Modulation of the membrane surrounding particulate cement and polyethylene in the rabbit tibia

Twenty-nine mature New Zealand white, female rabbits were divided into four groups. Using sterile technique, a 6 mm drill hole was made in the tibia 1 cm distal to the knee joint. The marrow was scooped out underneath the hole. The right tibia received Simplex particulate cement polymer (PMMA) (groups 1 and 2) or polyethylene particles (UHMWP) (groups 3 and 4). The left leg functioned as a prepared but non-implanted control. All animals were fed a standard diet; those in groups 1 and 3 received plain water, while groups 2 and 4 drank water in which sodium naproxen was dissolved (1.375 mg/ml). Animals were killed after 16 wk. The implant area was harvested and grown in tissue culture. The cumulative collection of tissue culture supernatants over 3 d was assayed for prostaglandin E2 (PGE2) via radioimmunoassay. PGE2 production was significantly higher in the membrane harvested from the right side containing particulate cement with no sodium naproxen (group 1) compared with controls (P less than 0.05). The ratio of PGE2 values for the right divided by the left side yielded higher values in group 1, compared with groups 2, 3 or 4 (P less than 0.01). Previous studies have suggested that particulate debris and PGE2 production are associated with arthroplasty loosening. This experiment has demonstrated that PGE2 production by the membrane surrounding particulate debris can be suppressed by the administration of oral sodium naproxen. Because non-steroidal anti-inflammatory drugs are known to inhibit prostaglandin synthesis in man, these agents may prove useful in retarding the bone loss associated with early prosthetic loosening.