We performed a study to determine the prevalence, severity, and natural history of pain in the thigh in patients who had a total hip replacement with a porous-coated anatomic prosthesis and to determine if there was an association between the radiographic findings and the pain in the thigh. We previously reported the results in these patients two years after the operation. The present study included ninety-four patients (101 total hip replacements) who had been followed for a minimum of five years--six of the original group of patients had died from unrelated causes, two had had a revision, and one had been lost to follow-up. The patients were assessed prospectively with use of the Harris hip score and a visual-analog scale for pain in the thigh. Two of us, who were blinded to the clinical result, reviewed the radiographs retrospectively for the morphology of the proximal aspect of the femur; distal fit; metaphyseal fill; alignment of the stem; shedding of beads; distal cortical hypertrophy; pedestal formation; cancellous condensations at the distal end of the porous ingrowth surface; and the presence, extent, and location of radiolucent lines around the femoral component. At five years, pain in the thigh was found in association with twenty-seven hips (27 per cent). The pain was new in fourteen thighs (14 per cent), unchanged (from that at the two-year follow-up examination) in eight (8 per cent), increased in four (4 per cent), and decreased in one (1 per cent). Pain in the thigh was most strongly associated with a poor Harris hip score and the presence of loose beads.