Hypernatraemia and Hyponatraemia 41 



increased by an abnormal breakdown of body protein or by 

 excessive protein in the diet. One hundred grams of protein 

 contain 16 g. of nitrogen, excreted as 34 g. or 570 m-osm. of 

 urea. Ten grams of sodium chloride provide 340 m-osm. It is 

 not unusual for unconscious patients to receive these amounts 

 in their feeds ; and their endogenous production of urea may 

 already be very large (Cooper etal., 1951). The hypernatraemic 

 patient of Natelson and Alexander (1955) presumably had 

 an osmotic diuresis when he was made worse with "non- 

 saline fluids", because these consisted partly of protein 

 hydrolysate equivalent to 100 g. of protein. In certain 

 neurological disturbances (Astrup, Gotzche and Neukirch, 

 1954; Whedon and Shorr, 1957) and in water deficiency itself 

 (Black, McCance and Young, 1944) the breakdown of body 

 protein may be greatly accelerated. 



To detect a water deficit, the minimum data required are 

 the estimated intake and output of water and solutes, and the 

 volume and concentration of the urine. A water deficit is 

 confirmed if, with the administration of water, the elevated 

 plasma [Na] falls. 



In many of the reports of cerebral hypernatraemia it is 

 impossible to decide from the data given what the water 

 balance was. The patients with hypernatraemia of Higgins 

 and his co-workers (1954) seem to have begun with a deficit of 

 water of about one litre. Subsequently their intake of water 

 may have been as little as two litres daily. Their exogenous 

 osmolar load was about 610 m-osm. We do not know what 

 was the total excretion; urine volumes and specific gravities 

 are not stated. The blood urea was high, and fell as the plasma 

 [Na] fell, when their intake of fluid was increased. In other 

 reports the data actually show there was a cumulative deficit 

 of water although the fact may have been disregarded 

 (Anthonisen, Hilden and Thomsen, 1954; Allott, 1957). 



Failure of thirst. Even when losses of water do go un- 

 recognized by the clinician, there is no danger of water 

 depletion as long as the patient responds normally with thirst 

 and is able to drink. For example, in uncomplicated diabetes 



