17 

Are There Safe Upper and Lower Limits for Serum Sodium and Serum Osmolality in Head-Injured Patients? Should I Use Hypertonic Saline or Fluid Restriction to Treat Hyponatremia?

Shon W. Cook, Pejman Cohan, Daniel F. Kelly

BRIEF ANSWER

Background and Literature Review: Safe Upper and Lower Limits for Serum Sodium and Serum Osmolality in Head-Injured Patients

No class I studies have addressed the safe range of serum sodium and serum osmolality in head-injured patients. For ethical reasons, it is unlikely that a prospective, randomized, controlled study will be done to determine definitively a safe upper or lower limit for serum sodium in head-injured patients. Currently available evidence is primarily class III, but some class II studies (not limited to head-injured patients) are available.

Significance of Hyponatremia

Acute hyponatremia is relatively common following head injury, occurring in 8% of moderately or severely injured patients in one retrospective series (class III data).² Areview of the literature since 1935 reveals multiple cases of symptomatic hyponatremia following neurologic surgery, sometimes accompanied by coma and seizures (class III data).³ Hyponatremia is often delayed in onset and may not appear for several days after an intracranial insult (class III data).⁴

Class III evidence suggests that hyponatremia may lower the seizure threshold,⁵ exacerbate cerebral edema,² and cause or contribute to delayed elevations in intracranial pressure.⁴ Conscious patients may initially develop such nonspecific symptoms as headache, nausea, emesis, and weakness. Worsening hyponatremia may lead to progressive neurologic deterioration and convulsions (class III data).⁶ Class II data in subarachnoid hemorrhage patients indicate that those with hyponatremia have a significantly higher risk of cerebral ischemia than those without hyponatremia, but this predisposition is apparently not associated with an increase in mortality.⁷

Predicting the serum sodium concentration at which an individual will develop symptoms is difficult, but the degree of hypo-osmolality and especially the rate of decrease seem to correlate in general terms with severity of symptoms (class II data).⁸ Many patients seem to become symptomatic or deteriorate neurologically when serum sodium levels fall below 130 mEq/L. In one series, all patients with a serum sodium level below 125 mEq/L had some degree of symptomatology (class II data).⁸ Major neurologic manifestations, including seizures and coma, are common when sodium concentrations fall below this level (class III data).^5,9 Class II and III evidence suggests that female gender (menstruant women), hypoxia, and young age may worsen the prognosis of hyponatremic encephalopathy.¹

Etiology of Hyponatremia

In the early 1950s, renal salt wasting was first reported in patients with intracranial disease. Because the pituitary-adrenal axis was intact, the excessive natriuresis was thought to be due to disruption of direct neural regulation of renal tubular activity and therefore was named CSW (class III data).^10–12 In 1957, however, SIADH was described (class III data),¹³ and it soon became the favored explanation for hyponatremia in neurosurgical patients (class III data).^2,14–16 The popularity of SIADH as a diagnosis in hyponatremic patients fueled recommendations to restrict fluid intake during the acute management of neurosurgical patients (class III data),¹⁷ including those with head injuries (class III data).¹⁸

In recent years it has become evident that many hyponatremic patients with acute brain disease are in fact relatively hypovolemic (class II and III data),^19–21 compatible with the original diagnosis of CSW. There has also been an increase in the number of reports describing such instances as hyponatremia with normal ADH (class III data)^22,23 and high plasma atrial natriuretic peptide (class III data)²²; delay or failure of correction with fluid restriction (class III data)^24,25; successful treatment with mineralocorticoids (class III data)^25,26 or with administration of fluids and salt (class II data)²⁷; and clinical biochemical results being more consistent with CSW than SIADH (class III data).28 Overall, these studies support the concept that CSW is a common cause of hyponatremia in patients with cerebral pathologies. Several articles nicely summarize the mounting evidence in favor of CSW over SIADH as the primary cause of hyponatremia in acutely ill neurologic patients.^{6,21, 29–34}

Differentiation Between Cerebral Salt Wasting and the Syndrome of Inappropriate Secretion of Antidiuretic Hormone

The differentiation between CSW and SIADH is critical because not only are the treatments different, but the incorrect choice of treatment may exacerbate conditions associated with hyponatremia. For example, a retrospective review of subarachnoid hemorrhage patients revealed that fluid restriction in patients with hyponatremia was associated with a higher risk of cerebral infarction (class III data).³⁵

The physiologic changes in these two entities can be differentiated based on the mechanism of each disorder. Simplistically, CSW is a renal loss of sodium, and with it water. Conversely, SIADH is retention of free water, with a compensatory attempt at volume reduction by excretion of sodium. See Table 17-1 for a summary of clinical and laboratory findings used to differentiate CSW from SIADH.

The key factor distinguishing CSW from SIADH is hypovolemia (class II).^21,27 Clinically, patients with CSW show signs of dehydration such as poor skin turgor, lack of hand vein distention when in a dependent position, and dry mucous membranes. Orthostatic hypotension develops if dehydration is severe. A rapid weight loss with a negative fluid balance also suggests hypovolemia. If invasive monitoring is available, it will show that patients with CSW have a low central venous pressure, a low pulmonary capillary wedge pressure (in the absence of cardiac or pulmonary disease), and a low plasma volume.

Table 17-1 Comparison of Findings in CSW and SIADH

Laboratory values consistent with CSW include markedly increased urinary sodium, mildly increased or normal urinary potassium, and increased or normal urine osmolality. The hematocrit and blood urea nitrogen (BUN) are elevated due to dehydration, and uric acid should be normal.

In SIADH, water is not excreted appropriately, so unlike CSW, the extracellular volume increases. Consequently, the glomerular filtration rate increases, the renin-angiotensin-aldosterone axis is suppressed, and atrial natriuretic peptide is secreted. Each of these responses leads to a paradoxical increasein urinary sodium excretion. The diagnosis is based on the findings of hypervolemic hyponatremia, serum hypo-osmolality, and an inappropriately high urine osmolality. A definitive diagnosis of SIADH can be made by observing an abnormal response to a water load test. However, because this test may exacerbate hyponatremia in patients who may already be volume-overloaded, it should be used judiciously in critically ill patients.

In a retrospective series of postoperative female neurosurgical patients in whom the average plasma sodium level reached 108 mEq/L, all 15 patients developed grand mal seizures, followed by respiratory arrest requiring intubation (class III data).⁵ At that time, the urinary sodium concentration and osmolality averaged 68 mEq/L and 501 mOsm per kilogram, respectively, suggesting inappropriate secretion of antidiuretic hormone. Net postoperative fluid retention averaged 7.5 L.

Hypernatremia

It is likely that hypernatremia in head-injured patients is most commonly iatrogenic, although unrecognized diabetes insipidus (DI) can also result in rapid loss of free water and hypernatremia. DI may be caused by cranial base fractures and resultant hypothalamic and/or pituitary stalk injuries. In one retrospective military series, posttraumatic hypernatremia was seen in 8% of head-injured patients; one third of the hypernatremic patients died within 30 days of injury (class III data).36 Aretrospective review of neurosurgical patients who developed DI (most of whom suffered from subarachnoid hemorrhage or severe head injury) reported a mortality rate of 72% (class III data).³⁷ The diagnosis of DI is typically made after polyuria of over 200 mL/h for 3 hours accompanied by a urine specific gravity of 1.005 or less and a rising serum sodium concentration.

Background and Literature Review: Treatment of Hyper- and Hyponatremia

Class I studies aimed at defining the optimal treatment modality for hyponatremia in head-injured patients are not currently available. This problem is due, at least in part, to the continued controversy about the underlying etiology of hyponatremia in this patient population. Class II and III studies are reviewed below.

Treatment of Cerebral Salt Wasting (Table 17-2)

The treatment of CSW should be dictated by the magnitude of the hyponatremia, by the presence and severity of clinical symptoms, and by the degree to which the patient is at risk for complications of hyponatremia. For head-injured patients with brain swelling and/or hemorrhagic lesions on computed tomography scan, hyponatremia of any degree should be aggressively treated because of the potential for worsening brain swelling. In addition, hyponatremia may lower the seizure threshold in patients who are already predisposed to seizures because of traumatic brain injury. If hyponatremia develops in the acute postinjury period in such patients, rapid correction to mild hyponatremia or to a normal serum sodium level is generally well tolerated.