Technical performance of lactate biosensors and a test-strip device during labour

Lactate in fetal blood has a high diagnostic power to detect fetal compromise due to hypoxia, as lactate allows an estimation of duration and intensity of metabolic acidemia. Biosensor technology allows an instantaneous diagnosis of fetal compromise in the delivery room. The goal of the current investigation is to define the preanalytical and analytical biases of this technology under routine conditions in a labour ward in comparison to test-strip technology, which allows measurement of lactate alone.

Three lactate biosensors (RapidLab 865, Siemens Medical Solutions Diagnostics, Bad Nauheim, Germany; Radiometer ABL625 and ABL 700, Radiometer Copenhagen, Denmark) and one test-strip device (Lactate Pro, Oxford Instruments, UK) were evaluated regarding precision in serial and repetitive measurements in over 1350 samples of fetal whole blood. The coefficient of variation (CV) and the standard deviation (SD) were calculated. The average value of all three biosensors was defined as an artificial reference value (refval). Blood tonometry was performed in order to test the quality of respiratory parameters and to simulate conditions of fetal hypoxia (pO (2): 10 and 20 mmHg).

The precision of serial measurements of all biosensors indicated a coefficient of variation (CV) between 1.55 and 3.16% with an SD from 0.042 to 0.053 mmol/L. The test-strip device (Lactate Pro) mounted to 0.117 mmol/L and 3.99% (SD, CV). When compared to our reference value (refval) ABL 625 showed the closest correlation of -0.1%, while Siemens RapidLab 865 showed an overestimation of +8.9%, ABL700 an underestimation of -6.2% and Lactate Pro of -3.7%.

For routine use all tested biosensors show sufficient precision. The test-strip device shows a slightly higher standard deviation. A direct comparison of measured lactate values from the various devices needs to be interpreted with caution as each method detects different lactate concentrations. Furthermore, the 40 min process of tonometry led to an increase of SD and coefficient of variation in all devices. This results in the important preanalytical finding that the precision of replicated measurements worsens significantly with time. The clinician should be aware of the type of analyser used and of preanalytical biases before making clinical decisions on the basis of lactate values.