ARTICLE IN PRESS O. Roualdes et al/ Biomaterials xox(2009)1-12 产 b d olemented with 10% FBS, 1% penicillin/streptomycin) for 24 h At that time, medium was replaced hie由ga media containing either alumina (a, b), zirconia(c, d)or no particles(control: e, f) and cultured for 72 h under standard conditions. After this period, cells were with Coomassie blue brilliant in order to observe cell morphology and spreading (original magnification= 40x, arrows indicate particles localization and bars indicate 10 um). gere more present in the intracellular compartment (respectively correlation r=0.5151, p< 0.05). All synovial lining cell reactions 42+0.07 and 3. 23+0. 42)than extra-cellularly (respectively observed were similar for both alumina and zirconia powders. 2.49+0.15 and 1.84+1.13). Thereafter, we compared the scores of In a second time, we evaluated the sub-synovial tissue reat total particles amount in the synovial lining cell layers and in the tion Rats who received particles showed a significantly increase ub-synovial tissue. Results presented in Fig. 8 showed a signifi- of the macrophages, histiocytes and foreign body giant cells cantly higher repartition of both powders in the synovial lining cell populations(Fig. 10). Most of these cells contained an important layers than in sub-synovial tissue (p< 0.001). This amount intra-cytoplasmic particle amount. The scores of number of decreased quickly in the depth of the sub-synovial tissu macrophages and of the total particles amounts were positively The injection of particles led to some cellular and tissue modi- correlated in the sub-synovial tissue, all injected animal being fications of the synovial membrane. At first, we analyzed the taken together (Spearman correlation r=0.5105, p< 0.05 and synovial lining cell reaction. It was characterized by a slight Pearson correlation r= 0.5414, p<0.01). However, like in the and 9). Indeed, animals injected with particles showed a synovial lymphocytes infiltration and no fibrosis Vascularisation appeared ning cell composed with more cell layers(p< 0.01)and the cells to be slightly but not significantly increased( Fig. 10)but there was ere bigger or more cuboidal (p < 0.001 and p<0.05 for alumina a positive correlation between the vascularisation score and the nd zirconia respectively)compared to the control group(Fig 9). macrophage amount(Spearman correlation r=0.5413, p<0.001 The addition of the scores of cell layer hyperplasia and cell hyper- and Pearson correlation r=0.4845, p<0.001). More impo trophia constituted an adequate score for the evaluation of the there was a stronger correlation between the score of the synovial lining cell reaction. In this study, it was increased for lining cell reaction and the score of macrophage animals that received alumina or zirconia powders (p<0.001 and (Spearman correlation r=0.7435, p<0.001 and Pearson 0.01 respectively) and positively correlated with the score tion r=0.7522, p<0.001). All this results were in accordance corresponding to the total particle amount found in the synovial with a moderate non-specific granulomatous response of the lining cell(Spearman correlation r=0.5200, p< 0.01 and Pearson synovial tissue. lease cite this article in press as: Roualdes O, et al, In vitro and in vivo evaluation of an., Biomaterials(2009), doi: 10.1016/ j biomaterials 2009. 11.107were more present in the intracellular compartment (respectively 3.42  0.07 and 3.23  0.42) than extra-cellularly (respectively 2.49  0.15 and 1.84 1.13). Thereafter, we compared the scores of total particles amount in the synovial lining cell layers and in the sub-synovial tissue. Results presented in Fig. 8 showed a signifi- cantly higher repartition of both powders in the synovial lining cell layers than in sub-synovial tissue (p < 0.001). This amount decreased quickly in the depth of the sub-synovial tissue. The injection of particles led to some cellular and tissue modi- fications of the synovial membrane. At first, we analyzed the synovial lining cell reaction. It was characterized by a slight increase of the cell layer hyperplasia and cell hypertrophia (Figs. 7 and 9). Indeed, animals injected with particles showed a synovial lining cell composed with more cell layers (p < 0.01) and the cells were bigger or more cuboidal (p < 0.001 and p < 0.05 for alumina and zirconia respectively) compared to the control group (Fig. 9). The addition of the scores of cell layer hyperplasia and cell hyper￾trophia constituted an adequate score for the evaluation of the synovial lining cell reaction. In this study, it was increased for animals that received alumina or zirconia powders (p < 0.001 and p < 0.01 respectively) and positively correlated with the score corresponding to the total particle amount found in the synovial lining cell (Spearman correlation r ¼ 0.5200, p < 0.01 and Pearson correlation r0 ¼ 0.5151, p < 0.05). All synovial lining cell reactions observed were similar for both alumina and zirconia powders. In a second time, we evaluated the sub-synovial tissue reac￾tion. Rats who received particles showed a significantly increase of the macrophages, histiocytes and foreign body giant cells populations (Fig. 10). Most of these cells contained an important intra-cytoplasmic particle amount. The scores of number of macrophages and of the total particles amounts were positively correlated in the sub-synovial tissue, all injected animal being taken together (Spearman correlation r ¼ 0.5105, p < 0.05 and Pearson correlation r0 ¼ 0.5414, p < 0.01). However, like in the control group, animals injected with ceramic particles showed no lymphocytes infiltration and no fibrosis. Vascularisation appeared to be slightly but not significantly increased (Fig. 10) but there was a positive correlation between the vascularisation score and the macrophage amount (Spearman correlation r ¼ 0.5413, p < 0.001 and Pearson correlation r0 ¼ 0.4845, p < 0.001). More importantly, there was a stronger correlation between the score of the synovial lining cell reaction and the score of macrophage amount (Spearman correlation r ¼ 0.7435, p < 0.001 and Pearson correla￾tion r0 ¼ 0.7522, p < 0.001). All this results were in accordance with a moderate non-specific granulomatous response of the synovial tissue. Fig. 6. Fibroblasts (a, c, e) and osteoblasts (b, d, f) were cultured in DMEM (supplemented with 10% FBS, 1% penicillin/streptomycin) for 24 h. At that time, medium was replaced with fresh culture media containing either alumina (a, b), zirconia (c, d) or no particles (control: e, f) and cultured for 72 h under standard conditions. After this period, cells were fixed and stained with Coomassie blue brilliant in order to observe cell morphology and spreading (original magnification ¼ 40, arrows indicate particles localization and bars indicate 10 mm). 8 O. Roualdes et al. / Biomaterials xxx (2009) 1–12 ARTICLE IN PRESS Please cite this article in press as: Roualdes O, et al., In vitro and in vivo evaluation of an..., Biomaterials (2009), doi:10.1016/ j.biomaterials.2009.11.107