CLIN. CHEM.
26/12,
1745-1747
(1980)
Effect of Dimethyl Sulfoxide on Serum Osmolality
Douglas N. Runckel and J. Robert Swanson
Consecutive values for serum osmolality measured in a
patient exceeded (548 mOsm) the previously reported
pathological
range and, indeed, exceeded that generally
considered compatible with life. The patient was receiving
large amounts of intravenous dimethyl sulfoxide as part
of an experimental protocol to control intracranial pressure
resulting from trauma. This compound increases serum
osmolality linearly with respect to concentration over the
range considered therapeutically significant, 0 to 10 mL/L.
The mechanism of its effect on osmolality readings is
discussed. It does not interfere with any values for those
routine measurements
made by continuous-flow
(SMAC,
Technicon).
Discussion
This case is an extreme example of the hyperosmolal state
(Table 1). Serum hyperosmolality
has been described in many
clinical
settings
R.M.,
a 34-year-old
white
man,
was hospitalized
because
of severe head trauma. Comatose on arrival, he was found to
have a large parietofrontal
skull fracture. He soon underwent
surgery for removal of a large intracranial hematoma, and a
pressure-monitoring
time.
He received
pressure,
including
intracranial
multiple
screw was inserted
drugs
dexamethasone
at that
to control
intracranial
and mannitol.
In addition,
he was included in an experimental
protocol to control
intracranial
pressure and cerebral edema through use of intravenous dimethyl sulfoxide (CH3)2SO.
He lived for nine days,
but did not regain
During the course of his hospitalization
strikingly
osmolality
consciousness.
he developed
some
abnormal
laboratory
findings,
including
a serum
that steadily increased
to 548 mOsm (Table 1).
Figure 1 shows his recorded
output does not include insensible
osmotically
active drugs.
fluid intake and output
loss), Figure
2 the input
of a comatose patient. Note that urine output was adequate
throughout
the hospitalization.
The serum did show progressively increasing values for urea nitrogen and creatinine,
of renal failure
or disturbance
and, in
Osmolalities
as calculated
by a formula
provided
by Dor-
wart and Chalmers (3), mOsmol/kg
= 1.75 Na + (glucose/iS)
+ (urea nitrogen/2.8) + 10.1, are compared to measured values
in Table 2. It can be seen that from the start there was a major
discrepancy between computed osmolalities and measured
values. This discrepancy, when it exceeds 40 mOsm, has been
associated
with a very poor prognosis
(4). Much of this
unexplained
osmotic activity can be attributed to administered osmotic
agents.
Mannitol,
which was administered
early
in the course, was probably totally cleared from the body by
the time the serum osmolality peaked on the 7th and 8th day
Table 1. Selected Clinical Laboratory Data on the
Patient during Hospitalization
a final concentration
of 10 mL/L. The two sera were then
analyzed
consecutively
by continuous
flow (SMAC, Technicon). The added compound
had no effect on values for any of
the 18 analytes measured
with this instrument.
Pooled human serum and saline were mixed with dimethyl
sulfoxide to give concentrations
of 0, 1, 2, 5, and 10 mL of dimethyl sulfoxide per liter. The osmolalities
of these solutions,
which were carefully sealed between runs and were analyzed
in duplicate,
were measured with a vapor-pressure
osmometer
osmometer.
some degree
addition, serum glucose increased progressively. Urine and
serum osmolalities were both measured on the day before the
patient’s death, and together they would indicate maintenance
of renal concentrating
ability.
of
A pooled specimen of human serum was divided into equal
portions; dimethyl sulfoxide was added to one portion to give
Figures
of those re-
(the
Experimental Data
and a freezing-point
many
this group. A review of this patient’s input and output record
(Figure 1) emphasizes
the difficulty in gauging the fluid needs
suggesting
Case Report
and this case incorporated
ported causes. Although cerebral lesions reportedly induce
diabetes insipidus through interference
with vasopressin
transport and excretion (1), Zierler, in his review (2), asserted
that relative dehydration, secondary to inadequate water intake, was far more commonly the cause of hyperosmolality
in
The results are presented
in
3a and b.
Serum
Day
1
Day 2
Day
6
Day
7
Day 8
Sodium, mmol/L
Potassium,
mmol/L
138
4.9
140
4.5
161
3.0
174
4.7
173
2.4
Chloride,
mmol/L
103
108
132
146
145
23
23
23
19
270
270
510
500
16
11
22
Bicarbonate,
mmol/L
Urea N, mg/L
Creatinine,
mg/L
Glucose, g/L
110
1.00
2.95
2.85
2.76
4.23
359
381
443
461(545) 548
Osmolality,
mOsm
Clinical Chemistry Section, Department
of Clinical Pathology,
University of Oregon Health Sciences Center, 3181S. W. Sam Jackson
Park Road, Portland, OR 97201.
Received May 22, 1980; accepted June 20, 1980.
Urine osmolality,
mOsm
CLINICAL
526
CHEMISTRY,
Vol. 26, No. 12, 1980
1745
450
425
Serum
.
a
Dayl
2
3
4
6
5
7
275
0%
.1%
.2%
0%
.1%
.2%
.3% 4%
.5% .6% .7%
Percent DMSO by Volume
.8%
.9%
.0%
.3%
.8%
.9%
1.0%
8
Day of Clinical
Course
Fig. 1. The patients intake
(#{149})
and output (0) of fluids during
the hospital course
(5). When we found
that the patient was receiving large
sulfoxide intravenously,
we attempted
compound
could account for these extraordinarily
high values for serum osmolality, 545 and 548
mOsm, well above the highest published value we could find
amounts
of dimethyl
to determine
if that
I
(6).
Dimethyl
sulfoxide
is a small, rapidly diffusing
molecule
with amphophilic
properties.
It freely permeates
biologic
membranes
and reaches maximal concentrations
in the blood
about 10 mm after cutaneous
application (7). It appears to
have free access to almost all body spaces, including bone and
brain (8). After intravenous
injection
in man, it has an
elimination half-time of four days (7), and most of the drug
appears to be excreted by the kidney unchanged
or as sulfone
(9). This patient received 840 g of dimethyl sulfoxide over his
eight-day course. If a four-day excretion half time is used and
divisions of less than one day are ignored, about 400 g of dimethyl sulfoxide would have accumulated
by the last day.
This would equal 0.57 g of dimethyl sulfoxide per 100 g of
lJ
Mernilol
Idsily do,.
oMso (doily
do,.
I,
roms)
I,
b
4%
.5%
.6%
7%
PercentOMSO by Volume in Water
Osmolalityat various concentrations of dimethyl sulfoxide
(DMSO) In pooled serum and saline
FIg.3.
a: as measured wIth the Wescor Vapor Pressure Osmometer 5100 B; b: as
meast,ed wIth the Advanced Instruments Freezing Point Osmometer-Dlgamatlc Model 3D
tissue, if equal distribution
over the entire body is assumed.
Dimethyl sulfoxide, however, reaches much lower concentrations in bone (7, 9) than it does in soft tissues, and as bone
makes up a substantial proportion of body weight, dimethyl
sulfoxide concentrations
would be expected to be about 0.9
to 1 g per 100 g of tissue. In addition, animal work (7, 9)
suggests
a relatively
greater
partitioning
into aqueous
than
lipid environments,
a factor that would further increase the
concentration
in serum. Figures 3a and 3b show that dimethyl
sulfoxide concentrations
of 10 g/kg, which we estimate in this
patient, would increase the osmolality by about 100 mOsm/kg.
The difference between measured osmolality and osmolality
calculated
from measurements
of normal
serum
constituents
Table 2. Measured vs Calculated Serum
Osmolailty for the Patient during Hospitalization
Calculated
Hospital day
7
Fig. 2. The patient’s dosage of osmotically active drugs during
the hospitalcourse
1748 CLINICALCHEMISTRY,Vol. 26, No. 12, 1980
1
3
6
7
8
Measured
Difference
mOsni
275
295
334
366
372
359
381
443
461
585
84
86
109
95
176
(Table 2) ranges from 84 to 176 mOsm/kg
in this patient.
Therefore,
the amount of increase in osmolality correlates well
with our estimate
of the dimethyl
sulfoxide concentration
in
the patient’s serum.
Although dimethyl sulfoxide has been documented to inhibit certain enzymes in vitro at high concentration (9), when
added to a control specimen it had no effect on the clinical
chemical results at 10 mL of dimethyl sulfoxide per liter of
serum.
Dimethyl sulfoxide tagged with 35S is significantly bound
to plasma protein (9), but this is not reflected with our in vitro
osmotic comparisons
between the osmolality of it in serum and
saline. In fact, freezing point depression osmometer measurements indicate a greater rate of osmolal increase with
concentration of dimethyl sulfoxide in serum as compared to
saline. Apparently,
for each molecular equivalent of dimethyl
sulfoxide bound to protein or other binding substances in
plasma, slightly more than one molecular equivalent
of
osmotically active solute is released.
There is disagreement over the suitability of the two major
types of clinical laboratory osmometer for measuring or detecting small hydrophilic molecules such as ethanol, methanol,
or acetone by their apparent osmotic effects on serum (10, 11).
That this dual-solvent system .does not truly reflect osmolality
as originally defined was discussed by Barlow (11). In addition, it is apparent that the “osmotic perturbation”
created
by adding to water a second compound with a non-negligible
vapor
pressure
is a function
of temperature
of measurement,
the phase, and vapor-pressure
values of the mixture relative
to temperature.
In contrast to all of the volatile fluids previously discussed, dimethyl sulfoxide has freezing and boiling
points-18.45
and 189 #{176}C,
respectively-that
are above those
of water.
If water
and dimethyl
sulfoxide
formed
a perfect
solution, the freezing point would be above 0 #{176}C,
creating an
apparent lowering in osmolality as measured by freezing point
depression. An examination
of the extremely non-ideal di.
methyl sulfoxide/water
phase diagram (12) shows why this
is not the case, even for very dilute concentrations. In the area
of clinical
interest,
either osmometer
concentration
is reflected
equivalently
by
type.
The calculated osmolality of a 10 mL/L solution of dimethyl
sulfoxide in water, if it is assumed that dimethyl sulfoxide is
a non-aggregating solute with negligible vapor pressure, is 141
mOsm. This value is very close to that measured in saline (139
mOsm by vapor pressure and 144 by freezing point depression).
In conclusion, this work demonstrates
that dimethyl sulfoxide can markedly increase serum osmolality, the increase
being linearly related to concentration
of dimethyl sulfoxide
in the clinically relevant region from 0 to 10 mL/L. Thus osmometry might offer an easy indirect measure of serum dimethyl
sulfoxide
concentration.
Dimethyl
sulfoxide
has no
effect on in vitro routine clinical chemical measurements
at
these concentrations,
and both freezing point and vapor
pressure osmometers give equivalent readings.
References
1. Engstrom, W.W.,and
Liebman, A.,Chronic hyperosmolarity
of
body fluids with a cerebral lesion causing diabetes insipidus and anterior pituitary insufficiency. Am. J. Med. 15, 180 (1953).
2. Zierler, K. L., Hyperosmolarity
in adults: A critical review. J.
Chronic Dis. 7, 1-23 (1958).
3. Dorwart, W. V., and Chalmers, L., Comparison of methods for
calculating serum osmolality from chemical concentrations,
and the
prognostic value of such calculations. Clin. Chem. 21, 190(1975).
4. Rubin, A. L., Braveman, W. S., Dexter, R. L., et al., The relationship between plasma osmolality and concentration
in disease states.
Clin. Res. Proc. 4, 129 (1956).
5. Mudge, G. H., Diuretics and other agents employed in the mobilization of edema fluid. In The Pharmacological
Basis of Therapeutics, 3rd ed., Goodman and Gilman, Eds., New York, NY, 1965, pp
829-830.
6. Stern, E. L., Serum osmolality in cases of poisoning. N. EngI. J.
Med. 21, 1026 (1974).
7. Kolb, K. H., Jaenicke, G., Kramer, M., and Schulze, P. E., Absorption, distribution, and elimination of labeled dimethyl sulfoxide
in man and animals. Ann. N.Y. Acad. Sci. 141, 85 (1967).
8. Denko, C. W., Goodman, R. M., Miller, R., and Donovan, T., Distribution of dimethyl sulfoxide-S
in the rat. Ibid., p 77.
9. Gerhards, E., and Gibian, H., The metabolism of dimethyl sulfoxide and its metabolic effects in man and animals. Ibid., p65.
10. Rocco, R. M., Volatiles and osmometry. Clin. Chem. 22, 399
(1976). Letter.
11. Barlow, W. K., Volatiles and osmometry (cont.). Clin. Chem. 22,
1231-1232 (1976). Letter.
12. Rasmussen, D. H., and MacKenzie, A. P., Phase diagram of the
system water-dimethyl
sulfoxide. Nature 200, 1315-1317 (1968).
CLINICAL CHEMISTRY, Vol. 26, No. 12, 1980
1747