Atta, Mohamed G. MD, MPH
Water retention and dilutional hyponatremia are frequent complications of advanced liver disease with cirrhosis. Ample data have demonstrated that the severity of hyponatremia and ascites are major determinants of disease severity and prognosis.
Mohamed G. Atta, MD,...Image Tools
When both hyponatremia and ascites are present, the risk of pretransplant death within 180 days may exceed 40%. Hence, expedited consideration of liver transplantation in cirrhotic patients with persistent ascites and hyponatremia, even in those who have a Model for End-Stage Liver Disease (MELD) score below 21, is suggested.
Hyponatremia also is associated with a high rate of neurologic disorders, infectious complications, and renal failure during the first month after transplantation in patients with cirrhosis, as well as lower three-month survival. Thus, the condition should be considered a risk factor not only for death before transplantation but also for impaired early transplantation outcome.
It has been demonstrated that incorporating sodium into the MELD score may provide a more accurate survival prediction than MELD score alone. Serum sodium has been shown to be an earlier and more sensitive test than creatinine to detect circulatory dysfunction resulting in renal failure or death.
Serum creatinine, a determinant of the MELD score, is subject to variation in cirrhotic patients due to decreased hepatic creatinine synthesis, increased tubular creatinine secretion, and decreased skeletal muscle mass. The Consensus Conference of the International Ascites Club established that elevation of serum creatinine occurs only after the onset of sodium retention and impaired free water excretion.
Hyponatremia appears to be a good surrogate marker for refractory ascites and should be taken into consideration when attempting to interpret prognostic factors in patients with cirrhosis. However, it should be noted that the severity of ascites and hyponatremia might be subject to manipulation based on the extent of treatment with diuretics and the restriction of salt and water intake.
In evaluating hyponatremia in cirrhotic patients, all conditions that could possibly cause low serum sodium concentration should be actively sought and treated:
* Solute deficit from extrarenal (diarrhea, vomiting, enteric and biliary fistulas, paracentesis, or overzealous use of lactulose) or renal losses (aggressive diuretic therapy).
* Ensuing volume depletion (as in therapeutic paracentesis without colloid supplementation causing stimulation of thirst and excessive free water intake).
* Presence of osmotically active solute (as in decompensated diabetes mellitus or treatment with mannitol to reduce brain edema).
* Water oversupply (in the setting of hypotonic fluids infusion).
* The diagnosis is purely clinical, based on a thorough history, careful physical examination, determination of fluid status, and measurement of serum and urine electrolytes and osmolality. A sudden gain in body weight points toward a water surplus condition, whereas a solute deficit should be considered if body weight is unchanged.
Although fluid restriction is the most commonly used strategy for treating hyponatremia in cirrhosis, limiting fluid consumption to an arbitrary amount is rarely effective. The ineffectiveness of this approach is expected, since many physicians don't account for insensible water loss, the portion of free water in consumed fluids, and, most importantly, the amount of solute in urine water when prescribing fluid restriction.
Using the urine/plasma electrolytes ratio (urine sodium + potassium/plasma sodium) is a simple bedside approach to overcome this issue and predict the efficacy of fluid restriction.
For example, when the ratio is less than 0.5, significant amounts of electrolyte-free water are excreted in urine, and fluid restriction should be very effective. In contrast, no electrolyte-free water is excreted when the ratio is equal to or more than 1.0, and fluid restriction will not be effective at any level.
Care should be exercised not to overcorrect serum sodium too rapidly in order to avoid the risks of osmotic demyelination syndrome, quadriplegia, coma, and death. Hypertonic saline is only indicated in symptomatic hyponatremia or within hours of liver transplantation to prevent the likelihood of an emergent rapid correction in the operating room.
As a vast majority of cirrhotic patients who present with hyponatremia also have ascites, there definitely is a role for sodium restriction and effective diuretic therapy in this patient population.
The combination of oral spironolactone and furosemide at a daily dose of 100 mg and 40 mg, respectively, is a good place to start. Care should be taken to obtain urine osmolarity in a random sample at the time of admission, and the dose of diuretics should be increased at 100-mg and 40-mg increments of spironolactone and furosemide, respectively, to a maximum of 400 mg and 160 mg if there is no decrease in body weight or a decline in urine osmolarity after two or three days.
Additionally, urinary excretion of sodium in excess of the dietary intake is a good predictor of weight loss in afebrile patients who have no diarrhea, as urine is the predominant route of sodium loss in the patient group.
If the urinary sodium concentration is less than 10 mmol/L and the urine volume is less than 1 L/day on diuretic treatment, it is recommended that the diuretic dose must be increased. The general thinking that urinary sodium measurements are of no clinical relevance in patients on diuretic treatment is a misconception.
A negative sodium balance (urinary sodium excretion in excess of dietary intake) with a weight loss of 0.5 kg/day in a patient with no pulmonary edema is a reasonable goal in the medical treatment of ascites in patients with cirrhosis.
Vasopressin Receptor Antagonists
Although vasopressin is a potent vasoconstrictor and pressor agent, it produces a dose-dependent biphasic response: vasoconstriction at low concentrations and vasodilation at high concentrations.
The vasodilator response of vasopressin is blocked by selective vasopressin V2-receptor antagonism, and patients with congenital nephrogenic diabetes insipidus who have a V2 receptor gene defect demonstrate a vasoconstrictor response to vasopressin and resistance to the effects of the selective V2-receptor agonist desmopressin. Thus, the vasodilatory response of vasopressin appears to be mediated by the V2 receptor.
When serum sodium correction is difficult to achieve using standard measures, selective vasopressin V2-receptor antagonists or the inhibition of vasopressin secretion with κ-opioid receptor agonists may be successful.
Though the latter approach induced aquaresis in animal models and in patients with cirrhosis, the aquaretic effect was not sustained in patients.
Blockade of the vasopressin V2 receptor, on the other hand, will induce an effective aquaresis in addition to blocking the potential for V2-mediated vasodilation. Nonpeptide V2 receptor antagonists have no agonist effects in humans, are orally bioavailable, and have a longer plasma half-life than endogenous vasopressin.
The first selective nonpeptide vasopressin V2 receptor antagonist was OPC-31260. Subsequently, other such antagonists with greater receptor affinity have been formulated. These agents promote significant aquaresis and an increase in serum sodium level. However, particular caution must be exercised to avoid rapid correction and severe dehydration, which may predispose the patient to renal insufficiency, encephalopathy, and osmotic demyelination syndrome.
Although fluid restriction and diuretic therapy have been the standard of care in this patient population for many years, the use of more recently introduced selective vasopressin V2 receptor antagonists has increased effectiveness, as it directly targets the pathophysiologic mechanism of hyponatremia in this group.
© 2012 Lippincott Williams & Wilkins, Inc.