Raman spectroscopy is a non-destructive analytical technique and previous results have shown that qualitative analysis of the lipid component of human atheromatous arteries is feasible. In this paper, we describe a quantitative analytical method for cholesterol and cholesteryl esters in human atherosclerotic plaques, combined with Raman spectroscopic results, using partial least-squares (PLS) regression, a statistical multivariate method based on factorial analysis. Twenty-nine human atherosclerotic pooled samples were studied and the results of Raman spectroscopy coupled with the PLS method were compared to biochemical results. The standard error of prediction was 16.1, 13.6, 1.9, 3.3 and 3.4 mg/g for total cholesterol, free cholesterol, palmitate cholesteryl, oleate cholesteryl and linoleate cholesteryl, respectively. The repeatability of Raman spectroscopy was found to be excellent. Our results show that Raman spectroscopy is a promising technique to obtain a consistent and non-destructive quantitative analysis of cholesterol and cholesteryl esters in human atherosclerotic lesions. In situ and in vivo analysis is a possibility in the near future.