OUP user menu

Myoinositol Administration Improves Survival and Reduces Myelinolysis After Rapid Correction of Chronic Hyponatremia in Rats

Stephen M. Silver MD, Barbara M. Schroeder, Richard H. Sterns MD, Amyn M. Rojiani MD, PhD
DOI: http://dx.doi.org/10.1097/01.jnen.0000195938.02292.39 37-44 First published online: 1 January 2006

Abstract

When chronic hyponatremia is rapidly corrected, reaccumulation of brain organic osmolytes is delayed and brain cell shrinkage occurs, leading to the osmotic demyelination syndrome (ODS). We hypothesized that treatment with myoinositol, a major organic osmolyte, could prevent ODS. Severe hyponatremia was induced in adult male rats by administration of arginine vasopressin and intravenous infusion of dextrose and water. Sixty-four hours after induction of hyponatremia, all animals underwent rapid correction of hyponatremia with infusion of hypertonic saline over 4 hours, increasing the serum sodium from 105 to 135 mM; half of the animals were also given myoinositol intravenously beginning 20 minutes before correction and continuing for 28 hours. Serum sodium concentrations were equivalent in both groups at all time points. At 7 days, 7 of 8 animals that received myoinositol survived compared with one of the 9 control animals (p < 0.01). In a second study, sodium was reduced to 106 mM over 64 hours in 24 animals and then corrected by 20 mM over 4 hours with concomitant loading and infusion of either mannitol (control) or myoinositol. Animals were killed 96 hours after correction of hyponatremia was begun. Myoinositol-treated animals had significantly fewer demyelinating lesions than mannitol (2.25 ± 1.1 versus 6.42 ± 1.4 lesions/brain, p < 0.03). We conclude that myoinositol administration improves survival and reduces myelinolysis after rapid correction of chronic hyponatremia in rats.

Key Words
  • Brain osmolarity
  • Central pontine myelinolysis
  • Demyelinating diseases
  • Hyponatremia
  • Myoinositol

Introduction

Brain swelling is minimized in hyponatremia by an adaptive loss of brain solutes (1-5). Electrolytes account for only approximately 60% of the adaptation; the remaining solutes are organic osmolytes, which include myoinositol and several amino acids. Loss of organic osmolytes is maximal approximately 2 days after the onset of hyponatremia in animals. When the serum sodium concentration is returned to normal, brain sodium and potassium content is restored relatively rapidly, but reaccumulation of other organic osmolytes is delayed for several days, leading to brain cell shrinkage and demyelination (1, 6-8). Pontine and extrapontine demyelinating brain lesions develop in humans within 48 hours of rapid correction of hyponatremia, leading to a course of neurologic deterioration and, in some cases, death. This course has been termed central pontine myelinolysis (9), pontine and extrapontine myelinolysis (10), or the osmotic demyelination syndrome (ODS) (11). Animal models that mirrored the clinical scenario were developed in dogs and rodents. The lesions in dogs closely resemble the human disease (12). In contrast, the demyelinating lesions observed after rapid correction of hyponatremia in rats, although histologically identical, are topographically different in that they are extrapontine in distribution. This animal model has been referred to as electrolyte-induced myelinolysis (13) or electrolyte-induced demyelination (14). Thus, despite a range of terms used for the human and animal models, the clinical sequence of events and its pathologic manifestations are identical except for limited topographic variance.

Because depletion of brain organic osmolytes may contribute to development of ODS, we hypothesized that exogenous administration of organic osmolytes might increase their content in brain and protect against development of ODS. Soupart et al have recently demonstrated that rapid correction of chronic hyponatremia in uremic animals resulted in less brain demyelination than in controls, and the major metabolic difference found in uremic animals after correction was a more rapid recovery of brain myoinositol (15, 16). We demonstrated that when plasma sodium is rapidly increased, concomitant myoinositol administration results in a near doubling of brain content of myoinositol within 3 hours compared with animals not given myoinositol (17). The purpose of this study was to investigate in rats the effect of myoinositol administration on mortality and severity of brain demyelination after rapid correction of chronic hyponatremia.

Materials and Methods

For all studies, male Sprague-Dawley rats (Harlan, Indianapolis, IN) weighing 275 to 350 g were used. Rats were housed in an animal facility according to National Institutes of Health and institutional animal care and use guidelines.

Surgical Insertion of an Indwelling Jugular Catheter

Our method, previously described (17), was based on that of Takeyama (18). After surgery, animals were housed individually for the duration of the experiment. This technique allowed continuous intravenous infusion and blood sampling in unanesthetized, unrestrained rats.

Experiment 1: Effect of Myoinositol Administration on Mortality After Rapid Correction of Hyponatremia

Summary of Experiment 1 Protocol

In the first experiment, severe hyponatremia was induced in adult male rats over 24 hours by administration of arginine vasopressin and intravenous infusion of dextrose and water and then was maintained for an additional 48 hours. All animals underwent rapid correction of hyponatremia with infusion of hypertonic saline over 4 hours; half of the animals also were given a bolus infusion of myoinositol 20 minutes before correction and a continuous infusion of myoinositol during correction of hyponatremia. Animals were observed for 7 days after correction of hyponatremia, and morbidity and mortality in animals that received myoinositol infusion were compared with those that did not.

Induction of Chronic Hyponatremia

On day 1, hyponatremia was induced in 2 animals at a time; one was assigned to a control group and one to a myoinositol-treated group. Indwelling catheters were placed in the 2 animals at approximately 8:00 AM. At approximately 2:00 PM, plasma was obtained for measurement of sodium concentration. An Alzet 2001 minipump (Durect Corp., Cupertino, CA) delivering aqueous arginine vasopressin (American Pharmaceutical Partners, Inc., Schaumburg, IL) at 20 mU/hour was placed subcutaneously, and an infusion with 2.5% glucose (Sigma Aldrich, St. Louis, MO) at a rate of 2.0 mL/hour was started and continued to provide a total volume of 0.1 mL/g body weight. Animals were allowed 0.1 mL/g body weight of a 10% glucose and 36% (w/v) sodium-free diet with 5% casein hydrolysate (ICN Biomedicals, Inc., Aurora, OH) added.

On day 2, at 2:00 PM, 24 hours after induction of hyponatremia had begun, plasma was again obtained for measurement of sodium concentration. Infusion was then started with 2.5% glucose at 1.5 mL/hour to provide a total volume 0.05 mL/g body weight, and animals were provided access to a diet with the same content as in day one at 0.05 mL/g body weight.

On day 3, at 2:00 PM, 40 hours after induction of hyponatremia had begun, an infusion of 2.5% glucose was started at a rate of 1.0 mL/hour to provide a total volume of 0.05 mL/g body weight. No oral diet was offered to the animals.

On day 4, at 8:00 AM, 64 hours after induction of hyponatremia had begun, plasma was obtained for sodium, glucose, urea, and myoinositol measurement. At this point, animals were divided into 2 groups, control and myoinositol-infused.

Precorrection Loading With Myoinositol

Group I (controls, n = 9) consisted of animals given an intravenous infusion of 5 mL of 105 mM NaCL (Sigma Aldrich) (the approximate sodium concentration of plasma) over 20 minutes. Group II (myoinositol-infused, n = 8) consisted of animals preloaded with 2.22 mmol myoinositol/kg body weight (Sigma Aldrich) in 5 mL of 105 mM NaCl over 20 minutes before rapid correction of hyponatremia.

Rapid Correction of Hyponatremia

In group I (controls), after the 20-minute loading infusion, serum sodium was rapidly corrected with an intravenous infusion of 0.75 M NaCl at a rate of 1.5 mL/hour for 4 hours. Animals were then given an intravenous infusion of 0.15 M NaCl for 24 hours at a rate of 0.25 mL/hour. Plasma was obtained at the beginning and after 1 and 4 hours of the hypertonic saline infusion and at end of the 24-hour infusion of 0.15 M saline.

In group II (myoinositol-infused), after the 20-minute loading infusion, sodium was rapidly corrected with an intravenous infusion of 0.75 M NaCl at a rate of 1.5 mL/hour for 4 hours followed by 0.15 M NaCl like in group I. Myoinositol was added to the infusate during rapid correction of hyponatremia and was administered at a rate of 2.22 mmol/kg per hour for 4 hours followed by 0.37 mmol/kg per hour for 24 hours in the 0.15 M NaCl infusion.

Postcorrection and Infusion Procedures for Animals in Experiment 1

After correction, rats were housed individually with free access to pelleted chow (PMI Nutrition International, LLC, Brentwood, NJ) and water and were assessed daily over a 7-day period for behavioral abnormalities and survival. Animals were observed immediately before correction of hyponatremia (day zero) and on days one and 4 through 7 after correction of hyponatremia. Based on these observations, animals were categorized into one of the following sets of behavioral characteristics: normal appearance, hyperirritability, and/or poor grooming; lethargy and/or ataxia; or moribund state. Each animal was assigned only one category each day based on the most severe behavioral abnormality they exhibited. Body weight was measured before catheter placement and after death or at the end of the experiment.

Experiment 2: Effect of Myoinositol on Brain Demyelination After Rapid Correction of Chronic Hyponatremia

Summary and Rationale of Experiment 2 Protocol

The purpose of this experiment was to determine the effect of myoinositol administration on brain-demyelinating lesions after rapid correction of hyponatremia. These lesions have been observed to occur by 4 days after correction. The protocol was modified from that in experiment 1 to attempt induction of less severe hyponatremia (less intravenous water was administered on day 3 of induction of hyponatremia). Animals were given less glucose in their diet and intravenous fluid than in experiment 1 because mild hyperglycemia was observed in experiment 1. All animals underwent rapid correction of hyponatremia with infusion of hypertonic saline over 4 hours; however, the hypertonic saline administration was adjusted for body weight instead of being given as a fixed volume like in experiment 1. Half of the animals also were given myoinositol intravenously like in the first experiment, and the other half of the animals (controls) received mannitol instead of myoinositol in an attempt to control for the increase in osmolality caused by myoinositol. Animals were observed for 4 days after correction of hyponatremia was initiated, and morbidity in animals that received myoinositol infusion was compared with those that received mannitol. Animals were killed before 4 days if they appeared near death. At 4 days, surviving animals were killed. Brain tissue was compared between the 2 groups for the presence and severity of demyelination.

Experiment 2: Methods

On day 1, hyponatremia was induced in 2 animals at a time. Animals were not preassigned to a group. Catheter placement, vasopressin administration, measurement of plasma sodium, and dietary allowance were performed like in experiment 1. An infusion with 1.5% glucose at a rate of 2.0 mL/hour was started and continued to provide a total volume of 0.09 mL/g body weight. Animals were allowed 0.08 mL/g body weight of a 5% glucose and 36% (w/v) sodium-free diet (ICN Biomedicals, Inc.) with 5% casein hydrolysate (ICN Biomedicals, Inc.) added.

On day 2, at 2:00 PM, 24 hours after induction of hyponatremia was begun, plasma was obtained for measurement of sodium concentration. Infusion was then started with 1.5% glucose at 1.5 mL/hour and continued to provide a total volume 0.05 mL/g body weight and animals were offered a diet with the same content as in day 1 at 0.05 mL/g body weight.

On day 3, at 2:00 PM, 40 hours after induction of hyponatremia was begun, an infusion of 1.5% glucose was started at a rate of 0.3 mL/hour and continued to provide a total volume of 0.0085 mL/g body weight. No oral diet was offered to the animals.

On day 4, at 8:00 AM, 64 hours after induction of hyponatremia was begun, plasma was obtained for sodium, glucose, urea, and myoinositol measurement. At this point, animals were randomized blindly into mannitol (control) and myoinositol-infused animals.

Precorrection Loading With Myoinositol or Mannitol

In group I (mannitol-infused controls = 12), animals were infused with 0.77 mmol/kg body weight of mannitol (Sigma Aldrich) in 5 mL of 110 mM NaCl over 20 minutes immediately before rapid correction of hyponatremia.

In group II (myoinositol-infused, n = 12), animals were infused with 2.22 mmol myoinositol/kg body weight in 5 mL of 110 mM NaCl over 20 minutes immediately prior to rapid correction of hyponatremia.

Rapid Correction of Hyponatremia

In group I (mannitol-infused), after the 20-minute loading infusion, serum sodium was rapidly corrected with an intravenous infusion of in 0.75 M NaCl at a rate of 5 mL/kg body weight/hour for 4 hours. Animals were then given an intravenous infusion of 0.15 M NaCl for 24 hours at a rate of 0.25 mL/hour. Mannitol was added to the infusate and was administered at a rate of 0.77 mmol/kg per hour for 4 hours in the hypertonic saline infusion followed by 0.13 mmol/kg per hour for 24 hours in the 0.15 M NaCl infusion. Plasma was obtained before and 1 and 4 hours after the hypertonic saline infusion and at the end of the 24-hour infusion of 0.15 M NaCl.

In group II (myoinositol-infused), after the 20-minute loading infusion, sodium was rapidly corrected followed by 0.15 M NaCl infusion like in group I. Myoinositol was added to the infusate and was administered at a rate of 2.22 mmol/kg per hour for 4 hours followed by 0.37 mmol/kg per hour for 24 hours.

Experiment 2: Postcorrection and Infusion Procedures

After correction, rats were housed individually with free access to pelleted chow (Sigma Aldrich) and water and were assessed daily over a 72-hour period (until 96 hours after correction of hyponatremia was initiated) for signs of behavioral abnormalities like in experiment 1. Animals that appeared near death before day 4 were killed. On day 4, surviving rats were killed by decapitation using a guillotine. The skull was opened using rongeurs and the brain was extracted from the cranial cavity and immediately immersed in 10% buffered formalin solution for later pathologic analysis. Body weight was measured before catheter placement and before death.

Analytical Procedures

Plasma Sodium Concentration

Plasma sodium concentration was measured by flame photometry (Instrumentation Laboratory Model 943, Boston, MA).

Plasma Myoinositol, Mannitol, Glucose, and Urea Concentration

The method for measurement of organic osmolyte concentration in plasma has been described previously (19). The content of myoinositol, urea, and glucose were determined by high-pressure liquid chromatography using a SugarPak II column (Waters Corp., Milford, MA). Maltose served as the internal standard. Levels of organic osmolytes were quantified on a Waters Millenium Chromatography System interfaced with Empower.

Brain Histopathologic Evaluation

Animals were killed by decapitation. Brains were removed from the cranial cavity and fixed in 10% neutral buffered formalin for at least 1 week before sectioning. The brainstem and cerebellum were separated from the cerebrum by a transverse section at the level of the upper midbrain. Four serial coronal sections were made through the cerebral hemispheres. Tissue was routinely processed for paraffin embedding. Coronal slices were embedded in paraffin such that all sections were obtained from the posterior surface of the specimen. Sections were stained by hematoxylin and eosin (H&E) and Luxol fast blue-H&E methods. Demyelinated lesions were counted in all 4 cerebral slices in each animal and expressed as a total within each experimental group. Histologic assessment of lesion severity was graded on a scale of 1 or 2, in which a grade of 1 represented one third or less involvement of a particular anatomic structure and a grade of 2 represented more than one third involvement. A lesion severity score/brain was determined by the sum of the grades assigned to each lesion observed/brain. Lesions were counted independently in each hemisphere. Based on previous studies (14), the following anatomic sites were evaluated: anterior commissure, basal ganglia (pencil fibers), external capsule, thalamus, and cerebral peduncles. All histopathologic evaluations were performed in a blinded fashion.

Statistical Methods

Plasma Values in Experiments 1 and 2

The plasma sodium, glucose, urea, myoinositol, and mannitol levels were compared between groups at each time point. Values that were normally distributed were compared using a t-test with significance set at a 2-sided alpha of 0.05. Values that were not normally distributed were compared by a Kruskal-Wallis test (for nonparametric data).

Survival in Experiment 1

The overall survival was compared using Kaplan-Meier survival analysis; animals were censored at the time of natural death (20).

Number and Severity of Brain Lesions in Experiment 2

The major outcome variables were the number and severity score of lesions on brain pathology. Because these numbers were not normally distributed, a Kruskal-Wallis test (for nonparametric data) was used to compare the groups.

Results

Experiment 1

Nine animals underwent induction and correction of hyponatremia without myoinositol. Eight animals underwent induction and correction of hyponatremia with myoinositol infusion; one animal died as a result of bleeding after indwelling catheter placement but before receiving myoinositol and before undergoing correction of hyponatremia. Plasma sodium levels were equivalent in the 2 groups of animals at all time points (Table 1). Plasma myoinositol levels were increased significantly in animals after receiving myoinositol infusion. There were small differences in plasma urea and glucose between the groups 24 hours after correction of hyponatremia.

View this table:
TABLE 1.

The survival of animals receiving myoinositol was significantly improved compared with controls (Fig. 1). By 7 days, the planned end of the protocol, one animal survived correction of hyponatremia without myoinositol treatment and one animal receiving myoinositol died. After a brief period of irritability immediately after correction of hyponatremia, the condition of the control animals progressively deteriorated. Their course was marked by poor grooming, ataxia, lethargy, and death (Table 2). Myoinositol-treated animals also were observed to be irritable and lethargic immediately after correction of hyponatremia. After this, most of the animals appeared healthy through the course of observation, although one animal died after 4 days and another appeared ill. Even if this animal was included as a death, the difference in survival between control and myoinositol-treated animals remains highly significant. Weight of the control animals at the time of death or at the end of 7 days was significantly less than animals treated with myoinositol (188 ± 11 vs 288 ± 4 g, p < 0.01).

View this table:
TABLE 2.
FIGURE 1.

Kaplan-Meier plot comparing survival of control versus myoinositol-treated animals in experiment 1. *, p < 0.05; #, p < 0.01 versus myoinositol-treated group by log rank test.

Experiment 2

Plasma values are shown in Table 3. Plasma sodium levels were equivalent between both groups throughout the experiment. Body weight was equivalent in the 2 groups before induction of hyponatremia (333 ± 20 g in myoinositol-treated animals vs 318 ± 13 g in controls). The glucose content of the infusion during induction of hyponatremia was less than that of experiment 1 and animals were less hyperglycemic. Animals were minimally less hyponatremic before correction (1-2 mM) than animals in experiment 1. The concentrations of mannitol and myoinositol after precorrection bolus and at 4 hours after initiation of correction were not equivalent; plasma myoinositol concentration was approximately 4 mM higher than mannitol concentration at these time points. Four hours after the start of the hypertonic saline infusion, plasma sodium had increased by approximately 20 mM.

View this table:
TABLE 3.

Animal behavior is described in Table 4. By day 4, 3 mannitol-treated animals had been killed because they appeared near death, and approximately half of the remaining animals demonstrated significant behavioral abnormalities. In contrast, only one myoinositol-treated animal had to be killed by day 4 and all but one of the remaining animals appeared healthy. Animal weights in the two groups (304 ± 22 g) in the myoinositol- and 292 ± 20 g in mannitol-treated animals) were not significantly different at the time of death.

View this table:
TABLE 4.

Histopathologic examination of the brains identified specific demyelinated areas in a topographic distribution, similar to that previously described (14). Lesions were frequently bilateral and fairly symmetric. They were typically restricted to white matter with almost no involvement of adjacent gray matter. The external capsule and adjacent pencil fibers of the striatum were most frequently involved. In addition to the extent of involvement of individual anatomic structures, mild lesions displayed only myelin pallor and some vacuolation. Severe and usually larger lesions frequently were well-circumscribed and also had reactive cells. Some lesions were seen to have endothelial hyperplasia and occasionally early neovascularization (Fig. 2A-D).

FIGURE 2.

(A) Marked loss of myelinated fibers can be seen in the external capsule (arrows), basal ganglia (*), and anterior commissure (short arrow). Luxol fast blue/hematoxylin & eosin (H&E)-stained section of a coronal slice through the right cerebral hemisphere of a mannitol-treated animal. Original magnification: 40×. (B) Anterior commissure with only a narrow rim of preserved myelin seen in the upper left corner (arrows). Pallor and focal vacuolation is seen within the demyelinated area. The lesion was graded as a severe lesion based on greater than two thirds of the anterior commissure being demyelinated. Luxol fast blue/H&E-stained section of the right cerebral hemisphere of a mannitol-treated animal. Original magnification: 200×. (C) Midbrain-thalamus junction with bilateral, large demyelinating lesions (severe lesion). Luxol fast blue/H&E-stained section of the most posterior coronal slice, cerebral hemisphere of a mannitol-treated animal. Original magnification: 40×. (D) At higher magnification, mild hypercellularity and prominent vessels (arrows) can readily be seen. Also note the presence of preserved and unaltered neurons in the adjacent tissue. Luxol fast blue/H&E-stained section of the most posterior coronal slice, cerebral hemisphere of a mannitol-treated animal. Original magnification: 400×.

Comparison of number and severity of demyelinating lesions between groups is shown in Figure 3A-B. There were significantly fewer demyelinating lesions in the myoinositol-treated animals (Fig. 3A). This difference was sustained when severity as well as frequency of lesions was considered (Fig. 3B).

FIGURE 3.

(A) Number of demyelinating lesions/brain in experiment 2. (B) Lesion score/brain based on number and severity of demyelinating lesions. Data are means ± standard error of mean. *, p < 0.03.

Discussion

In the present study, survival in rats after rapid correction of chronic hyponatremia is substantially improved by concomitant administration of myoinositol. Myoinositol administration appears responsible for this survival benefit because the rapid increase in serum sodium in myoinositol-treated animals was equivalent to controls. The model used to induce and correct hyponatremia in these studies yielded reproducible plasma sodium concentrations with little variability among animals. The only death related to uncorrected hyponatremia was caused by a bleeding complication from the intravenous catheter. The stability and uniformity of the model increases the likelihood that the observed difference between myoinositol-infused animals and controls is caused by the myoinositol. In experiment 2, myoinositol-treated animals appeared to have less behavioral abnormalities and had significantly fewer and less severe brain demyelinating lesions when compared with mannitol-infused controls.

The osmotic demyelination syndrome was the most likely factor leading to death in animals after rapid correction of hyponatremia. There was significantly more weight loss in controls and, although unmeasured, possibly significant metabolic abnormalities in the control group at the time of death. If so, we suspect that these abnormalities would be those seen in moribund animals and most likely the result, rather than the cause, of neurologic damage in this group. Several groups of investigators have demonstrated in experimental studies that rapid correction of hyponatremia of at least 3 days' duration results in weight loss, neurologic deterioration, and death and pathologic findings of demyelination (6, 12, 13, 21-23). The minimum increase in serum sodium concentration required to produce demyelinating lesions is approximately 15 mM a day. In experiment 1, animals that were corrected from serum sodium of 102 mM to 131 mM in 4 hours without myoinositol replacement had an 88% mortality rate after 6 days. In a study by Verbalis that also used intravenous infusion to correct serum sodium, 30% of rats with chronic hyponatremia (113 mM) corrected to levels of more than 140 mM over 6 hours died or were killed because of weight loss or decreased neurologic function, and 90% had neuropathologic evidence of demyelination (21). Sterns et al observed a 40% mortality rate 5 days after correction of chronic hyponatremia (95 mM) corrected by 25 mM in 2 hours to moderately hyponatremic levels (119 mM) by intraperitoneal hypertonic saline (22). In these studies, animals initially appeared well but developed neurologic symptoms the next day. A similar pattern occurred in our study, but the mortality rate of our control animals appeared higher and the onset of neurologic symptoms appeared in some animals after 24 hours. Soupart et al also observed neurologic symptoms within 24 hours in the majority of animals corrected by 20 to 25 mM/24 hours of rapid correction and a mortality rate comparable to ours (23).

One can speculate that a factor contributing to the relatively high mortality in our control animals in experiment 1 was compromised nutritional status. Our animals ate little during the induction of hyponatremia. They were not gavage fed protein and their weight during induction of hyponatremia did not change despite a presumed increase in total body water. Malnutrition may predispose to ODS, possibly by resulting in a decrease in available myoinositol or other osmolytes in brain or in plasma. Thus, myoinositol administration may have been especially efficacious in our animal model. However, brain and plasma myoinositol content in our model is comparable to other models of hyponatremia. The high survival rate of myoinositol-treated animals compared with controls in other studies indicates that myoinositol administration has a salutary effect on the course of ODS despite the relatively high mortality rate of our controls. If aggressive nutritional support was given in our study, analogous to treatment of patients, the mortality rate in control animals may have been less (24).

In experiment 2, mannitol was used in control animals to attempt to control for the osmolality increase induced by infusion of myoinositol. In our previous study (17), mannitol was given in a lower dose than myoinositol to account for metabolism of myoinositol, which resulted in mannitol levels that were approximately 6 mM lower than myoinositol levels. Thus, in the present study, the mannitol dose was doubled and myoinositol infusion decreased by one third with the goal of making these levels equivalent. In response to these changes, myoinositol levels did decrease but mannitol levels did not increase; thus, a 4 mM difference in plasma concentrations of these solutes remained before and during rapid correction of hyponatremia. However, this difference is relatively small. Hypothetically, the higher osmolality of the myoinositol-infused animals during rapid correction would tend to exacerbate demyelination, further supporting the contention that myoinositol is protective.

The increased survival and decreased brain demyelination observed in myoinositol-treated animals supports the hypothesis that depletion of organic osmolytes contributes to the development of osmotic demyelination. Consistent with this hypothesis, we recently demonstrated that intravenous infusion of myoinositol increases brain myoinositol content compared with controls within 200 minutes provided that plasma tonicity is also increasing (17). This result indicates that administered myoinositol increases brain myoinositol content rapidly when hyponatremia is rapidly corrected like it was in the present study. Further indirect evidence of the protective effect of myoinositol was provided by Soupart et al. They demonstrated in experimental studies that both correction of hyponatremia with urea and correction of hyponatremia in azotemic animals decreases the incidence of ODS (15). In azotemic animals undergoing rapid correction of hyponatremia, brain content of myoinositol is nearly 3 times higher than that of uncorrected controls within 2 hours after correction (16); the increment in brain myoinositol observed in these azotemic animals is comparable to what is observed when myoinositol is administered during correction of hyponatremia to nonazotemic animals. There has been a significant experience in Europe using urea to treat patients with hyponatremia in the setting of Syndrome of Inappropriate Anti-Diuretic Hormone secretion. The effect of urea administration on brain myoinositol uptake has not yet been studied in experimental models. It is unclear whether the effect of urea on the course of ODS is related to an effect on brain myoinositol or whether the effects of myoinositol and urea administration are additive in prevention of ODS.

It is not clear how myoinositol exerts its protective effect. It could protect merely by decreasing cell shrinkage. Regional differences in organic osmolyte content in the brain correlate to susceptibility to osmotic demyelination after rapid correction of hyponatremia, and these regions might be more affected by the increase in brain myoinositol content (25) after myoinositol supplementation. The blood-brain barrier is disrupted in osmotic demyelination (26), and myoinositol might prevent endothelial shrinkage, thereby avoiding disruption of the blood-brain barrier. However, although myoinositol is a significant brain osmolyte, it is only one of at least several, and might not be expected quantitatively to prevent demyelination as effectively as it did in our study. In addition to its role as an osmolyte, myoinositol might also function to prevent apoptosis. Recently, apoptotic markers have been identified in autopsy specimens from patients with central pontine myelinolysis (27). Alfieri et al have demonstrated that in porcine pulmonary endothelial cells exposed to hypertonicity, addition of betaine and myoinositol prevented apoptotic cell morphology and decreased caspase-3 activity (a marker for apoptotic activity) (28).

Our observations indicate that myoinositol administration prevents ODS after correction of chronic hyponatremia in rats. Myoinositol has few side effects and may have potential as an adjunctive therapy for patients undergoing correction of hyponatremia who are at risk of developing ODS. As V2 receptor blockers become available for the treatment of hyponatremia, more patients may be put at risk for ODS as a result of rapid correction of hyponatremia. Further study is needed to determine the ideal dose and timing of myoinositol administration and to better understand its effect on brain pathology and metabolism.

Acknowledgments

The authors thank Dr. Ruth Kouides for her expert assistance with statistical analysis of our data.

Footnotes

  • This work was supported by the Upstate New York Chapter of the National Kidney Foundation.

References

  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
View Abstract