**2.4 Statistical analysis**

164 From Preconception to Postpartum

Fetal blood was obtained by cordocentesis from the umbilical vein, a technique carried out without maternal premedication, and only under local anesthesia at the puncture site

Five hundred microlitres of umbilical venous blood were collected in a heparinised syringe for gasometric and acid-base analyses, pH, partial CO2 pressure (pCO2), bicarbonate

The concentrations of ketone bodies (aceto-acetate and beta-hydroxybutyrate), free fatty

Two hundred microlitres of umbilical venous blood were collected on sodium fluoride to measure glucose and lactate, and 200 µl were collected without anticoagulant to measure

The good quality of the sample was verified, in particular the absence of contamination by maternal blood or amniotic fluid (Forestier, 1988). An immediate check was made by determining the erythrocyte group iI, and then a Kleihauer test, red and white cell count

The serum hCG concentration of fetal blood was also measured, this test being proposed as a sensitive contamination marker due to the low hCG content of fetal serum, compared with

Fetal blood samples were stored in ice at +4°C and transported to the laboratory immediately. The blood pH and gas measurement and the deproteinisation of the whole blood with perchloric acid for the ketone bodies determination were carried out upon receipt of the samples; the deproteinisation supernatants were decanted and kept at -20°C. The blood was centrifuged at +4°C for the other analyses. The plasma glucose, lactate, cholesterol and hCG concentrations were determined immediately; the free fatty acid

The sample storage conditions were verified for the quantitative analysis carried out after

The gasometric and acid-base analyses (pH, pO2, pCO2, bicarbonate, SaO2) were carried out

The glucose and lactate concentrations were determined with the Ektachem 500 automated analyser (Kodak, New York, United States) by an enzymatic method using glucose oxidase (EC 1.1.3.4) and lactate oxidase (EC 1.13.12.4) and a measurement by reflectometry (reference

The total free fatty acid concentration was measured by a manual colorimetric enzymatic assay (Okabe et al., 1980), using an acyl-coenzyme A synthetase (EC 6.2.1.3), an acylcoenzyme A oxidase (EC 1.3.3.6) and a peroxidase (EC 1.11.1.7) (Biomérieux, Marcy l'Etoile,

interval in adult blood: 3.6 – 5.8 mmol/l for glucose and 0.7 – 2.1 mmol/l for lactate).

that of the maternal blood and / or amniotic fluid (Dommergues et al., 1993).

contents were determined subsequently on the plasma kept at -20°C.

using the ABL 300 analyser (Radiometer, Copenhagen, Denmark).

concentration, partial oxygen pressure (pO2) and oxygen saturation (SaO2).

acids (FFA) and cholesterol were evaluated on the same sample.

and a leukocyte differential count were carried out.

**2.2 Sampling procedures** 

(Daffos, 1983).

hCG.

freezing.

France).

**2.3 Analytical methods** 

The comparative study between the results of the control group and the pathological group was carried out using the Student's t-test (unpaired series) and the Mann-Whitney U test (non-parametric test). The search for a possible relation between different constituents measured in the umbilical venous blood was carried out by calculating the linear correlation coefficient r and the Spearman coefficient rs. The significance of the correlations was evaluated with Fisher's exact test.

In the control group, an analysis by linear regression was used to study any changes in the parameters according to gestational age (expressed as weeks of amenorrhea).

A p value below 0.05 was considered to be statistically significant.
