The impact of dissolution products and solution pH on in vitro behaviour of the bioactive glasses 45S5 and S53P4

Glass is a versatile material used in our everyday life. The chemical durability of glass depends on its composition. Some applications benefit from lower chemical durability. Glasses in a narrow compositional range are reactive in aqueous solutions and form hydroxyapatite on the surface. Such glasses are defined as bioactive glasses. Bioactive glasses are clinically used as a filler material due to diseases causing bone cavities. The formed hydroxyapatite bonds to bone apatite and new bone regenerates as the implanted bioactive glass dissolves. The dissolution occurs from the surface of the bioactive glass and is highly dependent on the environment. The environment inside and outside of a porous implant can differ (e.g., due to dissolution products) and the local environment can change depending on the implant site (e.g., fluid flow rate) or due to diseases (e.g., decrease of pH).

In vitro dissolution experiments are commonly conducted by immersing the bioactive glass in physiological-like solutions. Mimicking the circulating solutions in the human body in vitro contributes to an increased knowledge of the behaviour of bioactive glasses. Such experiments lead to better estimations of in vivo reactions, ultimately decreasing the number of in vivo tests needed before clinical applications.

This thesis investigated how dissolution products, fluid flow rate, and solution pH influence the in vitro behaviour of the bioactive glasses 45S5 and S53P4 in a dynamic environment. Also, the impact of solution pH and ion concentrations on the bioactive glass S53P4 in a static environment was studied. Dissolution experiments were performed in simulated body fluid, Tris buffer solution of different initial pH values, acetic acid-sodium hydroxide solution, and lactic acid. The impact of solution composition was studied using a cascade reactor, i.e., multiple reactors in series, to a continuous flow-through setup. The glass dissolution reactions were expressed as changes in the outflow solutions from each reactor and the glass samples in the reactors. Static tests were conducted by immersing the bioactive glass samples in extracts containing dissolution products from previous dissolutions. Also, two different flow rates were studied.

The reaction products, fluid flow rate, and pH influenced the in vitro reaction behaviour of the bioactive glasses. In the physiological pH range (7.4), the increasing ion concentrations hindered the bioactive glass dissolution, but typical reaction layers on the particle surfaces developed regardless. The results from the cascade reactors suggested that particles react differently depending on their location in a particle bed. Particles inside a bed or the interior of a porous implant would dissolve slower than the material on the outer surfaces. In contrast, particles in the bed dissolved similarly in lactic acid (pH 2), i.e., the dissolution products did not influence the release of alkali and alkaline earth ions.

Although the pH and ion concentrations increased with decreasing flow rate in the physiological pH range, the release rate normalised to the solution volume suggested slower release for the lower flow rate. In contrast, the fluid flow rate change did not similarly impact the release rate in the acidic solutions (pH 2). Thus, to avoid undesired cellular effects, the local fluid flow rates must be considered when tailoring bioactive glasses to clinical applications.

The key features of bioactive glasses are the silica-rich and hydroxyapatite layers that form on the surfaces in vitro and in vivo. The formation and nature of these reaction layers depended on the pH of the test solutions. Solutions with an alkaline pH (9) dissolved glass without typical reaction layers. Silicarich and calcium phosphate layers formed in dynamic and static solutions with a physiological-like pH (7.4). However, increased ion concentrations in the solutions resulted in mixed layers of silica and calcium phosphate. Static acidic pH (5) solutions resulted in a thick silica-rich layer with some calcium phosphate. In contrast, only a thick silica-rich layer formed in solutions with a lower pH (2).

The results of the thesis increased the knowledge of bioactive glass reactions and can be used to develop new compositions for new clinical applications. This thesis suggests that developing new experimental methods that better imitate the complex human body would further minimise the gap between in vitro and in vivo, ultimately decreasing the animal tests needed before clinical use.