Early failure in total hip arthroplasty.

Although some patients experience a success rate greater than 90% after total hip arthroplasty, others require revision surgery within 5 years after the index procedure. The purpose of our study was to analyze the failure mechanisms in patients who had revision surgery within 5 years after index total hip arthroplasty. We retrospectively reviewed 824 revision total hip arthroplasties performed in 692 patients from 1986-2001. Seven hundred forty-five patients had adequate data. Two hundred ninety-one (39%) patients had revisions within 5 years after index arthroplasty. Ninety-six (33%) patients had revision surgery for instability, 88 (30%) for aseptic loosening, 41 (14%) for infection, 14 (5%) for osteolysis, 44 (15%) for failed painful hemiarthroplasties, and eight (3%) for periprosthetic fractures. Early revisions for aseptic loosening decreased from 38% in the early period to 24% in the current period, whereas revisions for instability increased from 9% to 42%. We were alarmed that 39% of the revisions at our institution were performed during the first 5 years after index surgery. Although improved fixation methods decreased early revisions for loosening, early revisions for instability increased substantially during the same time. Steps to avoid short-term failure must be taken.