Plasma amyloid-β peptide (Aβ) levels have been suggested as a biomarker candidate for detecting incipient AD. Aβ peptides are known to be sensitive to distinct preanalytical sample handling, which calls for standardised preanalytical procedures. We investigated serum and plasma samples of 19 patients with no clinical signs of dementia for different preanalytical sample handlings. Both serum and plasma were analysed by the one-dimensional Aβ-SDS-PAGE/immunoblot, either immediately or after storage at room temperature for 24 and 48 h, respectively. The panel of Aβ1–37/38/39/40/42 and Aβ2–40 was evaluated. In both analytical matrices, sample storage led to a significant loss of measurable peptide levels. This effect was most pronounced during the first 24 h of storage and stronger in serum than in plasma. There were no significant differences between the distinct analysed Aβ peptide species regarding these results. The ratios of peptides (e.g. Aβ1–42/Aβ1–40 and Aβ1–42/Aβ1–38) displayed a higher stability under the influence of storage than each single peptide. In conclusion, plasma may be more appropriate than serum for analysing Aβ peptides for routine application. At least, the analysis should be done within 24 h and peptide ratios should be created to minimise artificial results.