This paper reports the findings of two investigations into methodological problems associated with the interpretation of blood lactate (BLa) in the sports sciences. In Experiment 1, brachial artery (A), antecubital venous (V) and fingertip capillary (C) blood samples were drawn simultaneously from nine subjects (mean age 21.1 +/- 1.3 years) during an incremental treadmill protocol and immediately assayed for BLa concentration. Experiment 2 investigated the extent of lactate concentration differences in whole blood (WB), lysed blood (LB) and plasma (P) measured using a YSI 23 AM analyser. In Experiment 1, a comparison of the mean BLa concentrations obtained from the three sites revealed no significant differences (P greater than 0.05). Correlations between BLa samples from different sites were very high, with r values ranging from 0.858 to 0.983. In Experiment 2, the mean lactate concentrations were: WB, 4.7 +/- 2.7 mM; LB, 5.0 +/- 3.0 mM; P, 7.0 +/- 3.8 mM. Plasma (P) values were significantly higher than WB and LB. Values from all sites were highly correlated with coefficients ranging from 0.963 to 0.987.
In conclusion: (1) Significant arterial and venous BLa concentration differences do not exist during incremental treadmill exercise. (2) As capillary BLa concentrations reflect arterial values, their use in laboratory and field settings is recommended. (3) Lactate concentration differences in whole blood, lysed blood and plasma will influence the assessment of performance at fixed lactate reference values. (4) If the inter-laboratory test procedures are to be standardized and results compared, precise reporting of lactate sampling and assay techniques is critical.