Statistical analysis
Logarithmic10 transformation of variables was performed
prior to statistical analysis, when necessary, then the
Kolmogorov-Smirnov test was used to determine whether the data were
parametric. Parametric continuous variables were presented in mean and
standard deviation (SD) while nonparametric continuous variables were
presented in median and interquartile range (IQR). Count and percentage
were used to present categorical variables. The following variables were
reported as z-scores corrected for gestational age: EFW, birth weight,
UA-PI, MCA-PI, CPR, and UtA-PI. Relationship between EFW z-score, UA-PI
z-score, MCA-PI z-score, CPR z-score, UtA-PI z-score,
log10 cCTG STV, and umbilical cord arterial pH was
determined using Pearson correlation analysis. To determine which of the
maternal characteristics, labour onset, indications of induction of
labour, EFW z-score, maternal-foetal Doppler indices,
log10 cCTG STV, and Dawes-Redman criteria were
significant predictors of umbilical cord arterial pH, univariate linear
regression analysis was used. To identify the variables that were
independent predictors of umbilical cord arterial pH, predictors with a
p value of < 0.1 were included in a multivariate linear
regression analysis with forward selection. Univariate and multivariate
logistic regression analyses were used to determine whether which of the
study factors mentioned above were significant predictors of umbilical
cord arterial pH < 7.1 and NICU/SCBU admission. The
statistical software package SPSS 20.0 (IBM SPSS Statistics for Windows,
Version 20.0. Armonk, NY: IBM Corp) was used for data analyses.