Total hip arthroplasty (THA) is among the most successful of elective surgical interventions; it has enabled drastic improvements in function and quality of life for millions of patients. THA is indicated for advanced hip disease resulting from osteoarthritis, avascular necrosis, slipped capital femoral epiphysis, congenital hip dysplasia, and rheumatoid arthritis.
Conventional hip replacement has the potential risks of postoperative dislocation, bone loss extending to the metaphyseal and diaphyseal femur, and gross migration of the femoral component. Stress shielding of the proximal femur can occur with rigid intramedullary fixation of stiff metallic THA stems. Conventional joint replacement does not provide a lasting solution to suit the dynamic lifestyle of a young active population. Revision surgery with conventional joint replacement is inevitable in the young population and is associated with high morbidity.
Hip resurfacing is an alternative form of hip arthroplasty that conserves proximal femoral bone. [1, 2, 3, 4, 5, 6] Bone conservation is attractive to young and active patients. Other advantages of hip resurfacing over conventional THA include added hip stability from the larger diameter of the prosthetic femoral component and more accurate restoration of leg length, femoral offset, and femoral anteversion.
Thus, proponents of hip resurfacing claim that its advantages include a greater range of motion of the hip joint after surgery, a lower rate of dislocation, less prosthetic bearing wear, easier revision surgery, and less likelihood of stress-shielding of the proximal femur. Hip resurfacing may therefore be particularly suitable for young and active patients.
The short-term failure of hip resurfacing from the 1960s through the 1980s was due to aseptic loosening, which was caused by poor internal fixation, generation of large volumes of biologically active particulate debris (leading to bone loss), technical errors in device implantation, and problems with design and manufacturing.
Volumetric polyethylene wear in initial metal-on-polyethylene resurfacing was often 10 times higher than for standard hip replacements. Blood and urine metal ion levels, metallosis, capsular lymphocytic aggregation, and hypersensitivity are still major concerns with metal-on-metal articulation.
The major mode of failure of first-generation hip resurfacing procedures was femoral neck fracture, caused by improper surgical technique, or osteolysis of the femoral neck from wear particles produced by the articulation. Femoral bone stock was well preserved in first-generation hip resurfacing implants, but significant acetabular bone was removed to accommodate the acetabular component and the cement mantle.
Only comparatively recently has resurfacing arthroplasty regained acceptance, with successful short- and medium-term results being achieved through design modifications and improvements in surgical technique (long-term safety and efficacy and revision surgery outcomes require further evaluation).
Hip resurfacing is being performed more frequently in the United Kingdom and Australia than in the United States. It accounts for 7.5% of all hip replacements in Australia and has a 2.2% revision rate per year, with femoral fracture being the most common reason for revision. In a comparison study however, United Kingdom researchers found no evidence that hip resurfacing provides better hip function or higher activity levels than total hip replacement in the first year after surgery. The clinical efficacy and cost-effectiveness of hip resurfacing in the long term have yet to be determined. 
A key attraction of hip resurfacing is that it addresses both the increasing expectations of the active older patients and the work and recreational activities of younger patients. It must be kept in mind, however, that the procedure should be performed only by surgeons trained in resurfacing. Ceramic bearings are currently precluded, because the thickness of ceramic required would not be bone-sparing on the acetabular side. In the future, ceramic-on-ceramic resurfacing may offer better wear properties and address the issues of metallosis and metal hypersensitivity.
Suitable candidates for resurfacing are patients younger than 60 years with good bone stock and a diagnosis of osteoarthritis.  Resurfacing can be done in patients with proximal femoral deformity when a standard stemmed prosthesis is difficult to place. It is a good alternative for patients with neuromuscular disorders because it is associated with less risk of hip dislocation.
Resurfacing is also an option in cases involving retained hardware that would be difficult to remove for a standard stem placement. It can be a choice in patients with conditions that pose a high risk for failure after standard THA, such as sickle-cell disease and alcoholism.
Advanced hip arthritis usually manifests with radiographic evidence of joint-space narrowing in the hip, osteophytes, periarticular cysts, and subchondral sclerosis. Typically, the spherical shape of the femoral head and the acetabular cavity is preserved or deformed slightly as a consequence of mechanical wear and secondary skeletal remodeling. Ideally, for hip resurfacing to be considered, there should be no significant anatomic deformity of the femoral head and acetabulum, particularly the former.
If the femoral head can be reamed so that the spherical contour can be reconstructed and if there is sufficient viable subchondral bone, then hip resurfacing is a reasonable option in the appropriate patient population. Mild dysplasia of the acetabulum can be reamed away, but a deformed femoral head may require excessive reaming and bone removal to seat a resurfacing femoral component.
Cystic degeneration of the femoral head and collapse after severe long-standing avascular necrosis can lead to anatomic deformities that can preclude hip resurfacing. In such cases, THA components should be available during surgery as a backup, and the patient should know that a THA may be performed instead.
Absolute contraindications for hip resurfacing include advanced age and postmenopausal status with osteoporosis, impaired renal function, and known metal hypersensitivity. Other contraindications include deficiency of the femoral head or neck bone stock, femoral neck or head cysts, severe hip dysplasia, and a small or bone-deficient acetabulum.
A hip with deficient femoral head bone stock after fracture or other conditions (eg, rapidly progressive osteoarthritis, or disappearing bone disease) cannot be resurfaced. Hip resurfacing cannot correct large inequalities in limb length or change horizontal femoral offset; accordingly, arthritic hips that are at least 1 cm shorter than the contralateral limb or that have low horizontal femoral offset may be better managed with a standard THA or extended offset stemmed implants.
Relative contraindications include inflammatory arthropathy, abnormal proximal femoral geometry, and extensive osteonecrosis. For successful implantation of the femoral component, the spherical shape of the femoral head must be preserved. Deformation of the femoral head resulting from severe trauma, hip dysplasia, extensive cystic degeneration, previous surgery or hardware, extensive collapse from avascular necrosis, and related conditions precludes hip resurfacing, making THA a more reasonable option.
Advantages of hip resurfacing
The advantages of hip resurfacing over conventional THA are that it restores physiologic biomechanics, it preserves bone stock, and it allows the patient to return to a more active lifestyle. Favorable survivorship with a return to high-demand activities after hip resurfacing may be advantageous in younger patients. However, patients should limit their involvement in sports to realistic levels that the implant will be able to tolerate. 
Although outcomes data on functional results from hip resurfacing versus replacement are mixed, proponent surgeons believe that patients who undergo hip resurfacing have superior kinematics and higher activity scores than those who undergo standard THA. They walk faster and have better hip abduction and postoperative range of motion in gait analysis. The physiologic pattern of femoral head loading after resurfacing facilitates better stress transfer by producing compressive forces rather than hoop stress. This may improve bone mineral density, prevent stress shielding, preserve the proximal femoral bone stock, and avoid periprosthetic bone loss. 
The use of metal-on-metal bearings may lead to marked reductions in periprosthetic bone loss, which was often extensive with first-generation polyethylene acetabular resurfacing components. Advances in design have led to the introduction of polar bearing implants, which have reduced surface asperity and provided superior fluid film lubrication for the bearing surfaces. The thin acetabular shells that have been developed allow the removal of less acetabular bone.
Hip resurfacing, by accommodating a large femoral head, is associated with a lower dislocation rate than conventional THA is. Finally, hip resurfacing may yield better proprioception than conventional THA because of preservation of proximal femoral bone.
Modern hip resurfacing devices consist of a cemented femoral component and a hydroxyapatite-coated acetabular component. Contemporary components have a metal acetabular bearing that requires the removal of little acetabular bone stock. Resurfacing preserves bone on the femoral side and avoids use of the intramedullary devices that are implanted in standard hip replacements.
With femoral failure, the acetabular component can be left in place and mated to a standard femoral stem with a large-diameter head. Revision of hip resurfacing is thus easier than revision of a failed THA. Other intraoperative advantages of hip resurfacing may include less blood loss and a reduced need for blood transfusion because the intramedullary femoral canal is not broached. The bone mineral density of the proximal femur is maintained or increased with resurfacing, whereas a significant decrease from stress-shielding can occur with THA.
Hemiresurfacing of the hip refers to resurfacing the femoral head alone while preserving the native acetabulum. It yields unpredictable pain relief, with 80% and 60% survivorship at 5 and 10 years, respectively. Acetabular cartilage wear or persistent pain is the primary cause for failure. Although these outcomes are inferior to those of standard THA performed by a skilled surgeon, they are comparable, and perhaps superior, to those of other femoral head-sparing procedures (eg, free vascularized fibular graft and proximal femoral osteotomy) and result in less morbidity.
Limitations of hip resurfacing
For resurfacing to be truly conservative, the surgeon should not remove any more bone from the acetabulum than would be required for a THA. Only relatively recently has this goal been addressed by the development of thinner acetabular shells that are designed for implantation without cement. For patients with low femoral head-neck ratio (<1.2), such as those with osteoarthritis secondary to Legg-Calvé-Perthes disease or slipped capital femoral epiphysis, removal of more acetabular bone is often necessary during hip resurfacing.
The lack of modularity of resurfacing devices reduces the ability to adjust leg length. The biomechanical results of hip resurfacing were dependent on the preoperative anatomy of the femur. Limb lengthening cannot be achieved, and horizontal offset is essentially unchanged by resurfacing. In contrast, both offset and limb lengthening can be increased reliably during THA. A biomechanical study recommended that leg length discrepancy be no greater than 1 cm and that the hip should have high horizontal femoral offset.
For hip resurfacing, body mass index (BMI) should be less than 35 kg/m2, though some studies have shown no prejudicial effect of BMI on the outcome after hip resurfacing. No difference in rehabilitation has been shown between hip resurfacing and THA.
Some patients may experience a noise after resurfacing; one study showed a 23% prevalence of clicking and a 4% prevalence of squeaking after metal-on-metal resurfacing. It is unclear if these joint noises are related to impingement, subluxation, or microseparation of the components.
Metal-on-metal resurfacing should be used with caution in women who want to become pregnant after hip arthroplasty and patients with definite metal allergy or mild compromise of renal function, even though a study of three women with metal-on-metal replacements and elevated ion levels showed no increase in ion levels in the umbilical cord blood and no adverse consequences for the infants.
Nevertheless, the elevation of serum metal ions remains a potential long-term concern for all patients, and until future studies clarify the significance of this phenomenon, any metal-on-metal bearing in the body should be implanted cautiously and only after proper patient counseling. Also, pseudotumors have been shown to occur in the periprosthetic joint space with metal-on-metal hip resurfacing, raising the concern that adverse host reactions to metal ion debris may be more widespread and a greater concern than earlier studies have identified. Finally, no study has shown hip resurfacing to be superior to modern total hip replacements with regard to durability and survivorship.
Some data suggest that modern hip resurfacing has a survival rate of 98% at 5-year follow-up. The most common reason for revision was fracture of the femoral neck, mostly occurring within the first 6 months. Men older than 65 years are at higher risk for revision than men younger than 65 years. Among women, those aged 55-64 years are at highest risk for revision. Patients with osteoarthritis have a lower incidence of early revisions compared to those with rheumatoid arthritis, osteonecrosis, and developmental dysplasia.
An article on short-term survivorship of resurfacing arthroplasty suggests the results are equivalent to traditional THA. It should be emphasized that the study is from the Finnish Registry and is limited to 3-year follow-up. A short-term study, based upon registry data, could only suggest early patterns of behavior rather than make definitive statements about expected outcomes. 
In a retrospective cohort study aimed at comparing 10-year cumulative mortality figures for patients undergoing metal-on-metal hip resurfacing and those undergoing cemented or uncemented THA for osteoarthritis in England from April 1999 to March 2012, Kendal et al determined that metal-on-metal hip resurfacing was associated with reduced long-term mortality as compared with either variety of THA.  The difference remained after confounding factors were adjusted for.
The range of movement is improved in all patients after hip resurfacing, with reported mean flexion of 92° preoperatively and 110° postoperatively.
At 5-year radiostereometric follow-up, resurfaced hips have shown no changes in mean acetabular component translation and rotation and femoral component translation, indicating that the implants are stable (at least over a 5-year period).
Some narrowing of the femoral neck can occur after resurfacing, and this phenomenon is not associated with any adverse clinical or radiologic outcome up to a maximum of 6 years after the initial operation. There is a significant association between narrowing in females and a valgus femoral neck-shaft angle. There is no significant association between the range of movement, position or size of the component or radiologic lucent lines, and narrowing of the neck.
Daniel et al reported a 99.8% survival rate with a follow-up of 1-8 years for 446 hip resurfacings performed in 384 osteoarthritis patients using the Birmingham hip resurfacing (BHR) approach. 
Treacy et al reported a series of 144 consecutive metal-on-metal resurfacings of hips (BHR) with a survival rate of 98% at a minimum of 5 years' follow-up. 
Amstutz et al reported on a series of 400 hip resurfacings performed in 355 patients using the Conserve Plus prosthesis and documented a 94% component survival rate at 4 years.  They also presented the risk index for failure.
Gravius et al described a series of 59 resurfacings with the Durom prosthesis in 52 patients and reported two revisions over a period of 25±10 months, which were attributable to femoral neck fracture and heterotopic ossification (HO). 
Nishii et al evaluated the 5-year results of metal-on-metal resurfacing arthroplasty in 45 patients with developmental dysplasia or dislocation of the hip (70%) and noted a 96% 5-year survival rate.  They reported two revisions, attributable to femoral neck fracture and septic loosening; they also had one femoral component loosening.
Steffen et al, in a study evaluating 5-year clinical follow-up and 7-year survival in a total of 610 BHR procedures done in 532 patients, revised 23 hips (3.8%), mainly because of femoral neck fracture (12), aseptic loosening (four) and possible metal debris (three).  They reported a 93% rate of excellent or good outcome (on the Harris hip score) at 5 years and 95% survival at 7 years. There were no patients with definite radiologic evidence of loosening or of narrowing of the femoral neck exceeding 10% of its width.
Hing et al, in a study evaluating the results of 230 BHR procedures in 212 patients at a mean follow-up of 5 years, reported a survival rate of 97.8% and noted that two revisions were done because of acetabular component loosing and suspected avascular femoral head necrosis. 
Radiologic review at 5-year follow-up showed one patient with progressive lucent line around the acetabular component and six with progressive lucent lines around the femoral component.  About 8% of femoral components migrated into varus, and those with lucent lines migrated more than the rest. The investigators reported a statistically significant association between the presence of radiolucent lines to superolateral notching of the femoral neck and reactive sclerosis at the tip of the femoral peg.
Amstutz et al evaluated metal-on-metal hybrid hip resurfacing for Crowe type I and II developmental dysplasia in 51 middle-aged patients (59 hips), concluding that medium-term results were disappointing with respect to femoral component durability. However, the fixation of the porous-coated acetabular components without adjuvant fixation was excellent despite incomplete lateral acetabular coverage of the socket. More rigorous patient selection and meticulous bone preparation are essential to minimize femoral neck fractures and loosening after this procedure.
Ball et al presented early results of conversion of 21 failed femoral components in 20 hip resurfacing patients and found no significant differences between the conversion arthroplasty group and the conventional arthroplasty group with regard to operating time, blood loss, or complication rates.  At 2-year follow-up, there were no significant differences between the two groups with regard to the mean Harris hip score; the University of California at Los Angeles pain, walking, and activity score; or the SF-12 score.
On radiographic assessment, the quality of component fixation and the alignment of the reconstruction were equivalent between the two groups.  There had been no instances of aseptic loosening of the femoral or the acetabular component in either group, and there had been no dislocations after conversion of a resurfacing arthroplasty.
Pascual-Garrido et al retrospectively reviewed the outcomes of 202 patients (206 hips) treated with BHR procedures (mean patient age, 51±8 years; mean follow-up, 4±1.6 years).  Outcomes were evaluated by recording the Harris Hip Score at 6 months, at 12 months, and yearly thereafter. The BHR procedure yielded significant postoperative improvement was significant, from a score of 62.9±10.6 before surgery to a score of 98.6±6.7 postoperatively. Nine patients (4%) had complications, and five 5 hips (2.4%) underwent revision surgery. Overall, men had better survival rates and outcomes than women (especially those with modest bone size)
Brooks reported a large (N=1333; mean age, 53.1 years) single-surgeon experience with the BHR system over a mean follow-up period of 4.3 years (range, 2-5.7 years). Of the 1333 patients 70% were male and 91% had osteoarthritis. Complications rates were low, and no destructive pseudotumors occurred. At up to 5.7 years, overall survivorship was 99.2%, and in males younger than 50 years, aseptic survivorship was 100%.