INTRODUCTION: Processing delays after blood collection is a common pre-analytical condition in large epidemiologic studies. It is critical to evaluate the suitability of blood samples with processing delays for metabolomics analysis as it is a potential source of variation that could attenuate associations between metabolites and disease outcomes. OBJECTIVES: We aimed to evaluate the reproducibility of metabolites over extended processing delays up to 48 h. We also aimed to test the reproducibility of the metabolomics platform. METHODS: Blood samples were collected from 18 healthy volunteers. Blood was stored in the refrigerator and processed for plasma at 0, 15, 30, and 48 h after collection. Plasma samples were metabolically profiled using an untargeted, ultrahigh performance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS) platform. Reproducibility of 1012 metabolites over processing delays and reproducibility of the platform were determined by intraclass correlation coefficients (ICCs) with variance components estimated from mixed-effects models. RESULTS: The majority of metabolites (approximately 70% of 1012) were highly reproducible (ICCs ≥ 0.75) over 15-, 30- or 48-h processing delays. Nucleotides, energy-related metabolites, peptides, and carbohydrates were most affected by processing delays. The platform was highly reproducible with a median technical ICC of 0.84 (interquartile range 0.68-0.93). CONCLUSION: Most metabolites measured by the UPLC-MS/MS platform show acceptable reproducibility up to 48-h processing delays. Metabolites of certain pathways need to be interpreted cautiously in relation to outcomes in epidemiologic studies with prolonged processing delays.
INTRODUCTION: Processing delays after blood collection is a common pre-analytical condition in large epidemiologic studies. It is critical to evaluate the suitability of blood samples with processing delays for metabolomics analysis as it is a potential source of variation that could attenuate associations between metabolites and disease outcomes. OBJECTIVES: We aimed to evaluate the reproducibility of metabolites over extended processing delays up to 48 h. We also aimed to test the reproducibility of the metabolomics platform. METHODS: Blood samples were collected from 18 healthy volunteers. Blood was stored in the refrigerator and processed for plasma at 0, 15, 30, and 48 h after collection. Plasma samples were metabolically profiled using an untargeted, ultrahigh performance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS) platform. Reproducibility of 1012 metabolites over processing delays and reproducibility of the platform were determined by intraclass correlation coefficients (ICCs) with variance components estimated from mixed-effects models. RESULTS: The majority of metabolites (approximately 70% of 1012) were highly reproducible (ICCs ≥ 0.75) over 15-, 30- or 48-h processing delays. Nucleotides, energy-related metabolites, peptides, and carbohydrates were most affected by processing delays. The platform was highly reproducible with a median technical ICC of 0.84 (interquartile range 0.68-0.93). CONCLUSION: Most metabolites measured by the UPLC-MS/MS platform show acceptable reproducibility up to 48-h processing delays. Metabolites of certain pathways need to be interpreted cautiously in relation to outcomes in epidemiologic studies with prolonged processing delays.
Entities:
Keywords:
Human plasma; Mass spectrometry; Metabolomics; Pre-analytical; Processing delay
Authors: Alexandra C Schrimpe-Rutledge; Simona G Codreanu; Stacy D Sherrod; John A McLean Journal: J Am Soc Mass Spectrom Date: 2016-09-13 Impact factor: 3.109
Authors: Joo Ern Ang; Victoria Revell; Anuska Mann; Simone Mäntele; Daniella T Otway; Jonathan D Johnston; Alfred E Thumser; Debra J Skene; Florence Raynaud Journal: Chronobiol Int Date: 2012-08 Impact factor: 2.877
Authors: Jared R Mayers; Chen Wu; Clary B Clish; Peter Kraft; Margaret E Torrence; Brian P Fiske; Chen Yuan; Ying Bao; Mary K Townsend; Shelley S Tworoger; Shawn M Davidson; Thales Papagiannakopoulos; Annan Yang; Talya L Dayton; Shuji Ogino; Meir J Stampfer; Edward L Giovannucci; Zhi Rong Qian; Douglas A Rubinson; Jing Ma; Howard D Sesso; John Michael Gaziano; Barbara B Cochrane; Simin Liu; Jean Wactawski-Wende; JoAnn E Manson; Michael N Pollak; Alec C Kimmelman; Amanda Souza; Kerry Pierce; Thomas J Wang; Robert E Gerszten; Charles S Fuchs; Matthew G Vander Heiden; Brian M Wolpin Journal: Nat Med Date: 2014-09-28 Impact factor: 53.440
Authors: Nicole H Tobin; Aisling Murphy; Fan Li; Sean S Brummel; Taha E Taha; Friday Saidi; Maxie Owor; Avy Violari; Dhayendre Moodley; Benjamin Chi; Kelli D Goodman; Brian Koos; Grace M Aldrovandi Journal: Metabolomics Date: 2021-06-23 Impact factor: 4.747