Our aim in this pilot study was to evaluate the fixation of, the bone remodelling around, and the clinical outcome after surgery of a new, uncemented, fully hydroxyapatite-coated, collared and tapered femoral component, designed specifically for elderly patients with a fracture of the femoral neck.
We enrolled 50 patients, of at least 70 years of age, with an acute displaced fracture of the femoral neck in this prospective single-series study. They received a total hip replacement using the new component and were followed up regularly for two years.
Fixation was evaluated by radiostereometric analysis and bone remodelling by dual-energy x-ray absorptiometry. Hip function and the health-related quality of life were assessed using the Harris hip score and the EuroQol-5D.
Up to six weeks post-operatively there was a mean subsidence of 0.2 mm (−2.1 to +0.5) and a retroversion of a mean of 1.2° (−8.2° to +1.5°). No component migrated after three months. The patients had a continuous loss of peri-prosthetic bone which amounted to a mean of 16% (−49% to +10%) at two years. The mean Harris hip score was 82 (51 to 100) after two years.
The two-year results from this pilot study indicate that this new, uncemented femoral component can be used for elderly patients with osteoporotic fractures of the femoral neck.
There is good evidence in support of using a primary hip replacement instead of internal fixation for a displaced fracture of the femoral neck in elderly patients who are mobile and have no severe cognitive impairment.1–5
Cemented femoral components are used both in total hip replacement (THR) and hemi-arthroplasty, and earlier studies have supported their use in patients with a fracture of the femoral neck, mainly because of decreased postoperative pain.6–8 However, the concept of inserting an uncemented femoral component, even in elderly patients, is attractive to many surgeons9 since the cementing process can induce cardiac arrhythmia and cardio-respiratory collapse.10 The mechanisms involved are not yet fully understood, but probably include pulmonary embolisation of bone marrow and particles of methylmethacrylate.11,12 The mortality rate in these elderly patients may therefore be higher after cemented rather than uncemented replacement.13,14 The potential advantage of using an uncemented femoral component is also related to the shorter duration of surgery, thereby reducing intra-operative bleeding and the risk of infection.6 The disadvantages include an increased risk of peri-prosthetic fracture, thigh pain and stress-shielding of the proximal femur.
A new uncemented femoral component (Biomet Fracture Stem; Biomet UK Ltd, Bridgend, United Kingdom) based on a successful earlier design,15,16 with a fully coated hydroxyapatite (HA) stem to induce fast ingrowth by osteoporotic bone17 and with a collar to resist excessive subsidence, has been developed specifically for the treatment of fractures of the femoral neck. Our aim in this pilot study was to evaluate the fixation of and bone remodelling around the new component in a representative sample of elderly patients with a displaced fracture of the femoral neck, using radiostereometric analysis (RSA)18,19 and dual-energy x-ray absorptiometry (DXA).20–22
Patients and Methods
This prospective, single-series study was carried out at the Orthopaedic Department of Danderyd Hospital, Stockholm, Sweden and ethical approval was obtained.
Between October 2005 and April 2008, 229 patients with a fracture of the femoral neck sustained less than 24 hours before admission were screened for participation in the study. The inclusion criteria were a displaced fracture (Garden type III or IV),23 age of at least 70 years, intact cognitive function (at least eight correct answers on a ten-item Short Portable Mental Status Questionnaire (SPMSQ)),24 the previous ability to walk independently with or without walking aids and a willingness to participate in the study. We excluded patients with a previous fracture in the same hip or a pathological fracture, those deemed not to be suitable for THR by the anaesthetist25 and those who for any other reason were unsuitable for participation in the trial (Fig. 1⇓). A research nurse gave the patients oral and written information about the study and written informed consent was obtained. The clinical details of the 50 patients are given in Table I⇓.
The Biomet Fracture Stem is a tapered, collared stem intended for uncemented fixation (Fig. 2⇓). It is made of a titanium alloy (Ti-6Al-4V) with a grit-blasted surface roughness of 7.5 μm to 10.0 μm with a straight 3° proximal-to-distal taper in two planes and a taper from the lateral shoulder to the medial calcar area. Its geometry, except for the collar, is identical to the Bi-Metric stem (Biomet UK Ltd).15,16 It has plasma-sprayed HA on the entire surface (thickness 65 μm to 95 μm, crystallinity 50% to 70%, purity > 95%) to promote fast ingrowth by osteoporotic bone26 and it is available in six sizes (7 mm to 17 mm, uneven sizes only), all with a neck-shaft angle of 140°.
Pre-operative planning was performed using digital software (mDesk; RSA Biomedical AB, Umeå, Sweden). A modular 32 mm cobalt-chrome femoral head was used in all patients and a cemented acetabular component (ZCA; Zimmer, Warsaw, Indiana). The operations were all done within the first 48 hours after admission by one of four surgeons (OGS, MOS, HSB, or TEA). A posterior approach was used, with repair of the posterior capsule and external rotator muscles. The femoral neck was resected using a template and the femur was reamed with reamers until contact with cortical bone was obtained. Thereafter, the proximal femur was prepared with broaches of increasing size until rotational stability was achieved, and leaving the final broach in place, the calcar was planed flush. Thus, the collar of the component rested on the calcar when it was fully seated and rotationally stable. In some hips, the fracture line was more distal than the ideal level for resection. In these hips, the collar did not rest on bone when the stem was rotationally stable so there was no collar-calcar contact. Before the final implant was inserted, between five and nine tantalum beads of 1.0 mm in diameter were inserted in the cancellous bone of the proximal femur. All the patients received intravenous tranexamic acid (Cyclokapron; Pfizer, Sollentuna, Sweden) pre-operatively to reduce bleeding. Prophylactic antibiotics (Cloxacillin, 2 g intravenously; Meda, Solna, Sweden) were given peri-operatively and in the first 24 hours postoperatively and dalteparin (Fragmin, 5000 units subcutaneously; AstraZeneca, Södertälje, Sweden) was given for ten days post-operatively as thromboprophylaxis. The patients were mobilised fully weight-bearing with crutches.
The primary endpoint of the study was migration of the femoral component. Secondary endpoints included changes in bone mineral density (BMD) in the zones of Gruen, McNeice and Amstutz27 and the clinical outcome. The study protocol is shown in Figure 1⇑.
Radiostereometric analysis (RSA).
This is a high-precision method of assessing three-dimensional (3D) micro-movement from calibrated stereoradiographs18 and is a standard technique for evaluating new implants since early migration can predict loosening.28 The RSA method in our study followed published guidelines.29 We took digitial calibrated radiographs (Bucky Diagnostic; Philips, Eindhoven, The Netherlands) using one fixed and one mobile Roentgen source (120 kV, 4 mAs to 6 mAs), and a uniplanar calibration cage (Uniplanar digital 43; RSA Biomedical AB). All the data were analysed using the UmRSA software (RSA Biomedical AB). The markers in the proximal femur form one segment and the centre of the prosthetic head, in combination with the tantalum marker beads in the stem, forms another. The 3D translations and rotations of the calculated centre of gravity of the stem in relation to the femoral bone segment were calculated at each follow-up visit and compared with the immediate post-operative measurements. We also measured the maximum total point movement (MTPM), which is the 3D translation vector of the marker in the femoral stem that has the largest movement and is seen as an indicator of the overall magnitude of migration. At 12 months, we performed two examinations 15 minutes apart on 25 patients with complete repositioning of the X-ray tubes and the calibration cage. We calculated the precision as the 99% confidence interval (CI) (sd 2.7) of the difference between these examinations. For translation along the x-(transverse), y-(vertical) and z-(anteroposterior (AP)) axes, this was 0.27, 0.19 and 0.52 mm, respectively. For rotation about the x-(flexion/extension), y-(ante-/retroversion) and z-(varus/valgus) axes, the values were 0.52°, 0.76° and 0.27°, respectively, and for the MTPM it was 0.74 mm. The precision for our RSA setting was similar to that previously reported.17,30 The mean error of rigid body fitting31 was used to evaluate the stability of the markers over time. We excluded examinations in which this value was > 0.3 mm because this indicated migration of the markers. The condition number31 is used to evaluate the distribution of the markers and a high value precludes accurate measurements of z-translation as well as segment rotation and MTPM. Therefore, in examinations in which the condition number exceeded 100, only transverse (x) and vertical (y) translations were calculated.
Bone mineral density (BMD) and radiological evaluation.
The BMD of the peri-prosthetic femur was measured over the seven Gruen zones27 in the frontal plane using DXA (DPX-L; Lunar Co., Madison, Wisconsin). During scanning, the patient was placed in the supine position with standard knee and foot supports with the femur in neutral rotation. The change in the peri-prosthetic BMD ratio in all zones, as well as the entire peri-prosthetic region (zones 1 to 7), was calculated by dividing the BMD value at each follow-up visit by the post-operative BMD and converting it to a percentage change. We had previously made two sequential measurements in ten patients with complete repositioning of the patients and the scanner.32 Postoperatively, we measured the BMD of the proximal femur of the healthy hip (World Health Organization (WHO) definition of the total hip33) for patients in Charnley class A,34 and vertebrae L1 to L4 (WHO lumbar spine33) of all patients, to assess their general bone mass. The BMD of the vertebrae was also measured at 24 months post-operatively.
Varus/valgus angle and fill35 of the stem in the femoral canal were measured on radiographs using the digital templating software. The proximal fill was measured at the upper border of the lesser trochanter and the distal fill 3 cm proximally from the tip of the component. Fill was defined as good when there was an 80% fill on the AP radiograph and 70% fill on the lateral radiograph.35 At 24 months, the presence of heterotopic ossification was evaluated according to the criteria of Brooker et al.36
Hip function was evaluated using the Harris hip score (HHS).37 This score has been validated for patients with fractures of the femoral neck.38 Health-related quality of life was assessed by the EQ-5D (EuroQoL),39 which uses five dimensions: mobility, self-care, usual activity, pain/discomfort and anxiety/depression. Each dimension is divided into three levels as follows: 1, no problems; 2, some problems; and 3, extreme problems. This generates 243 different ‘health states’ and the EQ-5D index score assigns each ‘health state’ to a value, ranging from −0.59, indicating the worst possible state of health, to 1, indicating full health. In this score, it is possible to have negative values, i.e., a perceived health state worse than death.
Pain from the hip was recorded using the Pain Numeric Rating Scale (PNRS),40 which is an 11-point (0 to 10) scale, in which 0 denotes no pain and 10 unbearable pain. After inclusion, but before surgery, we asked each patient to estimate their HHS, EQ-5D and PNRS during the previous week. The clinical outcome was also recorded at the follow-up visits (Fig. 1⇑).
With 20 patients, the study had a power (two-sided, p = 0.01) of more than 99% and 93% to detect a continuous migration in the MTPM and y-translation, respectively. These estimates were based upon a previous RSA study with the HA-coated version of the Bi-Metric stem17 in which the mean migration was 1.9 mm (sd 1.3) and the y-translation 0.2 mm (sd 0.2). We recruited 50 patients to allow for loss to follow-up and to allow for analysis of subgroups of patients with high and low BMD.
We used analysis of covariance (ANCOVA) to study covariates affecting migration of the implant and bone loss. First, in order to evaluate the effect of peri-prosthetic BMD on migration and bone loss, we used the MTPM and the change in BMD in zones 1 to 7 as dependent variables and gender, age, body mass index (BMI), stem size and immediate post-operative BMD in zones 1 to 7 as covariates. The median BMD (1.51 g/cm2) was used to divide the patients into two groups, those with high or low BMD (high vs low: mean BMD 1.77 g/cm2 (sd 0.33) and 1.35 g/cm2 (sd 0.11), respectively).
In the second analysis, we evaluated the effect of the preoperative BMD on bone loss in a subgroup of 36 patients. These patients had a healthy contralateral femur at inclusion and we had complete data at follow-up of 24 months. These patients were all in Charnley class A or C,34 but did not differ from the rest of the patients in regard to anthropometrical data. Post-operative DXA scans of the healthy hip and vertebrae L1 to L4 (WHO total hip and lumbar spine) were used as a proxy for the pre-operative BMD and were categorised by the T-score.33 This represents the number of standard deviations above or below the mean for a healthy 30-year-old adult of the same sex and ethnicity as the patient. Our patients were categorised as normal (T-score > −1 sd), osteopenic (−1 sd ≥ T-score > −2.5 sd) and osteoporotic (T-score ≤ −2.5 sd).33 In the analysis, the change in BMD in zones 1 to 7 was then used as the dependent variable and gender, age, BMI, stem size and T-score category of the hip or lumbar spine as covariates.
Between-group comparisons of continuous variables at follow-up were analysed by the Mann-Whitney U test and within-group comparisons between baseline and follow-up values by the Wilcoxon signed-rank test. We used the chi-squared test for nominal variables. A p-value ≤ 0.05 was considered to be significant. We used the statistical software PASW Statistics 18.0 for Windows (IBM Corporation, Armonk, New York).
The mean operating time was 81 minutes (59 to 140) and the mean intra-operative blood loss was 400 ml (100 to 1400). A total of 13 patients required one to two and three needed three to four units of blood transfusions. The mean length of stay was eight days (5 to 11), after which 17 patients returned home and 33 went to a rehabilitation centre.
Two femoral components were placed in 3° of varus and all the others were in neutral alignment. The fill of the femoral canal by the component was classified as good in 17 hips (34%) and there was collar-calcar contact in 33 hips (66%). At 24 months, 29 hips had no evidence of heterotopic ossification whereas 15 had class I to class II and six class III to class IV ossification.36
A total of 30 femoral components migrated beyond the detection limit within six weeks postoperatively and four had migrated further at a follow-up of three months, after which all had stabilised. The mean initial translation in the x-, y- and z-axes was −0.03 mm (sd 0.27), −0.16 mm (sd 0.48) and −0.31 mm (sd 0.60), respectively, with a corresponding rotation of −0.16° (sd 0.53), −1.16° (sd 1.88) and 0.01° (sd 0.62) at six weeks. The mean MTPM was 1.83 mm (sd 1.34) at six weeks. After this, the mean migration stopped except for rotation in flexion/extension (x), which indicated a small but statistically significant continuous migration between 12 and 24 months (Table II⇓, Fig. 3⇓). The RSA data are missing for the hip which was revised as the result of a deep infection. In one dislocating hip, the femoral component retroverted 15° and had an MTPM of 13 mm at six weeks. The component has been stable on all subsequent examinations, but because of the large initial migration, it was treated as an outlier and not included in the RSA analysis. Migration was significantly more pronounced in patients with a low peri-prosthetic BMD (Fig. 4⇓). At all the follow-ups, the MTPM and bone loss were found to be significantly higher in patients with a low BMD after controlling for gender, age, BMI and stem size (ANCOVA). The bone loss, but not the migration, was continuous between 12 and 24 months for patients with low BMD (t-test, p = 0.001).
There was a continuous decrease in the peri-prosthetic BMD in all zones except zone 4, with the highest rate of bone loss occurring during the first 12 months, in all zones except zone 4, with a mean reduction of total peri-prosthetic BMD of 16% (− 49% to + 10%) at 24 months (Table III⇓, Fig. 5⇓). The bone loss was greatest in zones 1 and 7, with decreases of 30% (− 61% to − 6%) and 26% (− 58% to + 25%), respectively, at 24 months. It was significantly related to both the initial post-operative BMD and the patients’ general bone mass (Figs 4⇑ and 6⇓). The mean BMD of vertebrae L1 to L4 at 24 months did not differ from that immediately post-operatively. Osteopenia and osteoporosis in the hip and lumbar spine were significantly related to peri-prosthetic bone loss after controlling for gender, age, BMI and stem size (total hip: p = 0.015 and p < 0.001, respectively; vertebrae L1 to L4: p = 0.021 and p = 0.027, respectively; ANCOVA).
There was a slight deterioration in function and an increased pain in the operated hip during the period of study (mean HHS and mean PNRS pre-fracture vs 24 months post-operatively; 87 (sd 11) and 0.4 (sd 1.3) vs 82 (sd 13) and 1.0 (sd 1.9) (Wilcoxon signed-rank test, p = 0.006 and p = 0.033, respectively) (Table IV⇓). The outcome was worse for patients in Charnley class C34 in whom degenerative disease in other joints and/or associated medical comorbidities affected the outcome. The mean HHS was 88 (100 to 52), 90 (91 to 80) and 76 (88 to 51) in Charnley classes A, B and C,34 respectively.
Health-related quality of life also declined during the study, but did not reach statistical significance (mean EQ-5D pre-fracture vs 24 months post-operatively; 0.71 (sd 0.23) vs 0.63 (sd 0.37), t-test, p = 0.112; Table IV⇑).
One femoral component was revised because of deep infection. It was removed and another uncemented component was implanted three weeks postoperatively. The infection has since resolved uneventfully. There was one intra-operative fracture of the greater trochanter which was reattached at the time. Post-operatively, in the same patient, an undisplaced femoral fracture at the tip of the femoral component was discovered, which was treated by protected weight-bearing for six weeks. At 18 months she presented with a deep infection which was treated by debridement and with antibiotics. Revision of the acetabular component was subsequently required for dislocation. The femoral component was not replaced and the RSA results showed that there has been no migration and clinical function was excellent. In all, seven hips (14%) dislocated, three of which had recurrent dislocation, requiring revision of the acetabular component. Two patients received antibiotics for a superficial wound infection.
In a cognitively intact series of elderly patients with a displaced fracture of the femoral neck treated by the insertion of a new HA-coated femoral component, all the components were stable after three months. The mean subsidence, retroversion and total migration were of the same, or smaller, magnitude than those reported in other RSA studies of the clinically successful use of uncemented femoral components.17,41 We also found a continuous decrease in the BMD around the components for up to two years post-operatively.
The femoral component was designed to be easy to use and is marketed with a unipolar or bipolar hemiarticulation. In clinical practice, a hemiarthroplasty is still the most common procedure for a displaced fracture of the femoral neck in elderly patients despite recent evidence that a THR gives better hip function.42 In our trial, we wished to exclude the possibility of acetabular erosion and the pain associated with this, and therefore used THR in all the patients.
A stepwise clinical introduction of new implants has been advocated.43 This involves pre-clinical testing, small prospective trials using high-precision methods such as RSA18 to assess fixation, larger multicentre trials and finally population-based register studies. In this trial, we have used two validated methods for the determination of the fixation of the implant and bone remodelling, namely RSA and DXA. To our knowledge, there is no other study which has used this combination to evaluate a new implant in osteoporotic patients with a fracture of the femoral neck, but our the results should be interpreted with caution since we have no control group. Ours was designed as a pilot study, and the next step will be to conduct a large randomised, controlled trial in which this implant is compared with a cemented component.
In order to assess the influence of the fracture on the clinical outcome scores, the patients were asked to report their pre-fracture status when they were enrolled in the study. It is obvious that the patients’ ability to record this correctly while awaiting urgent surgery may be questioned. It is, however, impossible to collect these data in a prospective manner and the method is regularly used in patients with a fracture of the hip.1,2,4 There is also indirect evidence that this method yields correct pre-fracture scores. When comparing our pre-fracture EQ-5D score with that of an age-matched Swedish reference population the estimated quality of life was similar (study EQ-5D versus Swedish reference age 80 to 88 years44; 0.71 vs 0.74), indicating that the patients had at least not overestimated their pre-fracture status.45
Our rate of dislocation was high despite using posterior repair and 32 mm heads in all cases. A high dislocation rate for the posterolateral approach after THR in such patients has been reported,45 and after the completion of the current trial we have subsequently reduced the dislocation rate by adopting the anterolateral approach.46
The use of uncemented components for THR is popular. In a study on younger patients with osteoarthritis, some performed as well, or better, than cemented components at follow-up at ten years.47 The rationale for using these devices for displaced fractures of the femoral neck in osteoporotic elderly patients, often with a stove-pipe femur, is mainly theoretical. During pressurisation, cement and fat embolism may occur,11,12 and can have an impact on mortality.13,14
There has been some evidence of an improved functional outcome in patients with a hip fracture when an uncemented Austin-Moore hemiarthroplasty was compared with a cemented Thomson component.48 However, a recent study has shown that modern uncemented HA-coated and cemented components gave similar results.49 Nevertheless, we acknowledge that there are several potential problems when using an uncemented component in this age group, particularly the increased risk of peri-prosthetic fracture.
In the latest report from the Swedish Hip Arthroplasty Register, modern HA-coated uncemented femoral components used for hemiarthroplasty in this elderly population were compared with modern cemented components and were associated with an increased risk of revision because of femoral fracture (odds ratio 3.8, 95% CI 2.0 to 7.1).50 Similar results were seen in the Australian Arthroplasty register in which uncemented femoral components in fracture patients had a significantly higher risk of revision compared with cemented components.51
We only had one fracture; a femoral crack distal to the tip of the component. We have not had any femoral split fractures or calcar fractures. However, all the surgeons who performed the operations in our study were experienced hip surgeons. The risk of femoral split fractures is perhaps also implant-specific. The most commonly used uncemented femoral component for fractures of the femoral neck in Sweden is a press-fit type (Corail; DePuy, Warsaw, Indiana), in which the technique is to use an increasing size of broach and to impact as much cancellous bone as possible.
Therefore, despite the fact that an HA-coated femoral component can work well in osteoporotic patients, there is still little evidence to recommend it. In fact, the rate of calcar split/peri-prosthetic fractures in this group of patients in published reports ranges between 0% and 17% and is typically around 4% (Table V⇓), a risk that is perhaps unacceptable when compared with the low risk of adverse events during cementation of a component.
The Biomet Fracture Stem is a modification of the original Bi-Metric design. It is marketed in uneven sizes only, has a single offset option and a collar. The first two factors are unnecessary limitations, particularly in respect to men, since more than one offset option is lacking and the high dislocation rate seen in our study could, in part, be attributed to this fact.
We cannot prove that the collar is either an advantage or disadvantage regarding migration. We believe, however, that it makes insertion of the implant safer by reducing the risk of causing split fractures of the calcar. The full HA coating is probably also advantageous in this group since it has strong osteoconductive properties. Despite the fact that we only attained a tight fill in the femoral canal in about one-third of the hips, all the stems stopped migrating after three months.
In RSA studies, the amount of migration which increases the risk of revision varies between implants and the method of fixation. For the cemented Lubinus SP I stem (Link, Hamburg, Germany), subsidence exceeding 1.2 mm during the first post-operative year was associated with a risk of revision of 50% within five to seven years.28 In those with uncemented femoral components, there is evidence that subsidence should be less than 1.0 mm to 1.5 mm and retroversion less than 3° during the first year in order to avoid revision surgery.52
HA coating has been advocated for uncemented femoral components since it decreases migration. In one study on patients with osteoarthritis, the MTPM of the Bi-Metric stem was 1.7 mm (HA coating) and 3.9 mm (no HA coating) after 12 months.17 The subsidence (0.2 mm at 12 months) did not differ in the groups, but the retroversion was smaller in the HA group. However, the Bi-Metric stem, with and without HA coating, has excellent long-term results, and therefore this small difference in migration is not of clinical relevance.53
Migration of femoral components in elderly patients with a fracture of the femoral neck has only been investigated once before. Klestil et al54 evaluated the migration of the uncemented Zweymüller femoral component in 23 such patients using Einzel-Bild-Roentgen-Analyse Femoral Component Analysis55 and found that active patients had a significant migration; 30% of all stems subsided by at least 2 mm. Whether this migration was continuous was not evident from the study. Although none required revision and there were no cases of calcar split fracture they did not recommend the use of this component for this group of patients because of high migration and a fear of early loosening.
In our study, we observed that some patients had a large initial migration, at most 13 mm in the MTPM (Table II⇑). Despite this, all the components stabilised. However, there was a non-significant tendency for continuous migration at 24 months, which could be attributed to local bone resorption. Stress-related bone loss around femoral components has been extensively studied,20–22 and is also related to the size of the component and, as in our investigation, the initial BMD of the patients.56–58 Investigation of bone remodelling in this group has, to our knowledge, only been done once before involving a new acetabular component.59 The remodelling around the femoral component in our study can therefore serve as a reference for other implants used in this age group. Our patients lost a significant amount of bone, but, up to two years, this had no obvious effect on the stability of the implant. However, we cannot rule out a negative effect of this bone loss in the longer term. We measured the loss of bone in the unoperated hip in approximately half of our patients, but there was no bone loss in these cases, suggesting that it was not due to a decrease in physical activity after fracture, for example. Recently, electrochemically-deposited HA, compared with plasma-sprayed HA as used in our study, has been found to yield a higher BMD in zone 1 in osteo-arthritic patients treated using an uncemented press-fit femoral component.60 A randomised trial would be required to investigate if this bone-sparing effect also occurs in patients with a fracture of the femoral neck.
We found a difference in migration of the component between patients with high or low peri-prosthetic BMD (Fig. 4⇑). Proximal femoral BMD was the strongest predictor for migration. Since the patients who were osteopenic or osteoporotic before the fracture, as estimated by the BMD of the contralateral femur, had also lost more proximal femoral bone at two years, this situation could lead to loosening of the component or predispose to a fracture.
Most data on modern uncemented femoral components (not Austin-Moore or Thompson types) in patients with a fracture of the femoral neck are derived from single series studies54,61–65 and there are only two trials in which HA-coated components were compared with other types (Table V⇑).49,66 Livesley et al66 compared an HA-coated Furlong hemiarthroplasty (JRI) with the conventional uncemented Austin-Moore hemiarthroplasty and found better functional results in the JRI group. However, the trial was not randomised and patients were allocated to treatment by week of admission. They also compared the HA-coated implant against an implant which we now know to be inferior. According to the Swedish Hip Arthroplasty Register the revision rate for the Austin-Moore hemiarthroplasty is high and its use is no longer recommended.67 Figved et al49 have performed the only randomised study which compared a modern cemented hemiarthroplasty (Spectron; Smith & Nephew, Memphis, Tennessee) with an HA-coated press-fit hemiarthroplasty (Corail). Their study showed similar HHS values in the two groups after one year. Patient selection was similar to that in our study, with a mean age of 83 years, and consisting of predominantly women. The hip function and health-related quality of life were similar to our results. They had more intra- and post-operative peri-prosthetic fractures in the uncemented group, compared with the cemented group (6% vs 2% after one year). They also excluded four patients during surgery because they could not achieve primary stability with uncemented fixation. Despite this, they concluded that both femoral components in the study could be used with good results in displaced fractures of the femoral neck.
In conclusion, the short-time results from our pilot study indicate that this new uncemented femoral component can be used for elderly patients with osteoporotic fractures of the femoral neck. Additional research should, in the context of multicentre, randomised trials, focus on comparing the outcome after using uncemented HA-coated femoral components and cemented components in the treatment of patients with a fracture of the femoral neck.
The authors would like to thank the research nurses H. Sjöö and P. Kelly-Pettersson for their invaluable help with inclusion of the patients and the clinical evaluations, and H.-J. Lundberg, Department of Nuclear Medicine, Danderyd Hospital for his technical assistance with the DXA analyses and E. Hofmann and L.-L. Widmark for all their help with the RSA examinations.
No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.
- Received October 15, 2010.
- Accepted December 9, 2010.
- © 2011 British Editorial Society of Bone and Joint Surgery