# Question 3

A 20 year old female in ICU following a diffuse axonal head injury develops a severe exacerbation of intracranial hypertension on day 3. She is mechanically ventilated, paralysed and sedated. Investigations  during a subsequent  episode of marked polyuria are summarised below.

 Variable Normal range pH 7.50 PaCO2 28mm Hg HCO3- 21mmol/L Standard base excess -1.5mmol/L Sodium 147mmol/L (135 -145) Potassium 3.2mmol/L (3.2 - 4.5) Chloride 110mmol/L (100 -110) Urea 3.0mmol/L (3.0 - 8.0) Creatinine 65Dmol/L (50- 100) Glucose 4.0mmol/L (3.0 – 6.0) Measured plasma osmolality 333mosmol/kg Urine osmolality 410mmol/L

1)  What is the most likely explanation for the polyuria?

3)  What action needs to be taken concerning the polyuria, and why?

4)  Describe the acid-base status

5)  What action needs to be taken concerning the ventilation, and why?

1.  Mannitol therapy

2.  There is increased measured plasma osmolality with an elevated osmolar gap. The

gap is 44 mosmol/kg, if we use a calculated osmolality of 1.86 × ([Na] + [K]) + [urea]

+ [glucose]. If we use the simple formula of 2 × [Na] + [urea] + [glucose] for calculated osmolality, the gap is 32 mosmol/kg. (There are also other formulae which are more difficult to remember). In the setting of treatment for an exacerbation of intracranial hypertension, the increased osmolar gap is likely to be due to mannitol administration. The high urinary osmolality rules out diabetes insipidus, and supports the diagnosis of mannitol induced polyuria.

3.  Mannitol should be ceased until the measured plasma osmolality is < 320 mosmol/kg.

There is no benefit and higher death rates in case controlled studies at higher induced osmolality.

4.  Acute (uncompensated) respiratory alkalosis.

5.  Minute ventilation should be reduced as soon as possible to return the PaCO2 to 35 –

40 mm Hg. Hypocapnia should be reserved for brief intermittent use to buy timeduring critical neurological events (eg pupillary dilation, new lateralising signs). Prolonged hypocapnia reduces cerebral blood flow and oxygenation, and eventually becomes ineffective as CSF pH returns towards normal.

## Discussion

This question is remarkably similar to Question 3.1 from the second paper of 2010, and Question 22.1from the second paper of 2011.

Let us dissect these results systematically.

1. The A-a gradient cannot be calculated
2. There is alkalaemia
3. The PaCO2 is contributing to the alkalosis
4. The SBE is -1.5, suggesting a normal acid-base balance; it and the bicarbonate is mildly decreased suggesting a trend towards metabolic acidosis
5. The metabolic compensation is appropriate. It is expected that the bicarbonate would decrease by 2mmol/L for every 10mmHg decrease in PaCO2, and thus we would expect to see a bicarbonate value around 22, which we do.
6. The anion gap is slightly raised:
(147+3.2) - (110+21) = 19.2
The delta ratio suggests that there is a combination of pure high anion gap metabolic acidosis and metabolic alkalosis here.
(19.2 - 12) / (24 - 21) = 2.4
7. The osmolar gap is increased:
333 - (2 × 147 + 3 + 4) = 32 (normal is under 10)

The head injury story and the polyuria together make one want to cry "DI!" but in actual DI the urinary osmolality is frightfully low, as torrents of dilute urine issue forth from the vasopressin-deficient patient. The urinary osmolality here is given, so that the candidate does not become confused. The history of a recent spike in ICP inflames the imagination, promoting the idea that someone may have given this patient some hypertonic syrup.

The physiological consequences of mannitol therapy are discussed in greater detail elsewhere.

So, "what action needs to be taken" concerning the polyuria?

Well. The mannitol should be ceased, for one. And then the tincture of time should be liberally applied. The mannitol will exit via the urine, taking with it some water, and the serum osmolality will settle into a more tolerable range.

The ventilation adjustment required to correct this hyperventilation is a decreased respiratory rate. No head-injured patient needs a PaCO2 of 28; hyperventilation in head injury is a last-ditch effort to decrease intracranial pressure, and it may reduce cerebral blood flow in the process.

## References

UpToDate has a nice article on the complications of mannitol therapy.

The specific features of it, being the diuresis and the resulting electrolyte derangement, are well explored in this ancient yellowed scroll from NEJM:

GIPSTEIN RM, BOYLE JD. HYPERNATREMIA COMPLICATING PROLONGED MANNITOL DIURESIS. N Engl J Med. 1965 May 27;272:1116-7.