Reference intervals of specific biomarkers are used to distinguish between healthy and diseased people. This enables an accurate interpretation of biomedical laboratory tests, which is a prerequisite for the proper treatment of patients. Reference intervals and target values are not only used for diagnosis, but also to monitor therapeutic treatments and to screen for risks of disease. Therefore, the continuous evaluation and fine-tuning of reference intervals, based on age-, gender- and ethnic-specific variations as well as on different methods, are essential. Physiological changes in biomedical parameters occur with age and are especially pronounced during childhood and adolescence due to constant growth, development as well as maturation, thereby causing alterations in metabolic processes , , . Thus, the establishment and validation of paediatric reference values is of high relevance. Studies like the Children’s Health Improvement through Laboratory Diagnostics program, the Lifestyle of Our Kids (LOOK) program, the German Health Interview and Examination Survey for Children and Adolescents (KiGGS), the Canadian Laboratory Initiative on Paediatric Reference Intervals (CALIPER) project (reviewed in ) as well as the Identification and prevention of Dietary- and lifestyle-induced health Effects in Children and infants (IDEFICs) project  contribute to meet this demand and close gaps in paediatric laboratory medicine. Although these studies all have a similar aim, the study protocols differ from each other. Nevertheless, the common denominator is the evaluation of reference intervals of different biomedical markers for age groups between 0 and 18 years. Recently, the MoYo (Motivating Young people to maintain a healthy life-style) project was performed. This was a school-based pilot study with an interventional design performed in Salzburg (Austria) from 2014 to 2015 . It has provided baseline data from healthy adolescents at the age of 14–17 years and facilitated an “a priori” approach for the evaluation of reference values. The purpose of this paper is, to evaluate the reference intervals of haematological and biochemical laboratory parameters for local, Caucasian adolescents in Salzburg, and to compare these with available paediatric reference data.
Materials and methods
The anthropometric and biomedical data were obtained from the baseline data of the MoYo-pilot project, which was performed in schools and therefore in a non-clinical setting. In total, 110 per se healthy adolescents of Caucasian origin in the age group of 14–17 years participated. All but one indicated at the beginning of the study that their health status was good and that they did not have any known diseases. For 102 adolescents (63 males, 39 females) haematological parameters as well as carbohydrate and lipid metabolism markers, hormones, vitamins and micronutrients were determined. A full list with all the analysed parameters can be found in Table 1. All participants and their legal guardians gave informed consent to participate in this study. This informed consent could be withdrawn at any point of the study and beyond and included the deletion of all available data when required. The study design was approved by the local Ethics Committee (county of Salzburg, Austria) and meets the Declaration of Helsinki.
Venipuncture and anthropometric measurements were carried out between 07:30 and 08:30 AM. The participants were instructed not to eat or drink anything except for water at least 10 h before the day of the blood sample collection. Blood samples were drawn using a 21G Safety blood collection set (Ref. 450081, Greiner bio-one GmbH, Kremsmünster, Austria) in combination with Vacuette tubes for serum (Ref. 455092, Greiner bio-one GmbH, Kremsmünster, Austria), EDTA- (Ref. 454086, Greiner bio-one GmbH, Kremsmünster, Austria) and Li-Heparin-plasma (Ref. 454083, Greiner bio-one GmbH, Kremsmünster, Austria) and handled according to pre-analytical guidelines . Serum samples were stored at room temperature in an upright position for 30 min to allow the sample to fully clot before further processing. Serum and plasma samples were centrifuged at 1500×g for 10 min at 22°C.
Haematological parameters were measured using EDTA-samples on the XS-1000i instrument (Sysmex GmbH, Vienna, Austria). Clinical chemistry parameters, such as carbohydrate and lipid metabolism markers, hormones, vitamins and micronutrients were determined from serum using either the Cobas c311 (Roche Diagnostics, Switzerland), the Vitros ECiQ® (Assista, Vienna, Austria) or diagnostically approved ELISAs (IBL International GmbH [Hamburg, Germany] and Biovendor GesmbH [Vienna, Austria]). Assessment of the optical density of these assays was performed on the Infinite 200 pro plate reader (Tecan, Grödig, Austria) (Table 1).
Anthropometric measurements for evaluating body mass index (BMI), waist-to-hip ratio (WHR) and waist-to-height ratio (WHtR) were performed with a portable calibrated stadiometer, a measuring tape and a segmental body composition analyser (Tanita BC-418, Tanita Corporation of America, Inc., IL, USA) commonly used in clinical research.
For biometric data acquisition, Excel 2010 (Microsoft Corporation, WA, USA) was used. The BMI was calculated as the ratio of the body weight to the square of body height (kg/m2). The WHR was calculated by dividing waist circumference by hip circumference and WHtR by dividing waist circumference by body height. Laboratory data were managed and validated by the laboratory software GLIMS (MIPS Diagnostics Intelligence Europe, Gent, Belgium).
For statistical analyses, the data were imported to IBM SPSS Statistics version 25 (IBM Corporation, NY, USA) and descriptive und inference statistical analyses were performed. The reference intervals were calculated according to the CLSI guidelines C28-A3c  after the data were normalized and outliers were identified and excluded using the Grubbs test. Before the 2.5 and the 97.5 percentiles were calculated, the data were weighted with respect to gender proportion (number of cases 120). In addition, the 90% confidence intervals (CIs) were computed for the upper and lower limit of the reference interval ±1.64×SD/√n. Differences between males and females were analysed conducting a multivariate analysis of variance.
The analysis of the biometrical data showed that the BMI is distributed according to the age- and gender-matched reference range. In males, the median BMI was 20.37 (25 percentile 18.73 and 75 percentile 22.67), in females the median BMI was 20.52 (25 percentile 18.14 and 75 percentile 22.74). The WHR was normally distributed in males (mean WHR±SD=0.85±0.05), but not in females (median WHR=0.77 (25 percentile 0.74 and 75 percentile 0.80). The evaluated median WHtR for male was 0.42 (25 percentile 0.40 and 75 percentile 0.44) and for female 0.41 (25 percentile 0.39 and 75 percentile 0.44).
The calculated reference intervals, including the 90% CI of the lower and upper limit from investigated laboratory parameters, are listed in Table 2. According to the Grubbs outlier test, 17 individuals (11 boys and 6 girls) with a higher score than the critical Grubbs cut-off in at least one parameter were excluded. In addition, it should be mentioned that one male diabetes patient in the age of 15 years participated in the study, who was also excluded from the calculation of reference ranges. To compare these with reference intervals of similar study populations (age, gender and ethnicity), a survey for published reference intervals was performed. The results for the reference intervals for the Austrian adolescents as determined in the presented study and those published of matched age ranges , , , , ,  are shown in Table 3.
The MoYo-project was performed as a feasibility study for health screening in a non-clinical setting and health promotion in schools, including an interventional study design. The implementation of sampling for anthropometrical and biomedical measurements in the school setting was successfully established and samples of healthy adolescents were analysed for haematological and biochemical parameters (Table 1). Even though the number of cases was comparatively low for the analysis in terms of gender-matched groups, the data set provides a good basis for the calculation of paediatric reference intervals according to the CLSI guidelines .
Over the last 15 years, different studies assessing paediatric reference intervals have been published. These evolved from projects like KiGGS (Germany; 1–18 years), CALIPER (Canada; 0–18 years), NHANES (US; all age groups), LOOK (Australia; 8, 10 and 12 years), AACB (Australia and New Zealand; all age groups), COPENHAGEN (Denmark; 5–20 years) (reviewed in ) and IDEFICS (Italy, Spain, Germany, Belgium, Sweden, Hungary and Estonia; 2–11 years) .
The age- and gender-based evaluation of the BMI of the Austrian adolescents investigated shows that in comparison to German  and Greek  adolescents the Austrian cohort had a slightly lower BMI. This highlights that the study cohort is within the healthy range of BMI. The current WHO BMI tables are based on age- and gender-specific percentiles and z-scores (http://www.who.int/growthref/who2007_bmi_for_age/en/). These were derived by pooling the data sets of the Health Examination Survey, and the Health and Nutrition Examination Survey . However, an age-, gender- and percentile-based BMI value calculation requires a different study set-up and was beyond the scope of the present study. The values for WHR and WHtR are provided as well, as these are described as discriminators for cardiovascular risk factors and metabolic syndrome diagnosis .
The present biomedical data analysis represents a synopsis of reference values for haematological and clinical biochemical parameters of adolescents in Europe and the Anglo-American region. Both the MoYo data set as well as the CALIPER data set , ,  were derived from healthy and normal-weight adolescents and were analysed “a priori”. As ethnical, environmental and lifestyle circumstances can influence the evaluation and establishment of reference intervals and the data of the CALIPER study are based on different ethnic groups , the MoYo data add value in regard of validation of reference intervals for Caucasian adolescents in the age of 14–17 years.
In terms of haematological data, Zierk et al. ,  used a two-fold different approach: first, Zierk et al. used an indirect approach by retrospectively evaluating haematological parameters from the laboratory database of the University Children’s Hospital Erlangen containing healthy and pathologic samples for data analysis; and secondly, they calculated continuous reference intervals to meet the demands of variations in age and sex. In the first study , they compared the reference intervals for erythrocytes, haemoglobin, haematocrit, mean corpuscular volume, mean corpuscular haemoglobin, mean corpuscular haemoglobin concentration and platelets to the 2.5th and 97.5th percentile (95% CI) of the data retrieved from the KiGGS study, which investigated, among others, haematological parameters in a cohort representative for healthy 1–18 year old German children “a priori”  (Table 3). The analysis of the haematological parameters in the different studies was performed either using a Sysmex haematological analyser ,  or a Coulter STKS haematological flow cytometer , and the results are in the same range. However, although the continuous reference interval approach is appropriate for specific parameters that are changing continuously with age (e.g. creatinine or uric acid , ), currently, the prevalent application for reference intervals is still the age-grouped reference range. One reason for that is that currently most laboratory IT systems are not designated for data interpretation using continuous reference intervals.
The CALIPER database (https://app3.ccb.sickkids.ca/caliper/caliperlogin) comprises data of healthy children between 0 and 18 years derived from a nation-wide, multi-centre study performed in hospitals across Canada. It contains validated reference intervals for over 100 biomedical blood tests for common biochemical, endocrine, metabolic disease and tumour biomarkers as well as vitamins , , , , , , . The correlation with selected biomedical parameters also determined in the MoYo-study reveals a high correlation of the values for both genders, even though the laboratory analysis was performed with different devices (Cobas c311, Vitros ECiQ® and diagnostically approved ELISAs in the MoYo-study vs. Abbott Architect ci4100, c8000 or i2000 in the CALIPER project) and the study population was different in terms of regional aspects. Ideally, measurement values and therefore reference ranges should not differ depending on the technique or equipment used. However, it is known from the literature as well as from reference guides that there are significant alterations of the reference values depending on the used methodology , , . Therefore, every clinical laboratory needs to validate the reference intervals for the parameters that are measured and released for diagnostic and therapeutic use , and it is of main importance to clearly indicate the methods and equipment used for clinical parameters tested. The availability of carefully established and validated databases for laboratory reference intervals and medical decision levels can serve not only as general reference work, but also as a standard for other laboratories.
The parameters proinsulin, lipoprotein (a) and adiponectin were neither analysed by Soldin et al.  nor by the CALIPER study. The cut-off for lipoprotein (a) is reported to be 30 mg/dL for 9.1–18-year-old persons . The upper limit of the MoYo-study population was much higher (171.39 mg/dL). The adiponectin concentration in the serum varies significantly with age, but generally the range of the adiponectin values of the Austrian cohort match with the year-of-age specified reference intervals of Lausten-Thomsen et al. . To the best of our knowledge, currently no reference intervals for proinsulin at fasting for 14–17-year-old patients exist. The analysis of the MoYo cohort revealed no significant gender differences and indicates a range of 0.62–8.34 pmol/L.
As already described, the MoYo study was performed as a pilot study and it should be mentioned that the small sample size is a limitation of this work. However, it is challenging to gain data of a per se healthy study cohort of people within this age range and the analysis of the data set in terms of reference limits gives valuable insights in both non-gender-matched and gender-matched reference intervals. Therefore, the data set provides a good basis for the calculation of paediatric reference intervals according to the CLSI recommendations .
To conclude, the results of this “a priori” study add value to the evaluation of paediatric reference intervals of Caucasian adolescents in terms of an increase of the spectrum and a broader validation of the available biomarkers.
We greatly acknowledge the coordination of the MoYo-project at the partner school by Edith Oberkofler and the competent assistance during sampling by Justine Gmachl-Baugartner, Renate Wiltsche and Liselotte Helminger.
Tahmasebi H, Higgins V, Fung AW, Truong D, White-Al Habeeb NM, Adeli K. Pediatric reference intervals for biochemical markers: gaps and challenges, recent national initiatives and future perspectives. EJIFCC 2017;28:43–63. PubMedGoogle Scholar
Soldin SJ, Wong EC, Brugnara C, Soldin OP. Pediatric reference intervals, 7th ed. Washington, DC: AACC Press, 2011:304. Google Scholar
Ahrens W, Bammann K, Siani A, Buchecker K, De Henauw S, Iacoviello L, et al. The IDEFICS cohort: design, characteristics and participation in the baseline survey. Int J Obes (Lond) 2011;35(Suppl. 1):S3–15. PubMedCrossrefGoogle Scholar
World Health Organization. WHO guidelines on drawing blood: best practices in phlebotomy. Geneva, Switzerland: WHO Press, 2010:125. Google Scholar
CLSI. Defining, establishing, and verifying reference intervals in the clinical laboratory; Approved guideline – third edition, in CLSI document C28-A3c. Wayne, PA: Institute CaLS 2008. Google Scholar
Colantonio DA, Kyriakopoulou L, Chan MK, Daly CH, Brinc D, Venner AA, et al. Closing the gaps in pediatric laboratory reference intervals: a CALIPER database of 40 biochemical markers in a healthy and multiethnic population of children. Clin Chem 2012;58:854–68. CrossrefWeb of ScienceGoogle Scholar
Bailey D, Colantonio D, Kyriakopoulou L, Cohen AH, Chan MK, Armbruster D, et al. Marked biological variance in endocrine and biochemical markers in childhood: establishment of pediatric reference intervals using healthy community children from the CALIPER cohort. Clin Chem 2013;59:1393–405. Web of ScienceCrossrefGoogle Scholar
Konforte D, Shea JL, Kyriakopoulou L, Colantonio D, Cohen AH, Shaw J, et al. Complex biological pattern of fertility hormones in children and adolescents: a study of healthy children from the CALIPER cohort and establishment of pediatric reference intervals. Clin Chem 2013;59:1215–27. CrossrefWeb of ScienceGoogle Scholar
Rosario AS, Kurth BM, Stolzenberg H, Ellert U, Neuhauser H. Body mass index percentiles for children and adolescents in Germany based on a nationally representative sample (KiGGS 2003–2006). Eur J Clin Nutr 2010;64:341–9. PubMedWeb of ScienceCrossrefGoogle Scholar
Bacopoulou F, Efthymiou V, Landis G, Rentoumis A, Chrousos GP. Waist circumference, waist-to-hip ratio and waist-to-height ratio reference percentiles for abdominal obesity among Greek adolescents. BMC Pediatr 2015;15:50. Web of SciencePubMedCrossrefGoogle Scholar
de Onis M, Onyango AW, Borghi E, Siyam A, Nishida C, Siekmann J. Development of a WHO growth reference for school-aged children and adolescents. Bull World Health Organ 2007;85:660–7. Web of ScienceCrossrefPubMedGoogle Scholar
Lee CM, Huxley RR, Wildman RP, Woodward M. Indices of abdominal obesity are better discriminators of cardiovascular risk factors than BMI: a meta-analysis. J Clin Epidemiol 2008;61:646–53. PubMedCrossrefWeb of ScienceGoogle Scholar
Zierk J, Arzideh F, Rechenauer T, Haeckel R, Rascher W, Metzler M, et al. Age- and sex-specific dynamics in 22 hematologic and biochemical analytes from birth to adolescence. Clin Chem 2015;61:964–73. CrossrefWeb of SciencePubMedGoogle Scholar
Dortschy R, Schaffrath-Rosario A, Scheidt-Nave C, Thierfelder W, Thamm M, Gutsche J, et al. Bevölkerungsbezogene Verteilungswerte ausgewählter Laborparameter aus der Studie zur Gesundheit von Kindern und Jugendlichen in Deutschland (KiGGS), in Beiträge zur Gesundheitsberichterstattung des Bundes, Koch-Institut R, Editor 2009, Robert Koch-Institut: Germany. 132. Google Scholar
Ceriotti F, Boyd JC, Klein G, Henny J, Queralto J, Kairisto V, et al. Reference intervals for serum creatinine concentrations: assessment of available data for global application. Clin Chem 2008;54:559–66. CrossrefWeb of ScienceGoogle Scholar
Bailey D, Bevilacqua V, Colantonio DA, Pasic MD, Perumal N, Chan MK, et al. Pediatric within-day biological variation and quality specifications for 38 biochemical markers in the CALIPER cohort. Clin Chem 2014;60:518–29. CrossrefWeb of ScienceGoogle Scholar
Bevilacqua V, Chan MK, Chen Y, Armbruster D, Schodin B, Adeli K. Pediatric population reference value distributions for cancer biomarkers and covariate-stratified reference intervals in the CALIPER cohort. Clin Chem 2014;60:1532–42. Web of ScienceCrossrefGoogle Scholar
Teodoro-Morrison T, Kyriakopoulou L, Chen YK, Raizman JE, Bevilacqua V, Chan MK, et al. Dynamic biological changes in metabolic disease biomarkers in childhood and adolescence: a CALIPER study of healthy community children. Clin Biochem 2015;48:828–36. Web of ScienceCrossrefPubMedGoogle Scholar
CLSI. Method procedure comparison and bias estimation using patient samples; approved guideline – Third edition, in CLSI document EP09-A3. Wayne, PA: Institute CaLS. 2013. Google Scholar
Langer C, Tambyrayah B, Thedieck S, Nowak-Gottl U. Testing for lipoprotein(a) concentration and apolipoprotein(a) phenotypes: method standardization and pediatric reference values. Semin Thromb Hemost 2011;37:810–3. Web of ScienceCrossrefPubMedGoogle Scholar
Lausten-Thomsen U, Christiansen M, Fonvig CE, Trier C, Pedersen O, Hansen T, et al. Reference values for serum total adiponectin in healthy non-obese children and adolescents. Clin Chim Acta 2015;450:11–4. PubMedCrossrefWeb of ScienceGoogle Scholar
About the article
Published Online: 2018-10-30
Published in Print: 2019-05-27
Author contributions: All the authors have accepted responsibility for the entire content of this submitted manuscript and approved submission.
Research funding: This work was supported by the County of Salzburg (grant number 2014-7081-004) and an Institutional Grant of the Salzburg University of Applied Sciences.
Employment or leadership: None declared.
Honorarium: None declared.
Competing interests: The funding organization(s) played no role in the study design; in the collection, analysis, and interpretation of data; in the writing of the report; or in the decision to submit the report for publication.
Citation Information: Clinical Chemistry and Laboratory Medicine (CCLM), Volume 57, Issue 6, Pages 891–900, ISSN (Online) 1437-4331, ISSN (Print) 1434-6621, DOI: https://doi.org/10.1515/cclm-2018-0715.
©2019 Barbara Bogner et al., published by De Gruyter, Berlin/Boston. This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License. BY-NC-ND 4.0