- Email: [email protected]
Plasma pancreatic and salivary-type amylase and immunoreactive trypsin concentrations: variations with age and reference ranges for children
265
Clinica Chimica Acta, 104 (1980) 265-273 @ Elsevier/North-Holland Biomedical Press
CCA
1385
PLASMA PANCREATIC AND SALIVARY-TYPE AMYLASE AND IMMUNOREACTIVE TRYPSIN CONCENTRATIONS: VARIATIONS WITH AGE AND REFERENCE RANGES FOR CHILDREN
MARY
D. O’DONNELL
a+* and N.J. MILLER
b
a Department of Medicine and Therapeutics, University College Dublin b Central Biochemistry Laboratory, Sir Patrick Dun’s Hospital, Dublin, (Received
November
13th,
and (Republic
of Ireland)
1979)
Summary The differential ar-amylase (EC 3.2.1.1) assay was applied to 166 control children in the age range 0.1-13 years. Circulating levels of both pancreatic and salivary-type amylase were very low (mean 20 U/l) in the first four months of life. Pancreatic levels increased gradually with age, reaching adult levels (mean 74 U/l) by the age of eight years. Salivary amylase levels showed a sharp rise in the 0.9-1.9 year period reaching maximum levels (mean 99 U/l) by age 5-6 years. While linear regression analysis showed significant correlation between age of subject and pancreatic and salivary amylase levels, no such correlation was evident between age and plasma immunoreactive trypsin levels over the age range studied. Plasma trypsin levels in children were lower than reported adult values. Reference ranges for pancreatic and salivary-type amylase and immunoreactive trypsin in children are presented. The importance of age-matching, when pancreatic amylase and plasma trypsin are being investigated in children, is emphasised.
Introduction The specific measurement of pancreatic (P) amylase in the presence of salivary (S)-type amylase in adult human serum [l], which requires no electrophoresis or chromatography, has been described. The method utilises an inhibitor protein with a specificity 100 times greater for human salivary than for human pancreatic amylase. Its use in the investigation of chronic pancreatitis and pancreatic insufficiency has been described [2,3]. Measurement of
*To
whom
correspondence shouldbeaddressed.
P-type amylase by this method also appears promising for the diagnosis of pancreatic insufficiency in children without the need for duodenal intubations [ 41. Since pancreatic and salivary amylase levels in children vary with age [ 5,6], it was desirable to study the normal variation with age for these enzymes using the inhibitor method and to establish reference values for children of different age groups. Plasma immunoreactive trypsin (IRT) values were also studied simultaneously, since reference values for children have not yet been established. Methods Subjects Control subjects were hospital in-patients ranging in age from 3 weeks to 13 years. No patient had any evidence of pancreatic or parotid gland disease, abdominal, renal or respiratory dysfunction or were receiving drug therapy. In all, there were 97 males and 69 females. Samples Non-fasting blood samples, 3-10 ml, were placed the plasma immediately separated after centrifugation
in heparinized tubes and and stored at -20°C.
Analyses Pancreatic and salivary-type amylase were determined by the method of O’Donnell et al. [ 11 and IRT by the radioimmunoassay kit (RIA-gnost Trypsin, Behringwerke-AG, Marburg, F.R.G.). Study of the effect of heparin on amylase assay In view of the presence of heparin in the plasma samples, it was important to establish at the outset the effect, if any, of added heparin on (a) P and S amylase activities and (b) the action of the inhibitor. This experiment was carried out as follows: S and P amylase and amylase-free serum were prepared as described previously [l]. S and P amylase were added separately to 0.2-ml samples of amylase-free serum. Heparin (Pullarin, 1000 U/ml, Evans Medical Ltd., U.K.) in the range 2-50 units was then added to enzyme-serum mixtures and tubes were incubated for 16 h at 4°C. They were analysed then for enzyme activities in the presence and absence of added inhibitor to the pre-incubation medium [ 11. Calculations We calculated the mean, standard deviation, skew and kurtosis for P-type amylase, S-type amylase and IRT for subjects in the various age groups by standard methods [7]. Reference ranges were calculated for each group as the 95% range from minus to plus two standard deviations from the mean. In cases where significant positive skew in distribution of data was found these ranges were calculated after logarithmic transformation of data. We sought linear correlations between age and concentrations of enzymes by regression analysis combined with analysis of variance. Differences between means were compared by the non-paired Student’s t test (n > 20) and by the Mann-Whitney U test (n < 20).
267
Results The effect of heparin (10250 U/ml serum) on salivary and pancreatic amylase activities are shown in Fig. 1. No change in enzyme activities was evident. Heparin was also shown to have no effect on the inhibition of either S or P-type amylase by the inhibitor. Table I shows the characteristics of the subjects according to their age group. No sex-related differences were noted for any of the enzymes studied in any age category. Hence, results for males and females were grouped together. Distribution of data Histograms of the three parameters (P-type amylase, S-type amylase and IRT) for the different age groups are shown in Fig. 2. Groups 1 and 2 were combined when testing for skewness. Significant positive skew was found for IRT values in all age groups. P-type amylase values were also positively skewed with the exception of the age group aged 2-4.9 years, in which values appeared normally distributed. Significance of skew is indicated in Fig. 2. Skewness was removed after logarithmic transformation of data with the exception of IRT values for the 8-13 age group. Significant skew (p < 0.01) and kurtosis (p < 0.01) was evident for IRT distribution in this group before and after logarithmic transformation. Salivary amylase values showed a normal distribution in all age groups. Reference ranges The mean and reference ranges within age groups for each of the enzymes is shown in Table II. Where positive skew in distribution was decreased by logarithmic transformation, the reference values quoted were derived from the log values. Variation of P-type amylase with age Fig. 3 shows plasma P-type amylase values obtained for our 166 control subjects between the ages of 0.1 and 13 years. Values averaged 20 U/l for children of mean age 2.3 _+0.7 months and increased gradually with values approaching (B)
(A)
,oo ,~... ......._.__._.__.. //--
0
,oo
80
80
2
. .
0-w
o-1, 25
250 Hepann Fig. in
1. the
amylase
Effect
of
presence per
test).
increasing (G)
and
concentrations
absence
(0)
.
of
(U/ml
serum
. .
.
.
.
l
.._........___.....
50
250
)
of heparin added
. ..-
.o-o--._o d-0
inhibitor.
on salivary (25
mU
(A)
and
salivary
pancreatic amylase
(B) and
amylase 28
mU
activities pancreatic
268 TABLE
I
DISTRIBUTION Age
OF
range
SEXES Mean
OF
age
IIOSPIT.4L
CONTROL
0.8
0.31
f 0.19
1 (i
0.9-
1.9
1.28
* 0.28
17
2
--
3.14
* 0.74
32
5
-7.9
6.0
* 0.92
42
8
~13
10.2
+ 1.36
59
AGE: 0.1
AGI.:
GROl’t’
5 9: 8 20/12 21/21 36/2:<
S-TYPEAMYLASE-•
IRT-l
0.8 YRS (16:
%hl.,
dn . .d
lg&g&_ 20 40 60 80
GROUP 2
TO
11,’
P-TYPEAMYLASE-• GROUP 1
ACCORDING
‘\
f S.D.
O.l-
4.9
CIIII,DREN
0 20 40
AGE: 0.9
-mm
60
i-i00 200 300 400
1,9 YRS !17i
0
II__
30 60 90 120
GROUP 3 12 r
:
:
8
2
4
vl
0
AGE: 2
__I
0 100 200 300 400 500 600
4.9 YRS (321
h 60
GROUP 4
90
& 0
120
AGE: 5
12
z4 2
60 90 120 150
. .
30
:
L 30
Jb
md11_1
30
60
90
O- 100 200 300 400 500 600
120 150
7.9 YRS (423 l
8 IIilamm 0 30 60 90 120 150
GROUP 5
AGE, 8
13
0 30
60 90 120 150 180 1521
YRS (59:
20 16
IRT(*g/11
AMYLASE (U/I) Fig. of
2.
Distribution
hospital
Enzyme
control
values
in parentheses.
for
of plasma children. groups
P-type
amylase,
Asterisks 1 and
2 were
denote
S-type
amylase
significant
combined
when
and immunorractive
skew testing
in distribution for
skewness.
trypsin (* p -< 0.05; Number
in a populatlorl **
P ‘
of subjects
0.01). is given
269 TABLE
II
MEANS AND REFERENCE RANGES FOR P-TYPE AMYLASE. REACTIVE TRYPSIN IN PLASMA OF CONTROL CHILDREN Age range (yrs)
S-TYPE
AhlYLASE
AND
IMMLJNO-
p-type amYlase fUrI) -
s-type amYlase (U/l)
IRT Wail)
mean 01)
reference range
mean (II)
reference range
mean (n)
reierence range e
26 b (16) 44 b (17) 61 IJ (32) 68.5 (42) 74 (59)
0- 64 11-124” 13- 89 22-169 a 31-l 51 a
(16) 19 c (17) 79 (32) 70 99.5 d (42) (58) 86
o- 45 7-I 51 I-140 3-196 5-l 66
186 169 209 192 198
91--340a 45--457 a 51-578 a 66G.454 a g&34*
0.060.8 0.9 - 1 .!f 2 ~ 4.9 5 - 7.9 8 -13
(16) (17) (24) (42) (52)
Reference ranges are from minus to plus 2 standard deviations from mean. a Data deriwd from logarithmic transformation where this decreased positive skew. b Significant difference 0, < 0.0005) in means when compared with oldest age group (8-13 c Significant difference @ .< 0.0005) in means when compared with all older age groups. d SignificantI?: higher (p < 0.005) than 24.9 year group. ’ Overall reference range for plasma IRT values in children (0.1-13 yrs) is 70-434 @g/l.
Yrs).
a plateau at 5-7.9 years. Maximum adult levels were shown by the S-10.9 and 11-13 year groups. Hence these two groups were pooled for statistical purposes. Significance values obtained from comparison of means are indicated in Table II. Mean P-type amylase levels in the 0.9-1.9 year group were not significantly different from those of the 2-4.9 year group. Hence a reference range of 11-124 U/l could be taken to cover the age range O-9-4.9 years. Linear regression analysis showed a significant positive correlation between age vs. plasma P-type amylase (r = 0.46; P < 0.001) from 0.1-7.9 years (Table III). No significant correlation with age was evident from 8-13 years. S-type amylase variation with age Plasma S-type amylase activity
160
was also low (19 * 13 U/l) at 2.3 4 0.7
1
140
l
-
; 2
1
l
120
l
i
$4EO-
T.
.
z
g
ao-
,x ic
l
T.. :
.
60-
.
.= . t: : .:. .
_ 0.8 -
39
4 -10
O,Q-I,9
2-
4.9
* MOllU?S
5-
?.Q
B - 70.9
11 -
13
IYeOrS AGE
Fig. 3. Plasma pancreatic-type
amylase levels (mean
f 1 S.D.) in control children
of different
age groups.
270
260 X
240
..*i. .:. :* ill
2.x 200 2 . J
160
$ -0
160
; 0
140
01
.. . ? . .
..
l%
.
lx!
0.
R ‘; VI
100
z
60
“O ii
60
.;
:: $ =t.. .. ‘.
40
.
:
20 0 0,s -3,9 L
4 -
10
* Months
0,9 -1,9 I L
2 -4.9
5 -7,9
8 -13
i
Y Years
AGE Fig.
Plasma
4.
Point
X not
salivary-type
included
amylase
levels
(mean
tl
S.D.)
in
control
children
of
different
age
groups.
in statistics.
months but increased sharply to reach 79 5 36 U/l by l-2 years. There was a slight rise thereafter with levels reaching a plateau by age 5-6 (Fig. 4). Significant positive correlation (r = 0.51; P < 0.001) was found for age vs. plasma correlation S-type amylase from 0.1-5.9 years. There was no significant between 6 and 13 years (Table III). Plasma IR T values Fig. 5 shows plasma IRT values for 151 analysis showed no significant correlation
TABLE
III
CORRELATIONS S-TYPE
BETWEEN
AMYLASE
AND
AGE
IRT
AND
CIRCULATING
IN CONTROL Age
CONCENTRATIONS
OF
CHILDREN
range
(yrs) Age
control subjects. Linear regression with age (r = 0.025). There was,
Carrel.
Significance
coeff.
P
0.46
*
vs. P-type
arnylase
O.l-
p-type
arnylase
8
s-type
arnylase
O.l-
s-type
amy1ase
6
IRT p-type
7.9 -13 5.9
4.15 8
N.S.
0.025
N.S.
0.44
vs.
IRT
* Identical
0.51
-13
O.lamylase
4.16
0.1-13
p values
were
obtained
using
logarithmic
transformation
of skewed
data
P-TYPE
AMYLASE.
271
0.8-3.9 \
4 --x,
Y Months
I ,
O,Q-$9
2 -4,9
5-7.9
0-IQ9
M-13
” Years
/
AGE Fig.
5. Plasma
immunoreactive
trypsin
levels
(mean
iI
SD.)
in control
children.
however, a significant positive correlation between plasma P-type amylase and IRT concentrations (r = 0.44; P < 0.001; Table III). Discussion Since heparin has been reported to precipitate amylase from saliva and serum of cystic fibrosis patients [S], the effect of heparin on the assay of both pancreatic and salivary amylase by the inhibitor method was studied. It is apparent from the results that heparin, when added to normal human serum in concentrations equivalent to those obtained in heparinized blood samples, had no effect on amylase assay by the inhibitor method. We also found that heparin in similar concentrations had no effect on amylase activity of serum from cystic fibrosis patients (results not shown). Since measurement of P-type amylase is likely to be used in detecting pancreatic malfunction, particularly in young children, it was essential to establish reference ranges for the various age groups (Table II). We confirmed the variation with age of circulating P-type amylase levels reported by Skude [5]. We found some very low values (<20 U/l) in the first four months of life. The value reached by age 8-11 (mean 73 U/l) was not significantly different from our adult male level (82 U/l) previously reported [ 11. These findings agree very well with those of Skude [ 51, who used an electrophoretic method of analysis. Salivary amylase levels were also very low up to 10 months (mean 19 U/l). Linear regression analysis indicated that maximum levels were reached by age 5-5.9 years. This is also in agreement with the results of Skude [ 51. We found some very low S-type amylase levels in all age groups (
272
adults [l]. A possible explanation for this is the fact that the adults studied were healthy ambulant subjects, in contrast to the hospitalised control children. It is also possible that circulating S-type amylase levels decline with advancing age. The majority of our adults were in the age range 40-50 years. Interestingly a small group of 16 ambulant teenage subjects showed significantly lower serum S-type amylase values (67 _+45 U/l) than the 5-13 year hospital in-patients (results not shown). We found no sex-related difference for P or S-type amylase levels in any of the age groups studied unlike the slight but significantly higher mean values for P-type amylase shown by adult females [l]. Reference ranges for serum pancreatic amylase in adults are 37-162 li,il (males) and 52-175 U/l (females). Corresponding ranges for serum S-type amylase are 9-172 U/l (males) and 9-196 U/l (females). These values were calculated following logarithmic transformation of skewed data [ 11. We found that plasma IRT levels did not vary over the age range studied (0.1-13 years). Our reference ranges for the entire group obtained after logarithmic transformation (70-434 pg/l; II = 151) are much wider than those proposed for adults by this method (150-400). It is apparent from inspection of values in Fig. 5 that many children in the various groups have plasma IRT levels less than 120 pg/l. The absence of properly established reference ranges for younger age groups may lead to some confusion in the investigation of pancreatic disease in children [ 9,101. In addition, the specificity of this assay for trypsin and its inhibited forms as claimed by Stagg and Wood [ 111 remains to be clarified. The normal levels obtained are 10 times higher than the values obtained by Geokas et al. [ 121, who use an antibody specific for cationic trypsin and trypsinogen. Until the reasons for these differences are elucidated the value of the Hoechst IRT assay for the specific measurement of pancreatic function remains in doubt. The reference ranges established for plasma P-type amylase should be useful in the diagnosis of pancreatic insufficiency in children over one year. Agematching is essential as is evident from the present results. Detection of decreased pancreatic function in children less than one year old would not be possible by this method, since normal children of this age have barely detectable levels of circulating pancreatic amylase. Acknowledgements Thanks are due to Mrs. B. McNulty of the National Children’s Hospital, Dublin, for collecting blood samples. Plasma immunoreactive trypsin kits were supplied by Hoechst Ltd. (U.K.) and the assistance of Dr. R.H. Rouse11 of Hoechst is acknowledged. This work was supported in part by the Cystic Fibrosis Association of Ireland. References 1
O’Donnell,
2
Hegaty.
J.E.,
M.D.,
3
Groarke,
J.. O’Donnell,
4
Kenny,
5
Skude.
D..
FitzGerald,
O’Donnell.
Cooke,
G. (1975)
A., Stand.
0.
M.D., M.D.,
and McGeeney,
K.F.
(1977)
McGeeney,
and
FitzGerald,
McGeeney,
Tempany,
E. and
,I. Gastroenterol.
K.Y. K.F.
and
FitzGerald.
McGeeney, 10,
Clin.
577-584
K.F.
(1978)
Chem. 0.
0.
23,
(1978)
(1980) Clin.
560-566 Gut
19,
35O--354
Ir. J. Med.
Chim.
Acta
89,
Sci.
(in Press)
429.-433
6
Otsuki,
7
Snedrcar, Univ.
M.,
Saeki.
S.,
G.W.
and
H..
Maeda,
M. and
W.G.
Baba.
(1967)
in
S. (1976)
Statistical
Clin. Methods.
Chrm. 6th
22.
439-444
ed.,
PP.
86-89,
Iowa
State
Press
8
Doggett.
R.F.
9
Crossley.
J.R..
Elliott.
P.,
Hodson.
10
Dandona.
11
Stagg,
12
Geokas, E83
Yuu,
Cochranr,
B.H.
and
and
M.C..
Harrison.
Wood, Largman.
G.M.
R.B. hf.,
and Bell.
T.P. C.,
(1973) Smith,
J..
Nature P.A.
Ramdial.
(1979) Brodrick,
Ann.
(London)
(1979) L. and
Clin. J.W.
1,
Batten,
Biochem. and
New
Lancet
Johnson,
J.C. 16.
Biol.
243.
251--252
472-474 (1979)
Lancet
1,
1032
147m-151 J.H.
(1979)
Am.
.J. Phvsiol.
236
(1).
E77
Clinica Chimica Acta, 104 (1980) 265-273 @ Elsevier/North-Holland Biomedical Press
CCA
1385
PLASMA PANCREATIC AND SALIVARY-TYPE AMYLASE AND IMMUNOREACTIVE TRYPSIN CONCENTRATIONS: VARIATIONS WITH AGE AND REFERENCE RANGES FOR CHILDREN
MARY
D. O’DONNELL
a+* and N.J. MILLER
b
a Department of Medicine and Therapeutics, University College Dublin b Central Biochemistry Laboratory, Sir Patrick Dun’s Hospital, Dublin, (Received
November
13th,
and (Republic
of Ireland)
1979)
Summary The differential ar-amylase (EC 3.2.1.1) assay was applied to 166 control children in the age range 0.1-13 years. Circulating levels of both pancreatic and salivary-type amylase were very low (mean 20 U/l) in the first four months of life. Pancreatic levels increased gradually with age, reaching adult levels (mean 74 U/l) by the age of eight years. Salivary amylase levels showed a sharp rise in the 0.9-1.9 year period reaching maximum levels (mean 99 U/l) by age 5-6 years. While linear regression analysis showed significant correlation between age of subject and pancreatic and salivary amylase levels, no such correlation was evident between age and plasma immunoreactive trypsin levels over the age range studied. Plasma trypsin levels in children were lower than reported adult values. Reference ranges for pancreatic and salivary-type amylase and immunoreactive trypsin in children are presented. The importance of age-matching, when pancreatic amylase and plasma trypsin are being investigated in children, is emphasised.
Introduction The specific measurement of pancreatic (P) amylase in the presence of salivary (S)-type amylase in adult human serum [l], which requires no electrophoresis or chromatography, has been described. The method utilises an inhibitor protein with a specificity 100 times greater for human salivary than for human pancreatic amylase. Its use in the investigation of chronic pancreatitis and pancreatic insufficiency has been described [2,3]. Measurement of
*To
whom
correspondence shouldbeaddressed.
P-type amylase by this method also appears promising for the diagnosis of pancreatic insufficiency in children without the need for duodenal intubations [ 41. Since pancreatic and salivary amylase levels in children vary with age [ 5,6], it was desirable to study the normal variation with age for these enzymes using the inhibitor method and to establish reference values for children of different age groups. Plasma immunoreactive trypsin (IRT) values were also studied simultaneously, since reference values for children have not yet been established. Methods Subjects Control subjects were hospital in-patients ranging in age from 3 weeks to 13 years. No patient had any evidence of pancreatic or parotid gland disease, abdominal, renal or respiratory dysfunction or were receiving drug therapy. In all, there were 97 males and 69 females. Samples Non-fasting blood samples, 3-10 ml, were placed the plasma immediately separated after centrifugation
in heparinized tubes and and stored at -20°C.
Analyses Pancreatic and salivary-type amylase were determined by the method of O’Donnell et al. [ 11 and IRT by the radioimmunoassay kit (RIA-gnost Trypsin, Behringwerke-AG, Marburg, F.R.G.). Study of the effect of heparin on amylase assay In view of the presence of heparin in the plasma samples, it was important to establish at the outset the effect, if any, of added heparin on (a) P and S amylase activities and (b) the action of the inhibitor. This experiment was carried out as follows: S and P amylase and amylase-free serum were prepared as described previously [l]. S and P amylase were added separately to 0.2-ml samples of amylase-free serum. Heparin (Pullarin, 1000 U/ml, Evans Medical Ltd., U.K.) in the range 2-50 units was then added to enzyme-serum mixtures and tubes were incubated for 16 h at 4°C. They were analysed then for enzyme activities in the presence and absence of added inhibitor to the pre-incubation medium [ 11. Calculations We calculated the mean, standard deviation, skew and kurtosis for P-type amylase, S-type amylase and IRT for subjects in the various age groups by standard methods [7]. Reference ranges were calculated for each group as the 95% range from minus to plus two standard deviations from the mean. In cases where significant positive skew in distribution of data was found these ranges were calculated after logarithmic transformation of data. We sought linear correlations between age and concentrations of enzymes by regression analysis combined with analysis of variance. Differences between means were compared by the non-paired Student’s t test (n > 20) and by the Mann-Whitney U test (n < 20).
267
Results The effect of heparin (10250 U/ml serum) on salivary and pancreatic amylase activities are shown in Fig. 1. No change in enzyme activities was evident. Heparin was also shown to have no effect on the inhibition of either S or P-type amylase by the inhibitor. Table I shows the characteristics of the subjects according to their age group. No sex-related differences were noted for any of the enzymes studied in any age category. Hence, results for males and females were grouped together. Distribution of data Histograms of the three parameters (P-type amylase, S-type amylase and IRT) for the different age groups are shown in Fig. 2. Groups 1 and 2 were combined when testing for skewness. Significant positive skew was found for IRT values in all age groups. P-type amylase values were also positively skewed with the exception of the age group aged 2-4.9 years, in which values appeared normally distributed. Significance of skew is indicated in Fig. 2. Skewness was removed after logarithmic transformation of data with the exception of IRT values for the 8-13 age group. Significant skew (p < 0.01) and kurtosis (p < 0.01) was evident for IRT distribution in this group before and after logarithmic transformation. Salivary amylase values showed a normal distribution in all age groups. Reference ranges The mean and reference ranges within age groups for each of the enzymes is shown in Table II. Where positive skew in distribution was decreased by logarithmic transformation, the reference values quoted were derived from the log values. Variation of P-type amylase with age Fig. 3 shows plasma P-type amylase values obtained for our 166 control subjects between the ages of 0.1 and 13 years. Values averaged 20 U/l for children of mean age 2.3 _+0.7 months and increased gradually with values approaching (B)
(A)
,oo ,~... ......._.__._.__.. //--
0
,oo
80
80
2
. .
0-w
o-1, 25
250 Hepann Fig. in
1. the
amylase
Effect
of
presence per
test).
increasing (G)
and
concentrations
absence
(0)
.
of
(U/ml
serum
. .
.
.
.
l
.._........___.....
50
250
)
of heparin added
. ..-
.o-o--._o d-0
inhibitor.
on salivary (25
mU
(A)
and
salivary
pancreatic amylase
(B) and
amylase 28
mU
activities pancreatic
268 TABLE
I
DISTRIBUTION Age
OF
range
SEXES Mean
OF
age
IIOSPIT.4L
CONTROL
0.8
0.31
f 0.19
1 (i
0.9-
1.9
1.28
* 0.28
17
2
--
3.14
* 0.74
32
5
-7.9
6.0
* 0.92
42
8
~13
10.2
+ 1.36
59
AGE: 0.1
AGI.:
GROl’t’
5 9: 8 20/12 21/21 36/2:<
S-TYPEAMYLASE-•
IRT-l
0.8 YRS (16:
%hl.,
dn . .d
lg&g&_ 20 40 60 80
GROUP 2
TO
11,’
P-TYPEAMYLASE-• GROUP 1
ACCORDING
‘\
f S.D.
O.l-
4.9
CIIII,DREN
0 20 40
AGE: 0.9
-mm
60
i-i00 200 300 400
1,9 YRS !17i
0
II__
30 60 90 120
GROUP 3 12 r
:
:
8
2
4
vl
0
AGE: 2
__I
0 100 200 300 400 500 600
4.9 YRS (321
h 60
GROUP 4
90
& 0
120
AGE: 5
12
z4 2
60 90 120 150
. .
30
:
L 30
Jb
md11_1
30
60
90
O- 100 200 300 400 500 600
120 150
7.9 YRS (423 l
8 IIilamm 0 30 60 90 120 150
GROUP 5
AGE, 8
13
0 30
60 90 120 150 180 1521
YRS (59:
20 16
IRT(*g/11
AMYLASE (U/I) Fig. of
2.
Distribution
hospital
Enzyme
control
values
in parentheses.
for
of plasma children. groups
P-type
amylase,
Asterisks 1 and
2 were
denote
S-type
amylase
significant
combined
when
and immunorractive
skew testing
in distribution for
skewness.
trypsin (* p -< 0.05; Number
in a populatlorl **
P ‘
of subjects
0.01). is given
269 TABLE
II
MEANS AND REFERENCE RANGES FOR P-TYPE AMYLASE. REACTIVE TRYPSIN IN PLASMA OF CONTROL CHILDREN Age range (yrs)
S-TYPE
AhlYLASE
AND
IMMLJNO-
p-type amYlase fUrI) -
s-type amYlase (U/l)
IRT Wail)
mean 01)
reference range
mean (II)
reference range
mean (n)
reierence range e
26 b (16) 44 b (17) 61 IJ (32) 68.5 (42) 74 (59)
0- 64 11-124” 13- 89 22-169 a 31-l 51 a
(16) 19 c (17) 79 (32) 70 99.5 d (42) (58) 86
o- 45 7-I 51 I-140 3-196 5-l 66
186 169 209 192 198
91--340a 45--457 a 51-578 a 66G.454 a g&34*
0.060.8 0.9 - 1 .!f 2 ~ 4.9 5 - 7.9 8 -13
(16) (17) (24) (42) (52)
Reference ranges are from minus to plus 2 standard deviations from mean. a Data deriwd from logarithmic transformation where this decreased positive skew. b Significant difference 0, < 0.0005) in means when compared with oldest age group (8-13 c Significant difference @ .< 0.0005) in means when compared with all older age groups. d SignificantI?: higher (p < 0.005) than 24.9 year group. ’ Overall reference range for plasma IRT values in children (0.1-13 yrs) is 70-434 @g/l.
Yrs).
a plateau at 5-7.9 years. Maximum adult levels were shown by the S-10.9 and 11-13 year groups. Hence these two groups were pooled for statistical purposes. Significance values obtained from comparison of means are indicated in Table II. Mean P-type amylase levels in the 0.9-1.9 year group were not significantly different from those of the 2-4.9 year group. Hence a reference range of 11-124 U/l could be taken to cover the age range O-9-4.9 years. Linear regression analysis showed a significant positive correlation between age vs. plasma P-type amylase (r = 0.46; P < 0.001) from 0.1-7.9 years (Table III). No significant correlation with age was evident from 8-13 years. S-type amylase variation with age Plasma S-type amylase activity
160
was also low (19 * 13 U/l) at 2.3 4 0.7
1
140
l
-
; 2
1
l
120
l
i
$4EO-
T.
.
z
g
ao-
,x ic
l
T.. :
.
60-
.
.= . t: : .:. .
_ 0.8 -
39
4 -10
O,Q-I,9
2-
4.9
* MOllU?S
5-
?.Q
B - 70.9
11 -
13
IYeOrS AGE
Fig. 3. Plasma pancreatic-type
amylase levels (mean
f 1 S.D.) in control children
of different
age groups.
270
260 X
240
..*i. .:. :* ill
2.x 200 2 . J
160
$ -0
160
; 0
140
01
.. . ? . .
..
l%
.
lx!
0.
R ‘; VI
100
z
60
“O ii
60
.;
:: $ =t.. .. ‘.
40
.
:
20 0 0,s -3,9 L
4 -
10
* Months
0,9 -1,9 I L
2 -4.9
5 -7,9
8 -13
i
Y Years
AGE Fig.
Plasma
4.
Point
X not
salivary-type
included
amylase
levels
(mean
tl
S.D.)
in
control
children
of
different
age
groups.
in statistics.
months but increased sharply to reach 79 5 36 U/l by l-2 years. There was a slight rise thereafter with levels reaching a plateau by age 5-6 (Fig. 4). Significant positive correlation (r = 0.51; P < 0.001) was found for age vs. plasma correlation S-type amylase from 0.1-5.9 years. There was no significant between 6 and 13 years (Table III). Plasma IR T values Fig. 5 shows plasma IRT values for 151 analysis showed no significant correlation
TABLE
III
CORRELATIONS S-TYPE
BETWEEN
AMYLASE
AND
AGE
IRT
AND
CIRCULATING
IN CONTROL Age
CONCENTRATIONS
OF
CHILDREN
range
(yrs) Age
control subjects. Linear regression with age (r = 0.025). There was,
Carrel.
Significance
coeff.
P
0.46
*
vs. P-type
arnylase
O.l-
p-type
arnylase
8
s-type
arnylase
O.l-
s-type
amy1ase
6
IRT p-type
7.9 -13 5.9
4.15 8
N.S.
0.025
N.S.
0.44
vs.
IRT
* Identical
0.51
-13
O.lamylase
4.16
0.1-13
p values
were
obtained
using
logarithmic
transformation
of skewed
data
P-TYPE
AMYLASE.
271
0.8-3.9 \
4 --x,
Y Months
I ,
O,Q-$9
2 -4,9
5-7.9
0-IQ9
M-13
” Years
/
AGE Fig.
5. Plasma
immunoreactive
trypsin
levels
(mean
iI
SD.)
in control
children.
however, a significant positive correlation between plasma P-type amylase and IRT concentrations (r = 0.44; P < 0.001; Table III). Discussion Since heparin has been reported to precipitate amylase from saliva and serum of cystic fibrosis patients [S], the effect of heparin on the assay of both pancreatic and salivary amylase by the inhibitor method was studied. It is apparent from the results that heparin, when added to normal human serum in concentrations equivalent to those obtained in heparinized blood samples, had no effect on amylase assay by the inhibitor method. We also found that heparin in similar concentrations had no effect on amylase activity of serum from cystic fibrosis patients (results not shown). Since measurement of P-type amylase is likely to be used in detecting pancreatic malfunction, particularly in young children, it was essential to establish reference ranges for the various age groups (Table II). We confirmed the variation with age of circulating P-type amylase levels reported by Skude [5]. We found some very low values (<20 U/l) in the first four months of life. The value reached by age 8-11 (mean 73 U/l) was not significantly different from our adult male level (82 U/l) previously reported [ 11. These findings agree very well with those of Skude [ 51, who used an electrophoretic method of analysis. Salivary amylase levels were also very low up to 10 months (mean 19 U/l). Linear regression analysis indicated that maximum levels were reached by age 5-5.9 years. This is also in agreement with the results of Skude [ 51. We found some very low S-type amylase levels in all age groups (
272
adults [l]. A possible explanation for this is the fact that the adults studied were healthy ambulant subjects, in contrast to the hospitalised control children. It is also possible that circulating S-type amylase levels decline with advancing age. The majority of our adults were in the age range 40-50 years. Interestingly a small group of 16 ambulant teenage subjects showed significantly lower serum S-type amylase values (67 _+45 U/l) than the 5-13 year hospital in-patients (results not shown). We found no sex-related difference for P or S-type amylase levels in any of the age groups studied unlike the slight but significantly higher mean values for P-type amylase shown by adult females [l]. Reference ranges for serum pancreatic amylase in adults are 37-162 li,il (males) and 52-175 U/l (females). Corresponding ranges for serum S-type amylase are 9-172 U/l (males) and 9-196 U/l (females). These values were calculated following logarithmic transformation of skewed data [ 11. We found that plasma IRT levels did not vary over the age range studied (0.1-13 years). Our reference ranges for the entire group obtained after logarithmic transformation (70-434 pg/l; II = 151) are much wider than those proposed for adults by this method (150-400). It is apparent from inspection of values in Fig. 5 that many children in the various groups have plasma IRT levels less than 120 pg/l. The absence of properly established reference ranges for younger age groups may lead to some confusion in the investigation of pancreatic disease in children [ 9,101. In addition, the specificity of this assay for trypsin and its inhibited forms as claimed by Stagg and Wood [ 111 remains to be clarified. The normal levels obtained are 10 times higher than the values obtained by Geokas et al. [ 121, who use an antibody specific for cationic trypsin and trypsinogen. Until the reasons for these differences are elucidated the value of the Hoechst IRT assay for the specific measurement of pancreatic function remains in doubt. The reference ranges established for plasma P-type amylase should be useful in the diagnosis of pancreatic insufficiency in children over one year. Agematching is essential as is evident from the present results. Detection of decreased pancreatic function in children less than one year old would not be possible by this method, since normal children of this age have barely detectable levels of circulating pancreatic amylase. Acknowledgements Thanks are due to Mrs. B. McNulty of the National Children’s Hospital, Dublin, for collecting blood samples. Plasma immunoreactive trypsin kits were supplied by Hoechst Ltd. (U.K.) and the assistance of Dr. R.H. Rouse11 of Hoechst is acknowledged. This work was supported in part by the Cystic Fibrosis Association of Ireland. References 1
O’Donnell,
2
Hegaty.
J.E.,
M.D.,
3
Groarke,
J.. O’Donnell,
4
Kenny,
5
Skude.
D..
FitzGerald,
O’Donnell.
Cooke,
G. (1975)
A., Stand.
0.
M.D., M.D.,
and McGeeney,
K.F.
(1977)
McGeeney,
and
FitzGerald,
McGeeney,
Tempany,
E. and
,I. Gastroenterol.
K.Y. K.F.
and
FitzGerald.
McGeeney, 10,
Clin.
577-584
K.F.
(1978)
Chem. 0.
0.
23,
(1978)
(1980) Clin.
560-566 Gut
19,
35O--354
Ir. J. Med.
Chim.
Acta
89,
Sci.
(in Press)
429.-433
6
Otsuki,
7
Snedrcar, Univ.
M.,
Saeki.
S.,
G.W.
and
H..
Maeda,
M. and
W.G.
Baba.
(1967)
in
S. (1976)
Statistical
Clin. Methods.
Chrm. 6th
22.
439-444
ed.,
PP.
86-89,
Iowa
State
Press
8
Doggett.
R.F.
9
Crossley.
J.R..
Elliott.
P.,
Hodson.
10
Dandona.
11
Stagg,
12
Geokas, E83
Yuu,
Cochranr,
B.H.
and
and
M.C..
Harrison.
Wood, Largman.
G.M.
R.B. hf.,
and Bell.
T.P. C.,
(1973) Smith,
J..
Nature P.A.
Ramdial.
(1979) Brodrick,
Ann.
(London)
(1979) L. and
Clin. J.W.
1,
Batten,
Biochem. and
New
Lancet
Johnson,
J.C. 16.
Biol.
243.
251--252
472-474 (1979)
Lancet
1,
1032
147m-151 J.H.
(1979)
Am.
.J. Phvsiol.
236
(1).
E77