Identification of an important potential confound in CSF AD studies: aliquot volume

Background: Cerebrospinal fluid (CSF) amyloid beta 1-42 (A beta 1-42), total tau (T-tau) and phosphorylated tau181 (P-tau) are finding increasing utility as biomarkers of Alzheimer's disease (AD). The purpose of this study was to determine whether measured CSF biomarker concentrations were affected by aliquot storage volume and whether addition of detergent-containing buffer mitigates any observed effects. Methods: AD and control CSF was distributed into polypropylene tubes in aliquots of different volumes (50-1500 mu L). A beta 1-42, T-tau and P-tau were measured with and without addition of Tween 20 (0.05%). Results: Measured concentrations of A beta 1-42 increased two-fold with aliquot storage volume. A volume increase of 10 mu L caused an A beta 1-42 increase of 0.95 pg/mL [95% confidence interval (CI) 0.36-1.50, p=0.02] in controls, and 0.60 pg/mL (CI 0.23-0.98 pg/mL, p=0.003) in AD samples. Following addition of Tween 20, the positive relationship between A beta 1-42 and aliquot volume disappeared. T-tau and P-tau were not significantly affected. Conclusions: CSF aliquot storage volume has a significant impact on the measured concentration of A beta 1-42. The introduction of a buffer detergent at the initial aliquoting stage may be an effective solution to this problem.