Magnesium screws have been clinically researched as an alternative for bone fixation with positive results for decades, in particular over the last 10 years [1], [2], [3], [4], [5], [6], [7]. Magnesium alloys possess many advantages both physically (stiffness, density, degradability in vivo) and biologically (biocompatibility, oseopromotion) for use in the field of bone healing, as reviewed by [8]. As a resorbable osteosynthesis material, magnesium should provide mechanical stability to the fracture during the initial phase of ossification and be reabsorbed at adequate speed, allowing bone remodeling [3,9]. However, once implanted magnesium degradation generates byproducts such as hydrogen gas [10], which in high concentrations can slow down the bone healing process (as reviewed by [11]).

Commonly it is assumed, that magnesium degradation does not significantly interfere with the bone healing process, as long as the degradation rate is moderate in the first weeks of implantation [9]. The ideal rate is yet to be defined, although the faster the degradation, the more bone cavities are formed in the surrounding bone, caused by the rapid expansion of hydrogen gas exceeding tissue diffusion volume [11]. On the other hand, there is a clinical demand for full bioabsorption of implants in certain indications, for example in the growing skeleton or in patients where a second operation is needed. Hence, magnesium degradation needs to be evaluated at least over most of the implant's lifetime to reliably predict the outcomes of different degradation rates on long-term bone healing. With this knowledge, the right alloying and surface conditions based on the indications to be addressed can be chosen.

Most magnesium alloys degrade at a slower pace than pure magnesium [12]]. A lean ZX00 magnesium alloy with a composition of Mg–0.45Zn–0.45Ca (in wt%) has shown slower and uniform degradation in both in vivo [3,13] and clinical studies [4]. However, finely dispersed temporary bone cavities can still be observed surrounding ZX00 screws [3,4,13], which can locally but mostly temporarily reduce biomechanical strength. Therefore, there is still a need to improve or rather shape the degradation process of ZX00 alloys to reduce or delay temporary interactions of the degradation process with the surrounding bone. The goal should be to allow for wider medical use where prolonged healing phases are needed.

(Electro-)chemical surface modification of magnesium is a promising field in the search for slower degradation rates. Plasma electrolytic oxidation (PEO), also called micro-arc oxidation (MAO), is an enhanced anodic oxidation process facilitated by short-lived plasma ignitions in an aqueous electrolyte and has already been applied successfully on different durable or even bioabsorbable metallic devices in the biomedical field [14]. The resulting surfaces can be designed to be bioinert to bioactive and tailored in roughness or porosity [15], which in right combination significantly improves the osteointegration and stability of devices such as screws or plates [16]. Furthermore, it was shown that PEO effectively decreases the degradation rate of Rare Earth based as well as other magnesium alloys, however not to an extent that the mere influence of the magnesium base alloy becomes subordinated in vivo [17]. In vitro, PEO has shown to increase the corrosion resistance of a magnesium alloy by manifold orders of magnitude [18], which shows PEO's potential for use with bioabsorbable magnesium alloys.

The present work aims to observe and evaluate the degradation behavior of ZX00 magnesium-based screws used for craniomaxillofacial (CMF) and orthopedic applications for the first time ever over a prolonged period of 18 months in a large animal model (forehead bones of minipigs) and to compare the different kinetics in degradation and bone healing between PEO surface-modified and unmodified implants manufactured from this alloy. Göttingen miniature pig, a research dedicated breed with close tissue similarity and scalable metabolism compared to humans was selected as in vivo model for the presented study accordingly [19,20].