Portal de Periódicos da CAPES

Overcorrection of hyponatremia is a medical emergency

2009; Elsevier BV; Volume: 76; Issue: 6 Linguagem: Inglês

10.1038/ki.2009.251

1523-1755

Richard H. Sterns, John K. Hix,

Renal function and acid-base balance

Overcorrection of hyponatremia is a medical emergency. Excessive correction usually results from the unexpected emergence of a water diuresis after resolution of the cause of water retention. The concurrent administration of desmopressin and 5% dextrose in water can be given to cautiously re-lower the serum sodium concentration when therapeutic limits have been exceeded. Nephrologists should be equally aggressive in correcting hyponatremia and in un-correcting it when their patients get too much of a good thing. Overcorrection of hyponatremia is a medical emergency. Excessive correction usually results from the unexpected emergence of a water diuresis after resolution of the cause of water retention. The concurrent administration of desmopressin and 5% dextrose in water can be given to cautiously re-lower the serum sodium concentration when therapeutic limits have been exceeded. Nephrologists should be equally aggressive in correcting hyponatremia and in un-correcting it when their patients get too much of a good thing. For a quarter century, nephrologists treating hyponatremia have struggled with a dilemma. On the one hand, some authors have written that symptomatic hypo-natremia has an unacceptably high mortality rate unless it is treated aggressively. On the other hand, some have linked overaggressive therapy to a crippling or fatal neurological complication known as osmotic demyelination syndrome, or central pontine myelinolysis (CPM). In the early days of this controversy, Kidney International devoted a Nephrology Forum to this conundrum, characterizing it as 'damned if we do and damned if we don't'.1.Berl T. Treating hyponatremia: damned if we do and damned if we don't.Kidney Int. 1990; 37: 1006-1018Abstract Full Text PDF PubMed Scopus (146) Google Scholar Three years ago, a Commentary in this journal revisited the controversy and took a strong stand, asserting that 'one must conclude that we are damned if we don't, or at least the patient is'.2.Kokko J.P. Symptomatic hyponatremia with hypoxia is a medical emergency.Kidney Int. 2006; 69: 1291-1293Abstract Full Text Full Text PDF PubMed Scopus (19) Google Scholar The article maintained that because of the high mortality risk, symptomatic hyponatremic patients 'should receive rapid correction of their sodium concentration to safe levels.' An increase of 8–10 mequiv. per liter in sodium concentration in the first 4–6 h was advised, followed by correction to about 120 mequiv. per liter in the next 24 h, with correction of the remaining deficit 'at a rate that improves serum concentration each 24 h by 50% of the desired final sodium concentration.' Emphasizing the additive risks of hypoxia in symptomatic hyponatremia, the Commentary recommended intensive care and endo-tracheal intubation if the arterial partial pressure of oxygen (PO2) falls below 70 mm Hg. While acknowledging CPM as a theoretical possibility, the writer concluded, 'our therapeutic approaches should not be paralyzed by fear that CPM will develop.' In support of these conclusions, the Commentary cited experimental findings suggesting that brain damage in hyponatremia is the result of hypoxia rather than overzealous therapy.3.Ayus J.C. Armstrong D. Arieff A.I. Hyponatremia with hypoxia: effects on brain adaptation, perfusion, and histology in rodents.Kidney Int. 2006; 69: 1319-1325Abstract Full Text Full Text PDF PubMed Scopus (64) Google Scholar In this issue of the journal, Gankam Kengne and colleagues4.Gankam Kengne F. Soupart A. Pochet R. et al.Re-induction of hyponatremia following rapid overcorrection of hyponatremia reduces mortality in rats.Kidney Int. 2009; 76: 614-621Abstract Full Text Full Text PDF PubMed Scopus (74) Google Scholar present experimental findings that are difficult to reconcile with the notion that the histologic lesions of osmotic demyelination syndrome result from hypoxia rather than excessive correction of hyponatremia. In the chronically hyponatremic rat, almost uniformly fatal brain lesions can be consistently induced by an increase of the serum sodium by more than 20 mequiv. per liter per day. Re-lowering the serum sodium concentration reverses the breakdown of the blood–brain barrier caused by rapid correction and prevents brain damage and mortality. Interestingly, the investigators find that opening of the blood–brain barrier, one of the proposed mechanisms for demyelination, may be an epiphenomenon, because glucocorticoids are equally effective in protecting the blood–brain barrier but relatively ineffective in preventing brain lesions or mortality. Clinical experience validates these experimental findings. Early manifestations of osmotic demyelination following excessive correction of hyponatremia have been reversed in individual case reports by re-lowering of the serum sodium concentration.5.Perianayagam A. Sterns R.H. Silver S.M. et al.DDAVP is effective in preventing and reversing inadvertent overcorrection of hyponatremia.Clin J Am Soc Nephrol. 2008; 3: 331-336Crossref PubMed Scopus (118) Google Scholar,6.Sterns R.H. Nigwekar S.U. Hix J.K. The treatment of hyponatremia.Semin Nephrol. 2009; 29: 282-299Abstract Full Text Full Text PDF PubMed Scopus (195) Google Scholar We routinely use desmopressin to prevent the serum sodium from increasing more than it should, and we have frequently administered 5% dextrose in water with desmopressin to re-lower the serum sodium after inadvertent overcorrection of symptomatic hyponatremia; our published and confirmatory unpublished experiences with this maneuver have shown that it is well tolerated.5.Perianayagam A. Sterns R.H. Silver S.M. et al.DDAVP is effective in preventing and reversing inadvertent overcorrection of hyponatremia.Clin J Am Soc Nephrol. 2008; 3: 331-336Crossref PubMed Scopus (118) Google Scholar If symptomatic hyponatremia carries a high risk of seizures, respiratory arrests, neurogenic pulmonary edema, and hypoxic exacerbation of cerebral edema leading to fatal herniation, how is it possible that patients do well when their serum sodium concentration is re-lowered into the 'unsafe' hyponatremic range? Perhaps we should reexamine the evidence. To be sure, even mild hyponatremia is associated with increased hospital mortality.7.Whelan B. Bennett K. O'Riordan D. Silke B. Serum sodium as a risk factor for in-hospital mortality in acute unselected general medical patients.QJM. 2009; 102: 175-182Crossref PubMed Scopus (69) Google Scholar We really do not understand why this is true, but fatal cases are rarely associated with clinical evidence of cerebral edema or, for that matter, osmotic demyelination. Studies suggesting that respiratory arrest culminating in brain damage or death is a common complication of symptomatic hyponatremia consist of cases 'referred' to the authors after the patients had already died or suffered irreversible brain damage.8.Ayus J.C. Arieff A.I. Chronic hyponatremic encephalopathy in postmenopausal women: association of therapies with morbidity and mortality.JAMA. 1999; 281: 2299-2304Crossref PubMed Scopus (171) Google Scholar Case series free of selection bias suggest that osmotic demyelination due to overcorrection of hyponatremia is more common than respiratory arrest from inadequately treated hyponatremia.6.Sterns R.H. Nigwekar S.U. Hix J.K. The treatment of hyponatremia.Semin Nephrol. 2009; 29: 282-299Abstract Full Text Full Text PDF PubMed Scopus (195) Google Scholar What is the evidence that hypoxia rather than overcorrection of hyponatremia causes demyelinating brain lesions? Ayus and colleagues demonstrated that severe hypoxia exacerbates brain edema in hyponatremic animals.3.Ayus J.C. Armstrong D. Arieff A.I. Hyponatremia with hypoxia: effects on brain adaptation, perfusion, and histology in rodents.Kidney Int. 2006; 69: 1319-1325Abstract Full Text Full Text PDF PubMed Scopus (64) Google Scholar If this occurs in humans, hypoxia from neurogenic pulmonary edema might plausibly cause a vicious cycle ending in death. However, these investigators were not able to cause demyelinating brain lesions with hypoxia.3.Ayus J.C. Armstrong D. Arieff A.I. Hyponatremia with hypoxia: effects on brain adaptation, perfusion, and histology in rodents.Kidney Int. 2006; 69: 1319-1325Abstract Full Text Full Text PDF PubMed Scopus (64) Google Scholar Exposure to severe hypoxia was uniformly and immediately fatal in hyponatremic animals. Exposure of normonatremic rats to prolonged severe hypoxia caused brain lesions in 23% of animals that were similar in distribution to those caused by rapid correction of chronic hyponatremia. Although the distribution of lesions was similar, their histologic characteristics were not. Unlike the lesions of osmotic demyelination, which spare neurons, hypoxia-induced lesions were associated with neuronal necrosis. Why intubate a hyponatremic patient because of a PO2 of 70 mm Hg (a value associated with an oxygen saturation above 95%)? In the study by Ayus et al., five mechanically ventilated hyponatremic rats with hypoxia of this severity had a significantly lower 'cerebral perfusion index' than controls or hyponatremic animals without hypoxia.3.Ayus J.C. Armstrong D. Arieff A.I. Hyponatremia with hypoxia: effects on brain adaptation, perfusion, and histology in rodents.Kidney Int. 2006; 69: 1319-1325Abstract Full Text Full Text PDF PubMed Scopus (64) Google Scholar The finding seems insufficient to merit a clinical mandate: the study was small; it provided no information on how (or whether) blood pressure and PCO2, keys to cerebral perfusion, were maintained at equivalent levels across groups; and the investigators used a technique whose validity has been challenged to estimate cerebral blood flow.9.Muller T.B. Haraldseth O. Jones R.A. et al.Combined perfusion and diffusion-weighted magnetic resonance imaging in a rat model of reversible middle cerebral artery occlusion.Stroke. 1995; 26 (discussion 457–458): 451-457Crossref PubMed Scopus (81) Google Scholar Certainly, mechanical ventilation can result in alterations in blood pressure and PCO2, factors that do influence cerebral blood flow (for better or worse).10.Noth U. Kotajima F. Deichmann R. et al.Mapping of the cerebral vascular response to hypoxia and hypercapnia using quantitative perfusion MRI at 3 T.NMR Biomed. 2008; 21: 464-472Crossref PubMed Scopus (44) Google Scholar Furthermore, a high rate of cerebral perfusion is not necessarily desirable. Increasing the serum sodium concentration from 111 to 128 mequiv. per liter over 3 h (a rate known to cause osmotic demyelination in the rat) dramatically increases cerebral blood flow without affecting PO2, PCO2, or blood pressure; the surge in blood flow is a prelude to breakdown of the blood–brain barrier.11.Adler S. Verbalis J.G. Williams D. Effect of rapid correction of hyponatremia on the blood-brain barrier of rats.Brain Res. 1995; 679: 135-143Crossref PubMed Scopus (43) Google Scholar Directing our attention to cerebral perfusion in mechanically ventilated animals obscures the most important clinical question: what is the consequence of mild to moderate hypoxia in spontaneously breathing hyponatremic patients? Even in patients with extremely low serum sodium concentrations, a linkage between hypoxia and adverse neurological outcomes has not been established. A recent study of all patients with serum sodium concentrations of 110 mequiv. per liter or less at our medical center identified only a handful with oxygen saturations below 90%; these subjects, like those without hypoxia, recovered uneventfully after treatment designed to limit correction to less than 12 mequiv. per liter in 24 h and 18 mequiv. per liter in 48 h.12.Vootukuru S. Kadam S. Kouides R. Sterns R. Outcomes of severe symptomatic hyponatremia and effect of hypoxia.J Am Soc Nephrol. 2007; 18 (abstr): 790AGoogle Scholar Our recent findings confirm the favorable outcomes in severe hyponatremia documented in an earlier study from the same center.13.Sterns R.H. Severe symptomatic hyponatremia: treatment and outcome. A study of 64 cases.Ann Intern Med. 1987; 107: 656-664Crossref PubMed Scopus (246) Google Scholar When it was fashionable to bring serum sodium concentrations of 105 mequiv. per liter or less into a 'safe range' above 120 mequiv. per liter (a goal that requires an increase of more than 15 mequiv. per liter within 24 h), half the patients suffered temporary or permanent post-therapeutic neurological complications.13.Sterns R.H. Severe symptomatic hyponatremia: treatment and outcome. A study of 64 cases.Ann Intern Med. 1987; 107: 656-664Crossref PubMed Scopus (246) Google Scholar We agree that fear of osmotic demyelination does not justify therapeutic inaction, but fear of complications from hyponatremic encephalopathy does not justify therapeutic excess. Regardless of how often it occurs, the fact that some patients with acute hyponatremia succumb to cerebral edema makes this a medical emergency. Fortunately, it now appears that an increase in serum sodium concentration by 4–6 mequiv. per liter is sufficient to rescue a patient from impending herniation.6.Sterns R.H. Nigwekar S.U. Hix J.K. The treatment of hyponatremia.Semin Nephrol. 2009; 29: 282-299Abstract Full Text Full Text PDF PubMed Scopus (195) Google Scholar Adopting a strategy of prompt but limited correction of hyponatremia, we believe, should prevent the most devastating consequences of hyponatremia while minimizing the chance of osmotic demyelination in any patient, particularly those with chronic hyponatremia. Overcorrection of hyponatremia should be viewed as a medical emergency. In most cases, excessive correction results from the unexpected emergence of a water diuresis after resolution of the cause of water retention (medications, hypovolemia, transient syndrome of inappropriate antidiuretic hormone, and so on).5.Perianayagam A. Sterns R.H. Silver S.M. et al.DDAVP is effective in preventing and reversing inadvertent overcorrection of hyponatremia.Clin J Am Soc Nephrol. 2008; 3: 331-336Crossref PubMed Scopus (118) Google Scholar,14.Mohmand H.K. Issa D. Ahmad Z. et al.Hypertonic saline for hyponatremia: risk of inadvertent overcorrection.Clin J Am Soc Nephrol. 2007; 2: 1110-1117Crossref PubMed Scopus (121) Google Scholar The surest way to avoid inadvertent overcorrection in a patient undergoing a water diuresis is to administer desmopressin (but only to patients who can be relied on to curtail their water intake, and probably not to patients with hyperacute hyponatremia developing over a few hours). If the serum sodium has already increased by more than 10–12 mequiv. per liter in 24 h or 18 mequiv. per liter in 48 h, the concurrent cautious administration of 5% dextrose in water in individual doses of 6 ml per kilogram body weight, given over an hour or two (calculated to reduce the serum sodium by 2 mequiv. per liter), monitored by serum sodium determinations after each dose (and before another is given), can be used to return the serum sodium concentration to a level below these limits of overcorrection. Subsequently, a state of controlled antidiuresis is maintained by the administration of desmopressin every 6–8 h, a slow increase of the serum sodium concentration as desired with water restriction plus hypertonic saline or salt tablets, or the periodic delaying of a dose of desmopressin to permit a brief water diuresis (Figure 1). Both untreated and overtreated hyponatremia can result in disastrous neurological complications. We need not choose between these evils. Nephrologists should be equally aggressive in correcting hyponatremia and in un-correcting it when their patients get too much of a good thing. RHS has been paid honoraria by Astellas and Otsuka.