Purpose: Osteochondral grafts are being commonly used to repair articular surface defects. The purpose is to achieve the normal architecture of hyaline cartilage with secure and seamless incorporation into recipient sites. However the details of the incorporation of these grafts have not yet been completely elucidated. The expectation is that graft union would involve the proliferation and/or migration of cells and the secretion of matrix and fibres into the graft-host cleft. The aim of this study was to determine the composition of the graft-to-host repair tissue and the integrity of the surfaces of the transplanted graft.

Method: The medial femoral condyle (WC) of the right knee of 12 adult male New Zealand White rabbits was exposed via a midline incision and medial arthrotomy under a general anaesthetic. A cylindrical 4mm diameter and 4mm long osteochondral graft was obtained using the T- handle harvester (MITEK COR System) and then it was reinserted into the same site. A groin-to-toe plaster of Paris cast was applied and the animals were allowed to recover. At weekly intervals, 3 animals were killed and the MFC was excised, fixed in 10% buffered formalin for a week and decalcified in Kristensen’s solution for another week. The specimens were dehydrated through graded ethanol and amyl acetate. Next, they were critical point dried in Blazers Critical Point Drier CPD 030 giving four 15-minute exchanges through liquid C02 before critical point drying. Finally, the specimens were mounted on aluminium stubs and sputter coated in a Polaron SC7640 Sputter coater for 90 seconds resulting in a layer of Gold/Palladium with an approximate thickness of 673 Ao. The samples were then viewed in the Hitachi S-300H scanning electron microscope.

Results: Cartilage-to-cartilage union was not observed at any time interval. Where cartilage union appeared to have occurred, this was due primarily to press fit or ‘surface weld’. In some cases, the adjoining graft and host surfaces revealed superficial fractures presumably caused, as grafts were malleted into place. There was bony union at the base in all cases. In the later time intervals this union had crept up towards the joint surface. The materials in the cleft between the graft and the recipient bed ranged from fibrous to bony elements. The graft surfaces were smooth like the surrounding normal articular cartilage at 1 and 2 weeks but fibrillated at 3 and 4 weeks.

Conclusions: These results appear to suggest that direct cartilage-to-cartilage healing may not occur following osteochondral grafting. Bone-to-bone healing appears to be universal and rapid and, materials ftom this source may be responsible for gap healing. The results also raise the possibility that the articular surfaces of grafts may deteriorate with time but the reasons are not apparent from this study.

Introduction: Following osteochondral transplantation for articular surface defects, union between graft and recipient bed cartilage may occur via two mechanisms. Healing could occur as a result of ingrowths of mesen-chymal cells derived from the subchondral bone. Direct cartilage-to-cartilage healing could occur as a result of chondrocyte proliferation and migration from the margins of graft and recipient bed. This latter mechanism depends upon the marginal chondrocytes surviving the transplantation process, remaining viable and then being capable of cell division as well as normal matrix production.

Aim: The purpose of this study was to investigate the viability of chondrocytes at the graft-recipient bed boundary using the Trypan Blue exclusion technique.

Method: Under general anaesthesia, the medial femoral condyle (MFC) of the right knee of 12 adult male New Zealand White rabbits was exposed via a midline incision and medial arthrotomy. A cylindrical 4mm diameter and 4mm long osteochondral graft was obtained using the T- handle harvester (MITEK COR System) and then it was reinserted into the same site. A groin-to-toe plaster of Paris cast was applied and the animals were allowed to recover. At weekly intervals, 3 animals were killed and the femoral artery of the operated leg was perfused with 10ml Trypan blue. The MFC was excised and fixed in 10% buffered formalin for 1 week. Thereafter, the specimens were decalcified in 10% Kristensen’s solution for 1 week, processed and then paraffin embedded. Sections ‘6u thick were obtained and examined with a light microscope. For each specimen, one section was counterstained with eosin before microscopy.

Results: The animals survived for the duration of the study and the wounds were well healed with no signs of infection. Joint effusion and synovitis were observed in the operated knees at weeks 1, 2 and 3. All grafts were in place and all had faint demarcating borders separating the graft from the surrounding recipient bed. In all cases, there was a zone of positively staining chondrocytes on the periphery of the graft and in the adjoining recipient bed. The zone of positively staining cells extended some considerable distance into the cartilage and affected all its layers. Chondrocytes at the periphery of osteochondral grafts and the adjoining recipient bed may not survive transplantation.

Discussion: This calls into question the ability to achieve direct union between the graft and the recipient bed cartilages. The likely causes of cell death are physical perturbation and direct contact between chondrocytes and blood or synovial fluid. The long-term survival of an osteochondral graft may be determined by whether or not boundary healing has occurred. In the absence of boundary healing, a graft could become bathed in synovial fluid. A pseudarthrosis of sorts could then form which may erode the graft, cause graft subsidence and/or ultimately result in graft death.

We present the results of a retrospective study of 561 consecutive Total Knee replacements performed by one consultant between January 1990 and June 2000 on 456 patients. The groups included 500 PFC, 40 PROM and 21 Kinematic TKRs. These were assessed using two postal questionnaires, a clinical evaluation and radiographic investigation. 464 patients survived to follow-up of which 421 returned the questionnaires and 292 were seen in clinic. 89% had good to excellent Oxford and clinical scores. There were no radiographically loose implants at the time of evaluation. The three types of prosthesis were then compared and a review of those that had undergone revision surgery.