IN this issue of Anesthesiology, Rozet et al.  1present the findings of a comprehensive study comparing two hyperosmolar agents, hypertonic saline (HS) and mannitol, for brain relaxation during craniotomy. This work calls into question whether sufficient information is now available to advocate substitution of mannitol with HS to promote brain relaxation during routine craniotomy.

It was discovered in the early 20th century that HS reduces brain bulk in animals.2Reports in humans were not forthcoming until the 1980s.3The neurosurgical and neuro–critical care communities have since explored the use of HS because of its ability to treat cerebral edema and intracranial hypertension. Although mannitol remains the recommended hyperosmolar therapy in settings such as severe traumatic brain injury,4HS is an appealing alternative because its reflection coefficient is superior to that of mannitol (1.0 vs.  0.9; i.e. , HS does not cross an intact blood–brain barrier).5,6HS also differs from mannitol in that it has little diuretic effect and thus should better maintain cerebral perfusion pressure.6At the same time, HS and mannitol have similarities. Both agents have potential to improve blood rheology.4–7This can serve to either reduce brain blood volume or improve flow through stenotic vessels. In contrast, in the presence of blood–brain barrier breakdown, both agents can accumulate over time in brain parenchyma and negate any beneficial effect or even increase intracranial pressure (ICP).5The question, therefore, is whether HS offers efficacy superior to mannitol without adding risk.

Numerous studies have investigated the efficacy of different HS formulations (2–30% wt/vol) in the setting of neurologic injury.6In contrast to the intraoperative investigation by Rozet et al. ,1the following studies compared mannitol with HS in neuro–critical care populations. Vialet et al.  8performed a randomized comparative study of 20% mannitol and 7.5% NaCl, administered in equal volumes (not equiosmolar loads; 20% mannitol is approximately equivalent to 3% NaCl), to control ICP in 20 patients with traumatic brain injury. ICP control was better in the HS group, which received the higher osmolar load. Harutjunyan et al.  9randomly assigned 40 patients at risk for intracranial hypertension to receive 7.2% NaCl–hydroxyethyl starch 200/0.5 or 15% mannitol. Both drugs were continuously infused as required to maintain ICP less than 15 mmHg. 7.2% NaCl–hydroxyethyl starch 200/0.5 was more effective and had a greater osmolar load than 15% mannitol. A similar elevation in cerebral perfusion pressure was seen with both drugs (approximately 10 mmHg). Schwarz et al.  10compared equiosmolar loads of 20% mannitol versus  7.5% NaCl–6.0% hydroxyethyl starch for control of intracranial hypertension in nine ischemic stroke patients with 30 episodes of elevated ICP. Patients were randomly assigned to one or the other therapy and subsequent doses were alternated, mannitol or HS. The authors found HS to be more efficacious in controlling ICP, but the small sample size and potential crossover effects create limitations in interpreting the data. In a subsequent study, they found that 10% NaCl was effective in controlling ICP refractory to mannitol therapy in ischemic stroke patients.11The use of HS as a rescue therapy for ICP refractory to mannitol has been reported by others.12These studies, while providing insight, did not adequately assess the comparative efficacy of HS versus  mannitol, due to either nonequiosmolar dosing or trial design.

In the operative arena, three previous investigations have been reported. These studies either did not use equiosmolar loads of mannitol and HS, or used a nonneurosurgical population.13–15De Vivo et al.  13studied 30 neurosurgical patients randomly assigned to three groups: 3% HS, HS–mannitol, or 18% mannitol alone. Therapy was started intraoperatively and continued through postoperative day 3. ICP was decreased by all three strategies. Although 3% HS and 20% mannitol have similar osmolarity, the volumes of the respective solutions infused differed (and thus so did the osmolar load). Changes in serum sodium and potassium were inconsequential, but HS alone preserved central venous pressure better, consistent with the larger volume of HS given. Gemma et al.  14studied 7.5% NaCl versus  20% mannitol in equal volumes of 2.5 ml/kg (but different osmolar loads) in 50 elective craniotomy patients. They found that both therapies had similar effects on mean arterial pressure, central venous pressure, ICP, and brain bulk and concluded that the therapies were equally efficacious. Erard et al.  15conducted a pharmacodynamic study comparing equiosmolar mannitol and HS in 30 nonhemorrhagic surgical patients. Patients were randomized to hypertonic 20% mannitol, 7.5% NaCl, or isotonic 0.9% NaCl. They found that the osmotic changes over time were the same for both mannitol and HS and greater than for 0.9% NaCl.

The study by Rozet et al.  1in this issue of Anesthesiology is the first human study, to our knowledge, to compare equiosmolar loads of mannitol (20% wt/vol; 1,098 mOsm/l) and HS (3% wt/vol; 1,024 mOsm/l) for craniotomy. This prospective randomized blinded study compared the effects of 5 ml/kg 20% mannitol (n = 20), and 5 ml/kg 3% HS (n = 20), on brain relaxation during craniotomy. Half of the patients in both groups had aneurysmal subarachnoid hemorrhage, and the rest had craniotomies for elective surgery. Secondary endpoints included brain oxygen metabolism and electrolyte changes. Both treatments had similar effects on brain relaxation scores. There was no difference between groups in cerebral arteriovenous oxygen or lactate differences, arterial pH, central venous pressure, blood glucose, or plasma and cerebrospinal fluid osmolality. Mannitol had a greater diuretic effect, but arterial blood pressure values remained similar between groups. The mannitol group experienced a decreased fluid balance and increased arterial lactate concentration compared with HS. The HS group experienced increased plasma and cerebrospinal fluid sodium concentrations. ICP was not measured in this study. Therefore, this study provides a valid comparison of the efficacy of mannitol and HS in providing brain relaxation during routine craniotomy.

A number of patient safety concerns arise with the increasing adoption of HS in the perioperative arena. One is the issue of safely administering hyperosmolar solutions (HS or mannitol) through a peripheral intravenous catheter. Most elective craniotomy patients do not have central venous lines (which were placed in all patients, Rozet et al.  1). At our institution, pharmacy guidelines limit peripheral HS administration (maximum 2% NaCl; less than 1,000 mOsm/l) because of concerns for phlebitis and extravasation, although mannitol is allowed. There is little evidence on which to base this practice, and to the extent of our knowledge, there are no adequate data to assess the safety of peripherally administered mannitol relative to HS.16There are case reports of forearm compartment syndrome from extravasated mannitol, although central venous catheters are certainly implicated in some extravasation injuries as well.17,18HS (23.4% NaCl) is a venous sclerosing agent used in dermatology to treat superficial varicosities, and is known to cause skin necrosis with extravasation.19 

Second, is a patient with baseline hyponatremia at risk for central pontine or extrapontine demyelination from the use of HS? Hyponatremic patients were specifically excluded from the study of Rozet et al.  1and most other HS trials. Although central pontine demyelination is attributed to rapid correction of hyponatremia with HS,20central pontine demyelination has not been reported in clinical trials testing HS for other indications (i.e. , fluid resuscitation, ICP control, or brain relaxation).6A pediatric trial examined the brain postmortem and found no evidence for central pontine demyelination.21 

Next, HS therapy results in an elevation of plasma sodium and often a hyperchloremic metabolic acidosis with ongoing therapy (unless formulated with acetate).6Finally, plasma osmolality is frequently pushed beyond the traditional 320-mOsm/l limit during the use of HS (as well as with mannitol, for that matter). Despite these concerns and others, there is little evidence for harm with HS from the existing literature (other than in the chronically hyponatremic patient).22Interestingly, Rozet et al.  1used a second 5-ml/kg bolus if needed to provide adequate brain relaxation (6 patients in each group received this). A patient could receive up to 10 ml/kg 3% HS or 20% mannitol (2 g/kg). Unless the clinical scenario is dire, this aggressive approach cannot be recommended without monitoring plasma osmolality. However, this practice does speak to the apparent safety of these agents and common practice by neuroanesthesiologists and neurointensivists to push beyond the conventional non–evidence-based safety limits. At the same time, Rozet et al.  1exposed only 20 patients to 3% HS, and therefore, the sample size was insufficient to allow comparison of adverse events, which are infrequent.

The authors have made a novel and major contribution to this field by comparing equiosmolar  doses of mannitol and HS for brain relaxation during craniotomy. One can conclude from this study that it is the osmotic effect of the agent that results in brain relaxation and that both agents are equivalent in this regard. This is supported by the previous literature cited above where the hypertonic agent given in greater osmolar dose had greater efficacy in decreasing ICP.

Future larger-scale trials will be needed to address the impact of HS versus  mannitol on clinical outcomes, probably by comparing longer-term use of the two agents for ICP control in traumatic brain injury and various types of stroke (subarachnoid hemorrhage, ischemic stroke, and intracerebral hemorrhage). Quantitative endpoints (i.e. , ICP rather than brain relaxation) will need to be included in these studies, and the issue of bolus administration versus  continuous infusion will need further investigation. The time is ripe for the neuroanesthesia and neuro–critical care communities to start registries of therapies such as HS that will allow for the detection and quantification of uncommon adverse events.

In conclusion, neuroanesthesiologists and neurointensivists should add HS to their pharmacologic armamentarium but should do so with caution with regard to route of administration, magnitude of sodium change, and total osmolar load. Is there sufficient reason to supplant the routine use of mannitol with HS for craniotomy or ICP control? Physiologic data such as those provided by Rozet et al.  1suggest that we are close. However, monitored exposure of a larger number of craniotomy patients to HS will be required to define relative safety.

* Departments of Anesthesiology and Medicine (Neurology), † Departments of Anesthesiology, Neurobiology, and Surgery, Duke University Medical Center, Durham, North Carolina.


Rozet I, Tontisirin N, Muangman S, Vavilala MS, Souter MJ, Lee LA, Kincaid MS, Britz GW, Lam AM: Effect of equiosmolar solutions of mannitol versus  hypertonic saline on intraoperative brain relaxation and electrolyte balance. Anesthesiology 2007; 107:697–704.
Weed LH, McKibben PS: Experimental alteration of brain bulk. Am J Physiol 1919; 48:531–55
Worthley LI, Cooper DJ, Jones N: Treatment of resistant intracranial hypertension with hypertonic saline: Report of two cases. J Neurosurg 1988; 68:478–81
Brain Trauma Foundation: Guidelines for the management of severe traumatic brain injury: Hyperosmolar therapy. J Neurotrauma 2007; 24 (suppl 1):S14–20
Brain Trauma Foundation
Lescot T, Degos V, Zouaoui A, Preteux F, Coriat P, Puybasset L: Opposed effects of hypertonic saline on contusions and noncontused brain tissue in patients with severe traumatic brain injury. Crit Care Med 2006; 34:3029–33
White H, Cook D, Venkatesh B: The use of hypertonic saline for treating intracranial hypertension after traumatic brain injury. Anesth Analg 2006; 102:1836–46
Muizelaar JP, Wei EP, Kontos HA, Becker DP: Cerebral blood flow is regulated by changes in blood pressure and in blood viscosity alike. Stroke 1986; 17:44–8
Vialet R, Albanese J, Thomachot L, Francois A, Bourgouin A, Alliez B, Martin C: Isovolume hypertonic solutes (sodium chloride or mannitol) in the treatment of refractory posttraumatic intracranial hypertension: 2 ml/kg 7.5% saline is more effective than 2 ml/kg 20% mannitol. Crit Care Med 2003; 31:1683–87
Harutjunyan L, Holz C, Rieger A, Menzel M, Grond S, Soukup J: Efficiency of 7.2% hypertonic saline hydroxyethyl starch 200/0.5 versus  mannitol 15% in the treatment of increased intracranial pressure in neurosurgical patients: A randomized clinical trial. Crit Care 2005; 9:R530–40
Schwarz S, Schwab S, Bertram M, Aschoff A, Hacke W: Effects of hypertonic saline-hydroxyethyl starch solution and mannitol in patients with increased intracranial pressure after stroke. Stroke 1998; 29:1550–5
Schwarz S, Georgiadis D, Aschoff A, Schwab S: Effects of hypertonic (10%) saline in patients with raised intracranial pressure after stroke. Stroke 2002; 33:136–40
Horn P, Munch E, Vajkoczy P, Hermann P, Quintel M, Schilling L, Schmiedek P, Schurer L: Hypertonic saline solution for control of elevated intracranial pressure in patients with exhausted response to mannitol and barbiturates. Neurol Res 1999; 21:758–64
De Vivo P, DelGaudio A, Ciritella P, Puopolo M, Chiarotti F, Mastronardi E: Hypertonic saline solution: A safe alternative to mannitol 18% in neurosurgery. Minerva Anestesiol 2001; 67:603–11
Gemma M, Cozzi S, Tommasino C, Mungo M, Calvi MR, Cipriani A, Garancini MP: 7.5% hypertonic saline versus  20% mannitol during elective neurosurgical supratentorial procedures. J Neurosurg Anesthesiol 1997; 9:329–34
Erard AC, Walder B, Ravussin P: Effects of equiosmolar load of 20% mannitol, 7.5% saline, and 0.9% saline on osmolarity, haemodynamics and plasma concentrations of electrolytes. Ann Fr Anesth Reanim 2003; 22:18–24
Suarez JI: Hypertonic saline for cerebral edema and elevated intracranial pressure. Cleveland Clin J Med 2004; 71 (suppl 1):S9–13
Edwards JJ, Samuels D, Fu ES: Forearm compartment syndrome from intravenous mannitol extravasation during general anesthesia. Anesth Analg 2003; 96:245–6
Schummer W, Schummer C, Bayer O, Muller A, Bredle D, Karzai W: Extravasation injury in the perioperative setting. Anesth Analg 2005; 100:722–7
Zimmet SE: Hyaluronidase in the prevention of sclerotherapy-induced extravasation necrosis. Dermatol Surg 1996; 22:73–7
Sterns RH, Riggs JE, Schochet SS: Osmotic demyelination syndrome following correction of hyponatremia. N Engl J Med 1986; 314:1535–42
Peterson B, Khanna S, Fisher B, Marshall L: Prolonged hypernatremia controls elevated intracranial pressure in head-injured pediatric patients. Crit Care Med 2000; 28:1136–43
Dubick MA, Wade CE: A review of the efficacy and safety of 7.5% NaCl/6% dextran 70 in experimental animals and in humans. J Trauma 1994; 36:323–30