SUCCESS AND FAILURE IN HIP RESURFACING – A PERSONAL PERSPECTIVE OVER 10 YEARS?

Abstract

In a personal series of over 560 hip resurfacing operations performed since 1995, 35 hips have required revision operations.

Only one has been for infection, a single stage revision was performed at 33 months following primary implantation, and the revised hip is still functioning well over 7 years later.

There have been five femoral neck fractures, all became symptomatic within 100 days of surgery. The first four were revised to cemented stems with large modular heads, the fifth was revised to an uncemented stem with large head. The first of these patients developed aseptic loosening at the bone/cement interface, and was revised at 18 months to a cemented Exeter stem with ceramic head and uncemented socket with plastic liner. At 8 years, the hip continues to function well clinically with no radiological signs of loosening.

The remaining four patients have all had satisfactory results at a mean follow up of 69.3 months (range 52 – 96 months)

Ten hips have required revision operations for femoral head loosening or collapse. The mean time after initial surgery was 54.6 months (range 23 – 107 months). In every case, the original socket was preserved and a stem with large modular head was used. The first two were using cemented stem and are still functioning well at 52 and 69 months. The next was with a Trifit uncemented stem, and remains satisfactory at 36 months. The remaining five have all been revised with a Plus Orthopaedics Zweymuller stem. The mean follow up is 13.8 months (range 5 – 33 months) and all are currently satisfactory.

There have been 18 sockets which required revision operations for loosening. The three earliest of these underwent revision surgery at 16, 17 and 27 months, and probably represent sockets which were inadequately seated at the time of implantation and failed to osseointegrate. This is evident on the initial post op radiographs in two patients. The third patient had unexplained discomfort in the hip from the outset, and eventually the hip was explored; at operation the socket was found to be loose after the hip had been opened and the head excised. Only one other patient required the head to be removed; the bone loss in the socket around the loose acetabular socket required revision with a larger socket than would allow the head to be preserved.

There was a cluster of five sockets which were revised between 50 and 57 months, the remaining five sockets were revised between 78 and 98 months. The timing of these failures suggests different modes of failure.

One patient had a revision with a dysplasia socket but no graft and 12 months later had clinical and radiological signs of loosening, but the symptoms settled before surgery was performed.

One patient developed a post operative infection and the implants were removed.

The remaining patients have a mean follow up of 36.5 months (range 8 – 61 months), with no signs of failure of the revised socket or the original femoral component.