Approximately 700 TKRs (Total Knee Arthroplasty) were performed in our department from 1992–2000. In 13 cases, patellar dislocation occurred post-operation - 12 with no trauma and one following a fall and trauma to the medial aspect of the knee. Dislocation occurred from 2 months following the operation until 1 year postoperative (average 4.5 months). None of the patients had malalignment of the components. The average femorotibial angle was 5° (range 8° valgus to 3° varus). No patient had preoperative valgus deformity. Postoperative range of motion was 105° (range 90–125°). All dislocations were treated by operative lateral relapse and medial capsular implication followed by 6 weeks of immobilization with a brace in full extension and then physiotherapy, range of motion, patellar taping and Vastus medialis strengthening. The patella stabilized in 11 cases following the operation. In 2 cases, dislocation recurred following the operation and they were operated on again where medialization of the tibial tubercle was performed. Both cases stabilized following the second operation. One case developed a stress fracture of the tibia at the end of the tubercle osteotomy which healed conservatively. All dislocations occurred in the IBII prosthesis. During the past 3 years since using the PFC Sigma and Legacy prosthesis, we have had no dislocations of the patella, probably for the following reasons:
Use of the mid vastus exposure. Geometry of the prosthesis. In only 10% of the replacements, patellar resurfacing was performed.
Forty-two revision knee replacements were performed in our department between 1992–2000. We report our experience in 18 cases of stiff knees with a range of motion from −5° – 75° (average 50°) where an oblique incision through the quadriceps tendon combined with medial capsular incision (the “wandering resident” incision) was used for exposure. This exposure allowed us to expose the stiff knee with no hazard of avulsion of the patellar tendon and with easy removal of the old prosthesis and implantation of the new one. In 5 of these cases, this exposure was used twice in two stage revisions of a septic prosthesis. Post-operative rehabilitation was slower, a knee brace was used in extension for 6 weeks and daily physiotherapy and CPM from 0°–70° only. Full range of motion was started after 6 weeks. Follow-up in 1–8 years (average 3.5 years). All patients had good clinical results with range of motion from 0°–110° (average 86°). One patient had a lag of 10° in active extension. The knee score of the American Knee Society ranged from 35–52 (average 40) and improved to 72–89 (average 84). In 3 cases, we used a non-constrained prosthesis (PCL) sacrificing condylar prosthesis), in 11 cases a constrained prosthesis (CCK type) and in 4 cases a rotating hinge prosthesis.
