or medial or lateral ulnar window were used to removed the uncemented stems still osteointegrated. All the bipolar operations used the Coonrad-Morrey prosthesis, but the last case a Discovery prosthesis. The operative tricks are described, the management of the extensor apparatus is discussed, the clinical outcomes (especially the extensor apparatus function, most often weak) and the radiographic outcomes are presented

The purpose of this study was to assess the stability of a developmental pelvic reconstruction system which extends the concept of triangular osteosynthesis with fixation anterior to the lumbosacral pivot point. An unstable Tile type-C fracture, associated with a sacral transforaminal fracture, was created in synthetic pelves. The new concept was compared with three other constructs, including bilateral iliosacral screws, a tension band plate and a combined plate with screws. The pubic symphysis was plated in all cases. The pelvic ring was loaded to simulate single-stance posture in a cyclical manner until failure, defined as a displacement of 2 mm or 2°. The screws were the weakest construct, failing with a load of 50 N after 400 cycles, with maximal translation in the craniocaudal axis of 12 mm. A tension band plate resisted greater load but failure occurred at 100 N, with maximal rotational displacement around the mediolateral axis of 2.3°.

The combination of a plate and screws led to an improvement in stability at the 100 N load level, but rotational failure still occurred around the mediolateral axis. The pelvic reconstruction system was the most stable construct, with a maximal displacement of 2.1° of rotation around the mediolateral axis at a load of 500 N.