Friction was studied in 67 retrieved cemented cups with 32 mm internal diameter. Friction was measured under 1.0 KN of static load. High molecular hyaluronic acid was adapted as a lubricant. Thirty cups were combined with alumina heads and 37 were combined with metal heads. The years cups were in situ was 7.5 (3.2–13.2) for alumina-polyethylene implants and 8.9 (1.5–15.7) for metal-polyethylene implants (p> 0.05).

The revision rate at 15 years follow-up was higher in metal-polyethylene (PE) implants (57%) than that of alumina-PE implants (40%) (p< 0.05). The prevalence of cup loosening was less in alumina-PE implants (12/30) than in metal-PE implants (29/37) (p< 0.01). Less wear was observed in alumina-PE implants (1.15+−0,80mm) than in metal-PE implants (1.62+−0.61mm) (p< 0.01). Less wear was observed in cups without loosening (alumina-PE implants: 1.84+−0.57mm, metal-PE implants: 1.75+−0.51mm) than in those with loosening (alumina-PE implants: 0.69+−0.56mm, metal-PE implants: 1.31+−0.73mm) in both types (alumina-PE implants: p< 0.01, metal-PE implants: p< 0.05). Less wear rate was observed in cups without loosening (alumina-PE implants: 0.11+−0.05 mm/year, metal-PE implants: 0.14+−0.05mm/year) than in those with loosening (alumina-PE implants: 0.17+−0.03 mm/year, metal-PE implants: 0.22+−0.09mm/year) in both types (alumina-PE implants: p< 0.01, metal-PE implants: p< 0.05). The coefficient of friction increased in proportion to the progress of cup wear in both types (alumina-PE implants: r2 =0.217, p< 0.01, metal-PE implants: r2 =0.183, p< 0.01). Relation between the coefficient of friction and stability of implants was not detected in both types, while alumina-PE implants had lower coefficient of friction (0.137+-0.056) than metal-PE implants (0.209+−0.098) (p< 0.01). The torque of metal-PE implants without stem loosening (0.137+−0.053) was larger than that of alumina-PE implants with stem loosening (0.274+−0.088) (p< 0.01).

The results suggest that wear has greater influence on stability of implants than the friction, whereas coefficient of friction increases in worn implants.

A method was developed to take radiographs showing the inner articulation of bipolar hip prostheses. By this method, wear was measured in 68 hips whose inner head diameter was 22 mm. Average annual wear rate was 0.17 mm. Osteolysis was observed in 25 hips (37%) and there was no difference between the annual wear rate of hips with and without osteolysis. Studying 19 retrieved prostheses, abrasion of the rim was deeper in hips with osteolysis than those without it. Wear rate of the inner articulation in bipolar hip prosthesis is much larger than that in Charnley’s prosthesis, as linear penetration into the articulation surface reduces the motion range of the inner articulation and this increases impingement and advances rim abrasion.