Predictive value of early radiographic findings in cementless total hip arthroplasty femoral components: an 8- to 12-year follow-up.

Total hip replacement was performed in 119 patients (124 hips) with an uncemented Ortholoc femoral component between 1982 and 1987. Eight patients (9 hips) died, and 13 patients (15 hips) were lost to follow-up, leaving 98 patients (100 hips) followed for 8 to 12 years. Pain scores and anteroposterior and lateral radiographs of the femoral component from the 2-year follow-up visit were evaluated. Twenty-four of the 100 hips were revised for aseptic loosening. Radiographic criteria were highly effective in predicting those cases that would require revision. Radiolucent lines occurred at a higher frequency in all Gruen zones of the revision group radiographs, and the differences were statistically significant in zones 1, 7, 8, and 14 (P < .001). Proximal radiolucent lines were present in all of the revision cases and in 45% of the nonrevision cases. Radiolucent lines occurring on the lateral view alone were a statistically significant predictor of future need for revision (P < .001). Presence of radiolucent lines around the porous coating was 100% sensitive and 55% specific in predicting future revision. Absence of radiolucent lines around the porous coating was 100% sensitive and 45% specific in predicting no need for revision in the first 8 years.