True resurfacing of the hip involves resurfacing of the worn joint with living tissue. Currently, autogenous cartilage transplantation to the hip holds great promise for the future, but is not reliable enough at this point to be applied clinically except in the most compelling of circumstances. Prosthetic surface replacement of the hip has been performed since the 1960's, with resurgence in interest in the 1990's with the advent of metal-metal bearings. Recent clinical results in 2006 and 2007 have raised concerns about the outcomes of this procedure. Issues concerning hip resurfacing include the following topics:

1. Incidence of hip dislocation following hip resurfacing
2. Femoral neck fracture following hip resurfacing
3. Technical inability to correct leg length and offset
4. Acetabular side failures
5. Difficulty of revising prosthetically resurfaced hips
6. Decline in use of metal-metal hip resurfacing among surgeons that had previously employed it more often
7. Additional risks for patients with osteonecrosis
8. Additional risks for patients with hip dysplasia
9. Additional risks for patients with femoroacetabular impingement
10. Additional risks for women
11. The problem of the disappearing femoral neck following hip resurfacing
12. Possible causes of persistent groin pain in some patients with hip resurfacing
13. Problems with uncemented femoral components in hip resurfacing
14. Problems with cemented femoral components in hip resurfacing
15. Cases of run-away wear and extreme serum metal levels
16. Anticipated problems with acetabular component designs currently under development

Each of these issues can be analyzed in turn. The first 8 issues are currently reviewed on this website.

Acknowledgements. Dr. Murphy wishes to acknowledge the experience and perspective of the following individuals: Benjamin E. Bierbaum, MD, Boston, Massachusetts; Reinhold Ganz, MD, Zurich, Switzerland; Donald Howie, MD, Adelaide, Australia; Joshua J. Jacobs, MD, PhD, Chicago, Illinois; Merrill Ritter, MD, Mooresville, Indiana.

1. Incidence of Hip Dislocation following Hip Resurfacing.

It has generally been assumed that the large femoral head used for hip resurfacing would naturally result in a lower dislocation rate than total hip arthroplasty. This assumption is based on the fact that a larger femoral head requires more displacement than a smaller femoral head to dislocate. In fact, the incidence of hip dislocation following hip resurfacing is not lower than that for many different methods of performing total hip arthroplasty, and in some cases, is actually higher. For example, in a prospective study of hip resurfacing reported at the American Academy of Orthopedic Surgeons in 2007, Mont et al report a dislocation rate following hip resurfacing of 1.8%.

Poster NO. P034
Modern Generation Metal-on-Metal Total Hip Resurfacings: Results of a prospective FDA-IDE Study. Mont MA, Amstutz HC, Boyd HS, Schmalzried TP, Vail TP, Sparling EA, Kennedy WR, Seyler TM, Goldberg V.

In addition, looking specifically at hip resurfacing for dysplasia, the hip dislocation rate has been reported to be 5% (3/59) with an incidence of revision for hip instability alone of over 3% (2/59).

Amstutz HC, Antoniades JT, Le Duff MJ. Results of metal-on-metal hybrid hip resurfacing for Crowe type-I and II developmental dysplasia. J Bone Joint Surg AM. 2007 Feb; 89(2):339-46.

By contrast, the incidence of hip dislocation for total hip arthroplasty (including many patients with hip dysplasia) has been reported to be 0.6% using an anterior exposure, 0% (0/194) using predominantly a transgluteal exposure with preservation of the posterior capsule and short rotators, and 0.54% (1/185) using a superior exposure with preservation of the posterior capsule and short rotators. These rates for the transgluteal and superior exposures are in patients who have no motion restrictions at all after surgery. Even using the posterior exposure with capsular and short-rotator repair the incidence of dislocation has been reported to be between 0 (0/395) and 0.8% (1/124) in one study and 0.85% (8/945) in another. All of these methods of performing total hip arthroplasty have a dislocation rate that is lower than that reported for hip resurfacing above. Thus, while it is true that many studies of total hip arthroplasty performed using a posterior exposure, without capsular repair, have dislocation rates that are higher than 1.8%. Most modern methods of performing total hip arthroplasty with preservation or repair of the posterior structures of very low dislocation rates. Each of these references is cited in order below with additional references of interest.

Matta JM. Anterior Approach THA, Proceedings of the Thirty-Fifth Open Meeting of the Hip Society, San Diego, California, Saturday, February 17, 2007

Murphy SB, Ecker TM, Tannast M. Two to 9 year clinical results of alumina ceramic-ceramic total hip arthroplasty. Clinical Orthopedics and Related Research. 453,pp97-102.2006.

Murphy SB, Ecker TM, Tannast M. THA performed using conventional and navigated tissue-preserving techniques. Clinical Orthopedics and Related Research. 453,pp.160-167. 2006.

Pellicci PM, Bostrom M, Poss R. Posterior Approach to Total Hip Replacement Using Enhanced Posterior Soft Tissue Repair. Clinical Orthopaedics and Related Research. 355, pp. 224-228, 1998.

Weeden SH, Paprosky WG, Bowlin JW. The early dislocation rate in primary total hip arthroplasty following the posterior approach with posterior soft-tissue repair. J Arthroplasty. 2003 Sep; 18(6):709-13.

Khatod M, Barber T, Paxton E, Namba R, Fithian D. An analysis of the risk of hip dislocation with a contemporary total joint registry. Clin Orthop Relat Res 447:19-23, 2006.

Why isn't the incidence of hip dislocation following hip resurfacing lower than that for total hip arthroplasty? The answer may lie with a combination of four contributing factors.

First, the ability to correct for abnormalities of leg length and offset are greatly limited with hip resurfacing. From a practical point of view then, this means that when the surgeon puts either trial or real implants in during a hip resurfacing, if the trial reduction demonstrates that the hip is too loose and unstable, there are very few options available to the surgeon to correct for the laxity, since putting the femoral prostheses further away from the femoral bone would require excessive support from bone cement and moving the cup out away from the prepared socket bone bed would render the cup unstable. Thus, excessive residual soft-tissue laxity is often a consequence of hip resurfacing.

Second, the head-neck ratio of hip resurfacing is poor. With respect to head-neck ratio, the typical ratio for a total hip arthroplasty with a 32mm head and a 10mm neck is 3.2 to 1. The head-neck ratio for hip arthroplasties with 36 and 40mm bearings is even greater. By contrast, the typical head-neck configuration for a hip resurfacing may be a head diameter of 52mm and a native neck of diameter of 44mm for example. This produces a head-neck ratio of only 1.18. A poor head-neck ratio leads to impingement between the neck and the pelvic bone or soft-tissue at the extremes of motion and impingement increases the likelihood of dislocation (as well as notching of the femoral neck, which can lead to fracture).

Third, acetabular components for hip resurfacing are generally not true hemispherical components. Most hip resurfacing acetabular components generally only form a 166 degree arc instead of a standard 180 degree arc. This reduced arc allows for the femoral head to dislocate more easily than it would otherwise.

Fourth, these three problems are compounded by the fact that most hip resurfacing is performed through a posterior exposure with release of the posterior capsule and short rotators. Whether repaired or not, with all other factors being equal, this surgical technique always has a higher incidence of dislocation than techniques that preserve the posterior capsule and short rotators.

The next logical question would be, "Why can't the posterior capsule and short rotators be repaired after a hip resurfacing just like after a total hip replacement using the same exposure?" The answer is that hip replacement performed through a posterior exposure does not require as much soft-tissue release as a hip resurfacing requires. Some surgeons go so far as to release the gluteus maximus tendon insertion during the exposure for hip resurfacing. This is a surgical technique that hasn't been routinely practiced for primary total hip arthroplasty in 30 years! This greater degree of soft-tissue release and mobilization makes strong repair of the posterior capsule and short rotators less likely. In addition, the other three factors contributing to hip instability following hip resurfacing still apply anyway.

Is it possible to argue the opposite, that hip resurfacing has a lower dislocation rate than total hip arthroplasty? To do so would require complete denial of modern surgical techniques with lower dislocation rates while simultaneously quoting historical studies of total hip arthroplasty with high dislocation rates from the past. Those studies are just that - historical.

Thus, while the incidence of hip dislocation is generally assumed to be less frequent with hip resurfacing, in fact, objective analysis clearly demonstrates that the opposite is the case.

2. Femoral neck fracture following hip resurfacing

The incidence of femoral neck fracture following hip resurfacing varies from 0 to more than 5% but is typically between 1 and 3% in the first 5 years. Fracture rates are generally twice as frequent in women than in men (AAOS 2007 Paper No. 123: Femoral Neck Fractures Following Metal-on-metal Total Hip Resurfacing. Michael A. Mont, MD, Ronal Emilio Delanois, MD, Johannes F Plate, BS, Thorsten M Seyler, MD.) For example, femoral neck fracture occurred in 1.9% (19 of 1016) of hips treated by resurfacing at an average follow up of less than 5 years (51 months) following surgery (AAOS 2007 Poster NO. P034, Modern Generation Metal-on-Metal Total Hip Resurfacings: Results of a prospective FDA-IDE Study. Mont MA, Amstutz HC, Boyd HS, Schmalzried TP, Vail TP, Sparling EA, Kennedy WR, Seyler TM, Goldberg V. Proceedings of the 2007 Annual Meeting of the American Academy of Orthopedics Surgeons). This means that the revision rate for femoral neck fracture alone at less than 5 years is nearly as high as that for all of the causes for revision combined at 9 years for many studies of young patients following total hip arthroplasty. In addition, the femoral neck fracture rate increases with time since the aggregate problem of femoral neck bone loss (a combination of stress shielding, osteolysis, neck notching from impingement, and femoral component loosening) following hip resurfacing is often progressive.

3. Technical inability to correct leg length and offset

The location of the femoral prosthesis of a hip resurfacing is limited by the location of the native femoral head, since the prosthesis is placed on the femoral head. Moving it to a substantially different position increases the risk of neck-notching (which leads to femoral neck fracture) and causes lack of bony support for the prosthesis (which leads to femoral component loosening). As such, the ability to properly correct for leg length and offset is greatly limited. This issue is discussed in an article by Silva et al ( Silva M, Lee KH, Heisel C, Dela Rosa MA, Schmalzried TP. The biomechanical results of total hip resurfacing arthroplasty. J Bone Joint Surg Am 86A(1):40-46, 2004 ). There are many negative consequences of this technical limitation. First, in patients with hip dysplasia, the femoral neck angle is sometimes too high (coxa valga), leading to very poor offset. Poor offset leads to a reduced lever-arm for muscles to pull against and this leads to higher muscle forces needed to compensate for the reduced lever-arm. The higher muscle forces lead to higher forces at the hip joint which produces higher wear, wear-associated debris, and wear debris-associated bone resorption which can lead to component loosening and neck fracture. The second major consequence is that the muscle may not be able to generate sufficient forces due to the poor lever arm which increases the likelihood of a persistent limp. The third major negative consequence is the restricted ability to balance the soft-tissue tension around the hip. This problem may be one of the factors that leads to a higher dislocation rate for hip resurfacing than for total hip arthroplasty.

4. Acetabular side failures

Acetabular components used for hip resurfacing need to be very thin to allow for a large femoral component. This means acetabular components are generally made as one piece. One-piece acetabular components are generally made of cobalt-chromium alloy because metal-metal bearing surfaces are made of cobalt-chromium alloy. Typically, total hip replacement acetabular components are modular with the metal shell being made of titanium alloy and the bearing being the surgeon’s bearing material of choice. These typical titanium shells have a screw hole in the middle for attachment of the insertion instrument and may or may not have other screw holes for screw fixation. When they are implanted, the surgeon can confirm that the cup is fully seated by looking through the hole or holes. If there is any concern about the integrity of the fixation between the cup and the bone, additional screws can be used to more firmly secure the cup before the bearing is inserted.

The thin, one-piece cobalt-chrome cups used for hip resurfacing are more difficult to insert than are the modular titanium cups used for most total hip arthroplasties. First, the surgeon cannot easily tell if the cup is fully inserted and second, if there is any concern about fixation between the cup and the bone, there is no opportunity to place screws through the cup itself. While there are a few designs that allow for screw fixation at the edge of the cup, this is a less satisfactory method of fixation than are screws placed into solid bone through the main part of the cup. The issues above apply to both hip resurfacing and large bearing metal-metal total hip arthroplasty where a one-piece, thin cup is used.

Hip resurfacing has a further technical problem though. That is, the cup diameter of a hip resurfacing is dictated by the size of the femoral head component and the size of the femoral head component is limited by the diameter of the patient’s femoral neck (because smaller head components will notch the femoral neck). This means that the acetabular cup component is forced to be larger than would otherwise be necessary even for larger diameter metal-metal total hip arthroplasty using the same cup component. The fact that hip resurfacing acetabular components, on average, are of larger diameter than total hip replacement components is documented in the same article by Silva et al. The larger diameter cup means that the pelvis has more difficulty grabbing and holding the cup than it would otherwise. This creates an additional problem. To illustrate this problem in practical terms, if you imagine that your hand is the bony acetabulum, and the cup is a ball, it would be much easier for you to grab and control a baseball with one hand than it would for you to grab and control a basketball. The same phenomenon applies to acetabular component fixation when the surgeon is forced to use a larger cup. The net result of one-piece cups combined with larger diameter cups is that failure of the cup to remain secure in the pelvis is more likely to occur with hip resurfacing. This may be why the Canadian hip registry is already showing a significantly higher incidence of failure of this type of cup design than well established, successful designs commonly used for total hip arthroplasty.

5. Difficulty of revising prosthetically resurfaced hips

The surgical disruption of the soft-tissues required to perform a hip resurfacing is tremendously greater than that currently required to perform a total hip arthroplasty. This is because the hip needs to be dislocated to perform the surgery and because the femoral head needs to be mobilized enough to replace the socket and then the head has to be mobilized even further to put a femoral head component on it. This is true no matter how “minimally invasive” hip resurfacing is presented to be or what surgical exposure is used. Most hips don’t need to be dislocated at all during total hip arthroplasty, let alone mobilized to that degree. This degree of soft-tissue dissection required for hip resurfacing can have a permanent affect causing the revision surgeon to reoperate through an excessively dissected soft-tissue envelope. Further, if a posterior exposure was used and the posterior capsule and short rotators didn’t heal properly, the likelihood of dislocation of a revision total hip replacement would be tremendously greater than that for a true primary total hip arthroplasty. Finally, if the reason for revision is for acetabular component failure, the bone stock loss on the acetabular side, combined with the fact the more bone, on average, was removed on the acetabular side during the first surgery, makes it likely that the patient would be far worse off than they would have been had the hip resurfacing never been done in the first place.

6. Decline in use of metal-metal hip resurfacing among surgeons who had previously employed it in a higher percentage of patients.

Many scientific papers over the past few years have focused on narrowing the indications for hip resurfacing in an effort to improve the clinical outcomes. The Hip Resurfacing Risk Index is a good example of this trend (Beale PE, Dorey FJ, LeDuff M, Gruen T, Amstutz HC. Risk factors affecting outcome of metal-on-metal surface arthroplasty of the hip. Clin Orthop 418: 87-93, 2004). Even the 2007 AAOS Poster P034 abstract states, “After changes were made in the prosthetic design, INDICATIONS, and technique, the overall complication rate was reduced.”

7. Additional risks for patients with osteonecrosis

Patients with osteonecrosis of the hip typically have a dead, collapsed segment of the femoral head. Resurfacing a hip with a dead, collapsed segment of the femoral head means that instead of being supported properly by bone, the femoral prosthesis of a hip resurfacing is being supported by a thicker mantle of bone cement. This would suggest that failure of hip resurfacing for osteonecrosis might be high. The Surface Arthroplasty Risk Index give 2 points of additional risk to hips with cystic lesions within the femoral head of greater than 1 centimeter. Virtually all hips with osteonecrosis have a bone defect that is that large or larger.

Beale PE, Dorey FJ, LeDuff M, Gruen T, Amstutz HC. Risk factors affecting outcome of metal-on-metal surface arthroplasty of the hip. Clin Orthop 418: 87-93, 2004.

A study specifically looking at the outcome of hip resurfacing for osteonecrosis demonstrated a 5% revision rate (3/56) at 4.9 years. Two of the three failures were for femoral component loosening.

Beaule PE, Amstutuz HC, Le Duff M, Dorey F. Surface arthroplasty for osteonecrosis of the hip: hemiresurfacing versus metal-on-metal hybrid resurfacing. J Arthroplasty. 2004 Dec;19(8 Suppl 3):54-8.

8. Additional risks for patients with hip dysplasia

Hip resurfacing for dysplastic hips presents several additional risks for early failure. First, these hips generally have the greatest abnormalities in leg length and offset and hip resurfacing has been shown to be incapable of properly dealing with these issues (see #3 above and Silva M, Lee KH, Heisel C, Dela Rosa MA, Schmalzried TP. The biomechanical results of total hip resurfacing arthroplasty. J Bone Joint Surg Am 86A(1):40-46, 2004 ).

Second, dysplastic hips typically have a smaller than average dimensions than other hips and smaller diameter femoral components have been shown to loosen at a higher rate than larger femoral components.

Third, smaller diameter components have been shown to have a higher wear rates than larger diameter components. This leads to higher serum metal ion levels in hips resurfaced with smaller components than hips resurfaced with larger components. ( AAOS 2007 Poster No. 008. Effects of Femoral Head Size on Serum Cobalt and Chromium Levels Following Hip Resurfacing. D Gordon Allan, MD, Brad Dyrstad, BS, Joseph C Milbrandt, PhD).

Fourth, a recent publication documents that hip dislocation following hip resurfacing, even for mild hip dysplasia (Crowe I and Crowe II) has been shown to occur at a very high rate (5%). (Amstutz HC, Antoniades JT, Le Duff MJ. Results of metal-on-metal hybrid hip resurfacing for Crowe type-I and II developmental dysplasia. J Bone Joint Surg AM. 2007 Feb; 89(2):339-46.)

Fifth, the same reference also shows a very high revision rate for hip resurfacing performed for hip dysplasia (12% (7/59) at 6 years).

Based on these factors, patients with hip dysplasia appear to be at significant risk for high wear, dislocation, and early failure.

Summary.

All of this information does not mean that hip resurfacing is never indicated. However, an increasing body of scientific evidence clearly raises questions about the prudence of commonly applying hip resurfacing for end-stage problems about the hip joint.

Other references:

http://www.coa-aco.org/coa_bulletin/issue_67_nov%10dec_2004/waddell_hip_resurfacing_arthroplasty.html

Keywords: hip dislocation and hip resurfacing, femoral neck fracture and hip resurfacing, revision and hip resurfacing, hip dysplasia and hip resurfacing, osteonecrosis and hip resurfacing, women and hip resurfacing, groin pain and hip resurfacing, failure of hip resurfacing, risks of hip resurfacing.