Improving long-term stability of retinol in dried blood spots and quantification of its levels via a novel LC-MS/MS method

Abstract

Vitamin A deficiency (VAD) is a major micronutrient deficiency in children. Although plasma and serum retinol levels are proposed as the key indicators of VAD, collecting and transporting plasma and serum are difficult and inconvenient in field studies. Dried blood spot (DBS) retinol has been used as an alternative to plasma retinol in several epidemiological and clinical studies. A limitation of methods that use DBS retinol is the instability and apparent loss of retinol in DBSs. Therefore, an accurate, reliable method for stabilizing retinol in DBSs and quantifying and comparing DBS retinol concentrations with equivalent plasma retinol levels is required. In this study, antioxidants on paper combined with vacuum treatment were found to greatly increase the stability of DBS retinol during 120 min of air drying and 30 days of room-temperature storage. A surrogate matrix of whole blood prepared using a mixture of human erythrocytes and 2% BSA in PBS was firstly used in DBS retinol determination based on the fact that retinol is excluded from erythrocytes. The method was linear in the concentration range of 0.04–300 μg/mL. Both the between-run (n = 5) and within-run (n = 6) precision (relative standard deviations, RSD%) were below 8.42%. The spiked recoveries at 3 concentrations ranged from 86.48 to 98.13%. The internal standard (IS)–normalized matrix factor (MF) was 99.72% with a RSD% of 10.50% (n = 3). The accuracy was calibrated using two National Institute of Standards and Technology (NIST) serum-generated calibrants at concentrations of 0.1962 and 0.3948 g/mL, and relative errors (RE% values) of 0.07% and 4.95% were found, respectively. A simple calibration model was first developed to convert DBS retinol concentration to the equivalent plasma retinol concentration, thereby enabling comparisons with clinical reference ranges and with studies using serum or plasma samples.