DOI: 10.1530/EJE-07-0338
European Journal of Endocrinology, Vol 157, Issue 5, 597-603
Copyright © 2007 by European Society of Endocrinology
Central laboratory reassessment of IGF-I, IGF-binding protein-3, and GH serum concentrations measured at local treatment centers in growth-impaired children: implications for the agreement between outpatient screening and the results of somatotropic axis functional testing
Berthold P Hauffa1,
Nils Lehmann2,
Markus Bettendorf3,
Otto Mehls3,
Helmuth-Günther Dörr4,
Nikolaus Stahnke5,
Heinz Steinkamp6,
Elfriede Said6,
Michael B Ranke7 and
participating members of the German KIGS Board/Medical Outcome Study Group
1 Division of Pediatric Endocrinology and Diabetology, University Children's Hospital, Heidelberg, Germany
2 Institute for Medical Informatics, Biometry and Epidemiology, University of Duisburg-Essen, Essen, Germany
3 University Children's Hospital, Heidelberg, Germany
4 University Hospital for Children and Adolescents, Erlangen, Germany
5 Endokrinologikum, Hamburg, Germany
6 Pfizer Pharma GmbH, Karlsruhe, Germany
7 Pediatric Endocrinology Section, University Children's Hospital, Tuebingen, Germany
(Correspondence should be addressed to B P Hauffa who is now at Abteilung für Pädiatrische Endokrinologie und Diabetologie, Zentrum für Kinderheilkunde und Jugendmedizin, Universität Duisburg-Essen, Hufelandstr. 55, D-45122 Essen, Germany Email: berthold.hauffa{at}uk-essen.de)
 |
Abstract
|
|---|
Background: Childhood GH deficiency, suspected in the presence of decreased height velocity and short stature, is usually characterized by low IGF-I and IGF-binding protein-3 (IGFBP-3) serum concentrations and is conventionally confirmed by diminished GH peak responses to pharmacological stimuli.
Objective: We evaluated the agreement between different IGF-I (IGFBP-3) assays in predicting GH deficiency and tested whether variability between growth factor screening and pharmacological testing could be diminished by reassessment of growth factor and GH peak concentrations in a single laboratory.
Design: Using the Tuebingen IGF-I (IGFBP-3) RIA, 317 (321) sera from children evaluated for growth disorders in 19 centers were reanalyzed. In 103 children with insulin hypoglycemia and arginine tests, we evaluated how the association between the outcome of growth factor screening and functional testing would change if different assays were employed.
Results: Locally measured IGF-I correlated better than IGFBP-3 with the results of the central laboratory (Tuebingen) assay (slope of the regression curve 1.05; 95% confidence interval (95% CI) 1.01–1.1 versus 1.18; 95% CI 1.09–1.3). Agreement between local and central laboratory assays in predicting GH deficiency was better for IGF-I than for IGFBP-3 assays (
=0.59 versus =0.47). The poor agreement between growth factor screening and GH pharmacological testing was not improved when hormone concentrations were remeasured in the central laboratory ( local=–0.0031, central=0.12).
Conclusions: In children with impaired growth, growth factor screening reflects different aspects of GH insufficiency than does functional testing. Agreement between these approaches is poor and could not be improved by reduction of assay-related variability.
|
Introduction
|
|---|
Growth hormone (GH) deficiency is suspected in children with decreased height velocity and short stature, characterized by low insulin-like growth factor (IGF-I) and IGF-binding protein-3 (IGFBP-3) serum concentrations in most patients, and conventionally confirmed by diminished GH peak responses to pharmacological stimuli (1). The agreement between the results of growth factor screening and GH pharmacological testing is highly variable (2–19). Part of this variability is introduced by the use of differently designed hormone assays, using different antibodies and standards, both for measurement of GH and IGF-I or IGFBP-3. We have shown previously that regional use of different GH assays and cutoff values increases the variability not only of hormone measurements, but also of diagnostic and therapeutic decisions regarding growth hormone deficiency (20). Similar restrictions could apply to the measurement of growth factors, often employed in screening for GH deficiency.
Therefore, we wanted to determine the agreement between IGF-I (IGFBP-3) serum concentrations measured in various treatment centers, and measurements performed on the same sera in one central laboratory. We further asked if the agreement of patient diagnostic assignment (‘GH deficient’ versus ‘non-GH deficient’) using growth factor screening alone, or comparing the results of growth factor screening and GH pharmacological testing, would change when different assays, each using its own established reference range, were employed.
|
Subjects and methods
|
|---|
Samples and subjects
Twenty-one German secondary and tertiary care centers diagnosing and treating children with growth disorders participated in the study. From January 1, 1998 to December 31, 1999, all consecutive children and adolescents who underwent functional testing of the somatotropic axis or evaluation of spontaneous GH secretion were considered for inclusion. First, IGF-I, IGFBP-3 concentrations (basal samples), and GH concentrations (basal and post-stimulation samples) were measured in each center, using the local assays. If serum was available from the basal sample and the post-stimulation sample containing the peak GH concentration, these were sent to the reference laboratory (University Children's Hospital, Tuebingen) for re-evaluation.
Here, 317 (321) basal sera whose IGF-I (IGFBP-3) concentrations had been measured locally were available. Measurement results were classified as low, normal or high using the local reference ranges, which differed between assays. Analyte concentrations were then remeasured in the central laboratory using the in-house Tuebingen RIAs for IGF-I and IGFBP-3.
Here, 790 GH peak sera of 434 patients were reassayed centrally. Ninety-one of these results (52 patients, 2 centers) were excluded from the analysis for various reasons (20). Thus, 699 peak GH sera of 382 children and adolescents (258 males and 124 females) from 19 centers were available for analysis. From these, a subgroup of 132 patients with the most abundant test combination (insulin hypoglycemia test and arginine test) were evaluated for changes in the assignment to the diagnostic group of GH deficiency, which occurred with the use of different assays. Patients were diagnosed as GH deficient if the GH concentration remained below the assay-specific cutoff value in both tests. Otherwise, the patients were categorized as non-GH deficient. In 103 of these patients, measurements of IGF-I and IGFBP-3 were available for comparison.
Ethical approval
The study protocol was reviewed and approved by the Ethics Committee for Medical Research of the University of Tuebingen.
Analytical methods
Assays for IGF-I and IGFBP-3 used in the participating centers (local assays) and their characteristics are detailed in Table 1. GH assays were used as already described (20). Samples were stored frozen for various lengths of time not exceeding 6 months at –20 °C in 300 µl aliquots and sent to the central laboratory at the University Children's Hospital, Tuebingen (M B R), where IGF-I, IGFBP-3, and GH concentrations were reassessed in duplicate using in-house assays, as already described (16). Analytes in serum as measured with the Tuebingen assays were reported to be stable under the storage conditions over a time period of up to 1 year.
View this table:
[in this window]
[in a new window]
|
Table 1 Characteristics and frequency of use of different insulin-like growth factor-I (IGF-I) and IGF-binding protein-3 (IGFBP-3) immunoassay methods employed by the participating centers.
|
|
The central laboratory (M B R) was blinded to the origin of the samples, the type of assay used, the hormone measurements obtained at the respective centers, and the cutoff levels employed for the diagnosis of GH deficiency at these centers. The evaluating center (B P H) was blinded to the name of the referring center.
Statistical methods
Analytical results obtained at the different local centers were combined and compared pairwise with the results of the central laboratory by the Bland–Altman plot (21) and the Passing–Bablok regression (22), using MedCalc software, Maria Verke, Belgium (23). These methods are nonparametric and do not require one assay to be assigned the role of an analytical gold standard. Whether the local assays and each of the assays used by the central laboratory agreed in the assignment of a patient to a diagnostic group (GH deficient versus non-GH deficient) was assessed using the McNemar test and the statistic (24). A value between 0.4 and 0.6 indicates moderate agreement, between 0.6 and 0.8 indicates clear agreement, and between 0.8 and 1.0 indicates high agreement. The latter analyses were performed using SAS software version 8.2, Cary, NC, USA (25).
|
Results
|
|---|
In general, central laboratory measurements of serum IGF-I concentrations showed a good agreement with the measurements obtained at the local laboratories (regression equation of central laboratory measurements (x) with pooled local measurements (y): y=–6.8+1.05x; Fig. 1A and C). There was no significant deviation of the regression line from linearity. Bland–Altman analysis showed a negligible bias of –1.5 ng/ml. The ±1.96 S.D. range of the measurement differences, however, was comparably large. It spanned 170.2 ng/ml and approached one-third of the observed concentration range, with greater differences between results at higher analyte concentrations. A slightly lesser agreement between local and central laboratory measurements was found for IGFBP-3 (regression equation: y=–0.9+1.18x; Fig. 1B and D). Bias was 0.36 mg/l, and the ±1.96 S.D. range was 3.0 mg/l, approaching about 45% of the IGFBP-3 concentration range of the samples. The regression line did not deviate from linearity. Regression equations of measurements obtained with individual local assays and the central laboratory assays are shown in Table 2.
View larger version (29K):
[in this window]
[in a new window]
|
Figure 1 Comparison of IGF-I (A and C) and IGFBP-3 (B and D) concentrations measured in 19 local centers and remeasured in a central laboratory (Tuebingen). In A and B, for each sample analyte, concentrations measured in the central laboratory (x-axis) are plotted against the concentrations measured in the local laboratories (y-axis). Solid lines represent the regression lines according to Passing and Bablok, with the 95% confidence intervals indicated by dotted lines. C and D display the Bland–Altman plots, where the mean of the local and central measurements (x-axis) is plotted against the difference of these values (y-axis). Horizontal lines indicate the mean and ±1.96 S.D. range of these differences.
|
|
View this table:
[in this window]
[in a new window]
|
Table 2 Regression equations and 95% confidence intervals (CI) of intercept and slope of the pairwise method comparisons of the local insulin-like growth factor-I (IGF-I) and IGF-binding protein-3 (IGFBP-3) measurements (y) versus the central laboratory (Tuebingen) assay results (x), calculated with the Passing–Bablok method (22). Because of similarities in assay design and lack of difference with regard to the regression equation, the data of both Nichols assays (and the data of the Mediagnost assay and the in-house assay Gießen) were pooled.
|
|
When either IGF-I or IGFBP-3 results were used separately for predicting GH deficiency (analyte decreased, GH deficiency possible; analyte normal or elevated, GH deficiency unlikely), agreement between predictions obtained with local and central laboratory assays was better for the IGF-I assay than for the IGFBP-3 assay (=0.59 versus =0.47). When IGF-I and IGFBP-3 measurements were combined to sort the outcomes of the growth factor screening into three categories (IGF-I and IGFBP-3 low, IGF-I or IGFBP-3 low, and IGF-I and IGFBP-3 normal or high), the distribution of the patients between these categories obtained with the use of local assays (Table 3) agreed moderately (=0.46, 95% CI 0.37–0.58) with the distribution seen with the use of the central laboratory assays.
View this table:
[in this window]
[in a new window]
|
Table 3 Comparison between the categorization of patients (n=312) according to their insulin-like growth factor-I (IGF-I) and IGF-binding protein-3 (IGFBP-3) concentrations measured in the local centers and remeasured in a central laboratory (Tuebingen), showing moderate agreement. *=kappa, 95% CI=95% confidence interval for .
|
|
The agreement between growth factor/binding protein screening (suspected GH deficiency: both analytes below reference range) and GH pharmacological testing (GH deficiency: GH below cutoff in two tests) was poor for a group of 103 children and adolescents, whether the comparison was based on the data obtained with the local or central laboratory assays. Agreement did not improve when the hormone concentrations remeasured in the central laboratory were used for comparison ( local=–0.0031; central=0.12; Table 4). When less stringent criteria were employed for suspected GH deficiency (either IGF-I or IGFBP-3 below reference range), agreement between results of growth factor screening and GH functional testing was still low, when local measurements were considered (IGF-I low: local=0.035; IGFBP-3 low: local=–0.0008). There was only a minor improvement of this agreement, when the analysis was based on the central laboratory measurements (IGF-I low: central=0.17; IGFBP-3 low: central=0.14).
View this table:
[in this window]
[in a new window]
|
Table 4 Comparison between the prediction of growth hormone deficiency in 103 children and adolescents with growth disorders using insulin-like growth factor-I (IGF-I) and IGF-binding protein-3 (IGFBP-3) concentrations (growth factor screening) and the results of growth hormone functional testing (insulin hypoglycemia and arginine infusion) measured in the local centers (A) and remeasured in a central laboratory (Tuebingen, B), showing lack of agreement. *=kappa, 95% CI=95% confidence interval for .
|
|
|
Discussion
|
|---|
We have shown that IGF-I and IGFBP-3 concentrations measured locally in 19 German treatment centers correlated well with the assays used in a central laboratory (Tuebingen assay). The correlation was better for IGF-I than for IGFBP-3. When children with growth disturbances were categorized according to their IGF-I and IGFBP-3 concentrations, a moderate agreement was found between local and central assays with regard to assigning individual patients to these categories. There was, however, no agreement concerning the assignment of patients to the diagnostic categories ‘growth hormone deficient’ or ‘non-growth hormone deficient’, between the results of growth factor screening and GH functional testing. The agreement did not improve when only hormone concentrations measured in the central laboratory were used for analysis.
Correlations between locally and centrally measured IGF-I and IGFBP-3 concentrations were better than those reported by us for GH, obtained from the same study population (20). This is due in part to a greater procedural standardization of commercially available IGF-I and IGFBP-3 assays. All commercially available IGF-I assays in this study used a human recombinant IGF-I standard calibrated against the International Reference Reagent WHO 87/518. Although this procedure has been criticized as a least common denominator needing improvement (26), it constitutes a more desirable situation than for GH, where different and outdated standards are still used for assay calibration. In growth factor measurement, major differences exist with regard to separation techniques. In our study, however, these differences did not prohibit a rather good agreement between assays, as judged by the similarity of the regression equations. In a previous study, almost identical regression curves were obtained, when IGF-I concentrations in the sera of 427 adult blood donors, measured with four of the assays used in this study (Tuebingen RIA, Mediagnost RIA, DSL IRMA, Nichols Advantage IRMA), were compared (27). Not surprisingly, this agreement between results obtained with different IGF-I assays was reflected in our study with a moderate agreement between the outcomes of diagnostic categorization, not only when IGF-I but also when IGFBP-3 measurements were considered.
IGF-I and IGFBP-3 measurements are commonly employed in the diagnosis of GH deficiency. In a 1995 survey, 88% of pediatric endocrinologists in the USA used IGF-I measurements in the diagnostic work up of children with suspected GH deficiency, at least occasionally or on a regular basis (28). Recent editions of major textbooks of endocrinology include the measurement of IGF-I and IGFBP-3, alone or in combination, in their algorithms for work up of children with short stature (29, 30), mostly for screening purposes. The appropriate interpretation of the results of this growth factor screening, however, is subject to considerable debate. If GH deficiency is defined by the failure of GH to rise above a predefined value in two independent pharmacological or physiological tests (1), then reported cutoff-dependent sensitivities of the IGF-I (IGFBP-3) measurements to predict GH deficiency range from 34 to 96% (14.9 to 97%) with a specificity ranging from 32 to 97.9% (50 to 97.9%) (5–7, 9, 11, 13, 15, 31–33). The ability of these measurements to serve as a surrogate parameter for GH functional testing in the diagnosis of GH deficiency has been estimated from being high (34) to nonexistent (15). Our data seem to support the latter notion, since there was no agreement between the results of growth factor screening and GH testing. However, we had used the most restrictive criteria for prediction of GH deficiency, requiring that both IGF-I and IGFBP-3 be low. But even if less stringent criteria were employed, agreement improved only slightly with all <0.2 and significant variability remained. This could be explained by the fact that some local investigators had used information for their decision to proceed with GH functional testing not contained in the IGF-I and IGFBP-3 data and unknown to us, e.g. a very poor growth rate with IGF-I and IGFBP-3 concentrations just above the lower limit of normal. This would generate a higher proportion than expected of patients discordantly combining IGF-I and IGFBP-3 values not indicative of GH deficiency with the results of GH functional testing compatible with GH deficiency. This observation implies that in many situations the information provided by IGF-I and IGFBP-3 may be canceled out by the clinical findings during the process of diagnostic decision making.
Further, biological and analytical factors have been noted to explain the poor agreement. Circulating concentrations of IGF-I and IGFBP-3 depend not only on GH secretion, but also on GH-independent causes like genetic factors (35, 36), nutritional status (37, 38), hepatic and renal function (39–41), or underlying diseases presenting with varying degrees of GH resistance. Some of these factors have no or little effect on GH concentration. This would explain the discordant combination of low IGF-I and IGFBP-3 concentrations and normal results of GH functional testing in many patients.
Thus, it cannot be expected that IGF-I or IGFBP-3 concentrations represent a direct mirror image of the GH secretory status. It has been hypothesized that in case analytical sources of IGF-I (IGFBP-3) variation could be minimized, the prediction of GH secretory status would improve (42). Our data do not support this hypothesis. When central laboratory assays were used instead of local assays, a numerical shift occurred between the patients allocated to discordant categories, but prediction of GH deficiency did not improve. This may be explained in part by the fact that intraindividual variation over time is greater for the results of GH functional testing than for IGF-I (IGFBP-3) concentrations (43). This would imply that overall agreement could rather be improved by standardization of the GH assays.
We conclude that the assay-related variability of results between local centers and a reference laboratory is lower for IGF-I than for IGFBP-3. In children with impaired growth, growth factor screening reflects aspects of GH insufficiency different from those reflected by functional testing. Agreement between these diagnostic approaches is poor and could not be improved by reduction of assay-related variability.
|
Acknowledgements
|
|---|
The participating investigators of the German KIGS Board/Medical Outcome Study Group are as follows: W Andler (Datteln), R Arndt (Neubrandenburg), M Bettendorf (Heidelberg), H J Böhles (Frankfurt/Main), H G Dörr (Erlangen), S Eisberg (Minden), B P Hauffa (Essen), O Hiort (Lübeck), E Korsch (Köln), M Mix (Rostock), K Mohnike (Magdeburg), M Morlot (Hannover), R Mühlenberg (Krefeld), C-J Partsch (Kiel), M B Ranke (Tuebingen), E Schönau (Köln), H-P Schwarz (München), G Simic-Schleicher (Bremen), K-P Ullrich (Gotha), T van Meegen-Freund (Aachen), and Ch Vogel/A Klinghammer (Chemnitz). This work was supported by a grant from Pharmacia Corporation, Erlangen, Germany to B P H and M B R. The authors thank Mrs K Weber for her expert technical assistance and R D Oades, Department of Child and Adolescent Psychiatry, University of Essen for helpful review of the manuscript.
|
References
|
|---|
1. Sizonenko PC, Clayton PE, Cohen P, Hintz RL, Tanaka T & Laron Z. 2001 11 137–165.2. Rosenfeld RG, Wilson DM, Lee PD & Hintz RL. Insulin-like growth factors I and II in evaluation of growth retardation. Journal of Pediatrics 1986 109 428–433.[CrossRef][Web of Science][Medline]3. Lee PD, Wilson DM, Rountree L, Hintz RL & Rosenfeld RG. Efficacy of insulin-like growth factor I levels in predicting the response to provocative growth hormone testing. Pediatric Research 1990 27 45–51.[Web of Science][Medline]4. Phillip M, Chalew SA, Kowarski AA & Stene MA. Plasma IGFBP-3 and its relationship with quantitative growth hormone secretion in short children. Clinical Endocrinology 1993 39 427–432.[Medline]5. Smith WJ, Nam TJ, Underwood LE, Busby WH, Celnicker A & Clemmons DR. Use of insulin-like growth factor-binding protein-2 (IGFBP- 2), IGFBP-3, and IGF-I for assessing growth hormone status in short children. Journal of Clinical Endocrinology and Metabolism 1993 77 1294–1299.[Abstract]6. Cianfarani S, Boemi S, Spagnoli A, Cappa M, Argirò G, Vaccaro F, Manca Bitti ML & Boscherini B. Is IGF-binding protein-3 assessment helpful for the diagnosis of GH deficiency. Clinical Endocrinology 1995 43 43–47.[Medline]7. Nunez SB, Municchi G, Barnes KM & Rose SR. Insulin-like growth factor I (IGF-I) and IGF-binding protein-3 concentrations compared to stimulated and night growth hormone in the evaluation of short children – a clinical research center study. Journal of Clinical Endocrinology and Metabolism 1996 81 1927–1932.[Abstract]8. Rasat R, Livesey JL, Espiner EA, Abbott GD & Donald RA. IGF-1 and IGFBP-3 screening for disorders of growth hormone secretion. New Zealand Medical Journal 1996 109 156–159.[Web of Science][Medline]9. Karachaliou F, Stamoyannou L, Maravelias K, Bartsocas CS & Koutselinis A. Serum levels of IGFBP-3: usefulness in diagnosis of GH deficiency and relationship to measurements of GH secretion in children. Journal of Pediatric Endocrinology and Metabolism 1996 9 169–174.10. Juul A, Kastrup KW, Pedersen SA & Skakkebaek NE. Growth hormone (GH) provocative retesting of 108 young adults with childhood-onset GH deficiency and the diagnostic value of insulin-like growth factor I (IGF-I) and IGF-binding protein-3. Journal of Clinical Endocrinology and Metabolism 1997 82 1195–1201.[Abstract/Free Full Text]11. Juul A & Skakkebæk NE. Prediction of the outcome of growth hormone provocative testing in short children by measurement of serum levels of insulin-like growth factor I and insulin-like growth factor binding protein 3. Journal of Pediatrics 1997 130 197–204.[CrossRef][Web of Science][Medline]12. Koch A & Dorr HG. Insulin-like growth factor-I and its binding protein-3 in serum: are they good screening properties for the diagnosis of growth hormone deficiency?. European Journal of Clinical Chemistry and Clinical Biochemistry 1997 35 379–385.[Web of Science][Medline]13. Tillmann V, Buckler JMH, Kibirige MS, Price DA, Shalet SM, Wales JKH, Addison MG, Gill MS, Whatmore AJ & Clayton PE. Biochemical tests in the diagnosis of childhood growth hormone deficiency. Journal of Clinical Endocrinology and Metabolism 1997 82 531–535.[Abstract/Free Full Text]14. Rikken B, Van Doorn J, Ringeling A, Van den Brande JL, Massa G & Wit JM. Plasma levels of insulin-like growth factor (IGF)-I, IGF-II and IGF-binding protein-3 in the evaluation of childhood growth hormone deficiency. Hormone Research 1998 50 166–176.[CrossRef][Web of Science][Medline]15. Mitchell H, Dattani MT, Nanduri V, Hindmarsh PC, Preece MA & Brook CGD. Failure of IGF-I and IGFBP-3 to diagnose growth hormone insufficiency. Archives of Disease in Childhood 1999 80 443–447.[Abstract/Free Full Text]16. Ranke MB, Schweizer R, Elmlinger MW, Weber K, Binder G, Schwarze CP & Wollmann HA. Significance of basal IGF-I, IGFBP-3 and IGFBP-2 measurements in the diagnostics of short stature in children. Hormone Research 2000 54 60–68.[CrossRef][Web of Science][Medline]17. Boquete HR, Sobrado PG, Fideleff HL, Sequera AM, Giaccio AV, Suarez MG, Ruibal GF & Miras M. Evaluation of diagnostic accuracy of insulin-like growth factor (IGF)-I and IGF-binding protein-3 in growth hormone-deficient children and adults using ROC plot analysis. Journal of Clinical Endocrinology and Metabolism 2003 88 4702–4708.[Abstract/Free Full Text]18. Badaru A & Wilson DM. Alternatives to growth hormone stimulation testing in children. Trends in Endocrinology and Metabolism 2004 15 252–258.[CrossRef][Web of Science][Medline]19. Darendeliler F, Spinu I, Bas F, Bundak R, Isguven P, Arslanoglu I, Saka N, Sukur M & Gunoz H. Reevaluation of growth hormone deficiency during and after growth hormone (GH) treatment: diagnostic value of GH tests and IGF-I and IGFBP-3 measurements. Journal of Pediatric Endocrinology and Metabolism 2004 17 1007–1012.20. Hauffa BP, Lehmann N, Bettendorf M, Mehls O, Dorr HG, Partsch CJ, Schwarz HP, Stahnke N, Steinkamp H, Said E, Sander S & Ranke MB. Central reassessment of GH concentrations measured at local treatment centers in children with impaired growth: consequences for patient management. European Journal of Endocrinology 2004 150 291–297.[Abstract]21. Bland JM & Altman DG. Comparing methods of measurement: why plotting difference against standard method is misleading. Lancet 1995 346 1085–1087.[CrossRef][Web of Science][Medline]22. Passing H & Bablok W. A new biometrical procedure for testing the equality of measurements from two different analytical methods. Journal of Clinical Chemistry and Clinical Biochemistry 1983 21 709–720.[Web of Science][Medline]23. Schoonjans F, Zalata A, Depuydt CE & Comhaire FH. MedCalc: a new computer program for medical statistics. Computer Methods and Programs in Biomedicine 1995 48 257–262.[CrossRef][Web of Science][Medline]24. Sachs L. Angewandte Statistik, Anwendung Statistischer Methoden 1992 Verlag Berlin, Heidelberg, New York25. SAS System Release 8.2 Edition 2002 SAS Institute Carey, NC26. Quarmby V, Quan C, Ling V, Compton P & Canova-Davis E. How much insulin-like growth factor I (IGF-I) circulates? Impact of standardization on IGF-I assay accuracy. Journal of Clinical Endocrinology and Metabolism 1998 83 1211–1216.[Abstract/Free Full Text]27. Ranke MB, Osterziel KJ, Schweizer R, Schuett B, Weber K, Robbel P, Vornwald A, Blumenstock G & Elmlinger MW. Reference levels of insulin-like growth factor I in the serum of healthy adults: comparison of four immunoassays. Clinical Chemistry and Laboratory Medicine 2003 41 1329–1334.[CrossRef][Web of Science][Medline]28. Wyatt DT, Mark D & Slyper A. Survey of growth hormone treatment practices by 251 pediatric endocrinologists. Journal of Clinical Endocrinology and Metabolism 1995 80 3292–3297.[Abstract]29. Reiter EO & Rosenfeld RG. Normal and aberrant growthEds.PR Larsen, HM Kronenberg, S Melmed & KS Polonsky. Williams Textbook of Endocrinology 2002 Saunders Philadelphia 1003–1114.30. Lifshitz F. Pediatric Endocrinology, Growth, Adrenal, Sexual, Thyroid, Calcium, and Fluid Balance Disorders 2006 CRC Press Inc. Boca Raton, USA31. Blum WF, Ranke MB, Kietzmann K, Gauggel E, Zeisel HJ & Bierich JR. A specific radioimmunoassay for the growth hormone (GH)-dependent somatomedin-binding protein: its use for diagnosis of GH deficiency. Journal of Clinical Endocrinology and Metabolism 1990 70 1292–1298.[Abstract/Free Full Text]32. Hasegawa Y, Hasegawa T, Takada M & Tsuchiya Y. Plasma free insulin-like growth factor I concentrations in growth hormone deficiency in children and adolescents. European Journal of Endocrinology 1996 134 184–189.[Abstract/Free Full Text]33. Juul A. Determination of insulin-like growth factor I in children: normal values and clinical use. Hormone Research 2001 55 94–99.[Web of Science][Medline]34. Blum WF & Schweizer R. Insulin-like growth factors and their binding proteinsEds.MB Ranke. Diagnostics of Endocrine Function in Children and Adolescents 2003 Karger Basel 166–199.35. Kao PC, Matheny AP Jr & Lang CA. Insulin-like growth factor I comparisons in healthy twin children. Journal of Clinical Endocrinology and Metabolism 1994 78 310–312.[Abstract]36. Harrela M, Koistinen H, Kaprio J, Lehtovirta M, Tuomilehto J, Eriksson J, Tolvanen L, Koskenvuo M, Leinonen P, Koistinen R & Seppälä M. Genetic and environmental components of interindividual variation in circulating levels of IGF-I, IGF-II, IGFBP-1, and IGFBP-3. Journal of Clinical Investigation 1996 98 2612–2615.[Web of Science][Medline]37. Clemmons DR & Van Wyk JJ. Factors controlling blood concentration of somatomedin C. Clinics in Endocrinology and Metabolism 1984 13 113–143.[CrossRef][Web of Science][Medline]38. Ho KY, Veldhuis JD, Johnson ML, Furlanetto R, Evans WS, Alberti KG & Thorner MO. Fasting enhances growth hormone secretion and amplifies the complex rhythms of growth hormone secretion in man. Journal of Clinical Investigation 1988 81 968–975.[Web of Science][Medline]39. Kratzsch J, Blum WF, Schenker E & Keller E. Regulation of growth hormone (GH), insulin-like growth factor (IGF)I, IGF-binding proteins -1, -2, -3 and GH binding protein during progression of liver cirrhosis. Experimental and Clinical Endocrinology and Diabetes 1995 103 285–291.40. Holt RI, Jones JS, Stone NM, Baker AJ & Miell JP. Sequential changes in insulin-like growth factor I (IGF-I) and IGF-binding proteins in children with end-stage liver disease before and after successful orthotopic liver transplantation. Journal of Clinical Endocrinology and Metabolism 1996 81 160–168.[Abstract]41. Blum WF. Insulin-like growth factors (IGFs) and IGF-binding proteins in chronic renal failure: evidence for reduced secretion of IGFs. Acta Paediatrica Scandinavica 1991 379 24–31.42. Van Wyk JJ & Bruton ET. Molecular basis for species and ligand specificity of a monoclonal antibody raised against human IGF-I. Endocrinology 1997 138 4521–4523.[Abstract/Free Full Text]43. Hilczer M, Smyczynska J, Stawerska R & Lewinski A. Stability of IGF-I concentration despite divergent results of repeated GH stimulating tests indicates poor reproducibility of test results. Endocrine Regulations 2006 40 37–45.[Medline]
Received 21 May 2007
Accepted 8 August 2007
Top
Abstract
Introduction
