1. Aetiology and classification
Hypernatraemia is defined as a plasma sodium of >145mmol/l. It is a much less common electrolyte abnormality than hyponatraemia.
In a recent spot analysis at our hospital, where we had 62,488 requests for U&Es, hyponatraemia was documented in approximately 10 per cent of patients, whereas hypernatraemia occurred in only 1.9 per cent. These figures may represent an exaggerated estimate, because high or low results might be repeated, but the relative frequencies are less likely to be distorted.
Given that sodium is the major contributor to plasma tonicity, patients with hypernatraemia are always hyperosmolar and virtually all will have a reduction of extracellular water rather than an increase in body sodium.
The likely cause will be different depending on whether the abnormality arises in patients in or outside of hospital; the cause is more likely to be iatrogenic if patients are hospitalised.
As is the case for patients with hyponatraemia, the differential diagnosis is generally guided by an assessment of volume status.
If a patient loses both salt and water then they are likely to become volume depleted and will therefore manifest hypotension and other features of dehydration.
If more water than sodium is lost they become hypernatraemic and the clinical scenarios outlined in the figure are usually clinically obvious.
If clinically significant hypernatraemia occurs in the context of hypervolaemia, it is invariably the case that large amounts of sodium have been administered (ie it is iatrogenic).
An example of this in adults is the administration of IV sodium bicarbonate to patients in renal failure. It should be noted that the hypernatraemia of primary aldosterone excess is minor and of no clinical consequence.
As is the case for patients with hyponatraemia, most adults with hypernatraemia are clinically euvolaemic (particularly those who present de novo with this electrolyte abnormality) and represent a group of patients who are exclusively water depleted.
Most patients will be asymptomatic, and it would be unusual to see significant symptoms unless the plasma sodium is in excess of 160mmol/l.
Above that level patients should be complaining of intense thirst and the absence of this, in a fully conscious patient, should indicate likely damage or lesions on the hypothalamus with probable associated diabetes insipidus (DI).
Symptoms of hypernatraemia are largely non-specific (anorexia, muscle weakness, restlessness, nausea and vomiting) but may be followed by lethargy, irritability, stupor and even coma.
The osmolar shift of water from the brain may induce cerebral shrinkage and intracerebral bleeding.
The algorithm should assist accurate assessment and guide treatment.
Patients who are volume depleted usually have an obvious cause and, unless the problem is within the kidney, they are likely to be avidly retaining sodium.
Similarly, although the hypervolaemic group will not be oedematous, it should be clear (at least in adults) that they have been given too much salt-containing medication.
The majority of patients do not fall into either of the above categories. They will have lost water or hypotonic fluid. They will only become hypernatraemic if fluid access is denied for some reason (such as elderly confused patients or patients with poor thirst sensation or an altered conscious state).
For this group, urinary sodium is likely to reflect intake and measurement will be unhelpful.
Such patients would (in the absence of DI) exhibit a high urine osmolality, and a value less than 500mosmol/kg would certainly suggest at least partial DI.
In patients with an inappropriately diluted urine, the response to 1(mu)g of IV desmopressin should separate a cranial from a nephrogenic cause.
Always consider lithium as a possible cause of acquired nephrogenic DI.
Correction of the underlying cause is an important first step in management but this can take time.
Reducing plasma sodium too quickly can be harmful, and unless the abnormality has developed within a few hours, where the serum sodium can be lowered by as much as 1mmol/l/hr, it is prudent to lower the sodium level by no more than 0.5mmol/l/hr or 10mol/l/day.
In the vast majority of patients it is water that needs to be replaced and this should be done by increasing oral intake where possible. If not, the usual fluids given are 5 per cent dextrose or 0.45 per cent (half normal) saline.
The more hypotonic the solution the slower the rate of infusion necessary but there may be little to choose between the two in practice.
In patients with hypovolaemia, normal saline should be infused until volume status is normalised and the further treatment should be as above.
The amount of IV fluid replacement is based on a calculation of the likely water deficit, and there are a number of different formulae to assist in this. They are all imprecise and so frequent monitoring of the changes in plasma sodium is vital.
The simple calculation of water deficit as being equal to total body water X (1 - (140 divided by serum Na+ concentration)) will only work if there is a loss of pure water, for example in diabetes with adypsia, and a more robust formula has been suggested (see box).
It is important to stress that frequent monitoring of vital signs and electrolytes will allow adjustments in infusion rates to be made until a target plasma sodium of 145mmol/l is reached.
Remember to allow for 500-1,500ml insensible loss to be replaced with the same infusate. Some patients may require in excess of 250ml/hr.
Mortality rates in patients with severe hypernatraemia can be as high as 70 per cent, so careful management is of the utmost importance.