Results
Baseline values regarding age, weight, plasma sodium, and plasma osmolality were similar in all groups (Table 1). Roughly two-thirds of all fluids were administered orally, and the remaining one-third was administered intravenously (Table 1 and Figure 1). Twenty-one women (15 nulliparous and 6 multiparous) developed hyponatraemia defined as plasma sodium ≤130 mmol/l during delivery (Table 2 and Figure 2). Plasma glucose above12 mmol/l was found in six of these hyponatraemic women. Reduction in plasma sodium was significantly correlated with the duration of labour and with total fluid volume administered during labour (Figure 3). Analysis by multivariate logistic regression showed that maternal hyponatraemia was significantly correlated with total fluid volume administered during labour, but not with epidural analgesia, or oxytocin administration (Table 3). The lowest maternal plasma sodium after delivery was 122 mmol/l, and the corresponding plasma sodium in the umbilical artery was 126 mmol/l. No participant developed signs of severe hyponatraemic encephalopathy. Maternal reduction in plasma sodium correlated with longer duration of second stage (Spearman's rank correlation, n = 218, r = 0.35, P < 0.001). Maternal reduction in plasma sodium was also significantly larger following instrumental vaginal delivery and emergency caesarean section for failure to progress compared with reduction in plasma sodium following spontaneous vaginal delivery (Mann–Whitney U test, n = 42, n = 202, P = 0.002). In women delivered by emergency caesarean section, plasma sodium even tended to be lower before caesarean section than immediately after delivery. However, this difference was not statistically significant (P = 0.2). No woman developed hyponatraemia following caesarean section. Umbilical arterial sodium concentration showed significant correlation with postpartum maternal values (P < 0.001) but was higher than maternal levels in all groups.
Figure 1  Oral and intravenous fluids during labour in the three fluid groups (mean values used). Mean sodium content of intravenous fluids was 69 mmol/l. In addition to the 130 women who received oxytocin intravenously, 93 women received Ringer's acetate during neuraxial analgesia, and 22 women received caloric supplement as glucose 50 or 100 mg/ml. Sport drinks are included among oral fluids called ‘others’. Ringer, Ringer's acetate; Glucose+, Glucose 50 or 100 mg/ml with Na 50 mmol/l; Glucose−, Glucose 50 or 100 mg/ml without electrolytes.
Figure 2  Fluids related to maternal hyponatraemia ≤130 mmol/l (mean values used). Hourly oral fluid intake was similar in all women (P = 0.65), but hourly intravenous infusion rates were higher in hyponatraemic women (P < 0.001). However, the resulting hourly fluid volumes were similar in all women (P = 0.46). Glucose was administered intravenously as energy supply to 22 women, of these 8 women developed hyponatraemia ≤130 mmol/l after having received a mean of 600 ml glucose intravenously (range 100–1500 ml). The 14 women who did not develop hyponatraemia received a mean of 664 ml (100–2000 ml) There was no significant difference between the groups (P = 0.8). Ringer, Ringer's acetate; Glucose+, Glucose 50 or 100 mg/ml with Na 50 mmol/l; Glucose−, Glucose 50 or 100 mg/ml without electrolytes.
Figure 3  Variation of maternal plasma sodium (Na change) during labour and caesarean section and total fluid volumes administered. Reduction in plasma sodium is significantly correlated with total fluid volume (P < 0.001).
Table 3  Interrelationships between maternal hyponatraemia (≤130 mmol/l) immediately postpartum and the various variables analysed with univariate (left) and multivariate (right) logistic regressions
Parameter  Total  Na ≤130  Na ≤130 (%)  Univariate logistic regression  Multivariate logistic regression*
OR (95% CI)  P value  OR (95% Cl)  P value
Fluid group**
<1000  113  1  0.9  1.00   1.00
1000–2500  87  4  4.6  6.8 (2.9–15.7)   5.2 (2.2–12.7)
>2500  61  16  26.2  46 (8.5–247)  <0.001  27 (4.6–162)  <0.001
Oxytocin
<5 units  221  10  4.5  1.00   1.00
≥5 units  40  11  27.5  8.0 (3.1–21)  <0.001  2.6 (0.9–7.4)  0.072
Epidural
No  172  3  1.7  1.00
Yes  89  18  20.2  14.3 (4.1–50)  <0.001  —  —
Parity
Multipara  148  6  4.1  1.00
Nullipara  113  15  13.3  3.6 (1.4–9.7)  0.011  —  —
Body mass index and age are not shown (P = 0.62 and P = 0.59, respectively, in the univariate analysis and are therefore not included in the multivariate analysis).
*  Parity was deleted from the multivariate analyses with P = 0.86 and epidural with P = 0.12.
**  Fluid groups as defined in Tables 1 and 2. The control group elective caesarean section is not included in the analysis.