Experience with knee arthroplasty.

Knee arthroplasty is a suitable way of treating severely affected joints. During the years 1976-1987, 160 knee arthroplasties were performed at the author's institution. Of 142 followed. 41 cases were of the unconstrained type (Marmor, Bechtol, and Oxford), while 101 cases were of the constrained type (St. Georg hinge and Endo-Model). In the unconstrained group, two infected cases were encountered (one cured, one fused) and three loosenings requiring revision. Prompt and intensive physiotherapy was necessary in this younger and less affected group of patients in order to achieve a functioning knee joint. The average improvement of range of motion was 15 degrees. In the constrained group, full restoration of the normal axis was achieved with full extension; flexion was greater than 100 degrees, and there was a stable joint. Three infections occurred, all in rheumatoid patients. There were two early deaths from unrelated causes. Three late dislocations of the patella, with the knees still functioning, were encountered in this older group of patients with more severely affected joints. Proper function of the unconstrained prosthesis is also questionable in relation to the average geometry of the artificial models in use and the geometry of the human body. This relation, if not accurate, might result in loosening even in a well-installed prosthesis. The constrained artificial joint possessing its own mechanism of function allows full restoration of the leg axis and early ambulation and weight bearing, which are essential for the elderly. Bearing in mind the method of fixation, the function, and the results of the prosthesis used, the implantation of a constrained prosthesis more suitable to the elderly is strongly advised.