Osmotherapy for management of raised intracranial pressure

Osmotherapy is a crude but effective means of desiccating the brain parenchyma and decreasing its contribution to intracranial pressure. This issue is interesting to the time-poor exam candidate because hypertonic saline has already been a subject of an SAQ (Question 4 from the first paper of 2007). In the more distant past (Question 8 from the first paper of 2001) the college demanded that you compare and contrast hypertonic saline with mannitol.  It is only a matter of time before they do it again.

Mechanism of osmotherapy for raised intracranial pressure

An excellent overview of this is afforded by Diringer et al (2013). In brief, osmotherapy works by the following mechanism:

Osmotic effects: 

  • Osmotherapy solutions change (increase) the osmolality of extracellular fluid
  • The change in extracellular osmolality results in the shift of fluid from the intracellular to the extracellular compartment
  • This has the effect of decreasing the volume of intracranial parenchyma 
  • The water which has shifted into the extracellular compartment then undergoes removal by osmotic diuresis
  • The reduction in intracranial parenchymal volume is not sustained because the neurons will respond by generating idiogenic osmoles

Non-osmotic effects:

  • Osmotherapy agents increase systemic blood volume, which leads to:
    • haemodilution
    • increased cardiac output
    • increased blood pressure.
  • Mannitol also reduces blood viscosity and supposedly improves rheology of red cells
  • Improved intracranial blood flow should result from this
  • The improvement in blood viscosity also decreases intracerebral blood volume

A comparison of osmotherapy agents

In brief:



Hypertonic saline

A Comparison of Mannitol and Hypertonic Saline Osmotherapy
  • Molecular weight: 182.17 g/mol
  • Reflection coefficient: 0.9
  • Sodium content: none
  • Osmolarity (20%): 1100 mOsm/L
  • Molecular weight: 58.45 g/mol
  • Reflection coefficient: 1.0
  • Sodium content 3400 mEq/L
  • Osmolarity: 6800 mOsm/L
  • Still fairly cheap
  • Rapid effect
  • Seems to have some sort of rheological benefit (increases red cell deformability)
  • Acts as a transient volume expander
  • May have a better effect on cerebral blood flow for a given reduction in ICP.
  • Cheap
  • Stable in storage
  • Very rapid effect
  • Seems to have some sort of intrinsic anti-inflammatory effect
  • May also have some rheological benefits
  • At least as potent as mannitol when it comes to reducing intracranial pressure
  • Less potential for hypovolemia than with mannitol- the diuretic effect is less potent
  • May have a better effect on cerebral blood flow for a given reduction in ICP.
  • Safe endpoint (serum sodium around 145-155) is easily monitored with serial ABGs.
  • Unstable in storage: at low temperatures and at altitude, it precipitates.
  • Medium for bacteria and fungus.
  • Causes a brief state of volume overload
  • Causes torrential diuresis and hypovolemia
  • Causes hyponatremia while in the serum, and hypernatremia after the inevitable diuresis
  • Endpoint is serum osmolality(320), which is cumbersome to measure
  • May cause ICP to "rebound" after prolonged use
  • Need for central venous access
  • No standards for which concentration to use, or how to give it
  • Hypokalaemia
  • Hyperchloraemic acidosis
  • Should not be used if the patient is chronically hyponatremic
  • Possible seizures due to wild fluctuations in serum sodium
  • Increase in circulating volume with risk of fluid overload.
  • Coagulopathy (APTT and INR)
  • Altered platelet aggregation.
  • May affect normal brain more that injured brain which theoretically may worsen herniation

In detail:

The history of osmotherapy stretches back to the early 20th century, when early cat-haters injected cats with hypertonic saline and noted that their thecal sacks had become flaccid, making it difficult to acquire CSF from them. Mannitol became popular later- if urea causes such splendid osmotic effects, then why not mannitol, Shenkin argued. Indeed, his patients demonstrated a rise in serum osmolality by 20-30 mOsm/L, and a fall in ICP by 30-60%. The physiological consequences of infusing 20% mannitol are discussed elsewhere, and the ICU fellowship candidate should be at least dimly aware of them.

Evidence in support of osmotherapy

The BTF guidelines (4th edition) are suprisingly ambivalent towards this practice. Their previous Level II and III recommendations were not carried forward from the previous edition because they were based on studies which do not meet their new (stricter) criteria. The BTF committee acknowledged the utility of this strategy (they were "universal in their belief" that hyperosmolar agents are useful) but did not make an actual recommendation in support of their use. 

As far as published literature goes, the BTF statement is good enough as an overview of the reasons behind their non-recommendation. For an even larger overview of the evidence, one can look to the excellent 2017 review by Witherspoon et al. 

In summary:

  • The 2007 Brain Trauma Foundation guidelines gave recommendations in support fo mannitol
  • The new BTF guidelines (4th edition)  have withdrawn some of their support, and merely restated the old recommendation.
  • This is because the study quality was sub-optimal
  • There were two RCTs and one cohort study in 
  • None met the criteria for level II or III recommendation
  • The ongoing use of osmotherapy derives support from the concept that ICP-guided therapy has a mortality benefit, but itself has been challenged.

Which is better, mannitol or normal saline?

Good question. Let us face it, with the silliness of glycerol and urea behind us, these two are the main contenders. We might appeal to the authority of higher beings for an answer. Indeed, the old 3rd edition of the  Brain Trauma Foundation guidelines will impress the casual reader with the steel of their firm conviction.
" Mannitol is effective in reducing ICP", they say, but as for hypertonic saline, "Current evidence is not strong enough". That was in 2007. What about now?

Well.  The (2016)  publication of the BTF Guidelines was again unable to make a firm recommendation in favour of one agent over another. Sure, many low-quality case control and observational studies support hypertonic saline. But to make recommendations on such weak foundations would undermine the quality of the BTF package as a whole, and discolour all the other recommendations with a blush of shame.

A 2008 trial ran the two substances head to head in an equiosmolar contest. Both substances "equally and durably reduced the ICP" in a case series of 20 stable TBI patients. The conclusion of the investigators was still in favour of mannitol, given the proposed "rheological effects" on the quality of blood flow. The belief in these effects is reasonably well founded in serial measurements of blood viscosity,which revealed that red cell "deformability" increases with mannitol therapy, thereby improving the passage of these soft flexible red cells through the microcirculation.

A 2011 review of all available literature on this matter, however, had scraped together some support for hypertonic saline, finding a small statistically significant benefit for its use among a slush of poorly designed heterogenous trials. Given that little extra evidence became available in the subsequent 2 years, a 2013 review reached the same conclusion.

Overall, it seems the trend these days is away from the mannitol, and towards the saline. Specifically, today's public seem to favour super-high osmolarity saline solutions, such as 20% saline. So, what are the advantages of saline over mannitol?

  • Its cheap
  • It does not cause massive diuresis and hypovolemia
  • It is easy to monitor therapy with blood gases (aiming for a Na+ level around 145-155)
  • It seems to have some sort of mysterious anti-inflmmatory properties which decreases the permeability of the injured blood-brain barrier

The use of hypertonic saline has been much maligned by allegations of platelet function impairment and increased risk of bleeding. Fortunately, one is unlikely to ever see these effects in a realistic clinical setting. Studies have demonstrated that one would need to replace 10% of their blood volume with hypertonic saline before one experiences any of these effects.


The only reference really needed for this are the excellent Brain Trauma Foundation Guidelines.

Oh's Intensive Care manual:

Chapter 43 (pp. 563) Cerebral protection by Victoria Heaviside and Michelle Hayes, and

Chapter 67 (pp. 765) Severe head injury by John A Myburgh.

Shenkin, Henry A., et al. "The use of mannitol for the reduction of intracranial pressure in intracranial surgery." Survey of Anesthesiology 8.5 (1964): 405-406.

Francony, Gilles, et al. "Equimolar doses of mannitol and hypertonic saline in the treatment of increased intracranial pressure*." Critical care medicine 36.3 (2008): 795-800.

Burke, Allan M., et al. "The effects of mannitol on blood viscosity." Journal of neurosurgery 55.4 (1981): 550-553.

Kamel, Hooman, et al. "Hypertonic saline versus mannitol for the treatment of elevated intracranial pressure: A meta-analysis of randomized clinical trials*."Critical care medicine 39.3 (2011): 554-559.

Nau, Roland. "Osmotherapy for elevated intracranial pressure." Clinical pharmacokinetics 38.1 (2000): 23-40.

Rickard, A. C., et al. "Salt or sugar for your injured brain? A meta-analysis of randomised controlled trials of mannitol versus hypertonic sodium solutions to manage raised intracranial pressure in traumatic brain injury." Emergency Medicine Journal (2013).

Lazaridis, Christos, et al. "High-Osmolarity Saline in Neurocritical Care: Systematic Review and Meta-Analysis*." Critical care medicine 41.5 (2013): 1353-1360.

Bhardwaj, Anish, and John A. Ulatowski. "Hypertonic saline solutions in brain injury." Current opinion in critical care 10.2 (2004): 126-131.

Arbabi, Saman, et al. "Hypertonic saline induces prostacyclin production via extracellular signal-regulated kinase (ERK) activation." Journal of Surgical Research 83.2 (1999): 141-146.

R LAWRENCE REED, I. I., et al. "Hypertonic saline alters plasma clotting times and platelet aggregation." Journal of Trauma-Injury, Infection, and Critical Care 31.1 (1991): 8-14.

Fink, Matthew E. "Osmotherapy for intracranial hypertension: mannitol versus hypertonic saline." CONTINUUM: Lifelong Learning in Neurology 18.3, Critical Care Neurology (2012): 640-654.

Diringer, Michael N. "New trends in hyperosmolar therapy?.Current opinion in critical care 19.2 (2013): 77.

Witherspoon, Briana, and Nathan E. Ashby. "The Use of Mannitol and Hypertonic Saline Therapies in Patients with Elevated Intracranial Pressure: A Review of the Evidence." Nursing Clinics of North America (2017).