|Home | About | Journals | Submit | Contact Us | Français|
Language: | French
The purpose of this study was to perform an objective clinical and radiological assessment of the thrust plate prosthesis (TPP). Fifty-three prostheses were evaluated clinically using the Harris hip score (HHS), visual analog scale (VAS), and radiographically before surgery, at the time of discharge, and postoperatively after on average of 8.09 (range 4.61–9.93) years. The average HHS significantly (p≤0.05) improved from 48 (range 18–77) points to 95 (range 46–100) points. The VAS revealed significant (p≤0.05) reduction of pain at rest and under load. Radiographic analysis showed a considerable potential for osteolysis under the thrust plate. Sixteen prostheses revealed signs of radiolucent zones. In general, there was a good clinical outcome with no major limitations in function. Radiographic changes under the thrust plate indicate an adaptation processes resulting from changed biomechanics. This study suggests that the TPP could be a good alternative in total hip replacement in younger patients.
Le propos de cette étude est d’évaluer cliniquement et radiologiquement une prothèse plaque (TPP). 53 prothèses plaques ont été évaluées cliniquement grâce au score de Harris et à l’échelle visuelle analogique et radiographiquement avant l’intervention, après la sortie de l’établissement, en post-opératoire avec une moyenne de 8,09 ans (de 4,61 à 9,93 ans). le score de Harris a augmenté de façon significative (p=0.05) passant de 48 (18–77) à 95 (46–100) points. L’évaluation de la douleur par échelle visuelle analogique (p=005) a montré une réduction de celle-ci. L’analyse radiographique a montré par contre un potentiel important de liserés sous la prothèse. 16 prothèses montrent des signes radiologiques d’appréciation difficile. ces prothèses démontrent qu’une bonne évolution clinique est possible avec peu de limitations fonctionnelles. La radiographie montre par contre des processus d’adaptation et de modifications osseuses secondaires aux sollicitations biomécaniques. Cette étude permet de penser que la prothèse TPP est une bonne alternative au remplacement de la hanche chez les sujets les plus jeunes.
The thrust plate prosthesis (TPP) was conceived by Huggler and Jacob in 1976. The aim of the development was to provide a physiological local force transmission into the calcar area by metaphyseal fixation to the proximal femur leaving the diaphyseal bone intact for further endoprosthetic replacement especially in younger patients . The third generation of this prosthesis was developed in 1994 and has been part of the endoprosthetic replacement protocol in our department since 1997. The thrust plate must rest in full contact with the strongest part of the medial cortical bone of the stump of the femoral neck. The oval shaped plate is fixed to a mandrel that has to be placed centrally in the femoral neck. Laterally, the stiff mandrel is firmly anchored with a single bolt to a lateral strap, which itself is fixed to the femur below the innominate bone by means of two cortical srews (Fig. 1). The TPP must be implanted with a physiological femoral-shaft angle of 125°–130° in order to allow the resultant hip–joint force to act physiologically in magnitude and direction. With this proximal load transmission giving rise to predominantly compressive forces, problems like bone remodelling and stress shielding in the trochanteric region caused by load sharing in diaphyseal implants following aseptic loosening should be avoided [6, 12]. Biomechanical studies as well as finite element analysis, have supported the concept that TPP transmits forces to femoral cortex identical to intact bone . However, the surgical approach differs in comparison to a femoral stem prosthesis. Since the strap has to be fixed to the lateral cortex of the diaphysis, a more distal approach is necessary. The tensor fasciae latae and the vastus lateralis must be split more distally. A few clinical and radiological short- to mid-term results have been published on this type of prosthesis since Suezawa reported a first series of 17 prostheses in 1983. This encouraged us to evaluate the late mid-term results of the TPP with special regard to functional and radiological outcome.
We performed a retrospective study of 46 patients (26 men and 20 women) with a mean age of 50 (range 30–65) years. Patients underwent TPP-implantation in our department between March 1997 and April 2002. Seven patients received the TPP bilaterally (i.e. total number of 53 prostheses). The right side was affected in 29 cases, the left side in 24 cases. The average clinical and radiological follow-up was 8.09 (range 4.61–9.93) years. The age limit was arbitrarily set at 60 years. Six older patients presenting a younger physical status were also included since they showed no health-related contraindications. Patients suffering from polyarthritis, ankylosing spondylitis, osteoporosis, pronounced coxa vara/valga, with a short and narrow femoral neck, or those who were unable to walk without crutches were excluded from the study. Eleven patients were lost for follow-up because of death (one patient) or change of residence (ten patients). Seven of the latter have not so far, required any revision surgery. Of the remaining patients no further information could be obtained. Eighteen patients suffered from a primary arthritis of the hip joint. In 35 cases, the indication for implantation of the TPP was secondary arthritis of the hip joint. Fifteen of these patients suffered from avascular necrosis of the femoral head (FHN), 17 from hip dysplasia (HD), two from osteoarthritis secondary to epiphysiolysis (ECF), and one developed avascular necrosis of the femoral head secondary to dysplasia of the hip. All patients underwent total hip arthroplasty through the posterolateral approach. In all cases a cementless, press-fit socket was used for acetabular replacement (Typ 2000, ESKA-Implants, Lübeck, Germany). In seven cases the acetabular shell was additionally fixed with one (six cases) or two (one case) screws and beyond that one case had an acetabuloplasty. The most commonly used combination, in 41 cases, consisted of a ceramic/polyethylene combination. In 12 cases a ceramic/ceramic combination was used. Postoperative management included early movement, and patients were allowed full weight-bearing immediately after the operation. Heterotopic ossification prophylaxis was given in all cases using Diclofenac 100 mg/d over a period of two weeks. In this prospective investigation clinical findings at the preoperative and follow-up examination were evaluated according to the Harris hip score (HHS) and the visual analog scale (VAS). Radiographic assessment was performed according to Gruber . Standardised radiographic views were necessary. A specific anteroposterior view of the pelvis with 20° internal rotation of the affected limb to evaluate the bone under the thrust plate and an axial view of the hip were performed. All hips were rated by two independent observers. No densitometry was used and if the scores were different, then the patient was reassessed and consensus was achieved. According to the radiological evaluation of the TPP by Gruber, there are seven different regions of interest in the proximal femur as shown in Fig. 2. The full cortical and cancellous bone structure of the proximal end of the femur was compared to the preoperative condition. The evaluation concentrated on radiolucencies as well as possible changes in the angle of the femoral shaft and prosthesis. Additional radiographic examination concentrated on heterotopic ossification. Heterotopic ossification was classified according to the Brooker classification which is based on AP radiographs and divided into four grades of increasing periarticular heterotopic bone formations with grade 4 showing a “bridging” bone formation between the pelvis and the proximal part of the femur .
The parameters for each patient were compared by Wilcoxon signed-rank test and the Mann-Whitney U test. The level of significance was chosen at p≤0.05.
In 53 hips a clinical and radiological follow-up examination could be performed. The mean preoperative HHS was 48 (range 18–77) points and postoperative values significantly (p≤0.05) improved to 95 (range 46–100) points. Twenty-six hips achieved 95–100 points and 47 hips achieved 91–100 points at the time of follow-up. two hips had 81–90 points and four hips gained less than 80 points. The mean range of motion assessed at clinical examination was 109.5° (SD 12.7°) on the treated side in the sagittal plane, 54.6° (SD 13.6°) in the frontal plane, and 51.9° (SD 19.0°) in the transverse plane. The operated leg was on average 0.06 cm (SD 0.45 cm) shorter than the other in all cases. In 43 cases no leg-length discrepancy could be detected.
Radiological examination showed a mean angle of the femoral shaft and prosthesis of 131° (SD 4.14°) at the time of implantation which changed by an average of 0.71° (SD 2.73°) in the valgus direction. The mean angle at the time of follow-up examination was significantly different at 132° (SD 4.76°) (p≤0.05). In ten hips radiolucencies occurred in zone A1 and in two hips in zone A2. In three hips radiolucencies appeared in both zones. In total, 16 prostheses revealed signs of radiolucent zones. In this study 13 hips had Brooker class I, and two had Brooker class II heterotopic ossification. Three hips showed ossifications of class III, and two hips were suffering from class IV ossifications.
Eight patients complained of slight to moderate pain over the lateral cortex. This required surgical revision of the fascia lata after four years in one case. The outcome was acceptable to the patient. There were no significant intraoperative complications, and no patient had a deep infection or deep vein thrombosis postoperatively. Two dislocations occurred in one case, conservative treatment with closed reduction was carried out, and this patient had no further complications. One patient with radiolucent trabecular formations in zones A1, A2, B, and C did not complain about any symptoms indicating a loosening process; therefore, the prosthesis was left in place. One TPP with a seam formation in zones A1 and B was also left in place with absence of any symptoms. The VAS revealed a significant (p≤0.05) reduction of pain at rest and under load in the operated limb. Kaplan-Meier survivorship using any part of the prosthesis removed for any reason as the end point was 100% after 9.93 years (Table 1).
With regard to the biomechanical concept, the TPP has a direct load transfer to the cortical bone of the femoral neck, particularly to the cortical bone of the calcar. According to Jacob and Huggler this force transmission is close to the physiological force transmission of the natural femur, and well known problems such as bone remodelling and stress shielding in the trochanteric region caused by load sharing in diaphyseal implants should be avoided. Adaptation of the bone and osseus integration were evaluated by Burgi and Schenk with promising radiological and histological results . Suezawa et al. reported a primary series of 17 TPP with a mean follow-up of five years showing no osteolysis in the proximal part of the femur . This accords with Huggler et al. who did not see any atrophic bone in the proximal femur in 162 TPP and with Krappel and Harland who reported on 20 TPP with a mean follow-up of 14 months [16, 22].
In contrast, Alt and Walker described atrophy of the medial cortical bone of the femoral neck in 12 of 49 cases after an average follow-up of 17 months . Menge, in his prospective study, reported a third of 268 TPP with a mean follow-up of four or seven years, respectively, having a marked bone remodelling or having a total loss of bony contact below the thrust plate only being fixed to the bone with the mandrel . Dunai and Menge also observed a marked bone remodelling in a third of 240 cases . Fink, in his study, saw five of 47 cases with osteolysis next to the thrust plate on the medial side of the femoral neck . Four of them also showed an osteolysis in the A1 zone. A seam formation of 2 mm was found in five of 52 cases within the first two years of follow-up.
Our results reveal 16 thrust plate prostheses with radiographic changes. In four cases a total loss of bony contact under the thrust plate was observed (A1 and A2 zones).
Bone resorption at the medial edge of the thrust plate may not only represent a sign of loosening but might also be seen as a phenomenon resulting from stress shielding, although this would challenge the biomechanical concept of Huggler and Jacob of a close to physiological force transmission to the proximal femur.
We agree with Ishaque et al. that radiological changes of the periprosthetic bone stock in TPP are related to physiological adaption processes resulting from changed biomechanics . In cases of stress shielding a progressive atrophy under the thrust plate and a concomitant sclerosis at the calcar region can be observed. In contrast, a loosening process might be characterised by a progressive radiolucency including zone B or C. In our study only three cases showed bone resorption in zones B or C but did not complain of any symptoms indicating a loosening process. Ishaque et al. reported radiolucencies in zones B or C in 18 cases . Ten of these were affected in both zones of which three had to undergo a revision of the prosthesis.
The evaluation of the inclination angle at the time of follow-up demonstrated that 49 hips were within the acceptable range of 120°–140°. Although the mean angle of the femoral shaft and prosthesis was greater than recommended by Huggler and Jacob (125°–130°), an average inclination angle of 132° can be considered highly satisfactory. Similar results were obtained in previous studies [9, 10, 18].
In our series 38% showed heterotopic ossification with two hips suffering from class IV ossifications. Zelle et al. reported 41% showing heterotopic ossifications in their study on TPP with no cases of grade IV . Ishaque et al. diagnosed heterotopic ossifications in 50% of their cases with 4% suffering from ankylosed grade IV . These results are similar to those obtained in our study.
Several studies have described short-term follow-up periods of two to four years. Fink reported on TPPs after inflammatory arthritis with a follow-up period of two years and a mean HHS of 86 points . He also reported on TPPs after femoral head necrosis with a mean HHS of 86 points after an average of 4.8 years . The clinical outcome in this study with an average HHS of 95 points after a mean follow-up period of eight years is very satisfactory and comparable to cementless stemmed endoprostheses. At follow-up examination, patients were asked to estimate their residual pain in the operated limb in rest and under load using a VAS ranging from 0 to 10. The residual pain was very low and averaged 0.43 (0–6) points in rest and 1.13 (0–8) points under load. This revealed a significant improvement. Therefore the functional results and subjective improvements were very satisfactory in this study.
The problem of lateral thigh pain has previously been reported and is associated with this type of prosthesis. Lateral pain has been mainly observed in young and thin patients and may decrease over time . In this series eight patients complained of lateral thigh pain. In one the pain turned out to be the major complaint and necessitated revision. In this case a gap of the fascia lata had to be closed with a satisfying result. The strap was left in place as it was clearly not the cause of the problem. Three cases suffering from strap-related trochanteric bursitis diagnosed by ultrasound examination benefited from local steroid infiltration. Four patients reported a painful lateral thigh within the first two years of operation which disappeared over time. With respect to this problem the volume of the strap was reduced with inauguration of a second generation of the TPP leading to a clear reduction of lateral thigh pain in subsequent cases . Some authors have suggested removing the strap for revision, but there are no reliable data available in the literature about this procedure. However, thigh pain is also a well-known problem in intramedullary fixed cementless stems [7, 23]. A non-physiological strain distribution due to press-fit fixation seems to be the cause of this phenomenon. The dislocation rate of 1.8% for the posterolateral approach in our study is low and comparable to stemmed endoprostheses, which are associated with rates between 0.7% and 4.8% [25, 27]. Fink et al. reported a comparable dislocation rate of 3.3% for the posterolateral approach . Zelle et al. had an unsatisfactory dislocation rate of 8.6%, but it has to be considered that all the patients with dislocations had undergone previous hip surgery .
We only saw two cases with radiolucencies in zones B or C, but without any clinical symptoms indicating a loosening process. This is remarkably low in contrast to other loosening rates reported in the literature (Table 2). Our results might be due to the fact that patients with polyarthritis, ankylosing spondylitis, osteoporosis, pronounced coxa vara/valga, with a short and narrow femoral neck, or those unable to walk without crutches were excluded from the study. Contrary to expectations of other researchers studying FHN who have suggested that the aetiological factors could lead to a premature loosening of the prosthesis, our data refute this point of view. FHN was diagnosed in nearly a third of all cases in our study and the mean follow-up time in this sub-group was 8.18 (range 6.79–9.73) years. Only one of them showed radiolucencies in zones A1 and B after seven years but clinically had no loosening symptoms (Fig. 3).
In summary, there are good indications for the TPP as an alternative to stemmed endoprostheses or resurfacing implants in view of the comparable functional and radiological results. Amstutz et al. reported on 400 hybrid resurfacing arthroplasties in younger patients (mean age 48 years) after an average follow-up of 3.5 years with a survival rate of 94.4% after four years. The mean HHS was 93.5 points and heterotopic ossification of grades III and IV were diagnosed in 10% of the cases . Witzleb et al. evaluated the clinical and radiographic outcome of 300 consecutively implanted Birmingham hip resurfacing arthroplasties. At a median follow-up of 24 months the Kaplan-Meier survival rate was 98%. Only six implant revisions had to be performed due to infection (two cases), malposition (one case), femoral neck fracture (one case), primary unstable cup (one case), and chronic pain (one case). The median Harris hip score improved from 51 points to 96 points at the last follow-up. No hip showed radiographic signs of aseptic implant loosening . Preliminary experiences with resurfacing hip arthroplasty for younger adults requiring hip arthroplasty seem to be encouraging. Especially in younger patients in which a resurfacing of the femoral head cannot be performed and in patients with FHN, the TPP is a successful alternative to preserve bone for future endoprosthetic replacement. The relatively high loosening rates and the radiolucent zones described in the literature should lead surgeons to careful follow-up examinations with sequential clinical and radiographic results.