Introduction: We recently showed that Amyloid Beta (A beta)(40) accumulates in erythrocytes and possibly causes cell damage as evidenced by an increased number of assumed injured low-density (kg/L) erythrocytes. Furthermore, we have suggested a separation technique to isolate and concentrate such damaged red blood cells for subsequent analysis. Objectives: We isolated high- and low-density erythrocytes and investigated the accumulation patterns of the A beta peptides (A beta(40), A beta(42), and A beta(43) ) in Alzheimer (AD), mild cognitive impairment (MCI), and Subjective Cognitive Impairment (SCI). Methods: Whole blood was fractionated through a density gradient, resulting in two concentrated high-and presumed injured low-density erythrocyte fractions. After cell lysis, intracellular A beta(40) , A beta 4(2), and A beta (43) were quantified by ELISA. Results: In both high- and low-density erythrocytes, A beta(40) displayed the lowest concentration in MCI, while it was equal and higher in AD and SCI. A beta(40) was detected at a 10-fold higher level than A beta(42), and in injured low-density erythrocytes, the lowest quantity of A beta(42) was found in AD and MCI. A beta(40) exhibited a 100-fold greater amount than A beta(43). and lighter erythrocytes of MCI subjects displayed less intracellular A beta(43) than SCI. Conclusion: Red blood cell accumulation patterns of A beta(40), A beta(42), and A beta(43) differ significantly between AD, MCI, and SCI. The data must be verified through larger clinical trials. It is, however, tenable that AP peptide distributions in erythrocyte subpopulations have the potential to be used for diagnostic purposes.