Objectives

Third-body wear is believed to be one trigger for adverse results with metal-on-metal (MOM) bearings. Impingement and subluxation may release metal particles from MOM replacements. We therefore challenged MOM bearings with relevant debris types of cobalt–chrome alloy (CoCr), titanium alloy (Ti6Al4V) and polymethylmethacrylate bone cement (PMMA).

Ceramic-on-ceramic alumina bearings (ALX) have demonstrated low wear with minimal biological consequences for almost four decades. An alumina-zirconia composite (BIOLOX-DELTATM) was introduced in 2000 as an alternative ceramic. This contains well-distributed zirconia grains that can undergo some surface phase transformations from tetragonal to monoclinic. We analyzed 5 cases revised at 1–7 years to compare to our simulator wear studies. For the retrieved DELTA bearings, two important questions were

how much tetragonal to monoclinic transformation was there in the zirconia phase and

The retrieval cases were photographed and logged with respect to clinical and revision details. The DELTA balls varied from 22mm to 36mm diameters. These had been mated with liner inserts varying by UHMWPE, BIOLOX-FORTE and BIOLOX-DELTA materials. Bearing features were analyzed for roughness by white-light interferometry, for wear by SEM, for dimensions by CMM and for transfer layers by EDS technique. Surface transformations on DELTA retrievals were mapped by XRD. The four combinations of 36mm diameter BIOLOX-FORTE and BIOLOX-DELTA were studied in a hip simulator, which was run in ‘severe’ micro-separation test mode to 5 million cycles. Wear rates, wear stripes, bearing roughness and wear debris were compared to the retrieval data.

In two DELTA ball cases, there were conspicuous impingement signs, stripe wear and black metallic smears. It is to be noted that the metal transfer sites (EDS) appeared to be from the revision procedures. The retrieved balls run with alumina liners showed monoclinic phase peaking at 32% on the particular surface and internal bore. On the fracture surface of case 1, the monoclinic content had increased to 40%. Various surface roughness indices were assessed on the bearings. The polished articular surfaces averaged roughness (Sa) of the order 3 nm, representing extremely smooth surfaces. The main wear zone was only marginally rougher (5 nm). In contrast the stripe wear zones had roughness of the order 55–140 nm.

In the laboratory, the DELTA bearings provided a 3–6 fold wear reduction compared to FORTE controls. Roughness of stripes increased to maximum 113nm on controls. Roughness of wear stripes showed FORTE with the highest and DELTA with the lowest values. DELTA bearings also revealed much milder wear by SEM imaging. Phase transformations showed peaks at < 30% for both main wear zone and stripe wear sites. It is hypothesized that the concentration of monoclinic phase reached a certain level due to compression contraint imposed by the alumina matrix. With implant wear, additional tetragonal grains of zirconia are exposed and these will also transform to tetragonal. This consistency between laboratory and retrieval studies confirmed the stable nature of the bearings. The BIOLOX-DELTA combination provides optimal potential for a clinically relevant reduction in stripe wear.

Zirconia (ZrO2) on Polyethylene (PE) has been used for18 years. However, a majority of clinical results expressed caution, citing surprisingly high wear and osteolysis. The most recent clinical study (Walters 2004) reports 20% higher wear with ZrO2/PE compared to CoCr/PE The PE wear may be the result of increased surface roughness of the ball, due to zirconia transformation from tetragonal to monoclinic phase. Impingement of metal cup backing on zirconia may result in high stress that drives the transformation. Our objective was to simulate the conditions most likely to transform the zirconia ball surface.

In phase one, the rim of a titanium cup was loaded against a zirconia ball with a static load range of 0.01kN to 10kN to simulate impingement-dislocation stress. The ball was cleaned with acid, ultrasonic cleaning, and then observed under SEM. Ball sections were made for XRD study. In phase two, the zirconia balls were pressure stressed in a hip simulator (static load range of 1kN to 4kN) for 600 cycles. Wear studies follow.

In phase one, the rim of a titanium cup was loaded against a zirconia ball with a static load range of 0.01kN to 10kN to simulate impingement-dislocation stress. The ball was cleaned with acid, ultrasonic cleaning, and then observed under SEM. Ball sections were made for XRD study. In phase two, the zirconia balls were pressure stressed in a hip simulator (static load range of 1kN to 4kN) for 600 cycles. Wear studies follow.

Pressure alone does not appear to transform zirconia in the articular surface. The phase two simulator studies include combinations of mechanical stress, environmental aging (autoclave) and lubricants (air, water, serum). The role of the heavy metal transfer may actually protect the zirconia surface from transformation phenomena.

1. The fibrillar networks of adult human articular cartilage, taken from femoral and acetabular specimens, have been systematically examined by scanning electron microscopy. The internal structures revealed by rupturing the tissue were compared with published findings from transmission electron microscope studies.

2. Though this technique demonstrated the internal fibrillar appearance of cartilage to a remarkable degree, it had several attendant limitations. On final drying, specimens generally exhibited shrinkage which varied within wide limits; this could have altered the internal architecture to some extent. In addition, the rupturing technique, which at the time of this investigation was the only satisfactory method of revealing the fibrillar cartilage structure, may well have had a great influence on the fibril orientations.

3. The fibrils revealed no characteristic collagen periodicity and were considerably thicker than those observed by transmission electron microscopy. It is suggested that a coating of mucin on the collagen fibrils might account for this.

4. At low magnifications the torn layers in the fractured surfaces extended radially from the calcified zone and turned obliquely at or near the articular surface to merge with the distinctly layered superficial zone, thus forming arcade-like structures. That these were not artefacts produced by the fracturing technique was shown by their similarity to the classical arcade pattern of light microscopy. However, the factor which governed the direction of these planes of weakness, be it collagen, mucopolysaccharides or cells, could not be satisfactorily determined.

5. At higher magnifications only three regions of distinct fibrillar organisation could be identified: 1) a surface layer consisting of a random fibrillar network; 2) a superficial zone composed of layers of fibrillar network, intersecting and overlapping in planes parallel to the surface; and 3) elsewhere below the superficial zone a network of virtually random fibrils which extended to the calcified region with apparently little variation in thickness or density. There was little variation from this pattern even in aged fibrillated specimens.

6. At the lower magnification range the scanning electron microscope has revealed the arcade pattern described by light microscopy, while at the higher magnifications the fibrillar organisation as seen by scanning electron microscopy correlated well with the concepts developed by transmission electron microscopy, that is, a random network of fibrils overlaid at the articular surface by a membrane-like system of bundled fibrils.

7. A possible role in the transmission of joint forces is outlined for the above fibrillar organisation.