How to determine and use base excess (BE) in perinatal medicine.

Background: Foetal hypoxia may lead to multi-organ failure and cerebral injury. Usually this process is accompanied by severe metabolic acidosis. The base excess (BE) determined in umbilical artery (UA) blood is the most appropriate parameter to evaluate metabolic acidosis. The correct determination of BE therefore is of paramount importance both for the jeopardised foetus and the obstetrician in litigation. In blood, BE is dependent on the oxygen saturation [sO2 (%)] of haemoglobin (cHb). Due to the normally low foetal pO2 values in UA blood (median: ca. 18 mmHg) sO2 is low as well; therefore computational correction of BE to - by definition - 100 % oxygen saturation seems to be mandatory. This paper presents an analysis of this complex problem in obstetrics.
Methods: pH, pCO2, pO2 and cHb were measured in UA blood of 6 302 term infants delivered spontaneously using equipments from Radiometer (Copenhagen). BE was computed according to the equation of Siggaard-Andersen actually used in many blood-gas analysers. sO2 (%) was computed for HbF using the algorithm of Ruiz et al. . The numerical correction of BE was achieved with an equation given again by Siggaard-Andersen . APGAR indices after 1 minute were used.
Results: Median BE in UA was - 4.6 and the mean was - 4.9 +/- 3.0 mmol / L, respectively. Correction of BE (BEoxy.) to the actual (calculated) oxygen saturation (%) leads (always) to lower values: a median BE (oxy.) in UA of - 7.4 and a mean of - 7.6 +/- 3.2 mmol / L, respectively. There is no correlation between BE and sO2 in UA blood: r = 0.0078, p = 0.532, n = 6 302 (mean oxygen saturation: 27.7 +/- 18.3 % ). The median cHb amounted to 15.2 and the mean to 15.0 +/- 2.6 g %, respectively. The median delta-BE,UA(BE - BEoxy.) amounts to 2.74 mmol / L; the maximum delta-BE reached 5.2 mmol / L in this sample. Correction of BE to 100 % oxygen saturation based on the (calculated) real oxygen saturation (%) leads to significantly (p = 0.0099) higher correlations with the APGAR index (1 min) and pCO2 in UA (p << 10 (-4)) as well.
Conclusion: Correction of BE in UA, i. e., correction of BE to 100 % oxygen saturation using the (calculated) actual oxygen saturation (%) of the blood sample is mandatory in perinatal medicine. Correction uniformly leads to lower BE values (median: 2.7 mmol / L) and significantly higher correlation coefficients with important clinical variables (e. g., the APGAR index).