From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
Hemoglobin Level Is Linked to Growth Hormone-Dependent Proteins in
Short Children
By Elina Vihervuori, Martti Virtanen, Hannu Koistinen, Riitta Koistinen, Markku Seppala, and Martti A. Siimes
Erythropoiesis was investigated in 32 children with short
stature and in eight children with skeletal dysplasia by
studying blood hemoglobin in relation to growth and to serum concentrationsof insulin-likegrowth factor I (IGF-ll, IGF
binding protein-3 (IGFBP-31, and erythropoietin(EPOIbefore,
during, and after 12 months of recombinant human growth
hormone (GH) treatment. Blood hemoglobin concentration
was positively correlated with relative body height and with
serum IGF-l and IGFBP-3 levels ( P = .001 to .02), but not with
the concentrations of EPO. The normal age-dependency of
hemoglobin was lacking. Hemoglobin levels and their responses to GH treatment were similar in the patients with
GH deficiency and those with normal GH secretion. Treat-
ment with GH acceleratedgrowth and elevated the concentrations of hemoglobin, IGF-I, and IGFBP-3. In the eight patients with skeletal dysplasia, body mass increased similarly,
but gain in height was l e u than in the other patients, and
the increase in hemoglobin was markedly pronounced. In
this group, the correlationsbetween hemoglobin, IGF-I, and
IGFBP-3 were extremely close (r = 0.80 to 0.85, P = .031 to
.008).These findings are in accord with earlier observations
from in vitro and animal studies, and suggest that the GHIGF axis is involved in the physiologicelevation of hemoglobin levels during childhood.
0 1996 by The American Society of Hematology.
G
binding proteins, IGF binding protein-3 (IGFBP-3).18 While
GH has a short half-life and its secretion is pulsatile,’’ IGFI and IGFBP-3 show little, if any, diurnal variation?’”’ Of
the six currently known IGF-binding proteins, the most abundant is IGFBP-3,22regulated by GH23and IGF-I.18*24
The present study was designed to determine the relations
between GH, growth, and erythropoiesis in humans. For this
purpose we investigated the associations between hemoglobin, EPO, IGF-I, IGFBP-3, and body growth during GH
treatment of 40 short children.
ROWTH HORMONE (GH) and its principal regulator,
insulin-like growth factor I (IGF-I), have been demonstrated to stimulate erythropoiesis under various in vitro conditions’.’’ and in animal ~tudies.’”’~
In man, the view that
GH plays a role in hematopoiesis is complicated by observations that individuals with GH deficiency are generally not
anemic. With respect to erythropoiesis, the effects of treatment with recombinant human GH have not been thoroughly
studied.
The mechanism linking increasing erythrocyte mass with
body growth is poorly understood. Erythropoietin (EPO), the
primary substance regulating erythropoiesis, was formerly
believed to increase during growth as a consequence of renal
hypoxia related to kidney growth.’’-’2However, experiments
on neonatal mice indicated that prevention of hypoxia by
hypertransfusion does not totally suppress erythropoiesis
during growth,” although it does so in adult mice.” Moreover, in rats, serum EPO decreased during growth, while
the increasing concentrations of serum IGF-I were linearly
correlated with total incorporation of iron into erythrocyte^.'^
These findings suggest that factors related to growth itself
are invoIved.’’-’3.’6
In vitro experiments have shown that physiologic concentrations of IGF-I stimulate human and murine erythroid prewhereas large
cursors, whether EPO is present or
In an experidoses of GH are required for the same
ment with murine erythroid colony-forming units,’ the relative effect of IGF-I was strongest in the absence of EPO and,
with increasing concentrations of EPO, the effect became
weaker. Furthermore, administration of IGF-I to neonatal
rats led to accelerated red blood cell production.‘4In a study
on hypophysectomized rats, both human GH and recombinant human IGF-I stimulated erythropoiesis and increased
the serum EPO concentration.” However, the stimulatory
effect on erythropoiesis occurred before the serum EPO levels had increased. In uremic patients with anemia, the serum
EPO concentration was within the normal range, but the
serum IGF-I level was elevated.” In these patients, IGF-I
was positively correlated with hematocrit values, but serum
EPO was not.
Treatment of short children with GH offered an opportunity to study the effects of GH on erythropoiesis in humans.
GH stimulates the circulating levels of IGF-I and one of its
Blood, Vol 87, No 5 (March 1). 1996: pp 2075-2081
MATERIALS AND METHODS
Subjects. Fifty-one children with short stature were given GH
medication between January 1992 and May 1993 at the Children’s
Hospital, University of Helsinki, or at the Aurora Hospital, Helsinki,
Finland. They were followed up during the first year of treatment.
The children with chronic medication other than GH were excluded,
leaving 20 boys and 20 girls for the study group. None of the patients
had hematologic abnormalities nor had they any treatable diseases
such as hypothyroidism or celiac disease. The patients’ age ranged
from 1.6 to 13.3 years, with a mean of 7.0 years (Table 1). Seven
patients had signs of early puberty during the study year. Two of
them were boys. At the end of the follow-up period, they were at
stage P2G2 according to the criteria of Tanner.*’
In eight children, the cause for their short stature was severe
skeletal dysplasia. In the other 32 short children, endogenous GH
secretion was evaluated by GH stimulation tests” or overnight sampling analysis. Maximal plasma GH concentrations formed a continuity from very low serum values to normal values (Fig 1). The
From the Children’s Hospital and the Department of Obstetrics
and Gynecology, University of Helsinki, Helsinki, Finland.
Submitted July IO, 1995; accepted October 16, 1995.
Supported by the Foundation for Pediatric Research, Helsinki,
the Nordisk Forsknings Komitt!, the University of Helsinki, and the
Academy of Finland.
Address reprint requests to Elina Vihervuori, MO, Children’s
Hospital, University of Helsinki, Stenbackinkatu I I , FIN-00290 Helsinki, Finland.
The publication costs of this article were defrayed in part by page
charge payment. This article must therefore be hereby marked
“advertisement” in accordance with 18 U.S.C. section 1734 solely to
indicate this fact.
0 1996 by The American Society of Hematology.
0006-4971/96/8705-~28$3.00/0
2075
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
VIHERVUORI ET AL
2076
Table 1. Baseline Characteristics of the Short Children. Mean (range)
All
N
Female/male
Age at start (yr)
Relative height (SD units)
Growth velocity (cm/yr)
Relative weight (96 of normal)
Hemoglobin (g/dL)
Hemoglobin (SD units)
Serum IGF-I (nmol/L)
Serum IGF-I (SD units)
Serum IGFBP-3 (ng/mL)
Serum IGFBP-3 (SD units)
Serum erythropoietin (mU/mL)
40
20/20
7 (1.6-13.3)
-3.3 (-6.9--1.1)
5.6 (2.8-10.3)
104 (68-159)
12.55 (10.5-14.5)
-0.6 (-3.4-1.9)
12.3 (1.0-30.0)
-1.0 (-2.4-1.5)
4222 (1311-7614)
0.1 (-2.1-3.3)
14.9 (4.5-61.0)
Patients With Skeletal
Dysplasia’
Patients With Other Causes
for Short Staturet
8
3/5
8.0 (5.1-10.6)
-5.4 (-6.9--3.9)*
4.0(2.9-5.5)§
134 (117-159)*
11.88 (10.5-13.1
)§
-1.6 (-3.4-0.2)
13.9 (9.0-21.0)
-1.1 (- 1.5--0.5)
4372 (2258-5772)
-0.0 (-2.0-1.5)
17.1 (11.0-30)
32
17/15
6.7(1.6-13.3)
-2.8(-5.2--1.1)
5.9 (2.8-10.3)
97 (68-130)11
12.72 (11.3-14.5)
-0.3 (-3.1-1.9)
12.0 (i.o-3o.o)n
-1.0 (-2.4-1.5)
4184 (1311-7614)n
0.1 (-2.i-3.3)n
14.4 (4.5-61.0)
* Symbols indicate differences from the other 32 patients.
t Symbols indicate differences related to diagnoses tested with ANOVA and the post hoc Student-Neuman-Keulstest.
Significant at P < ,001 (independent samples t-test).
§ Significant at P < .05.
/I Patients with normal G H secretion weighed less than the other patients ( P= .05).
1Patients with GH deficiency had the lowest values ( P< .05).With regard to the other characteristics listed, no diagnosis-related differences
were noted.
*
tests were repeated if values were below 10 ng/mL. None of the
children had complete absence of GH secretion; in the eight children
considered to be GH-deficient, the mean maximum GH response to
stimulation tests was 5.8 (range, 1.0 to 9.9) ng/mL, and on average,
they grew 4.6 (range, 2.8 to 6.0) cm/year. For comparisons, the
patients were divided into four categories (Table 2).
The study was approved by the Ethics Committee of the Children’s Hospital, University of Helsinki. Informed consent was obtained from the children and their parents.
Protocol. The study protocol included physical examination and
blood tests on the day when treatment was started, and at 3, 6, and
12 months of GH medication. In addition, 23 patients allowed a
blood sample to be taken 1 week after the start of treatment. The
investigations included body weight and height measurements with
a Harpenden stadiometer, and the laboratory tests consisted of measurements of blood hemoglobin, hematocrit, reticulocyte count, and
serum concentrations of ferritin, EPO, IGF-I, and IGFBP-3. The
study protocol was followed strictly in 36 of the 40 patients. In four
I
Patients
Fig 1. Individual growth responses (bars) and maximal values in
GH stimulation tests (0)of the 32 patients. The patients are arranged
in ascending order according to the maximum response in growth
hormone stimulation tests. The lower edga of the simple range bar
represents growth velocity before GH treatment, and the upper edga
represents growth velocity during the first 12 months of GH treatment. The patients with skeletal dysplasia are not included.
patients the protocol was followed for 6 months; three of these had
skeletal dysplasia.
GH treatment. The indication for GH treatment was short stature. The recombinant human GH (Genotropin; Kabi Pharmacia,
Stockholm, Swedeflumatrope; Lilly, Indianapolis, IN/Norditropin;
Novo Nordisk Gentofte, DenmarWSaizen; Serono Nordic, Geneva,
Switzerland) was given as daily subcutaneous injections. The mean
dose was 0.13 (range, 0.09 to 0.27) IU/kg/d or 3.2 (range, 1.8 to
6.5) IU/m2/d.Five patients (one girl with the Tumer syndrome, two
children with prenatal growth retardation, and two children who
were likely GH deficient) were given iron medication (3 mg/kg/d)
during the study because of signs of iron deficiency.*’
Biochemical analyses. Blood hemoglobin concentrations and
hematocrit were measured with a Coulter Counter T 890 (Coulter,
Miami, FL). Serum IGF-I concentration was determined by radioimmunoassay (RIA) (Incstar CO, Stillwater, MN) after acid extraction
chromatography. The sensitivity of the assay was 2 nmol/L, the
intraassay variation 7% to 8%, and the interassay variation 11% to
14%. Serum IGFBP-3 was measured by immunofluorometric
assay.” The serum samples were diluted 1: 100 and the measurement
range covered 0.6 to 650 ng/mL. The intraassay variation was 3.6%
to 6.2% and the interassay variation 5.4% to 11%. EPO concentrations were determined by RIA (Incstar CO).At concentrations greater
than the detection limit of 10 mU/mL, the intraassay coefficient of
variation was 7% to 10% and the interassay variation ranged from
8% to 12%. Reticulocytes were counted with a flow cytometer.”
Serum ferritin was measured by RIA (Ferritin Ria E(lt, Kodak Clinical Diagnostics, Amersham, UK). The total circulating erythrocyte
volume was estimated in mL according to the following formula:
hematocrit x body weight (kg) x proportion of body weight assumed
for blood volume (6.5 mL per kg).30
Datu frunsfonnations. Because the age of the patients ranged
from 2 to 14 years, we transformed all age-dependent values into
standard deviation (SD) units. The transformation of body height
was based on Finnish growth standards” and, consequently, the
transformed height is referred to as relative height. The mean relative
height before treatment was -3.3 SD units (Table 1).
Because in the reference series h e m o g l ~ b i nIGF-1,”
, ~ ~ and IGFBP-
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
GROWTH HORMONE AND ERYTHROPOIESIS
2077
Table 2. Diagnoses of the 40 Study Patients
Potential Etiology of Short Stature
Patient Category
GH deficiency likely (8patients)
GH secretion normal (18 patients)
Borderline findings concerning GH secretion (6 patients)
Skeletal dysplasia (8patients)
Organic GH deficiency as a consequence of cranial irradiation (for treatment of
leukemia/cerebellar glioma): 2 patients
Idiopathic GH deficiency: 6 patients
Small for gestational age: 8 patients
Constitutional growth delay or familiar short stature: 6 patients
Turner syndrome: 2 patients
Russell-Silver syndrome: 2 patients
Partial GH deficiency as a consequence of cranial irradiation (for treatment of
cerebral ependymoma): 1 patient
Turner syndrome: 1 patient
Small for gestational age: 1 patient
Achondroplasia: 7 patients
Spondyloepiphyseal- and metaphyseal dysplasia: 1 patient
3" were age-dependent, these results were also transformed. The
conversion of hemoglobin was based on a large body of FinnishAmerican data.3zThe conversions of IGF-I and IGFBP-3 were based
on the values we obtained from 139 healthy children between 0.2
and 15.9 years of age. The samples had been obtained when the
children visited the general outpatient clinic because of allergy, a
psychosomatic problem, minor surgery, or follow-up of these. The
reference values were calculated for 2-year cohorts. In the transformation procedure, values were interpolated between the mean ages
of the 2-year cohorts.
The EPO values were not converted into SD units, because EPO
concentration remains stable after infancy.16 The distributions of
EPO and ferritin concentrations were skewed, and these data were,
therefore, subjected to logarithmic transformation.
Data presentation and statistical analyses. Data are presented
in the text as means (95% confidence intervals) and, for the sake of
clarity, in the figures as means ? standard error of mean (SEM).
The total ranges of baseline values are given in Table 1. Both simple
and multiple linear regression analyses were performed. We considered the various diagnostic categories, maximal serum value observed in GH stimulation tests, IGF-I, IGFBP-3, EPO, ferritin, age,
sex, puberty, body mass index, relative height, and growth velocity
as independent variables to explain the variation of hemoglobin
level. Linear associations were tested with Pearson's correlation coefficient. To compare the different diagnostic categories, analysis of
variance (ANOVA) and the post hoc Student-Neuman-Keuls test
were used. Values for the patients with skeletal dysplasia and the
32 other patients were compared by the two-tailed independentsamples t test. Paired samples t test was used to test changes in
ferritin level. Differences were considered significant at P < .05.
and negatively associated with age (P < .001). The best
four-variable model (Rz= .61) also included serum ferritin
(P = .060) in the list of variables that had positive associations with hemoglobin. Transformation of hemoglobin into
SD units did not change the results.
Effect of GH on growth and hemoglobin, IGF-I, and
IGFBP-3 concentrations. In the 40 patients, the average
growth rate of 5.5 (4.8 to 6.1) cm per year before GH treatment increased to 9.8 (9.0 to 10.6) cm per year. Correspondingly, the relative height increased by a mean of 1.0 (0.8 to
1.1) SD units during the 1-year study period. Figure 3 illustrates the respective changes for the two subgroups.
Hemoglobin increased by 0.42 (0.16 to 0.68) g/dL during
15-
-2
'
IGF-I
-1: 0
IGFBP-3
:1 :-2 . 0 ..2. .
'
'
'
Height
'
-6
. . -4.
'
Erythropoietin
-2
' '
..
'
0'
10
100
. .'."..*.r..*.*.
: .
*
*
.....
1oc
..
..
0
I
.-
..
'.
. .-..
.
.I
RESULTS
Correlation of blood hemoglobin concentration with body
height and with serum IGF-I and IGFBP-3. In the 32 patients with GH secretion ranging from normal to pathological, blood hemoglobin concentration (as g/dL) correlated
positively with relative height (r = 0.37, P = .037), with
IGF-I (r = 0.37, P = .OM), and with IGFBP-3 (r = 0.50, P
= .003), but was independent of EPO ( P = .48). If the eight
patients with skeletal dysplasia were included, the correlations were even closer (Fig 2). Multiple regression analysis
was performed on 40 patients. The best three-variable model
explained 56% of the variation in hemoglobin level. According to this model, hemoglobin was positively associated
with IGFBP-3 (P= ,002) and body height in cm (P< .OOl)
I
P
a
12
't.4
I
o
-2
Y
y:
Fig 2. Scattergram matrix between hemoglobin, IGF-I, IGFBP-3,
EPO, and relative body height bdora GH treatment in the 40 children
with short datura. Blood hemoglobin is expressed as gldL; IGF-I,
IGFBP-3, and height as SD units; and EPO as mUlmL on a logarithmic
scala. Only significant regression lines are drawn. Paarson correlation
coflicients: hemoglobin versus IGF-I: r = 0.37, P = .022; hemoglobin
versus IGFBP-3: r = 0.52, P = .001; hemoglobin versus height r =
0.52, P = .001; IGF-l versus IGFBP-3: r = 0.65, P < .OOl.
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
2078
'L.
VlHERVUORl ET AL
1
0
6
12
0
6
12
-2 1
0
6
12
0
6
12
Time, months
Fig 3. Relative height (SD units), blood hemoglobin (g/dL), serum
IGF-l (SD units), serum IGFBP-3 (SD units), and serum EPO (mu/
mL) concentrations during the first year of GH treatment. Values are
means 2 SEM. (A) Relative height. The eight patients with skeletal
Patientswith other causes for short stature, GH secredysplasia (0).
tion ranging from normal to pathological (01.(B) Hemoglobin. Symbols as above. (C) IGF-l (A)and IGFBP-3 (AI, all 40 patients. The
patients with skeletal dysplasia did not differ from the other patients.
(D) EPO in the 23 patients with a measurement at 1 week. One of
these patients had skeletal dysplasia.
the 12 months in the 40 patients. Initially, however, it fell
by 0.24 (0.0 to 0.48) g/dL, then rose by 0.47 (0.23 to 0.72)
g/dL (Fig 3B). In contrast, IGF-I increased by 0.7 (0.4 to
1.1) SD units and IGFBP-3 by 1.7 (1.3 to 2.1) SD units
during the first week of GH treatment; they both continued
to increase throughout the study period (Fig 3C). During the
first week, the reticulocyte count increased by 46 (4% to
88)% (n = 18) and EPO increased by 6.6 (3.9% to 9.4)%
(n = 23) (Fig 3D). The increase in EPO concentration during
the first week of treatment correlated positively with the
subsequent change in hemoglobin (r = .52, P = .012). The
(geometric) mean ferritin level decreased from 26 to 18 pg/
L during the first 6 months of GH treatment ( P < .001), and
the number of patients with ferritin below 12 pg/L increased
from three to eight. None of the changes observed in blood
or in serum correlated with the body growth response.
When IGF-I and IGFBP-3 began to increase, the associations with hemoglobin were transiently lost. After 6 months,
however, as before treatment, hemoglobin again correlated
with IGF-I and IGFBP-3 (r = 0.48 to 0.63, P = .001 to
.002).
Extremely close associations between hemoglobin and
IGF-I or IGFBP-3 in the eight children with skeletal dysplasia. These patients were shorter ( P < .001), their weightto-height ratios were higher ( P < .001) (Table l), and they
received larger GH doses per body surface area than the
other patients ( P = .058) (Fig 4). In these eight patients, the
increase in hemoglobin was marked (1.32; 0.85 to 1.80 g/
dL), being greater than in any of the other 32 -patients(P <
.001, Fig 3B). Despite the larger doses of GH, they grew
less than the other patients (6.7 v 10.8 c d y r , P < ,001, or
relative height 0.5 v 1.1 SD units, P = .003, Fig 3A). The
estimated total erythrocyte mass increased by 150 mL. in the
patients with skeletal dysplasia, while in the other patients
the increase was 100 mL (P = .020). There were no significant differences from the other patients in the levels or increments of IGF-I, IGFBP-3, or EPO. After treatment, the association between hemoglobin and relative height was evident
only if the patients with skeletal dysplasia were excluded
(with this group included: r = .15, P = .37, with this group
excluded: r = .51, P = .004).
No relation between endogenous GH secretion and blood
hemoglobin level or other changes observed during treatment. As expected, the initial IGF-I and IGFBP-3 levels
were lowest in the patients with GH deficiency (Table l ) ,
but the groups did not differ in respect of hematologic measurements. No differences were observed regarding increases
in relative height, hemoglobin, IGF-I, IGFBP-3, or EPO in
the three subgroups classed according to GH secretion. Exclusion of the patients with signs of puberty did not change
the results.
DISCUSSION
The present study demonstrates that three parameters of
growth, body height, IGF-I, and IGFBP-3, correlate closely
with blood hemoglobin level. The results strongly suggest
that the GH-IGF-I axis participates in the regulation of human erythropoiesis. In addition to corroborating the results
of earlier studies conducted in vitro and in ani mal^,'.'^ the
present results substantiate the previous anecdotal observations. 17,-33,34
There is a fundamental difference in blood hemoglobin
Post
Pre
13
r = 0.85
P = 0.008
l O - 1
14
1
2
3
4
IGF-I. SDS
IGF-I, SDS
15
0
r = 0.80
P = 0.017
13
12
11
10
-2
-1
0
1
2
3
IGFBP-3, SDS
4
IGFBP-3.SDS
Fig 4. Associations between blood hemoglobin and serum IGF-l
and IGFBP-3 concentrationsin the eight patientswith skeletal dysplasia before and after OH therapy. The latter values are averages of
the values at 6 months and at 12 months of OH treatment.
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
2079
GROWTH HORMONE AND ERYTHROPOIESIS
level between adults and growing children. In adulthood,
aged erythrocytes are constantly replaced by equal quantities
of new erythrocytes to maintain the hemoglobin level. During growth, in contrast, each new erythrocyte generation has
to be more numerous than the previous, just to maintain a
stable hemoglobin level. Yet, the mean blood hemoglobin
concentration normally increases from about 11.0 g/dL at
the age of 2 months to about 13.5 g/dL in adult women and
to 15.5 g/dL in adult men. Our results suggest that the GHIGF-I axis plays a role in growth of the continuously renewed
erythrocyte mass. If GH promotes the increase in the erythrocyte mass during childhood, the system may be adjusted to
overproduction of erythrocytes and a gradual increase in
hemoglobin concentration. In light of the proposed possibility, it is not surprising that the children with growth failure
did not demonstrate the normal age-dependency of hemoglobin, but they did demonstrate dependency on body height.
EPO, by reacting against any renal hypoxia, is unquestionably the principal factor preventing anemia. But although
blood hemoglobin concentration remains constantly within
the normal range, the GH-IGF-I system may elevate it further
during growth periods. This increase would appear to guarantee better oxygen supply to growing tissues. In adulthood,
the role of this latter mechanism in the regulation of erythropoiesis would be of minor importance. The notion that GHIGF-I axis and EPO appear to have parallel functions is
supported by observations from an in vitro study on mice in
which, at normal EPO levels, about 40% of erythropoiesis
depended on EPO alone, another 30% or less on the additive
effect of IGF-I, and the remaining 30% or more on other
unidentified serum factor^.^
During GH treatment, erythropoiesis and rapid body
growth increase iron needs, and serum ferritin concentration
decreases.” In the light of this observation, the slight increase in hemoglobin concentration observed seems even
more remarkable. In early male puberty, the decrease in
ferritin that precedes accelerated
may be related to
early increase in GH secretion, and significant increase in
hemoglobin concentration occurs later during pubertal maturation.3s It is not impossible that our observation of decreased ferritin levels in prepubertal children given GH is
related to a similar mechanism as that seen in early puberty
under physiological conditions. Further studies are needed
to address this possibility.
The present study proved that the role of GH-IGF-I system
in the control of hematopoiesis is obvious. However, the
mechanism through which the GH-IGF-I system acts on erythroid progenitor cells is not clear. In a recent study, Ratajczak et a136 suggested that IGF-I does not have a direct
effect on proliferation of erythroid progenitor cells, and other
studies have suggested that, in hematopoietic cells, IGF-I
acts principally by inhibiting a p o p t o ~ i s . ~ ~ . ~ ~
In our patients, endogenous GH secretion varied from
apparently normal to remarkably subnormal, although none
had complete absence of GH secretion. In keeping with earlier observation^,^^^^ the baseline hemoglobin level was not
significantly lower in the children with GH deficiency than
in the other patients. In our study, these children were hardly
distinguishable from the other patients in their 1-year growth
response to GH treatment. This finding is also supported by
several other studies.40” In fact, we observed no important
differences between the subgroups classed according to GH
secretion, but we cannot rule out the possibility that such
differences may exist. As the diagnosis of GH deficiency is
a matter of subjective
it may be argued that
very few, if any, of our patients had unequivocal GH deficiency. From the statistician’s point of view, the present
study was not designed to detect similarities. For that purpose, the sample size would have had to be larger.
A remarkable hemoglobin response to GH treatment was
observed in the children with skeletal dysplasia. Because of
their chondrodystrophic bone disease, predominantly achondroplasia, they were not only short but their trunMimb ratio
was disproportionate for age. In achondroplasia, there are
mutations in the gene encoding fibroblast growth factor receptor-3.” Growth deformities in the spines and limbs of
those affected reduce their growth potential, their final height
usually being within the range of 112 to 145 ~ m . 4These
~
rare conditions may help us understand the relationship between body growth and erythropoiesis. In our eight patients,
GH treatment resulted in a slight body growth response only,
whereas the other responses to GH, such as erythropoiesis,
seemed unaffected, for their erythrocyte mass increased well.
IGF-I and IGFBP-3 increased in a similar manner both in
the patients with skeletal dysplasia and in the other patients,
but the association between hemoglobin and these GH-related parameters was clearer in the former patients.
The role of EPO as an indicator of hypoxia and a regulator
of erythropoiesis is well documented, and in anemic patients,
serum EPO concentration is known to correlate negatively
with hemoglobin
However, a correlation between
serum EPO and blood hemoglobin concentrations has not
been reported in subjects with normal hemoglobin levels.I6
In keeping with this, we found no such correlation. As expected from previous studies, EPO increased during the first
week of GH
The sodium-retaining action of
GH causes fluid r e t e n t i ~ n ~and,
’ . ~ ~most likely, expansion of
plasma volume. From these observations it is possible to
speculate that initial hemodilution due to this phenomenon
would explain the early temporary decline in hemoglobin
concentration that coincided with the elevation in EPO concentration. The fact that the initial elevation of EPO correlated with the subsequent elevation of hemoglobin concentration confirms the involvement of EPO in the observed
changes.
ACKNOWLEDGMENT
We thank Dr I. Sipila and Dr J. Maenpa (Aurora Hospital, Helsinki) for collaboration in examinations of the patients. We are grateful to Dr L. Hagenas (Karolinska Hospital, Stockholm, Sweden) for
cooperation concerning the patients with skeletal dysplasia and to
Celtrix Pharmaceuticals, Santa Clara, CA, for recombinant IGFBP3 for developing the assay. We also thank Dr S. Kontiainen and Dr
J. Maenpaa (Aurora Hospital, Helsinki) for supplying serum samples
from healthy children.
REFERENCES
1. Golde DW, Bersch N, Li CH: Growth hormone: species-spe-
cific stimulation of erythropoiesis in vitro. Science 196:1 1 12, 1977
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
2080
2. Claustres M, Chatelain P, Sultan C: Insulin-like growth factor
I stimulates human erythroid colony formation in vitro. J Clin Endocrinol Metab 6578, 1987
3. Merchav S, Tatarsky I, Hochberg Z: Enhancement of erythropoiesis in vitro by human growth hormone is mediated by insulinlike growth factor I. Br J Haematol 70:267, 1988
4. Akahane K, Tojo A, Urabe A, Takaku F Pure erythropoietic
colony and burst formation in serum-free culture and their enhancement by insulin-like growth factor I. Exp Hematol 15:797, 1987
5. Sawada K, Krantz SB, Dessypris EN, Koury ST, Sawyer ST:
Human colony-forming units-erythroid do not require accessory
cells, but do require direct interaction with insulin-like growth factor
I andor insulin for erythroid development. J Clin Invest 83:1701,
1989
6. Correa PN, Axelrad AA: Production of erythropoietic bursts
by progenitor cells from adult human peripheral blood in an improved serum-free medium: Role of insulinlike growth factor I.
Blood 76:2823, 1991
7. Boyer SH, Bishop TR, Rogers OC, Noyes AN, Freilin LP,
Hobbs S: Roles of erythropoietin, insulin-like growth factor 1, and
unidentified serum factors in promoting maturation of purified murine erythroid colony-forming units. Blood 80:2503, 1992
8. Barak Y, Zadik A, Karov Y, Hahn T: Enhanced response of
human circulating erythroid progenitor cells to hGH and to IGF-I
in children with insufficient growth hormone secretion. Pediatr Res
32:282, 1992
9. Sanders M, Sorba S, Dainiak N: Insulin-like growth factors
stimulate erythropoiesis in serum-substituted umbilical cord blood
cultures. Exp Hematol 21:25, 1993
IO. Geffner ME, Yeh DY, Landaw EM, Scott ML, Stiehm ER,
Bryson YJ, Israele V: In vitro insulin-like growth factor-I, growth
hormone, and insulin resistance occur in symptomatic human immunodeficiency virus-1-infected children. Pediatr Res 34:66, 1993
11. Sanengen T, Halvosen S: Regulation of erythropoiesis during
rapid growth. Br J Haematol 61:273, 1985
12. Kurtz A, Zapf J, Eckardt KU, Clemons G, Froesch ER, Bauer
C: Insulin-like growth factor I stimulates erythropoiesis in hypophysectomized rats. Proc Natl Acad Sci USA 85:7825, 1988
13. Kurtz A, Matter R, Eckardt KU, Zapf J: Erythropoiesis, serum
erythropoietin, and serum IGF-I in rats during accelerated growth.
Acta Endocrinol (Copenh) 122:323, 1990
14. Phillips AF, Persson B, Hall K, Lake M, Skottner A, Sanengen T, Sara VR: The effects of biosynthetic insulin-like growth
factor- I supplementation on somatic growth, maturation, and erythropoiesis on the neonatal rats. Pediatr Res 23:298, 1988
15. Erslev A, Silver R, Coaro J, Paist S, Cobbs E: The effect of
sustained hypertransfusion on haematopoiesis, in Murphy MJ (ed):
In Vitro Aspects of Erythropoiesis. New York, NY, Springer, 1978,
P 58
16. Pressac M, Morgant G, Farnier MA, Aymard P Enzyme
immunoassay of serum erythropoietin in healthy children: reference
values. Ann Clin Biochem 28:345, 1991
17. Urena P, Bonnardeaux A, Eckardt KU, Kurtz A, Driieke TB:
Insulin-like growth factor I: A modulator of erythropoiesis in uraemic patients? Nephrol Dial Transplant 7:40, 1992
18. Jflrgensen JOL, Blum WF, MflllerN, Ranke MB, Christiansen
JS: Short-term changes in serum insulin-like growth factors (IGF)
and IGF binding protein 3 after different modes of intravenous
growth hormone (GH) exposure in GH-deficient patients. J Clin
Endocrinol Metab 72582, 1991
19. Albertsson-Wikland K, Rosberg S: Methods of evaluating
spontaneous growth hormone secretion, in Ranke MB (ed): Functional Endocrinologic Diagnostics in Children and Adolescents.
Mannheim, Germany, J & J Verlag, 1992, p 76
VIHERVUORI ET AL
20. Clemmons DR, Van Wyk JJ: Factors controlling blood concentration of somatomedin C. Clin Endocrinol Metab 13:113, 1984
21. JflrgensenJOL, Blum WF, Msller N, Ranke MB, Christiansen
JS: Circadian pattems of serum insulin-like growth factor (IGF) 11
and IGF binding protein 3 in growth hormone-deficient patients and
age- and sex-matched normal subjects. Acta Endocrinol (Copenh)
123:257, 1990
22. Blum WF, Ranke MB: Insulin-like growth factor binding
proteins (IGFBPs) with special reference to IGFBP-3. Acta Paediatr
Scand 36755, 1990 (suppl)
23. Baxter RC, Martin JL: Radioimmunoassay of growth hormone-dependent insulinlike growth factor binding protein in human
plasma. J Clin Invest 78:1504, 1986
24. Camacho-Hubner C, Clemmons DR, D’Ercole AJ: Regulation
of insulin-like growth factor (IGF) binding proteins in transgenic
mice with altered expression of growth hormone and IGF-I. Endocrinology 129:1201, 1991
25. Tanner JM: Physical growth and development, in Campbell
AGM, McIntosh N (eds): Forfar and Ameil’s Textbook of Pediatrics.
Edinburgh, Scotland, Churchill Livingstone, 1992, p 389
26. Ranke MB, Haber P: Growth hormone stimulation tests, in
Ranke MB (ed): Functional Endocrinologic Diagnostics in Children
and Adolescents. Mannheim, Germany, J & J Verlag, 1992, p 61
27. Vihervuori E, Sipila I, Siimes MA: Increases in hemoglobin
concentration and iron needs in response to growth hormone treatment. J Pediatr 125:242, 1994
28. Koistinen H, Seppala M, Koistinen R: Different forms of
insulin-like growth factor-binding protein-3 detected in serum and
seminal plasma by immunofluorometric assay with monoclonal antibodies. Clin Chem 40:531, 1994
29. Lee LG, Chen CH, Chiu LA: Thiazole orange: A new dye
for reticulocyte analysis. Cytometry 7508, 1986
30. Green R, Charlton R, Seftel H, Bothwell T, Mayet F, Adams
B, Finch C, Layrisse M: Body iron excretion in man. Am J Med
45:336, 1968
31. Soma R, Tolppanen EM, Perheentupa J: Variation of growth
in height and weight of children. Acta Paediatr Scand 79:498, 1990
32. Dallman PR, Siimes MA: Percentile curves for hemoglobin
and red cell volume in infancy and childhood. J Pediatr 94:26, 1979
33. Anttila R, Koistinen R, Seppila M, Koistinen H, Siimes MA:
Insulin-like growth factor I and insulin-like growth factor binding
protein 3 as determinants of blood hemoglobin concentration in
healthy subjects. Pediatr Res 36:745, 1994
34. Wilson DM, Stene MA, Killen JD, Hammer LD, Litt IF,
Hayward C, Taylor CB: Insulin-like growth factor binding protein3 in normal pubertal girls. Acta Endocrinol (Copenh) 126:381, 1992
35. Anttila R, Siimes MA: Development of iron status and response to iron medication in pubertal boys. J Pediatr Gastroenterol
Nutr (in press)
36. Ratajczak MZ, Kuczynski WI, Onodera K, Moore J, Ratajczak J, Kregenow DA, DeRiel K, Gewirtz AM: A reappraisal of
the role of insulin-like growth factor I in the regulation of human
hematopoiesis. J Clin Invest 94:320, 1994
37. Williams GT, Smith CA, Spooncer E, Dexter TM, Taylor
DR: Haematopoietic colony stimulating factors promote cell survival
by suppressing apoptosis. Nature 343:76, 1990
38. Muta K, Krantz SB: Apoptosis of human erythroid colonyforming cells is decreased by stem cell factor and insulin-like growth
factor I as well as erythropoietin. J Cell Physiol 156:264, 1993
39. Ardizzi A, Guzzaloni G, GNgni G, Moro D, Calo G, Mazzilli
G, Tonelli E, Morabito F: (The effect of GH on erythropoiesis in
vivo,) Effetto, in vivo, del GH sull’eritropoeisi. Minerva Endocrinol
1833, 1993
40. Genentech Collaborative Study Group: Idiopathic short stat-
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
GROWTH HORMONE AND ERYTHROPOIESIS
ure: Results of a one-year controlled study of human growth hormone treatment. J Pediatr 115:713, 1989
41. Hopwood NJ, Hintz RL, Gertner JM, Attie KM, Johanson
AJ, Baptista J, Kuntze J, Blizzard Rh4, Cara JF, Chemausek SD,
Kaplan SL, Lippe BM, Plotnick LP, Saenger P: Growth response
of children with non-growth-hormone deficiency and marked short
stature during three years of growth hormone therapy. J Pediatr
123:215, 1993
42. Zadik Z, Chalew S, Zung A, Landau H, Leiberman E, Koren
R, Voet H, Hochberg Z, Kowarski A: Effect of long-term growth
hormone therapy on bone age and pubertal maturation in boys with
and without classic growth hormone deficiency.J Pediatr 125:189,1994
43. Silink M: Alternative methods of diagnosis of growth hormone deficiency. J Pediatr Endocrinol 5:43, 1992
44. Rousseau F, Bonaventure J, Legeal-Mallet L, Pelet A, Rozet
J-M, Maroteaux P, Le Merrer M, Munnich A: Mutations in the
gene encoding fibroblast growth factor receptor-3 in achondroplasia.
Nature 371:252, 1994
2081
45. Horton WA, Rotter JI, Rimoin DL, Scott CI, Hall JG:
Standard growth curves for achondroplasia. J Pediatr 93:435,
1978
46. Brown MS, Phibbs RH, Garcia JF, Dallman PR: Postnatal
changes in erythropoietin levels in untransfused premature infants.
J Pediatr 103:612, 1983
47. Eckardt KU, Hartmann W, Vetter U, Pohlandt F, Burghardt
R, Kurtz A: Serum immunoreactive erythropoietin of children in
health and disease. Eur J Pediatr 149:459, 1990
48. Peschle C, Rappaport IA, Sasso GF, Gordon AS, Condorelli
M: Mechanism of growth hormone (GH) action on erythropoiesis.
Endocrinology 91:511, 1972
49. M ~ l l e J,
r J~rgensenJOL, Ovensen P, M~llerN, Laursen T,
Christiansen JS: Effects of growth hormone on body fluid homeostasis. J Pediatr Endocrinol 5:79, 1992
50. Herlitz H, Jonsson 0, Bengtsson BA: Effect of recombinant
human growth hormone on cellular sodium metabolism. Clin Sci
(Colch) 86:233, 1994
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
1996 87: 2075-2081
Hemoglobin level is linked to growth hormone-dependent proteins in
short children
E Vihervuori, M Virtanen, H Koistinen, R Koistinen, M Seppala and MA Siimes
Updated information and services can be found at:
http://www.bloodjournal.org/content/87/5/2075.full.html
Articles on similar topics can be found in the following Blood collections
Information about reproducing this article in parts or in its entirety may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests
Information about ordering reprints may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#reprints
Information about subscriptions and ASH membership may be found online at:
http://www.bloodjournal.org/site/subscriptions/index.xhtml
Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American
Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.
Copyright 2011 by The American Society of Hematology; all rights reserved.