Introduction: Hip simulator tests, analyses of retrieved components as well as radiostereometric measurements revealed that cups made of highly linked polyethylene show a much better wear performance than cups of conventional polyethylene. However, to the best of our knowledge, histomorphological studies of tissues from the surrounding of highly cross – linked polyethylene components have not been reported yet. The aim of this study was to examine such tissues for particles, released from highly cross – linked polyethylene cups and to compare the findings with those of conventional polyethylene.

Material and methods: So far, periprosthetic tissues retrieved at revisions of 11 total hip endoprostheses with highly cross – linked polyethylene cups could be analysed. The revisions became necessary 3 to 50 months after implantation because of cup loosening (4), stem loosening (1), infection (3), periprosthetic fracture of the femur (1), multiple dislocations (1) and periarticular ossification (1). The findings were compared with those of 5 artificial joints (2 ABG, 2 Müller Cup older design, 1 Metalback pressfit) with conventional polyethylene cups and 54 to 231 months of function. 5 μm sections were made from the tissues and conventionally stained with HE and van Gieson. Morphometric measurements were done using objectives 10 and 40 of an Olympus microscope and the ‘Analysis’ program of Soft Ware Imaging GmbH. In the Durasul TM – cases, the total amount and the total area of particles were ascertained while in the cases used for comparison only random CX 40 samples could be measured because of the much higher content of particles.

Results: The DurasulTM cases showed in contrast to the cases with conventional polyethylene cups, no distinct foreign body reaction. In four of the DurasulTM cases no particles could be detected at all and in the remaining seven particles could be found only in a few areas. Their number per section was between 6 and 1208, their total area per mm² section ranged between 0,03 and 6,99 × 10 5.mm² In the conventional polyethylene cases the number of particles per section was between 2832 and 71447, their total area per mm² section ranged between 1,06 and 25,91 × 10 3mm²! The average size of the DurasulTM particles was clearly bigger than the size of the conventional Polyethylene.

Discussion: The measurements in tissues of early revisions show that DurasulTM releases much less particles into the surrounding than PE and cause rather no foreign body reaction. Accordingly, the burden of the tissue with polyethylene debris is much lower while the particle size is bigger with DurasulTM than with PE.

Metal-on-metal bearing surfaces have been reintroduced for use in total hip replacement, despite concerns regarding the potential risks posed by metallic by-products. We have compared periprosthetic tissues from metal-on-metal and metal-on-polyethylene hip replacements at revision surgery with control tissues at primary arthroplasty.

Tissues were obtained from 9 control, 25 contemporary metal-on-metal, 9 CoCr-on-polyethylene and 10 titanium-on-polyethylene hip replacement arthroplasties. Each was processed for routine histology with Haematoxylin and Eosin. Quantitative stereological analysis was performed at the light microscopic level.

Metal-on-metal sections showed more surface ulceration and this was correlated with the density of inflammation in the deeper tissues layers. Metal-on-metal tissues displayed a pattern of well-demarcated tissue layers, which were rarely seen in metal-on-polyethylene cases. In metal-on-polyethylene cases, the inflammation was predominantly histiocytic. Metal-on-metal cases by contrast showed a lymphocytic infiltrate with abundant plasma cells. Metal-on-metal tissues showed a striking pattern of peri-vascular inflammation with prominent lymphocytic cuffs especially deep to areas of surface ulceration. Levels of inflammation were higher in cases revised for failure than in those retrieved at autopsy or exploratory surgery. Total replacement and surface replacement designs of metal-on-metal arthroplasty showed similar histological changes. Plasma cells were not seen in any of the metal-on-polyethylene cases. The differences between the patterns of inflammation and cellular infiltration seen in metal-on-metal and metal-on-polyethylene tissues were highly statistically significant.

The pattern and type of inflammation in periprosthetic tissues from metal-on-metal and metal-on-poly-ethylene arthroplasties is very different. Our findings support the conclusion that metal-on-metal articulations are capable of generating a form of immunological response to metallic wear debris that has not been described previously. The incidence and clinical implications of these immunological responses in failed metal-on-metal joints are unknown.