The controversy about what fluids to use in critically ill children to prevent hyponatraemia is ongoing. We congratulate Singhi and Jayashree for their contribution to this debate(1). There is however one point in their paper that needs clarification. In Fig.1, a line showing electrolyte free water (EFW) intake and expected fall in serum sodium has been drawn. It runs diagonally across the figure suggesting that a linear relationship between the two is expected. The data source for this assumption has not been provided.

Fig.1 Electrolyte free water (EFW) and fall in sodium.

Interestingly, in their data on children who developed hyponatraemia, the biggest falls in serum sodium were seen in the patients who received less EFW (r = – 0.57). The message can be best conveyed by drawing the regression line. We have scanned in their figure and have drawn the regression line in the attached figure. This graph depicts data on the relationship of EFW and fall in sodium in patients who developed hyponatraemia. It will be crucial to know whether this trend is seen in the other patients also (those who did not develop significant hyponatraemia), to test the hypothesis that fall in sodium is inversely related to EFW. It may be possible to understand whether the group that developed hyponatraemia is different from those who did not develop hyponatraemia and we can then ‘focus on the inherent properties of the patient’s physiology, rather than the inherent properties of the fluid being used’ as suggested by Choong in the accompanying editorial(2).