Reference values for serumgrowth factor binding prot

e

Clinical Biochemistry 45 (2012) 1621

Contents lists available at SciVerse ScienceDirect

Clinical Bioc

evh Ajou University, Republic of Koreai Yonsei University Yonsei Wonju College of Medicine, Republic of Koreaj Wonkang University, Republic of Koreak University of Ulsan, Republic of Koreal Ewha Womans University, Republic of Koream Inje University, Republic of Korean Chonnam National University, Republic of Koreao Chonbuk National University, Republic of Koreap Chosun University, Republic of Koreaq Chungnam National University, Republic of Korear Chungbuk University, Republic of Koreas Hallym University, Republic of Korea

a b s t r a c ta r t i c l e i n f o

Article history:Received 16 May 2011Received in revised form 27 September 2011Accepted 3 October 2011Available online 12 October 2011

Keywords:IGF-IIGFBP-3KoreanChildAdolescent

Objective: Measurements of serum insulin-like growth factor-I (IGF-I) and IGF binding protein-3 (IGFBP-3) areutilized in the diagnostic work-up and clinicalmanagement of childrenwith growth disorders.We designed thisstudy to establish the reference values of serum IGF-I and IGFBP-3 levels according to age, sex and pubertal stagein Korean children and adolescents.Methods: For the study, 1378 healthy Korean children and adolescents aged 0 to 17 years (722 boys, 656 girls)were randomly selected. Blood samples were collected, and the stored sera were assayed for IGF-I and IGFBP-3using immunoradiometric assay (IRMA, Immunotech). The R 2.8.1 program (Bell Laboratories) was used to gen-erate reference percentile curves for IGF-I and IGFBP-3 according to age, sex, and pubertal stageResults: Serum IGFBP-3 level was higher in girls compared to that in boys of the same ages throughout the pu-bertal period, whereas IGF-I was only higher for girls younger than 13 years of age. Serum levels of IGF-I andIGFBP-3 increased steadily with age in the prepubertal stage, followed by a progressive decline thereafter.Peak levels of serum IGF-I and IGFBP-3 were observed two years earlier in girls compared to those in boys(13 vs. 15 years of age, respectively). Serum IGF-I and IGFBP-3 concentrations were highest at Tanner stageIV in boys and girls, with a subsequent decline.Conclusions:Our reference value model based on age, sex, and pubertal stage can improve the diagnostic utilityof IGF-1 and IGFBP-3 levels in the evaluation and management of Korean children and adolescents with growth Corresponding author at: Department of Pediatrics, Cversity, 250 Sungsan-ro, Sudaemoon-gu, Seoul, 120-752

E-mail address: kimho@yuhs.ac (H.-S. Kim).

0009-9120/$ see front matter 2011 The Canadian Sdoi:10.1016/j.clinbiochem.2011.10.003Keimyung University, Republic of Koreaf Korea University, Republic of Koreag Seoul National University, Republic of Koreac Konkuk University, Republic of Koread Kyungpook National University, Republic of KoSei Eun Hyun a, Byung Churl Lee b, Byung Kyu Suh b, So Chung Chung c, Cheol Woo Ko d, Heung Sik Kim e,Kee Hyoung Lee f, Sei Won Yang g, Choong Ho Shin g, Jin Soon Hwang h, Duk Hee Kim a, Baek Keun Lim i,Jong Duck Kim j, Han-Wook Yoo k, Hye Soon Kim l, Woo Yeong Chung m, Mi Jung Park m, Young Jong Woo n,Chan Jong Kim n, Dae-Yeol Lee o, Eun Young Kim p, Jin Ho Choi q, Heon Seok Han r,Il Tae Hwang s, Ho-Seong Kim a,a Department of Pediatrics, Institute of Endocrinology, College of Medicine, Yonsei University, Republic of Koreab The Catholic University of Korea, Republic of Korea

realevels of insulin-like growth factor-I and insulin-likeein-3 in Korean children and adolescentsj ourna l homepage: www.e lsdisorders. 2011 The Canadi

ollege of Medicine Yonsei Uni-, Korea. Fax: +82 2 393 9118.

ociety of Clinical Chemists. Publishehemistry

i e r .com/ locate /c l inb iocheman Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved.

Introduction

The insulin-like growth factor (IGF) signaling system is comprisedof the IGF-I, -II and insulin ligands, IGF binding proteins (IGFBPs), anda family of transmembrane receptors including the insulin, IGF-I and

d by Elsevier Inc. All rights reserved.

IGF-II receptors [13]. The IGFs are important factors in the regulationof cell proliferation, differentiation, cell survival and metabolic activ-ity, thereby modulating somatic growth. The six classical IGFBPs(IGFBP-1 to -6) bind IGF-I and IGF-II, but not insulin, with high afn-ity [4]. The major functions of IGFBPs are to transport IGFs, prolongtheir half-lives and regulate the availability of free IGFs for interac-tions with IGF receptors, thereby modulating the effects on growthand differentiation [35]. In serum, most of the IGF-I circulates as a150-kDa complex, which is comprised of IGF-I, IGFBP-3 and a boundacid-labile subunit. All three subunits are positively regulated by

mild symptoms just like cough and rhinorrhea. The aim of our study

a

b

0100200300400500600700800900

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

IGF-

I (n

g/mL)

Age (years)

97th

90th75th

50th25th10th

3rd

0100200300400500600700800900

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18IG

F-I

(ng/m

L)Age (years)

97th90th75th50th25th

10th3rd

Fig. 1. Reference percentile curves for IGF-I according to age in Korean boys (a) and

a

b

0

1

2

3

4

5

6

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

IGFB

P-3

(ug/m

L)IG

FBP-

3 (u

g/mL)

Age (years)

97th90th75th50th25th10th3rd

0

1

2

3

4

5

6

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18Age (years)

97th90th75th50th25th10th

3rd

Fig. 2. Reference percentile curves for IGFBP-3 according to age in Korean boys (a) and

17S.E. Hyun et al. / Clinical Biochemistry 45 (2012) 1621growth hormone in vivo, and their levels in the circulation are di-rectly correlated with body growth [6].

The clinical use of IGF-I and IGFBP-3 measurements has primarilybeen for diagnosing or excluding growth hormone (GH) deciencyand for monitoring GH therapy [710]. The proper means for estab-lishing a diagnosis of GH deciency continues to be highly controver-sial. Measurement of random serum GH concentrations is virtuallyuseless for establishing a diagnosis of GH deciency because GH se-cretion is pulsatile throughout the day and night. Therefore, measure-ment of GH secretory capacity has relied on the use of physiologic orpharmacologic stimuli, and these provocative tests have been thebasis for diagnosing GH deciency in a clinical setting [8]. An alterna-tive approach to the diagnosis of GH deciency relies on measure-ments from IGF-related assays, which can accurately reect the GHstatus of the patient [911]. Furthermore, these assays have thegreat advantage that IGF-I and IGFBP-3 normally circulate in serumat high concentrations and remain relatively constant during theday. IGF-I and IGFBP-3 concentrations have been shown to relate tothe maximal GH concentrations in serum during provocative tests[12,13]. Although measurements of IGF-I and IGFBP-3 are useful fordiagnosis and follow-up of patients with acromegaly, the sensitivityand specicity of IGF-I or IGFBP-3 in the diagnosis of GH deciencyin children is a matter of controversy [14,15]. It is important to recog-nize the potential limitations of IGF-related assays because IGF-I andIGFBP-3 concentrations are highly age-dependent and may be inu-enced by nutrition and other factors [16]. Moreover, these peptidesshow marked variation even in healthy children, especially youngerones. It is thus imperative to achieve uniform interpretations ofIGF-I and IGFBP-3 levels by improving the standardization andinter-laboratory comparability of analysis procedures [17]. This in-cludes the establishment of normative data from representativepopulations using valid assays and considering non-GH-related fac-tors that are known to inuence IGF-I and IGFBP-3 concentrationslike age, sex, pubertal stage, and race.

The objective of the present study was to develop cross-sectionalreference values for IGF-I and IGFBP-3 in a normal population ofKorean children and adolescents.

Methods

Subjects and measurements

In total, 1378 Korean children and adolescents aged 0 to 17 years(722 boys, 656 girls) participated in this study. These subjects wererandomly selected from patients who visited outpatient clinics at 19

Table 1Demographic and anthropometric characteristics of subjects.

Boys (n=722) Girls (n=656)

Age (years) 9.04.5 8.64.6Height SD score 0.11.0 0.01.0Weigh SD score 0.01.7 0.32.3Body mass index (kg/m2) 19.91.6 20.51.3Serum IGF-I (ng/mL) 215.7156.3 252.4165.8Serum IGFBP-3 (ug/mL) 2.590.76 2.730.75medical college hospitals due to minor illnesses such as commoncold or mild gastroenteritis. They were ambulatory and had only

girls (b).girls (b).

Table 2IGF-I percentile values (ng/mL) according to age in Korean boys and girls.

Age(year)

Boys Girls

n 3rd 10th 25th 50th 75th 90th 97th n 3rd 10th 25th 50th 75th 90th 97th

0 34 6.6 8.7 15.7 36.7 80.3 134.3 268.8 19 5.1 7.1 40.0 52.9 84.7 130.3 186.51 35 12.8 18.1 25.3 56.3 82.2 94.3 127.2 25 39.6 43.6 59.3 76.4 117.0 153.7 195.32 36 8.1 26.0 40.6 83.0 124.2 161.4 182.3 34 24.1 39.4 63.2 105.0 155.6 170.8 206.03 37 32.2 38.0 63.6 97.2 134.2 204.7 255.4 36 26.2 46.9 96.3 141.8 162.5 214.7 294.94 33 7.9 54.9 77.6 129.1 152.4 202.4 253.8 31 29.9 43.8 102.6 151.4 205.3 239.7 359.45 41 42.0 74.3 114.3 153.1 201.7 227.2 245.1 42 70.2 115.9 135.3 169.3 203.5 237.6 300.36 58 26.6 58.7 96.4 142.6 177.5 211.7 515.2 53 70.3 97.1 147.7 192.2 231.9 280.5 330.97 47 90.6 120.3 138.8 177.3 192.5 241.3 268.8 61 102.3 138.5 163.1 213.5 255.4 321.8 460.18 59 50.3 123.6 137.5 184.8 224.5 250.9 265.9 78 93.5 150.6 196.7 232.1 287.0 339.3 441.39 53 60.2 109.2 159.7 183.0 232.0 278.6 324.9 65 91.0 155.8 199.4 242.9 280.5 356.9 499.810 69 119.4 148.6 177.3 211.9 257.8 306.4 454.7 57 50.5 113.8 204.8 275.4 482.2 599.6 807.411 55 96.2 137.1 187.8 250.9 331.9 484.2 599.1 49 156.1 181.3 252.1 368.8 521.6 624.9 657.512 42 102.1 184.4 207.2 263.4 495.1 660.0 681.0 32 196.2 294.8 359.5 446.8 564.5 670.6 812.013 37 145.2 175.1 290.3 399.2 538.7 586.7 647.8 23 238.0 332.6 442.4 496.4 538.8 587.7 819.514 27 111.4 152.6 326.5 457.2 558.6 609.9 770.5 25 135.1 231.1 340.3 419.2 508.3 589.7 662.415 28 215.7 283.7 359.4 430.3 530.3 566.8 614.6 15 138.5 204.1 309.7 427.5 517.8 590.4 607.816 16 112.5 181.4 279.6 348.8 497.8 557.4 665.7 7 253.9 253.9 350.7 421.5 489.3 536.6 536.617 8 237.3 237.3 312.5 365.8 447.4 475.6 475.6 2 370.2 370.2 370.2 395.5 420.9 420.9 420.9

18 S.E. Hyun et al. / Clinical Biochemistry 45 (2012) 1621was to construct a reference model for a normal population, so onlychildren with height and weight within2 SD of the populationmean were included [18]. Subjects with known endocrinologic disor-ders such as growth retardation, precocious puberty, diabetes, hypo-glycemia, and other chronic medical illness were excluded. TheEthical Committees of the participating Universities approved thestudy. Informed consent was obtained from the parents of eachchild and from all appropriately aged children.

All subjects had their height and weight measured. Body massindex was calculated as weight (kg) divided by height (m) squared.The height and weight standard deviation scores (SDS) were calculat-ed using the Korean children and adolescents growth standard [18].Pubertal staging was determined by endocrinologists at each hospitalaccording to Tanner stage criteria based on breast development ingirls and testicular volume in boys.

Blood samples were drawn at each participating hospital, and thesera were separated after centrifugation of coagulated blood andstored at 70 C until assay. Measurements of IGF-I and IGFBP-3levels were conducted using immunoradiometric assay (IRMA, Immu-notech, Marseilles, France) at one central laboratory within 3 monthsafter collection. IRMA is the assay based on the reversible and non-

covalent binding of an antigen by a specic antibody and labeled with

Table 3IGFBP-3 percentile values (ug/mL) according to age in Korean boys and girls.

Age(year)

Boys

n 3rd 10th 25th 50th 75th 90th 97th

0 34 0.62 0.89 1.26 1.56 1.76 2.05 2.211 35 1.15 1.29 1.45 1.73 1.94 2.11 2.222 36 1.12 1.26 1.57 1.81 2.10 2.34 2.593 37 1.18 1.33 1.74 1.92 2.23 2.50 2.734 33 1.20 1.52 1.74 2.14 2.44 2.60 3.125 41 1.62 1.88 2.00 2.22 2.44 2.90 3.056 58 1.25 1.68 2.02 2.33 2.50 2.98 3.437 47 1.62 1.99 2.25 2.52 2.68 3.03 3.498 59 1.82 2.05 2.35 2.57 2.81 3.42 3.739 53 2.06 2.23 2.47 2.60 3.11 3.63 4.1910 69 1.72 2.03 2.50 2.74 3.13 3.68 4.1011 55 2.12 2.32 2.66 3.06 3.34 3.55 4.9712 42 2.18 2.26 2.74 3.09 3.43 3.66 4.1613 37 2.20 2.77 3.02 3.27 3.69 4.17 4.4914 27 1.65 2.47 3.10 3.29 3.69 4.16 4.3515 28 2.35 2.73 2.99 3.25 3.71 4.00 4.8016 16 2.15 2.33 2.79 3.18 3.31 4.01 4.1517 8 2.46 2.46 2.84 3.20 3.34 4.23 4.23a radioactive nuclide 125I as a tracer. In IGF-I assay, sample and the dis-sociation buffer and tracer are added to themonoclonal antibody coatedtubes successively, and incubated for 1 h at 1825 C with shaking. Twomonoclonal antibodies highly specic for IGF-I with no cross-reactivityfor IGF-II bind two different epitopes of IGF-I, producing the sandwich-like antibody complex. After incubation, the content of tube is removedand bound radioactivity is measured. The bound radioactivity is directlyproportional to the IGF-I concentration in the sample. IGFBP-3 assay re-quires a 1:100 predilution of serumwith an assay-specic diluent. Pre-diluted sample and tracer are added to the coated tubes successively,and incubated for 3 h at 1825 C with shaking. Likewise with IGF-Iassay, sandwich complexes with two monoclonal antibodies againsttwo different epitopes of IGFBP-3 are formed. The intra-assay andinter-assay coefcients of variation were 6.3% and 6.8% for IGF-1, and6.0% and 9.5%, respectively, for IGFBP-3.

Statistical analysis

The serum concentrations of IGF-1 and IGFBP-3 according toage and pubertal stage in both sexes were recorded from thethird percentile to the 97th percentile using SAS 9.1. Additionally,

we constructed reference percentile curves for IGF-I and IGFBP-3

Girls

n 3rd 10th 25th 50th 75th 90th 97th

19 0.95 1.11 1.32 1.61 2.05 2.10 2.1225 1.43 1.53 1.69 1.89 2.07 2.36 2.5634 0.71 1.32 1.65 1.92 2.29 2.47 2.5436 1.27 1.49 1.78 2.13 2.50 2.77 2.8731 1.17 1.53 2.01 2.34 2.54 2.74 3.3442 1.80 1.97 2.17 2.42 2.77 3.01 3.0753 1.56 1.99 2.24 2.48 3.00 3.38 3.5661 1.81 2.07 2.38 2.74 3.14 3.40 3.9378 2.09 2.26 2.58 2.92 3.29 3.71 3.9865 2.10 2.31 2.57 2.85 3.39 3.86 4.2457 1.39 2.19 2.58 3.08 3.59 4.27 4.4549 2.40 2.60 2.85 3.16 3.59 4.04 4.3732 2.22 2.76 2.93 3.39 3.87 4.11 4.3423 2.57 2.97 3.05 3.50 3.76 3.99 4.3625 2.54 2.86 3.10 3.43 3.76 3.99 4.0815 2.14 2.65 3.06 3.37 3.55 3.75 5.797 2.53 2.53 2.85 3.14 3.52 4.34 4.342 3.22 3.22 3.22 3.23 3.24 3.24 3.24

was only higher in girls younger than 13. A steeper increase in serumIGF-I level was observed in girls between 9 and 10 years of age and in

Table 4IGF-I percentile values (ng/mL) according to pubertal stage in Korean boys and girls.

Tannerstage

Boys Girls

n 3rd 10th 25th 50th 75th 90th 97th n 3rd 10th 25th 50th 75th 90th 97th

1 514 12.8 40.3 89.5 147.4 198.3 245.0 283.3 429 32.5 59.8 120.7 173.9 229.4 280.5 351.42 85 124.1 162.6 226.9 289.8 372.5 464.7 575.7 76 90.8 180.9 232.5 286.7 363.5 499.8 581.93 41 128.5 195.9 241.3 380.0 460.1 599.1 664.9 54 162.6 275.0 361.7 448.5 558.0 610.6 670.64 25 68.5 265.9 402.9 538.7 614.6 689.9 770.5 43 204.1 340.3 413.5 502.7 584.9 662.4 818.85 57 112.5 262 318 443.9 543.0 577.8 641.9 54 138.5 277.6 353.0 449.2 518.8 610.3 700.7

Table 5IGFBP-3 percentile values (ug/mL) according to pubertal stage in Korean boys and girls.

Tannerstage

Boys Girls

n 3rd 10th 25th 50th 75th 90th 97th n 3rd 10th 25th 50th 75th 90th 97th

1 514 1.22 1.53 1.89 2.32 2.67 3.12 3.63 429 1.32 1.64 2.07 2.45 2.86 3.25 3.782 85 2.01 2.33 2.72 3.03 3.32 3.63 4.14 76 2.08 2.43 2.62 3.12 3.52 4.05 4.243 41 2.47 2.74 3.10 3.27 3.55 4.13 4.49 54 2.57 2.80 2.98 3.38 3.85 4.05 4.574 25 1.56 2.85 3.27 3.60 3.80 4.09 4.23 43 2.54 2.86 3.06 3.40 3.83 4.11 4.415 57 2.33 2.61 2.97 3.17 3.64 4.16 4.35 54 2.54 2.87 3.10 3.38 3.74 4.20 4.34

19S.E. Hyun et al. / Clinical Biochemistry 45 (2012) 1621based on age and pubertal stage in both sexes using the quantileregression analysis by R 2.8.1 program (Bell Laboratories). The in-dependent t-test was used to compare the levels of age-matchedboys and girls. Multiple linear regression analysis was used to esti-mate the effects of age and pubertal stage on the levels of IGF-1and IGFBP-3. We considered a p value less than 0.05 to be statisti-cally signicant.

Results

Demographic and anthropometric characteristics of subjects

The height standard deviation score was 0.11.0 and 0.01.0in boys and girls, respectively. And the weight standard deviationscore was 0.01.7 and 0.32.3 in boys and girls, respectively. BMI(kg/m) was 19.91.6 and 20.51.3 in boys and girls, respectively(Table 1).

Distributions of serum IGF-1 and IGFBP-3 levels according to age

The reference values for IGF-1 and IGFBP-3 according to age were

constructed with percentile denitions as curves (Figs. 1 and 2) and

Table 6Mean and SD for IGF-I level (ng/mL) according to age in Korean boys and girls.

Age(year)

Boys

n 2SD 1SD Mean 1SD 2SD

0 34 5.00 5.00 63.71 139.34 214.981 35 13.00 25.76 57.92 90.08 122.252 36 5.00 34.97 85.71 136.45 187.203 37 25.09 31.00 117.13 209.17 301.214 33 11.07 66.93 122.79 178.65 234.515 41 33.87 94.34 154.81 215.28 275.756 58 7.00 51.88 150.71 249.54 348.377 47 74.77 123.28 171.79 220.30 268.828 59 77.50 130.74 183.98 237.23 290.479 53 60.08 126.35 192.62 258.88 325.1510 69 59.80 142.25 224.71 307.16 389.6211 55 70.40 138.41 276.42 414.42 552.4312 42 89.00 167.86 348.41 528.95 709.5013 37 97.70 249.16 400.62 552.09 703.5514 27 71.53 252.54 433.56 614.57 795.5815 28 214.79 324.43 434.08 543.72 653.3716 16 75.44 225.34 375.24 525.14 675.0417 8 204.53 287.54 370.55 453.56 536.57boys 11 and 12 years of age. In contrast, serum IGFBP-3 levels increasedsteadily. Tables 6 and 7 show reference values for IGF-I and IGFBP-3,presented as meanSD.

Distribution of serum IGF-1 and IGFBP-3 levels according to pubertal stage

The reference values for IGF-1 and IGFBP-3 according to pubertalstage were constructed with percentile denitions as curves (Figs. 3and 4) and tables (Tables 4 and 5). Serum IGF-I concentrations werehighest at Tanner stage IV in boys and girls with subsequent declinesin both sexes. Serum IGFBP-3 levels showed the same pattern in bothtables (Tables 2 and 3). Overall, serum IGF-I and IGFBP-3 levelswere increased continuously from early childhood into adolescence.Peaks of serum IGF-I and IGFBP-3 levelswere observed two years earlierin girls than they were in boys (13 vs. 15 years of age, respectively).After pubertal peaks, IGF-I and IGFBP-3 levels declined with increasingage. The serum IGFBP-3 level was higher in girls compared to that inboys of the same ages throughout the pubertal period, whereas IGF-Isexes.

Girls

n 2SD 1SD Mean 1SD 2SD

19 5.00 20.91 62.84 104.76 146.6825 5.98 46.69 92.40 138.11 183.8234 6.00 50.76 108.09 165.41 222.7436 12.15 63.19 138.52 213.85 289.1831 3.26 79.03 154.81 230.58 306.3642 61.82 116.60 171.37 226.14 280.9253 38.13 117.75 197.38 277.00 356.6261 51.90 137.18 222.47 307.76 393.0478 70.60 157.25 243.90 330.55 417.2065 58.02 152.71 247.40 342.10 436.7957 6.00 141.27 337.23 533.20 729.1649 78.04 237.07 396.10 555.13 714.1632 166.64 317.67 468.71 619.75 770.7923 249.07 368.78 488.48 608.18 727.8825 158.09 288.80 419.51 550.22 680.9215 130.31 273.24 416.17 559.10 702.037 226.02 319.38 412.74 506.11 599.472 323.85 359.70 395.55 431.40 467.25

differences or imprecise staging for pubertal maturation in boys. Itis well known that pubertal stage inuences the serum IGF-1 andIGFBP-3 levels more closely does than chronological age. The age ofonset and tempo of puberty can vary, even in healthy children. There-fore, if the reference values of serum IGF-I and IGFBP-3 are basedsolely on chronological age in the pubertal period, the interpretationof serum levels of IGF-I and IGFBP-3 may be misleading. It is recom-mended that serum IGF-I and IGFBP-3 levels be interpreted usingthe reference values based on pubertal stage as well as chronological

Girls

2SD n 2SD 1SD Mean 1SD 2SD

2.33 19 0.84 1.23 1.61 1.99 2.382.40 25 1.32 1.61 1.91 2.20 2.492.66 34 0.90 1.41 1.92 2.44 2.952.78 36 1.18 1.65 2.13 2.61 3.093.00 31 1.31 1.78 2.26 2.73 3.203.11 42 1.66 2.05 2.44 2.83 3.223.33 54 1.52 2.04 2.56 3.08 3.603.39 62 1.66 2.21 2.76 3.31 3.873.62 78 1.91 2.43 2.96 3.48 4.003.89 65 1.71 2.32 2.94 3.55 4.164.05 57 1.59 2.34 3.10 3.85 4.614.22 49 2.16 2.71 3.26 3.81 4.364.39 32 2.31 2.85 3.38 3.92 4.464.60 23 2.56 3.02 3.48 3.94 4.404.55 25 2.52 2.96 3.40 3.85 4.294.44 15 1.77 2.56 3.35 4.14 4.934.20 7 2.12 2.70 3.28 3.85 4.434.24 2 3.20 3.22 3.23 3.24 3.25

20 S.E. Hyun et al. / Clinical Biochemistry 45 (2012) 1621Discussion

In this study, reference values of serum IGF-I and IGFBP-3 wereestablished in a large cohort of Korean children and adolescents. Pre-vious studies that have accomplished this in different populationsincluding those of Spain, Sweden, Denmark, Germany, Turkey, andChina have shown that serum IGF-1 and IGFBP-3 concentrationswere inuenced by sex, age, and pubertal stage [1925]. We con-ducted a survey in healthy Korean children and adolescents aged0 to 18 years and showed that IGF-1 and IGFBP-3 levels are indeedinuenced by age, sex and sexual maturation.

Age and sex are the representative variables that are associatedwith changes in serum IGF-I and IGFBP-3 concentrations. SerumIGF-I and IGFBP-3 concentrations are low at birth and rise progres-sively in males and females during childhood. With the onset ofpuberty, the increase in IGF-I concentration accelerates and peaksaround 1214 years of age in girls and around 1316 years inboys [1925]. In contrast, serum IGFBP-3 levels have been reportedto reach a peak in adolescents aged between 13 and 16.5 years [1925].Following puberty, IGF-1 and IGFBP-3 concentrations decline and thenmaintain a constant level during early adulthood in both boys and

Table 7Mean and SD for IGFBP-3 level (ug/mL) according to age in Korean boys and girls.

Age(year)

Boys

n 2SD 1SD Mean 1SD

0 34 0.68 1.09 1.51 1.921 35 1.02 1.36 1.71 2.052 36 0.98 1.40 1.82 2.243 37 1.09 1.51 1.93 2.364 33 1.22 1.66 2.11 2.555 41 1.44 1.86 2.27 2.696 58 1.28 1.79 2.30 2.827 47 1.61 2.05 2.50 2.948 60 1.63 2.13 2.63 3.129 54 1.69 2.24 2.79 3.3410 69 1.61 2.22 2.83 3.4411 55 1.89 2.48 3.06 3.6412 43 1.66 2.34 3.02 3.7013 37 2.11 2.73 3.35 3.9714 28 2.07 2.69 3.31 3.9315 28 2.25 2.80 3.35 3.8916 16 2.04 2.58 3.12 3.6617 8 2.13 2.65 3.18 3.71girls [2629]. The similar changing patterns of serum IGF-I and IGFBP-3 during puberty have also been observed in some longitudinal studies[28,29]. In the present study, peaks of serum IGF-I and IGFBP-3 levelswere observed at 13 years of age in girls and 15 years of age in boys,which is compatible with the results of previous studies. We observedthat serum IGF-I levels increase steeply between the ages of 9 and10 years in girls and 11 and 12 years in boys. Onset of a steeper increaseand peak in IGF-I levels occurs two years earlier in girls than it does inboys, which seems to be consistent with the time lag for peak growthrates between sexes. For IGFBP-3 levels, no similar steep increase wasobserved. Results from previous studies demonstrating that IGF-I ismore sensitive to GH than IGFBP-3 suggest a possible explanation forthis nding [26,30].

In addition to sex and age, pubertal stage is an important factorinuencing the changes in IGF-1 and IGFBP-3 levels in children andadolescents. This phenomenon seems to occur due to the increasesin gonadal steroids and GH secretion [31]. It has been reported thatIGF-I levels peak at Tanner stage IV in girls and at stage V in boys[19,20,23]. Peak IGFBP-3 levels were observed at Tanner stage IV inboth sexes in previous studies [10,23,26]. In contrast, our resultsshowed that peak IGF-I and IGFBP-3 levels occurred at Tanner stageIV in both sexes, which is similar to the results reported in some ofthe literature [21]. These differences might be explained by ethnicage.Many studies that have been performed in different populations,

including ours, have shown that the reference values of serum IGF-I

a

400500600700800900

F-I

(ng/m

L) 97th90th75th50th

25thb

0100200300

1 2 3 4 5

IGIG

F-I

(ng/m

L)

Tanner Stage

10th

3rd

0100200300400500600700800900

1 2 3 4 5Tanner Stage

97th

90th

75th50th25th10th

3rd

Fig. 3. Reference percentile curves for IGF-I according to pubertal stage in Korean boys(a) and girls (b).

Tanner Stage

L 75th

21S.E. Hyun et al. / Clinical Biochemistry 45 (2012) 1621and IGFBP-3may be somewhat different. These differences may resultfrom variability in ethnicity, methodology of measurement, and dif-

Tanner Stage

0

1

2

3

1 2 3 4 5

IGFB

P-3

(ug/m 50th25th

10th3rd

Fig. 4. Reference percentile curves for IGFBP-3 according to pubertal stage in Koreanboys (a) and girls (b).b

4

5

)

97th90tha

0

1

2

3

4

5

1 2 3 4 5

IGFB

P-3

(ug/m

L)97th90th

75th50th25th10th3rdferences between laboratories. However, these studies show a similarchanging pattern for serum IGF-I and IGFBP-3 levels during puberty.This is the reason why it is necessary to construct an ethnicity-specic model to establish the reference values of IGF-1 and IGFBP-3.Accurate and reliable screening and monitoring of serum IGF-1 andIGFBP-3 concentrations are possible with ethnicity-specic data.

In conclusion, establishment of the normal ranges of serum IGF-Iand IGFBP-3 levels is essential for diagnostic work-up and manage-ment of children with growth problems. These reference values forserum IGF-I and IGFBP-3 according to age, sex and pubertal stagemay be helpful for assessing and managing Korean children and ado-lescents with growth disorders.

Acknowledgments

This work was supported by a grant from the Korean Academy ofMedical Sciences (to HK).

References

[1] Daughaday WH, Rotwein P. Insulin-like growth factors I and II. Peptide, messen-ger ribonucleic acid and gene structures, serum and tissue concentrations. EndocrRev 1989;10:6891.

[2] Baxter RC, Martin JL. Binding proteins for the insulin-like growth factors: struc-ture, regulation and function. Prog Growth Factor Res 1989;1:4968.

[3] Jones JI, Clemmons DR. Insulin-like growth factors and their binding proteins:biological actions. Endocr Rev 1995;16:334.

[4] Le Roith D, Bondy C, Yakar S, Liu JL, Butler A. The somatomedin hypothesis: 2001.Endocr Rev 2001;22:5374.

[5] Oh Y, Muller HL, Lamson G, Rosenfeld RG. Insulin-like growth factor (IGF)-independent action of IGF-binding protein-3 in Hs578T human breast cancercells. Cell surface binding and growth inhibition. J Biol Chem 1993;268:1496471.

[6] Kim HS, Rosenfeld RG, Oh Y. Biological roles of insulin-like growth factor bindingproteins (IGFBPs). Exp Mol Med 1997;29:8596.[7] Rosenfeld RG, Wilson DM, Lee PDK, Hintz RL. Insulin-like growth factors I and II inthe evaluation of growth retardation. J Pediatr 1986;109:42833.

[8] Rosenfeld RG, Cohen P. Disorders of growth hormone/insulin-like growth factorsecretion and action. Sperling MA. Pediatric Endocrinology, 3 rd edPhiladelphia,Pennsylvania: Saunders; 2008. p. 254334.

[9] Zapf J, Walter H, Froesch ER. Radioimmunological determination of insulin likegrowth factor I and II in normal subjects and in patients with growth disordersand extrapancreatic tumor hypoglycemia. J Clin Invest 1981;68:132130.

[10] Ranke MB, Schweizer R, Elmlinger MW, Weber K, Binder G, Schwarze CP, et al.Signicance of basal IGF-I, IGFBP-3 and IGFBP-2 measurements in the diagnosticsof short stature in children. Horm Res 2000;54:608.

[11] Blum WF, Ranke MB, Kietzmann K, Gauggel E, Zeisel HJ, Bierich JR. A specicradioimmunoassay for the growth hormone-dependent somatomedin-bindingprotein: its use for diagnosis for GH deciency. J Clin Endocrinol Metab 1990;70:12928.

[12] Kristrom B, Jansson C, Rosberg S, Albertsson-Wikland K. Growth response togrowth hormone (GH) treatment relates to serum insulin-like growth factor I(IGF-I) and IGF-binding protein-3 in short children with various GH secretioncapacities. Swedish Study Group for Growth Hormone Treatment. J Clin Endocri-nol Metab 1997;82:288998.

[13] Wikland KA, Kristrom B, Rosberg S, Svensson B, Nierop AF. Validated multivariatemodels predicting the growth response to GH treatment in individual shortchildren with a broad range in GH secretion capacities. Pediatr Res 2000;48:47584.

[14] Bussieres L, Souberbielle JC, Pinto G, Adan L, Noel M, Brauner R. The use of insulin-like growth factor 1 reference values for the diagnosis of growth hormonedeciency in prepubertal children. Clin Endocrinol (Oxf) 2000;52:7359.

[15] Tillmann V, Buckler JM, Kibirige MS, Price DA, Shalet SM, Wales JK, et al. Biochem-ical tests in the diagnosis of childhood growth hormone deciency. J Clin Endocri-nol Metab 1997;82:5315.

[16] Clemmons DR, Klibanski A, Underwood LE, McArthur JW, Ridgway EC, Beitins IZ,et al. Reduction of plasma immunoreactive somatomedin C during fasting inhumans. J Clin Endocrinol Metab 1981;53:124750.

[17] Ranke MB, Feldt-Rasmussen U, Bang P, Baxter RC, Camacho-Hubner C, ClemmonsDR, et al. How should insulin-like growth factor I be measured? A consensusstatement. Horm Res 2001;55:1069.

[18] Korea Center for Disease Control and Prevention, The Korean Pediatric Society,The Committee for the Development of Growth Standard for Korean Childrenand Adolescents. 2007 Korean children and adolescents growth standard (com-mentary for the development of 2007 growth chart). [Government report online].Seoul: Division of Chronic Disease Surveillance; 2007 Nov. Available from: URL;//http://www.cdc.go.kr/. and URL;//http://www.pediatrics.or.kr/.

[19] Argente J, Barrios V, Pozo J, Munoz MT, Hervas F, Stene M, et al. Normative data forinsulin-like growth factors (IGFs), IGF-binding proteins, and growth hormone-binding protein in a healthy Spanish pediatric population: age- and sex-relatedchanges. J Clin Endocrinol Metab 1993;77:15228.

[20] Lofqvist C, Andersson E, Gelander L, Rosberg S, Blum WF, Albertsson Wikland K.Reference values for IGF-I throughout childhood and adolescence: a model thataccounts simultaneously for the effect of gender, age, and puberty. J Clin Endocri-nol Metab 2001;86:58706.

[21] Elmlinger MW, Kuhnel W, Weber MM, Ranke MB. Reference ranges for two auto-mated chemiluminescent assays for serum insulin-like growth factor I (IGF-I) andIGF-binding protein 3 (IGFBP-3). Clin Chem Lab Med 2004;42:65464.

[22] Lofqvist C, Andersson E, Gelander L, Rosberg S, Hulthen L, Blum WF, et al. Refer-ence values for insulin-like growth factor-binding protein-3 (IGFBP-3) and theratio of insulin-like growth factor-I to IGFBP-3 throughout childhood and adoles-cence. J Clin Endocrinol Metab 2005;90:14207.

[23] Bereket A, Turan S, Omar A, Berber M, Ozen A, Akbenlioglu C, et al. Serum IGF-Iand IGFBP-3 levels of Turkish children during childhood and adolescence: establish-ment of reference ranges with emphasis on puberty. Horm Res 2006;65:96105.

[24] Kong AP, Wong GW, Choi KC, Ho CS, Chan MH, Lam CW, et al. Reference values forserum levels of insulin-like growth factor (IGF-1) and IGF-binding protein 3(IGFBP-3) and their ratio in Chinese adolescents. Clin Biochem 2007;40:10939.

[25] Xu S, Gu X, Pan H, Zhu H, Gong F, Li Y, et al. Reference ranges for serum IGF-1 andIGFBP-3 levels in Chinese children during childhood and adolescence. Endocr J2010;57:2218.

[26] Juul A, Dalgaard P, Blum WF, Bang P, Hall K, Michaelsen KF, et al. Serum levels ofinsulin-like growth factor (IGF)-binding protein-3 (IGFBP-3) in healthy infants,children, and adolescents: the relation to IGF-I, IGF-II, IGFBP-1, IGFBP-2, age,sex, body mass index, and pubertal maturation. J Clin Endocrinol Metab1995;80:253442.

[27] Juul A, Bang P, Hertel NT, Main K, Dalgaard P, Jorgensen K, et al. Serum insulin-likegrowth factor-I in 1030 healthy children, adolescents, and adults: relation to age,sex, stage of puberty, testicular size, and body mass index. J Clin Endocrinol Metab1994;78:74452.

[28] Cara JF, Roseneld RL, Furlanetto RW. A longitudinal study of the relationship ofplasma somatomedin-C concentration to the pubertal growth spurt. Am J DisChild 1987;141:5624.

[29] Casazza K, Higgins PB, Fernandez JR, Goran MI, Gower BA. Longitudinal analysisof the insulin-like growth factor system in African-American and EuropeanAmerican children and adolescents. J Clin Endocrinol Metab 2008;93:491723.

[30] Blum WF, Albertsson-Wikland K, Rosberg S, Ranke MB. Serum levels of insulin-like growth factor I (IGF-I) and IGF binding protein 3 reect spontaneous growthhormone secretion. J Clin Endocrinol Metab 1993;76:16106.

[31] Ho KK, O'Sullivan AJ, Weissberger AJ, Kelly JJ. Sex steroid regulation of growthhormone secretion and action. Horm Res 1996;45:6773.

Reference values for serum levels of insulin-like growth factor-I and insulin-like growth factor binding protein-3 in Korean children and adolescentsIntroductionMethodsSubjects and measurementsStatistical analysis

ResultsDemographic and anthropometric characteristics of subjectsDistributions of serum IGF-1 and IGFBP-3 levels according to ageDistribution of serum IGF-1 and IGFBP-3 levels according to pubertal stage

DiscussionAcknowledgmentsReferences