On loosening and revision in total hip arthroplasty

Abstract

Periprosthetic bone resorbtion (osteolysis) and aseptic loosening constitute the major long-term complication in total hip arthroplasty. It has been suggested that osteolysis is the result of the action of osteoclasts generated by prosthetic wear-debris-triggered macrophages. In an electron-microscopic study of biopsies retrieved from periprosthetic osteolytic lesions, however, we were unable to show more osteoclasts than what would be expected in normal adult bone. Macrophages and other phagocytic cells seem to resorb the inflammatory interface membrane, but not bone.

Septic loosening is the second most important total hip arthroplasty complication, and it has been hypothesized that the current incidence of low-virulent infections is underestimated because of problems of differential diagnosis. This may entail inadequate handling of the patient in revision surgery. In a histological study of ultrasound-guided biopsies of the pseudocapsule, we found that infection eluded histological detection, and that it was difficult to differentiate between inflammation caused by a low-virulent microorganism and inflammation caused by particulate wear debris.

Revision total hip arthroplasty often yields poorer clinical results and durability than primary surgery does. Technically demanding surgery in deranged bone is the main reason. In an epidemiological study of revisions using the Swedish National Hip Arthroplasty Register, with re-revision as end-point, we found that early revisions in particular had an increased risk of re-revision; but that the risk was lower for first revisions performed in university/regional hospitals than for those performed in the central and rural hospitals. For early revisions, both cup and stem, cemented fixation afforded poorer results than uncemented. Femoral impaction bone grafting did not yield better results than conventional recementation.

In a radiographic study of re-cemented revisions using modem cementing techniques without bone grafting, we found early radiographic loosening of both components. The reasons were probably inferior surgical technique and that cancellous bone loss offered poor prerequisites for cemented fixation. Revision surgery in large acetabular defects is particularly difficult, and the results can be discouraging. We studied clinical and radiographic results of a stemmed acetabular reconstruction device inserted for severe acetabular bone loss, at five orthopaedic centers in Sweden. Early radiographic loosening necessitated subsequent revision in a majority of the patients.

In addition, we studied expectations and outcomes in revision patients. Post-operative dissatisfaction may originate from excessive pre-operative expectations regarding the functional result. Expectations often originated from sources other than the surgeon responsible, and patients were generally not aware of the fundamental differences between primary surgery and revision.

Osteolysis and aseptic loosening are a multifactorial syndrome in which the roles of osteoclast-mediated bone resorbtion and infection are unclear. Poor prognosis after early revisions highlights the importance of high-quality index procedures. Uncemented fixation can be considered for early revisions, even though, as with many newer methods such as bone grafting, our present lack of information on long-term outcome prompts extra care. Pre-operative information to revision patients should be improved.