The results of the in vitro tests indicate that adding increasing quantities of tricalcium phosphate to the lactic acid polymer stimulates both the proliferation of human osteogenous cells and the synthesis of the extracellular matrix. Cell density increased with the percentage of β–TCP in the composite, showing that cell proliferation was promoted by the presence of β–TCP. The results of the collagen assays in the extracellular matrix showed clearly that the quantity of collagen per cell increased in the same way, indicating that the β–TCP in the polymer matrix stimulated cell activity. The composite material with the highest mineral content (60 % by mass) showed the best behavior. We know that healing and consolidation of the bone by endochondral ossification are directly linked to the recruitment and phenotype expression of osteogenous cells at the site. Composite materials thus seem more favorable for bone consolidation than pure polymer materials and have been shown in vitro to improve biological behavior. These results confirm hypotheses on PLA/TCP [26–28, 37] materials and, more generally, the absorbable polyester/calcium phosphate composites presented in the literature [22–25, 29–32, 38], where the presence of minerals in the polymer matrix promoted the proliferation and phenotype expression of osteogenous cells in vitro. These findings were also corroborated by the histological examinations, which indicated clearly that pure PLA implants were systematically surrounded by a layer of conjunctival tissue containing large numbers of macrophages, irrespective of the length of implantation, whereas bone tissue grew in direct contact with composite implants during the first few months and matured gradually until the end of the study period with no marked macrophage activation. The composite material under investigation is thus more favorable for bone consolidation in vivo than pure PLA and has identical properties to those of pure β–TCP ceramics.

This result is not particularly surprising if we consider the properties of the materials in the composite separately. We know that pure lactic acid-based polymers are not cytotoxic or inflammatory in vitro [10–13, 39, 40] but are capable of causing an inflammatory reaction in vivo [9–13, 15], as confirmed by clinical observations [16–21]. Pure β–TCP, on the contrary, promotes bone healing by osteoconduction, becoming surrounded by healthy bone tissue containing active osteogenous cells, with no intermediate fibrous layer. It is, thus, conceivable that composite materials such as those we studied have a “hybrid” biological behavior that becomes closer to that of pure tricalcium phosphate as the mineral content increases. What mechanisms are involved?

Pure lactic acid polymer implants are absorbed slowly, over several years, depending on the volume of the implant [19, 41]. The main mechanism is internal hydrolysis [42], resulting in the sudden release of partially degraded polymer and monomer particles [9] at the end of the resorption process, which acidifies the peri-implant region [43]. During this phase, large numbers of active macrophages, containing particles resulting from hydrolytic degradation, are found within fibrous inflammatory tissue on the implant surface [10–15, 19]. If no macrophage activation mechanism is established it may be assumed that the monomer concentration [9, 43] (drop in pH) and the number of fine “phagocytizable” [44] particles contribute to this reaction. We observed significant resorption of composite material implants, and thus of polymer, without any inflammation. At 6 months, the surface was eroded to a depth of several hundred micrometers but there were only very few macrophages, some of which contained polymer particles. It cannot, therefore, be considered that activation of the cells during resorption of the polymer matrix was solely due to the presence of these particles. We know that the acidification that accompanies the degradation of lactic acid polymers may also set off an inflammatory reaction [43]. In a systematic in vitro study, Lin [28] measured the pH of solutions containing a variety of PDLLA/TCP composite materials. The pH stabilized at 2.3 for a pure polymer material and 5 for all the materials containing mineral, which shows the effect of β–TCP on the acidity of the solution. This observation was interpreted by the author as a result of the slower degradation kinetics of composite materials as compared to pure polymers. However, other authors have confirmed our histological finding that degradation kinetics increased with a higher mineral content [23, 26, 38], which leads us to observe that β–TCP causes an increase in both the degradation kinetics of the composite material and the pH of the solution. This may be interpreted by noting that the TCP particles increase the permeability of the material, resulting in a more homogeneous hydrolysis of the polymer matrix in a liquid medium, whereas hydrolysis of pure polymer materials starts from the surface. β–TCP is alkaline in solution and may neutralize the acid functions of the hydrolyzed polymer chains. It is also soluble in an acid medium and dissolves in the hydrolyzed polymer, thus buffering the acidification of the medium. This may also account for the high level of osteoblast activity observed, as human osteogenous cells are stimulated by the dissolution products of materials containing calcium and phosphate ions, like β–TCP [13, 30]. More generally, osteoblast cells activity and healing time may be linked to the concentration of dissolved calcium and phosphate ions and, thus, the dissolution kinetics of ionic materials. This interpretation accounts for our observations and is in agreement with widespread findings in the literature.