We scanned 25 left knees in healthy human subjects using MRI. Multiplanar reconstruction software was used to take measurements of the inferior and posterior facets of the femoral condyles and the trochlea.

A ‘basic circle’ can be defined which, in the sagittal plane, fits the posterior and inferior facets of the lateral condyle, the posterior facet of the medial condyle and the floor of the groove of the trochlea. It also approximately fits both condyles in the coronal plane (inferior facets) and the axial plane (posterior facets). The circle fitting the inferior facet of the medial condyle in the sagittal plane was consistently 35% larger than the other circles and was termed the ‘medial inferior circle’. There were strong correlations between the radii of the circles, the relative positions of the centres of the condyles, the width of the condyles, the total knee width and skeletal measurements including height. There was poor correlation between the radii of the circles and the position of the trochlea relative to the condyles.

In summary, the condyles are approximately spherical except for the inferior facet medially, which has a larger radius in the sagittal plane. The size and position of the condyles are consistent and change with the size of the person. However, the position of the trochlea is variable even though its radius is similar to that of the condyles. This information has implications for understanding anterior knee pain and for the design of knee replacements.

Cite this article: Bone Joint J 2014;96-B:1623–30.

Since the Oxford knee was first used unicompartmentally in 1982, a small number of bearings have fractured. Of 14 retrieved bearings, we examined ten samples with known durations in situ (four Phase 1, four Phase 2 and two Phase 3). Evidence of impingement and associated abnormally high wear (> 0.05 mm per year) as well as oxidation was observed in all bearings. In four samples the fracture was associated with the posterior radio-opaque wire. Fracture surfaces indicated fatigue failure, and scanning electron microscopy suggested that the crack initiated in the thinnest region. The estimated incidence of fracture was 3.20% for Phase 1, 0.74% for Phase 2, 0.35% for Phase 3, and 0% for Phase 3 without the posterior marker wire. The important aetiological factors for bearing fracture are impingement leading to high wear, oxidation, and the posterior marker wire. With improved surgical technique, impingement and high wear should be prevented and modern polyethylene may reduce the oxidation risk. A posterior marker wire is no longer used in the polyethylene meniscus. Therefore, the rate of fracture, which is now very low, should be reduced to a negligible level.

National registers compare implants by their revision rates, but the validity of the method has never been assessed. The New Zealand Joint Registry publishes clinical outcomes (Oxford knee scores, OKS) alongside revision rates, allowing comparison of the two measurements. In the two types of knee replacement, unicompartmental (UKR) had a better knee score than total replacement (TKR), but the revision rate of the former was nearly three times higher than that of the latter. This was because the sensitivity of the revision rate to clinical failure was different for the two implants. For example, of knees with a very poor outcome (OKS < 20 points), only about 12% of TKRs were revised compared with about 63% of UKRs with similar scores.

Revision therefore is not an objective measurement and should not be used to compare these two types of implant. Furthermore, revision is much less sensitive than the OKS to clinical failure in both types and therefore exaggerates the success of knee replacements, particularly of TKR.

Retrieval studies have shown that the use of fully congruent meniscal bearings reduces wear in knee replacements. We report the outcome of 143 knees with anteromedial osteoarthritis and normal anterior cruciate ligaments treated by unicompartmental arthroplasty using fully congruous mobile polyethylene bearings. At review, 34 knees were in patients who had died and 109 were in those who were still living. The mean elapsed time since operation was 7.6 years (maximum 13.8). We established the status of all but one knee.

There had been five revision operations giving a cumulative prosthetic survival rate at ten years (33 knees at risk) of 98% (95% CI 93% to 100%). Considering the knee lost to follow-up as a failure, the ‘worst-case’ survival rate was 97%. No failures were due to polyethylene wear or aseptic loosening of the tibial component. One bearing which dislocated at four years was reduced by closed manipulation.

The ten-year survival rate is the best of those reported for unicompartmental arthroplasty and not significantly different from the best rates for total knee replacement.

Many designs of unicompartmental knee replacement show early and mid-term failure due to polyethylene wear. We studied the wear rate of congruent polyethylene meniscal bearings retrieved from failed Oxford unicompartmental knee replacements.

We examined 16 bearings, 0.8 to 12.8 years after implantation, measuring their thickness and comparing it with that of 14 unused bearings. The mean rate of penetration, which included the effects of wear at both upper and lower surfaces, was 0.036 mm per year (maximum 0.08). Bearings as thin as 3.5 mm wore no faster than thicker models, but ten with evidence of impingement had greater wear. The six bearings with no impingement showed a mean rate of penetration of 0.01 mm per year.

In unicompartmental knee replacement, careful implantation of fully congruous meniscal bearings can avoid failure due to polyethylene wear.

Using a new, non-invasive method, we measured the patellofemoral force (PFF) in cadaver knees mounted in a rig to simulate weight-bearing. The PFF was measured from 20° to 120° of flexion before and after implanting three designs of knee prosthesis.

Medial unicompartmental arthroplasty with a meniscal-bearing prosthesis and with retention of both cruciate ligaments caused no significant change in the PFF. After arthroplasty with a posterior-cruciate-retaining prosthesis and division of the anterior cruciate ligament, the PFF decreased in extension and increased by 20% in flexion. Implantation of a posterior stabilised prosthesis and division of both cruciate ligaments produced a decrease in the PFF in extension but maintained normal load in flexion.

There was a direct relationship between the PFF and the angle made with the patellar tendon and the long axis of the tibia. The abnormalities of the patellar tendon angle which resulted from implantation of the two total prostheses explain the observed changes in the PFF and show how the mechanics of the patellofemoral joint depend upon the kinematics of the tibiofemoral articulation.