An overlooked aspect on metabolic acidosis at birth: blood gas analyzers calculate base deficit differently.

OBJECTIVE

Metabolic acidosis (MA) at birth is commonly defined as umbilical cord arterial pH < 7.0 plus base deficit (BD) ≥ 12.0 mmol/L. Base deficit is not a measured entity but is calculated from pH and Pco(2) values, with the hemoglobin (Hb) concentration [Hb] included in the calculation algorithm as a fixed or measured value. Various blood gas analyzers use different algorithms, indicating variations in the MA diagnosis. The objective was therefore to calculate the prevalence of MA in blood and extracellular fluid with algorithms from three blood gas analyzer brands relative to the Clinical and Laboratory Standards Institute (CLSI) algorithm.

DESIGN

Comparative study.

SETTING

University hospital.

SAMPLE: Arterial cord blood from 15 354 newborns.

MAIN OUTCOME MEASURE: Prevalence of MA.

METHODS

Blood was analyzed in a Radiometer ABL 735 analyzer. Base deficit was calculated post hoc with algorithms from CLSI and Corning and Roche blood gas analyzers, and with measured and fixed (9.3 mmol/L) values of [Hb].

RESULTS

The prevalence of BD ≥12.0 mmol/L in blood was with the CLSI algorithm 1.97%, Radiometer 5.18%, Corning 3.84% and Roche 3.29% (CLSI vs. other; McNemar test, p < 0.000001). Likewise, MA prevalences were 0.58, 0.66, 0.64 and 0.64%, respectively (p≤ 0.02). Base deficit ≥ 12.0 mmol/L and MA rates were lower in extracellular fluid than in blood (p≤ 0.002). Algorithms with measured or fixed Hb concentration made no differences to MA rates (p≥ 0.1).

CONCLUSIONS

The neonatal metabolic acidosis rate varied significantly with blood gas analyzer brand and fetal fluid compartment for calculation of BD.