Correspondence
Hyponatremia
N Engl J Med 2000; 343:886-888September 21, 2000
- Article
To the Editor:
In commenting on hyponatremia secondary to the absorption of sodium-free irrigant solutions from an operative site, Adrogué and Madias (May 25 issue)1 refer only to men who are undergoing prostate surgery. However, the syndrome may occur in women as a complication of endometrial ablation.2 Adrogué and Madias state that the cause of the symptoms remains unclear. In fact, the pathogenesis of the major clinical manifestations has recently been described: the respiratory depression and brain edema are due either to ammonia intoxication (when glycine is the irrigating solution) or to hyponatremia with cerebral edema.2 It is extremely important to differentiate between these entities, since the treatments are different. In patients with hyponatremic encephalopathy, hypertonic sodium chloride is efficacious, whereas in those with ammonia-induced toxicity, the preferred therapy is respiratory support, with hemodialysis if there is renal impairment.2
Adrogué and Madias state that “water restriction will ameliorate all forms of hyponatremia.” In fact, although water restriction has been recommended as a therapy for hyponatremia, the procedure has never been evaluated and its efficacy is unproved, particularly in symptomatic patients. In a recent prospective evaluation, patients with chronic symptomatic hyponatremia were treated with water restriction; all had respiratory insufficiency and permanent brain damage.3
3 ReferencesAllen I. Arieff, M.D.
University of California, San Francisco, San Francisco, CA 94143J. Carlos Ayus, M.D.
Baylor College of Medicine, Houston, TX 770301
Adrogue HJ ,Madias NE . Hyponatremia. N Engl J Med 2000;342:1581-1589
Full Text | Web of Science | Medline2
Ayus JC ,Arieff AI . Glycine-induced hypo-osmolar hyponatremia. Arch Intern Med 1997;157:223-226
CrossRef | Web of Science | Medline3
Ayus JC ,Arieff AI . Chronic hyponatremic encephalopathy in postmenopausal women: association of therapies with morbidity and mortality. JAMA 1999;281:2299-2304
CrossRef | Web of Science | Medline
To the Editor:
I am concerned that in Table 1 of the review by Adrogué and Madias, the hyponatremia associated with central nervous system disorders (including hemorrhage) is classified as being due to the syndrome of inappropriate secretion of antidiuretic hormone. Hyponatremia develops in up to 30 percent of patients with subarachnoid hemorrhage, and in some patients it may indeed be due to the syndrome of inappropriate secretion of antidiuretic hormone. Many patients, however, have the fundamentally different condition of cerebral salt wasting,1 in which excessive renal sodium loss produces hyponatremia with volume depletion. The pathogenesis of this condition is poorly understood, but it probably includes increases in concentrations of brain natriuretic peptide and other natriuretic factors and the direct effects of disturbed renal autonomic innervation.2,3 Although formal measurements of plasma volume are difficult to perform in routine clinical practice, cerebral salt wasting may often be distinguished from the syndrome of inappropriate secretion of antidiuretic hormone by the presence of clinical signs of dehydration, low central venous pressure, an increased hematocrit and serum protein concentration, normal or increased serum osmolality, and markedly increased urinary sodium concentrations. Cerebral salt wasting is treated with sodium and volume replacement, sometimes with the addition of fludrocortisone.4 The use of fluid restriction in such patients, who frequently have cerebral arterial spasm and disturbed microvascular regulation, courts disaster in the form of cerebral infarction.
4 ReferencesAndrew J. Martin, F.R.C.S.
King's College Hospital, London SE5 9RS, United Kingdom1
Peters JP ,Welt LG ,Sims EAH ,Orloff J ,Needham J . A salt-wasting syndrome associated with cerebral disease. Trans Assoc Am Physicians 1950;63:57-64
Medline2
Harrigan MR . Cerebral salt wasting syndrome: a review. Neurosurgery 1996;38:152-160
CrossRef | Web of Science | Medline3
Berendes E ,Walter M ,Cullen P , et al. Secretion of brain natriuretic peptide in patients with aneurysmal subarachnoid haemorrhage. Lancet 1997;349:245-249
CrossRef | Web of Science | Medline4
Mori T ,Katayama Y ,Kawamata T ,Hirayama T . Improved efficiency of hypervolemic therapy with inhibition of natriuresis by fludrocortisone in patients with aneurysmal subarachnoid hemorrhage. J Neurosurg 1999;91:947-952
CrossRef | Web of Science | Medline
To the Editor:
It is true, as Adrogué and Madias state, that direct measurement of the serum sodium concentration with the ion-selective electrode eliminates the artifactual finding of pseudohyponatremia, but direct ion-specific electrodes, which actually assess activity rather than concentration, are not the most common tools used to measure sodium in clinical laboratories today. For a variety of reasons (related to mainly cost and maintenance), instruments designed to perform multiple tests with a high sample throughput often measure sodium by indirect ion-specific electrodes, which determine the sodium concentration rather than its activity. The most recent report of the College of American Pathologists1 indicates that the majority of clinical laboratories report their sodium-survey results performed with ion-specific electrodes after diluting the samples, making the analysis subject to pseudohyponatremia.
Pseudohyponatremia occurs when the nonaqueous volume of plasma is markedly increased, usually as a result of severe hyperlipidemia or a high concentration of a paraprotein, either of which leads to low concentrations of sodium in the total plasma.2,3 The indirect ion-specific electrode and flame photometry both include dilution of samples before the analysis and thus measure the concentration in the total plasma volume. The activity of sodium in the aqueous fraction of plasma, as measured by the direct use of ion-specific electrodes, is not affected by high triglyceride or protein concentrations. Pseudohyponatremia has not been eliminated, and clinicians still need to be aware of which sodium methods are being used by the clinical laboratory if their patients have high triglyceride or protein values.
3 ReferencesDavid E. Bruns, M.D.
University of Virginia Medical School, Charlottesville, VA 22908Jack H. Ladenson, Ph.D.
Mitchell G. Scott, Ph.D.
Washington University School of Medicine, St. Louis, MO 631101
Participant summary report: surveys 2000. Northfield, Ill.: College of American Pathologists, 2000.
2
Apple FS ,Koch DD ,Grave SS ,Ladenson JH . Relationship between direct-potentiometric and flame-photometric measurement of sodium in blood. Clin Chem 1982;28:1931-1935
Web of Science | Medline3
Maas AHJ ,Siggaard-Andersen O ,Weisberg HF ,Zijlstra WG . Ion-selective electrodes for sodium and potassium: a new problem of what is measured and what should be reported. Clin Chem 1985;31:482-485
Web of Science | Medline
To the Editor:
Adrogué and Madias state that “an increase of 100 mg per deciliter (5.6 mmol per liter) in the serum glucose concentration decreases serum sodium by approximately 1.7 mmol per liter.” This statement is based on the correction factor described by Katz,1 which is used to correct hyponatremia caused by hyperglycemia.
In a recent study, Hillier et al.2 infused somatostatin to block endogenous insulin secretion in six healthy subjects, and plasma glucose was increased to more than 600 mg per deciliter within one hour by an infusion of 20 percent dextrose. The dextrose infusion was stopped, and insulin was given until the plasma glucose concentration decreased to 140 mg per deciliter; plasma glucose and serum sodium concentrations were measured every 10 minutes. Overall, the mean decrease in serum sodium concentrations averaged 2.4 mmol per liter for every increase in the glucose concentration of 100 mg per deciliter. This value was significantly greater than the commonly used value of 1.6 mmol per liter (P=0.02). The difference becomes even more relevant in the presence of higher concentrations of plasma glucose (>400 mg per deciliter).
Thus, we support the proposal of Hillier et al.2 that a decrease of 2.4 mmol of sodium per liter for every increase of 100 mg per deciliter in the glucose concentration is a more practical and accurate value than the accepted value of 1.6 mmol per liter.3
3 ReferencesAjit Singh Kashyap, M.D.
Surekha Kashyap, M.D.
Armed Forces Medical College, Pune 411 040, India1
Katz MA . Hyperglycemia-induced hyponatremia: calculation of expected serum sodium depression. N Engl J Med 1973;289:843-844
Full Text | Web of Science | Medline2
Hillier TA ,Abbott RD ,Barrett EJ . Hyponatremia: evaluating the correction factor for hyperglycemia. Am J Med 1999;106:399-403
CrossRef | Web of Science | Medline3
Kashyap AS . Hyperglycemia-induced hyponatremia: is it time to correct the correction factor? Arch Intern Med 1999;159:2745-2746
CrossRef | Web of Science | Medline
To the Editor:
In their review of the management of hyponatremia, Adrogué and Madias recommend the use of hypertonic (3 percent) saline for certain patients with symptomatic hypotonic hyponatremia and the use of hypotonic (0.45 percent) saline in a patient with hypovolemic hyponatremia. We believe that symptomatic hyponatremia does not usually require rapid correction and that hypertonic or hypotonic saline is rarely needed in the management of this disorder.
Recent editorials, by Knochel in an American journal1 and by Lane and Allen in a British journal,2 have also advised that hypertonic saline be used for the management of severe hyponatremia. Lane and Allen recommend raising the serum sodium concentration by 1 to 2 mmol per liter per hour until symptoms resolve. Knochel states that fluid restriction should be abandoned in the treatment of symptomatic hyponatremia because the outcome in one study was universally tragic. In our experience, patients who are hospitalized with hyponatremia, whether symptomatic or not, often have a good response over a period of several days to moderate restriction of fluids when the cause is dilutional. The evidence against the gradual rate of correction (approximately 0.5 mmol per liter per hour) associated with fluid restriction is not sound enough to abandon this well-established practice. The treatment groups in the American study1 referred to by Knochel were poorly controlled, and the outcome in patients treated with water deprivation may have been due to the differences in the selection of patients or underlying abnormalities. Neither the initial plasma sodium value nor the rate of correction was correlated with the outcome. Since most cases of severe hyponatremia are dilutional, it seems illogical to treat this condition with exogenous sodium when there is no sodium deficiency, rather than to allow controlled loss of the water excess.
2 ReferencesCharles van Heyningen, F.R.C.Path.
Geoffrey V. Gill, F.R.C.P.
Ian D. Watson, F.R.C.Path.
University Hospital Aintree, Liverpool L9 7AL, United Kingdom1
Knochel JP . Hypoxia is the cause of brain damage in hyponatremia. JAMA 1999;281:2342-2343
CrossRef | Web of Science | Medline2
Lane N ,Allen K . Hyponatremia after orthopaedic surgery. BMJ 1999;318:1363-1364
CrossRef | Web of Science | Medline
To the Editor:
Adrogué and Madias present the management of hyponatremia as if it were a sodium-deficit state, which very often it is not. Just as the goal of treating hypernatremia is the repletion of water deficits, the ultimate goal in the treatment of hyponatremia should be the loss of the retained water. Furthermore, in their approach to the correction, Adrogué and Madias entirely ignore the excretion of solutes and water. The formulas presented in Table 2 of their article will, in fact, predict the change in serum sodium with the administered solution, but only if there is no concomitant urine output. The composition of the urine can significantly affect the rate of correction, particularly when furosemide is administered, which increases electrolyte-free water losses. In the case of the patient with postoperative hyponatremia, the authors calculate that the administration of 180 ml of 3 percent sodium chloride will increase the serum sodium concentration by 3 mol per liter in a period of three hours. This value is correct if there is no urine output. However, if after the administration of furosemide, the patient excretes 2 liters of urine containing 60 mmol of sodium per liter and 20 mmol of potassium per liter, the rate of increase in the serum sodium concentration — 6 mmol per liter over the three-hour period — will be twice as large as the rate predicted on the basis of the formula of Adrogué and Madias. Such a rate of urine and solute excretion is very much in the range expected after the administration of furosemide in a euvolemic patient with a normal glomerular filtration rate who is receiving a saline infusion. The difference in the correction rate may not be of importance in a patient with acute postoperative hyponatremia that requires rapid correction. However, it is potentially dangerous in a patient whose hyponatremia is of longer duration. Thus, urinary electrolytes should be assessed in patients with symptomatic hyponatremia, because ultimately the serum sodium concentration will be corrected only in patients who excrete electrolyte-free water. The authors acknowledge this point in their description of common errors in management when they allude to the excretion of saline and retention of water in patients with the syndrome of inappropriate secretion of antidiuretic hormone.
Tomas Berl, M.D.
University of Colorado School of Medicine, Denver, CO 80262Author/Editor Response
The authors reply:
To the Editor: Contrary to the claim of Arieff and Ayus, in Table 1 of our review we list sodium-free irrigant solutions used in hysteroscopy, laparoscopy, or transurethral resection of the prostate as causes of hyponatremia. We also believe that the pathogenesis of the symptoms associated with the hyponatremia of massive absorption of sodium-free irrigant solutions remains unclear. In response to the final criticism of Arieff and Ayus, we would like to point out that our complete sentence read as follows: “Although water restriction will ameliorate all forms of hyponatremia, it is not the optimal therapy in all cases.” We then went on to explain why.
Table 1 of our review does not classify the hyponatremia associated with central nervous system disorders as being due exclusively to the syndrome of inappropriate secretion of antidiuretic hormone. Rather, it lists central nervous system disorders as one of the leading causes of this syndrome. We concur with Dr. Martin that hyponatremia develops in some patients with central nervous system disorders in association with cerebral salt wasting, a poorly understood condition of renal sodium loss, hypovolemia, and water retention. Patients with this condition do not meet the diagnostic criteria for the syndrome of inappropriate secretion of antidiuretic hormone.
We agree with Bruns et al. that clinicians should inquire whether a clinical laboratory uses methods other than direct ion-specific electrode to measure serum sodium. If this is the case, practitioners should still consider the possibility of pseudohyponatremia in the appropriate clinical setting.
Kashyap and Kashyap point out recent experimental observations of a decrease in serum sodium in response to hyperglycemia that was considerably larger than the standard correction factor.1 In the study they cite, approximately 800 ml of 20 percent dextrose in 0.45 percent saline was infused over a period of 60 minutes. The authors acknowledged that dilution of the serum sodium concentration by the infusate could have led to an overestimation of the degree of translocational hyponatremia. Additional experiments will be required.
In response to van Heyningen et al.: we advocate a slow-paced approach to persistent asymptomatic hyponatremia and rapid but controlled correction of symptomatic hyponatremia. Symptomatic hyponatremia reflects cerebral edema and therefore requires the prudent use of hypertonic saline, as we explained in our review. We know of no evidence that moderate restriction of fluid results in the gradual and safe correction of symptomatic hyponatremia at a rate of approximately 0.5 mmol per liter per hour.
We are puzzled by Dr. Berl's assertion that in one of our case reports we treat dilutional hyponatremia as a sodium-deficit state rather than a water-retention state. Our formulas, like conventional formulas, consider the patient as a closed system, with no gain or loss of water and electrolytes other than the administered infusate during the infusion. Because renal losses and extrarenal losses (e.g., those resulting from nasogastric suction) can be substantial in the course of treatment, frequent monitoring of the serum sodium concentration is required in order to adjust the fluid prescription. However, we consider an assessment of urinary electrolytes to be impractical and unnecessary during the administration of a hypertonic sodium solution.
1 ReferencesHoracio J. Adrogué, M.D.
Veterans Affairs Medical Center, Houston, TX 77030Nicolaos E. Madias, M.D.
New England Medical Center, Boston, MA 021111
Hillier TA ,Abbott RD ,Barrett EJ . Hyponatremia: evaluating the correction factor for hyperglycemia. Am J Med 1999;106:399-403
CrossRef | Web of Science | Medline
- Citing Articles (3)
Citing Articles
1
Akwasi Afriyie Boateng, Krishnan Sriram, Michael M. Meguid, Martin Crook. (2010) Refeeding syndrome: Treatment considerations based on collective analysis of literature case reports. Nutrition 26:2, 156-167
CrossRef2
Michael L. Moritz, Juan Carlos Ayus. 2009. Diabetes Insipidus and SIADH. , 261-286.
CrossRef3
Scott A Rivkees. (2008) Differentiating appropriate antidiuretic hormone secretion, inappropriate antidiuretic hormone secretion and cerebral salt wasting: the common, uncommon, and misnamed. Current Opinion in Pediatrics 20:4, 448-452
CrossRef
