Crevice or pitting corrosion of surgical implant metals is generally associated with the presence and activity of the chloride ion.' Because of the ubiquitous nature of the chloride ion in vivo, corrosion testing of materials for implantation is generally performed in physiologic saline solution^.^-^ However, the internal milieu is a complicated assortment of enzymes, proteins, and metabolically active cells in an electrolyte containing chloride ions. Because of our interest in metal sensitivity which is a reaction to metal-protein c~mplexes,~ we became interested in the question of whether the presence of proteins in an electrolyte solution would influence the corrosion processes. Samitz and Katz6 have shown that in some cases the solubilization of nickel from stainless steel is greater in whole blood than in physiologic saline. The results of electrochemical corrosion experiments presented in this report indicate that the presence of proteins can influence both the rates and'mechanisms of corrosion of surgical-grade stainless steel.
Accelerated corrosion tests were performed using an apparatus similar to that previously described.' One-centimenter segments of 3.2 mm diam rods were pushed through a polyethylene gasket in the bottom of a plastic beaker. The test specimen was made anodic to a stainless-steel strip with a dc power supply. Current flow and applied potential were measured between the specimen and the strip. Stainless-steel (316L) rods in the coldworked (Steinman pins from Richards Mfg. Co.) and annealed (Allegheny-Ludlum steel) condition were used wlth the surface electropolished or 600-grit-sanded. Metal loss was determined by weight with an analytic balance. The solutions used were USP saline (Abbott) to which calf serum (GIBCO) was added to a concentration of 0, 1, or 10%. The concentration of nickel in the solutions was measured using dimethylglioxime in accordance with ASTM D-1886.8