The effects of surface processing on in-vivo corrosion of Nitinol stents in a porcine model.

UNLABELLED

A major limitation with current assessments of corrosion in metallic medical devices is the lack of correlation between in-vitro and in-vivo corrosion performance. Therefore, the objective of this study was to elucidate the relationship between pitting corrosion measured by breakdown potentials (E) in ASTM F2129 testing and corrosion resistance in-vivo. Four groups of Nitinol stents were manufactured using different processing methods to create unique surface properties. The stents were implanted into iliac arteries of minipigs for six months and explanted for corrosion analysis. Scanning electron microscopy and energy dispersive X-ray spectrometry analyses indicated that stents with a thick complex thermal oxide (420nm) and high corrosion resistance in-vitro (E=975±94mV) were free from detectable corrosion in-vivo and exhibited no changes in Ni/Ti ratio when compared to non-implanted controls. This result was also found in mechanically polished stents with a thin native oxide (4nm; E=767±226mV). In contrast, stents with a moderately thick thermal oxide (130nm) and low corrosion resistance in-vitro (E=111±63mV) possessed corrosion with associated surface microcracks in-vivo. In addition, Ni/Ti ratios in corroded regions were significantly lower compared to non-corroded adjacent areas on explanted stents. When stents were minimally processed (i.e. retained native tube oxide from the drawing process), a thick thermal oxide was present (399nm) with low in-vitro corrosion resistance (E=68±29mV) resulting in extensive in-vivo pitting. These findings demonstrate that functional corrosion testing combined with a detailed understanding of the surface characteristics of a Nitinol medical device can provide insight into in-vivo corrosion resistance.

STATEMENT OF SIGNIFICANCE

Nitinol is a commonly used material in the medical device industry. However, correlations between surface processing of nitinol and in-vivo corrosion has yet to be established. Elucidating the link between in-vivo corrosion and pre-clinical characterization can aid in improved prediction of clinical safety and performance of nitinol devices. We addressed this knowledge gap by fabricating nitinol stents to possess distinct surface properties and evaluating their corrosion susceptibility both in-vitro and after six months of in-vivo exposure. Relationships between stent processing, surface characterization, corrosion bench testing, and outcomes from explanted devices are discussed. These findings highlight the importance of surface characterization in nitinol devices and provide in-vitro pitting corrosion levels that can induce in-vivo corrosion in nitinol stents.