The GSB total knee arthroplasty. A medium- and long-term follow-up and survival analysis.

From 1981 to 1987, 77 GSB-II total knee arthroplasties were implanted in 65 patients. There were 23 men and 42 women aged on average 60 years old (range 30-85 years). The diagnosis was osteoarthritis (OA) in 21 knees, rheumatoid arthritis (RA) in 44 knees, and other in 12 knees. A clinical and radiological follow-up was performed in two stages after a mean of 6.7 years (61 knees) and 14.8 years (22 knees) to assess the medium- and long-term results and to determine if deterioration had occurred after mid-term follow-up. A survival analysis was done with two endpoints: (1) revision, and (2) revision, moderate or severe pain and lost to follow-up (worst-case scenario). At the last follow-up 36 patients (44 knees) had died, 2 patients (2 knees) refused examination, and 3 patients (3 knees) were lost to follow-up. Six knees had been revised for malposition (1.3%), septic (3.9%) and aseptic (2.6%) loosening. The mean Knee Society score after 6.7 and 14.8 years was 85 points (OA 82 points, RA 87 points). Lateralisation, subluxation or dislocation of the patella was present in 8 of 17 knees at the last follow-up. The 6- and 15-year survival rates with revision as the endpoint were 95% (CI 89%-100%) and 87% (CI 65%-100%), respectively. For the worst-case scenario, the 6- and 15-year survival rates were 95% (CI 89%-100%) and 56% (CI 0%-100%), respectively. The medium- and long-term results of the GSB-II total knee arthroplasty were good, and a decline in the knee score did not occur beyond the mid-term follow-up. Patella complications were abundant, and a marked decrease in implant survival was noted when moderate or severe pain and lost to follow-up were included as endpoints.