Use of different anticoagulants in test tubes for analysis of blood lactate concentrations: Part 2. Implications for the proper handling of blood specimens obtained from critically ill patients

Abstract

Objectives: a) To test the hypothesis that the measurement of the circulating lactate concentration is influenced by the anticoagulant in the test tube that contains the blood sample; b) to test the hypothesis that the measurement of the circulating lactate concentration is influenced by the tissue used for analysis.

Design: A prospective, controlled study.

Setting: A critical care research laboratory, a 20-bed intensive care unit (ICU), and the general wards.

Subjects: Twenty-three ICU and ward patients with hyperlactatemia and 19 healthy volunteers.

Interventions: Blood samples were collected for determination of blood lactate concentration.

Measurements and main results: Venous blood samples (12 mL) were obtained from each of the 19 normal subjects and each 12-mL specimen was evenly divided into six aliquot portions (six test tubes). Experiment 1: Of the six tubes, two tubes were set aside for experiment 2. The other four tubes were used to test four anticoagulants (one anticoagulant per tube). The anticoagulants tested were: sodium heparin; EDTA; lithium heparin; and sodium citrate. Lactate concentrations were analyzed using an ion-selective, amperometric electrode that we have previously validated. There were no statistically significant differences between the lactate concentrations derived from blood samples stored in sodium heparin, EDTA, or lithium heparin (p > .05; n = 19; Student-Newman-Keuls' multiple comparisons test). The lactate concentration of blood stored in sodium citrate, however, was lower than all other anticoagulants (p < .001; n = 19; Student-Newman-Keuls' multiple comparisons). Experiment 2: Of the remaining two test tube samples from each subject, one tube contained sodium heparin and the other tube did not contain an anticoagulant. Each of these two tubes was centrifuged at 50 degrees F (10 degrees C) for 15 mins to obtain plasma and serum samples. Lactate concentrations were measured in the serum and plasma and compared with those concentrations found in whole blood samples from the tube containing sodium heparin from experiment 1. The plasma and serum lactate concentrations were consistently higher than the whole blood lactate values from the same specimen (p < .05; n = 42; Student-Newman-Keuls' multiple comparisons test). Since experiment 1 involved the collection of blood from healthy volunteers with normal lactate concentrations, we chose to investigate whether this discordance between plasma or serum and whole blood was dependent on the lactate concentration. To answer this question, we studied 23 patients with known hyperlactatemia and found that in subjects with a lactate concentration of < 2.2 mmol/L, there was a difference of 0.11 mmol/L in the mean values between plasma and whole blood concentrations (p < .0004; n = 19; paired t-test). In subjects with a lactate concentration of > 2.2 mmol/L, there was a difference of 0.14 mmol/L (p < .0001; n = 23; paired t-test) in the mean values between plasma and whole blood. In all samples at all concentrations, there was no significant difference between serum vs. plasma samples (p > .05; Student-Newman-Keuls' test).

Conclusions: a) Sodium citrate, as an anticoagulant, caused lower lactate concentrations to be measured as compared with heparin or EDTA; b) the measurement of lactate concentrations in plasma or serum samples yields a higher value than the concentration found in the original whole blood specimen.