DEVELOPING CORRELATIONS BETWEEN FIBRINOGEN ADSORPTION AND SURFACE-PROPERTIES USING MULTIVARIATE-STATISTICS

A multivariate model based on the partial least squares algorithm (PLS ) was constructed in order to establish a correlation between the surf ace properties of common polymeric materials and the amount and retent ion of fibrinogen adsorbed from a complex mixture. Surface characteriz ation was performed by means of static secondary ion mass spectroscopy (SIMS), electron spectroscopy for chemical analysis (ESCA), and by co ntact angle measurements of several liquids on those materials. I-125- fibrinogen was adsorbed from a 1% plasma solution in buffer and the am ount adsorbed after 2 h was determined. After 5 days of residence time in buffer, the adsorbed fibrinogen was eluted with a 1% solution of t he surfactant sodium dodecyl sulfate (SDS). The percent of fibrinogen that remained on the surfaces after elution is referred to as fibrinog en retention. Correlations between surface properties and the amounts of fibrinogen adsorbed or fibrinogen retention were established. These models also show the most important variables that are related to the protein behavior on these surfaces. (C) 1994 John Wiley & Sons, Inc.