Exploring magnesium corrosion in a medical environment

Abstract

Magnesium (Mg) and its alloys are at the forefront of an emerging series of biomaterials know as biodegradable metals. The ability to safely degrade in the body makes them ideal for temporary medical applications. Mg alloys have already shown good clinical success in orthopaedic and cardiovascular applications, suggesting this field is maturing from research consideration into a clinical reality.However, there is still considerable work required to fully characterise, understand, and control the rate of corrosion in the medical environment (i.e. biocorrosion). If the rate of biocorrosion is too high, the implant may fail before the body has finished healing. There are also material specific issues, such as the too rapid release of hydrogen gas, a by-product of Mg corrosion.As such, this thesis aims explore the corrosion of Mg alloys in a medical environment. A detailed literature survey considered the methods by which laboratory immersion tests (i.e. in vitro) attempt to replicate or mimic the corrosion from live animal or clinical studies (i.e. in vivo). A number of recommendations were made for the composition of the immersion solution, and the experimental parameters applied during the test (i.e. solution temperature), in an effort to improve the standard of in vitro tests and make them more reliable and comparable to in vivo biocorrosion.Following these recommendations, it was desirable to gather some in vivo biocorrosion data for comparison to in vitro data. However, before the samples could be implanted they first need to be sterilised. Two potential sterilisation techniques were applied to four Mg alloys, and their effect on the biocorrosion rate was measured in vitro. As neither gamma irradiation nor ethylene oxide influence the corrosion of the Mg alloys, these were deemed applicable techniques for future use.Specimens of three Mg alloys were sterilised with gamma irradiation and implanted subcutaneously into male and female Sprague-Dawley rats for either one or four weeks. The specimens were CT-imaged prior to sacrifice to determine the amount of hydrogen gas in a pocket around the sample. The corrosion rate and the size of the hydrogen pockets decreased with increasing implantation time. The corrosion rate for two of the alloys was also significantly higher in the female rats, as compared to the male rats. The female rats in these groups were also associated with a higher incidence of localised corrosion. In general, however, the corrosion was quite uniform across the samples surfaces.The results from this in vivo trial were compared to two common simulated body fluids (SBFs) in vitro. Hanks’ Balanced Salt Solution (HBSS) was the superior model, attributed to the fact that HBSS contains PO42-, which forms an essential part of the corrosion product layer in vivo. HBSS was selected as the basis for the second part of the in vitro trial, which introduced an amino acid, and protein, in concentrations similar to solutions in the body, into the solution to study their effect. It was expected that making the solution more similar to the bodily fluids would yield better comparisons to in vivo data. In fact, the opposite occurred. Both the amino acids, and the protein, increased the corrosion rate by a factor of ~3, and produced significantly more localised corrosion. However, it was also noted that the subcutaneous in vivo environment did not contain a body of solution but instead could be considered ‘damp’ or ‘wet’, so that the in vivo environment was not equivalent to total immersion of the specimen in a body solution. A smaller volume of solution would be expected to decrease the rate of corrosion in vivo. One of the alloys studied (ZX00) was formed into wire, in order to study Mg alloys in a new application: as surgical tacks. However, cold drawing the extruded ZX00 into wire introduced iron (Fe) impurity particles into the surface of the specimens. These impurity particles increased the corrosion rate, and a variety of techniques were considered in an effort to remove them. Mechanically cleaning the wire with SiC grinding paper was found to reduce the corrosion rate and produce an even surface finish.This wire was then formed into surgical tacks, a device commonly used during laparoscopic hernia repair. The prototype absorbable Mg tack was compared against commercially available Titanium (ProTacktm), and PLGA (AbsorbaTacktm) tacks. The fixation strength of the tacks was tested in situ by fixing strips of mesh to porcine abdominal muscle tissue. The Mg tack performed statistically similarly to both commercial tacks.Finally, a reflection on the in vitro assessment of Mg biocorrosion was conducted, with a focus on practical experience and the associated lessons that could be drawn from the experiences. A series of updated recommendations and thoughts are presented, in an effort to improve future in vitro immersion testing.