Conclusions
Our results indicate that current policies of fluid administration during labour may cause significant hyponatraemia in a large proportion of women. Hyponatraemia during labour has been recognised and studied many decades ago; it is therefore surprising that hyponatraemia during labour still occurs.29,30
One reason might be that studies in obstetric literature indicate that higher fluid volumes improve labour outcome.7,8 In addition, midwives focus their attention on caloric requirements believed to be greatly increased during labour, and consequently, encourage oral or intravenous energy supplies.9 Moreover, women are informed through media that high volumes of fluids are essential to preserve health.10 Tolerance to a water load is, however, diminished during labour; therefore, even moderate fluid volumes may cause hyponatraemia, as experienced by several women in our study.17
The highly significant correlation of hyponatraemia with prolonged duration of second stage and instrumental delivery could indicate a causal relationship, but the influence of hyponatraemia on uterine contractility has not been studied. Our results advocate further studies in this area.
Although no severe symptoms of hyponatraemic encephalopathy occurred in any mother or infant in our study, the high prevalence of hyponatraemia imply that hyponatraemic encephalopathy could occur in the larger population of labouring women and their offspring.
We suggest that oral fluid intake during labour should be recorded, and intravenous administration of hypotonic fluids be avoided. The policy of liberal fluid administration should be questioned. Obstetricians and midwives should recognise that hyponatraemia during labour is not uncommon, potentially harmful but easily avoidable, and inform pregnant women accordingly.

Disclosure of interests
None declared.

Contribution to authorship
V.M. planned the study with assistance from L.I., L.B., and M.R. V.M. conducted the study. L.B. analysed the data. All authors wrote the paper.

Details of ethics approval
The study was approved by the ethical committee at the University of Linköping, Sweden (M91-06). All study participants signed informed consent before being enrolled in the study.

Funding
The study was supported by a grant from Kalmar County Research and Development Committee. The study was planned and performed, the results analysed, and the article written without any involvement of the funders.