Clinical Chemistry 43: 2397-2402, 1997;
(Clinical Chemistry. 1997;43:2397-2402.)
© 1997 American Association for Clinical Chemistry, Inc.
Age-specific distribution of plasma amino acid concentrations in a healthy pediatric population
Nathalie Lepage,
Nancy McDonald,
Louis Dallaire and
Marie Lamberta
Service de Génétique Médicale, Département de Pédiatrie, Hôpital Sainte-Justine and Université de Montréal, Montréal, QC, Canada.
a Address correspondence to this author at: Service de génétique médicale, Hôpital Ste-Justine, 3175 Côte Ste-Catherine, Montréal, QC, Canada H3T 1C5. Fax 514-345-4766; e-mail lamberma{at}ere.umontreal.ca
 |
Abstract
|
|---|
Reference values were determined for 23 plasma free amino acids from
measurements done in 148 healthy children ranging from 0 to 18 years of
age. Amino acid analysis was performed by ion-exchange chromatography.
We propose a graphic form of presenting the age-specific distribution
of plasma amino acid concentrations where the 10th, 50th, and 90th
quantiles are illustrated. Although each amino acid possesses its own
pattern of distribution, we can identify five different profiles. Nine
amino acids (alanine, arginine, asparagine, methionine, ornithine,
phenylalanine, proline, threonine, and tyrosine) demonstrate a decrease
in their concentrations during the first year of life; their
concentrations then tend to increase throughout childhood and
adolescence. Nine others (cystine, glutamine, glycine, histidine,
isoleucine, leucine, lysine, tryptophan, and valine) show a steady
increase throughout infancy, childhood, and adolescence. Five amino
acids (aspartic acid, citrulline, glutamic acid, serine, and taurine)
do not follow these two common profiles. For the first time, quantile
curves are produced to illustrate the age-dependent variation of amino
acid concentrations from infancy to adulthood. This alternative way of
presenting amino acid concentrations may facilitate the follow-up of
patients with inborn errors of amino acid metabolism.
|
Introduction
|
|---|
Amino acid analysis by automated ion-exchange chromatography is a
procedure extensively used for diagnosis and follow-up of inborn errors
of metabolism. It is a technique that provides both highly reproducible
and sensitive results. Normative values for plasma amino acid
concentrations have been reported in adults (1)(2)(3)(4), in
infants ages 0–2 months (5) and 1–5 months
(6), and in older children ages 6–18 years
(1)(7)(8). These data are usually
presented in the form of mean concentrations with individual SDs for
the selected population
(1)(2)(4)(6)(7)(8).
However, no data describe the age-specific distribution of amino acid
concentrations from infancy to adulthood.
In the present paper, we propose an alternative method of analyzing
results that takes advantage of all amino acid values. The distribution
of amino acid concentrations is presented graphically from birth to 18
years of age, showing the 10th, 50th, and 90th quantile curves. Thus,
the amino acid values of a patient with a particular metabolic disease
can be easily compared from infancy to adulthood with those of the
control population of the appropriate age.
|
Materials and Methods
|
|---|
clinical specimens
Blood samples were collected in heparinized tubes from children of
both sexes ranging from ages 0 to 18 years after a 12-h fast (Table 1
). They were recruited
frompatients undergoing minor elective surgery at Hôpital
Sainte-Justine. Those known to have metabolic, renal, hepatic, cardiac,
or muscular diseases were excluded from the study. Those who had
presented a serious acute disease 3 months or less before surgery or
who showed failure to thrive were also excluded. The study was approved
by the Hospital Ethics Committee, and informed consent was obtained
from the parents and (or) patients. Because only six children under age
1 year were recruited, we included in our study the results from 28
children ages 0 to 12 months for whom we received blood samples and
who, after careful revision of their medical charts, showed no evidence
of metabolic, renal, hepatic, cardiac, or muscular disorders.
All plasma samples were analyzed fresh or stored at -80 °C until
assayed. Before analysis, 2-aminoethyl cysteine hydrochloride
(Calbiochem) was added to plasma and used as an internal calibrator.
Then, plasma samples were deproteinized according to Mondino et al.
(9) with sulfosalicylic acid (ICN Biomedicals) and brought
to pH 2.2 with LiOH (Baker Chemical). After filtration, samples were
ready for analysis.
amino acid analysis
Analysis was performed with the Beckman System 7300 High
Performance Amino Acid Analyzer, according to manufacturer's
specifications. Briefly, this analyzer utilizes cation-exchange
chromatography with a step buffer elution and with ninhydrin detection.
Absorbance of amino acid–ninhydrin complexes is detected at 570 nm for
primary amino groups and at 440 nm for proline and hydroxyproline.
Amino acid concentration is automatically calculated on the basis of
both the calibrators and the internal calibrator.
Buffers designed for the Beckman System 7300 Amino Acid Analyzer, the
amino acid calibrators, and ninhydrin were all purchased from Beckman
Instruments. Tryptophan, glutamine, asparagine (Calbiochem), and
5-aminolevulinic acid (Sigma) were added to complete the amino acid
calibrator mix.
cvs
To establish the mean CV for the individual amino acid
concentrations obtained with the Beckman System 7300 High Performance
Amino Acid Analyzer, six blood samples were used, and each sample was
analyzed on five different occasions. Conservation of the deproteinized
sample was at -80 °C. The mean CVs were calculated as follows: For
every amino acid, the CVs were calculated for each blood sample; the
mean CV was then computed for each amino acid.
statistical methods
Parametric statistical analyses were performed with SAS
statistical software [release 6.09 (1989), SAS Institute].
Nonparametric analyses were done with the Mathematica program [release
2.2 (1988), Wolfram Research]
First, a descriptive univariate analysis was undertaken to assess the
quality of the data, and to identify missing data, lack of continuity,
or nonnormal distribution characteristics. Hence t-tests
were performed on all amino acid concentrations, verifying for sex
difference. Significant statistical differences were obtained only for
isoleucine (P = 0.01) and for lysine (P
= 0.03) when data from the 148 control children were analyzed. However,
no significant statistical difference for these two amino acids was
observed (isoleucine, P = 0.4; lysine,
P = 0.08) when data from the 114 children over age 1
year were analyzed. Therefore, sex was not taken into consideration for
subsequent analyses.
Next, a regression analysis was performed on each amino acid
concentration to determine its polynomial relation to age
(10). Where the assumptions needed to apply regression
were met, the 10th and 90th quantiles were calculated. Then, the
two-sided 90% confidence bands for the 10th and 90th quantile curves
were computed (10). Because they would have been almost
superimposed on the 50th quantile curve, the two-sided confidence bands
were not calculated for the median curve.
Finally, when the assumptions needed to apply regression failed
(alanine, arginine, asparagine, aspartic acid, citrulline, cystine,
glutamic acid, isoleucine, leucine, methionine, ornithine,
phenylalanine, proline, serine, taurine, threonine, tyrosine, and
valine), parametric methods became inappropriate and we turned to
kernel estimators to calculate the 10th, 50th, and 90th quantiles.
Again, the two-sided 90% confidence bands were obtained
(11).
Both methods, as expected by definition, led to 10% of amino acid
concentrations being higher than the 90th quantile value, and to 10%
of concentrations being lower than the 10th quantile value.
|
Results and Discussion
|
|---|
This study presents the age-specific distribution, as well as the
10th, 50th, and 90th quantile curves, for concentrations of 23 amino
acids. To verify the precision of our technique, we determined mean CVs
for each amino acid tested. As shown in Table 2
, these mean CVs range from
4.0% to 12.1%. These results indicate that our technique is
satisfactory considering that methods with manual pipetting, such as
the one used for amino acid analysis, usually yield CVs between 10%
and 15% (12).
Values of amino acids tend to vary up to 18 years of age (Figs. 1
to 5). The concentrations
rangefrom undetectable to 1000 µmol/L and each amino acid possesses its
own maximum values. For convenience of illustration, we have used four
different scales with respect to the maximum concentration shown by
each amino acid (scale 1: maximum <100 µmol/L; 2: <250 µmol/L; 3:
<500 µmol/L; and 4: <1000 µmol/L). A unique pattern of
distribution is demonstrated for each amino acid. However, we can
identify common tendencies. As shown in Fig. 1a
–i, concentrations of
this group of amino acids tend to decrease during the first years of
life, then increase steadily up to 18 years of age. Alanine (a),
arginine (b), asparagine (c), methionine (d), ornithine (e),
phenylalanine (f), proline (g), threonine (h), and tyrosine (i) belong
to this first group. Two amino acids, serine and taurine, show an
initial reduction in their concentrations followed by stable
concentrations (Figs. 2
a and 2b).
Aspartic acid (Fig. 3
a) and glutamic acid (Fig. 3b
) demonstrate decreasing values
throughout the studied period. Figs. 4
and
5 display amino acids with no initial decrease in their
concentrations. Citrulline (Fig. 4
) shows a modest two-step increment:
The first increase occurs from 0 to 3 years of age, and the second from
13 to 15 years. The last group includes nine amino acids showing
steadily increasing concentrations throughout infancy, childhood, and
adolescence: cystine, glutamine, glycine, histidine, isoleucine,
leucine, lysine, tryptophan, and valine (Figs. 5a
to 5i). Table 3
gives the numerical values for the 10th, 50th, and 90th
quantiles of plasma amino acid concentrations at four different ages: 6
months, 2 years, 6 years, and 16 years.
View larger version (41K):
[in this window]
[in a new window]
|
Figure 1. Age-specific distributions of plasma concentrations of
alanine (a), arginine (b), asparagine
(c), methionine (d), ornithine (e),
phenylalanine (f), proline (g), threonine
(h), and tyrosine (i) from 148 control children.
The upper curve represents the 90th quantile (solid
line) with its 90% confidence bands (dotted lines).
The lower curve represents the 10th quantile with its 90%
confidence bands, and the line enclosed by the 10th and the 90th
quantiles represents the 50th quantile of amino acid
concentrations.
|
|
View larger version (15K):
[in this window]
[in a new window]
|
Figure 2. Age-specific distributions of plasma concentrations of
serine (a) and taurine (b) from control children
ages 0 to 18 years.
Curves are as defined in Fig. 1
.
|
|
View larger version (14K):
[in this window]
[in a new window]
|
Figure 3. Age-specific distributions of plasma concentrations of
aspartic acid (a) and glutamic acid (b) from
control children ages 0 to 18 years.
Curves are as defined in Fig. 1
.
|
|
View larger version (20K):
[in this window]
[in a new window]
|
Figure 4. Age-specific distributions of plasma concentrations of
citrulline from control children ages 0 to 18 years.
Curves are as defined in Fig. 1
.
|
|
View larger version (41K):
[in this window]
[in a new window]
|
Figure 5. Age-specific distributions of plasma concentrations of
cystine (a), glutamine (b), glycine
(c), histidine (d), isoleucine (e),
leucine (f), lysine (g), tryptophan
(h), and valine (i) from control children ages 0
to 18 years.
Curves are as defined in Fig. 1
.
|
|
The potential for comparison of these results with others is limited
because previous reports did not take into account the 10th, 50th, and
90th quantile values, and amino acid reference values were usually
presented as mean concentrations with their individual SDs for the
selected population. Scott et al. (13) studied amino acid
concentrations in infants at five different ages: 11 days, 21 days, 7
weeks, 11 weeks, and 15 weeks. The age distribution from 0 to 4 months
that can be estimated from the study is consistent with our results
except for alanine, asparagine, glycine, and lysine. In the Scott
study, alanine and asparagine showed a continuous increase in
concentrations, whereas we observed a decrease up to age 2 years for
asparagine (Fig. 1c
) and up to age 3 years for alanine (Fig. 1a
). Scott
et al. found a decrease in concentration for glycine and lysine
(13), whereas we observed an increase in concentration
(Fig. 5c
and g). However, Janas et al. (5) measured amino
acid concentrations in infants of three different ages: 2 weeks, 4
weeks, and 8 weeks. In comparison with our results, discordant profiles
are only seen for arginine and cystine: arginine showed a slight
increase, whereas we observed a decrease (Fig. 1d
), and cystine showed
a slight decrease, whereas we observed an increase in concentration
(Fig. 5a
). Finally, a third report compared amino acid concentrations
at ages 8 and 16 years (7). The conclusions reached in
that study were similar to ours except for phenylalanine, which showed
a decrease in concentration where we found an increase (Fig. 1f
). Thus,
our results compare adequately with previous reports.
Many analytes, including plasma amino acids, show an age-related
distribution of their concentrations. This underlines the importance of
using appropriate reference values when working with a pediatric
population. Quantile curves that illustrate the age-dependent variation
of the median (50th quantile) and the extremes (10th and 90th
quantiles) of plasma amino acid concentrations should be a useful tool
for the follow-up of patients with disorders of amino acid metabolism.
Variations in amino acid concentrations according to age should be
considered by clinical chemists studying amino acids in children.
|
Acknowledgments
|
|---|
Statistical analysis was performed by Marie-Claude Guertin from the
Départment de Mathématiques et Statistiques de
l'Université de Montréal. These statistical services were
made possible through special funding from the Fonds de Recherche en
Santé du Québec (FRSQ) and the Interservice Club Council
(Telethon of Stars) granted to the Group on Evaluative, Clinical, and
Epidemiologic Research at Hôpital Sainte-Justine. We thank Louise
Lortie and the anesthesiologists of Hôpital Sainte-Justine for
their help in recruiting subjects.
|
References
|
|---|
-
Armstrong MD, Stave U. A study of plasma free amino acid levels. II. Normal values for children and adults. Metabolism 1973;22:561-569.
[ISI][Medline]
[Order article via Infotrieve]
-
Perry TL, Hansen S. Technical pitfalls leading to errors in the quantitation of plasma amino acids. Clin Chim Acta 1969;25:53-58.
[ISI][Medline]
[Order article via Infotrieve]
-
Caballero B, Gleason RE, Wurtman RJ. Plasma amino acid concentration in healthy elderly men and women. Am J Clin Nutr 1991;53:1249-1252.
[Abstract/Free Full Text]
-
Moller SE. Quantification of physiological amino acids by gradient ion-exchange high-performance liquid chromatography. J Chromatogr 1993;613:223-230.
[ISI][Medline]
[Order article via Infotrieve]
-
Janas LM, Picciano MF, Hatch TF. Indices of protein metabolism in term infants fed human milk, whey-predominant formula, or cow's milk formula. Pediatrics 1985;75:775-784.
[Abstract/Free Full Text]
-
Lindblad BS, Alfven G, Zetterstrom R. Plasma free amino acid concentrations of breast-fed infants. Acta Paediatr Scand 1978;67:659-663.
[ISI][Medline]
[Order article via Infotrieve]
-
Gregory DM, Sovetts D, Clow CL, Scriver CR. Plasma free amino acid values in normal children and adolescents. Metabolism 1986;35:967-969.
[ISI][Medline]
[Order article via Infotrieve]
-
Walker V, Mills GA. Quantitative methods for amino acid analysis in biological fluids. Ann Clin Biochem 1995;32:28-57.
-
Mondino A, Bongiovanni G, Fumero S, Rossi L. An improved method of plasma deproteination with sulphosalicilic acid for determining amino acids and related compounds. J Chromatogr 1972;74:255-263.
[ISI][Medline]
[Order article via Infotrieve]
-
Royston P. Constructing time-specific reference ranges. Stat Med 1991;10:675-690.
[ISI][Medline]
[Order article via Infotrieve]
-
Ducharme GR, Gannoun A, Guertin MC, Jequier JC. Reference values obtained by kernel-based estimation of quantile regressions. Biometrics 1995;51:1105-1116.
[ISI][Medline]
[Order article via Infotrieve]
-
Sacher RA McPherson RA eds. Widmann's clinical interpretation of laboratory tests 10th ed. 1991:871 FA Davis Co. Philadelphia. .
-
Scott PH, Sandham S, Balmer SE, Wharton BA. Diet-related reference values for plasma amino acids in newborns measured by reversed-phase HPLC. Clin Chem 1990;36:1922-1927.
[Abstract]
The following articles in journals at HighWire Press have cited this article:
|
|
|
|
 
D. A van Waardenburg, C. T de Betue, Y. C Luiking, M. Engel, and N. E Deutz
Plasma arginine and citrulline concentrations in critically ill children: strong relation with inflammation
Am. J. Clinical Nutrition,
November 1, 2007;
86(5):
1438 - 1444.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
H. A.B. Smith, J. A. Canter, K. G. Christian, D. C. Drinkwater, F. G. Scholl, B. W. Christman, G. D. Rice, F. E. Barr, and M. L. Summar
Nitric oxide precursors and congenital heart surgery: A randomized controlled trial of oral citrulline
J. Thorac. Cardiovasc. Surg.,
July 1, 2006;
132(1):
58 - 65.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
D. Li, L. Pickell, Y. Liu, Q. Wu, J. S Cohn, and R. Rozen
Maternal methylenetetrahydrofolate reductase deficiency and low dietary folate lead to adverse reproductive outcomes and congenital heart defects in mice
Am. J. Clinical Nutrition,
July 1, 2005;
82(1):
188 - 195.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
M. L. Summar, J. V. Gainer, M. Pretorius, H. Malave, S. Harris, L. D. Hall, A. Weisberg, D. E. Vaughan, B. W. Christman, and N. J. Brown
Relationship Between Carbamoyl-Phosphate Synthetase Genotype and Systemic Vascular Function
Hypertension,
February 1, 2004;
43(2):
186 - 191.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
D. L. Pearson, S. Dawling, W. F. Walsh, J. L. Haines, B. W. Christman, A. Bazyk, N. Scott, and M. L. Summar
Neonatal Pulmonary Hypertension- Urea-Cycle Intermediates, Nitric Oxide Production, and Carbamoyl-Phosphate Synthetase Function
N. Engl. J. Med.,
June 14, 2001;
344(24):
1832 - 1838.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
