Q J Med 2003; 96: 421-426
© 2003 Association of Physicians
Solute loss plays a major role in polydipsia-related hyponatraemia of both water drinkers and beer drinkers
From the 1Department of Internal Medicine, Bracops Hospital, and 2Research Unit for the Study of Hydromineral Metabolism, Erasme University Hospital, Free University of Brussels, Brussels, Belgium
Received 13 September 2002 and in revised form 10 February 2003
 |
Summary |
|---|
 Top Summary IntroductionMethodsResultsDiscussionReferences |
|---|
Background: Polydipsia-related hyponatraemia is generally considered an acute dilutional state.
Aim: To determine whether solute loss plays a role in the pathogenesis of polydipsia-related hyponatraemia.
Design: Prospective uncontrolled study
Methods: We studied routine biochemical volume-related parameters before and after 2 l isotonic saline infusion over 24 h, in 10 consecutive hyponatraemic polydipsia patients (mean age 55 ± 11 years; 6 beer drinkers and 4 compulsive water drinkers) with initial urinary osmolality <220 mosm/kg H2O. In five of these patients, we measured balance data over 24 h.
Results: Mean initial plasma protein concentration in the 10 studied polydipsia patients was 7 ± 0.7 g/dl, unexpectedly high for an acute dilutional state. Mean plasma sodium concentration increased from 126 ± 5 mmol/l before saline, to 135 ± 5 mmol/l after infusion of 2 l isotonic saline (p < 0.01). Balance data in five polydipsia patients showed a mean decrease of 1.6 kg of their initial body weight and a mean salt retention of 406 mosm.
Discussion: Polydipsia-related hyponatraemia is a mixed disorder, in which about half of sodium decrease is due to solute loss. This explains the apparent paradox of a normal plasma protein concentration, despite the increase in body weight due to water intoxication.
|
Introduction |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Hyponatraemia related to polydipsia is generally considered to be an acute dilutional state in which water intake overwhelms renal excretory capacity.1–2 Solute intake affects the capacity of the kidney to excrete a water load, and low solute intake contributes to susceptibility to hyponatraemia in beer drinkers.3–6
Analysing the biochemical data of 10 hyponatraemic polydipsia patients at admission, we were surprised to observe a normal mean plasma protein concentration, despite a reduction of mean plasma sodium of >10%. This is not what one would expect in an acute dilutional state.7
We know of only one polydipsia case report4 that included balance data. It concerned a severe hyponatraemia in a polydipsia patient who was also taking diuretics. Significant saline retention was reported during saline infusion. The changes in his body weight, as a measure of the degree of water intoxication, were not given. We prospectively measured balance data in five polydipsia patients without diuretic treatment, to investigate whether significant solute loss is involved in the pathogenesis of the hyponatraemia of both compulsive water drinkers (CWD) and beer drinkers (BE).
|
Methods |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
We studied routine biochemical volume-related parameters in 10 consecutive hyponatraemic polydipsia patients, (55 ± 11 years; 6 BE and 4 CWD) admitted to the ward of a 200-bed general hospital. These hyponatraemic polydipsia patients were hospitalized for symptoms such as drowsiness, weakness, confusion and falling. In some patients, seizures could not be completely excluded, but if seizures occurred, it was a few hours before admission. The range of plasma sodium was 115–130 mmol/l. The history of these patients included a context of fluid intake exceeding 4 l a day. None of them took diuretics. Biochemical parameters, including a urinary spot analysis, were evaluated at admission (t0), and after infusion of 2 l isotonic saline over 24 h. The urinary spot analysis allowed the calculation of the fractional excretion (FE%) of filtered sodium, potassium, chloride, urea and uric acid: FEx=Ux/PxxPcreat/Ucreatx100. During this procedure, all patients were following a mild water restriction (<1 l/day) and a normal hospital diet, containing about 120 mosm NaCl/day. Only patients with urinary osmolality <220 mosm/kg H2O, with plasma potassium >3.4 mmol/l and without inflammatory biochemical characteristics were included. We limited our study to polydipsia patients without hypokalaemia, considering that solute retention (NaCl) took only place in the extracellular fluid. If we had included patients with a potassium deficit, our method to estimate changes in extracellular volume would have been less accurate, due to transcellular fluid shift and electrolyte fluxes. Plasma protein concentration is a easily-measured parameter for indirect estimation of the plasma volume in hyponatraemic patients.7 Haematocrit determination by the routine Coulter Counter method is inaccurate in hyponatraemic patients.8
In five polydipsia patients (3 BE and 2 CWD) without obvious gastrointestinal losses, we had the opportunity to study balance data during the same procedure of isotonic salt infusion. The amount of sodium and chloride retention was calculated as the sum of infused (616 mosm) and oral ingested (estimated mean 120 mosm) saline, minus urinary sodium and chloride, measured during the 24 h urine collection. We noted that urinary potassium losses matched the oral potassium intake and therefore, potassium was not taken into account for solute balance. Statistical analysis was carried out by means of the paired t-test.
|
Results |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
In the whole group of 10 patients, mean plasma protein concentration at t0 was 7 ± 0.7 g/dl (Table 1). Their plasma sodium significantly increased after infusion of 2 l isotonic saline, while their mean FENa and mean FE osm did not significantly change (Table 1). Mean FENa increased from 0.5% to 1% (NS), due to a large natriuresis at t24 in three polydipsia patients, who at t24 already had corrected their plasma sodium. When these patients were excluded, mean FENa at t24 became 0.6%, nearly the same value as observed at t0, indicating salt retention. Mean supine blood pressure and mean supine heart rate at t0 were not significantly different from those at t24 (Table 1).
|
The clinical context of the five patients studied for balance data can be summarized as follows. Patient 1 was a 59-year-old male beer drinker, admitted after a fall. His history was not suggestive of seizures. Patient 2 was a 60-year-old woman with known schizophrenia. She was also a well-known water drinker, previously hospitalized for hyponatremia. She came in after a fall, without seizures. Patient 3 was a 61-year-old female, hospitalized for weakness and confusion. During hospitalization, the psychiatrist diagnosed psychotic disease. Patient 4 was a 36-year-old beer drinker, admitted for regression of his cognitive functions. He asked for medical help in stopping his alcohol abuse. Finally, patient 5 was a 50-year-old male beer drinker, who presented with a fall, just before he was transported to the ward. None of these patients presented with overt signs of volume depletion. Mean supine blood pressure and mean supine heart rate were not significantly different at t0, compared to the corresponding values at t24 (Table 2). However, in one patient, supine blood pressure (80/60 mmHg) was low at admission, but rose to 110/70 mmHg after 2 l isotonic saline. Another patient presented with supine blood pressure of 120/80 mmHg, a rather low value for this patient, rising to 180/80 mmHg at t24. In two of these five patients, both upright and supine blood pressures were measured and no significant orthostatic hypotension was observed. Although it seems likely that some of these patients, particularly the beer drinkers, had had a poor food intake for several days at least, none of them presented with cachexia, or was aware of a specific imposed low-salt or low-protein diet.
|
The balance data in the five polydipsia patients showed a mean retention of 406 mosm solute (NaCl) and a mean decrease of 1.6 kg in body weight. This resulted in a mean increase of 11 mmol/l in plasma sodium (Table 2). Individual balance data demonstrated that polydipsia-related hyponatraemia is generally a mixed disorder, in which both variable degrees of solute depletion and water excess play a role (Table 3). In one patient (patient 3 in Table 3), we did not observe a fall, but rather an increase of 1.1 kg in body weight, during the 2 l isotonic saline infusion procedure. This patient also showed the largest saline retention of the five studied patients.
|
|
Discussion |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
In dilutional hyponatraemia, we expect low plasma protein concentrations, as the consequence of an expanded extracellular space.7 However, we found mean plasma protein to be 7 ± 0.7 g/dl in our 10 studied polydipsia patients, despite a decrease in plasma sodium of >10%. We found in an earlier study that SIADH patients with a similar degree of hyponatraemia have a mean plasma protein concentration of 6.7 ± 0.8 g/dl.9 Interestingly, in SIADH patients the plasma protein concentration does not change greatly after infusion of 2 l isotonic saline.9,10 Such patients already have a mildly expanded extracellular space.7 Polydipsia patients, however, show a significant decrease in their plasma protein concentration after infusion of 2 l isotonic saline (p < 0.02; Table 1), a response also seen in normonatraemic controls and in salt-depleted hyponatraemic patients.9 The hypothesis that polydipsia-related hyponatraemia is not a pure dilutional state is further strengthened by the increase in plasma sodium together with the absence of clear FENa and FEosm increases after isotonic saline (Table 1). Balance data, measured in three BE and two psychotic CWD patients, confirmed that both water retention and electrolyte loss contribute to the hyponatremia (Table 2). Individual analysis of the data showed that in some patients water retention was predominant, and in others solute loss was the most important determinant of the hyponatremia (Table 3). Most cases are of ‘mixed origin’, and the mean loss of body weight (1.6 kg), together with the mean value of solute retention (406 mosm NaCl; Table 2) suggest that water retention and solute loss contribute in roughly equal proportions to pathogenesis. Since we lack balance data over a longer period than the first 24 h, we can not exclude the possibility that the real solute deficit could be somewhat overestimated, particularly in patients who already normalized their plasma sodium at t24. We can assume that the kidney lacked sufficient time to excrete the infused sodium load.
It is difficult to judge the cause of the solute deficit in polydipsia-related hyponatremia. The patient histories, studying the balance data, do not suggest gastrointestinal solute loss. Our patients seemed well-nourished and were not following a low-protein, low-salt or vegetarian diet. It seems likely that beer drinkers have decreased food intake, but it remains unclear why CWDs are solute-depleted. An alternative explanation could be that the solute losses of polydipsia patients are caused by a renal pathway. Defence mechanisms against extracellular fluid expansion, including inhibition of the renin-angiotensin-aldosterone system and stimulation of natriuretic peptides, could be involved in solute loss of such patients.
Figure 1 shows what theoretically happens when in steady state, a polydipsia patient with a plasma sodium of 134 mmol/l and a plasma protein concentration of 6.6 g/% (patient 2 in Table 3), gains 2.6 l water or loses 450 mosm NaCl. Both manipulations on their own, contribute to the observed hyponatraemia and the combined effect of both features results in a pronounced hyponatraemia (calculated value of plasma sodium = 115 mmol/l; measured value at t0 = 117 mmol/l). However, the effects of these theoretical manipulations on the plasma protein concentration are opposite. Water loading of 2.6 l in this theoretical example induces a fall in plasma protein concentration of 9.6%, but the loss of 450 mosm NaCl leads to an increase in plasma protein concentration of 13.3%. The final result of the combined operation is an increase in plasma protein concentration of 3.7% (Figure 1). This means that the expected final plasma protein concentration is 3.7% higher than 6.6 g/dl, or 6.8 g/dl, corresponding fairly well to the measured value of 6.7 g/dl at t0. These opposite forces on plasma protein concentration may explain the apparent paradox of a normal plasma protein concentration, despite the increase in body weight due to water intoxication. This mathematical approach to the effects of water loading and solute loss on the plasma protein concentration does not take into account eventual transcellular ion fluxes (as can occur in hypokalaemia) or changes in capillary permeability.
|
Balance data over a longer period than the first 24 h, would have been of interest. In the case described by Demanet et al.,4 which was a case of severe hyponatraemia (103 mmol/l) of mixed origin (beer potomania and diuretics), sodium balance was positive for nearly 300 mmol sodium the first day and for about 550 mmol after 2 days, under an infusion of about 300 mmol Na per day. Further infusion of 150 mmol Na per day during the next two days, only slightly increased total sodium balance to 600 mmol. However, balance data over a longer period are not easy to obtain in clinical practice, because such patients often show new polydipsic behaviour as soon as their infusion is stopped and they regain their mobility. In patient 2, for example (Table 3 and Figure 1), body weight at t48 remained unchanged compared to t24, despite new polydipsic behaviour and a diuresis of 6.2 l/24 h. Plasma sodium concentration slightly decreased from 134 to 133 mmol/l and plasma protein concentration slightly increased from 6.6 to 6.7 g/dl. In another patient, body weight was available at t72 and was 0.1 kg higher than at t24. A third patient regained 2 kg in body weight at t48, but showed also new polydipsic behaviour. In this patient, diuresis between t24 and t48 was 4.75 l. Plasma sodium decreased from 133 to130 mmol/l and so did the plasma protein concentration, falling from 7.3 to 7.0 g/dl.
The patient described in Figure 1 improved his plasma sodium concentration by 17 mmol/l in 24 h, a correction rate that can be considered somewhat too fast, with regard to the danger of central pontine myelinolysis. This dangerous complication is generally observed in patients with plasma sodium <115 mmol/l11 (excluded from our study) and in patients with other predisposing risk factors as alcohol abuse, hypokalaemia, poor nutritional state, liver disease or burns,12 which was not the case in our patient, who showed no neurological deterioration. However, this example demonstrates that hypertonic saline can easily result in excessive correction rates, and should be used with caution. Our nine other patients did not exceed a correction rate of 12 mmol/l per 24h.
Individual analysis of data suggests that polydipsia could be a dynamic model, in which water intoxication predominates in the early stage and solute loss in a later stage. We suspect that the stage of water intoxication is often of limited duration and can easily be missed, especially when early urinary data are not available, or when attention is not focused on the detection of polyuria and on the follow-up of an initially measured body weight.
We conclude that hyponatremia related to polydipsia is a mixed disorder, in which about half of the sodium decrease is due to solute loss. This explains the apparently paradox of a normal mean plasma protein concentration, despite the increase in body weight, due to water intoxication.
|
Acknowledgments |
|---|
This work was supported by a grant from the Fonds National de la Recherche Scientifique (FNRS) (1.5.164.98F – 1.5.141.00F).
|
Footnotes |
|---|
Address for correspondence to Dr G. Decaux, Research Unit for the study of hydromineral metabolism, Department of General Internal Medicine, University Hospital Erasme, 808 Route de Lennik, 1070 Brussels, Belgium. e-mail: guy.decaux{at}skynet.be
|
References |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
1. Rose BD. Hypoosmolal States—Hyponatremia. In: Jeffers JD, Navrozov M, eds. Clinical Physiology of Acid-Base and Electrolyte Disorders, 5th edn. New York, McGraw-Hill, 2001:699–723.
2. Sterns RH, Ocdol H, Schrier RW, Narins RG. In: Narins RG, ed. Clinical disorders of fluid and electrolyte metabolism. Hyponatremia: pathophysiology, diagnosis and therapy, 5th edn. New York, McGraw-Hill, 1995:615–883.
3. Fenves AZ, Thomas S, Knochel JP. Beer potomania: two cases and review of the literature. Clin Nephrol 1996; 45:61–4.[Web of Science][Medline]
4. Demanet JC, Bonnyns M, Bleiberg H, Stevens-Rockmans C. Coma due to water intoxication in beer drinkers. Lancet 1971; 2:1115–17.[Medline]
5. Hilden T, Svendsen TL. Electrolyte disturbances in beer drinkers. Lancet 1975; 2:245–6.[Web of Science][Medline]
6. Thaler S, Teitlebaum I, Berl T. ‘Beer Potomania’ in non-beer drinkers: effect of low dietary solute intake. Am J Kidney Dis 1998; 31:1028–31.[Web of Science][Medline]
7. Namias B, Soupart A, Kornreich A, Decaux G. In human patients, vascular water retention during DDAVP-related hyponatremia occurs mainly in the plasma volume and not in the erythrocyte. J Lab Clin Med 1996; 128:612–17.[Medline]
8. Decaux G, Efira A, Dhaene M, Unger J. Interference of serum tonicity with the measurement of red cell mean corpuscular volume. Acta Haematol 1982; 97:62–6.
9. Musch W, Decaux G. Utility and limitations of biochemical parameters in the evaluation of hyponatremia in the elderly. Intern Urol Nephrology 2001; 32:475–93.[CrossRef][Medline]
10. Musch W, Thimpont J, Vandervelde D, Verhaeverbeke I, Berghmans T, Decaux G. Combined fractional excretion of sodium and urea better predicts response to saline in hyponatremia than do usual clinical and biochemical parameters. Am J Med 1995; 99:348–55.[CrossRef][Web of Science][Medline]
11. Cluitmans FHM, Meinders AE. Management of severe hyponatremia: rapid or slow correction? Am J Med 1990; 88:161–6.[CrossRef][Web of Science][Medline]
12. Soupart A, Decaux G. Therapeutic recommendations for management of severe hyponatremia: current concepts on pathogenesis and prevention of neurologic complications. Clin Nephrol 1996; 46:149–69.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  A M Goudie, D S Tunstall-Pedoe, M Kerins, and J Terris Exercise-associated hyponatraemia after a marathon: case series J R Soc Med, July 1, 2006; 99(7): 363 - 367. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||