Prediction of outcome of total hip arthroplasty based on initial postoperative radiographic analysis. Matched, paired comparisons of failed versus successful femoral components

Abstract

All patients treated by total hip arthroplasty (THA) from 1971 to 1980 were analyzed to assess cause and rate of revision. Multiparameter radiographic analysis of failed femoral components ( FFC ) and matched control THAs was performed to determine predictability of failure, based on initial postarthroplasty and subsequent follow-up radiographic examinations. Rate of revision for all causes was 10.4% (35 THAs ) for 335 THAs in 270 patients with a follow-up period averaging 5.0 (range, 2-10) years. Cause of revision was component loosening in 7.2% (24 THAs ) [23 (6.9%) with femoral component loosening, 3 (0.9%) progressing to stem fracture, 3 (0.9%) with associated acetabular loosening, and 1 (0.3%) with isolated acetabular loosening]; infection in 1.5% (5 THAs ); and 1.8% (6 THAs ) comprised other causes. Older, inactive women, light in body weight (not including 42 THAs for rheumatoid arthritis), had fewer clinical failures. Second revision rate after revision for component loosening was 29% after an average follow-up period of 3.1 years. Paired comparison of initial postoperative and time-of-failure radiographs of 15 patients with THAs requiring revision for femoral component loosening versus initial postoperative and equivalent follow-up radiographs of 15 age-, sex-, weight-, activity-, diagnosis-, prosthesis-, and follow-up-matched patients with clinically successful THAs revealed the FFC group to be worse with regard to the following parameters: on initial postoperative radiographs, (a) femoral and acetabular bone-cement radiolucency (p less than 0.006 and p less than 0.02, respectively) and (b) when analyzed together, femoral bone-cement radiolucency, calcar-collar contact, femoral metal-cement radiolucency, femoral cement mantle adequacy, cement adequacy distal to the femoral component stem tip, and femoral component stem position (p less than 0.006); and on failure/equivalent radiographs, (a) femoral metal-cement radiolucency (p less than 0.01) and (b) when analyzed together, femoral metal-cement and bone-cement radiolucency, subsidence, cortical hypertrophy, calcar resorption, and sclerosis (p less than 0.006).