Little is known about the long-term outcome of mobile-bearing total ankle replacement (TAR) in the treatment of end-stage arthritis of the ankle, and in particular for patients with inflammatory joint disease. The aim of this study was to assess the minimum ten-year outcome of TAR in this group of patients.

We prospectively followed 76 patients (93 TARs) who underwent surgery between 1988 and 1999. No patients were lost to follow-up. At latest follow-up at a mean of 14.8 years (10.7 to 22.8), 30 patients (39 TARs) had died and the original TAR remained in situ in 28 patients (31 TARs). The cumulative incidence of failure at 15 years was 20% (95% confidence interval (CI) 11 to 28). The mean American Orthopaedic Foot and Ankle Society (AOFAS) ankle–hindfoot score of the surviving patients at latest follow-up was 80.4 (95% CI 72 to 88). In total, 21 patients (23 TARs) underwent subsequent surgery: three implant exchanges, three bearing exchanges and 17 arthrodeses. Neither design of TAR described in this study, the LCS and the Buechel–Pappas, remains currently available. However, based both on this study and on other reports, we believe that TAR using current mobile-bearing designs for patients with end-stage arthritis of the ankle due to inflammatory joint disease remains justified.

Cite this article: Bone Joint J 2013;95-B:1656–61.

Introduction

Dislocation after primary total hip arthroplasty (THA) is a devastating and frequent postoperative complication. Many risk factors for dislocation have been identified, however, thus far there has been no consensus whether inflammatory arthritis is a risk factor for dislocation or not. We carried out a prospective study assessing the prevalence of dislocation within 2 years after primary total hip arthroplasty for osteoarthrosis and inflammatory arthritis.

Patients and Methods

Between 1996 and 1999 312 patients (342 hips) with either a primary or a posttraumatic osteoarthrosis (OA group) and 59 patients (69 hips) with rheumatoid arthritis or other forms of inflammatory arthritis (IA group) were operated. One single type of prosthesis was implanted (EPF-PLUS^® cup and SL-PLUS^® stem) using an anterior approach. All dislocations in the two years following surgery were recorded. Both diagnostic groups were compared for known risk factors such as old age, female gender, prior hip surgery, and experience of the surgeon. Radiographs were examined for avulsion fractures of the tip of the trochanter and signs of loosening. The abduction and anteversion angles of the acetabular component were measured. Statistical analysis was performed with the Chi-square test and Student’s t-test.

Results

The dislocation rate for inflammatory arthritis patients was significantly greater than that in patients with osteoarthrosis: 10. 1% (7 hips) in the IA group, 2. 9% (10 hips) in the OA group (p = 0. 006). There were no other differences in risk factors favouring dislocation in the IA group, such as old age, female gender, prior hip surgery, experience of the surgeon, trochanteric fractures or malposition of the prosthetic components. All dislocations in the IA group were posterior and occurred without any kind of trauma. In contrast, nearly half of the dislocations in the OA group were anterior and two were of traumatic origin.

Discussion

Taking into account the fact that there are no differences in known risk factors for dislocation between our two groups and no differences in complication rate, except for dislocation, we can say that inflammatory arthritis has to be considered an independent risk-factor for dislocation after primary total hip arthroplasty. It may be that inferior quality of the (pseudo) capsule and the muscles stabilising the hip joint due to inflammatory arthritis leads to inadequate soft tissue tension. Another factor can be the concomitant impairments in rheumatoid patients, such as impairments of the upper extremity, ipsilateral knee or contralateral hip, leading to hyperflexion in the operated hip causing a posterior dislocation without trauma.

Introduction

Multiple factors contributing to an elevated risk for dislocation after total hip arthroplasty (THA) have been identified. Patient-related risk factors that have been identified are prior hip surgery, old age and female gender. However, there have been no prospective reports whether inflammatory arthritis (IA) is an independent risk factor.

Material and methods

From January 1996 to December 1999 427 primary total hip arthroplasties were carried out using one type of uncemented prosthesis: a hydroxyapatite coated EPF-PLUS^® cup and a SL-PLUS^® stem (PLUS Endoprothetik AG, Rotkreuz-CH). A 28 mm. ball head was used in every hip. To evaluate whether IA is a risk factor for dislocation the incidence and cause of early (< 2 year post-surgery) dislocation in IA hips was compared with those carried out for osteoarthrosis (OA). There were 341 THAs in 311 patients with OA and 69 THAs in 59 patients with IA (mainly rheumatoid arthritis) included in this study. The remaining 17 THAs were for various other reasons and excluded from this study.

Statistical analysis of the dichotomous variables was carried out by the chi-square test and the Fisher’s exact test, Student’s t-test was used for the analysis of continuous variables.

Results and discussion

Both groups were comparable with respect to the following risk factors: gender, approach (either straight-lateral or anterolateral), position of the acetabular component and experience of the surgeon. Mean age was lower in the IA group than in the OA group: 61. 0 vs 68. 1 years. Furthermore, the incidence of prior hip surgery was higher in the OA group.

The incidence of dislocation was 7 out of 69 (10. 1%) in IA hips and 10 out of 341 (2. 9%) in OA hips (p=0. 006). All dislocations in IA where posterior, in OA 5 were posterior and 4 were anterior (1 unknown). No other mechanical factors leading to an increased instability of the hip in IA, such as trochanteric fractures, could be identified. Due to the relatively small numbers a statistical difference in the direction of dislocation could be identified (p= 0. 088).

So, IA has to be considered as an independent risk factor for dislocation after THA. Both the polyarticular impairments and the lower quality of the soft tissues in IA could explain this elevated risk. To reduce the incidence of dislocation in IA it therefore seems advisable to pay detailed attention the soft tissues and the position of the prosthetic components in IA at the time of surgery. Also, consideration should be given to the use of an acetabular component with an elevated rim.

Introduction

Total hip arthroplasty (THA) is the only successful treatment for patients in whom the hip joint is destroyed by inflammatory arthritis. Due to the effects of both the disease and its treatment elevated rates of complications and of aseptic loosening have been described. Whether with modern uncemented hip prostheses the results can be improved is not fully known. Therefore, we decided to carry out a prospective study.

Material and methods

At the introduction of a new press-fit acetabular component a prospective study on the results of uncemented THA was started. From 1995 to 1999 85 primary THAs were carried out in 72 patients (57 women, 15 men) suffering from inflammatory arthritis. Diagnosis: RA 76, adult-onset Still 4, JIA 3, miscellaneous 2. Mean age at operation was 60 years (SD 15. 9). Clinically, the Harris Hip Score was used for evaluation. Radiographically, the Larsen classification was used and at follow-up radiolucencies and signs of migration were registered.

The implants used in this study were the EPF-PLUS® acetabular component and the SL-PLUS® femoral component (PLUS Endoprothetik AG, Rotkreuz-CH). The EPF-PLUS® acetabular component is a novel modular press-fit cup. Its shell has a triple radius profile on cross-section, thereby creating a gradual lowering of the polar part of the cup. This produces a small gap of about 2 mm. between the acetabulum and the pole of the cup. Therefore, forces are mainly transmitted to the peripheral part of the acetabulum, leading to an enhanced primary stability. Originally, the shell had a gritblasted surface for osseointegration. Since 1996, the outer surface of the cup has been coated with a ground layer of pure titanium and a superficial layer of a crystalline hydroxyapatite (Ti-HA). The gritblasted version was used in 14 hips (1995–1996), the Ti-HA coated version in 71 hips. In 68 hips the cup was implanted by press-fit fixation, in 17 screws were added.

Results

Larsen classification was as follows: 0-1: 6; 3: 17; 4: 60; 5: 2. At follow-up, 5 patients (6 THAs) were deceased. Deep infection required revision in one hip. Another revision was carried out for recurrent dislocations. One grit-blasted cup developed late subsidence and was revised almost 4 years postoperatively. One Ti-HA coated cup failed early due to severe acetabular bone loss. Finally, one femoral component developed a varus tilt and became symptomatic 2 years after implantation, requiring revision. In all 62 patients with 74 THAs in follow-up cup and stem are functioning well, both clinically and radiographically. Harris Hip Score increased from 36. 6 (SD 17. 7) to 87. 9 (SD 11. 8).

Discussion

The medium-term results of the EPF®-PLUS cup show that its primary stability is good and that, as the polar gap rapidly disappears, osseointegration is secure. Only severe acetabular deficiency appears to be a contra-indication for this implant. Also, the SL-PLUS® stem performs well in inflammatory arthritis.

We carried out arthrodesis of the radiolunate joint in 46 wrists (38 patients) for pain and ulnar translation of the carpus because of rheumatoid (42) or psoriatic arthritis (4). At follow-up, three patients had died and in three (1 bilateral) an additional midcarpal arthrodesis had been undertaken. The remaining 32 patients (39 wrists) were evaluated after a mean of five years.

The clinical results were good with a mean visual analogue score of 8.3 for pain, 7.2 for hand function and 9 for overall satisfaction. Except for palmar flexion, mobility was equal to or better than before operation. Radiologically, there was deterioration of the midcarpal joint with an increase in the Larsen score from 1.8 to 2.7 (p < 0.001), some decrease in carpal height and recurrence of carpal translation.

Radiolunate arthrodesis gives good clinical results at five years although there is some deterioration radiologically.