In all patients treated with as-polymerized poly( L-lactide) (PLLA), a swelling at the site of implantation was observed after three years of implantation. These swellings seem to be related with degrading PLLA and the formation of particles of high crystallinity. To avoid these complications, poly(96%L-, 4%D-lactide) (PLA96) was developed that possesses lower crystallinity that probably results in a faster and more complete degradation. To study the cause of the swelling of PLLA implants and to study the degradation of PLA96 long-term implantation studies are required. Considering the very slow degradation rate of as-polymerized PLLA, in vitro predegradation was performed at elevated temperatures (90 degrees C) to simulate long-term physiological degradation. In this study a comparison was made between the histopathological reaction to non-degraded and predegraded PLLA, PLA96 and polyethylene (PE) discs implanted subcutaneously in rats. Animals were sacrificed after a postoperative period varying from 4 to 52 weeks. Chemical, light- and electron microscopical analysis and semi-quantitative measurements were performed. Based on the chemical analysis, the degradation rate of PLA96 was higher compared with PLLA. The histological reaction to non-degraded PLLA and PLA96 discs was very mild. The histological reaction to the predegraded implants was qualitatively similar to the reaction to the non-degraded implants, however, quantitatively an increase was noted. A number of predegraded PLLA and PLA96 discs showed an increase of volume with implantation time caused by the formation of fields of polymer debris accompanied by a granulomatous inflammatory reaction. The debris zone was found to consist of both polylactide polymer fragments and small remnants of degenerated cells. From our results it can be concluded that, when compared to PLLA, the degradation of PLA96 is enhanced. Subcutaneously implanted predegraded PLLA a nd PLA96 discs can induce a swelling similar to th at observed with PLLA implants in patients. So, in vitro predegradation followed by in vivo implantation might be used as a model to predict late complications during clinical use.