Chondrocalcinosis

A SURVIVAL ANALYSIS OF AN INDEPENDENT SERIES

We describe the outcome of a series of 124 Oxford meniscal-bearing unicompartmental arthroplasties carried out for osteoarthritis of the medial compartment. They had been undertaken more than ten years ago in a non-teaching hospital in Sweden by three surgeons. All the knees had an intact anterior cruciate ligament, a correctable varus deformity and full-thickness cartilage in the lateral compartment. Thirty-seven patients had died; the mean time since operation for the remainder was 12.5 years (10.1 to 15.6).

Using the endpoint of revision for any cause, the outcome for every knee was established. Six had been revised (4.8%). At ten years there were 94 knees still at risk and the cumulative survival rate was 95.0% (95% confidence interval 90.8 to 99.3). This figure is similar to that reported by the designers of the prosthesis and to the best published results for independent series of total knee replacement. If patients are selected appropriately, this implant is a reliable treatment for anteromedial osteoarthritis of the knee.

J Bone Joint Surg [Br] 2001;83-B: 191-4.

Received 26 January 2000; Accepted after revision 11 August 2000

In 1978 Goodfellow and O'Connor1 introduced the Oxford prosthesis, with congruent mobile bearings, for arthroplasty of the knee. In 1982, the first unicompartmental replacement with the Oxford prosthesis was performed. The implant was designed in the belief that the large areas of contact provided by the congruous articulation would diminish polyethylene wear2,3 and improve the long-term survival of unicompartmental arthroplasty. White, Ludkowski and Goodfellow4 defined the clinicopathology of `anteromedial osteoarthritis' and suggested that its anatomical features made it suitable for unicompartmental replacement.

Murray, Goodfellow and O'Connor5 reported a cumulative survival rate of 98% (confidence interval (CI) 95 to 100) at ten years for the designer's own series of 144 arthroplasties performed for anteromedial osteoarthritis. In 1995 Lewold et al6 described the results of 699 Oxford replacements (medial and lateral) enrolled in the Swedish Knee Arthroplasty Register between 1983 and 1992. They found a cumulative survival rate at six years of only 89%. Most of the failures (70%) had occurred in the first two years after surgery and dislocation of the bearing was found to be the commonest cause of failure. They cast doubt on whether the good results reported by the designer could be achieved elsewhere and suggested that to validate this a well-documented series from an independent centre was required.

We now report the long-term survival rate of the Oxford Knee in a series of patients with anteromedial osteoarthritis in which the operations were performed by three surgeons in a non-teaching hospital in Sweden.

Patients and Methods

Between 1983 and 1999, 395 knees (315 patients) were treated using the medial Oxford unicompartmental knee arthroplasty. By March 1999, 130 knees (109 patients) had been operated on more than ten years previously. Of these, 124 (103 patients) had primary anteromedial osteoarthritis with an intact anterior cruciate ligament and this group is the subject of our survival analysis. Details of the six medial-compartment arthroplasties performed for different indications and excluded from the study are given in Table I. The mean age at operation was 69.6 years (50.7 to 85.7). There were 54 women and 49 men and they underwent 82 unilateral, nine synchronous bilateral and 12 staged bilateral procedures.

Selection of patients. All the knees included in the study group had primary osteoarthritis of the medial compartment (Ahlback's radiological grades 2 or 3)7 with full-thickness preservation of articular cartilage in the lateral compartment as seen on preoperative radiographs. Fibrillation of the cartilage of the lateral compartment, observed at operation, and even focal erosions, if limited to the medial margin of the lateral femoral condyle, were not considered to be contraindications. The state of the patellofemoral compartment was not used as a criterion for selection and no patient was rejected because of patellofemoral degeneration. All knees had a functionally intact anterior cruciate ligament, but superficial damage to its synovial covering was not considered to be a contraindication, nor was the presence of chondrocalcinosis.

Patients with fixed flexion of more than 15 were excluded. Varus deformity was not a contraindication unless it did not correct passively. Age and weight did not influence selection. These qualifications are the same as those used by Murray et al5 when reporting the ten-year survival of the designer's series of medial, unicompartmental knee arthroplasties. The two studies are therefore comparable.

Operative technique. The procedures were carried out by three surgeons, although mostly by one (96, UCGS). The articular surface of the metal femoral component of the Oxford knee is spherical and that of the metal tibial component, flat. A fully congruent, mobile polyethylene bearing is inserted between the two and the metal components are cemented into place. The thickness of the bearings varied between 3.5 and 11.5 mm.

In the first 121 operations, the Oxford Knee Phase I (Biomet Ltd, Bridgend, UK) was used. The condyles were prepared by a standard method8 but accurate balancing of the flexion and extension gaps was difficult to achieve. In January 1989, the Phase-II prosthesis was introduced and succeeding procedures, three in this series, were performed using this implant. The basic design was not changed, but the non-articular surface of the femoral implant in the phase-II prosthesis was altered to allow the use of a rotary bone mill. Bone was removed in measured steps of 1 mm thickness, which facilitated more accurate balancing of tension in the ligaments. Ligamentous release was never required.

Review of patients. Every patient still living was contacted, and the status of their knees was established. The mean time to follow-up for the 74 unrevised knees in 60 patients was 12.5 years. For the 37 patients (44 knees) who had died, we used records from the hospital and local doctors to establish the state of every implant at the time of death. At the end of the review no patient was lost to follow-up. Positive evidence that a revision procedure had not been performed was obtained before the outcome of any knee was recorded.

Analysis of data. Using revision of the prosthesis for any cause as the endpoint, a life table was constructed and the 1-year to 16-year survival rates were determined. The 95% CI was calculated using the method of Peto et al.8

Results

Revision operations had been performed on six knees because of dislocation of the bearing in three, aseptic loosening in two and infection in one (Table II). Other procedures were carried out (Table III) on a further two knees which were not counted as revisions. In neither of these cases was the prosthesis removed, and both functioned well after surgery. Table IV is the life table for all causes of revision. The ten-year cumulative survival, when there were still 94 knees at risk, was 95.0% (95% CI 90.8 to 99.3). These figures represent the `worst-case' scenario as no patient was lost to follow-up. The survival rate has remained the same (95.0%) up to the 16th year, but the confidence intervals have widened as the numbers at risk have decreased.

Discussion

The survival rate of 95% of our series at ten years (CI 90.8 to 99.3) is not significantly different (log-rank test, p = 0.9) from the 97.7% (CI 93 to 100) reported by Murray et al.5

Our findings can be compared with those of the designer's series because the selection criteria employed were the same. It is not possible, however, to make a straightforward comparison with the cases reported by Lewold et al6 from the Swedish Knee Arthroplasty Register. In that project, no data were collected on the criteria for surgery and there was no authority over the participants, who applied their own indications for arthroplasty. The high rate of early failure reported in their study may be attributed either to inappropriate selection or to faulty technique.

Our results were achieved in knees with primary anteromedial osteoarthritis in which the anterior cruciate ligament was functioning normally and in which the varus deformity was passively correctable. In our study, and in the designer's series, such knees constituted about one in four of all those requiring replacement arthroplasty. By contrast, some surgeons have found that 70% to 90% of osteoarthritic knees are suitable for unicompartmental arthroplasty.9,10 It seems likely that our criteria differed from those used by some of the surgeons whose cases were reported to the Swedish Knee Register. Precise indications are important in unicompartmental arthroplasty, more particularly in a prosthesis with a mobile bearing than in one in which it is fixed.

The patients included in the Knee Arthroplasty Register were treated at 19 centres (including our own) from the time that the prosthesis was first introduced to Sweden. The technique of implantation of the Oxford Knee is demanding and very different from that of other unicompartmental implants. The data from the Swedish National Register show that the method has not been successful everywhere and suggest that it should only be employed by surgeons who have been appropriately trained in its use.

Our review confirms that the use of a meniscal-bearing prosthesis can provide good long-term results of unicompartmental replacement, probably by avoiding failure from polyethylene wear.2,3 Ninety of the knees in our series have been functioning for more than ten years and none has failed from this cause. We conclude that the Oxford Unicompartmental meniscal-bearing arthroplasty is a reliable treatment for anteromedial osteoarthritis of the knee, when correct indications are applied and appropriate surgical experience is available. The results which can be achieved for ten-year survival are better than those reported for fixed-bearing designs and are as good as those of total knee arthroplasty.

References

1. Goodfellow J, O'Connor J. The mechanics of the knee and prosthesis design. J Bone Joint Surg [Br] 1978;60-B:358-69.

2. Argenson JN, O'Connor JJ. Polyethylene wear in meniscal knee replacement: a one to nine-year retrieval analysis of the Oxford knee. J Bone Joint Surg [Br] 1992;74-B:228-32.

3. Psychoyios V, Crawford RW, O'Connor JJ, Murray DW. Wear of congruent meniscal bearings in unicompartmental knee arthroplasty: a retrieval study of 16 specimens. J Bone Joint Surg [Br] 1998;80-B:976-82.

4. White SH, Ludkowski PF, Goodfellow JW. Anteromedial osteoarthritis of the knee. J Bone Joint Surg [Br] 1991;73-B:582-6.

5. Murray DW, Goodfellow JW, O'Connor JJ. The Oxford unicompartmental arthroplasty: a ten-year survival study. J Bone Joint Surg [Br] 1998;80-B:983-9.

6. Lewold S, Goodman S, Knutson K, Robertsson 0, Lidgren L. Oxford meniscal bearing knee versus the Marmor knee in unicompartmental arthroplasty for arthrosis: a Swedish multi-centre survival study. J Arthroplasty 1995;10:722-31.

7. Ahlback S. Osteoarthrosis of the knee: a radiographic investigation. Acta Radiol 1968;Suppl 277:7-72.

8. Peto R, Pike MC, Armitage P, et al. Design and analysis of randomised clinical trials requiring prolonged observation of each patient. Br J Cancer 1977;35:1-40.

9. Christensen NO. Unicompartmental prosthesis for gonarthrosis: a nine-year series of 575 knees from a Swedish hospital. Clin Orthop 1991,273:165-9.

10. Larsson SE, Larsson S, Lundkvist S. Unicompartmental knee arthroplasty. Clin Orthop 1988;232:174-81.

U. C. G. Svard, A. J. Price

From Skaraborgs Sjukhus Karnsjukhuset, Skovde, Sweden

U. C. G. Svard, MD, Consultant Orthopaedic Surgeon

Department of Orthopaedic Surgery, Skaraborgs Sjukhus Karnsjukhuset, 451 85 Skovde, Sweden.

A. J. Price, FRCS, Research Fellow

Oxford Orthopaedic Engineering Centre, Nuffield Department of Orthopaedic Surgery, Oxford Orthopaedic Engineering Centre, Nuffield Orthopaedic Centre, Windmill Road, Headington. Oxford OX3 7LD, UK.

Correspondence should be sent to Mr A. J. Price at 81 Warwick Street, Iffley Fields, Oxford OX4 1SZ, UK,

We wish to thank Dr G. Gudmundsson and Dr B. Tjornstrand for allowing us to study their patients. We also thank Helena Svensson for her invaluable help during this study.

The author or one or more of the authors have received or will receive benefits for personal or professional use from a commercial party related directly or indirectly to the subject of this article. In addition, benefits have been or will be directed to a research fund, foundation, educational institution, or other non-profit organisation with which one or more of the authors are associated.

Copyright British Editorial Society of Bone & Joint Surgery Mar 2001
Provided by ProQuest Information and Learning Company. All rights Reserved