+ MODEL

Kaohsiung Journal of Medical Sciences (2018) xx, 1e6

Available online at www.sciencedirect.com

ScienceDirect

journal homepage: http://www.kjms-online.com

Original Article

Evaluation of the effects of glucose on osmolal

gap using freezing point depression and vapor
pressure methods
Kuei-Ying Lin a,b, Po-Chih Chen a,c, Szu-Chia Chen d,e,f, Yeou-Lih Huang a,c,
Hung-Chun Chen e,f,*

a
Department of Medical Laboratory Science and Biotechnology, College of Health Sciences,
Kaohsiung Medical University, Kaohsiung, Taiwan
b
Department of Laboratory Medicine, Kaohsiung Municipal Ta-Tung Hospital, Kaohsiung,
Taiwan
c
Department of Laboratory Medicine, Kaohsiung Medical University Hospital, Kaohsiung
Medical University, Kaohsiung, Taiwan
d
Graduate Institute of Clinical Medicine, College of Medicine, Kaohsiung Medical University,
Kaohsiung, Taiwan
e
Division of Nephrology, Department of Internal Medicine, Kaohsiung Medical University
Hospital, Kaohsiung Medical University, Kaohsiung, Taiwan
f
Faculty of Medicine, College of Medicine, Kaohsiung Medical University, Kaohsiung, Taiwan

Received 17 January 2018; accepted 2 March 2018

KEYWORDS Abstract The measurement of serum osmolality, and the calculation of osmolal gap (OG)
Osmolal gap; from a calculated osmolality are widely used in clinical and emergency medicine. In this study,
Glucose; the possible effects of blood glucose on OG were investigated by freezing point depression and
Freezing point vapor pressure methods. The concentrations of sodium, glucose, blood urea nitrogen and os-
depression method; molalities of 2640 samples were measured. There were two methods for calculating serum
Vapor pressure osmolality: freezing point depression method (n Z 2399) and vapor pressure method
method (n Z 241). The OG was positively associated with glucose in glucose 110e450 mg/dL
(r Z 0.191, p < 0.001) and glucose > 450 mg/dL (r Z 0.372, p < 0.001), but not in
glucose < 110 mg/dL (r Z 0.017, p Z 0.711) in freezing point depression method. However,
OG had no correlation with glucose regardless of glucose level in vapor pressure method. In
freezing point depression method, compared with the groups of glucose <110 and 110
e450 mg/dL, the group with glucose >450 mg/dL had higher OG (p < 0.001) and higher prev-
alence of OG > 10 mOsm/Kg H2O (p < 0.001). Our study demonstrated that OG is impacted by

Conflicts of interest: All authors declare no conflicts of interests.

* Corresponding author. Division of Nephrology, Department of Internal Medicine, Kaohsiung Medical University Hospital, Kaohsiung
Medical University, 100 Tzyou First Road, Kaohsiung, 807, Taiwan.
E-mail address: chenhc@kmu.edu.tw (H.-C. Chen).

https://doi.org/10.1016/j.kjms.2018.03.002
1607-551X/Copyright ª 2018, Kaohsiung Medical University. Published by Elsevier Taiwan LLC. This is an open access article under the CC
BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Please cite this article in press as: Lin K-Y, et al., Evaluation of the effects of glucose on osmolal gap using freezing point depression and
vapor pressure methods, Kaohsiung Journal of Medical Sciences (2018), https://doi.org/10.1016/j.kjms.2018.03.002
 + MODEL
2 K.-Y. Lin et al.

increasing blood glucose concentration using freezing point depression method, special atten-
tion should be made to blood glucose concentrations when using freezing point depression
method to determine OG.
Copyright ª 2018, Kaohsiung Medical University. Published by Elsevier Taiwan LLC. This is an
open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/
by-nc-nd/4.0/).

Introduction Collection of demographic and laboratory data

Osmolality is the total concentration exerted by soluble
materials in the blood. It is generally influenced by 3 major
species including: sodium (Na), Glucose (Glu), and blood
urea nitrogen (BUN) [1]. Normal osmolality is of about
275e295 mOsm/kg H2O, where osmolality of greater than
300 mOsm/kg H2O is indicative of concentrated blood os-
moles. The possible cause of this increase in osmolality can
include: dehydration, high sodium, high blood sugar,
alcohol, methanol poisoning, and uremia [2]. The clinical
determination of osmolality is mainly used for diagnostic
and assessments for diseases, such as for diabetes patients
in the emergency room [3]. The determination of osmo-
lality can be utilized for emergency patients to assess low
blood sodium, dehydration, and kidney functions.
The difference between measured osmolality and theo-
retically calculated osmolality is termed osmolal gap (OG)
(OG Z measured osmolality
calculated osmolality). The
osmolality is calculated as: calculated osmolality Z
2*Na þ Glu/18 þ BUN/2.8 [1]. The OG has specific meanings
for diagnosis, treatment, and prognosis for the clinical
evaluation of unmeasurable osmolalities. The calculated
value is generally 10 mOsm/kg H2O less than the measure
value, resulting in an OG of 10 mOsm/kg H2O [4e7]. A OG
value greater than 10 mOsm/kg is a critical clinical value
for indications of (1) assessment of serum water content in
hyponatremia with hyperlipidemia or hyperglycemia; and
(2) the assessment of ingestion of substances such as
ethanol, methanol, and acetone [8].
Currently, there is no literature that had suggested the
correlation between OG and Glu. There are two common
methods of determining osmolality: (1) freezing point
depression method and (2) vapor pressure method. This
study utilizes both of these aforementioned methods for
determining the effect of Glu on OG.
Data including age, and sex were recorded from the pa-
tients’ medical records or interviews. Serum osmolality,
GLU, BUN, creatinine and Na were analyzed on the same
blood specimen. Osmolality was measured by freezing point
depression method with Advance 3250 (Advanced Instru-
ment, USA) and vapor pressure method with APRO Vapor
Pressure Osmometer (Wascor, EliTech Group, USA). The
testing GLU and BUN reagents were prepared with Unicel
DxC 800 Synchron Clinical System and SYNCHRON SYS-
TEMS AQUA CAL 1, 2 (Beckman Coulter Inc, USA). The So-
dium concentrations in blood were quantified by Unicel
DxC 800 Synchron Clinical System (Beckman Coulter Inc,
USA). The value of estimated glomerular filtration rate
(eGFR) was calculated using the 4-variable equation in the
Modification of Diet in Renal Disease (MDRD) study [9].
Statistical analysis
All statistical analyses were performed using SPSS software
for Windows version 19.0 (SPSS Inc. Chicago, USA). Data
were expressed as percentages, means  standard de-
viations, and means  standard error of the mean for OG.
The differences between groups were checked by Chi-
square test for categorical variables, or by independent t-
test for continuous variables. The relationship between two
continuous variables was assessed by a bivariate correlation
method (Pearson’s correlation). Multiple comparisons
among the study groups were performed by one-way anal-
ysis of variance (ANOVA) followed by post hoc test adjusted
with a Bonferroni correction. A p value of less than 0.05 was
considered to indicate statistical significance.
Results

Subjects and methods
Study patients and design
This study was conducted at a medical center and a
regional hospital in southern Taiwan, and enrolled 2640
patients from January 2011 to December 2016. The medical
orders and results that occurred within 6 years were obtain
from the clinical database and retrospectively analyzed in
this study. Samples with alcohols were excluded from this
study. The study protocol was approved by the Institutional
Review Board of Kaohsiung Medical University Hospital, and
all participants provided written informed consent to
participate in this study. The methods were carried out in
accordance with the approved guidelines.
A total of 2640 serum samples were included. There were
2399 OG samples using freezing point depression method
and 241 samples using vapor pressure method, respectively.
The mean age was 65.1  21.1 years. Table 1 shows the
comparison of clinical characteristics among the study
groups using freezing point depression method and vapor
pressure method.
Comparison of glucose, BUN and Na in patients with
OG& and >10 mOsm/Kg H2O using different
methods
Table 2 showed the comparison of glucose, BUN and Na in
patients with OG & and >10 mOsm/Kg H2O using different
methods. In freezing point depression method, compared

Please cite this article in press as: Lin K-Y, et al., Evaluation of the effects of glucose on osmolal gap using freezing point depression and
vapor pressure methods, Kaohsiung Journal of Medical Sciences (2018), https://doi.org/10.1016/j.kjms.2018.03.002
 + MODEL
Glucose effect on OG 3

Table 1 Comparison of clinical characteristics between freezing point depression method and vapor pressure method.
Parameters Freezing point depression Vapor pressure p All (n Z 2640)
method (n Z 2399) method (n Z 241)
Age (years) 64.5  21.4 70.4  17.7 <0.001 65.1  21.1
Male (%) 52.9 57.7 0.159 53.4
Osmolality (mOsm/Kg H2O) 298.0  23.9 293.9  34.2 0.068 297.6  25.0
OG (mOsm/Kg H2O) 12.8  8.3 1.2  0.8 <0.001 11.7  0.2
OG > 10 mOsm/Kg H2O (%) 51.4 22.8 <0.001 48.8
Glucose (mg/dL) 321.1  259.5 392.1  290.3 <0.001 327.6  263.2
Glucose subgroup
< 110 mg/dL (%) 465 (19.4%) 34 (14.1%) <0.001 499 (18.9%)
110e450 mg/dL (%) 1299 (54.1%) 115 (47.7%) 1414 (53.6%)
>450 mg/dL (%) 635 (26.5%) 92 (38.2%) 727 (27.5%)
BUN (mg/dL) 31.2  29.9 35.4  29.6 0.036 31.5  29.9
Na (mEq/L) 134.5  7.9 129.7  11.2 <0.001 134.1  8.4
eGFR < 60 mL/min/1.73 m2 (%) 51.1 58.3 0.039 51.9
Abbreviations: OG, osmolality gap; BUN, blood urea nitrogen; Na, sodium; eGFR, estimated glomerular filtration rate.

Table 2 Comparison of glucose, BUN and Na in patients with OG& and >10 mOsm/Kg H2O using different methods.
Parameters OG & 10 mOsm/Kg H2O OG > 10 mOsm/Kg H2O p
Freezing point depression method
Glucose (mg/dL) 204.6  151.4 431.5  28.9 <0.001
BUN (mg/dL) 26.6  28.9 34.5  30.3 <0.001
Na (mEq/L) 135.1  7.7 133.9  8.0 <0.001
Vapor pressure method
Glucose (mg/dL) 358.6  262.4 505.3  349.0 0.005
BUN (mg/dL) 32.6  28.8 44.9  30.7 0.006
Na (mEq/L) 129.2  11.6 131.5  9.8 0.175
Abbreviations: BUN, blood urea nitrogen; Na, sodium.

with OG & 10 mOsm/Kg H2O, OG > 10 mOsm/Kg H2O had
higher glucose (p < 0.001), higher BUN (p < 0.001) and
higher Na level (p < 0.001). In vapor pressure method,
compared with OG & 10 mOsm/Kg H2O, OG > 10 mOsm/Kg
H2O had higher glucose level (p Z 0.005) and higher BUN
(p Z 0.006), but no significance between Na (p Z 0.175).
Correlation between OG and glucose, BUN and Na in
different methods
Correlation between OG and glucose, BUN and Na was
shown in Table 3. The OG was positively associated with
glucose (r Z 0.385, p < 0.001) and BUN (r Z 0.080,
p < 0.001). However, OG had no correlation with Na
(r Z 0.001, p Z 0.943) in freezing point depression method.
Besides, in vapor pressure method, OG was positively
correlated with glucose (r Z 0.254, p < 0.001) and BUN
(r Z 0.190, p Z 0.003), but not Na (r Z 0.101, p Z 0.118).
Further, we compared correlation of OG and glucose in
freezing point depression method (r Z 0.385) and vapor
pressure method (r Z 0.254) [10], and the difference was
achieving significant (p Z 0.031) in different methods.
Correlation between OG and glucose in different
methods and different glucose level
Table 4 showed the correlation between OG and glucose,
BUN and Na. The OG was positively associated with glucose

Table 3 Correlation between OG and glucose, BUN and Na in different methods.

Parameters Freezing point depression method Vapor pressure method
Pearson’s r p Pearson’s r p
Glucose (mg/dL) 0.385 <0.001 0.254 <0.001
BUN (mg/dL) 0.080 <0.001 0.190 0.003
Na (mEq/L) 0.001 0.943 0.101 0.118
Values expressed as r.
Abbreviations: OG, osmolality gap; BUN, blood urea nitrogen; Na, sodium.

Please cite this article in press as: Lin K-Y, et al., Evaluation of the effects of glucose on osmolal gap using freezing point depression and
vapor pressure methods, Kaohsiung Journal of Medical Sciences (2018), https://doi.org/10.1016/j.kjms.2018.03.002
 + MODEL
4 K.-Y. Lin et al.

Table 4 Correlation between OG and glucose in different methods and different glucose level.
Glucose (mg/dL) Freezing point depression method Vapor pressure method
Pearson’s r p Pearson’s r p
<110 (n Z 499) 0.017 0.711
0.210 0.233
110e450 (n Z 1414) 0.191 <0.001 0.062 0.507
>450 (n Z 727) 0.372 <0.001 0.123 0.232
Values expressed as r.
Abbreviations: OG, osmolality gap.

in glucose 110e450 mg/dL (r Z 0.191, p < 0.001) and
glucose > 450 mg/dL (r Z 0.372, p < 0.001), but not in
glucose < 110 mg/dL (r Z 0.017, p Z 0.711) in freezing
point depression method. However, OG had no correlation
with glucose regardless of glucose level in vapor pressure
method.
Comparison of OG and OG > 10 mOsm/Kg H2O in
different glucose level using different methods
Comparison between OG and OG > 10 mOsm/Kg H2O in
different glucose level using different methods was
demonstrated in Table 5. In freezing point depression
method, compared with the groups of glucose <100 and
110e450 mg/dL, the group with glucose >450 mg/dL had
higher OG (p < 0.001) and higher percentage of OG > 10
mOsm/Kg H2O (p < 0.001). Besides, in vapor pressure
method, compared with the group of glucose 110e450 mg/
dL, the group with glucose >450 mg/dL had higher OG
(p < 0.001) and higher percentage of OG > 10 mOsm/Kg
H2O (p Z 0.041).
Comparison of OG and OG > 10 mOsm/Kg H2O in
different renal function using different methods
Table 6 showed the comparison of OG and OG > 10 mOsm/
Kg H2O in different renal function using different methods.
In freezing point depression method, compared with pa-
tients with eGFR S 60 mL/min/1.73 m2, the values of OG
(p Z 0.002) and the percentage of OG > 10 mOsm/Kg H2O
were higher in patients with eGFR < 60 mL/min/1.73 m2
(p < 0.001). In vapor pressure method, compared with
patients with eGFR S 60 mL/min/1.73 m2, the values of
OG (p < 0.001) and the percentage of OG > 10 mOsm/Kg
H2O (p Z 0.013) were higher in patients with
eGFR < 60 mL/min/1.73 m2.
Discussion
Osmolality is calculated by Na, Glu and BUN concentra-
tions, however, the actual measured osmolality will also be
affected by these factors. This study was established to

Table 5 Comparison of OG and OG > 10 mOsm/Kg H2O in different glucose level using different methods.
Parameters Glucose <110 mg/dL Glucose 110e450 mg/dL Glucose >450 mg/dL p
Freezing point depression method
OG (mOsm/Kg H2O) 9.7  0.6 10.7  0.3 19.2  0.5*,y <0.001
OG > 10 mOsm/Kg H2O (%) 29.7 43.3* 83.6*,y <0.001
Vapor pressure method
OG (mOsm/Kg H2O)
1.7  2.1
1.2  1.2 5.3  1.2*,y <0.001
OG > 10 mOsm/Kg H2O (%) 17.6 17.4 31.5y 0.041
Abbreviations: OG, osmolality gap.
*p < 0.05 compared with glucose <110 mg/dL.
y
p < 0.05 compared with glucose 110e450 mg/dL.

Table 6 Comparison of OG and OG > 10 mOsm/Kg H2O in different renal function using different methods.
Parameters eGFR S 60 mL/min/1.73 m2 eGFR <60 mL/min/1.73 m2 p
Freezing point depression method
OG (mOsm/Kg H2O)
1.6  1.3 3.6  1.1 0.002
OG > 10 mOsm/Kg H2O (%) 50.0 68.7 <0.001
Vapor pressure method
OG (mOsm/Kg H2O) 13.5  0.4 15.4  0.3 <0.001
OG > 10 mOsm/Kg H2O (%) 15.3 29.2 0.013
Abbreviations: OG, osmolality gap.

Please cite this article in press as: Lin K-Y, et al., Evaluation of the effects of glucose on osmolal gap using freezing point depression and
vapor pressure methods, Kaohsiung Journal of Medical Sciences (2018), https://doi.org/10.1016/j.kjms.2018.03.002
 + MODEL
Glucose effect on OG
determine if OG will be affected by Glu, and found that
high blood Glu can result in the increase OG.
Currently, there is no literature that had suggested the
correlation between OG and Glu. Our study indicated that
OG has moderate and low correlation to Glu when deter-
mined by freezing point depression and vapor pressure
methods, respectively (freezing point depression: r Z
0.385, p < 0.001; vapor pressure: r Z 0.254, p < 0.001).
Both of the measurement methods exhibited low OG cor-
relation to BUN and no OG correlation to Na. Furthermore,
in freezing point depression method, the results suggest
that samples with higher Glu can result in higher OG.
Therefore, the samples were separated into 3 groups ac-
cording to their Glu (normal: GLU < 110 mg/dL; high: GLU:
110e450 mg/dL; very high: > 450 mg/dL) for determining
their correlation with OG measured by each of the
methods. It was determined that OG obtained with
freezing point depression had statistically significant cor-
relations in high and very high Glu groups, but not in
normal Glu group. It means that the OG determined by
freezing point depression method is impacted by
increasing blood glucose concentration. It is recom-
mended that special attention should be made to Glu
concentrations when using freezing point depression to
determine OG.
To compare the two osmolality measurement methods,
vapor pressure requires less sample volume than freezing
point depression (10 vs. 20 mL, respectively). The cost of
performing vapor pressure measurement is also lower.
However, freezing point depression is more commonly used
in industries because it requires less frequent calibrations,
is easier to use; and because vapor pressure cannot be
utilized for measuring alcohol containing solution. Due to
insufficient budgeting, some clinical laboratories have no
instruments for measuring osmolality and all assessments
are performed with estimated calculations. However, some
patient assessment requires the determination of OG, thus,
the actual measurement of osmolality is required. There-
fore, calculated osmolality alone cannot be used for
determining a reliable diagnosis by the physician. Condi-
tions such as patients with chronic renal failure, alcohol
poisoning, and the use of mannitol requires the determi-
nation of OG [8]. The results in our study indicated that
high blood Glu can increase OG, where it could not be
determined with calculations alone.
For the treatment of alcohol poisoning patients, the
severity of the poisoning is determined by the consump-
tion volume as well as the duration between alcohol
ingestion and treatment for poisoning. Severe alcohol
poisoning can lead to blindness and death. Therefore, the
early diagnosis of alcohol poisoning is very important for
patient outcomes. The OG can be used for the diagnosis of
alcohol poisoning, where osmolality measurement is
required. Therefore, the lack of actual osmolality mea-
surement can lead to delayed treatment and worsening of
patient conditions. The results in our study indicated that
increased blood Glu can increase the OG. Therefore,
alcohol poisoning patients with high blood Glu could be
determined with inaccurate OG. Therefore, it is recom-
mended that the clinical use of OG for alcohol poisoning
patient should be performed in combination with blood
Glu assessment. Therefore, the use of OG for the diagnosis
Please cite this article in press as: Lin K-Y, et al., Evaluation of the effects of glucose on osmolal gap using freezing point depression and
vapor pressure methods, Kaohsiung Journal of Medical Sciences (2018), https://doi.org/10.1016/j.kjms.2018.03.002
5
of alcohol poisoning should be performed with care for
improving patient diagnosis and treatment.
Because large OG have been observed in renal failure
[11,12], we also assessed the renal function of each pa-
tients by use of the MDRD equation to estimate the eGFR.
In our study, OG and OG > 10 mOsm/Kg H2O was higher in
the group with eGFR <60 mL/min/1.73 m2. Therefore,
renal insufficiency appeared to affect the estimate of
OG.
There are several limitations to this study. First, the
lack of correlation between OG and glucose in different
glucose sub-groups may caused by the fact that the num-
ber of patients for vapor pressure method was very small
compared to that of freezing point depression method.
Besides, lack of measurement of toxic alcohols in each
patients and basal difference of sodium concentration
may also influence the results. Second, we use the
freezing point depression and vapor pressure methods to
measure osmolalities, and then may not generalize to
other institutions that do not use these methods. Finally,
intra-individual variation of osmolality may be noted, and
influence the OG data [13]. However, the biological vari-
ation in hours would be lower than what is observed in
weeks. In our study, the measurements were done in
hours.
Our study demonstrated that high blood Glu can result
in the increase OG when blood Glu is elevated. Further-
more, the impact of Glu on OG is more apparent when
measured with freezing point depression method. How-
ever, 90% of current clinical laboratories utilize freezing
point depression for measuring osmolality. Therefore, if
the freezing point depression was used in the determina-
tion of osmolality, special attention should be paid to
blood Glu concentrations for avoiding inaccurate OG
measurements.
References
[1] Heavens KR, Kenefick RW, Caruso EM, Spitz MG, Cheuvront SN.
Validation of equations used to predict plasma osmolality in a
healthy adult cohort. Am J Clin Nutr 2004;100:1252e6.
[2] Lynd LD, Richardson KJ, Purssell RA, Abu-Laban RB,
Brubacher JR, Lepik KJ, et al. An evaluation of the osmole gap
as a screening test for toxic alcohol poisoning. BMC Emerg Med
2008;8:5.
[3] Mruk B. Renal safety of iodinated contrast media depending
on their osmolarity e current outlooks. Pol J Radiol 2016;81:
157e65.
[4] Serum osmolality. N Engl J Med 1984;310:1608e10.
[5] Smithline N, Gardner Jr KD. Gapseanionic and osmolal. JAMA
1976;236:1594e7.
[6] Penney MD, Walters G. Are osmolality measurements clinically
useful? Ann Clin Biochem 1987;24:566e71.
[7] Lord RC. Osmosis, osmometry, and osmoregulation. Postgrad
Med J 1999;75:67e73.
[8] Kraut JA, adn Kurtz I. Toxic alcohol ingestions: clinical fea-
tures, diagnosis, and management. Clin J Am Soc Nephrol
2008;3:208e25.
[9] Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D. A
more accurate method to estimate glomerular filtration rate
from serum creatinine: a new prediction equation. Modifica-
tion of diet in Renal Disease Study Group. Ann Intern Med
1999;130:461e70.
 + MODEL
6 K.-Y. Lin et al.

[10] Diedenhofen B, Musch J. Cocor: a comprehensive solution for
the statistical comparison of correlations. PLoS One 2015;10:
e0121945.
[11] Hoffman RS, Smilkstein MJ, Howland MA, Goldfrank LR. Osmol
gaps revisited: normal values and limitations. J Toxicol Clin
Toxicol 1993;31:81e93.
[12] Shaikh G, Sehgal R, Sandhu S, Vaddineni S, Fogel J,
Rubinstein S. Changes in osmol gap in chronic kidney disease:
an exploratory study. Ren Fail 2014;36:198e201.
[13] Fraser CG, Cummings ST, Wilkinson SP, Neville RG, Knox JD,
Ho O, et al. Biological variability of 26 clinical chemistry
analytes in elderly people. Clin Chem 1989;35:783e6.

Please cite this article in press as: Lin K-Y, et al., Evaluation of the effects of glucose on osmolal gap using freezing point depression and
vapor pressure methods, Kaohsiung Journal of Medical Sciences (2018), https://doi.org/10.1016/j.kjms.2018.03.002