Clinical Chemistry 45: 1842-1849, 1999;
(Clinical Chemistry. 1999;45:1842-1849.)
© 1999 American Association for Clinical Chemistry, Inc.
Serum Sialic Acid in a Random Sample of the General Population
Maritta Pönniö1,2,
Hannu Alho3,
Seppo T. Nikkari4,
Ulf Olsson5,
Ulf Rydberg2 and
Pekka Sillanaukee1,2,4,a
1
Pharmacia & Upjohn Diagnostics AB, Alcohol Related Diseases, SE-112 87 Uppsala, Sweden.
2
Karolinska Institute, Department of Neuroscience,
Medical School, 10401 Stockholm, Sweden.
3
National Public Health Institute, Alcohol Research
Center, P. O. Box 719, FIN-00101 Helsinki, Finland and Research Unit
of Alcohol Diseases, University of Helsinki, 00100 Helsinki,
Finland.
4
University of Tampere Medical School and Tampere
University Hospital, Department of Clinical Chemistry, 33101 Tampere,
Finland.
5
Swedish University of Agricultural Sciences, Department
of Statistics, Data Processing and Extension Education, P. O. Box
7013, S-750 07 Uppsala, Sweden.
a Address correspondence to this author at: Oy Finnish Immunotechnology, Ltd., Lenkkeilijänkatu 8, FIN-33520 Tampere, Finland. Fax 358 3 3138 7050; e-mail pekka.sillanaukee{at}fitltd.net
 |
Abstract
|
|---|
Background: The serum sialic acid (SA) concentration has
been reported to be a potentially useful but nonspecific disease
marker. We wanted to study which factors influence SA
concentration in a well-characterized healthy population.
Methods: SA was determined in 97 women and 96 men with a
colorimetric Warren method.
Results: The mean ± SD concentrations of SA were 634
± 109 (95% confidence interval, 612–656) and 630 ± 106 (95%
confidence interval, 608–651) mg/L for women and men, respectively.
The serum SA showed a significant positive association with body mass
index and with systolic and diastolic blood pressure among both women
and men. SA also correlated significantly with the use of
contraceptive pills and age among women and with smoking among men.
Conclusions: Our study suggests that SA does not increase with
age in men but appears to increase with female menopause. The
strong positive association with blood pressure may explain why SA
predicts cardiovascular mortality.
|
Introduction
|
|---|
Sialic acids (SAs)1
are acetylated derivatives of
neuraminic acid. They are attached to nonreducing residues of the
carbohydrate chains of glycoproteins and glycolipids. The
suggested biological functions of SA include: (a)
stabilizing the conformation of glycoproteins and cellular membranes;
(b) assisting in cell to cell recognition and interaction;
(c) contributing to membrane transport; (d)
affecting the function of membrane receptors by providing binding sites
for ligands; (e) influencing the function, stability, and
survival of blood glycoproteins; and (f)
regulating the permeability of the basement membrane of glomeruli
(1).
Increased SA concentrations
have been observed in several diseases, e.g., tumors, myocardial
infarction, diabetes, inflammatory disorders, and alcoholism
(2)(3)(4)(5)(6)(7)(8). The clinical usefulness of serum SA determination in
inherited SA storage diseases is well established. Serum SA is also
increased during inflammatory processes because of increased
concentrations of richly sialylated acute phase glycoproteins. There
are data suggesting a positive relationship between serum SA and stroke
and cardiovascular mortality (9). Cancer patients have
increased SA concentrations, which correlate positively with the degree
of metastasis and are useful in monitoring treatment (2).
Thus, several different mechanisms may lead to increased SA
concentrations in various pathological conditions. The nonspecificity
of serum SA limits its clinical usefulness. Nevertheless, when combined
with other markers, SA concentrations are helpful in disease screening
and follow-up, as well as in monitoring of treatment.
The aim of the present work was to determine reference values for
serum SA in women and men among a well-characterized healthy
population. We also assessed the effect of age, body mass index
(BMI), blood pressure, smoking, and hormonal disturbances on serum SA
concentrations. There are different methods for measuring SA:
colorimetric, enzymatic, fluorescence, and chromatographic
(2)(10)(11). We used a slightly
modified version of the colorimetric thiobarbiturate assay by Warren
(12).
|
Materials and Methods
|
|---|
study population
The present study population is part of a national health program
conducted and coordinated by the National Public Health Institute of
Finland (KTL). The study was conducted according to the Helsinki
Declaration of Human Experimentation, and it was approved by the
Ethical Committee of Primary Health Clinics in Finland.
The study was conducted in five geographic areas in Finland at the
beginning of 1997. The total study population consisted of 11 500
subjects. The survey included a self-administrated questionnaire with
156 questions about health-related factors and socioeconomic
background. The subpopulation for the present study was selected using
the following exclusion criteria: (a) alcohol consumption
during the week before entering the study of >105 g ethanol or an
average weekly alcohol consumption during the past year of >105 g
ethanol; (b) being drunk more than once a month during the
last year; (c) pregnancy; (d) cerebrovascular
disease; (e) coronary heart disease or cardiac
insufficiency; (f) cancer; (g)
lung emphysema; (h) biliary calculi or infection;
(i) rheumatoid arthritis; (j) kidney or
urinary tract disease; (k) diabetes; and (l) use
of the following medications during the past month before entering the
study: sleeping pills, sedatives, antidepressants, or anticoagulants.
The average alcohol consumption was calculated assuming that beer, long
drinks, spirits, and wine contained 12 g of alcohol per
serving, and cider and low alcoholic wine have 4 g. For the
present study, 100 men and 100 women were randomly selected from the
larger population after implementing the exclusion criteria so that
each of the five age groups (18–30, 31–40, 41–50, 51–60, and
61–75) had an equal number of subjects, divided equally between both
genders. Sera from 97 women and 96 men were available for the study.
The sera were stored at -70 °C until use.
questionnaire and clinical measurements
Every patient answered a questionnaire on social and behavioral
factors (e.g., age, occupation, education, smoking status, and
drinking) and somatic diseases. Blood pressure was measured after 15
min of rest in a sitting position from the right arm of the subject by
trained nurses using standardized Hg manometers. The pressure was
recorded twice, and the second value was used in the present analysis.
Subjects were fasting 4 h before sample collection.
biochemical analysis
Serum SA was measured with a slightly modified version of the
method of Warren (12). In comparison to the original Warren
method, smaller volumes of samples and reagents were used, and
photometry was carried out at 
= 549 and 513 nm, instead of 549
and 532 nm. The modified Warren method used was performed as follows.
Known concentrations of N-acetylneuraminic acid were diluted
in redistilled water. The sample (100 µL) was hydrolyzed with 10 µL
of 1.5 mol/L sulfuric acid, mixed, and incubated for 60 min at
80 °C. The tubes were cooled to room temperature in tap water.
Subsequently, 50 µL of 0.2 mol/L sodium metaperiodate was added in
10.5 mol/L 85% orthophosphoric acid. The tubes were mixed and
incubated for 20 min at room temperature. Next, 500 µL of 0.77 mol/L
sodium meta-arsenite was added in 56 mmol/L sulfuric acid, and the
tubes were mixed until the the yellow-brown color disappeared. Then,
1.5 mL of 42 mmol/L thiobarbituric acid in 506 mmol/L sodium sulfate
was added, and the tubes were incubated in a boiling water bath for 15
min and then cooled to room temperature in tap water. Next, 2 mL of
cyclohexanone was added. The tubes were mixed vigorously and
centrifuged for 3 min at 500g. The supernatant was
transferred into new tubes, and the absorbance was read at 549 and 513
nm. The difference between A549 and
A513 was calculated, and the sample
concentration was read from the calibration curve. The measuring range
used was 5–125 mg/L of N-acetylneuraminic acid.
Carbohydrate-deficient transferrin (CDT) was determined by a double
antibody kit, CDTect RIA (Pharmacia & Upjohn AB, Diagnostics, Uppsala,
Sweden) according to the manufacturer's instructions. Serum
-glutamyltransferase (GGT) was assayed by a routine clinical
laboratory method. Serum SA was analyzed in duplicates, and CDT and GGT
were analyzed as single replicates.
statistical methods
The relationships between continuous variables were studied using
standard regression and correlation methods. Relationships between SA
and classification variables or combinations of classification
variables with continuous variables were modeled as general linear
models (13). SAS (1989) software was used for all analyses
and graphs (14). Mean values for groups, adjusted for the
effects of other variables, were calculated as least squares means
using the GLM procedure of the SAS (1991) package (13).
Piecewise regression models (15) were used to study how SA
may depend on age. The model y = 
+ ßx
+ d(x -
x0) + 
was fitted to the data using
an iterative algorithm where all parameters are simultaneously
estimated. Here, x is age and y is the logarithm
of the SA concentration. This model implies that there is a linear
relationship between x and y but that different
lines are used before and after some changepoint
x0, which is also estimated. If the
estimate of the parameter is not significant, there would be no
indication of a firm "age of change"; otherwise,
x0 is interpreted as the age at which
there is a change in the relationship between x and
y.
|
Results
|
|---|
descriptive statistics of serum sa
The mean ± SD concentrations of SA were 634 ± 109 mg/L
(95% confidence interval, 612–656 mg/L) in women and 630 ± 106
(95% confidence interval, 608–651) mg/L in men. The 90% and 95%
quantiles were 764 and 835 mg/L for women, and 767 and 809 mg/L
for men. Serum SA concentration was slightly skewed to the right for
both sexes. Therefore, the natural logarithm (ln) of SA concentration
was also used in the regression analysis.
association of serum sa with age
Table 1
shows the age dependency of serum SA, systolic blood pressure,
diastolic blood pressure, and BMI. In Fig. 1
, a piecewise regression analysis of SA with age is shown. It
appears that the mean serum concentrations of SA start to increase in
women at about 45 years of age, which is close to the mean age of
menopause (Figs. 1A
and 2A
; Table 1
). The break in the line is statistically significant
(P <0.05). In men, no statistically significant increase of
SA with age was observed, nor was there any detectable "age of
change" (Figs. 1B
and 2B
; Table 1
). Systolic and diastolic blood
pressure increased significantly with age in both genders.
View larger version (12K):
[in this window]
[in a new window]
|
Figure 1. Piecewise regression analysis of SA with age for women
(A) and men (B).
LnSA, natural logarithm of SA.
|
|
View larger version (13K):
[in this window]
[in a new window]
|
Figure 2. Box-plots of SA of different age groups for women
(A) and men (B).
The box extends from the 25th to 75th centile, with the
horizontal line at the median. The vertical
lines show the range. The dots in the figures are
observations outside the range ± 1.5 times the interquartile
range (75th-25th centile).
|
|
univariate analysis of serum sa with other variables
Univariate regression analysis was performed with SA as the
dependent variable. There was no essential difference in the main
results when lnSA was used instead of SA as the dependent variable. SA
concentrations correlated positively with age in women
(r = 0.28, P <0.01) but not in men. SA
correlated significantly with BMI (women: r = 0.38,
P <0.001; men: r = 0.22, P
<0.05), with systolic blood pressure (women: r = 0.48,
P <0.001; men: r = 0.25, P
<0.05), and with diastolic blood pressure (women: r =
0.43, P <0.001; men: r = 0.26, P
<0.05; Fig. 3
). SA and smoking showed a significant correlation in male
subjects (r = 0.24, P <0.05) but not in
females; in men, this trend was also seen when the number of cigarettes
per day (r = 0.18, P = 0.09) was used
as the independent variable. The age- and BMI-adjusted SA
concentrations of male smokers were significantly (P <0.05)
higher than those of nonsmokers. Mean alcohol consumption last week and
year, GGT, and CDT did not correlate significantly with SA.
View larger version (21K):
[in this window]
[in a new window]
|
Figure 3. Correlation of SA with BMI (A and
B), systolic blood pressure (C and
D), and diastolic blood pressure (E and
F) in female and male subjects, respectively.
|
|
multiple linear regression analysis of serum sa and associated
variables
A stepwise regression analysis was done with serum SA as the
dependent variable and the following independent variables: age,
smoking, BMI, systolic blood pressure, diastolic blood pressure, and
mean alcohol consumption during the last week and last year.
In women, when SA was adjusted for age and BMI, it correlated with
systolic (P <0.01) and diastolic (P <0.05)
blood pressure and with mean blood pressure (P <0.001). The
corresponding analysis for male subjects showed a correlation with
systolic blood pressure (P <0.05) and mean blood pressure
(P <0.05) but not with diastolic blood pressure.
association of serum sa with hormones in women
Table 2
summarizes the results concerning hormonal factors. The serum
SA concentrations were adjusted for age and BMI. Serum SA
concentrations were increased by the use of contraceptive pills
(P <0.05) and hormone treatment (P <0.05).
|
Discussion
|
|---|
Our reference material consisted of a randomly selected sample of
the general population. Our primary goal was to determine the
reference serum SA concentrations for women and men. We also studied
the effect of age, BMI, blood pressure, smoking, and hormonal factors
on serum SA.
The mean serum SA concentrations did not differ between female and male
subjects, and they were in the range of 0.52–0.73 g/L, which is in
accordance with earlier findings (2). The SA concentration
was somewhat skewed to the right, as also reported previously by
Lindberg et al. (16).
We show that the serum total SA concentration increases with age in
women but not in men. The reason for the increase among women seems to
be menopause. This is at variance with some earlier studies, which have
shown an increase of serum SA for both women and men and no effect of
menopause (16)(17)(18). On the other hand, there are also
studies where serum SA concentrations were not correlated with age for
either sex (19)(20)(21). Because several diseases are known to
increase serum SA concentrations (2), one explanation for
the increase of serum SA with age would be a higher frequency of
subclinically diseased individuals among the elderly. Several
SA-containing acute phase reactants also increase with age, e.g.,
fibrinogen, C-reactive protein, 1-acid glycoprotein,
1-antichymotrypsin, 1-antitrypsin, and
haptoglobulin (22)(23).
Serum SA concentrations and BMI showed a positive relationship. Other
studies have also shown a similar association
(16)(24)(25). The fact that BMI is
known to be associated with several other factors, including blood
pressure, underlines the importance of taking it into consideration
when assessing the independent importance of SA on the other factors.
We also found that smoking increased the serum SA concentration in men
but not in women. This is in accordance with a previous study where
young male smokers had increased SA concentrations, but the increase
was not seen in women (17). There is no apparent explanation
for why smoking increases serum SA in men but not in women.
In the present study, we made the primary observation that in otherwise
healthy men and women, serum SA correlated positively with blood
pressure, even after adjusting for age and BMI. A recent study of a
smaller healthy population of men and women (n = 46 and n =
54, respectively) has also reported an independent correlation of serum
total SA with systolic and diastolic blood pressure, but only in women
(26). In contrast, in male and female diabetic subjects,
there was no independent association of serum SA with mean blood
pressure (24). Blood pressure is considered an important
cardiovascular risk factor. The finding that blood pressure correlates
positively with serum SA may explain in part why SA predicts
cardiovascular mortality (9)(16). Other
cardiovascular risk factors are also known to be reflected in serum SA
concentration, such as the SA-containing acute phase protein fibrinogen
(27). In fact, there is a general chronic inflammatory
process involving many acute phase proteins in atherosclerosis
(28). Moreover, both SA and the risk of coronary heart
disease increase after menopause (29).
In conclusion, this study demonstrates that blood pressure and
serum SA are associated in both women and men independently of age and
BMI. This implicates one mechanism behind the earlier observation about
the association between SA and cardiovascular risk. A possible
explanation of these findings is that serum SA reflects acceleration of
the atherosclerotic process by increased blood pressure, which is a
known independent risk factor for atheroslerosis. In women, age and the
use of contraceptive pills are factors increasing serum SA. In men,
smoking increases serum SA. The biological mechanism(s) causing
increased serum SA in various diseases, including cardiovascular
disease, are far from clear. The above-mentioned factors should be
taken into consideration when choosing control groups for clinical
studies where serum SA is analyzed. Furthermore, because
contraceptives, smoking, BMI, and other factors, including menopause,
increase SA, it is difficult to give reference values of serum SA
separately for different age groups. On the other hand, the present
study supports the conclusion that SA is relatively stable among men
25–64 years of age and among women 25–54 years of age.
|
Acknowledgments
|
|---|
This study is supported partly by the Finnish Alcohol Research
Foundation and partly by the Elli and Elvi Oksanen fund under the
auspices of the Finnish Cultural Foundation.
|
Footnotes
|
|---|
1 Nonstandard abbreviations: SA, sialic acid; BMI, body mass index; CDT, carbohydrate-deficient transferrin; and GGT, -glutamyltransferase. 
|
References
|
|---|
-
Schauer R, Kelm S, Reuter G, Roggentin P, Shaw L. Biochemistry and role of sialic acids. Rosenberg A eds. Biology of the sialic acids 1995 Plenum Publishing Corp. New York. .
-
Sillanaukee P, Pönniö M, Jääskeläinen IP. Occurrence of sialic acids in healthy humans and different disorders. Eur J Clin Investig 1999;29:413-425.
[ISI][Medline]
[Order article via Infotrieve]
-
Romppanen J, Eskelinen M, Tikanoja S, Mononen I. Total and lipid bound serum sialic acid in benign and malignant breast disease. Anticancer Res 1997;17:1249-1254.
[ISI][Medline]
[Order article via Infotrieve]
-
Paszkowska A, Berbec H, Semczuk A, Cybulski M. Sialic acid concentration in serum and tissue of endometrial cancer patients. Eur J Obstet Gynecol Reprod Biol 1998;76:211-215.
[ISI][Medline]
[Order article via Infotrieve]
-
Crook M, Haq M, Haq S, Tutt P. Plasma sialic acid and acute-phase proteins in patients with myocardial infarction. Angiology 1994;45:709-715.
-
Crook MA, Couchman S, Tutt P. Plasma fibrinogen and its relationship to plasma sialic acid in non-insulin-dependent diabetes mellitus. Blood Coagul Fibrinolysis 1996;7:586-589.
[ISI][Medline]
[Order article via Infotrieve]
-
Dogan H, Pasaoglu H, Ekinciler OF, Tatlisen N. A comparative study of total protein, total and lipid associated serum sialic acid levels in patients with Behcet's disease and control groups. Acta Ophthalmol 1992;70:790-794.
-
Sillanaukee P, Pönniö M, Seppä K. Sialic acid–new potential marker of alcohol abuse. Alcoholism Clin Exp Res 1999;23:1039-1043.
[ISI][Medline]
[Order article via Infotrieve]
-
Lindberg G, Råstam L, Gullberg B, Eklund GA. Serum sialic acid concentration predicts both coronary heart disease and stroke mortality: multivariate analysis including 54 385 men and women during 20.5 years follow-up. Int J Epidemiol 1992;21:253-257.
[Abstract/Free Full Text]
-
Waters PJ, Lewry E, Penncock CA. Measurement of sialic acid in serum and urine: clinical applications and limitations. Ann Clin Biochem 1992;29:625-637.
-
Crook M. The determination of plasma or serum sialic acid. Clin Biochem 1993;26:31-38.
[ISI][Medline]
[Order article via Infotrieve]
-
Warren L. The thiobarbituric acid assay of sialic acids. J Biol Chem 1959;234:1971-1975.
[Free Full Text]
-
Littell RC, Freund RJ, Spector PC. SAS system for linear models, 3d ed 1991 SAS Institute, Inc. Cary, NC. .
-
SAS/Stat user's guide, version 6. Cary, NC: SAS
Institute, Inc., 1989..
-
Seber GAF, Wild CJ. Nonlinear regression 1989 Wiley New York. .
-
Lindberg G, Eklund GA, Gullberg B, Råstam L. Serum sialic acid concentration and cardiovascular mortality. Br Med J 1991;302:143-146.
-
Lindberg G, Råstam L, Gullberg B, Eklund GA, Törnberg S. Serum sialic acid concentration and smoking: a population based study. Br Med J 1991;303:1306-1307.
-
Crook M, Collins D, Lumb P, Fogelman I, Treolar A. The relationship between the female menopause and serum sialic acid, a known cardiovascular risk factor. Eur J Obstet Gynecol Reprod Biol 1998;76:185-187.
[ISI][Medline]
[Order article via Infotrieve]
-
Crook MA, Tutt P, Simpson H, Pickup JC. Serum sialic acid and acute phase proteins in type 1 and 2 diabetes mellitus. Clin Chim Acta 1993;219:131-138.
[ISI][Medline]
[Order article via Infotrieve]
-
Crook MA, Tutt P, Pickup JC. Elevated serum sialic acid concentration in NIDDM and its relationship to blood pressure and retinopathy. Diabetes Care 1993;16:57-60.
[Abstract]
-
Crook MA, Earle K, Morocutti A, Yip J, Viberti G, Pickup JC. Serum sialic acid, a risk factor for cardiovascular disease, is increased in IDDM patients with microalbuminuria and clinical proteinuria. Diabetes Care 1994;17:305-310.
[Abstract]
-
Crook MA, Treloar A, Haq M, Tutt P. Serum total sialic acid and acute phase proteins in elderly subjects. Eur J Clin Chem Clin Biochem 1994;32:745-747.
[ISI][Medline]
[Order article via Infotrieve]
-
Milman N, Graudal N, Andersen HC. Acute phase reactants in the elderly. Clin Chim Acta 1988;176:59-62.
[ISI][Medline]
[Order article via Infotrieve]
-
Chen J, Gall M, Yokoyama H, Jensen J, Deckert M, Parving H. Raised serum sialic acid concentration in NIDDM patients with and without diabetic nephropathy. Diabetes Care 1996;19:130-134.
[Abstract]
-
Yokoyama H, Jensen JS, Jensen T, Deckert T. Serum sialic acid concentrations is elevated in IDDM especially in early diabetic nephropathy. J Intern Med 1995;237:519-523.
[ISI][Medline]
[Order article via Infotrieve]
-
Crook M, Lumb P, Andrews V, Swaminathan R. Serum total sialic acid, a reputed cardiovascular risk factor, and its relationship to lipids, plasma fasting insulin, blood pressure and body mass index in normal individuals. Clin Sci 1998;95:53-57.
[Medline]
[Order article via Infotrieve]
-
Meade TW, Chakrabarti R, Haines AP, North WRS, Stirling Y, Thompson SG. Haemostatic function and cardiovascular death: early results of a prospective study. Lancet 1980;1:1050-1053.
[ISI][Medline]
[Order article via Infotrieve]
-
Stuart J, George AJ, Davies AJ, Aukland A, Hurlow RA. Haematological stress syndrome in atherosclerosis. J Clin Pathol 1981;34:464-467.
[Abstract/Free Full Text]
-
Witteman J, Grobbee D, Kok F, Valkenburg H. Increased risk of atherosclerosis in women after the menopause. BMJ 1989;298:642-644.
The following articles in journals at HighWire Press have cited this article:
|
|
|
|
 
B. M.R. Appenzeller and R. Wennig
Altered Distribution of Transferrin Isoforms According to Serum Storage Conditions
Clin. Chem.,
November 1, 2005;
51(11):
2159 - 2162.
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
