Received on April 14, 2008
Accepted on August 7, 2008

General Clinical Chemistry

Effects of Delayed Sample Processing and Freezing on Serum Concentrations of Selected Nutritional Indicators

¹ National Center for Environmental Health, Centers for Disease Control and Prevention, Atlanta, GA
² Emory University, Atlanta, GA

^* To whom correspondence should be addressed. E-mail: CPfeiffer{at}cdc.gov.

BACKGROUND: Environmental conditions during sample processing, shipping, and storage are often suboptimal, particularly in less developed countries. We used samples from US volunteers to investigate the effects of delayed whole blood (WB) processing and delayed freezing of serum on selected nutritional indicators.

METHODS: WB tubes (n = 35) were either stored at 32 °C for up to 3 days before serum separation or centrifuged within 2 h of collection; serum samples were stored at 11 °C for up to 14 days to simulate delayed shipping. We assessed analyte stability by comparing results with data from optimally prepared/stored serum samples (<2 h on the clot, frozen at -70 °C) and by using clinical-acceptability criteria based on combined analytical imprecision and intraindividual biologic variability.

RESULTS: Clinically acceptable changes in concentration varied from 3%–15%. Delayed WB processing did not unacceptably affect concentrations of carotenoids and vitamins B₁₂, D, and E; however, we obtained clinically unacceptable changes for ferritin (+9%), soluble transferrin receptor (sTfR) (+5%), and folate (-30%) after 1 day, and for vitamin A (-10%) after 3 days. Delayed freezing of serum did not affect concentrations of ferritin, sTfR, carotenoids, and vitamins A, B₁₂, and E; however, we obtained clinically unacceptable changes for vitamins C (-20%) and D (+7%) after 7 days and for folate after 14 days (-22%).

CONCLUSIONS: Despite substantial delays in WB processing or in the freezing of serum samples, most nutritional indicators showed remarkable stability. This information is important for both the design of field studies and the use of residual samples subjected to suboptimal preanalytical factors.