Printer Friendly

Age- and sex-specific dynamics in 22 hematologic and biochemical analytes from birth to adolescence.

The interpretation of laboratory test results in pediatrics is performed in the context of age- and sex-dependent dynamics, as physiological development leads to changes in many of the analytes measured, particularly in the first years of life and during puberty. To reflect inter- and intraindividual variation in laboratory tests, clinical decision making is generally guided by reference intervals, which are defined as the 2.5th and 97.5th percentiles of a healthy population's distribution (1-4).

Partition into discrete age groups for both males and females is commonly performed to represent the age and sex dependence of laboratory analytes when reference intervals are used. Age groups are selected using visual inspection and statistical tests to approximate change in analyte concentration with age (5, 6). However, discrete age groups do not adequately reflect the continuous changes of biological development and thus cannot always represent the exact extent and onset of age-dependent dynamics. In analogy with other developmental quantities routinely specified in relation to age (e.g., weight- and height-for-age charts), a continuous description would seem to be a more appropriate approach for laboratory analytes (7, 8).

Such an approach is restricted by the requirement of a large number of samples from healthy children (9). According to the generally accepted recommendation of an IFCC expert group, approximately 120 samples are needed to establish reliable reference intervals for a homogenous population (1,2); creation of continuous age-dependent reference intervals with these procedures would require many more samples to account for variation in analyte concentration with age. Because access to blood samples from healthy children is limited by ethical and practical constraints, this procedure is infeasible in many settings (3, 9). Newborn and infant children are most affected by these restrictions, although precise age-adjusted reference intervals are especially important in these age groups because of the significant age-related contribution to pediatric morbidity and pronounced physiological development with consecutive changes in laboratory analytes. Great efforts have been undertaken to address these generally recognized issues (10, 11). The Canadian Laboratory Initiative on Pediatric Reference Intervals (CALIPER) [4] provides sex-specific reference intervals from birth to adulthood for > 80 biochemical analytes derived from a population of > 8600 healthy children (6, 12-15). The continuous dynamics for many of the examined analytes are acknowledged and carefully considered in the selection of age partitions. Separation into age groups with significant differences, however, invariably leads to a discontinuous representation of change with age. In the German Health Interview and Examination Survey for Children and Adolescents (KiGGS), continuous reference intervals are reported for 28 analytes on the basis of a representative German cohort of 14255 children (16, 17). However, the critical age group of newborns and infants was excluded from blood sampling in the KiGGS survey for the ethical and practical reasons mentioned above.

In a recent pilot study, we have shown an alternative method for the determination of continuous reference intervals that avoids these obstacles (18). Reference intervals are calculated from a laboratory database containing a mixture of healthy and pathologic samples; the distribution of supposedly healthy samples is estimated from the whole data set with a statistical approach and used to calculate continuous reference intervals. Comparison with reference intervals determined with conventional methods from the KiGGS survey showed a high concordance of reference limits and their age-dependent dynamics, thereby proving the feasibility of this indirect method and making it a viable alternative to conventional approaches when these are limited.

The purpose of the present report is to apply the indirect approach to a broader panel of hematologic and biochemical analytes (Table 1). The resulting continuous reference intervals should demonstrate the gradual influences of age and sex and therefore allow more precise clinical decision making on the basis of laboratory results. The effect on the classification of laboratory samples as healthy or pathologic is analyzed by comparing the original categorization (performed with conventional reference intervals) to categorization performed with the newly established continuous reference intervals.


We analyzed measurements of 22 analytes performed during clinical care of patients in the Department of Pediatrics and Adolescent Medicine, University Hospital Erlangen, a tertiary care center covering the entire spectrum of pediatrics.


The population in Germany--and our hospital's patient population--is composed predominantly of white individuals; stratification according to ethnicity was not performed.


Analytes were selected according to clinical relevance and methodologic requirements. We considered analytes for which a large number of measurements would allow application of the indirect approach used and where the availability of sex-stratified continuous reference intervals from birth to adulthood might benefit clinical decision making. This resulted in the identification of 13 biochemical analytes [sodium, chloride, potassium, calcium, magnesium, phosphate, creatinine (enzymatic), aspartate transaminase (AST), alanine transaminase (ALT), [gamma]-glutamyl transferase ([gamma]-GT), alkaline phosphatase, lactate dehydrogenase (LDH), and total protein] and 9 hematology analytes [hemoglobin, hematocrit, mean red cell hemoglobin (MCH), mean red cell hemoglobin concentration (MCHC), mean red cell volume (MCV), red cell count, red cell distribution width, white cell count, and platelet count] (see Supplemental Table 1, which accompanies the online version of this article at http:// All the measurements of these analytes performed for inpatients and outpatients aged [less than or equal to] 18 years, including patients from intensive care units and specialty units, were retrieved from the laboratory's database. The time period examined spanned January 2004 to April 2013 (biochemical analytes except for AST, ALT, [gamma]-GT) and April 2008 to April 2013 (hematologic analytes, AST, ALT, and [gamma]-GT) to provide a maximum number of samples measured with the same methods. Analysis of [gamma]-GT activity was performed for children aged 6 months to 18 years, as the number of available measurements allowed valid application of the algorithm in this interval only. For each analyte, 45978 to 210239 measurements (from 11 162 to 34807 distinct individuals) were available for analysis (Table 1).


Measurements of biochemical analytes were performed on a Cobas Integra 800 (Roche Diagnostics), and blood counts were measured on a Sysmex XE-2100 (Sysmex Europe). Method details are available in online Supplemental Table 1, and precision data are presented in online Supplemental Table 2. The stability of each analyte over time during the study period is demonstrated by stable monthly median values (see online Supplemental Tables 3 and 4).


We calculated continuous reference intervals with an indirect method described and validated previously (18); method details are available in the online Supplement. The approach is based on the assumption that the input data set consists of a mixture of parametrically distributed samples from healthy individuals and random pathologic samples (i.e., the proportion of nonpathologic samples is modeled with a statistical distribution, whereas the pathologic samples are assumed to be scattered randomly). When applied to a sufficiently large data set, these 2 sample sets can be reliably distinguished by a statistical algorithm. We estimate the underlying parametric distribution of healthy samples, and its 2.5th and 97.5th percentiles define the reference interval. The algorithm comprises 3 basic steps for each analyte: (a) partition of input data into age groups; (b) generation of discrete reference intervals for these groups; and (c) conversion of these discrete reference intervals into continuous reference intervals.

Partition into age groups. We split the samples in overlapping age groups to achieve a sample count of 500-2500 per group, depending on the analyte (Table 1) (18). Samples in each age group were selected from different patients. Because the number of available samples is variable and depends on the age period examined, each age group encompasses a time interval of a different length. Separation into overlapping age groups resulted in 38-197 groups (Table 1).

Generation of discrete reference intervals. We applied an indirect method for reference interval determination--an expanded version of an algorithm developed by Arzideh and colleagues (19-21)--to each age group. Briefly, a smoothed kernel density function is estimated for the distribution of the data. The "central" part of this distribution is assumed to represent the main part of the healthy population and is defined by truncation points in the Box--Cox transformed data set with an optimization method. A gaussian distribution of the central part is estimated, and its 2.5th and 97.5th percentiles are calculated to obtain the reference intervals.

Conversion into continuous reference intervals. For each analyte, we generated parametric curves for the 2.5th, 50th, and 97.5th percentiles. We used the R software's smooth.spline function to convert the reference limits into a continuous interval (22) and chose the number of degrees of freedom by visual inspection of the curves generated. Two different curves were created for most analytes to adequately represent the different dynamics of the neonatal period and infancy on the one hand and toddlerhood to adolescence on the other hand.


A comparison of the generated reference intervals to results from KiGGS (17) and CALIPER (6) is shown in online Supplemental Figs. 1 and 2. In KiGGS, hematology analyses were performed on a Cell-Dyn 3500 (Abbott) and biochemical analyses were performed on a Hitachi 917 (Roche Diagnostics); in CALIPER, analyses were performed on an Architect c8000 (Abbott). The same methods were used in general, except for phosphate, magnesium, and calcium (phosphomolybdate, Arsenazo, and Arsenazo III, respectively, in CALIPER).

The hypothetical effect of the newly established continuous reference intervals on classification of samples in the examined data set as healthy or pathologic is quantified in Table 2. We compared classification with the current categorization method (using conventional reference intervals from published literature or the manufacturer) to categorization performed with the newly established continuous reference intervals. Patients with multiple measurements of a certain analyte were classified as healthy if all their measurements for that analyte were also classified as healthy.


We calculated continuous age- and sex-specific reference intervals for 13 biochemical analytes and 9 hematology analytes. Graphical representations of the reference intervals are provided in Fig. 1A (creatinine), Fig. 2 (biochemical analytes without creatinine), and Fig. 3 (hematologic analytes). Data tables containing age-specific upper and lower reference limits to enable integration of the calculated reference intervals into laboratory information systems are available in the online Supplemental Database.

The majority of the analytes showed substantial age-specific dynamics, especially in the first months and years of life and after the onset of puberty. These age-dependent changes could only be approximated by reference intervals for distinct age groups, as cutoff points that would allow partition into separate age groups are nonexistent. The continuous change in creatinine reference intervals with age and the progressive divergence of male and female reference intervals are highlighted in Fig. 1A. The upper and lower reference limits underlying the reported continuous reference intervals for creatinine concentration are presented in Fig. 1B to demonstrate that a separation into age groups would be arbitrary and not due to biological changes in analyte concentration. Fig. 1C shows a comparison of continuous reference intervals for creatinine concentration to age-grouped reference intervals from CALIPER and highlights the problem of representing the age dependence of creatinine concentration with distinct age groups. Deficiencies become especially apparent at age group margins in infancy, e.g., when creatinine reference limits change by a factor of 3 [from 0.32-0.92 mg/dL (0-14 days) to 0.10-0.36 mg/dL (15 days to <2 years)]. Detailed consideration of sex-specific differences shows that these can be observed after the onset of puberty, but not before, in many analytes (alkaline phosphatase, ALT, AST, [gamma]-GT, LDH, hematocrit, hemoglobin, red cell count). Examples of this correlation are highlighted in Fig. 4, which shows sex-specific differences in alkaline phosphatase activity and hemoglobin and creatinine concentration in relation to Tanner stage PH2 (i.e., the appearance of pubic hair as a physical indicator of puberty onset).


Some analytes showed similar patterns of age-dependent change. A decline in concentration during the first months and years of life and a subsequent rise, with higher concentrations in males than in females after puberty, was observed in hemoglobin concentration, hematocrit, red cell count,

and creatinine concentration (Figs. 3 and 1A). A fast decline in activity of AST, ALT, and LDH during infancy was followed by a slower decline until 18 years of age, with a lower activity in females after puberty (Fig. 2). Red cell indices (MCH, MCHC, and MCV), red cell distribution width, and white cell count decreased continuously in infancy and stabilized afterward without substantial sex-specific differences. Plasma electrolytes and total protein concentration likewise exhibited age-specific changes in concentration but no substantial sex-specific differences. Platelet count rose during the first months of life and decreased afterward, and a higher platelet count was observed in females than in males in infancy and adolescence. Alkaline phosphatase showed the most complex pattern of change in analyte concentration over time: a decline in the first 4 years of life was followed by a rise with sex-specific onset, extent, peak, and subsequent decline. The increase in activity observed in females had an earlier onset, peak, and decline but was less pronounced than the rise in activity in males.


Comparison of the reference intervals provided with results from CALIPER and KiGGS showed consistent upper and lower reference limits and onset of age-dependent changes (see online Supplemental Figs. 1 and 2). Major differences, however, were observed between magnesium reference intervals in the CALIPER trial and those reported herein; comparison of our results to findings from the KiGGS survey and other sources (23, 24), however, showed no such differences and confirmed the reported reference intervals.

Use of the continuous reference intervals in comparison to the currently used reference intervals (mainly supplied by manufacturers or from published literature) resulted in a substantial reduction in the number of samples considered pathologic. The decrease in samples classified as pathologic was most pronounced for blood count analytes (except red cell distribution width) and less pronounced for most biochemical analytes. The proportion of samples classified as pathologic increased for AST, ALT, potassium, and LDH. Differences in classification of test results as healthy or pathologic with either the continuous reference intervals or the currently used reference intervals are summarized in Table 2.




We report continuous reference intervals for 22 important hematologic and biochemical analytes from birth to adulthood. Unlike most reference intervals currently used, which are valid for discrete age intervals stratified according to sex, we present a continuous classification strategy, allowing a precise representation of age- and sex-dependent change during development. This is possible because the analysis of a comprehensive clinical laboratory database with an indirect approach has made a large number of measurements available for evaluation. The number of samples analyzed (45978 to 210239, depending on the analyte) eliminates the need to approximate age-dependent change in analyte concentration with separate age groups. Accordingly, our data show gradual changes in analyte concentration over time rather than abrupt changes at well-defined time points (Figs. 1A, 2, and 3). Likewise, analysis of the discrete reference limits underlying the reported continuous reference intervals shows the absence of physiologically sensible cutoff points (Fig. 1B). This reflects the notion that separation into age groups is a technical limitation that can be overcome when a sufficient number of samples are available (7, 8).

The physiological developments leading to variation with age include the effects of increased creatinine production with muscle growth and maturation of renal function on creatinine concentrations (Fig. 1A). Transition from fetal to adult erythropoiesis and the physiologic anemia of infancy are accompanied by gradual changes in hemoglobin concentration, red cell count, red cell indices (MCV, MCH), and red cell distribution width (Fig. 3). Additionally, sex-dependent changes during and after puberty can be observed in many analytes (Figs. 2 and 4). The continuous nature of these changes is well established and is generally incorporated into clinical decision making on the basis of laboratory test results (3, 10, 25, 26). However, the availability of continuous reference intervals allows more precise consideration of these dynamics and better differentiation of change due to physiological development and change due to disease. Direct comparison of continuous reference intervals to age-grouped reference intervals highlights these advantages (Fig. 1C; online Supplemental Fig. 2). The potential benefits of continuous reference intervals could be limited in practice by current approaches to laboratory test result reporting that might in fact complicate interpretation. Therefore, improved forms of result reporting are necessary. Suitable representations include graphical result display (e.g., as in growth charts) or reporting of z-scores/percentiles instead of absolute values. Although the technical basis exists for such representations, they are rarely incorporated into current clinical practice. The continuous reference intervals provided can therefore serve as an incentive for laboratory software manufacturers to implement new strategies for result display.

Experimental application of the continuous reference intervals, in contrast to the reference intervals used previously, leads to substantial shifts in classification of samples as pathologic or not (Table 2). These data have to be interpreted with caution, as they are highly specific to our hospital population and the analytical methods used. Furthermore, the same data set that has been used to generate reference intervals has also been examined for shifts in classification; a decrease in samples classified as pathologic is therefore to be expected. However, 2 observations are noteworthy. First, the extent of change in classification is substantial, ranging from a 38% decrease to a 9% increase in the number of samples considered pathologic and a 23% decrease to a 14% increase in patients having at least 1 measurement considered pathologic. Second, the shifts in classification are more distinct in analytes exhibiting pronounced variation with age (e.g., hemoglobin, red cell indices, creatinine concentration) than in those analytes that do not (e.g., plasma electrolytes). These findings highlight the clinically relevant limitations of current reference intervals in representing the dynamics encountered in pediatric laboratory medicine.

Reference intervals in our study were determined with an indirect approach; the population analyzed consisted of inpatients and outpatients from a pediatric tertiary care center. The use of indirect methods for reference interval estimation is controversial, and the capability of these methods to correctly identify the subset of nonpathologic samples has been challenged (3, 7, 27-29). Therefore, validation of our results in other patient cohorts and with other analytical instruments is strongly warranted. However, concerns related to indirect methods are partly due to algorithm-specific restrictions, which include the assumption of an underlying normal gaussian distribution (i.e., a symmetrical distribution) of nonpathologic samples (28, 30). The algorithm we used does not assume a normal gaussian distribution of healthy samples but a Box-Cox distribution, which is not symmetrical and is often used in statistics to represent skewed data (31). Furthermore, we have validated the method in a previous study and shown that it generates reference intervals comparable to those of direct approaches, independent of the proportion of pathologic samples in the setting of a comprehensive hospital population (18). This is confirmed by comparison of our results to findings from CALIPER and KiGGS, showing a high concordance of reference limits and their age-dependent dynamics (see online Supplemental Figs. 1 and 2). The single analyte displaying major differences--magnesium reference intervals from the CALIPER trial--has been shown to be nontransferable in a CLSI-based transference study (32), explaining the observed disagreement. Furthermore, indirect approaches have been applied for reference interval calculation in the context of predominantly conventional methods: in the CALIPER trial, samples for children < 1 year old were collected from the maternity ward and outpatient clinics from "apparently healthy children" rather than from healthy children recruited from the community (6). On the other hand, the KiGGS survey covered aspects of pediatric health from birth to adolescence, yet blood samples were obtained only from children > 1 year old (17). The unique practical and ethical challenges of pediatric laboratory medicine therefore require the application of indirect and innovative approaches for reference interval calculation.

In the present study, we examined a data set of 22 frequently measured analytes from a single laboratory. The reference intervals were established in a German population of mainly white origin on a Sysmex XE-2100 (blood count) and a Cobas Integra 800 (clinical chemistry). The reported values are directly applicable only for this population and these analytical platforms, in which case the data tables published online enable integration into laboratory information systems to allow automatic classification of test results (see online Supplemental Database). However, transference of the published reference intervals according to guidelines is possible and allows their use in global populations and platforms (1). Moreover, the indirect approach exemplified here can be applied universally given a sufficiently large number of samples, which will allow validation of our results and testing of the presented approach in other patient cohorts. Furthermore, analysis of rare analytes or special patient subgroups can be performed with aggregate data from multiple centers. This will enable us to meet the particular challenges in the establishment of pediatric reference intervals, which are limiting the interpretation of laboratory test results in children and adolescents.


We report the sex- and age-dependent dynamics of 22 common hematologic and biochemical analytes from birth to adulthood and provide reference intervals for their interpretation. The complex dynamics in many analytes cannot be adequately represented by separation into age groups; the continuous description provided can therefore improve clinical decision making when interpreting pediatric laboratory test results.

Author Contributions: All authors confirmed they have contributed to the intellectual content of this paper and have met the following 3 requirements: (a) significant contribution to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; and (c) final approval of the published article.

Authors' Disclosures or Potential Conflicts of Interest: No authors declared any potential conflicts of interest.

Role of Sponsor: No sponsor was declared.

Acknowledgments: We thank the members of the working group on reference values of the German Society for Clinical Chemistry and Laboratory. Medicine ("AG Richtwerte der DGKL") for their valuable input.


(1.) CLSI. Defining, establishing, and verifying reference intervals in the clinical laboratory; approved guideline-third edition. Wayne (PA): CLSI; 2008. CLSI document C28-A3.

(2.) Solberg HE. The theory of reference values Part 5. Statistical treatment of collected reference values. Determination of reference limits. Clin Chim Acta 1984;137: 95-114.

(3.) Ceriotti F. Establishing pediatric reference intervals: a challenging task. Clin Chem 2012;58:808-10.

(4.) Jones G, Barker A. Reference intervals. Clin Biochem Rev 2008;29:S93-7.

(5.) Harris EK, Boyd JC. On dividing reference data into subgroups to produce separate reference ranges. Clin Chem 1990;36:265-70.

(6.) 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.

(7.) Loh TP, Antoniou G, Baghurst P, Metz MP. Development of paediatric biochemistry centile charts as a complement to laboratory reference intervals. Pathology (Phila)2014;46:336-43.

(8.) Ertl D-A, Gleiss A, Sagmeister S, Haeusler G. Determining the normal range for IGF-I, IGFBP-3, and ALS: new reference data based on current internal standards. Wien Med Wochenschr 1946 2014;164:343-52.

(9.) Brinkworth RS, Whitham E, Nazeran H. Establishment of paediatric biochemical reference intervals. Ann Clin Biochem 2004;41:321-9.

(10.) Pediatric reference intervals: critical gap analysis and establishment of a national initiative. Clin Biochem 2006;39:559-60.

(11.) Schnabl K, Chan MK, Gong Y, Adeli K. Closing the gaps in paediatric reference intervals: the CALIPER initiative. Clin Biochem Rev 2008;29:89-96.

(12.) 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.

(13.) 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.

(14.) Raizman JE, Cohen AH, Teodoro-Morrison T, Wan B, Khun-Chen M, Wilkenson C, et al. Pediatric reference value distributions for vitamins A and E in the CALIPER cohort and establishment of age-stratified reference intervals. Clin Biochem 2014;47:812-5.

(15.) 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.

(16.) Kohse KP, Thamm M. KiGGS: the German survey on children's health as data base for reference intervals. Clin Biochem 2011;44:479.

(17.) Dortschy R, Schaffrath Rosario A, Scheidt-Nave C, Thierfelder W, Thamm M, Gutsche J, et al. Bevolkerungs-bezogene verteilungswerte ausgewahlter laborparameter aus der studie zur gesundheit von kindern und jugendlichen in Deutschland (KiGGS). Berlin: Robert Koch-Institut; 2009.

(18.) Zierk J, Arzideh F, Haeckel R, Rascher W, Rauh M, Metzler M. Indirect determination of pediatric blood count reference intervals. Clin Chem Lab Med 2013;51:86372.

(19.) Arzideh F, Wosniok W, Haeckel R. Reference limits of plasma and serum creatinine concentrations from intra-laboratory data bases of several German and Italian medical centres: comparison between direct and indirect procedures. Clin Chim Acta 2010;411:215-21.

(20.) Arzideh F, Wosniok W, Haeckel R. Indirect reference intervals of plasma and serum thyrotropin (TSH) concentrations from intra-laboratory data bases from several German and Italian medical centres. Clin Chem Lab Med 2011;49:659-64.

(21.) Arzideh F, Brandhorst G, Gurr E, Hinsch W, Hoff T, Roggenbuck L, et al. An improved indirect approach for determining reference limits from intra-laboratory data bases exemplified by concentrations of electrolytes. Laboratoriums Medizin 2009;33:52-66.

(22.) R Development Core Team. R: a language and environment for statistical computing. Vienna (Austria); 2012.

(23.) De Kieviet W, Slaats EH, Abeling NG. Pediatric reference values for calcium, magnesium and inorganic phosphorus in serum obtained from Bhattacharya plots for data from unselected patients. J Clin Chem Clin Biochem 1986;24:233-42.

(24.) Wu A. Tietz clinical guide to laboratory tests, 4th Ed. St. Louis, MO: Saunders; 2006.

(25.) Guignard J-P, Drukker A. Why do newborn infants have a high plasma creatinine? Pediatrics 1999;103:e49.

(26.) Hann I, Arceci R, Smith O. Pediatric hematology, 3rd Ed. Malden, MA: Blackwell; 2006.

(27.) Solberg HE. Using a hospitalized population to establish reference intervals: pros and cons. Clin Chem 1994;40:2205-6.

(28.) Shaw JLV, Cohen A, Konforte D, Binesh-Marvasti T, Colantonio DA, Adeli K. Validity of establishing pediatric reference intervals based on hospital patient data: a comparison of the modified Hoffmann approach to CALIPER reference intervals obtained in healthy children. Clin Biochem 2014;47:166-72.

(29.) Bock BJ, Dolan CT, Miller GC, Fitter WF, Hartsell BD, Crowson AN, et al. The data warehouse as a foundation for population-based reference intervals. Am J Clin Pathol 2003;120:662-70.

(30.) Hoffmann RG. Statistics in the practice of medicine. JAMA 1963;185:864-73.

(31.) Sakia RM. The Box-Cox transformation technique: a review. J R Stat Soc Ser Stat 1992;41:169-78.

(32.) Estey MP, Cohen AH, Colantonio DA, Chan MK, Marvasti TB, Randell E, et al. CLSI-based transference of the CALIPER database of pediatric reference intervals from Abbott to Beckman, Ortho, Roche and Siemens clinical chemistry assays: direct validation using reference samples from the CALIPER cohort. Clin Biochem 2013; 46:1197-219.

(33.) Finne E, Bucksch J, Lampert T, Kolip P. Age, puberty, body dissatisfaction, and physical activity decline in adolescents. Results of the German Health Interview and Examination Survey (KiGGS). Int J Behav Nutr Phys Act 2011;8:119.

(34.) Klein G, Heiduk M, Paege I, Kliem C. Reference intervals for ALAT, ASAT, ALP, GGT, LDH, sTfR, and ferritin from infancy to childhood. Clin Chem 2005;51: A118.

(35.) Soldin SJ, Brugnara C, Wong EC. Pediatric reference intervals. 6th ed. Washington (DC): AACC Press; 2007.

(36.) Thomas L. Labor und diagnose. 6th ed. Frankfurt (Germany): TH-Books; 2005.

Jakob Zierk, [1] Farhad Arzideh, [2] Tobias Rechenauer, [1] Rainer Haeckel, [3] Wolfgang Rascher, [1] Markus Metzler, [1] and Manfred Rauh [1] *

[1] Department of Pediatrics and Adolescent Medicine, University Hospital Erlangen, Erlangen, Germany; [2] Department of Statistics, University of Bremen, Bremen, Germany; [3] Bremer Zentrum fur Laboratoriumsmedizin, Klinikum Bremen Mitte, Bremen, Germany.

* Address correspondence to this author at: Department of Pediatrics and Adolescent Medicine, University Hospital Erlangen, Loschgestr. 15,91054 Erlangen, Germany. Fax +499131/85-33700; e-mail

Preliminary results of this submission were presented at the XIII International Congress of Pediatric Laboratory Medicine, 2014, Istanbul, Turkey.

Received February 15, 2015; accepted April 27, 2015.

Previously published online at DOI: 10.1373/clinchem.2015.239731

[4] Nonstandard abbreviations: CALIPER, Canadian Laboratory Initiative on Pediatric Reference Intervals; KiGGS, German Health Interview and Examination Survey for Children and Adolescents; AST, aspartate transaminase; ALT, alanine transaminase; [gamma]-GT, [gamma]-glutamyltransferase; LDH, lactate dehydrogenase; MCH, mean red cell hemoglobin; MCHC, mean red cell hemoglobin concentration; MCV, mean red cell volume.
Table 1. Number of samples and distinct patients available for each
analyte and number of groups after partition into age groups. (a)

                                Distinct patients
Analyte                         Samples   Total    Male

  ALT                           51 574    12 842    6777
  Alkaline phosphatase          63 270    14 904    7865
  AST                           52 039    13714     7250
  Creatinine                    205 051   34 807   18 539
  [gamma]-GT                    45 978    11 162    5884
  LDH                           59 845    10 489    5546
  Plasma calcium                210 239   34 300   18217
  Plasma chloride               200 807   32 822   17 478
  Plasma magnesium              69 991    12 448    6568
  Plasma phosphate              138235    26 061   13 753
  Plasma potassium              209165    33 798   17 984
  Plasma sodium                 204 909   33 728   17 936
  Total protein                 164 954   29 717   15 833
  Hematocrit                    167079    32 274   16 996
  Hemoglobin                    167077    32 276   16997
  MCH                           167076    32 274   16 997
  MCHC                          167058    32 269   16 995
  MCV                           167079    32 275   16 997
  Platelets                     166 894   32 271   16 995
  Red cell count                167081    32 276   16 997
  Red cell distribution width   166230    32 257   16 985
  White cell count              167071    32 274   16 996

                                Distinct patients
Analyte                         Female   Age      Group
Biochemical                              groups   size

  ALT                            6047       52    1000
  Alkaline phosphatase           7025       89    1000
  AST                            6446       59    1000
  Creatinine                    16 131     197    1000
  [gamma]-GT                     5262       38    1000
  LDH                            4923       93     500
  Plasma calcium                15951       62    2500
  Plasma chloride               15220      172    1000
  Plasma magnesium               5855       50    1000
  Plasma phosphate              12210      139    1000
  Plasma potassium              15 683     184    1000
  Plasma sodium                 15 666     181    1000
  Total protein                 13 758     145    1000
  Hematocrit                    15209      153    1000
  Hemoglobin                    15210      153    1000
  MCH                           15208      153    1000
  MCHC                          15205      155    1000
  MCV                           15209      153    1000
  Platelets                     15207      153    1000
  Red cell count                15210      153    1000
  Red cell distribution width   15203      153    1000
  White cell count              15209       51    2500

(a) Group size denotes the number of samples in each age group.

Table 2. Proportion of samples and patients considered
pathologic. (a)
                                Current [RI.sub.b]   Proportion of
                                                     patients, %

Analyte                         Reference            Current   New
Biochemical                                            RI      RI

  ALT                           Klein et al. (34)      25      36
  Alkaline phosphatase          Package insert         20      19
  AST                           Klein et al. (34)      23      27
  Creatinine                    Package insert         34      27
  LDH                           Soldin et al. (35)     22      32
  Plasma calcium                Soldin et al. (35)     23      20
  Plasma chloride               Soldin et al. (35)     18      16
  Plasma magnesium              Package insert         23      23
  Plasma phosphate              Thomas (36)            19      19
  Plasma potassium              Soldin et al. (35)     14      17
  Plasma sodium                 Soldin et al. (35)     20      15
  Total protein                 Package insert         24      24
  Hematocrit                    Soldin et al. (35)     50      30
  Hemoglobin                    Soldin et al. (35)     48      28
  MCH                           Soldin et al. (35)     36      14
  MCHC                          Soldin et al. (35)     35      14
  MCV                           Soldin et al. (35)     42      19
  Platelets                     Soldin et al. (35)     42      27
  Red cell count                Soldin et al. (35)     43      31
  Red cell distribution width   Soldin et al. (35)     18      32
  White cell count              Soldin et al. (35)     41      30

                                Proportion   Proportion of samples, %

Analyte                         Absolute     Current   New   Absolute
Biochemical                      change        RI       RI     change

  ALT                              10          14       23        9
  Alkaline phosphatase             -1          24       17       -7
  AST                               4          18       20        2
  Creatinine                       -6          25       16       -9
  LDH                              10          18       26        8
  Plasma calcium                   -3          26       18       -8
  Plasma chloride                  -2          22       19       -3
  Plasma magnesium                  0          26       23       -4
  Plasma phosphate                  1          24       21       -3
  Plasma potassium                  3          19       22        3
  Plasma sodium                    -5          26       19       -7
  Total protein                    -0          20       16       -4
  Hematocrit                      -20          56       18      -38
  Hemoglobin                      -20          52       15      -37
  MCH                             -22          30        9      -21
  MCHC                            -22          42       17      -25
  MCV                             -23          40       13      -26
  Platelets                       -15          35       17      -17
  Red cell count                  -13          38       16      -22
  Red cell distribution width      14          17       20        3
  White cell count                -11          42       27      -15

(a) Patients were classified healthy with regard to a specific
analyte if all their measurements for that analyte were also
classified healthy, and pathologic otherwise.

(b) RI, reference interval.
COPYRIGHT 2015 American Association for Clinical Chemistry, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2015 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Pediatric Clinical Chemistry
Author:Zierk, Jakob; Arzideh, Farhad; Rechenauer, Tobias; Haeckel, Rainer; Rascher, Wolfgang; Metzler, Mark
Publication:Clinical Chemistry
Article Type:Report
Geographic Code:4EUGE
Date:Jul 1, 2015
Previous Article:Tandem mass spectrometric determination of atypical 3[beta]-hydroxy-[[DELTA].sup.5]-bile acids in patients with...
Next Article:Noninvasive detection of activating estrogen receptor 1 (ESR1) mutations in estrogen receptor-positive metastatic breast cancer.

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters |