Total hip arthroplasty (THA) results in good outcomes in function and risk for revision in older patients. However, in young, active patients, it results in an increased rate of revision and poorer outcomes. Modern metal-on-metal hip resurfacing arthroplasty (HRA) is described as an appropriate treatment of hip osteoarthritis in young, active patients. The selection of an appropriate prosthesis is critical for this patient demographic. This review compares the functional results of THA and HRA and focuses on range of motion, activity level, groin pain, patient satisfaction, restoration of normal hip anatomy, and gait.
Total hip arthroplasty (THA) has long been considered the treatment of choice for osteoarthritis of the hip in older patients. This procedure results in consistently good outcomes in function and risk for revision in this patient demographic. However, the same procedure performed in young, active patients results in an increased rate of revision and less favorable outcome of those revision proceedures.
Modern metal-on-metal hip resurfacing arthroplasty (HRA), despite having a higher overall rate of revision and less evidence-based literature supporting its use in all demographics, is perceived by patients as being a safer, more effective treatment that results in a greater range of motion than THA. In the literature, HRA is often described as an appropriate treatment of hip osteoarthritis in young, active patients. In Australia, 50% of HRA is performed in patients who are less than 55 years of age.
The active lifestyle of younger patients places additional stresses on hip prostheses for a prolonged period of time that are not encountered in older patients. Furthermore, young active patients are less tolerant of compromised function and, therefore, selection of an appropriate prosthesis that provides good functionality and durability is critical in these patients. Data that compare the functional results of HRA and THA across different patient demographics and activity levels give surgeons the ability to make adequately informed, patient-based decisions regarding prosthesis selection.
We have examined the literature to prepare a review of published studies that compare the functional results of THA and HRA. Specific outcomes such as range of motion, activity level, groin pain, patient satisfaction, and restoration of normal hip anatomy and gait are addressed separately.
The authors systematically reviewed the literature on hip resurfacing outcomes using the PubMed bibliographic database. An initial search was performed to identify all articles that might be relevant to the review by collecting all entries with the keywords “hip resurfacing,” “resurfacing arthroplasty,” “hip resurfacing versus total hip” and “functional outcome hip resurfacing.” The initial keyword search yielded 713 articles. The abstracts of these articles were searched to determine whether they were suitable for inclusion in this review. In examining the references generated by this process, articles were selected that discussed the functional outcome of HRA compared with the functional outcome of conventional total hip replacement, with preference given to studies in which patients were matched for age, gender, and preoperative function. Of the 713 articles found using the keyword search, 46 articles were selected for further evaluation. No preference was given to articles in which a specific femoral head size was used for either HRA or conventional THA. Review articles and the bibliographies of each reference were also searched to find additional articles that appeared relevant.
Range of Motion
Range of motion (ROM) is particularly important for younger patients who wish to return to a highly active lifestyle following joint replacement. Limited ROM may be a consequence of impingement, which may cause subluxation and hence high levels of wear and early failure.
In vitro studies including both cadaver and computer simulation studies consistently show that HRA results in reduced ROM when compared with conventional THA. Bengs and colleagues evaluated 3 contemporary hip resurfacing systems and compared 20 different movements (10 with zero femoral anteversion, and 10 with 20 degrees femoral anteversion) with those of 5 conventional hip replacement systems. Overall, the hip resurfacing systems resulted in less ROM than the conventional THA systems, with the conventional THA having significantly more ROM in 12 of the 20 movements tested. The summed mean arcs of motion in the sagittal, coronal, and axial planes for the HRA group were 135, 78, and 115 degrees, compared with 174, 87, and 150 degrees for the zero anteversion group, and 158, 90, and 147 degrees for the groups with 20 degrees of anteversion. These findings are consistent with those of Kluess and colleagues, who showed that ROM for 8 designs of hip resurfacing prosthesis tested in 3 different leg positions were on average 31 to 48 degrees less than for conventional hip replacements using a 32-mm head diameter. In both studies, neck-on-cup impingement was the cause of the observed reduction in ROM. Incavo and colleagues attempted to eliminate all patient-related variables by using a combination cadaver/computer simulation. The investigators found that, with controlled patient variables, THA was able to restore normal ROM more effectively than HRA. Surface replacement showed minor deficits in extension and significant reductions in flexion and internal rotation at 90° compared with the natural hip. The investigators concluded that decreased ROM for the HRA group was attributed to a smaller head-neck ratio or head-neck offset at points of impingement.
The translation of the results from these in vitro studies to the clinical situation is limited because they do not accurately mimic the complex nature of the in vivo implanted hip. Differences may be expected because of variation in hip anatomy and musculotendinous attachments, as well as subtle differences in surgical approach. Furthermore, fear of instability for hips treated with THA and the benefits of complete capsular release in HRA that overcomes preoperative soft tissue contracture may also cause discrepancies between the results of in vitro and in vivo studies.
Clinical studies report that the ROM for THA and HRA is similar or even better for HRA ( Table 1 ). Vail and colleagues, in a study of 52 patients (57 hips) with resurfacing and 84 patients (93 hips) with cementless THAs, found that, after controlling for age, gender, and preoperative differences, the resurfacing group had significantly higher ROM scores than did the cementless THA group after a mean follow-up of 3 years. However, Lavigne and colleagues, in a single-blind randomized study using digital photography of hip motion, failed to find a difference between patients assigned to the HRA group and the THA group at 1-year follow-up. In this study, patient demographics and preoperative ROM were similar. Le Duff and colleagues also found no difference in ROM between patients treated bilaterally, with an HRA on one side and a conventional THA in the contralateral limb to control patient variability. The investigators reported that the ROM for both implant types was consistent with the ROM seen in normal, undiseased hips. It is possible that, although THA can result in significantly greater ROM in the laboratory setting, this increased ROM is unable to be achieved in patients with normal to average flexibility, thereby resulting in a similar clinical ROM for THA and HRA.
|Study||No. of Patients||Mean Follow-up (mo)||Type of Evaluation||HRA||THA||P Value|
|Lavigne et al, 2010||24 HRA |
|12||Total arc of motion |
Arc of rotation
|Le Duff et al, 2009||35 HRA |
|88||Abduction/adduction Angle |
|71 (45–85) |
|70 (50–95) |
|Vail et al, 2006||52 HRA |
In summary, although the geometry of hip resurfacing components may limit their ROM in the laboratory setting, clinically patients may expect to achieve equivalent, if not better, ROM following HRA. If patients do experience a decreased ROM as a consequence of impingement, then subluxation and edge loading may occur, which can lead to a higher wear rate and early failure.
Many young patients, once relieved of their arthritic pain, may return to a physically active lifestyle after hip arthroplasty.
Most clinical studies report that patients return to high levels of activity after surgery. Six out of 7 studies that compare postoperative activity levels between HRA and THA report that patients have significantly increased activity levels when treated with HRA compared with THA. A summary of postoperative activity scores is presented in Table 2 . Vail and colleagues found significantly higher postoperative UCLA activity scores for patients in the HRA group compared with the THA group. Although the outcomes were controlled for age, gender, and preoperative clinical scores, there was variation in the demographic profile and preoperative hip score. Mont and colleagues, in a comparison of 54 consecutive HRAs and 54 consecutive THAs, found significantly higher postoperative activity levels in the HRA group (see Table 2 ). In the study by Mont and colleagues, the patients were matched according to demographics and preoperative Harris hip score; however, patients were not matched according to preoperative activity level. It was subsequently found that, before surgery, patients treated with HRA had a significantly higher weighted activity score, which, given that preoperative activity level has a positive influence on functional outcome, makes interpretation of these findings challenging. Pollard and colleagues, in a study comparing the outcome of hip resurfacing with THA in patients that were matched for age, gender, body mass index (BMI, calculated as weight in kilograms divided by the square of height in meters), and preoperative activity score, found higher University of California, Los Angeles (UCLA) activity scores in the resurfaced patients (see Table 2 ). Significantly more patients treated with HRA ran, participated in sport, or performed heavy manual labor than those treated with THA, a finding that is also reported by Vendittoli and colleagues and Pattyn and De Smet. These findings are further supported by Zywiel and colleagues, who evaluated patients that were matched in age, gender, BMI, and preoperative activity level. Pre- and postoperative activity levels were measured using a weighted activity questionnaire. At follow-up, the resurfacing patients had a significantly higher mean weighted activity score than those treated with THA (10.0 points, range 1.0–27.5 points, compared with 5.3 points, range 0–12.0 points).
|Study||No. of Patients||Mean Follow-up (mo)||WOMAC||OHS||HHS||UCLA/Activity Score|
|Mont et al, 2009||54 HRA |
|39 (24–60) |
|—||—||—||—||52 (33–71) |
|90 (50–100) |
|3 (0–15) a |
2 (0–6) a
|11.5 (0–32) a |
7 (0–21) a
|Killampalli et al, 2009||58 HRA |
|5 y (4–7)||—||—||44.4 (31–57) |
|16.6 (12–31) |
|—||—||4.2 (1–8) |
|6.7 (3–10) |
|Hall et al, 2009||33 HRA |
|6||—||—||38.1 (36.0–40.2) |
|18.6 (16.3–20.8) |
|Vendittoli et al, 2010||109 HRA |
|2 y||52.7±15.4 |
|Sandiford et al, 2010||141 HRA |
|6.1 (0–56) |
|37 (13–57) |
|15 (12–35) |
|54.1 (7–97) |
|96.8 (59–100) |
|Lavigne et al, 2010||24 HRA |
|12||46.5 (26–79) |
|3.0 (0–12) |
|—||—||—||—||—||8.0 (5–10) |
|Rahman et al, 2010||329 HRA||6.6 y (5–9.2)||47.9 (5–96)||6.9 (0–58)||38.3 (16–60)||15.9 (12–46)||51.3 (7–91)||94.3 (24–100)||4.7 (1–9)||7.5 (3–10)|
|Pollard et al, 2006||54 HRA |
|61 (52–71) |
|—||—||—||15.9 (12–42) |
|—||—||9 (6–10) |
|8.4 (4–10) |
|Fowble et al, 2009||50 HRA |
|38 (24–50) |
|—||—||—||—||46 (25–59) |
|97 (81–100) |
|4.2 (2–7) |
|8.2 (4–10) |
|Zywiel et al, 2009||33 HRA |
|42 (25–68) |
|—||—||—||—||52 (28–71) |
|91 (32–100) |
|2.1 (0–6) a |
2.3 (0–6) a
|10 (1–27.5) a |
5.3 (0–12) a
In the studies described here, the postoperative clinical scores such as Harris hip score, satisfaction score and pain score were similar between patients treated with HRA and those treated with THA, despite significant differences in postoperative activity level. This finding highlights the lack of sensitivity of common scoring systems caused by a ceiling effect that makes it difficult to distinguish between patients at the upper end of the scoring scale. Furthermore, it highlights the value of weighted activity scores in evaluating postoperative functional outcome, a sentiment that is shared by Zywiel and colleagues.
A critical aspect in achieving a good functional outcome is the correct positioning of both the femoral and acetabular components, which is more challenging in HRA. More specifically, the difficulties encountered during implantation of the acetabular component can result in prominence of the acetabular component anteriorly, which has been associated with groin pain caused by irritation of the Iliopsoas tendon. In addition, failure to recognize and treat underlying femoroacetabular impingement lesions can result in ongoing impingement of the femoral neck in flexion.
There are currently few studies that compare the incidence of groin pain between HRA and THA. Bin Nasser and colleagues evaluated the incidence of groin pain after metal-on-metal hip resurfacing for 116 patients with a mean follow-up of 26 months. Twenty-one patients (18%) reported persistent groin pain, with a mean pain score of 5.19 out of 10. Deep anterior groin pain that was aggravated by activity was reported in all of these patients. In patients reporting groin pain, 52% required analgesics and 57% reported that the pain limited their activities. Although patients reporting groin pain improved after surgery, their improvement was less than in those without groin pain.
A similar incidence of groin pain was reported recently by Bartelt and colleagues following a retrospective review of 282 hips treated with either HRA or THA at a mean follow-up of 14 months. Those treated with a conventional metal-on-polyethylene THA bearing reported a 7% incidence of groin pain compared with 18% if treated by metal-on-metal HRA. Of the patients receiving a metal-on-metal THA, 15% reported groin pain. The investigators suggest that the higher incidence of groin pain in the HRA group may have been a consequence of hypersensitivity to the metal-on-metal bearings or greater impingement of the psoas tendon. In this study, younger patients more frequently reported pain; however, a gender association was not found because of the large number of male patients enrolled in this study. The investigators propose that the higher activity and expectation levels in younger patients may make them more likely to report pain after surgery.
The incidence of groin pain reported for HRA in these 2 studies is considerably higher than that reported for conventional metal-on-polyethylene THA. It should be noted that in neither of these two studies was reference made to implant choice, surgical approach or implant orientation. Although these issues are important in all arthroplasty, they are particularly important in resurfacing arthroplasty.
Patient Satisfaction, Quality of Life, and Postoperative Pain
There are several validated scoring systems such as Harris hip score, Oxford hip score, Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) and UCLA activity scores that are used to evaluate the functional outcome of HRA and THA. Although these scoring systems provide useful information regarding many aspects of functional outcome, they are limited in their ability to measure satisfaction from the perspective of the patient.
Several studies report survey-based measures of patient satisfaction, and, although differences in the methods of measuring satisfaction vary between studies, the data provide us with some insight into the patient-perceived outcome of hip replacement. In general, studies report high patient satisfaction levels for both HRA and THA. Even studies in which no difference in satisfaction was observed between patients treated with THA and HRA, overall satisfaction reported was high for both treatments.
Hall and colleagues measured patient satisfaction in a case-matched control study and found that patients treated with HRA were more likely to report excellent or very good pain relief at 6 months, and also reported better heavy lifting ability at 6 months. Overall, both HRA and THA resulted in similar, very high patient satisfaction. This study also found that good preoperative function had a positive effect on postoperative satisfaction. Using an 11-point Likert scale, Mont and colleagues also observed that postoperative pain was similar following HRA and THA, with both procedures resulting in low pain scores of 1.4 and 1.6 for HRA and THA respectively.
Vendittoli and colleagues also found high satisfaction rates for both HRA and THA, with 99% of patients in both the HRA and THA groups reporting that they were satisfied or very satisfied with their procedure. Pain scores measured in this study were also similar between the 2 treatment groups. The only difference found was that a higher percentage of HRA patients experienced pain on rising from a chair (21% compared with 8% for THAs) at the earliest time point of 6 months. In a comparison of the return to function following HRA and THA, Stulberg and colleagues also found a trend for resurfacing patients to have increased pain during the early postoperative period (6 weeks). However, by 24 months there was no difference in pain level between patients having undergone HRA or THA.
Quality of life scores measured by Pollard and colleagues were better for patients treated with HRA. Mean EuroQol 5 part questionaire (EQ-5D) scores were 0.78 and 0.9 for THA and HRA respectively. Similarly, mean EuroQol visual analogue scale (EQ-VAS) scores were 69.3 and 82.3 for THA and HRA respectively. However, there was no preoperative evaluation of hip score in this study and it is therefore possible that patients in the HRA group had better preoperative hip function, which, as was found in the study by Hall and colleagues, is a positive contributor to postoperative outcome. In contrast, a recent study by Garbuz and colleagues found no difference in quality of life as measured by Short form health survey-12 questions (SF-12), WOMAC, and Quality of Life Outcome Index 5 Questions (PAT 5-D) scores, nor did they find any difference in UCLA activity score. However, this study compared HRA with a large-head metal-on-metal THA.
In summary, studies report high patient satisfaction for both THA and HRA. This satisfaction is most likely the result of high postoperative activity level and good pain relief. However, additional factors, such as patient perception and expectation, are also likely to contribute to the observed satisfaction scores.
A summary of the functional scores from 9 recent studies that report the outcome of HRA compared with THA is provided in Table 2 . The functional scores most frequently reported include Oxford hip score, Harris hip score, SF-12 scores, WOMAC, and UCLA activity scores. The functional scores reported are consistent between these studies. In 8 of the 9 studies, the HRA group had better preoperative function than the THA. All of the 9 studies reported an improvement in functional score after surgery for both HRA and THA; however, functional scores were consistently similar between the 2 procedures. Factors that are likely to influence functional outcome include gender, age, and preoperative function.
Restoration of Normal Hip Anatomy and Gait
It is important to restore normal hip biomechanics during the hip arthroplasty procedure to maximize ROM and to avoid leg length and gait abnormalities, dislocation, and increased wear, all of which can contribute to implant failure and ultimately patient dissatisfaction.
There are a small number of studies that have directly compared the outcome of THA and HRA in terms of restoration of normal hip anatomy and limb length. The data from these studies are contradictory, which, given their small number, makes it difficult to draw firm conclusions regarding their outcomes.
Three studies present similar results. Girard and colleagues, in a radiographic analysis of 49 hip resurfacings and 55 THAs, found that femoral offset and limb length were increased when compared with the contralateral limb following THA. In contrast, femoral offset and limb length were reduced following HRA. Limb length inequality was restored in 86% of resurfacings compared with only 60% of THA. The investigators concluded that biomechanical reconstruction of the femur is more reproducible with HRA. Ahmad and colleagues presented similar findings in a comparison of femoral offset and limb length between 28 THAs and 28 HRAs. THA resulted in increased femoral offset, whereas HRA resulted in reduced change in limb length. Also presenting similar findings, Robb and colleagues concluded that, although there was no difference in restoration of leg lengths, HRA more accurately restores femoral and total offset than cemented THA. In contrast, Loughead and colleagues found a greater increase in femoral length following HRA combined with a greater reduction in femoral offset, and concluded that HRA does not restore limb length or offset more accurately than THA.
Gait analysis has been used in an attempt to differentiate any functional differences between HRA and THA. Some gait analysis studies provide evidence to suggest that HRA results in equivalent or better walking speed, abductor moment, and extensor moment than THA.
Mont and colleagues compared walking speed, abductor, and extensor moments between 15 patients treated with THA, 15 patients treated with HRA, and 10 patients with untreated osteoarthritis of the hip after a mean follow-up of 13 months, and found that patients treated with HRA were able to walk significantly faster than patients treated with THA and had gait parameters that were closer to normal. Shrader and colleagues reported similar findings in a pilot study evaluating walking speed and stair negotiation in 7 patients treated with THA, 7 treated with HRA and 7 normal controls. Patients treated with HRA had more normal patterns of movement with greater improvement in hip abduction and extension moments than patients treated with THA. Stair negotiation was also improved in the HRA group. Lavigne and colleagues reported equivalent gait speed at both normal walking speed and fast walking speed for 48 patients treated with THA and 48 patients treated with HRA when evaluated at a mean of 14 months. Evaluation of postural balance was also similar between the groups. Patients in both treatment groups reached most control group values by 3 months after surgery. Similar findings were presented by Shimmin and colleagues in a comparison of patients with a postoperative Harris hip score of 100 treated with THA and HRA using the normal contralateral hip as a control. Gait at a fast walking or jogging pace, as well as abductor strength and stair climbing speed, was similar between the groups.
In summary, further studies that evaluate restoration of hip and limb anatomy and their effect on gait are necessary to be able to draw accurate conclusions on the benefits of HRA compared with THA in young, active patients. However, the limited data currently available suggest that the closer approximation of the proximal femoral anatomy combined with the conservation of proximal femoral bone stock achieved with HRA may result in a more normal gait in these patients.