The Neurological Complications of Hypotonic States

Sources: Steele A, et al. Postoperative hyponatremia despite near-isotonic saline infusion: A phenomenon of desalination. Ann Intern Med 1997;126:20-25; Laureno R, Karp BI. Myelinolysis after correction of hyponatremia. Ann Intern Med 1997;126:57-62.

Severe hyponatremia leading to convulsions, cardiopulmonary arrest, and death or permanent brain damage after elective surgery in healthy women was reported by Arieff (N Engl J Med 1986;314:1529-1535) and engendered much discussion as to its causes. Arieff concluded that postoperative hyponatremia was due to two factors: first, and perhaps essential, the infusion of excessive amounts of electrolyte-free water (5% dextrose in water or hypotonic saline) and second, actions of antidiuretic hormone to prevent excretion of water. Steele et al, however, observed that some patients who died of severe hyponatremia 1-3 days after surgery had not received electrolyte-free water in amounts needed to cause hyponatremia. Therefore, to establish the basis for postoperative hyponatremia, they prospectively studied 22 women who had elective gynecologic surgery. Plasma electrolyte levels were measured when anesthesia was induced and 24 hours later. Data on water and electrolyte balance were collected for the same 24-hour period. All patients received intravenous fluids, half as isotonic saline and half as Ringer lactate. Patients were not allowed any fluids by mouth.

At the time of induction of anesthesia, plasma sodium concentration was 140 ± 1 mmol/L; 24 hours later, it decreased in 21 of 22 patients (mean decrease, 4.2 ± 0.4 mmol/L; P < 0.001); the lowest level was 131 mmol/L in two patients. In all patients, urine remained hypertonic. Therefore, the study showed that postoperative hyponatremia is the result of two factors: 1) addition of electrolyte-free water by infusion, renal generation, or both; and 2) the presence of antidiuretic hormone to prevent excretion of electrolyte-free water.

These results suggest that perioperative fluid management should follow these or similar guidelines.

1. Hypotonic IV fluids should not be given in the perioperative period unless the patient is hypernatremic;

2. The minimum volume of isotonic fluid to maintain normal hemodynamics should be infused.

3. Plasma sodium concentration should be checked if more than 2-3 L of hypertonic urine (specific gravity > 1.020) has been excreted in the first 24 hours.

Laureno and Karp provide a useful review of myelinolysis, also called osmotic demyelination syndrome, a neurologic disorder that can occur after too rapid correction of hyponatremia; defined as faster than 12 mmol/L in 24 hours (Sterns RH, et al. N Engl J Med 1986; 314:1535-1542). In some cases, however, myelinolysis has followed correction of serum sodium within the safe guidelines. It may be that prolonged hyponatremia and/or alcoholism predisposed these individuals to demyelination despite a low rate of correction. Therefore, the reviewers urge moderation in treating even severe hyponatremia as long as the patient is clinically stable. In cases with convulsions, stupor, or coma, the administration of saline may be necessary to reverse signs of encephalopathy, but the rate of correction should be kept below 10 mmol/L during any 24-hour period, if possible.


Regular readers may well note that the editors have previously reported on the neurological disorders of hypotonic states (Neuro Alert 1994;12:33-40) and their correction (Neuro Alert 1995;14:27-28). Although Ayus and Arieff maintained that brain damage and postoperative hyponatremia were both common and had a predilection for women (Neurology 1996;46:323-328), most clinicians including ourselves have found it to be relatively uncommon. In fact, a recent publication (Wijdicks EFN, Larson TS. Ann Neurol 1994;35:626-628) stated that during a 16-year interval from 1976 to 1992 at the Mayo Clinic, there were no cases of fatal postoperative hyponatremia in women. Nevertheless, postoperative hyponatremia in women resulting in death or permanent brain damage and the osmotic demyelination syndrome remain as iatrogenic conditions that can be prevented.

The development of hyponatremia after surgery requires that electrolyte-free water be added to the extracellular fluid. Steele et al found that, in their patients, most of the water was generated by the kidney. At the same time factors such as stress, pain, or medication triggered the release of antidiuretic hormone. In this state, the infusion of isotonic fluid did not prevent mild hyponatremia. Had hypotonic fluids been infused, the result would have been clinically significant or even catastrophic hyponatremia.

The study of Steele et al does not answer the question as to why women are so much more likely than men to suffer brain damage in hyponatremic states. Ayus and Arieff have previously speculated that estrogens inhibit sodium and potassium ATPase activity in brain and thereby impair extrusion of sodium from the brain as a defense against hyponatremia. They also speculate that an elevation of vasopressin in the postoperative state may decrease cerebral perfusion and brain oxygen utilization because of a possible difference in vascular reactivity to vasopressin between females and males. Whatever the mechanisms of hyponatremic injury, the fact remains that although postoperative hyponatremia occurs equally in men and women, the morbidity appears primarily in women.

The review of osmotic demyelination summarized above emphasizes that the management of the hyponatremic patient involves comparing the risk of disability or death from untreated hyponatremia with the risk of myelinolysis from correction of hyponatremia. Currently, the decision whether to treat a patient with hyponatremia in a passive manner with water restriction or with an active therapeutic intervention such as the infusion of hypertonic NaCl is primarily based on the presence or absence of neurologic signs and symptoms. The presence of CNS symptoms has been shown to be associated with brain edema on both neuroradiologic studies and at autopsy. In symptomatic patients, active therapy is necessary to prevent brain herniation and cardiorespiratory arrest. In an asymptomatic hyponatremic patient, water restriction is usually the appropriate therapy. Thus, the physician must asses the severity of symptoms and, if possible, the duration of hyponatremia before he takes corrective action. He must also realize that it may be impossible to define consistently a level of correction of the serum sodium that is completely free of the risk of neurologic complications. In the end, as always, the physician must use his best clinical judgment. —jjc