OBJECTIVE: Biobank material is frequently used in epidemiological studies, but long-term storage of serum at -25 degrees C may reduce the quality of the samples. Knowledge about the stability of components in biological samples is fundamental for the interpretation of such studies. MATERIAL AND METHODS: We investigated the stability of seven biological components in serum samples stored at -25 degrees C for 25 and 2 years compared with 1-month-old samples. Specimens from 130 blood donors from each group were randomly selected among men without a cancer diagnosis during the follow-up time. We compared the distribution, dispersion and localization of medians and means, and established reference intervals of the components. RESULTS: The study demonstrated non-significant and numerically small differences in the levels of sodium, calcium, iron and creatinine over time. Differences between mean values for uric acid (-7.6%), potassium (+26.4%) and bilirubin (-59.4%) between fresh and 25-year-old samples indicated that sample handling before freezing and degradation during long-term storage may introduce a considerable bias for vulnerable components. CONCLUSIONS: There is large variation in level stability between different serum components in serum stored at -25 degrees C. Differences in sample handling before freezing may introduce bias on vulnerable components. The present study supports a routine of careful matching of cases and controls on storage time in epidemiological studies when biobank material is used.
OBJECTIVE: Biobank material is frequently used in epidemiological studies, but long-term storage of serum at -25 degrees C may reduce the quality of the samples. Knowledge about the stability of components in biological samples is fundamental for the interpretation of such studies. MATERIAL AND METHODS: We investigated the stability of seven biological components in serum samples stored at -25 degrees C for 25 and 2 years compared with 1-month-old samples. Specimens from 130 blood donors from each group were randomly selected among men without a cancer diagnosis during the follow-up time. We compared the distribution, dispersion and localization of medians and means, and established reference intervals of the components. RESULTS: The study demonstrated non-significant and numerically small differences in the levels of sodium, calcium, iron and creatinine over time. Differences between mean values for uric acid (-7.6%), potassium (+26.4%) and bilirubin (-59.4%) between fresh and 25-year-old samples indicated that sample handling before freezing and degradation during long-term storage may introduce a considerable bias for vulnerable components. CONCLUSIONS: There is large variation in level stability between different serum components in serum stored at -25 degrees C. Differences in sample handling before freezing may introduce bias on vulnerable components. The present study supports a routine of careful matching of cases and controls on storage time in epidemiological studies when biobank material is used.
Authors: Grant R Aitken; Paul J Roderick; Simon Fraser; Jennifer S Mindell; Donal O'Donoghue; Julie Day; Graham Moon Journal: BMJ Open Date: 2014-09-29 Impact factor: 2.692
Authors: Joy Simpson; Andrew D Smith; Abigail Fraser; Naveed Sattar; Mark Callaway; Robert S Lindsay; Debbie A Lawlor; Scott M Nelson Journal: J Clin Endocrinol Metab Date: 2016-09-20 Impact factor: 5.958
Authors: A Mander; C Gouttefangeas; C Ottensmeier; M J P Welters; L Low; S H van der Burg; C M Britten Journal: Cancer Immunol Immunother Date: 2010-01-06 Impact factor: 6.968