United KingdomBirmingham hip resurfacing: The prevalence of failure
Robert Jones and Agnes Hunt Orthopaedic Hospital, Gobowen, Oswestry SY10 7AG, UK
Carrothers AD1, Gilbert RE, Jaiswal A, Richardson JB.
J Bone Joint Surg Br. 2010 Oct
Despite the increasing interest and subsequent published literature on hip resurfacing arthroplasty, little is known about the prevalence of its complications and in particular the less common modes of failure. The aim of this study was to identify the prevalence of failure of hip resurfacing arthroplasty and to analyse the reasons for it.
From a multi-surgeon series (141 surgeons) of 5000 Birmingham hip resurfacings we have analysed the modes, prevalence, gender differences and times to failure of any hip requiring revision. To date 182 hips have been revised (3.6%).
- The most common cause for revision was a fracture of the neck of the femur (54 hips, prevalence 1.1%),
- followed by loosening of the acetabular component (32 hips, 0.6%),
- collapse of the femoral head/avascular necrosis (30 hips, 0.6%),
- loosening of the femoral component (19 hips, 0.4%),
- infection (17 hips, 0.3%), pain with aseptic lymphocytic vascular and associated lesions (ALVAL)/metallosis (15 hips, 0.3%),
- loosening of both components (five hips, 0.1%),
- dislocation (five hips, 0.1%) and
- malposition of the acetabular component (three hips, 0.1%).
- In two cases the cause of failure was unknown.
Comparing men with women, we found the prevalence of revision to be significantly higher in women (women = 5.7%; men = 2.6%, p < 0.001). When analysing the individual modes of failure, women had significantly more revisions for loosening of the acetabular component, dislocation, infection and pain/ALVAL/metallosis.
The mean time to failure was 2.9 years (0.003 to 11.0) for all causes, with revision for fracture of the neck of the femur occurring earlier than other causes. There was a significantly shorter time to failure in men compared with women.
Between July 1997 and November 2002, the Oswestry Outcome Centre independently and prospectively collected data on 5000 Birmingham Hip Resurfacings performed by 141 surgeons, at 84 hospitals.
Of their 107 retrieved specimens 59% had femoral fractures within the bone inside the femoral component, with osteonecrosis being the most frequent cause of fracture-related failures. The cement mantle, the depth of penetration of cement within the resurfacing head, bone density and the clearance between the reamed head and the femoral component have all been associated with implant survival. With cementing there is the potential for thermal necrosis of the cancellous bone in the reamed head. Too much cement can lead to thermal necrosis, whereas an insufficient amount may cause mechanical failure and particle-induced osteolysis.
The long-term success of uncemented acetabular components in THR is well established. Loosening of the acetabular components of hip resurfacing has not been reported as a particular problem. Amstutz et al published a series of 400 Conserve plus hip resurfacings. From this group only one hip (0.3%) was revised for acetabular loosening, but acetabular radiolucency was observed in 32% of cases, 26% in a single zone and 6% in two zones. In a multicentre review of 200 hip resurfacings at a mean of 19.5 months (3 to 47), Kim et al found that 5% (10 of 200) were revised for loosening of the acetabular component. The surgical learning curve was felt to be the explanation and the authors anticipated that further surgical experience would lead to a decrease in this high early failure rate.
There is a surgical learning curve associated with hip resurfacing and it is generally accepted to be a more complex procedure than THR. Both De Hann et al and Kim et al noted higher rates of failure due to malpositioning of the acetabular component, attributing this to the technical challenges of performing resurfacing. It has been stated that most failures occur during the learning curve, but McBryde et al found that revision was not associated with surgeon, surgeon experience or surgical approach. Furthermore, there is still no consensus on the number of resurfacings needed to overcome the learning curve. De Smet, Campbell and Gill summarising the Ghent advanced hip resurfacing course 2009 stated that an orthopaedic surgeon should have a minimum experience of 200 conventional THRs before starting hip resurfacing. Opinion varied on the number of resurfacings needed to overcome the learning curve.
Despite the increasing interest and subsequent published literature on hip resurfacing arthroplasty, little is known about the prevalence of its complications. This is in part due to their relative infrequency and published studies containing small numbers of patients. Hip resurfacing has its own unique set of complications, including fracture of the neck of the femur. It is important to understand the prevalence of complications as well as the risk factors in order to appropriately select patients and adequately inform them prior to hip resurfacing arthroplasty.
Robert Jones and Agnes Hunt Orthopaedic Hospital, Gobowen, Oswestry SY10 7AG, UK
Carrothers AD1, Gilbert RE, Jaiswal A, Richardson JB.
J Bone Joint Surg Br. 2010 Oct
Despite the increasing interest and subsequent published literature on hip resurfacing arthroplasty, little is known about the prevalence of its complications and in particular the less common modes of failure. The aim of this study was to identify the prevalence of failure of hip resurfacing arthroplasty and to analyse the reasons for it.
From a multi-surgeon series (141 surgeons) of 5000 Birmingham hip resurfacings we have analysed the modes, prevalence, gender differences and times to failure of any hip requiring revision. To date 182 hips have been revised (3.6%).
- The most common cause for revision was a fracture of the neck of the femur (54 hips, prevalence 1.1%),
- followed by loosening of the acetabular component (32 hips, 0.6%),
- collapse of the femoral head/avascular necrosis (30 hips, 0.6%),
- loosening of the femoral component (19 hips, 0.4%),
- infection (17 hips, 0.3%), pain with aseptic lymphocytic vascular and associated lesions (ALVAL)/metallosis (15 hips, 0.3%),
- loosening of both components (five hips, 0.1%),
- dislocation (five hips, 0.1%) and
- malposition of the acetabular component (three hips, 0.1%).
- In two cases the cause of failure was unknown.
Comparing men with women, we found the prevalence of revision to be significantly higher in women (women = 5.7%; men = 2.6%, p < 0.001). When analysing the individual modes of failure, women had significantly more revisions for loosening of the acetabular component, dislocation, infection and pain/ALVAL/metallosis.
The mean time to failure was 2.9 years (0.003 to 11.0) for all causes, with revision for fracture of the neck of the femur occurring earlier than other causes. There was a significantly shorter time to failure in men compared with women.
Between July 1997 and November 2002, the Oswestry Outcome Centre independently and prospectively collected data on 5000 Birmingham Hip Resurfacings performed by 141 surgeons, at 84 hospitals.
Of their 107 retrieved specimens 59% had femoral fractures within the bone inside the femoral component, with osteonecrosis being the most frequent cause of fracture-related failures. The cement mantle, the depth of penetration of cement within the resurfacing head, bone density and the clearance between the reamed head and the femoral component have all been associated with implant survival. With cementing there is the potential for thermal necrosis of the cancellous bone in the reamed head. Too much cement can lead to thermal necrosis, whereas an insufficient amount may cause mechanical failure and particle-induced osteolysis.
The long-term success of uncemented acetabular components in THR is well established. Loosening of the acetabular components of hip resurfacing has not been reported as a particular problem. Amstutz et al published a series of 400 Conserve plus hip resurfacings. From this group only one hip (0.3%) was revised for acetabular loosening, but acetabular radiolucency was observed in 32% of cases, 26% in a single zone and 6% in two zones. In a multicentre review of 200 hip resurfacings at a mean of 19.5 months (3 to 47), Kim et al found that 5% (10 of 200) were revised for loosening of the acetabular component. The surgical learning curve was felt to be the explanation and the authors anticipated that further surgical experience would lead to a decrease in this high early failure rate.
There is a surgical learning curve associated with hip resurfacing and it is generally accepted to be a more complex procedure than THR. Both De Hann et al and Kim et al noted higher rates of failure due to malpositioning of the acetabular component, attributing this to the technical challenges of performing resurfacing. It has been stated that most failures occur during the learning curve, but McBryde et al found that revision was not associated with surgeon, surgeon experience or surgical approach. Furthermore, there is still no consensus on the number of resurfacings needed to overcome the learning curve. De Smet, Campbell and Gill summarising the Ghent advanced hip resurfacing course 2009 stated that an orthopaedic surgeon should have a minimum experience of 200 conventional THRs before starting hip resurfacing. Opinion varied on the number of resurfacings needed to overcome the learning curve.
Despite the increasing interest and subsequent published literature on hip resurfacing arthroplasty, little is known about the prevalence of its complications. This is in part due to their relative infrequency and published studies containing small numbers of patients. Hip resurfacing has its own unique set of complications, including fracture of the neck of the femur. It is important to understand the prevalence of complications as well as the risk factors in order to appropriately select patients and adequately inform them prior to hip resurfacing arthroplasty.
Last edited by a moderator:
