| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Hospital for Children and Adolescents (O.M., L.D., M.A.S.), Departments of Medicine (E.J.) and Clinical Genetics (I.K.), University of Helsinki, FIN-00029 Helsinki, Finland
Address all correspondence and requests for reprints to: Outi Mäkitie, M.D., Hospital for Children and Adolescents, Helsinki University Hospital, Stenbäckinkatu 11, P.O. Box 281, FIN-00029 Helsinki, Finland. E-mail: outi.makitie{at}huch.fi
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Retrospective analysis of hematological data of 114 patients showed that the severity of the anemia and macrocytosis in CHH varies with age. The anemia was most severe in early childhood. A prospective study of 21 patients with CHH showed that height correlates with hemoglobin (P = 0.006) and mean corpuscular volume of red blood cells (P < 0.0001). The individual hemoglobin levels correlated with the GH parameters [P = 0.035 for insulin-like growth factor I (IGF-I) and P = 0.002 for IGF-binding protein-3], and the mean corpuscular volume of red blood cell values correlated with fetal hemoglobin. Bone marrow cultures obtained from six patients with CHH showed reduced or totally absent erythroid colony formation, which was not influenced by GH or IGF-I in vitro or by GH treatment in vivo.
In patients with CHH, we observed an association between erythropoiesis and growth. We conclude that body growth and erythropoiesis share common regulators. One of these is the GH-IGF-I axis; other factors, as not yet identified, may also be important.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Defective erythropoiesis, presenting in the peripheral blood as anemia, is an integral feature of CHH (6). In a previous study of 88 patients, we found a history of childhood anemia, usually in early infancy, in 73% of the retrospectively evaluated patients. The anemia had often been macrocytic and never microcytic. Fourteen patients had presented with severe anemia; most of them had recovered spontaneously before the age of 3 yr (6). The blood cell alterations may result from defects in cell proliferation and/or maturation, as impaired colony growth of erythroid progenitors has been observed in bone marrow cultures from these patients (7).
GH and insulin-like growth factor I (IGF-I) may modulate not only growth, but also erythropoiesis. In hypophysectomized rats, both GH and IGF-I stimulated erythropoiesis (8). In addition, the individual hemoglobin level is reported to be highly dependent on the IGF-binding protein-3 (IGFBP-3) concentration in healthy human subjects (9). Furthermore, in children with short stature, treatment with GH accelerated growth and elevated the concentration of hemoglobin (10). In seven patients with skeletal dysplasia (achondroplasia or spondyloepimetaphyseal dysplasia) the association between the hemoglobin response and the GH status was extremely close (10). Recently, an association between erythropoiesis and GH status has also been observed in adults with GH deficiency (11).
Fanconi’s anemia and Diamond-Blackfan anemia are genetic conditions in which both skeletal growth and erythropoiesis are disturbed, suggesting that they are partly regulated by identical factors. These have, however, not been evaluated in detail. In the present study we used CHH as a clinical model to further elucidate the interrelationship between body growth and erythropoiesis. More evidence was obtained for the role of GH in the regulation of both these variables.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Subjects. The hospital records of 114 patients with CHH (46 males and 68 females) were reviewed to trace any age-dependent trends in hemoglobin concentrations and mean corpuscular volume of red blood (MCV) values. The methods of ascertainment and diagnostics of the patients have been explained previously (5). There were 30 and 48 patients with 1, 26 and 25 patients with 2, 28 and 17 patients with 3, and 30 and 12 patients with 4–7 hemoglobin and MCV values, respectively, included in the analysis.
Methods. In the retrospective series, data were collected for each patient on seven occasions: at 2 months (1.5–2.5 months), 6 months (5–7 months), 1 yr (10–12 months), 2 yr (1–3 yr), 7 yr (5–10 yr), 15 yr (14–16 yr), and 18 yr or more. If several measurements were available for the age period, the hemoglobin values recorded nearest to the age point were used during the first year and after 14 yr; the lowest hemoglobin values for the age period were used between 1–10 yr. The corresponding MCV values were recorded. Values were pooled (male and female values combined) and calculated for the 10th, 50th, and 90th percentiles in each age category. Because of skewness, percentiles were used instead of SD units. The percentile curves were compared with the age-adjusted hemoglobin and MCV values.
Prospective analysis of hematological and growth- related parameters
Subjects. The 21 consecutive patients with CHH treated at the out-patient clinics of the Hospital for Children and Adolescents, University of Helsinki, between June 1997 and June 1998, were included in the study. There were 5 males and 16 females of ages ranging from 1.0–16.7 yr (median, 5.9 yr). Their diagnoses had been based on short-limbed short stature, sparse hair, generalized laxity of joint ligaments, and the characteristic skeletal abnormalities seen in radiographs (2, 12).
Methods. The patients were clinically evaluated at the
out-patient clinic. The physical examination included measurements of
body weight and height with a Harpenden stadiometer (Holtain Ltd.,
London, UK). The laboratory tests consisted of measurements of blood
hemoglobin, hematocrit, MCV, reticulocyte count, fetal hemoglobin
(Hb-F) concentration, and serum concentrations of erythropoietin (EPO),
IGF-I, IGFBP-3, iron, transferrin, ferritin, and haptoglobin. Bone
marrow aspirate samples were obtained from the posterior iliac crest
under general anesthesia (five children) or from the sternum under
local anesthesia (one adult). Bone marrow cultures were obtained from
six patients: from three patients as part of this study (patients 5,
13, and 14 in Tables 1–3
),
from two patients for clinical evaluation of anemia (patients 1 and 22
in Tables 1–3) and from one adult patient for evaluation of
immunodeficiency and hypersedimentation (patient 23 in Tables 2 and 3).
|
|
|
Blood hemoglobin concentrations, hematocrit, and MCV were measured with a Coulter counter T 890 (Coulter Electronics, Miami, FL). Reticulocytes were counted with a FACScan flow cytometer (Becton Dickinson and Co., San Jose, CA). The Hb-F concentration was measured by ion exchange high performance liquid chromatography (Diamat, Bio-Rad Laboratories, Inc., Hercules, CA) and expressed as a percentage of the total hemoglobin. The analytical sensitivity was 0.1%. EPO concentrations were determined by an in-house RIA (Medix Clinical Laboratory, Espoo, Finland). The functional sensitivity of the assay was 8 U/L. The intraassay coefficient of variation was 5–10% throughout the measuring range (8–1000 U/L). The interassay coefficients of variation for levels of 17 and 115 U/L were 10% and 8.3%, respectively.
Serum IGF-I concentration was determined by RIA (INCSTAR Corp., Stillwater, MN) after acid extraction chromatography. The sensitivity of the assay was 2 nmol/L, the intraassay variation 8–10%, and the interassay variation 10–15%. Serum IGFBP-3 was measured by RIA (1277 GammaMaster, EG&G Wallac, Inc., Turku, Finland) with an IGFBP-3 RIA kit (Nichols Institute Diagnostics, San Juan Capistrano, CA). The serum samples were diluted 1:250, or 1:500 for concentrations above 7 mg/L. The measurement range covered 0.25–14.0 mg/L. The intraassay variation ranged from 5.6–6.9%, and the interassay variation ranged from 12–17%.
The serum iron concentration was measured colorimetrically with a Hitachi 911 analyzer (Hitachi Ltd., Tokyo, Japan). The interassay variation was 2.8% at the level of 19.7 µmol/L and 3.0% at the level of 51.4 µmol/L. The serum transferrin concentration was determined by an immunoturbidometric method with the Hitachi 911 analyzer (Hitachi Ltd., Tokyo, Japan), antihuman transferrin from Orion Diagnostica (Helsinki, Finland), and an IMPRO calibrator (Labquality, Helsinki, Finland) calibrated with the BCR CRM 470. The interassay variation was 3.2% at the level of 0.97 g/L and 3.7% at the level of 2.67 g/L. Transferrin saturation was calculated by dividing 3.825 x serum iron (µmol/L) by serum transferrin (grams per L). Serum ferritin was measured by a chemiluminometric method with ACS 180 immunoanalyzer (Chiron Corp., Halstead, UK). The haptoglobin concentration was determined by an immunoturbidometric method with the Hitachi 911 analyzer, antihuman haptoglobin (Orion Diagnostica), and the IMPRO calibrator (Labquality) calibrated with the BCR CRM 470.
Bone marrow cultures
Erythroid and granulocyte-macrophage progenitors (CFU-GM) from bone marrow aspirates or blood were cultured according to the method of Guilbert and Iscove (13) as previously described (14). The culture medium consisted of 0.9% methyl cellulose, 20% FCS, 1% delipidated and deionized BSA, 10-4 mol/L 2-mercaptoethanol, 310 µg/mL fully iron-saturated human transferrin, 10% human phytohemagglutinin-stimulated leukocyte-conditioned medium, and Iscove’s modified Dulbecco’s medium. The concentration of mononuclear cells was 0.5 x 105/mL. The growth of erythroid colonies was stimulated with recombinant human EPO (2 U/mL; Cilag AG, Schaffhausen, Switzerland), and colonies for erythroid progenitors were scored on the 14th day of culture. Granulocyte-macrophage colony formation was stimulated with 5637-conditioned medium and colonies were scored on the 14th day of culture. GH was added to the basic culture medium at concentrations of 0, 4, 40, 400, and 4000 µU/mL, and IGF-I was added at concentrations of 0, 0.2, 2, 20, and 200 ng/mL to study their effects on colony formation. Healthy bone marrow transplant donors served as controls.
Data transformations
All age- and/or sex-dependent hematological values were transformed into SD units of the age-matched reference population. The heights were compared with age- and sex-adjusted Finnish norms (15) and expressed as SD scores. The conversion of hemoglobin and MCV was based on a large series of Finnish-American data (16, 17). The conversion of IGF-I and IGFBP-3 was based on values obtained from 139 healthy Finnish children (18). Values outside ±2 SD of the normal mean were regarded as abnormal. The Hb-F concentration was expressed as a percentage of the hemoglobin concentration (Hb-F%); after infancy this value is not age dependent.
Statistical analyses
Simple regression analysis (StatView 4.51 for MacIntosh, 1992–1995 Abacus Concepts, Inc., Berkeley, CA) was used for statistical analyses. P < 0.05 was considered significant.
Ethical considerations
The study was approved by the ethics committee of the Hospital for Children and Adolescents, University of Helsinki.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We observed anemia and macrocytosis in half of the patients with
CHH (Fig. 1). Anemia and macrocytosis
were more frequent in early childhood. However, the median of the
hemoglobin SD score was close to the lower limit of
normal throughout childhood until adolescence, when the reference mean
was reached. After the early years, the degree of macrocytosis remained
constant through childhood. Macrocytosis was present in 50% of the
adult patients.
|
There was extreme variability of the phenotype. The height
SD score ranged from -9.4 to -2.4 (median, -5.1) and was
not age dependent (r = 0.41). The levels of IGF-I and IGFBP-3 were
usually within age-specific norms, although the IGFBP-3 values were
closer to the lower limit of normal (median, -1.0 SD
score) and were below the limit in four patients. No significant
correlations were observed between the IGF-I (r = 0.43;
P = 0.052) or IGFBP-3 (r = 0.34; P
= 0.1) and height. Most of the patients had anemia and/or macrocytosis
(medians for hemoglobin, -1.8 SD score; for MCV,
+1.6 SD score; Table 1
).
We observed a correlation between the hematological parameters and body
growth and related parameters. The degrees of anemia (r = 0.57;
P = 0.006) and macrocytosis (r = -0.82;
P < 0.0001) correlated strongly with height (Fig. 2). Hemoglobin and MCV were also related
to each other (r = -0.51; P = 0.018). The
transferrin level (r = 0.68; P = 0.0004), but not
the other parameters of iron metabolism or haptoglobin, correlated with
the relative height.
|
). The EPO concentration was elevated in
15 of 19 patients (median, 80 U/L; range, 17–440 U/L; reference range,
12–35 U/L). It correlated with hemoglobin SD
score (r = -0.69; P = 0.0007). The serum iron,
transferrin, transferrin saturation, ferritin, and haptoglobin levels
varied from low to elevated, but did not correlate with the degree of
anemia (Table 1).
|
). The correlation remained significant
even after omission of height as a confounding parameter (r =
0.68; P = 0.001). MCV did not correlate with the
parameters of iron metabolism, except for the serum transferrin
concentration (r = -0.53; P = 0.017). In the
patients with macrocytosis, we observed no evidence of hemolysis and no
correlation between MCV and the reticulocyte percentage (r =
-0.21) or the haptoglobin concentration (r = -0.22).
|
Erythroid colony formation was markedly impaired in all patients;
few if any rudimentary colonies were seen (Table 2). Growth was
defective even in the patients with normal or near-normal hemoglobin
concentrations. Addition of GH or IGF-I to the culture mixture did not
affect the in vitro growth of the erythroid colonies (Table 3). In two patients (no. 1 and 14), the hemopoietic progenitors were
cultured before and during treatment with recombinant human GH. The
treatment did not enhance the in vitro growth of the
erythroid colonies.
In all patients, the granulocyte-macrophage growth was also markedly
defective. These cells either did not grow or grew only as CFU-GM
clusters, i.e. as colonies consisting of only a few cells
(Table 2). Addition of GH or IGF-I to the culture plates slightly
increased the numbers of CFU-GM clusters for the patients with CHH and
the numbers of CFU-GM colonies for the normal controls (Table 3).
Treatment with recombinant human GH also caused an improvement in the
growth of the CFU-GM clusters (Table 3).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The results of the present study throw further light on the association between the regulation systems involved in body growth and erythropoiesis. The pleiotropic features of CHH include severe growth failure due to metaphyseal dysplasia and impaired erythropoiesis, probably due to defective progenitor cell proliferation. Due to the population history, the CHH patients in Finland are all likely to carry an identical mutation. The variability in the phenotypic features thus reflects the interaction with other genes rather than variation in the CHH gene itself. The genes involved in the regulation of body growth and erythropoiesis vary with age. Fetal growth is regulated by several factors, but is virtually independent of GH, which becomes the main modulator of body growth after the first year (22). Growth during adolescence is related to both GH and sexual hormones, testosterone in males and estrogens in females. Erythropoiesis in the fetus is stimulated by hypoxia; after birth, the hemoglobin level rapidly decreases as the hypoxic stimulus subsides. During childhood and adolescence, the red cell mass again increases with body growth and the simultaneous rise in hemoglobin concentration, but the factors responsible for these remain unknown. GH may be one of the regulators. At male puberty, GH and sex steroids together cause the pubertal growth spurt and are probably related to the additional rise in the hemoglobin level. However, most of the pubertal growth spurt takes place in 2 or 3 yr, whereas the rise in hemoglobin concentration in healthy boys takes 5 or 6 yr (23). These observations indicate that other factors besides GH and androgens are necessary for the adjustment of hemoglobin. Further speculation is difficult, because very little is known about the regulation of the hemoglobin spurt even in healthy boys (9).
We have previously shown that the growth failure in CHH has its onset prenatally and is progressive; height deviates from normal during the first year and again at puberty, but remains stable in the period between these ages (24). The reason for the early deviation is unknown, but the growth deficit during adolescence is explained by a weak or absent pubertal growth spurt. Both of these represent periods of rapid growth and cell proliferation; impaired proliferation is likely to have a major effect, especially during these phases of growth. Erythropoiesis in CHH may be similarly dependent on age, as the most severe anemic episodes take place in early infancy and thereafter subside spontaneously (6). The present results indicate that hemoglobin is lowest during the first 2 yr of life, increases in midchildhood, and remains stable until puberty. After puberty, the hemoglobin level almost reaches the normal range in most patients, suggesting that the normal pubertal spurt of hemoglobin takes place even in male subjects in the same way as in healthy subjects (23). The degree of macrocytosis also varies with age, although the pattern differs from that of hemoglobin: macrocytosis is most severe in early childhood, but remains constant thereafter, even after puberty.
Further evidence for an association between body growth and erythropoiesis was obtained from the prospective study of the 21 patients with CHH. Both the severity of the anemia and the degree of macrocytosis again closely associated with relative height. The most marked changes were observed in the patients with the most severe growth failure. However, only the hemoglobin concentration was associated with the parameters of the GH-IGF-I axis, whereas the degree of macrocytosis seemed to be influenced by other factors. Even the retrospectively observed changes in hemoglobin level with age seem to correlate with the GH secretion pattern in CHH patients and support the role of GH in the regulation of erythropoiesis.
However, no stimulatory effect on erythroid progenitor cells from the patients could be obtained with GH or IGF-I in vitro. The standard bone marrow cultures were abnormal in all of the patients studied regardless of the other clinical or hematological features, providing further evidence for the suggestion that the defect lies in the hemopoietic progenitor cells (7). This defect could not be overcome by the addition of GH or IGF-I to the culture medium, which suggests that the impaired progenitor cell proliferation is not due to subnormal levels of GH or IGF-I. In fact, GH-related parameters were subnormal in only four patients, and in three of them, GH deficiency was excluded by the arginine stimulation test (peak values, 41.7, 27.6, and 22.3 mU/L). We have previously shown that in the bone marrow cultures of patients with CHH, the growth of erythroid colonies can be stimulated with plasma from a patient with aplastic anemia. The present results suggest that this stimulatory effect is not mediated by GH or IGF-I. In the controls no such stimulation was expected, because the culture medium itself contains sufficient amounts of growth factors for normal progenitor cell proliferation. A mild stimulatory effect of GH and IGF-I was observed in the growth of CFU-GM clusters in vitro and also during GH treatment. It is possible that a longer period of GH treatment would eventually have stimulated erythroid progenitor cell proliferation as well. GH treatment might be indicated at least in the patients with the most severe growth failure and anemia. We recently started GH treatment in two patients with exceptionally severe growth failure (patients 1 and 14); the results of the treatment will not only provide information on the effect of GH on body growth, but also further elucidate the effect of GH on the hematological parameters in this disorder.
The degree of macrocytosis, even though closely related to the severity of the growth failure, was not dependent on GH or IGF-I. In addition, there were no findings of hemolysis or increased reticulocytosis. In most of the patients, however, the proportion of fetal hemoglobin was increased and correlated closely with the MCV SD score. The clinical manifestations of CHH have been suggested to result from a generalized defect in cellular proliferation (4, 7). In chondrocytes, this results in impaired growth of the tubular bones. Macrocytosis may reflect the same proliferative defect in the erythrocytes. The biological function of the gene that is defective in CHH is unknown. Its characterization, however, will provide new insights into the regulatory mechanisms of body growth and erythropoiesis.
The increased levels of EPO confirm that the anemia is not the result of impaired secretion of EPO. On the other hand, even in patients with a normal hemoglobin concentration, EPO levels were supranormal, suggesting that increased stimulation of erythropoiesis is needed to maintain an adequate hemoglobin level. EPO alone is not a sufficient stimulus in the patients with the most severe growth failure and anemia. GH and IGF-I seem to act synergistically with EPO.
In conclusion, the present study provides data concerning defective erythropoiesis in patients with CHH. The severity of the defect varies with the degree of the growth failure. The degree of anemia is age dependent and partially regulated by the GH-IGF-I axis. The degree of macrocytosis is also dependent on age and relative height, but not on GH-related parameters. It may reflect the pathogenesis, the generalized proliferative defect, of CHH at the cellular level. The results suggest that body growth and erythropoiesis share common regulators, even other than the GH-IGF-I axis.
|
Footnotes |
|---|
Received April 27, 1999.
Revised September 21, 1999.
Accepted October 18, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
