Journal of Diabetes and Its Complications 21 (2007) 374 – 380
Body fluid abnormalities in severe hyperglycemia in patients on chronic
dialysis: theoretical analysis
Antonios H. Tzamaloukasa,b,4, Todd S. Ingc, Kostas C. Siamopoulosd, Mark Rohrscheibb,
Moses S. Elisaf d, Dominic S.C. Rajb, Glen H. Murataa,b
b
a
New Mexico VA Health Care System, Albuquerque, NM, USA
Department of Internal Medicine, University of New Mexico School of Medicine, Albuquerque, NM, USA
c
Department of Internal Medicine, Hines VA/Loyola University Medical Center, Hines, IL, USA
d
Department of Internal Medicine, Medical School, University of Ioannina, Ioannina, Greece
Received 21 April 2006; accepted 10 May 2007
Abstract
Hyperglycemic syndromes cause disturbances in the tonicity of body fluids, the distribution of body water between major body fluid
compartments, and the external balance of body solute and water. The unique feature of dialysis-associated hyperglycemia (DH) is that, during
its development, it can cause changes exclusively in the internal balance of body solute (hypertonicity) and fluids (intracellular volume
contraction and extracellular volume expansion) without affecting the external balance of water or solute. This makes DH the proper substrate
for studying predictions of the changes in tonicity and extracellular volume caused by hyperglycemia because these predictions fail, by and
large, to account for changes in the external balance of sodium, potassium, and water observed in hyperglycemic syndromes occurring in
patients with preserved renal function. The predictions suggest that the baseline state of extracellular volume and the degree of hyperglycemia
are major factors influencing the magnitude of abnormalities in the tonicity and extracellular volume resulting from DH, while transfers of solute
between the intracellular compartment and the extracellular compartment have relatively smaller effects. Edematous patients are at risk for
greater hypertonicity and larger increases in their extracellular volume than euvolemic—or, even less, hypovolemic—patients with the same
degree of hyperglycemia. Studies reporting the treatment of DH with only insulin therapy can be used to test these theoretical predictions and to
analyze the relationship between solute and fluid abnormalities and clinical manifestations.
Published by Elsevier Inc.
Keywords: Hyperglycemia; End-stage renal failure; Tonicity; Extracellular volume
1. Introduction
Hyperglycemia causes disturbances in the external
balance of body water and solutes, the distribution of
body water between the intracellular compartment (IC)
and the extracellular compartment (EC), the serum
concentrations of solutes, and the body’s acid–base
balance. In patients with preserved renal function, hyperglycemia leads to profound losses of water, sodium, and
4 Corresponding reviewer. New Mexico VA Health Care System,
Albuquerque, NM 87108, USA. Tel.: +1 505 265 1711x2418; fax: +1 505
256 2888.
E-mail address: [email protected]
(A.H. Tzamaloukas).
1056-8727/07/$ – see front matter. Published by Elsevier Inc.
doi:10.1016/j.jdiacomp.2007.05.007
potassium through osmotic diuresis (Brodsky, Rappaport,
& West, 1950; Gennari & Kassirer, 1974; Seldin &
Tarail, 1950). These losses may dominate the clinical
picture (Daugirdas, Kronfol, Tzamaloukas, & Ing, 1989;
Feig & McCurdy, 1977; Foster, 1974; Pollock & Arieff,
1980). Consequently, the importance of fluid replacement
during management of hyperglycemia is well recognized
(Foster & McGarry, 1983; Kitabchi et al., 2001, 2003),
even when there is no clinical evidence of severe
hypovolemia (Adrogue, Barrero, & Eknoyan, 1989).
Osmotic diuresis is either absent or minimal in
dialysis-associated hyperglycemia (DH). Axelrod (1975)
proposed that clinical manifestations of severe hyperglycemia differ between patients with preserved renal
A.H. Tzamaloukas et al. / Journal of Diabetes and Its Complications 21 (2007) 374 – 380
function and patients with advanced renal failure because
of the absence of osmotic diuresis in the latter group.
Subsequent studies provided an understanding of the
pathogenesis of the main clinical manifestations of DH
and principles of treatment.
We analyzed the association between body fluid
abnormalities and clinical manifestations of DH and
investigated factors that could affect the development
and severity of clinical manifestations. In this report, we
addressed the theoretical aspects of body fluid abnormalities in DH. In a second review, which follows, we
analyzed published reports on DH, and we compared the
findings of these reports to the findings of studies of
hyperglycemia in patients with intact renal function and
to predictions of theoretical analysis.
2. Effects of hyperglycemia on internal solute–water
balance
Hyperglycemia causes an increase in EC solute and a
shift of water from the IC into the EC (Petersen, Khalaf,
Bocker, & Kerr, 1993; Seldin & Tarail, 1950). Water
shift is a consequence of the osmotic principle, which
states that osmolalities in the IC and the EC are equal in
steady state (Appelboom, Brodsky, Tuttle, & Diamond,
1958; Maffly & Leaf, 1959; Peters, 1944). As a result of
the osmotic principle, body water is distributed between
the IC and the EC in proportion to the total amount of
solute present in each compartment (Darrow & Yannett,
1935). This principle of body water distribution constitutes the basis of theoretical analyses and therapeutic
schemes of fluid shifts and osmolality changes under
various abnormal osmotic states (Adrogue & Madias,
1997, 2000a, 2000b; Axelrod, 1975; Kurtz & Nguyen,
2005; Tzamaloukas, 1983).
In addition to the principle of body fluid distribution,
body fluid changes in hyperglycemia are influenced by
three other processes, namely, osmotic diuresis, water
intake as a result of thirst, and acid–base disturbances
that might accompany hyperglycemia. The magnitude of
changes in body fluids that result from the principle of
body fluid distribution can be predicted (see below). In
contrast, the magnitude of changes in body fluids
resulting from the other three processes present in severe
hyperglycemia is, by and large, unpredictable. Since most
dialysis patients have absent or negligible renal function,
DH is unique in that changes in body solute and fluid
during its development or its treatment with insulin
administration alone can be the exclusive consequence
of the principle of body fluid distribution. This particular
circumstance allows us to compare changes observed and
predicted by theoretical models of hyperglycemia, which
are based exclusively on the principle of body fluid
distribution without taking osmotic diuresis or fluid
intake into consideration.
375
2.1. Formulas for calculating tonicity and extracellular
volume changes in hyperglycemia
The following formulas represent clinically useful
approximations of serum tonicity ([Ton]) and total osmolality ([Osm]):
½Ton ¼ 2½Na þ ½Glu
ð1Þ
½Osm ¼ 2½Na þ ½Glu þ ½Urea;
ð2Þ
where [Na] is serum sodium concentration; [Glu] is serum
glucose concentration, and [Urea] is serum urea concentration (all in mmol/l) (McCurdy, 1970). The tonicity or
effective osmolality formula is an indirect measure of the
ability of serum to shrink or swell cells that are suspended in
it. This formula provides a better approximation of the effect
of DH on body fluid distribution than total osmolality
because urea, often in high concentrations in dialysis
patients to bring about elevated total serum osmolality
values, is distributed in total body water and does not
engender osmotic fluid shifts in steady state (Kapsner &
Tzamaloukas, 1991).
Correction of hyperglycemia causes a reduction in
tonicity. This finding was first demonstrated indirectly by
balance studies in patients with intact renal function who
were treated for severe hyperglycemia (Tomkins &
Dormandy, 1971) and then directly in patients with DH
who were treated with insulin therapy only. Predicting
tonicity at the end of treatment (when serum glucose
concentration is normalized) is important in fluid management. Eq. (2) is not appropriate for this prediction. In
practice, the prediction is made by the formula D[Na]/
D[Glu], where D[Na] is the change in serum sodium
concentration and D[Glu] is the corresponding change in
serum glucose concentration during the development or the
correction of hyperglycemia. When glucose and sodium
concentrations are both expressed in the International
System of Units, the relation among changes in serum
glucose concentration, serum sodium concentration, and
serum tonicity level is expressed by the equation (Tzamaloukas & Avasthi, 1986):
D½Ton ¼ D½Glu 2D½Na:
ð3Þ
During the development of hyperglycemia, a gain in
EC solute (EC hypertonicity) causes an osmotic fluid
shift from the IC into the EC, with resulting dilution of
serum sodium concentration. Under these circumstances,
the change in EC volume is derived from baseline
(euglycemic) serum sodium concentration and the final
(hyperglycemic) serum sodium concentration as follows
(Feig & McCurdy, 1977): If [Na]1 and [Na]2 are serum
sodium concentrations at the time of euglycemia and at
the time of hyperglycemia, respectively; V 1 and V 2 are
the corresponding EC volumes; and DV is the change in
376
A.H. Tzamaloukas et al. / Journal of Diabetes and Its Complications 21 (2007) 374 – 380
EC volume during treatment (i.e., DV=V 2V 1), assuming
that the amount of sodium remains constant in the EC:
½Na1 V1 ¼ ½Na2 ðV1 þ DV Þ
ð4Þ
from which
DV =V1 ¼ ½Na1 =½Na2 1:
ð5Þ
Eq. (5) expresses the fractional increase in EC volume
from euglycemic baseline value as a consequence of the
development of hyperglycemia.
3. Tonicity in DH
3.1. Modeling of DH
The quantitative changes in serum tonicity and in body
fluids that result from hyperglycemia can be predicted by
models based on the principle of body fluid distribution.
Fig. 1 shows three schematic stages of hyperglycemia
developing in an anuric patient. In this figure, ordinates
depict IC and EC volumes, abscissas depict osmolalities,
and areas of rectangles (products of ordinates and abscissas)
depict the total amounts of solute in the IC and in the EC.
Fig. 1A shows euglycemic baseline stage. Fig. 1B shows the
hypothetical stage of the instantaneous addition of glucose
to the EC without any change in the amount of total body
water and prior to any osmotic movement of fluid. Fig. 1C
illustrates the final steady-state stage after the net movement
of water from the IC into the EC has ceased because IC and
EC tonicities have now become equal.
The model depicted in Fig. 1 is based on the assumptions
that: total IC solute is the same in (A), (B), and (C); total EC
solute is the same in (B) and (C); and the principle of body
fluid distribution is the sole determinant of the final body
fluid distribution in (C). Under these assumptions, which are
necessary and sufficient for characterizing the body as a
bperfectQ osmometer (Tzamaloukas, 1983), IC and EC
solute conservation equations between (B) and (C) can be
derived with D[Ton] and DV/V 1 as dependent variables.
Fig. 1 is consistent with published models of body fluid
changes brought about by hyperglycemia (Axelrod, 1975;
Crandall, 1974; Jenkins & Larmore, 1974; Katz, 1973;
Robin et al., 1979; Tzamaloukas, Kyner, & Galley, 1987). A
critical value for this model is the amount of solute (glucose)
added to the EC in Fig. 1B. The total solute added is equal
to D[Glu]1V 1. The relation between D[Glu]1 and D[Glu] is
as follows:
D½Glu1 ¼ D½Gluð1 þ DV =V1 Þ:
ð6Þ
In comparison to the euglycemic stage (A), hyperglycemia developing in an anuric patient causes (C) an
increase in tonicity (D[Ton]), a rise in EC volume (DV), and
a corresponding fall in IC volume in steady state. The
clinical question raised is whether these changes in tonicity
and body water distribution lead to any clinical manifestations. Intuitively, the magnitude of D[Ton] and DV should
be among the determinants of clinical manifestations.
Quantitative predictions of D[Ton] and DV follow.
3.2. Prediction of the change in tonicity caused by DH
Fig. 1. Schematic stages of the development of hyperglycemia in an anuric
patient. (A) Euglycemic baseline stage. (B) Hypothetical stage of
instantaneous addition to the EC of an amount of glucose equal to
D[Glu]1V 1, where V 1 is the baseline extracellular volume prior to any
osmotic water exchange between the IC and the EC. (C) Stage of osmotic
equilibration after the osmotic movement of water from the IC to the EC.
Total body water (extracellular volume plus intracellular volume) is
constant between (A), (B), and (C). Total intracellular solute (the area
defined as intracellular volume times tonicity) is constant between (A), (B),
and (C). Total extracellular solute (the area defined as extracellular volume
times tonicity) is constant between (B) and (C).
The expression D[Na]/D[Glu] has a negative sign
because the directions of changes in serum sodium and
glucose concentrations are opposite during the development
or the correction of hyperglycemia (Eq. (3)). A D[Na]/
D[Glu] value of 2.8 mmol/l per 100 mg/dl (or 0.5 mmol/
mmol) was initially proposed (Welt, 1959). This value
implies no change in tonicity from hyperglycemia because
osmolalities in a 0.5-mmol/kg H2O monovalent sodium salt
solution (with complete dissociation) and a 1-mmol/kg H2O
glucose solution are equal. Arithmetic values of D[Na]/
D[Glu] higher than 2.8, reported in certain clinical studies,
cannot be attributed to the osmotic translocation of intracellular fluid into the EC because they indicate decreases in
tonicity as hyperglycemia develops. Changes in the external
A.H. Tzamaloukas et al. / Journal of Diabetes and Its Complications 21 (2007) 374 – 380
balance of body solutes and water resulting in relative water
excess cause, in most probability, these high D[Na]/D[Glu]
values (Tzamaloukas, 1982).
The main contribution of the Katz (1973) formula is the
recognition that the numerical value of the ratio D[Na]/
D[Glu] must be lower than 2.8 to reflect the rise in tonicity
as a result of hyperglycemia. Katz proposed a D[Na]/D[Glu]
value of 1.6 mmol/l per 100 mg/dl (0.29 mmol/mmol).
The predicted value from this formula rate of increase in
tonicity (Eq. (3)) is 2.2 mOsm/kg per 100-mg/dl increase in
serum glucose concentration. The Katz formula assumes
normal baseline EC volume and a serum glucose concentration approximately equal to 56 mmol/l (1000 mg/dl). This
formula and its modifications (Crandall, 1974; Jenkins &
Larmore, 1974; Moran & Jamison, 1985; Robin et al., 1979;
Roscoe, Halperin, Rolleston, & Goldstein, 1975; Tzamaloukas et al., 1987) do not take into account the contributions of osmotic diuresis and water intake to changes in
serum tonicity in hyperglycemia.
Modifications of the Katz formula either express deviations from perfect osmometric behavior or are consequences
of the perfect osmometer hypothesis. Deviations from
perfect osmometric behavior address exchanges of solute
between the IC and the EC during the development of
hyperglycemia. One modification of the Katz formula
estimated a D[Na]/D[Glu] value between 1.35 and
1.50 mmol/l per 100 mg/dl by analyzing the entry of
solute (glucose) during the development of hyperglycemia
into the IC of tissues, such as the brain, that do not depend
on insulin for glucose uptake (Roscoe et al., 1975). A
second modification of the Katz formula computed the
effects on the ratio D[Na]/D[Glu] of potassium exit from the
IC into the EC during the development of hyperglycemia
(Moran & Jamison, 1985). In addition to being relatively
small (Tzamaloukas, Moran, & Jamison, 1986), solute
movements in these two modifications of the Katz formula
have offsetting effects on the net fluid shift between the IC
and the EC.
Unlike the influences discussed, consequences of the
perfect osmometer hypothesis can have pronounced effects
on the change in tonicity in DH. This category of factors
affecting the degree of hypertonicity in anuric hyperglycemia includes the baseline state of the EC volume and
the degree of hyperglycemia. The ratio of IC volume to EC
volume in euglycemia is the most powerful determinant of
the ratio D[Na]/D[Glu] (Moran & Jamison, 1985; Tzamaloukas et al., 1987). Hyponatremia develops in hyperglycemia because of the osmotic translocation of IC fluid
into the EC. If the IC did not exist, serum sodium
concentration would not change during hyperglycemia
(D[Na]/D[Glu]=0), and the rise in serum tonicity would
attain a theoretical maximum that would be equal to the rise
in serum glucose concentration expressed in millimoles per
kilogram (Eq. (3)). The transfer of fluid from the IC into the
EC attenuates the effect of hyperglycemia on EC tonicity.
Attenuation is large (the value of the ratio D[Na]/D[Glu]
377
approaches 2.8 mmol/l per 100 mg/dl) in EC volume
depletion, when the IC is large in relation to the EC, and
small (the value of the ratio D[Na]/D[Glu] approaches zero)
in markedly edematous states, when the IC is less large in
relation to the EC.
Progressive hyperglycemia has an effect on the relationship D[Na]/D[Glu] similar to the effect of edema (Robin et
al., 1979; Tzamaloukas et al., 1987) because the IC
becomes progressively smaller and the EC becomes
progressively larger as hyperglycemia progresses. However, the effect of progressive hyperglycemia becomes
noticeable only at extremely high blood glucose concentrations (Robin et al., 1979).
Finally, a recent analysis calculated the effect of
hyperglycemia on serum sodium using the Katz formula
plus potential changes in total exchangeable sodium,
potassium, and water; Gibbs–Donnan equilibrium; osmotic
coefficients of sodium salts; osmotically inactive sodium
and potassium salts; and body solutes other than sodium or
potassium salts (Kurtz & Nguyen, 2005). In summary, the
ratio D[Na]/D[Glu], and therefore the change in tonicity
caused by hyperglycemia, cannot be represented by a
single value in DH, but should be expected to vary
considerably according to the underlying state of the EC
volume, degree of hyperglycemia, and changes in other
body solutes.
Fig. 2 shows the effects of different states of EC
volume on the ratio D[Na]/D[Glu] (Tzamaloukas et al.,
1987) in hypothetical patients with varying euglycemic EC
volumes but with the same IC volume. In these models, a
rise in serum glucose concentration from 5.6 mmol/l
(100 mg/dl) to 20 mmol/l (360 mg/dl) produces D[Na]/
D[Glu] values of 0.41 mmol/mmol (2.30 mmol/l per
Fig. 2. Prediction of the ratio D[Na]/D[Glu] in progressive hyperglycemia in
hypothetical anuric patients with the same euglycemic intracellular volume
(28 l) and euglycemic extracellular volume: 14 l in euglycemia, 7 l in severe
volume deficit, 28 l in edema, 42 l in anasarca, and 56 l in extreme anasarca
(Tzamaloukas et al., 1987). D[Na]/D[Glu] values of 1 mmol/mmol and
5.6 mmol/l per 100 mg/dl are equivalent.
378
A.H. Tzamaloukas et al. / Journal of Diabetes and Its Complications 21 (2007) 374 – 380
Fig. 3. Prediction (Tzamaloukas et al., 1987) of the change in serum
tonicity secondary to progressive hyperglycemia in hypothetical anuric
patients with the same euglycemic intracellular volume (28 l) and with
euglycemic extracellular volume varying between severe volume deficit
(7 l) and extreme anasarca (56 l).
100 mg/dl) in volume depletion and 0.08 mmol/mmol
(0.45 mmol/l per 100 mg/dl) in extreme anasarca.
Fig. 3, drafted from Eq. (3), shows the effect of different
states of EC volume and of the degree of hyperglycemia on
tonicity in the hypothetical subjects depicted in Fig. 2. The
low numerical values of the ratio D[Na]/D[Glu] in patients
with edematous states (Fig. 2) are translated into high values
of the corresponding D[Ton] (Fig. 3).
4. Internal volume shifts in DH
Eq. (5) allows the calculation of the fractional increase in
EC volume (DV/V 1) during the development of hyperglycemia. The corresponding fractional decrease in IC
volume can also be calculated (Tzamaloukas et al., 1987).
Fig. 4 shows the fractional changes in EC and IC volumes in
hypothetical patients depicted in Figs. 2 and 3.
The change in tonicity from the euglycemic stage to the
hyperglycemic stage (from A to C in Fig. 1) is determined
by the fractional change in IC volume because the IC solute
remains constant during the development of hyperglycemia.
The greater is the fractional dehydration of the IC, the
greater will be the rise in tonicity. The same change in
tonicity is alternatively determined by the fractional dilution
of the EC volume. The greater is the fractional dilution of
the EC, the greater is the attenuation of tonicity from Fig. 1B
to Fig. 1C (the lower is the rise in tonicity from Fig. 1A to
Fig. 1C). Thus, fractional changes in IC and EC volumes
have opposite effects on the rise in tonicity during the
development of hyperglycemia.
Fig. 4 shows that for the same degree of hyperglycemia,
the fractional increase in EC volume is highest in states of
EC volume depletion and lowest in situations of extreme
anasarca, while the corresponding fractional reduction in IC
Fig. 4. Prediction (Tzamaloukas et al., 1987) of the fractional increase in
extracellular volume and the fractional decrease in intracellular volume
secondary to progressive hyperglycemia in hypothetical anuric patients with
the same euglycemic intracellular volume (28 l) and with euglycemic
extracellular volume varying from severe volume deficit (7 l) to extreme
anasarca (56 l).
volume is highest in states of extreme edema and lowest in
states of EC volume depletion. These fractional volume
changes in body fluid compartments are consistent with the
changes in tonicity shown in Fig. 3.
The magnitude of total EC volume expansion caused by
a given degree of hyperglycemia can be found by multiplying fractional EC volume increase, calculated by Eq. (5),
Fig. 5. Prediction (Tzamaloukas et al., 1987) of the total increase in
extracellular volume secondary to progressive hyperglycemia in hypothetical anuric patients with the same euglycemic intracellular volume (28 l)
and with euglycemic extracellular volume varying between severe volume
deficit (7 l) and extreme anasarca (56 l).
A.H. Tzamaloukas et al. / Journal of Diabetes and Its Complications 21 (2007) 374 – 380
by euglycemic EC volume when the value of the latter is
known. Euglycemic EC volumes differed greatly among the
states shown in Figs. 2–4, while euglycemic IC volume was
the same for all hypothetical patients. Fig. 4 predicts that the
magnitude of osmotic volume shift from the IC to the EC
must be greatest in patients with extreme anasarca and
lowest in those with EC volume depletion to cause the
depicted fractional changes in IC volume, even though the
fractional increase in EC volume is lowest in states of
extreme anasarca.
Fig. 5 shows the magnitude of EC volume gains,
calculated by multiplying the fractional rises in EC volume
shown in Fig. 4 by the euglycemic EC volume of each
hypothetical hyperglycemic patient depicted in Figs. 2–4.
The findings of Fig. 5 confirm the predictions from the
fractional IC and EC volume changes depicted in Fig. 4. For
the same degree of hyperglycemia, the increase in EC
volume is substantially larger in volume-expanded states
than in volume-contracted states. In a euvolemic patient
with an EC volume of 14 l and an IC volume of 28 l at the
time of euglycemia (5.6 mmol/l or 100 mg/dl), the predicted
increase in EC volume for a rise in blood glucose
concentration equal to 66.7 mmol/l (1200 mg/dl) and a
final serum glucose value of 72.3 mmol/l (1300 mg/dl)
is 2.3 l.
Note that Figs. 2–5 compare the different EC volume
states with the same degree of hyperglycemia. The amount
of glucose required to raise serum glucose concentration to a
given level (D[Glu]1V 1 in Fig. 1) is progressively larger as
baseline EC volume (V 1) increases progressively from a
state of EC volume depletion to one of extreme anasarca
(Figs. 2–5). The differences shown in Figs. 2–5 are caused
by the different amounts of glucose required to generate the
same level of hyperglycemia. The results would be different
if the same amount of glucose were added to the EC in each
of the volume states. For example, if the amount of glucose
required to raise serum glucose concentration by 66.7 mmol/
l (1200 mg/dl) in a euvolemic patient were added to the EC
of the patient with extreme anasarca in Figs. 2–5, D[Glu]
would amount to only 8.3 mmol/l (150 mg/dl).
5. Conclusions
This theoretical analysis concluded that DH causes
hypertonicity and EC volume expansion. The magnitudes
of the increases in tonicity and in EC volume in DH
depend on the degree of hyperglycemia and on states of
EC volume at the time of euglycemia. When subjected to
the same degree of hyperglycemia, edematous patients are
at higher risk for severe hypertonicity and larger EC
volume expansion than euvolemic patients and, all the
more so, than volume-depleted patients. Edematous
patients are, therefore, at a higher risk of developing
clinical manifestations caused by hyperglycemia-induced
fluid abnormalities.
379
Acknowledgment
This work was supported by the New Mexico VA Health
Care System.
References
Adrogue, H. J., Barrero, J., & Eknoyan, G. (1989). Salutary effects of
modest fluid replacement in the treatment of adults with diabetic
ketoacidosis. Use in patients without extreme volume deficit. Journal of
the American Medical Association, 262, 2108 – 2113.
Adrogue, H. J., & Madias, N. E. (1997). Aiding fluid prescription for
dysnatremias. Intensive Care Med, 23, 309 – 316.
Adrogue, H. J., & Madias, N. E. (2000). Hypernatremia. New England
Journal of Medicine, 342, 1493 – 1499.
Adrogue, H. J., & Madias, N. E. (2000). Hyponatremia. New England
Journal of Medicine, 342, 1581 – 1589.
Appelboom, J. W., Brodsky, W. A., Tuttle, W. S., & Diamond, J. (1958).
The freezing point of mammalian tissues after sudden heating in boiling
distilled water. Journal of General Physiology, 41, 1153 – 1169.
Axelrod, L. (1975). Response of congestive heart failure to correction of
hyperglycemia in the presence of diabetic nephropathy. New England
Journal of Medicine, 293, 1243 – 1245.
Brodsky, W. A., Rapapport, S., & West, C. D. (1950). The mechanism of
glycosuric diuresis in diabetic man. Journal of Clinical Investigation,
29, 1021 – 1032.
Crandall, E. D. (1974). Serum sodium response to hyperglycemia. New
England Journal of Medicine, 290, 465.
Darrow, D. C., & Yannett, H. (1935). The change in the distribution of body
water accompanying increase and decrease in extracellular electrolyte.
Journal of Clinical Investigation, 14, 266 – 275.
Daugirdas, J. T., Kronfol, N. O., Tzamaloukas, A. H., & Ing, T. S. (1989).
Hyperosmolar coma: Cellular dehydration and the serum sodium
concentration. Annals of Internal Medicine, 110, 855 – 857.
Feig, P. U., & McCurdy, D. K. (1977). The hypertonic state. New England
Journal of Medicine, 297, 1444 – 1454.
Foster, D. W. (1974). Insulin deficiency and hyperosmolar coma. Advances
in Internal Medicine, 19, 159 – 174.
Foster, D. W., & McGarry, J. D. (1983). The metabolic derangements and
treatment of diabetic ketoacidosis. New England Journal of Medicine,
309, 159 – 169.
Gennari, F. J., & Kassirer, J. P. (1974). Osmotic diuresis. New England
Journal of Medicine, 291, 714 – 719.
Jenkins, P. G., & Larmore, C. (1974). Hyperglycemia-induced hyponatremia. New England Journal of Medicine, 290, 573.
Kapsner, C. O., & Tzamaloukas, A. H. (1990). Understanding serum
electrolytes. How to avoid making mistakes. Postgraduate Medicine,
90, 151 – 161.
Katz, M. A. (1973). Hyperglycemia-induced hyponatremia: Calculation of
expected serum sodium depression. New England Journal of Medicine,
289, 843 – 844.
Kitabchi, A. E., Umpierrez, G. E., Murphy, M. B., Barrett, E. J., Kreisberg,
R. A., Malone, J. I., et al. (2001). Management of hyperglycemic crises
in patients with diabetes. Diabetes Care, 24, 131 – 153.
Kitabchi, A. E., Umpierrez, G. E., Murphy, M. B., Barrett, E. J., Kreisberg,
R. A., Malone, J. I., et al. (2003). Hyperglycemic crises in patients with
diabetes mellitus. Diabetes Care, 26 (Suppl. 1), S109 – S117.
Kurtz, I., & Nguyen, M. K. (2005). Evolving concepts in the quantitative
analysis of the determinants of the plasma water sodium concentration
and the pathophysiology of the dysnatremias. Kidney International, 68,
1982 – 1993.
Maffly, R. H., & Leaf, A. (1959). The potential of water in mammalian
tissues. Journal of General Physiology, 42, 1257 – 1275.
McCurdy, D. K. (1970). Hyperosmolar hyperglycemic nonketotic diabetic
coma. Medical Clinics of North America, 54, 683 – 699.
380
A.H. Tzamaloukas et al. / Journal of Diabetes and Its Complications 21 (2007) 374 – 380
Moran, S. M., & Jamison, R. L. (1985). The variable hyponatremic
response to hyperglycemia. Western Journal of Medicine, 142, 49 – 53.
Peters, J. P. (1944). Water exchange. Physiological Reviews, 24, 491 – 531.
Petersen, K. G., Khalaf, A. N., Bocker, J., & Kerr, L. (1993). Glucoseinduced water movement from the intracellular to the extracellular
space and its influence on calculations of glucose metabolism. European Journal of Clinical Investigation, 23, 266 – 269.
Pollock, A. S., & Arieff, A. I. (1980). Abnormalities of cell volume
regulation and their functional consequences. American Journal of
Physiology, 239, F195 – F205.
Robin, A. P., Ing, T. S., Lancaster, G. A., Soung, L. S., Sparagana, M., Geis,
W. P., et al. (1979). Hyperglycemia-induced hyponatremia: A fresh
look. Clinical Chemistry, 25, 496 – 497.
Roscoe, J. M., Halperin, M. L., Rolleston, F. S., & Goldstein, M. B. (1975).
Hyperglycemia-induced hyponatremia: Metabolic considerations in
calculation of serum sodium concentration. Canadian Medical Association Journal, 112, 452 – 453.
Seldin, D. W., & Tarail, R. (1950). The metabolism of glucose and
electrolytes in diabetic acidosis. Journal of Clinical Investigation, 29,
552 – 565.
Tomkins, A. M., & Dormandy, T. L. (1971). Osmolal pattern during
recovery from diabetic coma. Lancet, 2, 952 – 955.
Tzamaloukas, A. H. (1982). Prediction of the change in extracellular
sodium concentration in hyperglycemia complicated by severe losses of
water and solute. Clinical Chemistry, 28, 1722 – 1724.
Tzamaloukas, A. H. (1983). A working model of the perfect osmometer hypothesis in anuria. Mineral and Electrolyte Metabolism, 9,
93 – 98.
Tzamaloukas, A. H., & Avasthi, P. S. (1986). Effect of hyperglycemia on
serum sodium concentration and tonicity in outpatients on chronic
dialysis. American Journal of Kidney Diseases, 7, 477 – 482.
Tzamaloukas, A. H., Kyner, W. T., & Galley, W. R., Jr. (1987).
Determinants of osmotic phenomena created by an isolated change in
extracellular solute in anuria. Mineral and Electrolyte Metabolism, 13,
117 – 125.
Tzamaloukas, A. H., Moran, S. M., & Jamison, R. L. (1986). Clinical
importance of the determinants of the hyponatremic response to
hyperglycemia. Western Journal of Medicine, 144, 89 – 90.
Welt, L. G. (1959). Clinical disorders of hydration and acid–base balance.
(2nd ed.) Boston7 Little, Brown, and Co.