Statistical analysis
The incidence of metabolic acidosis varies between 1.3% to 5.7% and
the confidence intervals are very wide. For this reason, no power
calculation was conducted for this study. Based on the study of
Wiberg-Itzel et al ., a sample size of 3000 women was chosen to
achieve a sufficiently large number of the predefined outcomes for the
calculations in a reasonable timeframe (9).
Receiver Operating Characteristic (ROC) curves with the respective areas
under the curve (AUC) were calculated for the relevant outcomes.
Spearman rank correlation coefficients were calculated. Kruskal-Wallis
or Fishers exact test were used when appropriate. To investigate a
possible relationship between the discrepancies in values obtained by
either the two BGAs or POCs linear regression and one sample T-test
before a potential Bland-Altman plot was constructed. Analyses were
performed using SPSS 26.0 (SPSS, Chicago, IL, USA). For the calculation
of sensitivity, specificity, positive (LR+) and negative likelihood
ratios (LR-), with their 95% confidence intervals, Medcalc.org was
used.