The Journal of Arthroplasty 30 (2015) 1160–1166
Contents lists available at ScienceDirect
The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org
Birmingham Hip Resurfacing: A Single Surgeon Series Reported at a Minimum of 10 Years Follow-Up Akshay Mehra, FRCS (Tr and Orth) a, Fiona Berryman, PhD b, Gulraj S. Matharu, BSc (Hons), MBChB, MRCS b, Paul B. Pynsent, PhD b, Eric S. Isbister, FRCS (Tr and Orth) c a b c
Alexandra Hospital, Redditch, United Kingdom The Royal Orthopaedic Hospital, Birmingham, United Kingdom New Cross Hospital, Wolverhampton, United Kingdom
a r t i c l e
i n f o
Article history: Received 9 December 2014 Accepted 28 January 2015 Keywords: hip resurfacing metal-on-metal outcomes revision surgery survival
a b s t r a c t We report outcomes on 120 Birmingham Hip Resurfacings (BHRs) (mean age 50 years) at a minimum of tenyears follow-up. Cases were performed by one surgeon and included his learning curve. Six hips were revised, with no revisions for infection, dislocation, or adverse reaction to metal debris. Ten-year survival was 94.2% (95% confidence interval (CI) 88.8%–98.7%) for all revisions and 96.1% (95% CI 91.5%–99.8%) for revisions for aseptic loosening. Gender (P = 0.463) and head size (P = 0.114) did not affect revision risk. Mean post-operative Harris hip score was 84.0. Contrary to previous independent reports, good outcomes into the second decade were achieved with the BHR in both men and women. Longer term follow-up will confirm whether these promising outcomes in women continue. © 2015 Elsevier Inc. All rights reserved.
In young and active patients, total hip arthroplasty (THR) using metal-on-polyethylene bearings has been associated with unsatisfactory outcomes and high failure rates due to wear debris resulting in osteolysis and component loosening [1,2]. Hard-on-hard bearings, such as metal-on-metal and ceramic-on-ceramic, were subsequently introduced as a low wearing alternative arthroplasty option for these high demand patients [3]. Hip resurfacing had also been an attractive concept as a bone conserving and more functional arthroplasty option to THR. Although hip resurfacing was originally described in the 1950s [4], over the decades attempts by numerous surgeons using a metalon-polyethylene articulation were unsuccessful due to early fixation failures and high polyethylene wear rates [5–7]. Metal-on-metal hip resurfacing was subsequently developed to address these major wear and fixation issues. Metal-on-metal hip resurfacing provides an alternative to THR, and appears best suited to younger patients given that it preserves femoral bone stock, therefore theoretically allowing more straightforward revision surgery [8,9]. Over recent years concerns have mounted regarding adverse reactions to metal debris (ARMD) associated with metal-on-metal hip arthroplasties. This condition is the sequelae of large amounts of metal debris released from metal-on-metal articulations due to wear and One or more of the authors of this paper have disclosed potential or pertinent conflicts of interest, which may include receipt of payment, either direct or indirect, institutional support, or association with an entity in the biomedical field which may be perceived to have potential conflict of interest with this work. For full disclosure statements refer to http://dx.doi.org/10.1016/j.arth.2015.01.042. Reprint requests: Akshay Mehra, FRCS (Tr and Orth), Alexandra Hospital, Redditch, United Kingdom. E-mail address: [email protected] (A. Mehra). http://dx.doi.org/10.1016/j.arth.2015.01.042 0883-5403/© 2015 Elsevier Inc. All rights reserved.
corrosion, with ARMD resulting in high short-term failure of certain metal-on-metal hip arthroplasties [10–12]. It has become apparent that the outcomes of metal-on-metal hip resurfacing are dependent on various patient, surgeon, and implant factors [13]. Women, small femoral component size, malposition of the acetabular component, patients with hip dysplasia, and certain implant designs are reported risk factors for hip resurfacing failure [11,12,14–16]. Hip resurfacing usage has significantly declined from 10.8% of all primary hip arthroplasties in England and Wales in 2006 to 1.3% in 2012 [17]. Whilst some designs have performed poorly and subsequently been withdrawn [17–19], the Birmingham Hip Resurfacing (BHR) remains the most commonly implanted resurfacing device worldwide [20]. Good to excellent outcomes are reported for the BHR by the designing surgeons [21,22] at up to 15-years follow-up, and by independent centres at 10-years [23–26]. However, despite achieving good outcomes in men some of these independent centres have observed significantly inferior results in women leading them to recommend against performing hip resurfacing in women [23,25]. The study aims were to determine the survival, radiological, and functional outcomes of the first 120 BHRs performed by a single surgeon at a minimum of 10 years follow-up. Patients and Methods Study Design, Patient Selection Criteria and Demographics A retrospective review of prospectively collected data was performed on all consecutive BHRs (Smith & Nephew, Warwick, United Kingdom) implanted at one district general hospital between
A. Mehra et al. / The Journal of Arthroplasty 30 (2015) 1160–1166
1997 and 2001. All operations were performed by a single surgeon (ESI). The operating surgeon had trained with the designing surgeon during the early 1990s, and to date has implanted over 1000 BHRs independently. This cohort includes the surgeon’s first BHRs, and therefore his learning curve with this procedure. Patients were considered eligible for BHR if they were young (men 65 years and under, and women 60 years and under) and maintained an active lifestyle (including sports participation and/or manual employment), with relatively normal hip morphology and likely to require revision surgery during their lifetime. At the time these operations were performed, diagnoses other than primary osteoarthritis (such as avascular necrosis and inflammatory arthritis) were not considered absolute contraindications for BHR provided patients met all of the other selection criteria. Other surgeons have similarly performed hip resurfacing in patients with these alternative diagnoses early in their respective cohorts [21,22,25,26]. However, the indications at this centre have subsequently been modified over recent years to be more selective and exclude patients with inflammatory arthritis. Contraindications for BHR included patients with impaired renal function, or hip morphology requiring significant correction of offset, leg length and/or where there was a large femoroacetabular size mismatch [27]. Patient demographics are summarised in Table 1 with the study cohort comprising 120 consecutive BHRs implanted in 103 patients. All data presented were collected from the hospital database, patient case notes, and pelvic radiographs.
Surgical Technique and Postoperative Regimen All surgeries were performed using a posterior approach to the hip joint [28]. Care was taken to achieve good exposure to allow satisfactory component alignment. The intention was to achieve 45° of acetabular component inclination, an anteversion aligned with the native acetabulum, and slight femoral valgus. All patients with acetabular dysplasia underwent bone grafting using autograft (acetabular reamings). Femoral neck notching and excessive loading of the femoral neck during preparation was avoided to minimise the risk of fracture. Cysts and areas of avascular necrosis were curetted to healthy bone in all cases. Defects of up to 30% were accepted though in the area of the superior head/neck junction only smaller defects were accepted. Additional cement rather than bone graft was used to fill any defects. Low viscosity cement was used on the femoral side filling the component to just above the chamfer line. Postoperatively all patients were allowed to mobilise full weight bearing with crutches except those requiring bone grafting for acetabular dysplasia or where there was concern over acetabular press-fit or proximal femoral bone stock. This latter group was kept partial weight bearing for 6 to 12 weeks. Patients received both mechanical (TED Table 1 Summary of the Study Cohort. Study Group (n = 120 hips) Gender Age Diagnosis
Follow-up time Femoral component size
Male Female Mean (range) in years Primary osteoarthritis Avascular necrosis Developmental dysplasia Other causes Not documented Rheumatoid arthritis Mean (range) in years 42 mm 46 mm 50 mm 54 mm 58 mm Not documented
63 (52.5%) 57 (47.5%) 50 (28 to 63) 68 (56.6%) 17 (14.1%) 14 (11.6%) 9 (7.5%) 8 (6.6%) 4 (3.3%) 10.8 (10.0 to 14.0) 28 (23.3%) 28 (23.3%) 30 (25.0%) 26 (21.6%) 3 (2.5%) 5 (4.1%)
1161
anti-embolism stocking, Kendall Health Care Group, Miami, Florida) and chemical (Warfarin) thromboprophylaxis for 6 weeks postoperatively, with the latter commenced the day before BHR surgery. All patients underwent clinical review in the outpatient clinic at 6 weeks, 3 months, and 1 year postoperatively, and then every subsequent 2 years postoperatively. These consultations included clinical examination and anteroposterior pelvic radiographs, however blood and urine analyses to determine metal ion concentrations were not performed. In addition, the Oswestry Outcome Centre independently posted functional outcome scores to all patients at annual intervals postoperatively as previously described [29]. Outcomes of Interest Outcomes of interest at a minimum of 10-years follow-up were implant survival, radiological outcome, and functional outcome. Radiographs were assessed for acetabular component inclination and anteversion angles using the open source software, ImageJ [30]. Inclination was measured directly on the radiographs using the angle between the tear drop line and the long axis of an ellipse projected on the circular opening of the cup; the length of the short and long axes of the fitted ellipse was also measured automatically for the calculation of the version angle (Fig. 1). Radiographic anteversion was computed using the Lewinnek method [31] (Fig. 1). This method has been shown to have good inter-observer and intra-observer reliability with the measurements obtained similar to those from computerised tomography scanning [32]. All radiographs were also analysed for signs suggestive of implant failure. The femoral component was considered to have evidence of loosening if there was a radiolucent line N2 mm in any of the three zones described by Amstutz et al [33]. Acetabular loosening was defined as a radiolucent line N2 mm in two or more zones as described by DeLee and Charnley [34]. Femoral neck narrowing of greater than 10% was considered significant as previously described [35]. Any osteolysis around the femoral or acetabular component was recorded. Functional outcomes were assessed using the Harris hip score (HHS; 0–100) [36] and a patient satisfaction score (0–4). The scoring for patient satisfaction after BHR surgery was as follows: 4 = extremely pleased, 3 = pleased, 2 = no different, 1 = worse than before, 0 = much worse than before [29]. Statistical Analysis All statistical analysis was performed using the programme R (R Foundation for Statistical Computing, Vienna, Austria) [37]. Cumulative BHR survival was determined using the Kaplan–Meier method, with the Peto method used to calculate the lower 95% confidence interval (CI). The endpoint for survival analysis was revision surgery, defined as removal or exchange of either the femoral or the acetabular component, or both. Patients not undergoing revision surgery were censored after their last contact with the hospital or after death. A Cox’s proportional hazards model was used to compare the differences in BHR survival distributions for each of the covariates recorded. A multivariate model was constructed, and then covariates that were not significantly influential were systematically removed from the model to identify those having the greatest influence on survival. Depending on data distribution either the median and interquartile range (IQR) or the mean and range were used. The level of significance was set at P b 0.05 with CIs set at the 95% level. Results Survival Analysis and Factors Affecting Survival Of the 120 BHRs implanted in 103 patients, 13 hips (12 patients) were lost to follow-up and 9 hips (8 patients) died during the study
1162
A. Mehra et al. / The Journal of Arthroplasty 30 (2015) 1160–1166
Fig. 1. The measurements necessary to calculate acetabular component inclination and version using the Lewinnek method [31].
period. This left 98 BHRs (83 patients) available for analysis. Of these, 6 hips (6 patients) have required revision. Therefore 92 BHRs (77 patients) in this cohort were available to determine clinical outcomes at a minimum of 10-years following arthroplasty. One of the patient deaths was secondary to a pulmonary embolism 13 days following BHR. None of the other deaths were related to the index surgery and none of these patients underwent further surgery on their BHR prior to death. Of the 6 hips revised (4 females and 2 males), 4 were performed for aseptic loosening (3 femoral and 1 acetabular) at a mean of 4.8 years (range 1.9–8.1 years) following BHR. In addition, 1 BHR was revised for a traumatic femoral neck fracture and 1 revision was performed for stiffness secondary to heterotopic ossification. There were no early femoral neck fractures, deep infections, or dislocations in the remaining cohort and no patient with a surviving BHR is currently under investigation for ARMD. No patient has undergone any further surgical intervention apart from the six revised hips. All postoperative complications are outlined in Table 2. One patient sustained an intraoperative posterior acetabular wall fracture during cup impaction, though a satisfactory press-fit was still achieved. No fixation was required, the BHR implants were stable, and the fracture subsequently healed with no adverse outcomes. The surviving 92 hips (77 patients) were followed up for a minimum of 10-years (mean 10.8 years; range 10.0 to 14.0 years) from index BHR surgery. Cumulative 10-year survival for the whole cohort (n = 120) with revision for any indication as the endpoint was 94.2% (95% CI 88.8%–98.7%; Fig. 2) and 96.1% (95% CI 91.5%–99.8%) when revision for aseptic loosening was used as the endpoint. Cumulative 10-year survival for males and females was 96.0% (95% CI 90.7%–100%) and 92.5% (95% CI 85.8%–99.9%) respectively. Cox-proportional hazards modelling demonstrated a 53% reduction in risk of revision in males, but neither
gender (P = 0.463) nor femoral component head size (P = 0.114) reached statistical significance. Details of Failures and Revision Surgery Overall six patients required revision of either the femoral and/or acetabular component (Table 3). Three patients required revision for femoral component loosening at 4.5, 4.7, and 8.1 years postoperatively. Radiologically all three patients with femoral loosening showed a typical slow progression of tilting into varus with inferior displacement. The intraoperative findings in each case were consistent with avascular necrosis of the femoral head though this diagnosis was not established after histopathological analysis (Fig. 3A and B). All failed femoral components were revised to cemented stems with modular heads and retention of the BHR acetabular component. The tissues and bone in all cases were healthy with no evidence of metallosis, soft-tissue damage or osteolysis.
Table 2 Postoperative Complications Not Requiring Further Surgery. Complication
Number of Hips
Trochanteric bursitis Deep vein thrombosis Pulmonary embolism (fatal) Superficial wound infection Intraoperative posterior acetabular wall fracture
4 (3.3%) 2 (1.7%) 1 (0.8%) 1 (0.8%) 1 (0.8%)
Fig. 2. Kaplan–Meier survival curve for 120 Birmingham Hip Resurfacings (revision for any reason as the endpoint). Six hips revised in total. Shaded area represents the upper and lower limits of the 95% confidence intervals. Black line represents an implant failure rate of 1% per year.
A. Mehra et al. / The Journal of Arthroplasty 30 (2015) 1160–1166
1163
Table 3 Clinical Details of Birmingham Hip Resurfacings Requiring Revision Surgery (n = 6). Patient Number
Age/Sex Diagnosis
Femoral Head Size (mm)
Time to Revision
Revision Indication
Revision Performed
Outcome after Revision
Traumatic femoral neck fracture
Cemented Exeter stem/Duraloc cementless cup/MoP Cemented CPT stem/modular MMT head/cup retained/MoM
4.5 years
Femoral component loosening
Cemented CPT stem/modular MMT head/cup retained/MoM
50
4.7 years
Femoral component loosening
Cemented CPT stem/modular MMT head/cup retained/MoM
52F OA
46
8.1 years
Femoral component loosening
Cemented CPT stem/modular MMT head/cup retained/MoM
47F Perthes
42
9.1 years
Stiffness secondary to HO
Cemented CPT stem/modular MMT head/cup retained/MoM
No symptoms or adverse radiological features 8-years post-revision. No symptoms or adverse radiological features 1-year post-revision. Subsequently lost to follow-up with no blood metal ions. No symptoms or adverse radiological features 10-years post-revision. Normal ultrasound but high blood metal ions (Co 977 nmol/l and Cr 386 nmol/l). No symptoms or adverse radiological features 9-years post-revision. Normal ultrasound but borderline raised blood metal ions (Co 131 nmol/l and Cr 72 nmol/l). Low grade pain 7-years post-revision for last 6-months. Cup (zone 1) and stem (zone 1 and 7) lucencies. Normal ultrasound with negative aspiration on culture. Blood Co 166 nmol/l and Cr 125 nmol/l. Likely to require re-revision over next 1-year. No symptoms or adverse radiological features 5-years post-revision. Good range of motion but poor abductor function. Awaiting blood metal ion testing.
1
58F OA
42
1.9 years
Acetabular component loosening
2
60F OA
42
2.8 years
3
59M OA
46
4
41M AS
5
6
M = male; F = female; OA = osteoarthritis; MoM = metal-on-metal; MoP = metal-on-polyethylene; AS = ankylosing spondylitis; HO = heterotopic ossification; Co = cobalt; Cr = chromium.
One patient required revision for acetabular component loosening at 1.9 years. There was a failure to gain adequate initial press-fit leading to failure of primary osseo-integration (Fig. 4A). The cup failed by rotation into a more open position with slight proximal migration progressively over a period of 1 year (Fig. 4B). At revision there was no intraoperative evidence of ARMD, and both BHR components were revised. Another patient sustained a traumatic femoral neck fracture at 2.8 years postoperatively and underwent a femoral component revision (cemented femoral stem with a modular head). There was one case of severe heterotopic ossification (Brooker, Grade 4) (Fig. 5A). This required excision of heterotopic bone at 9.1 years postoperatively because of functional restriction, under the care of another surgeon. Although the components were well fixed, during the course of the procedure it was deemed necessary to divide the femoral neck to gain access to the joint requiring revision of the femoral component (Fig. 5B). The trochanteric osteotomy subsequently united. Radiological and Functional Outcomes Radiographs were available for all surviving BHRs at a minimum of 10-years follow-up. The median acetabular inclination and anteversion were 48.1° (IQR 43.8°–52.0°) and 9.7° (IQR 5.2°–15.0°) respectively at latest follow-up with no change in these measurements since initial implantation (Fig. 6). One non-revised BHR demonstrated evidence of radiological femoral loosening around the stem with slight tilt into varus five-years after initial arthroplasty. These appearances have remained stable since and the patient is asymptomatic. None of the other follow-up radiographs demonstrated evidence of component loosening, migration, osteolysis, femoral neck narrowing or notching. Functional outcomes were available for 82 hips (67 patients) at a minimum of 10-years follow-up. The mean patient satisfaction score following BHR was 3.7 (range 2–4), with 95.2% of patients either pleased or extremely pleased after surgery. The mean preoperative HHS was 44.5 (range 28–60), which improved to a mean of 84.0 (range 19–100) at a minimum of 10-years follow-up. Nine patients (10 BHRs) had poor HHSs (less than 70); mean HHS 52 (range 19–67). Two of these patients had trochanteric bursitis. The remaining
7 patients (8 BHRs) had persistent pain with no specific cause for symptoms identified despite thorough investigation (including blood tests for inflammatory markers and metal ion concentrations, radiographs and cross-sectional hip imaging, and spinal imaging). These patients remain under regular clinical review. Outcomes in Patients Lost to Follow-Up Of the 13 hips (12 patients) lost to follow-up within 10-years of BHR, 12 hips (11 patients) underwent clinical review at a mean of 4.4 years (range 0.5–8.3 years) following BHR. At this latest review, all patients had their BHRs in situ, with no post-operative complications or radiological abnormalities, and good HHSs recorded in all but one patient (HHS 67 at 4-years follow-up). All of these patients subsequently moved out of the region, including abroad, therefore further clinical follow-up was not possible despite making several attempts to contact each patient. One patient (1 BHR) lost to follow-up did not attend any of their follow-up appointments after hospital discharge despite multiple attempts made to contact this patient. The outcome in this patient since surgery is therefore unknown. Discussion The present study reports the results for 120 BHRs performed by a single surgeon including his learning curve, with a minimum of 10years follow-up. Good survival (94.2% for all revisions and 96.1% for aseptic loosening) and functional outcomes (mean HHS 84.0 with 95.2% patients satisfied) were observed in a young cohort of patients (mean age 50 years) with no revisions performed for ARMD. These results are comparable to those reported from the designing surgeons (10-year survival of 96.3%–97.4%) [21,22], and compare favourably to those from other independent centres (10-year survival of 87.1%–94.5%) [23–26] and registry data (9-year survival of 91.9%) [17]. Higher failure rates for hip resurfacing have consistently been reported in women. In the Australian Joint Registry the revision rate in women was 2.3 times higher than in men at 7-years [38]. The 10-year survival of BHRs implanted in females at independent centres ranges
1164
A. Mehra et al. / The Journal of Arthroplasty 30 (2015) 1160–1166
Fig. 3. (A) Left BHR demonstrating femoral loosening 2-years postoperatively; (B) Pelvic radiograph following revision of the femoral component only.
from 73.9%–89.1% [23–26,39], which has led some of these authors to recommend against performing hip resurfacing in women [23,25]. Even reports from the designing surgeons [21,22] suggest that results are inferior in females compared to males despite still being well within published recommendations for continued implant usage [40]. Although women have factors which may increase their risk of failure, such as hip dysplasia and the need for smaller femoral component head sizes [15,21], the observation that good results can be obtained in women by designing surgeons suggests that patient selection and operative technique are important factors for success [21,22]. A recent study supported the continued use of the BHR in women provided they have primary osteoarthritis, femoral head anatomy sufficient to allow the use of above a 46 mm sized component, and they meet the other indications for hip resurfacing [21]. The present study findings also support the use of the BHR in women with good survival observed (92.5% at 10-years), though it is recognised that longer term follow-up is required to determine whether these findings continue. This is despite having a high proportion of women (48%) in the cohort (previous studies 26%–41%) [23–26], as well as a higher proportion of patients with femoral head sizes of 46 mm or under (47%) compared to previous studies (up to 39%) [21,24,25]. The commonest causes for hip resurfacing revision recorded in the National Joint Registry of England and Wales were pain (4.78 revisions per 1000 patient-years), ARMD (3.55), and aseptic component loosening (2.92) [17]. In the present cohort no patient underwent revision for pain or ARMD, with aseptic loosening the commonest reason for
Fig. 4. (A) Inadequate initial press fit of the acetabular component; (B) Acetabular component failure 2-years following surgery.
revision (accounting for 4 of 6 revisions). For all revisions performed in this series the bone and soft-tissues were healthy intraoperatively with no signs of metallosis, debris-mediated granuloma, or osteolysis. Revision on the femoral side to a stemmed femoral component was straight forward in all cases, with revision on the acetabular side accomplished with minimal bone loss when using modern instrumentation. However, in light of recent concerns regarding the use of largediameter metal-on-metal stemmed hips [41], revision for femoral component failure at this centre now includes revision of the acetabular component and the use of a non-metal-on-metal bearing. Femoral neck fracture is a unique and usually early complication following hip resurfacing [42]. There were no cases of early femoral neck fracture in the present cohort, with only one traumatic fracture occurring at 3 years postoperatively. In addition, none of the radiographs demonstrated evidence of femoral neck notching from the index surgery. These results compare favourably with registry data (1.57 revisions for fracture per 1000 patient-years) [17] and previous studies which report a prevalence of this complication in up to 2% [25,42]. It is suspected that careful preparation of the femoral head, cementing technique, and positioning of the femoral component are important in reducing the frequency of femoral neck fracture following hip resurfacing [43–45].
A. Mehra et al. / The Journal of Arthroplasty 30 (2015) 1160–1166
1165
Fig. 6. Scatter plot of orientation of the acetabular component (inclination vs. version), estimated using the Lewinnek method [31] with optimal zone (rectangle) for positioning as defined by Grammatopoulos et al [14]. Radiographs of the index hip resurfacing were not available for analysis for one of the six revisions. There were 38 (40%) hips in the optimal zone, 3 of which were revised. There were 58 (60%) hips outside the optimal zone, 2 of which were revised. A Fisher’s exact test shows no significant difference between the proportion being revised within or outside the optimal zones for version and inclination.
Fig. 5. (A) Brooker grade 4 heterotopic ossification 9-years following surgery; (B) Pelvic radiograph 4-years following revision surgery.
Component position is thought to play an important role in the long-term success of hip resurfacing. Evidence suggests that deviation of acetabular inclination and/or version from a safe-zone can lead to increased production of wear debris and subsequent implant failure [12,14,16]. De Haan et al demonstrated that acetabular components with inclination angles greater than 55° combined with small femoral component sizes were likely to generate higher blood metal ion concentrations [16]. The most likely mechanism responsible is decompression of the contact patch necessary for fluid film lubrication between the components resulting in edge loading and increased bearing wear [46,47]. The subtle nuances related to acetabular component position were not fully understood at the time when BHRs in this series were implanted (1997–2001) which may explain some of the positioning variability observed. The position of the acetabular component was in the recommended safe-zone [14] in 40% of BHRs (Fig. 6), with other studies reporting up to 88% of BHR acetabular components were implanted in this safe-zone [24,26]. It is suspected that the results in our series are related to the surgical technique employed during these early cases. The surgeon aimed to achieve 45° of acetabular component inclination, but component anteversion was aligned with the native acetabulum rather than aiming for a pre-determined target. When assessing inclination alone, 88% of patients in this series had components within the safe-zone for inclination. Therefore inadequate anteversion (b 10°) was responsible for most patients with components outside the combined inclination and version safe-zone (Fig. 6). Although presently no adverse outcomes have been observed in the
subgroup with components outside the recommended acetabular safe-zone, these patients will continue to be followed up given the increased potential risk of developing ARMD [14], with a low threshold for performing blood metal ion sampling and/or cross-sectional imaging. We anticipate that by using improved surgical techniques since the importance of cup position in metal-on-metal hip resurfacing was established will result in lower wear and long-term failure rates with the BHRs more recently implanted at this centre. This study has recognised limitations. As it is retrospective this study is subject to the limitations associated with this particular type of study. A proportion of patients were lost to follow-up, whilst some who did undergo clinical and radiographic review at 10-years failed to complete or return their functional outcome questionnaires. Unfortunately this is the reality of performing any long-term outcome study and has been experienced by others reporting 10-year outcomes for the BHR [23–26]. In addition, as BHR patients are young and active there is evidence to suggest that patients initially considered lost to follow-up actually have good outcomes [48], rather than elderly THR cohorts who may have worse outcomes compared to patients that continue to be reviewed [49]. Although outcomes in women following BHR were promising in this cohort compared to those reported by independent centres [23–26,39], it is recognised that the present findings may be influenced by numerous factors. These include the surgeons training and experience with the procedure, patient selection criteria, or even selection bias. Furthermore, although this study presents outcomes at a minimum 10-year follow-up, it is recognised that the cohort is relatively small, therefore the good outcomes observed in women are potentially subject to type II statistical error. This highlights the importance of longer term follow-up of this cohort, as well as validation of these findings by other non-designing surgeon centres. Finally, in line with national recommendations [19] BHR patients were not routinely investigated for evidence of ARMD. Although no BHR patient has been revised or currently is under investigation for ARMD, it is understood that longer-term follow-up is important. Current concerns with metal-on-metal hip resurfacing relate to problems in the short-term, such as fracture, early component loosening, ARMD and undiagnosed pain. The benefits of low-wear in the long-term have yet to be established. Evidence suggests patient selection, implant design, and surgical technique need to be optimised to minimise early failure and ensure good long-term results. The present single surgeon series confirms good clinical outcomes continue into
1166
A. Mehra et al. / The Journal of Arthroplasty 30 (2015) 1160–1166
the second decade for the BHR. Although our findings suggest that good outcomes in women can be achieved with the BHR, longer term followup is required to confirm whether these promising results continue. In our view, the BHR remains the procedure of choice for eligible young active patients who are likely to require revision during their lifetime. Acknowledgment We wish to acknowledge Mr. Eric Robinson at the Oswestry Outcome Centre for his help in collecting the functional outcome scores. The authors would also like to thank The Royal College of Surgeons of England and the Arthritis Research Trust for providing one of the authors with funding in the form of a Surgical Research Fellowship. References 1. Callaghan JJ, Forest EE, Olejniczak JP, et al. Charnley total hip arthroplasty in patients less than fifty years old: a twenty-five-year follow-up note. J Bone Joint Surg Am 1998;80-A:704. 2. Mäkelä KT, Eskelinen A, Pulkkinen P, et al. Results of 3,668 primary total hip replacements for primary osteoarthritis in patients under the age of 55 years. Acta Orthop 2011;82:521. 3. Fisher J, Jin Z, Tipper J, et al. Tribology of alternative bearings. Clin Orthop Relat Res 2006;453:25. 4. Charnley J. Arthroplasty of the hip. A new operation. Lancet 1961;1:1129. 5. Freeman MA, Bradley GW. ICLH surface replacement of the hip. An analysis of the first 10 years. J Bone Joint Surg (Br) 1983;65:405. 6. Wagner H. Surface replacement arthroplasty of the hip. Clin Orthop Relat Res 1978; 134:102. 7. Amstutz HC, Clarke IC, Christie J, et al. Total hip articular replacement by internal eccentric shells: the “tharies” approach to total surface replacement arthroplasty. Clin Orthop Relat Res 1977;128:261. 8. Hing C, Back D, Shimmin A. Hip resurfacing: indications, results and conclusions. Instr Course Lect 2007;56:171. 9. Ball ST, Le Duff MJ, Amstutz HC. Early results of conversion of a failed femoral component in hip resurfacing arthroplasty. J Bone Joint Surg Am 2007;89-A:735. 10. Pandit H, Glyn-Jones S, McLardy-Smith P, et al. Pseudotumors associated with metalon-metal hip resurfacings. J Bone Joint Surg (Br) 2008;90-B:847. 11. Langton DJ, Jameson SS, Joyce TJ, et al. Early failure of metal-on-metal bearings in hip resurfacing and larger-diameter total hip replacement: a consequence of excess wear. J Bone Joint Surg (Br) 2010;92-B:38. 12. Langton DJ, Joyce TJ, Jameson SS, et al. Adverse reaction to metal debris following hip resurfacing: the influence of component type, orientation and volumetric wear. J Bone Joint Surg (Br) 2011;93-B:164. 13. Haddad FS, Thakrar RR, Hart AJ, et al. Metal-on-metal bearings: the evidence so far. J Bone Joint Surg (Br) 2011;93-B:572. 14. Grammatopoulos G, Pandit H, Glyn-Jones S, et al. Optimal acetabular orientation for hip resurfacing. J Bone Joint Surg (Br) 2010;92-B:1072. 15. Glyn-Jones S, Pandit H, Kwon YM, et al. Risk factors for inflammatory pseudotumour formation following hip resurfacing. J Bone Joint Surg (Br) 2009;91-B:1566. 16. De Haan R, Pattyn C, Gill HS, et al. Correlation between inclination of the acetabular component and metal ion levels in metal-on-metal hip resurfacing replacement. J Bone Joint Surg (Br) 2008;90-B:1291. 17. National Joint Registry for England and Wales. 10th Annual Report. Available from URL http://njrcentre.org.uk; 2013. [last accessed on 20th January 2015]. 18. Medical and Healthcare Products Regulatory Agency (MHRA). Medical Device Alert: ASR™ hip replacement implant manufactured by DePuy International Ltd. MDA/ 2010/069. Available from URL http://www.mhra.gov.uk/; 2010. [last accessed on 20th January 2015]. 19. Medical and Healthcare Products Regulatory Agency (MHRA). Medical Device Alert: all metal-on-metal (MoM) hip replacements. MDA/2012/036. Available from URL http://www.mhra.gov.uk/; 2012. [last accessed on 20th January 2015]. 20. Birmingham Hip Resurfacing System. Available from URL http://www.smith-nephew.com/professional/products/all-products/bhr-birmingham-hip-resurfacing/. [last accessed on 20th January 2015].
21. Matharu GS, McBryde CW, Pynsent WB, et al. The outcome of the Birmingham Hip Resurfacing in patients aged b50 years up to 14 years post-operatively. Bone Joint J 2013;95-B:1172. 22. Daniel J, Pradhan C, Ziaee H, et al. Results of Birmingham hip resurfacing at 12 to 15 years. Bone Joint J 2014;96-B:1298. 23. Coulter G, Young DA, Dalziel RE, et al. Birmingham hip resurfacing at a mean of ten years: results from an independent centre. J Bone Joint Surg (Br) 2012; 94-B:315. 24. Holland JP, Langton DJ, Hashmi M. Ten-year clinical, radiological and metal ion analysis of the Birmingham Hip Resurfacing: from a single, non-designer surgeon. J Bone Joint Surg (Br) 2012;94-B:471. 25. Murray DW, Grammatopoulos G, Pandit H, et al. The ten-year survival of the Birmingham hip resurfacing: an independent series. J Bone Joint Surg (Br) 2012;94-B:1180. 26. Van Der Straeten C, Van Quickenborne D, De Roest B, et al. Metal ion levels from wellfunctioning Birmingham Hip Resurfacings decline significantly at ten years. Bone Joint J 2013;95-B:1332. 27. Daniel J, Pradhan C, Ziaee H. Patient selection and timing of operation. In: McMinn DJW, editor. Modern Hip Resurfacing. 1st ed. London: Springer; 2009. p. 163. 28. McMinn DJW. Patient positioning and exposure. In: McMinn DJW, editor. Modern Hip Resurfacing. 1st ed. London: Springer; 2009. p. 189. 29. Gilbert RE, Cheung G, Carrothers AD, et al. Functional results of isolated femoral revision of hip resurfacing arthroplasty. J Bone Joint Surg Am 2010;92-A:1600. 30. Rasband WS, Image J. US National Institutes of Health, Bethesda, Maryland, USA. http://imagej.nih.gov/ij/. [1997–2014. Last accessed on 20th January 2015]. 31. Lewinnek GE, Lewis JL, Terr R, et al. Dislocation after total hip arthroplasties. J Bone Joint Surg Am 1978;60-A:217. 32. Nho JH, Lee YK, Kim HJ, et al. Reliability and validity of measuring version of the acetabular component. J Bone Joint Surg (Br) 2012;94-B:32. 33. Amstutz HC, Beaule PE, Dorey FJ, et al. Metal on metal hybrid surface arthroplasty: two to six year follow up study. J Bone Joint Surg Am 2004;86-A:28. 34. DeLee JG, Charnley J. Radiological demarcation of cemented sockets in total hip replacement. Clin Orthop Relat Res 1976;121:20. 35. Hing CB, Young DA, Dalziel RE, et al. Narrowing of the neck in resurfacing arthroplasty of the hip: a radiological study. J Bone Joint Surg (Br) 2007;89-B:1019. 36. Mahomed NN, Arndt DC, McGrory BJ, et al. The Harris hip score: comparison of patient self-report with surgeon assessment. J Arthroplasty 2001;16:575. 37. R Development Core Team. R: A Language and Environment for Statistical Computing. Vienna, Austria: the R Foundation for Statistical Computing. Available from URL http://www.R-project.org/. [last accessed on 20th January 2015]. 38. Australian National Joint Registry Annual Report 2013. Available from URL http://aoanjrr. dmac.adelaide.edu.au/annual-reports-2013. [last accessed on 20th January 2015]. 39. Reito A, Puolakka T, Elo P, et al. Outcome of Birmingham hip resurfacing at ten years: role of routine whole blood metal ion measurements in screening for pseudotumours. Int Orthop 2014;38:2251. 40. Guidance in the Selection of Prostheses for Primary Total Hip Replacement. Technology Appraisal Guidance — No. 2. National Institute for Health and Clinical Excellence. Available from URL http://www.nice.org.uk; 2003. [last accessed on 20th January 2015]. 41. Smith AJ, Dieppe P, Vernon K, et al, National Joint Registry of England and Wales. Failure rates of stemmed metal-on-metal hip replacements: analysis of data from the National Joint Registry for England and Wales. Lancet 2012;379:1199. 42. Shimmin AJ, Back D. Femoral neck fractures following Birmingham hip resurfacing. J Bone Joint Surg (Br) 2005;87-B:463. 43. Davis ET, Olsen M, Zdero R, et al. A biomechanical and finite element analysis of femoral neck notching during hip resurfacing. J Biomech Eng 2009;131:041002. 44. Beaule PE, Matar WY, Poitras P, et al. 2008 Otto Aufranc Award: component design and technique affect cement penetration in hip resurfacing. Clin Orthop Relat Res 2009;467:84. 45. Davis ET, Olsen M, Zdero R, et al. Femoral neck fracture following hip resurfacing: the effect of alignment of the femoral component. J Bone Joint Surg (Br) 2008;90-B:1522. 46. Kwon YM, Glyn-Jones S, Simpson DJ, et al. Analysis of wear of retrieved metal-onmetal hip resurfacing implants revised due to pseudotumours. J Bone Joint Surg (Br) 2010;92-B:356. 47. Langton DJ, Jameson SS, Joyce TJ, et al. The effect of component size and orientation on the concentrations of metal ions after resurfacing arthroplasty of the hip. J Bone Joint Surg (Br) 2008;90-B:1143. 48. Matharu GS, McBryde CW, Treacy RB, et al. Impact of active patient follow-up on worst-case implant survival analysis. Hip Int 2013;23:259. 49. Murray DW, Britton AR, Bulstrode CJ. Loss to follow-up matters. J Bone Joint Surg (Br) 1997;79-B:254.
Contents lists available at ScienceDirect
The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org
Birmingham Hip Resurfacing: A Single Surgeon Series Reported at a Minimum of 10 Years Follow-Up Akshay Mehra, FRCS (Tr and Orth) a, Fiona Berryman, PhD b, Gulraj S. Matharu, BSc (Hons), MBChB, MRCS b, Paul B. Pynsent, PhD b, Eric S. Isbister, FRCS (Tr and Orth) c a b c
Alexandra Hospital, Redditch, United Kingdom The Royal Orthopaedic Hospital, Birmingham, United Kingdom New Cross Hospital, Wolverhampton, United Kingdom
a r t i c l e
i n f o
Article history: Received 9 December 2014 Accepted 28 January 2015 Keywords: hip resurfacing metal-on-metal outcomes revision surgery survival
a b s t r a c t We report outcomes on 120 Birmingham Hip Resurfacings (BHRs) (mean age 50 years) at a minimum of tenyears follow-up. Cases were performed by one surgeon and included his learning curve. Six hips were revised, with no revisions for infection, dislocation, or adverse reaction to metal debris. Ten-year survival was 94.2% (95% confidence interval (CI) 88.8%–98.7%) for all revisions and 96.1% (95% CI 91.5%–99.8%) for revisions for aseptic loosening. Gender (P = 0.463) and head size (P = 0.114) did not affect revision risk. Mean post-operative Harris hip score was 84.0. Contrary to previous independent reports, good outcomes into the second decade were achieved with the BHR in both men and women. Longer term follow-up will confirm whether these promising outcomes in women continue. © 2015 Elsevier Inc. All rights reserved.
In young and active patients, total hip arthroplasty (THR) using metal-on-polyethylene bearings has been associated with unsatisfactory outcomes and high failure rates due to wear debris resulting in osteolysis and component loosening [1,2]. Hard-on-hard bearings, such as metal-on-metal and ceramic-on-ceramic, were subsequently introduced as a low wearing alternative arthroplasty option for these high demand patients [3]. Hip resurfacing had also been an attractive concept as a bone conserving and more functional arthroplasty option to THR. Although hip resurfacing was originally described in the 1950s [4], over the decades attempts by numerous surgeons using a metalon-polyethylene articulation were unsuccessful due to early fixation failures and high polyethylene wear rates [5–7]. Metal-on-metal hip resurfacing was subsequently developed to address these major wear and fixation issues. Metal-on-metal hip resurfacing provides an alternative to THR, and appears best suited to younger patients given that it preserves femoral bone stock, therefore theoretically allowing more straightforward revision surgery [8,9]. Over recent years concerns have mounted regarding adverse reactions to metal debris (ARMD) associated with metal-on-metal hip arthroplasties. This condition is the sequelae of large amounts of metal debris released from metal-on-metal articulations due to wear and One or more of the authors of this paper have disclosed potential or pertinent conflicts of interest, which may include receipt of payment, either direct or indirect, institutional support, or association with an entity in the biomedical field which may be perceived to have potential conflict of interest with this work. For full disclosure statements refer to http://dx.doi.org/10.1016/j.arth.2015.01.042. Reprint requests: Akshay Mehra, FRCS (Tr and Orth), Alexandra Hospital, Redditch, United Kingdom. E-mail address: [email protected] (A. Mehra). http://dx.doi.org/10.1016/j.arth.2015.01.042 0883-5403/© 2015 Elsevier Inc. All rights reserved.
corrosion, with ARMD resulting in high short-term failure of certain metal-on-metal hip arthroplasties [10–12]. It has become apparent that the outcomes of metal-on-metal hip resurfacing are dependent on various patient, surgeon, and implant factors [13]. Women, small femoral component size, malposition of the acetabular component, patients with hip dysplasia, and certain implant designs are reported risk factors for hip resurfacing failure [11,12,14–16]. Hip resurfacing usage has significantly declined from 10.8% of all primary hip arthroplasties in England and Wales in 2006 to 1.3% in 2012 [17]. Whilst some designs have performed poorly and subsequently been withdrawn [17–19], the Birmingham Hip Resurfacing (BHR) remains the most commonly implanted resurfacing device worldwide [20]. Good to excellent outcomes are reported for the BHR by the designing surgeons [21,22] at up to 15-years follow-up, and by independent centres at 10-years [23–26]. However, despite achieving good outcomes in men some of these independent centres have observed significantly inferior results in women leading them to recommend against performing hip resurfacing in women [23,25]. The study aims were to determine the survival, radiological, and functional outcomes of the first 120 BHRs performed by a single surgeon at a minimum of 10 years follow-up. Patients and Methods Study Design, Patient Selection Criteria and Demographics A retrospective review of prospectively collected data was performed on all consecutive BHRs (Smith & Nephew, Warwick, United Kingdom) implanted at one district general hospital between
A. Mehra et al. / The Journal of Arthroplasty 30 (2015) 1160–1166
1997 and 2001. All operations were performed by a single surgeon (ESI). The operating surgeon had trained with the designing surgeon during the early 1990s, and to date has implanted over 1000 BHRs independently. This cohort includes the surgeon’s first BHRs, and therefore his learning curve with this procedure. Patients were considered eligible for BHR if they were young (men 65 years and under, and women 60 years and under) and maintained an active lifestyle (including sports participation and/or manual employment), with relatively normal hip morphology and likely to require revision surgery during their lifetime. At the time these operations were performed, diagnoses other than primary osteoarthritis (such as avascular necrosis and inflammatory arthritis) were not considered absolute contraindications for BHR provided patients met all of the other selection criteria. Other surgeons have similarly performed hip resurfacing in patients with these alternative diagnoses early in their respective cohorts [21,22,25,26]. However, the indications at this centre have subsequently been modified over recent years to be more selective and exclude patients with inflammatory arthritis. Contraindications for BHR included patients with impaired renal function, or hip morphology requiring significant correction of offset, leg length and/or where there was a large femoroacetabular size mismatch [27]. Patient demographics are summarised in Table 1 with the study cohort comprising 120 consecutive BHRs implanted in 103 patients. All data presented were collected from the hospital database, patient case notes, and pelvic radiographs.
Surgical Technique and Postoperative Regimen All surgeries were performed using a posterior approach to the hip joint [28]. Care was taken to achieve good exposure to allow satisfactory component alignment. The intention was to achieve 45° of acetabular component inclination, an anteversion aligned with the native acetabulum, and slight femoral valgus. All patients with acetabular dysplasia underwent bone grafting using autograft (acetabular reamings). Femoral neck notching and excessive loading of the femoral neck during preparation was avoided to minimise the risk of fracture. Cysts and areas of avascular necrosis were curetted to healthy bone in all cases. Defects of up to 30% were accepted though in the area of the superior head/neck junction only smaller defects were accepted. Additional cement rather than bone graft was used to fill any defects. Low viscosity cement was used on the femoral side filling the component to just above the chamfer line. Postoperatively all patients were allowed to mobilise full weight bearing with crutches except those requiring bone grafting for acetabular dysplasia or where there was concern over acetabular press-fit or proximal femoral bone stock. This latter group was kept partial weight bearing for 6 to 12 weeks. Patients received both mechanical (TED Table 1 Summary of the Study Cohort. Study Group (n = 120 hips) Gender Age Diagnosis
Follow-up time Femoral component size
Male Female Mean (range) in years Primary osteoarthritis Avascular necrosis Developmental dysplasia Other causes Not documented Rheumatoid arthritis Mean (range) in years 42 mm 46 mm 50 mm 54 mm 58 mm Not documented
63 (52.5%) 57 (47.5%) 50 (28 to 63) 68 (56.6%) 17 (14.1%) 14 (11.6%) 9 (7.5%) 8 (6.6%) 4 (3.3%) 10.8 (10.0 to 14.0) 28 (23.3%) 28 (23.3%) 30 (25.0%) 26 (21.6%) 3 (2.5%) 5 (4.1%)
1161
anti-embolism stocking, Kendall Health Care Group, Miami, Florida) and chemical (Warfarin) thromboprophylaxis for 6 weeks postoperatively, with the latter commenced the day before BHR surgery. All patients underwent clinical review in the outpatient clinic at 6 weeks, 3 months, and 1 year postoperatively, and then every subsequent 2 years postoperatively. These consultations included clinical examination and anteroposterior pelvic radiographs, however blood and urine analyses to determine metal ion concentrations were not performed. In addition, the Oswestry Outcome Centre independently posted functional outcome scores to all patients at annual intervals postoperatively as previously described [29]. Outcomes of Interest Outcomes of interest at a minimum of 10-years follow-up were implant survival, radiological outcome, and functional outcome. Radiographs were assessed for acetabular component inclination and anteversion angles using the open source software, ImageJ [30]. Inclination was measured directly on the radiographs using the angle between the tear drop line and the long axis of an ellipse projected on the circular opening of the cup; the length of the short and long axes of the fitted ellipse was also measured automatically for the calculation of the version angle (Fig. 1). Radiographic anteversion was computed using the Lewinnek method [31] (Fig. 1). This method has been shown to have good inter-observer and intra-observer reliability with the measurements obtained similar to those from computerised tomography scanning [32]. All radiographs were also analysed for signs suggestive of implant failure. The femoral component was considered to have evidence of loosening if there was a radiolucent line N2 mm in any of the three zones described by Amstutz et al [33]. Acetabular loosening was defined as a radiolucent line N2 mm in two or more zones as described by DeLee and Charnley [34]. Femoral neck narrowing of greater than 10% was considered significant as previously described [35]. Any osteolysis around the femoral or acetabular component was recorded. Functional outcomes were assessed using the Harris hip score (HHS; 0–100) [36] and a patient satisfaction score (0–4). The scoring for patient satisfaction after BHR surgery was as follows: 4 = extremely pleased, 3 = pleased, 2 = no different, 1 = worse than before, 0 = much worse than before [29]. Statistical Analysis All statistical analysis was performed using the programme R (R Foundation for Statistical Computing, Vienna, Austria) [37]. Cumulative BHR survival was determined using the Kaplan–Meier method, with the Peto method used to calculate the lower 95% confidence interval (CI). The endpoint for survival analysis was revision surgery, defined as removal or exchange of either the femoral or the acetabular component, or both. Patients not undergoing revision surgery were censored after their last contact with the hospital or after death. A Cox’s proportional hazards model was used to compare the differences in BHR survival distributions for each of the covariates recorded. A multivariate model was constructed, and then covariates that were not significantly influential were systematically removed from the model to identify those having the greatest influence on survival. Depending on data distribution either the median and interquartile range (IQR) or the mean and range were used. The level of significance was set at P b 0.05 with CIs set at the 95% level. Results Survival Analysis and Factors Affecting Survival Of the 120 BHRs implanted in 103 patients, 13 hips (12 patients) were lost to follow-up and 9 hips (8 patients) died during the study
1162
A. Mehra et al. / The Journal of Arthroplasty 30 (2015) 1160–1166
Fig. 1. The measurements necessary to calculate acetabular component inclination and version using the Lewinnek method [31].
period. This left 98 BHRs (83 patients) available for analysis. Of these, 6 hips (6 patients) have required revision. Therefore 92 BHRs (77 patients) in this cohort were available to determine clinical outcomes at a minimum of 10-years following arthroplasty. One of the patient deaths was secondary to a pulmonary embolism 13 days following BHR. None of the other deaths were related to the index surgery and none of these patients underwent further surgery on their BHR prior to death. Of the 6 hips revised (4 females and 2 males), 4 were performed for aseptic loosening (3 femoral and 1 acetabular) at a mean of 4.8 years (range 1.9–8.1 years) following BHR. In addition, 1 BHR was revised for a traumatic femoral neck fracture and 1 revision was performed for stiffness secondary to heterotopic ossification. There were no early femoral neck fractures, deep infections, or dislocations in the remaining cohort and no patient with a surviving BHR is currently under investigation for ARMD. No patient has undergone any further surgical intervention apart from the six revised hips. All postoperative complications are outlined in Table 2. One patient sustained an intraoperative posterior acetabular wall fracture during cup impaction, though a satisfactory press-fit was still achieved. No fixation was required, the BHR implants were stable, and the fracture subsequently healed with no adverse outcomes. The surviving 92 hips (77 patients) were followed up for a minimum of 10-years (mean 10.8 years; range 10.0 to 14.0 years) from index BHR surgery. Cumulative 10-year survival for the whole cohort (n = 120) with revision for any indication as the endpoint was 94.2% (95% CI 88.8%–98.7%; Fig. 2) and 96.1% (95% CI 91.5%–99.8%) when revision for aseptic loosening was used as the endpoint. Cumulative 10-year survival for males and females was 96.0% (95% CI 90.7%–100%) and 92.5% (95% CI 85.8%–99.9%) respectively. Cox-proportional hazards modelling demonstrated a 53% reduction in risk of revision in males, but neither
gender (P = 0.463) nor femoral component head size (P = 0.114) reached statistical significance. Details of Failures and Revision Surgery Overall six patients required revision of either the femoral and/or acetabular component (Table 3). Three patients required revision for femoral component loosening at 4.5, 4.7, and 8.1 years postoperatively. Radiologically all three patients with femoral loosening showed a typical slow progression of tilting into varus with inferior displacement. The intraoperative findings in each case were consistent with avascular necrosis of the femoral head though this diagnosis was not established after histopathological analysis (Fig. 3A and B). All failed femoral components were revised to cemented stems with modular heads and retention of the BHR acetabular component. The tissues and bone in all cases were healthy with no evidence of metallosis, soft-tissue damage or osteolysis.
Table 2 Postoperative Complications Not Requiring Further Surgery. Complication
Number of Hips
Trochanteric bursitis Deep vein thrombosis Pulmonary embolism (fatal) Superficial wound infection Intraoperative posterior acetabular wall fracture
4 (3.3%) 2 (1.7%) 1 (0.8%) 1 (0.8%) 1 (0.8%)
Fig. 2. Kaplan–Meier survival curve for 120 Birmingham Hip Resurfacings (revision for any reason as the endpoint). Six hips revised in total. Shaded area represents the upper and lower limits of the 95% confidence intervals. Black line represents an implant failure rate of 1% per year.
A. Mehra et al. / The Journal of Arthroplasty 30 (2015) 1160–1166
1163
Table 3 Clinical Details of Birmingham Hip Resurfacings Requiring Revision Surgery (n = 6). Patient Number
Age/Sex Diagnosis
Femoral Head Size (mm)
Time to Revision
Revision Indication
Revision Performed
Outcome after Revision
Traumatic femoral neck fracture
Cemented Exeter stem/Duraloc cementless cup/MoP Cemented CPT stem/modular MMT head/cup retained/MoM
4.5 years
Femoral component loosening
Cemented CPT stem/modular MMT head/cup retained/MoM
50
4.7 years
Femoral component loosening
Cemented CPT stem/modular MMT head/cup retained/MoM
52F OA
46
8.1 years
Femoral component loosening
Cemented CPT stem/modular MMT head/cup retained/MoM
47F Perthes
42
9.1 years
Stiffness secondary to HO
Cemented CPT stem/modular MMT head/cup retained/MoM
No symptoms or adverse radiological features 8-years post-revision. No symptoms or adverse radiological features 1-year post-revision. Subsequently lost to follow-up with no blood metal ions. No symptoms or adverse radiological features 10-years post-revision. Normal ultrasound but high blood metal ions (Co 977 nmol/l and Cr 386 nmol/l). No symptoms or adverse radiological features 9-years post-revision. Normal ultrasound but borderline raised blood metal ions (Co 131 nmol/l and Cr 72 nmol/l). Low grade pain 7-years post-revision for last 6-months. Cup (zone 1) and stem (zone 1 and 7) lucencies. Normal ultrasound with negative aspiration on culture. Blood Co 166 nmol/l and Cr 125 nmol/l. Likely to require re-revision over next 1-year. No symptoms or adverse radiological features 5-years post-revision. Good range of motion but poor abductor function. Awaiting blood metal ion testing.
1
58F OA
42
1.9 years
Acetabular component loosening
2
60F OA
42
2.8 years
3
59M OA
46
4
41M AS
5
6
M = male; F = female; OA = osteoarthritis; MoM = metal-on-metal; MoP = metal-on-polyethylene; AS = ankylosing spondylitis; HO = heterotopic ossification; Co = cobalt; Cr = chromium.
One patient required revision for acetabular component loosening at 1.9 years. There was a failure to gain adequate initial press-fit leading to failure of primary osseo-integration (Fig. 4A). The cup failed by rotation into a more open position with slight proximal migration progressively over a period of 1 year (Fig. 4B). At revision there was no intraoperative evidence of ARMD, and both BHR components were revised. Another patient sustained a traumatic femoral neck fracture at 2.8 years postoperatively and underwent a femoral component revision (cemented femoral stem with a modular head). There was one case of severe heterotopic ossification (Brooker, Grade 4) (Fig. 5A). This required excision of heterotopic bone at 9.1 years postoperatively because of functional restriction, under the care of another surgeon. Although the components were well fixed, during the course of the procedure it was deemed necessary to divide the femoral neck to gain access to the joint requiring revision of the femoral component (Fig. 5B). The trochanteric osteotomy subsequently united. Radiological and Functional Outcomes Radiographs were available for all surviving BHRs at a minimum of 10-years follow-up. The median acetabular inclination and anteversion were 48.1° (IQR 43.8°–52.0°) and 9.7° (IQR 5.2°–15.0°) respectively at latest follow-up with no change in these measurements since initial implantation (Fig. 6). One non-revised BHR demonstrated evidence of radiological femoral loosening around the stem with slight tilt into varus five-years after initial arthroplasty. These appearances have remained stable since and the patient is asymptomatic. None of the other follow-up radiographs demonstrated evidence of component loosening, migration, osteolysis, femoral neck narrowing or notching. Functional outcomes were available for 82 hips (67 patients) at a minimum of 10-years follow-up. The mean patient satisfaction score following BHR was 3.7 (range 2–4), with 95.2% of patients either pleased or extremely pleased after surgery. The mean preoperative HHS was 44.5 (range 28–60), which improved to a mean of 84.0 (range 19–100) at a minimum of 10-years follow-up. Nine patients (10 BHRs) had poor HHSs (less than 70); mean HHS 52 (range 19–67). Two of these patients had trochanteric bursitis. The remaining
7 patients (8 BHRs) had persistent pain with no specific cause for symptoms identified despite thorough investigation (including blood tests for inflammatory markers and metal ion concentrations, radiographs and cross-sectional hip imaging, and spinal imaging). These patients remain under regular clinical review. Outcomes in Patients Lost to Follow-Up Of the 13 hips (12 patients) lost to follow-up within 10-years of BHR, 12 hips (11 patients) underwent clinical review at a mean of 4.4 years (range 0.5–8.3 years) following BHR. At this latest review, all patients had their BHRs in situ, with no post-operative complications or radiological abnormalities, and good HHSs recorded in all but one patient (HHS 67 at 4-years follow-up). All of these patients subsequently moved out of the region, including abroad, therefore further clinical follow-up was not possible despite making several attempts to contact each patient. One patient (1 BHR) lost to follow-up did not attend any of their follow-up appointments after hospital discharge despite multiple attempts made to contact this patient. The outcome in this patient since surgery is therefore unknown. Discussion The present study reports the results for 120 BHRs performed by a single surgeon including his learning curve, with a minimum of 10years follow-up. Good survival (94.2% for all revisions and 96.1% for aseptic loosening) and functional outcomes (mean HHS 84.0 with 95.2% patients satisfied) were observed in a young cohort of patients (mean age 50 years) with no revisions performed for ARMD. These results are comparable to those reported from the designing surgeons (10-year survival of 96.3%–97.4%) [21,22], and compare favourably to those from other independent centres (10-year survival of 87.1%–94.5%) [23–26] and registry data (9-year survival of 91.9%) [17]. Higher failure rates for hip resurfacing have consistently been reported in women. In the Australian Joint Registry the revision rate in women was 2.3 times higher than in men at 7-years [38]. The 10-year survival of BHRs implanted in females at independent centres ranges
1164
A. Mehra et al. / The Journal of Arthroplasty 30 (2015) 1160–1166
Fig. 3. (A) Left BHR demonstrating femoral loosening 2-years postoperatively; (B) Pelvic radiograph following revision of the femoral component only.
from 73.9%–89.1% [23–26,39], which has led some of these authors to recommend against performing hip resurfacing in women [23,25]. Even reports from the designing surgeons [21,22] suggest that results are inferior in females compared to males despite still being well within published recommendations for continued implant usage [40]. Although women have factors which may increase their risk of failure, such as hip dysplasia and the need for smaller femoral component head sizes [15,21], the observation that good results can be obtained in women by designing surgeons suggests that patient selection and operative technique are important factors for success [21,22]. A recent study supported the continued use of the BHR in women provided they have primary osteoarthritis, femoral head anatomy sufficient to allow the use of above a 46 mm sized component, and they meet the other indications for hip resurfacing [21]. The present study findings also support the use of the BHR in women with good survival observed (92.5% at 10-years), though it is recognised that longer term follow-up is required to determine whether these findings continue. This is despite having a high proportion of women (48%) in the cohort (previous studies 26%–41%) [23–26], as well as a higher proportion of patients with femoral head sizes of 46 mm or under (47%) compared to previous studies (up to 39%) [21,24,25]. The commonest causes for hip resurfacing revision recorded in the National Joint Registry of England and Wales were pain (4.78 revisions per 1000 patient-years), ARMD (3.55), and aseptic component loosening (2.92) [17]. In the present cohort no patient underwent revision for pain or ARMD, with aseptic loosening the commonest reason for
Fig. 4. (A) Inadequate initial press fit of the acetabular component; (B) Acetabular component failure 2-years following surgery.
revision (accounting for 4 of 6 revisions). For all revisions performed in this series the bone and soft-tissues were healthy intraoperatively with no signs of metallosis, debris-mediated granuloma, or osteolysis. Revision on the femoral side to a stemmed femoral component was straight forward in all cases, with revision on the acetabular side accomplished with minimal bone loss when using modern instrumentation. However, in light of recent concerns regarding the use of largediameter metal-on-metal stemmed hips [41], revision for femoral component failure at this centre now includes revision of the acetabular component and the use of a non-metal-on-metal bearing. Femoral neck fracture is a unique and usually early complication following hip resurfacing [42]. There were no cases of early femoral neck fracture in the present cohort, with only one traumatic fracture occurring at 3 years postoperatively. In addition, none of the radiographs demonstrated evidence of femoral neck notching from the index surgery. These results compare favourably with registry data (1.57 revisions for fracture per 1000 patient-years) [17] and previous studies which report a prevalence of this complication in up to 2% [25,42]. It is suspected that careful preparation of the femoral head, cementing technique, and positioning of the femoral component are important in reducing the frequency of femoral neck fracture following hip resurfacing [43–45].
A. Mehra et al. / The Journal of Arthroplasty 30 (2015) 1160–1166
1165
Fig. 6. Scatter plot of orientation of the acetabular component (inclination vs. version), estimated using the Lewinnek method [31] with optimal zone (rectangle) for positioning as defined by Grammatopoulos et al [14]. Radiographs of the index hip resurfacing were not available for analysis for one of the six revisions. There were 38 (40%) hips in the optimal zone, 3 of which were revised. There were 58 (60%) hips outside the optimal zone, 2 of which were revised. A Fisher’s exact test shows no significant difference between the proportion being revised within or outside the optimal zones for version and inclination.
Fig. 5. (A) Brooker grade 4 heterotopic ossification 9-years following surgery; (B) Pelvic radiograph 4-years following revision surgery.
Component position is thought to play an important role in the long-term success of hip resurfacing. Evidence suggests that deviation of acetabular inclination and/or version from a safe-zone can lead to increased production of wear debris and subsequent implant failure [12,14,16]. De Haan et al demonstrated that acetabular components with inclination angles greater than 55° combined with small femoral component sizes were likely to generate higher blood metal ion concentrations [16]. The most likely mechanism responsible is decompression of the contact patch necessary for fluid film lubrication between the components resulting in edge loading and increased bearing wear [46,47]. The subtle nuances related to acetabular component position were not fully understood at the time when BHRs in this series were implanted (1997–2001) which may explain some of the positioning variability observed. The position of the acetabular component was in the recommended safe-zone [14] in 40% of BHRs (Fig. 6), with other studies reporting up to 88% of BHR acetabular components were implanted in this safe-zone [24,26]. It is suspected that the results in our series are related to the surgical technique employed during these early cases. The surgeon aimed to achieve 45° of acetabular component inclination, but component anteversion was aligned with the native acetabulum rather than aiming for a pre-determined target. When assessing inclination alone, 88% of patients in this series had components within the safe-zone for inclination. Therefore inadequate anteversion (b 10°) was responsible for most patients with components outside the combined inclination and version safe-zone (Fig. 6). Although presently no adverse outcomes have been observed in the
subgroup with components outside the recommended acetabular safe-zone, these patients will continue to be followed up given the increased potential risk of developing ARMD [14], with a low threshold for performing blood metal ion sampling and/or cross-sectional imaging. We anticipate that by using improved surgical techniques since the importance of cup position in metal-on-metal hip resurfacing was established will result in lower wear and long-term failure rates with the BHRs more recently implanted at this centre. This study has recognised limitations. As it is retrospective this study is subject to the limitations associated with this particular type of study. A proportion of patients were lost to follow-up, whilst some who did undergo clinical and radiographic review at 10-years failed to complete or return their functional outcome questionnaires. Unfortunately this is the reality of performing any long-term outcome study and has been experienced by others reporting 10-year outcomes for the BHR [23–26]. In addition, as BHR patients are young and active there is evidence to suggest that patients initially considered lost to follow-up actually have good outcomes [48], rather than elderly THR cohorts who may have worse outcomes compared to patients that continue to be reviewed [49]. Although outcomes in women following BHR were promising in this cohort compared to those reported by independent centres [23–26,39], it is recognised that the present findings may be influenced by numerous factors. These include the surgeons training and experience with the procedure, patient selection criteria, or even selection bias. Furthermore, although this study presents outcomes at a minimum 10-year follow-up, it is recognised that the cohort is relatively small, therefore the good outcomes observed in women are potentially subject to type II statistical error. This highlights the importance of longer term follow-up of this cohort, as well as validation of these findings by other non-designing surgeon centres. Finally, in line with national recommendations [19] BHR patients were not routinely investigated for evidence of ARMD. Although no BHR patient has been revised or currently is under investigation for ARMD, it is understood that longer-term follow-up is important. Current concerns with metal-on-metal hip resurfacing relate to problems in the short-term, such as fracture, early component loosening, ARMD and undiagnosed pain. The benefits of low-wear in the long-term have yet to be established. Evidence suggests patient selection, implant design, and surgical technique need to be optimised to minimise early failure and ensure good long-term results. The present single surgeon series confirms good clinical outcomes continue into
1166
A. Mehra et al. / The Journal of Arthroplasty 30 (2015) 1160–1166
the second decade for the BHR. Although our findings suggest that good outcomes in women can be achieved with the BHR, longer term followup is required to confirm whether these promising results continue. In our view, the BHR remains the procedure of choice for eligible young active patients who are likely to require revision during their lifetime. Acknowledgment We wish to acknowledge Mr. Eric Robinson at the Oswestry Outcome Centre for his help in collecting the functional outcome scores. The authors would also like to thank The Royal College of Surgeons of England and the Arthritis Research Trust for providing one of the authors with funding in the form of a Surgical Research Fellowship. References 1. Callaghan JJ, Forest EE, Olejniczak JP, et al. Charnley total hip arthroplasty in patients less than fifty years old: a twenty-five-year follow-up note. J Bone Joint Surg Am 1998;80-A:704. 2. Mäkelä KT, Eskelinen A, Pulkkinen P, et al. Results of 3,668 primary total hip replacements for primary osteoarthritis in patients under the age of 55 years. Acta Orthop 2011;82:521. 3. Fisher J, Jin Z, Tipper J, et al. Tribology of alternative bearings. Clin Orthop Relat Res 2006;453:25. 4. Charnley J. Arthroplasty of the hip. A new operation. Lancet 1961;1:1129. 5. Freeman MA, Bradley GW. ICLH surface replacement of the hip. An analysis of the first 10 years. J Bone Joint Surg (Br) 1983;65:405. 6. Wagner H. Surface replacement arthroplasty of the hip. Clin Orthop Relat Res 1978; 134:102. 7. Amstutz HC, Clarke IC, Christie J, et al. Total hip articular replacement by internal eccentric shells: the “tharies” approach to total surface replacement arthroplasty. Clin Orthop Relat Res 1977;128:261. 8. Hing C, Back D, Shimmin A. Hip resurfacing: indications, results and conclusions. Instr Course Lect 2007;56:171. 9. Ball ST, Le Duff MJ, Amstutz HC. Early results of conversion of a failed femoral component in hip resurfacing arthroplasty. J Bone Joint Surg Am 2007;89-A:735. 10. Pandit H, Glyn-Jones S, McLardy-Smith P, et al. Pseudotumors associated with metalon-metal hip resurfacings. J Bone Joint Surg (Br) 2008;90-B:847. 11. Langton DJ, Jameson SS, Joyce TJ, et al. Early failure of metal-on-metal bearings in hip resurfacing and larger-diameter total hip replacement: a consequence of excess wear. J Bone Joint Surg (Br) 2010;92-B:38. 12. Langton DJ, Joyce TJ, Jameson SS, et al. Adverse reaction to metal debris following hip resurfacing: the influence of component type, orientation and volumetric wear. J Bone Joint Surg (Br) 2011;93-B:164. 13. Haddad FS, Thakrar RR, Hart AJ, et al. Metal-on-metal bearings: the evidence so far. J Bone Joint Surg (Br) 2011;93-B:572. 14. Grammatopoulos G, Pandit H, Glyn-Jones S, et al. Optimal acetabular orientation for hip resurfacing. J Bone Joint Surg (Br) 2010;92-B:1072. 15. Glyn-Jones S, Pandit H, Kwon YM, et al. Risk factors for inflammatory pseudotumour formation following hip resurfacing. J Bone Joint Surg (Br) 2009;91-B:1566. 16. De Haan R, Pattyn C, Gill HS, et al. Correlation between inclination of the acetabular component and metal ion levels in metal-on-metal hip resurfacing replacement. J Bone Joint Surg (Br) 2008;90-B:1291. 17. National Joint Registry for England and Wales. 10th Annual Report. Available from URL http://njrcentre.org.uk; 2013. [last accessed on 20th January 2015]. 18. Medical and Healthcare Products Regulatory Agency (MHRA). Medical Device Alert: ASR™ hip replacement implant manufactured by DePuy International Ltd. MDA/ 2010/069. Available from URL http://www.mhra.gov.uk/; 2010. [last accessed on 20th January 2015]. 19. Medical and Healthcare Products Regulatory Agency (MHRA). Medical Device Alert: all metal-on-metal (MoM) hip replacements. MDA/2012/036. Available from URL http://www.mhra.gov.uk/; 2012. [last accessed on 20th January 2015]. 20. Birmingham Hip Resurfacing System. Available from URL http://www.smith-nephew.com/professional/products/all-products/bhr-birmingham-hip-resurfacing/. [last accessed on 20th January 2015].
21. Matharu GS, McBryde CW, Pynsent WB, et al. The outcome of the Birmingham Hip Resurfacing in patients aged b50 years up to 14 years post-operatively. Bone Joint J 2013;95-B:1172. 22. Daniel J, Pradhan C, Ziaee H, et al. Results of Birmingham hip resurfacing at 12 to 15 years. Bone Joint J 2014;96-B:1298. 23. Coulter G, Young DA, Dalziel RE, et al. Birmingham hip resurfacing at a mean of ten years: results from an independent centre. J Bone Joint Surg (Br) 2012; 94-B:315. 24. Holland JP, Langton DJ, Hashmi M. Ten-year clinical, radiological and metal ion analysis of the Birmingham Hip Resurfacing: from a single, non-designer surgeon. J Bone Joint Surg (Br) 2012;94-B:471. 25. Murray DW, Grammatopoulos G, Pandit H, et al. The ten-year survival of the Birmingham hip resurfacing: an independent series. J Bone Joint Surg (Br) 2012;94-B:1180. 26. Van Der Straeten C, Van Quickenborne D, De Roest B, et al. Metal ion levels from wellfunctioning Birmingham Hip Resurfacings decline significantly at ten years. Bone Joint J 2013;95-B:1332. 27. Daniel J, Pradhan C, Ziaee H. Patient selection and timing of operation. In: McMinn DJW, editor. Modern Hip Resurfacing. 1st ed. London: Springer; 2009. p. 163. 28. McMinn DJW. Patient positioning and exposure. In: McMinn DJW, editor. Modern Hip Resurfacing. 1st ed. London: Springer; 2009. p. 189. 29. Gilbert RE, Cheung G, Carrothers AD, et al. Functional results of isolated femoral revision of hip resurfacing arthroplasty. J Bone Joint Surg Am 2010;92-A:1600. 30. Rasband WS, Image J. US National Institutes of Health, Bethesda, Maryland, USA. http://imagej.nih.gov/ij/. [1997–2014. Last accessed on 20th January 2015]. 31. Lewinnek GE, Lewis JL, Terr R, et al. Dislocation after total hip arthroplasties. J Bone Joint Surg Am 1978;60-A:217. 32. Nho JH, Lee YK, Kim HJ, et al. Reliability and validity of measuring version of the acetabular component. J Bone Joint Surg (Br) 2012;94-B:32. 33. Amstutz HC, Beaule PE, Dorey FJ, et al. Metal on metal hybrid surface arthroplasty: two to six year follow up study. J Bone Joint Surg Am 2004;86-A:28. 34. DeLee JG, Charnley J. Radiological demarcation of cemented sockets in total hip replacement. Clin Orthop Relat Res 1976;121:20. 35. Hing CB, Young DA, Dalziel RE, et al. Narrowing of the neck in resurfacing arthroplasty of the hip: a radiological study. J Bone Joint Surg (Br) 2007;89-B:1019. 36. Mahomed NN, Arndt DC, McGrory BJ, et al. The Harris hip score: comparison of patient self-report with surgeon assessment. J Arthroplasty 2001;16:575. 37. R Development Core Team. R: A Language and Environment for Statistical Computing. Vienna, Austria: the R Foundation for Statistical Computing. Available from URL http://www.R-project.org/. [last accessed on 20th January 2015]. 38. Australian National Joint Registry Annual Report 2013. Available from URL http://aoanjrr. dmac.adelaide.edu.au/annual-reports-2013. [last accessed on 20th January 2015]. 39. Reito A, Puolakka T, Elo P, et al. Outcome of Birmingham hip resurfacing at ten years: role of routine whole blood metal ion measurements in screening for pseudotumours. Int Orthop 2014;38:2251. 40. Guidance in the Selection of Prostheses for Primary Total Hip Replacement. Technology Appraisal Guidance — No. 2. National Institute for Health and Clinical Excellence. Available from URL http://www.nice.org.uk; 2003. [last accessed on 20th January 2015]. 41. Smith AJ, Dieppe P, Vernon K, et al, National Joint Registry of England and Wales. Failure rates of stemmed metal-on-metal hip replacements: analysis of data from the National Joint Registry for England and Wales. Lancet 2012;379:1199. 42. Shimmin AJ, Back D. Femoral neck fractures following Birmingham hip resurfacing. J Bone Joint Surg (Br) 2005;87-B:463. 43. Davis ET, Olsen M, Zdero R, et al. A biomechanical and finite element analysis of femoral neck notching during hip resurfacing. J Biomech Eng 2009;131:041002. 44. Beaule PE, Matar WY, Poitras P, et al. 2008 Otto Aufranc Award: component design and technique affect cement penetration in hip resurfacing. Clin Orthop Relat Res 2009;467:84. 45. Davis ET, Olsen M, Zdero R, et al. Femoral neck fracture following hip resurfacing: the effect of alignment of the femoral component. J Bone Joint Surg (Br) 2008;90-B:1522. 46. Kwon YM, Glyn-Jones S, Simpson DJ, et al. Analysis of wear of retrieved metal-onmetal hip resurfacing implants revised due to pseudotumours. J Bone Joint Surg (Br) 2010;92-B:356. 47. Langton DJ, Jameson SS, Joyce TJ, et al. The effect of component size and orientation on the concentrations of metal ions after resurfacing arthroplasty of the hip. J Bone Joint Surg (Br) 2008;90-B:1143. 48. Matharu GS, McBryde CW, Treacy RB, et al. Impact of active patient follow-up on worst-case implant survival analysis. Hip Int 2013;23:259. 49. Murray DW, Britton AR, Bulstrode CJ. Loss to follow-up matters. J Bone Joint Surg (Br) 1997;79-B:254.
