Skip to main content

Quality Control of Serum and Plasma by Quantification of (4E,14Z)-Sphingadienine-C18-1-Phosphate Uncovers Common Preanalytical Errors During Handling of Whole Blood.

Liu X, Hoene M, Yin P, Fritsche L, Plomgaard P, Hansen JS, Nakas CT, Niess AM, Hudemann J, Haap M, Mendy M, Weigert C, Wang X, Fritsche A, Peter A, Häring HU, Xu G, Lehmann R

Publication: Clin Chem (2018)PMID: 29567661View on PubMed

Purpose

This paper identified sphingosine-1-phosphate (S1P-d18:2) and lactate as biomarkers for blood processing delays in plasma and investigated the effects of delayed centrifugation, exercise, and patient condition on these biomarkers. The effects of clot time, clot activator, post-clot storage, patient condition, and use of a specialized collection team on levels of S1P-d18:2 in serum were also investigated.

Conclusion

UPLC-MS analysis identified a total of 188 ion masses that could be used as biomarkers for delayed processing at room temperature of which sphingosine-1-phosphate (S1P-d18:2) and lactate were the most promising candidates. Both S1P-d18:2 and lactate in plasma increased with storage of blood at room temperature for 2 h or 4 h and lactate levels increased with storage at 4˚C for 4 h. The authors found threshold values of ≤0.085 µg/mL of S1P-d18:2 and ≤1.306 µg/mL of lactate to be indicative of high quality plasma. Although S1P-d18:2 levels were unaffected by exercise and remained within the expected ranges for high quality plasma for patients undergoing extreme physiologic stress, lactate levels were higher than the threshold in some ICU specimens and in healthy patients after circuit training to exhaustion. Levels of S1P-d18:2 did not differ between plasma stored for 6 months and that stored for 5 years.

Levels of S1P-d18:2 were significantly higher in serum collected without clot activator that was allowed to clot for 60 min than in serum collected with clot activator and allowed a 30 min clot time.  In specimens collected without clot activator, storage for 2 or 4 h after the initial 60 min clot time resulted in a significant increase in S1P-d18:2 levels. Comparison of S1P-d18:2 levels in serum specimens in a biobank showed that those levels of S1P-d18:2 were lower and more consistent when collected by a specialized team than when collected by the clinician. Importantly, in the 869 serum specimens collected at eight hospitals for seven different biobanks using the same SOP had similar levels of S1P-d18:2 and levels were comparable among the patient diagnoses.

Studies (2)

Purpose

This study identified lactate and S1P-d18:2 as biomarkers of delayed plasma processing and investigated the effects of patient condition and exercise on the suitability of these biomarkers. Blood was drawn into 10 mL EDTA tubes from 109 healthy patients, 6 patients with severe sepsis, 6 patients with liver cirrhosis, 1 patient with liver cirrhosis and sepsis, and 4 patients who had undergone cardiopulmonary resuscitation.  Specimens were immediately split into 5 aliquots and stored for 2 or 4 h at room temperature (22˚C) or on ice (4˚C). After storage, plasma was obtained by centrifugation at 3100 x g at 4˚C, aliquoted, and stored at -80˚C. Proteins were identified by UPLC-Triple Quadrupole MS. Biomarkers were identified using the blood from 30 healthy patients and validated using the blood from 79 healthy individuals. The effects of extreme physiological condition were investigated using specimens from patients with severe sepsis and/or liver cirrhosis and patients who had undergone cardiopulmonary resuscitation. The effect of exercise was investigated by collecting specimens from seven healthy patients before and immediately after exercise to complete exhaustion and from ten patients before and after 1, 2, and 3 h of cycling at 50% maximum oxygen volume. The ability of S1P-d18:2 to serve as a biomarker for processing delay was further tested using plasma from three different biobanks, two of which processed blood in accordance with an SOP that required immediate cooling of blood to 4˚C followed by plasma isolation within 2 h (197 specimens total) and the third biobank which specified plasma isolation as soon as possible (99 specimens). The effects of long term storage of plasma were evaluated in 50 plasma specimens isolated within 2 h of collection and stored at -80˚C for 5 years.

Summary of Findings

Pre-centrifugation storage at room temperature for 2 h and 4 h affected the levels of 9.2% and 16.1% of identified ion masses, respectively, compared to specimens processed immediately. In contrast, pre-centrifugation storage at 4˚C affected the levels of only 0.5% and 1.8% of ion masses after 2 and 4 h, respectively. Analysis showed that there were a total of 188 ion masses that could be used as biomarkers for delayed processing at room temperature, of which, S1P-d18:2 and lactate were the most promising candidates. Further investigation of these biomarkers in the plasma of 79 healthy individuals showed an expected significant increase in S1P-d18:2 and lactate with storage at room temperature for 2 h or 4 h (P<0.0001, all), but also found a significant increase in lactate levels when specimens were stored at 4˚C for 4 h (P<0.002). The authors identified the cut-off values for good quality plasma to be ≤0.085 µg/mL of S1P-d18:2 and ≤1.306 µg/mL of lactate. Although S1P-d18:2 levels remained below the cut-off values for good plasma in patients who had undergone extreme physiologic stress and in healthy patients, post-exercise; lactate levels were higher than the cut-off in some ICU patient specimens and specimens from individuals collected post-circuit training to exhaustion, making it a poor biomarker for specimen processing. In all 199 plasma specimens from the two biobanks that processed blood within 2 h, S1P-d18:2 levels were below the established cut-off value, indicating high quality, but many plasma specimens from the biobank in which processing delays were as long as 9 h had levels of S1P-d18:2 above the quality threshold. Importantly, there were no differences in S1P-d18:2 levels between plasma stored 6 months and that stored for 5 years. S1P-d18:2 levels were lower in ICU patients than in healthy patients. S1P-d18:2 levels were modestly correlated with hematocrit (r2=0.45, P=0.003), indicating that lower levels of S1P-d18:2 in ICU patients could be attributed to anemia. Levels of S1P-d18:2 in the NIST standard reference for plasma were below the threshold, indicating high quality.

  • Biospecimens

    FluidBlood
    FluidPlasma
  • Preservation Types

    Frozen
  • Diagnoses

    Normal
    Other diagnoses
    Cirrhosis
  • Platforms

    ProteinLC-MS or LC-MS/MS
  • Pre-analytical Factors

    ClassificationPre-Analytical FactorValue(s)
    PreaquisitionDiagnosis/ patient condition
    Healthy
    Liver cirrhosis
    Liver cirrhosis and sepsis
    Severe sepsis
    Underwent cardiopulmonary resuscitation
    Biospecimen AcquisitionTime of biospecimen collection
    After 1 h cycling
    After 2 h cycling
    After 3 h cycling
    After exercise to exhaustion
    Before exercise
    Biospecimen Aliquots and ComponentsCentrifugation
    Centrifugation delays investigated
    StorageStorage duration
    2 h
    4 h
    5 years
    6 months
    6-9 h
    <4 h
    StorageStorage temperature
    22˚C
    4˚C