Skip to main content
NCBI home page
HHS Author Manuscripts logoLink to HHS Author Manuscripts
. Author manuscript; available in PMC: 2023 May 3.
Published in final edited form as: JAMA Pediatr. 2022 Jul 1;176(7):699–714. doi: 10.1001/jamapediatrics.2022.0794

Current State of Pediatric Reference Intervals and the Importance of Correctly Describing the Biochemistry of Child Development A Review

Alicia N Lyle 1, Fidelia Pokuah 1, Dennis J Dietzen 1, Edward C C Wong 1, Amy L Pyle-Eilola 1, John S Fuqua 1, Alison Woodworth 1, Patricia M Jones 1, Lara J Akinbami 1, Luigi R Garibaldi 1, Hubert W Vesper 1
PMCID: PMC10155856  NIHMSID: NIHMS1847211  PMID: 35467725

Abstract

IMPORTANCE

Appropriately established pediatric reference intervals are critical to the clinical decision-making process and should reflect the physiologic changes that occur during healthy child development. Reference intervals used in pediatric care today remain highly inconsistent across a broad range of common clinical biomarkers.

OBSERVATIONS

This narrative review assesses biomarker-specific pediatric reference intervals and their clinical utility with respect to the underlying biological changes occurring during development. Pediatric reference intervals from PubMed-indexed articles published from January 2015 to April 2021, commercial laboratory websites, study cohorts, and pediatric reference interval books were all examined. Although large numbers of pediatric reference intervals are published for some biomarkers, very few are used by clinical and commercial laboratories. The patterns, extent, and timing of biomarker changes are highly variable, particularly during developmental stages with rapid physiologic changes. However, many pediatric reference intervals do not capture these changes and thus do not accurately reflect the underlying biochemistry of development, resulting in significant inconsistencies between reference intervals.

CONCLUSIONS AND RELEVANCE

There is a need to correctly describe the biochemistry of child development as well as to identify strategies to develop accurate and consistent pediatric reference intervals for improved pediatric care.

Hypotehtical case scenario: a 1-week-old infant has an abnormal thyrotropin (previously, thyroid-stimulating hormone) laboratory result on the state newborn screening test. At the 10-day-old follow-up visit, the infant’s primary care physician obtains a thyrotropin result of 9.4 mIU/L from the local laboratory, along with a pediatric reference interval (PRI) of 0.7 to 4.8 mIU/L for this age range. Based on these laboratory results, the physician tells the family that the infant has congenital hypothyroidism and begins treatment with levothyroxine. After performing additional literature reviews, the physician finds that PRIs published in the scientific literature and used in clinical laboratories vary greatly and that normal thyrotropin concentrations can measure as high as 27.2 mIU/L in the first month after birth. The physician realizes that he may have started the infant taking a medication that was not necessary.1,2 The family is upset that they were given erroneous information and treatment. The infant is referred to a pediatric endocrinologist to discuss how and when to safely stop levothyroxine based on current clinical guidelines to avoid rebound hypothyroidism.3

PRIs are critical for correct clinical decision-making.4–6 This case presents just 1 example of how inconsistent PRIs can impact clinical decision-making. Inconsistency between PRIs applies to a broad range of biomarkers. For example, a PRI study published for hematology biomarkers showed that 63% of the patient samples measured at the local hospital would change classification from normal to disease depending on the reference interval used.7 The physical development of children into adults requires appropriate growth and maturation of the body and its organ systems, which is accompanied by fluctuations in biomarker concentrations at various stages of healthy child development. In turn, PRIs must accurately reflect these dynamic biochemical and physiological changes in healthy children to ensure correct diagnosis and treatment of disease in children.

Despite notable efforts throughout the past 30 years, the PRIs currently used in pediatric care remain highly inconsistent.7–9 To address these inconsistencies, reviews and guidelines about PRIs typically focus on the technical aspects of creating reference intervals, such as statistical methods, population characteristics, and analytical considerations.4,9–11 In this review, we assess the consistency of PRIs reported on selected analytes in the context of the biological changes occurring during child development and identify strategies to develop accurate and consistent PRIs for pediatric care.

Methods

Literature Search Strategy

This review includes PRIs from PubMed-indexed articles published from January 2015 to April 2021. Transference studies and abstractswere not included. Literature search terms included, but were not limited to, pediatric reference intervals and/or children or pediatric in combination with the analytes listed below. After review, 18 articles were selected for inclusion. PRIs published by major commercial laboratories and study cohorts, such as the Canadian Laboratory Initiative on Pediatric Reference Intervals (CALIPER), were collected from websites.12 A member of the American Association for Clinical Chemistry (AACC) Pediatric Reference Interval Working Group (D.J.D) collected additional PRI data from 9 clinical practices and tertiary care, academic, and pediatric hospitals. References published prior to 2015 were included if referenced in selected PRI books or reviews.2,13–15

Analytes commonly measured in the pediatric clinical setting were used as examples for this review and included the thyroid axis hormones free thyroxine (FT4) and thyrotropin, the kidney function biomarker cystatin C, insulin-like growth factor 1 (IGF-1), total testosterone, estradiol (E2), and the iron-status indicators hemoglobin and ferritin. Each PRI source was examined for information on the laboratory method used, population surveyed, statistical approaches used, covariates, and published reference intervals.

Analysis of Analyte-Specific PRI

The use of PRIs for a specific biomarker in clinical practice is complicated by the availability of multiple PRIs from different sources. There is no common data source for deriving PRIs that uses a well-defined population representative of healthy US children. Many PRIs are from populations of other countries. PRIs reported by clinical laboratories often do not include information about the population from which they were derived.

Biomarker concentrations for most PRIs in this study differ across sources, in part, owing to variably calibrated analytical systems and differences in analytical system performance characteristics.16 However, these kinds of differences among analytical systems should not affect the patterns of biomarker changes at different developmental stages, such as the spike in thyrotropin in the first days of life. We observed inconsistent patterns for biomarker changes across PRIs (Figure 1 and Figure 2), partially attributable to the number of age groups used to establish PRIs. Age groups should reflect developmental stages during which biomarker concentrations are relatively constant. The number of age groups are highly variable across PRIs and range between 1 and 12 for hemoglobin and ferritin, for example. The number of individuals per age group is also highly variable and sometimes does not meet the minimum of 120 individuals per stratum to obtain robust estimates, as recommended by Clinical and Laboratory Standards Institute guideline document EP-28 (eTable 1 in the Supplement).4

Figure 1. Thyrotropin and Free Thyroxine (FT4) Pediatric Reference Interval (PRI) Patterns.

Figure 1.

Open in a new tab

PRI patterns are heterogeneous and illustrate dynamic changes at different ages. A, Thyrotropin PRIs from birth to age 12 months from 4 PRI sources. B, FT4 PRIs from birth to age 12 months corresponding with the same 4 PRI sources as thyrotropin panel A.

Figure 2. Insulin-like Growth Factor 1 (IGF-1) and Ferritin Pediatric Reference Interval (PRI) Patterns.

Figure 2.

Open in a new tab

PRI patterns are heterogeneous and illustrate dynamic changes at different ages. A, IGF-1 PRIs from birth to age 18 individuals from 4 PRI sources. years for males and female individuals from 2 PRI sources. B, Ferritin PRIs from birth to age 18 years for male

Other factors that contribute to variability across PRIs for certain biomarkers are diurnal variability and menstrual cycle, inclusion and exclusion criteria used, and statistical procedures. Most studies do not provide details about these factors. Discussions of how these factors influence PRIs are addressed in detail by other reviews and are outside the scope of this review.9,17–21 This review investigated how well dynamic changes in biomarkers during child development are reflected in recently published PRIs and PRIs currently used in pediatric care.

Results

Biomarkers of Thyroid Function: Free Thyroxine and Thyrotropin

Normal thyroid gland function in children is critical for early neurocognitive development and is important for growth and development from birth through adolescence, making early identification and treatment of thyroid diseases crucial. Hyperthyroidism accounts for 15% of pediatric thyroid disorders. Hypothyroidism affects 1 in 2000 to 4000 newborns, and screening is part of all newborn screening programs in the United States.22,23 Physiologic thyrotropin and FT4 serum concentrations surge in the first few days of life and normalize with the maturation of the thyrotropin/FT4 axis throughout the next few months.24,25

For thyroid function analytes, a large number of inconsistent PRIs were found. We identified 21 PRIs for FT4 and 21 PRIs for thyrotropin (Table 1).26–29,31–34 Four PRIs for FT4 and 5 for thyrotropin were derived from the same prospective study cohort (CALIPER) using the same statistical approaches but different assays. Information about the study populations for other PRIs was sometimes unavailable. PRIs stratified by sex were provided for 38% (8 of 21 studies) and 48% (10 of 21 studies) of the FT4 and thyrotropin studies, respectively. Only the CHILDx study30 described additional screening criteria used to ensure inclusion of healthy children with normal thyroid function, such as values for thyroid peroxidase autoantibodies.

Table 1.

Pediatric Reference Intervals for Male and Female Individuals From the Literature and Major US Commercial Clinical Laboratories: FT4 and Thyrotropin

Source Age range Sample size Reference ranges Analytical equipment
FT4, ng/dL
Higgins et al.,26 2018; CALIPER database12 0−<2 wk 60 0.97–5.63 ITROS 5600 (Ortho)
2 wk-<5 y 136 0.89–2.27
5–19 y 434 0.76–1.33
Soldin et al.,15 2011 0–3 d N = 119 2.0–5.0 VITROS ECi (Ortho)
3–30 d 0.9–2.2
31 d-18 y 0.8–2.0
Karbasy et al.,27 2016; CALIPER database12 0-<20 d 40 1.35–4.48 DxI and Access2 (Beckman Coulter)
20 d-<3 y 215 0.74–1.38
3−<19 y 455 0.61–1.06
Bohn et al.,1 2016; CALIPER database12 0−<1 mo 46 1.24–3.89 Cobas 8000 e602 (Roche)
1-<12 mo 79 1.09–1.71
1-<19 y 491 1.01–1.63
Lim,28 2017 N = 1978 Cobas 6000 (Roche)
10–12 y Male: 298; female: 298 Male: 1.04–1.69; female:
0.98–1.64
13–15 y Male: 386; female: 294 Male: 0.97–1.68; female:
0.95–1.53
16–18 y Male: 360; female: 342 Male: 1.04–1.72; female:
0.95–1.62
Gunapalasingham et al.,29 2019 N = 2407 Cobas 6000 (Roche)
6.0–9.9 y Male: 395; female: 471 Male: 0.99–1.18; female:
1.01–1.19
10.0–14.9 y Male: 416; female: 605 Male: 0.95–1.17; female:
0.93–1.14
15.0–18.9 y Male: 165; female: 359 Male: 0.94–1.20; female:
0.92–1.15
Wong et al.,2 2020 0-≤5 y Male: 208; female: 208 Male: 0.85–2.82; female:
0.85–2.82		 Cobas 6000 (Roche)
5-≤15 y Male: 238; female: 238 Male: 0.81–2.10; female:
0.81–2.15
15-≤20 y Male: 28; female: 36 Male: 0.77–2.82; female:
0.78–1.46
Wong et al.,2 2020; CALIPER database12 5-<15 d 66 1.05–3.21 Architect i2000 (Abbott)
15-<30 d 55 0.68–2.53
30 d-<1 y 270 0.89–1.7
1-<19 y 952 0.89–1.37
Soldin et al.,15 2011 1–3 d Male: 24; female: 38 Male: 0.80–2.78; female:
0.88–1.93		 IMx Analyzer (Abbott)
4–30 d Male: 73; female: 62 Male: 0.48–2.32; female:
0.61–1.93
1–12 mo Male: 52; female: 54 Male: 0.76–2.0; female:
0.88–1.84
1–5 y Male: 100; female: 117 Male: 0.90–1.59; female:
1.02–1.72
6–10 y Male: 104; female: 101 Male: 0.81–1.68; female:
0.82–1.58
11–15 y Male: 101; female: 100 Male: 0.92–1.57; female:
0.79–1.49
16–18 y Male: 110; female: 101 Male: 0.92–1.53; female:
0.83–1.44
Soldin et al.,15 2011 N = 6023 Architect ci8200 (Abbott)
<2 mo Male: 189; female: 160 Male: 0.78–1.83; female:
0.69–1.85
2-<12 mo Male: 125; female: 138 Male: 0.71–1.56; female:
0.69–1.49
12–24 mo Male: 190; female: 202 Male: 0.82–1.32; female:
0.73–1.41
2-<5 y Male: 286; female: 253 Male: 0.80–1.32; female:
0.79–1.38
5-<10y Male: 516; female: 655 Male: 0.78–1.29; female:
0.77–1.32
10-<15y Male: 722; female: 1003 Male: 0.69–1.23; female:
0.66–1.22
15-<20y Male: 507; female: 802 Male: 0.67–1.22; female:
0.67–1.22
Soldin et al.,15 2011 N = 1459 Architect ci8200 (Abbott)
0–12 mo Male: 36; female: 43 Male: 0.92–1.83; female:
0.85–1.59
1–5 y Male: 101; female: 93 Male: 0.86–1.62; female:
0.91–1.44
6–10 y 139 0.84–1.47
11–14y 161 0.78–1.31
15–20 y 163 0.79–1.34
La'ulu et al.,30 2016 6 mo-6 y 840 1.40–2.70 Equilibrium dialysis-liquid chromatography-tandem mass spectrometry
7–17 y 1373 1.10–2.00
Soldin et al.,15 2011 1–3d Male: 24; female: 38 Male: 1.16–2.95; female:
1.09–2.09		 Immuno I (Bayer)
4–30 d Male: 73; female: 62 Male: 0.78–2.25; female:
0.85–2.09
1–12 mo Male: 52; female: 54 Male: 1.00–2.17; female:
1.09–2.02
12 mo-5 y Male: 100; female: 117 Male: 1.16–1.71; female:
1.24–1.86
6–10 y Male: 104; female: 101 Male: 1.09–1.86; female:
1.0–1.71
11–15 y Male: 101; female: 100 Male: 1.16–1.71; female:
1.0–1.63
16–18 y Male: 110; female: 101 Male: 1.16–1.71; female:
1.09–1.63
Soldin et al.,15 2011 1 mo-18 y 1000 0.8–2.1 Isotope dilution tandem mass spectrometry in negative mode with ultracentrifugation
1–3 d ND Male: 0.97–1.87; female:
0.93–1.44		 Dimension RxLAnalyzer (Siemens)
4–30 d Male: 0.78–1.52; female:
0.81–1.44
1–12 mo Male: 0.89–1.48; female:
0.93–1.40
1–5 y Male: 0.97–1.25; female:
1.01–1.32
6–10 y Male: 0.93–1.32; female:
0.93–1.25
11–15 y Male: 0.97–1.25; female:
0.89–1.20
16–18 y Male: 0.97–1.25; female:
0.93–1.25
Wong et al.,2 2020 1–30 d 47 2.00–2.13 ADVIA Centaur (Siemens)
31–60 d 58 0.96–1.69
61 d-1 y 317 0.84–1.52
1–5 y 2722 0.91–1.48
6–10 y 3452 0.88–1.45
11–14 y 3429 0.82–1.39
15–18 y 1019 0.81–1.40
Soghier et al.,13 2019 5-<15 d ND 1.05–3.21 DELFIA immunofluorometric system (PerkinElmer)
15-<30 d 0.68–2.53
30 d-<1 y 0.89–1.70
1–19 y 0.89–1.37
ARUP Laboratories31 25–30 Weeks’ gestation ND 0.5–3.3 Quantitative equilibrium dialysis/HPLC-MS/MS
31–36 Weeks’ gestation 1.3–4.7
Birth-1 wk 2.2–5.3
2–3 wk 0.9–4.0
1–5 mo 1.1–2.2
6 mo-6 y 1.4–2.7
7–17 y 1.1–2.0
≥18 y 1.1–2.4
LabCorp32 0–3 d Varies 0.66–2.71 Electrochemiluminescence
immunoassay
4–30 d 0.83–3.09
31 d-12 mo 0.48–2.34
13 mo-5 y 0.85–1.75
6–10 y 0.90–1.67
11–19 y 0.93–1.60
>19 y 0.82–1.77
Mayo Clinic Laboratories33 0–5 d >120 Per age range per
sex		 0.9–2.5 Electrochemiluminescence
Immunoassay
6 d-2 mo 0.9–2.2
3–11 mo 0.9–2.0
1–5 y 1.0–1.8
6–10 y 1.0–1.7
11–19 y 1.0–1.6
Quest Diagnostics34 <1 mo ND Not established Immunoassay
1–23 mo 0.9–1.4
2–12 y 0.9–1.4
13–20 y 0.8–1.4
>20 y 0.8–1.8
Thyrotropin, mIU/L
Bohn et al.,1 2019; CALIPER database12 0-<1 mo 50 1.23–27.2 Cobas 8000 e602 (Roche)
1-<12 mo 86 1.03–6.80
1-<15 y 339 1.12–5.01
15-<19 y 148 0.68–4.09
Wong et al.,2 2020; CALIPER database12 N = 600 Cobas 6000 (Roche)
0-≤5 y Male: 189; female: 189 Male: 0.84–6.22; female:
0.84–6.22
5-≤10 y Male: 47; female: 61 Male: 1.18–5.33; female:
0.48–4.81
10-≤15 y Male: 118; female: 118 Male: 0.76–4.20; female:
0.76–4.20
15-≤20 y Male: 26; female: 35 Male: 0.64–5.37; female:
0.38–2.82
Gunapalasingham et al.,29 2019 N = 2409 Cobas 6000 (Roche)
6.0–9.9 y Male: 395; female: 471 Male: 0.18–0.30; female:
0.16–0.30
10.0–14.9 y Male: 416; female: 605 Male: 0.17–0.30; female:
0.17–0.31
15.0–18.9 y Male: 165; female: 359 Male: 0.17–0.29; female:
0.16–0.30
Lim,28 2017 N = 1044 Cobas 8000 e602 (Roche)
10–12 y Male: 298; female: 298 Male: 1.09–8.50; female:
0.61–7.84
13–15 y Male: 386; female: 294 Male: 0.49–5.90; female:
0.63–7.39
16–18 y Male: 360; female: 342 Male: 0.63–6.31; female:
0.55–6.49
Higgins et al.,26 2018; CALIPER database12 0-<1 wk 51 0.71–31.5 VITROS 5600 (Ortho)
1 wk-<19 y 600 0.86–5.7
Soldin et al.,15 2011 0–3 d 119 1.0–20.0 VITROS ECi,
chemilluminescent immunoassay (Ortho)
3–30 d 0.5–6.5
1–5 mo 0.5–6.0
6 mo-18 y 0.5–4.5
Karbasy et al.,27 2016; CALIPER database12 0-<12 y 423 0.79–5.85 DxI and Access2 (Beckman Coulter)
12-<19 y 182 0.68–3.35
Wong et al.,2 2020; CALIPER database12 4 d-<6 mo 139 0.73–4.77 Architect i2000 (Abbott)
6 mo-<14 y 640 0.7–4.17
14-<19 y 259 0.47–3.41
Soldin et al.,15 2011 N = 1459 Architect ci8200 (Abbott)
0–12 mo Male: 71 Male: 0.9–5.4; female: 0.9–5.4
1–5 y Male: 152; female: 155 Male: 0.7–4.5; female: 0.7–4.8
11–14 y Male: 93; female: 201 Male: 0.6–3.6; female: 0.5–4.1
Soldin et al.,15 2011 N = 6023 Architect ci8200 (Abbott)
<2 mo Male: 212; female: 171 Male: 1.1–6.3; female: 1.1–5.5
2-<12 mo Male: 138; female: 152 Male: 1.0–4.9; female: 1.1–4.5
12–24 mo Male: 197; female: 203 Male: 0.9–4.8; female: 1.0–4.4
2-<5 y Male: 297; female: 264 Male: 0.8–4.4; female: 0.9–4.1
5-<10 y Male: 537; female: 697 Male: 0.8–4.1; female: 0.9–4.1
10-<15 y Male: 746; female: 1063 Male: 0.8–4.0; female: 0.7–3.7
15-<20 y Male: 547; female: 877 Male: 0.6–3.6; female: 0.5–3.6
Soldin et al.,15 2011 1–30 d Male: 63; female: 89 Male: 0.5–16.0; female: 0.7–13.1 IMx analyzer (Abbott)
1 mo-5 y Male: 95; female: 95 Male: 0.6–7.1; female: 0.5–8.1
6–18 y Male: 95; female: 96 Male: 0.4–6.0; female: 0.4–5.8
Soghier et al.,13 2019 4 d-<1 y ND 0.73–4.77 House RIA; TSH immunoradiometric assay (Dynotest)
6 mo-<14 y 0.7–4.17
14-<19 y 0.47–3.41
Soldin et al.,15 2011 0–1 mo Male: 84; female: 62 Male: <6.5; female: <6.0 Immunofluorescent TSH Kit (Delphia)
1–12 mo Male: 114; female: 103 Male: <4.1; female: <4.0
1–3 y Male: 128; female: 126 Male: <3.0; female: <3.3
4–6 y Male: 109; female: 82 Male: <3.0; female: <2.8
7–12 y Male: 112; female: 107 Male: <3.1; female: <2.9
13–18 y Male: 106; female: 106 Male: <3.1; female: <3.0
Soldin et al.,15 2011 1–30 d Male: 63; female: 89 Male: 0.7–16.1; female: 1.0–13.7 Immuno I
1 mo-5 y Male: 95; female: 95 Male: 0.8–7.5; female: 0.7–8.6
6–18 y Male: 95; female: 96 Male: 0.6–6.4; female: 0.6–6.2
Soldin et al.,15 2011 N = 1459 ACS 180 using Chiron
Diagnostic TSH-3 kit
0-<2 y 36 0.7–4.5
2-<7 y 149 0.4–3.2
7-<13 y 128 0.3–2.7
13-<18 y 123 0.4–1.9
Soldin et al.,15 2011 1–30 d ND Male: 0.6–12.8; female: 0.8–10.8 Dimension RxL Analyzer (Siemens)
1 mo-5 y Male: 0.7–6.0; female: 0.6–6.8
6–18 y Male: 0.45–5.1; female: 0.5–4.9
Wong et al.,2 2020 1–30 d 48 1.08–11.8a ADVIA Centaur (Siemens)
31–60 d 56 0.68–12.56a
61 d-1 y 321 0.62–7.3a
1–5 y 2782 0.75–6.57a
6–10 y 3531 0.79–6.0a
11–14 y 3487 0.72–5.77a
15–18 y 1086 0.63–6.28a
ARUP Laboratories31 Cord blood ND 2.000–40.000 Quantitative
Electro-chemiluminescent
Immunoassay, 3rd Generation
0–3 d 5.170–14.600
4–30 d 0.430–16.100
1–24 mo 0.620–8.050
2–6 y 0.540–4.530
7–11 y 0.660–4.140
12–19 y 0.530–3.590
LabCorp32 0–6 d ND 0.0000007–0.0000152 Electrochemiluminescence
immunoassay
7 d-3 mo 0.00000072–0.000011
>3–12 mo 0.00000073–0.00000835
1–5 y 0.0000007–0.00000597
6–10 y 0.0000006–0.00000484
>10 y 0.000000450–0.0000045
Mayo Clinic Laboratories33 0–5 d >120 Per age range per sex 0.7–15.2 E immunoassay TSH method
(Cobas)
6 d-2 mo 0.7–11.0
3–11 mo 0.7–8.4
1–5 y 0.7–6.0
6–10 y 0.6–4.8
11–19 y 0.5–4.3
Quest Diagnostics34 Premature infants ND 0.20–27.90 Ultrasensitive 3rd generation
1–2 d 3.20–34.60
3–4 d 0.70–15.40
1–11 mo 0.80–8.20
1–19 y 0.50–4.30
Open in a new tab

Abbreviations: FT4, free thyroxine; HPLC-MS/MS, high-performance liquid chromatography-tandem mass spectrometry; ND, not disclosed; RIA, radioimmunoassays; TSH, thyroid-stimulating hormone.

SI conversion factors: To convert FT4 to picomoles per liter, multiply by 12.871; thyrotropin to milli-international units per liter, multiply by 1.

a

Values are reported as milligrams per liter.

The total number of individuals and number of age groups differed across PRIs (eTable 1 in the Supplement). Many PRIs captured the neonatal surges in FT4 and thyrotropin; however, others failed to do so. This is a result of the first age groups spanning anywhere from 0 to 3 days to 0 to 5 years for FT4 and0 to3 days to0 to 11 years for thyrotropin (Figure 1 and eFigure in the Supplement). Nine published PRIs did not provide information about the number of individuals included. CALIPER studies reported instrument-specific PRIs for the Cobas (Roche), Architect (Abbott), Vitros (Ortho), and DxL (Beckman) systems, illustrating how a lack of assay standardization can result in instrument-specific PRIs. Despite use of the same cohort and statistics, CALIPER PRIs still varied in the number of age groups, individuals per age group, and extent of changes between age groups (Table 1).

Of the 9 clinical laboratories that were surveyed (D.J.D.), 7 of 9 reported thyrotropin intervals indicative of the neonatal surge (eTable 2 in the Supplement). Only 3 laboratories use PRIs published in the literature, while 6 laboratories do not, suggesting that these laboratories did not adopt published PRIs and established PRIs internally. As observed with published PRIs, heterogeneity in patterns across clinical laboratory PRIs appear to stem from differences in age grouping, some of which inadequately reflect the dynamic changes that occur in childhood (eTable 2 in the Supplement). When comparing PRIs for thyrotropin and the corresponding FT4 PRIs from the same laboratory, age grouping differed for 7 of 9 laboratories. Considering that FT4 and thyrotropin are highly regulated through feedback loops, this observation suggests that FT4 and thyrotropin PRIs used by some laboratories do not appropriately reflect the underlying physiologic pathway (eTable 2 in the Supplement).

Sex Hormone Biomarkers: E2 and Testosterone

E2 and testosterone are required for normal pubertal development and sexual maturation. The secondary sex characteristics that develop during puberty are characterized according to Tanner stage.35 Clinically, E2 and total testosterone concentrations are assessed in the evaluation of pituitary diseases, irregular menses, suspected polycystic ovary syndrome, hypogonadism, and other diseases related to sexual development.36

E2 concentrations in female individuals are typically high at birth and decline rapidly throughout the first few weeks of life.37 They decrease further in the first year and remain low until puberty, which typically occurs after age 8 years. Concentrations rise progressively throughout puberty and may demonstrate cyclic variation even before menarche. After menarche, E2 concentrations vary with the menstrual cycle.37

Serum total testosterone concentrations are high in male individuals at birth, decline rapidly in the first week of life, and then increase again, peaking at about age 2 to 3 months, a phenomenon termed the mini-puberty of infancy.38 By approximately age 6 months, total testosterone concentrations decline and remain low throughout the prepubertal years. In male individuals, puberty typically begins after age 9 years when total testosterone concentrations start gradually rising and reach adult levels by the later stages of puberty, typically in the midteenage years. In female individuals, total testosterone concentrations are low throughout infancy and childhood and increase modestly during puberty but remain much lower than in male individuals. Importantly, total testosterone concentrations exhibit diurnal patterns, being highest in the morning, and these diurnal variations are most pronounced in pubertal male individuals.36

For E2 and total testosterone, we identified 15 PRIs and 12 PRIs, respectively (Table 239-48). About half of these PRIs (6 for E2 and 7 for total testosterone) were derived from the CALIPER study using the same statistical approaches. All studies reported sex-specific PRIs. Seven PRIs for E2 and 4 for total testosterone provided data by Tanner stage. None of the E2 PRIs provided information about the time point in the menstrual cycle when the sample was collected, nor was time of day information provided to account for total testosterone diurnal variations.

Table 2.

Pediatric Reference Intervals for Male and Female Individuals From the Literature and Major US Commercial Clinical Laboratories: E2 and Total Testosterone

Source Age range and Tanner stage Sample size Reference ranges Analytical equipment
E 2 pg/ml
Wong et al.,2 2020; CALIPER database12 Male: 15 d-<1 y; female: 15 d-<1 y Male: 197;female: 197 <25  Architect Í2000SR (Abbott)
Male: 1-<11 y; female: 1-<9 y Male: 331; female: 261 Male: <13; female: <10
Male: 11-<13 y; female: 9-<11 y Male: 81; female: 81 Male: <26; female: <48
Male: 13-<15 y; female: 11-<12 y Male: 80; female: 87 Male: <28; female: <94
Male: 15-<19 y; female: 12-<14 y Male: 37; female: 83 Male: <38; female: 11–172
Female: 14-<19 y Female: 175 Female: <255
T1 Male: 27; female: 22 Male: <19; female: <20
T2 Male: 38; female: 23 Male: <18; female: <26
T3 Male, 23; female: 26 Male: <21; female: <86
T4 Male: 47; female: 57 Male: <35; female: 13–141
T5 Male: 16; female: 30 Male: 17–34; female: 19–208
Higgins et al.,26 2018; CALIPER database12 0-<1 mo Male: 68; female: 68 44.2–376  VITROS 5600 (Ortho)
1 mo-<6 y Male: 196; female: 196 <6.36–48.2
6-<11y Male: 144; female: 144 <6.36–18.5
11-<15 y Male: 105; female: 86 Male: 6.92–39.6; female: 18.6–93.6
15-<19 y Male: 88; female: 85 Male: 13.2–39.2; female: 12.2–181
Bogner 201939 14–17y Male: 63; female: 39 Male: 6.929–38.507; female: 6.780–106.897  VITROS ECiQ (Ortho)
Karbasy et al.,27 2016; CALIPER database12 0-<1 y Male: 174; female: 174 <20–52.51  DxI (for age >1 y) and Access2 (for age <1 y) (Beckman Coulter)
1-<12 y Male: 321; female: 321 <20-<20
12-<19 y Male: 95; female: 96 Male: <20–42.83; female: <20–198.2
Bohn et al.,1 2019; CALIPER database12 0-<1 mo Male: 50; female: 50 2.54–96  Cobas 8000 e602 (Roche)
1 mo-<10 y Male: 267; female: 267 <4.90
Male: 10-<19 y; female: 10-<14 y Male: 157; female: 76 Male: <4.90–36.5; female: <4.90–68.1
Female: 14-<19 y Female: 86 Female: 14.6–248
Bae et al.,40 2019 <1 y Male: 124; female: 156 Male: <20.484; female: <22.064  LC-MS/MS
1–5 y Male: 206; female: 47 Male: <19.449; female: <17.243
5–10 y Male: 649; female: 557 Male: <10.078; female: <11.849
10–15 y Male: 731; female: 629 Male: <28.330; female: <228.084
T1 Male: 266; female: 224 <10.078
T2 Male: 75; female: 99 Male: <11.005; female: <83.410
T3 Male: 63; female: 74 Male: <32.743; female: <227.894
T4 Male: 73; female: 68 Male: <35.412; female: <271.697
T5 Male: 114; female: 90 Male: <52.764; female: <269.027
Wong et al.,2 2020 7-<10 y Male: 94; female: 65 Male: <6; female: <35  LC-MS/MS
10-<13 y Male: 95; female: 120 Male: <10; female: <87
13-<16 y Male: 81; female: 127 Male: 1–36; female: 9–248
16-<18 y Male: 49; female: 129 Male: 3–34; female:
2–266
T1 Male: 134; female: 158 Male: <8; female: <55
T2 Male: 60; female: 75 Male: <9; female: 2–133
T3 Male: 53; female: 100 Male: 1–35; female: 12–277
T4, T5 Male: 74; female: 108 Male: 3–35; female: 2–259
Ankarberg-Lindren et al.,41 2018 ND Male (TV, mL): 1–2: 9; female (TBS): 1: 15 Male: <0.54–3.81; female: <0.54–1.91  GC-MS/MS
Male (TV, mL): 3–6: 9; female (TBS): 2: 15 Male: <0.54–5.99; female: 1.63–12.26
Male (TV, mL): 8–12: 14; female (TBS): 3: 13 Male: 0.82–19.34; female: 10.08–160.45
Male (TV, mL): 15–25: 9; female (TBS): 4–5: 16 Male: 6.81–25.88; female: 24.24–211.93
Soldin et al.,15 2011 Male: 0-<19 y; female: 0-<6 y Male: 132; female: 50 Male: <20–40; female: <20–53  DPC Immulite 1000 (Siemens)
Female: 6-<11 y Female: 103 Female: <20–59
Female: 11-<15 y Female: 61 Female: <20–87
Female: 15-<19 y Female: 55 Female: <20–111
Wu,14 2006 30–60d Male: 10–32; female: 5–50  RIA
6 mo-10 y <15
T1 Male: 3–15; female: 5–10
T2 ND Male: 3–10; female: 5–115
T3 Male: 5–15; female: 5–180
T4 Male: 3–40; female: 25–345
T5 Male: 15–45; female: 25–410
Mayo Clinic42 T1 (>14 d-prepubertal) Male: LOD-13; female: L0D-20  LC-MS/MS
T2 Male: LOD-16; female: LOD-24
T3 ND Male: LOD-26; female: L0D-60
T4 Male: LOD-38; female: 15–85
T5 Male: 10–40; female: 15–350
ARUP Laboratories31 7–9 y Male: <7.0; female: <36.0  LC-MS/MS
10–12y Male: <11.0; female: 1.0–87.0
13–15 y Male: 1.0–36.0; female: 9.0–249.0
16–17 y Male: 3.0–34.0; female: 2.0–266.0
T1 (>14 d-prepubertal) Male: <8.0; female: <56.0
T2 Male: <10.0; female: 2.0–133.0
T3 Male: 1.0–35.0; female: 12.0–277.0
T4, T5 Male: 3.0–35.0; female: T5: 2.0–259
LabCorp43 Male: 0.5–10 y; female: 1–9 y <15.0  LC-MS/MS
T1: male: <9.8 y; female: <9.2y T1: male: 5.0–11.0; female: 5.0–20.0
T2: male: 9.8–14.5 y; female: 9.2–13.7 y T2: male: 5.0–16.0; female: 10.0–24.0
T3: male: 10.7–15.4y; female: 10.0–14.4 y ND T3: male:5.0–25.0; female: 7.0–60.0
T4: male: 11.8–16.2 y; female: 10.7–15.6 y T4: male:10.0–36.0; female: 21.0–85.0
T5: male: 12.8–17.3 y; female: 11.8–18.6 y T5: male:10.0–36.0; female: 34.0–170.0
LabCorp44 1–10 y ND Male: 0–20.0; female: 6.0–27.0  Electrochemiluminescence immunoassay analyzer method (Roche)
Quest Diagnostics45 1–9 y Male: <4; female: <16  LC-MS/MS
10–11y Male: <12; female: <65
12–14 y Male: <24; female: <142
15–17 y Male: <31; female: <283
Total testosterone, ng/dL
Wong et al.,2 2020; CALIPER database12 Male: 4 d-<6 mo; female: 4 d-<9 y Male: 72; female: 125 Male: 8.6–299; female: 1.15–62.0 Architect Í2000SR (Abbott)
Male: 6 mo-<9 y; female: 9-<13 y Male: 61; female: 102 Male: <35.7; female: <28.2
Male: 9-<11 y; female: 13-<15 y Male: 45; female: 79 Male: <23.3; female: 10.4–44.4
Male: 11-<14y; female: 15-<19 y Male: 89; female: 110 Male: <444; female: 14.1–45.0
Male: 14-<16y Male: 78 Male: 36.0–632
Male: 16-<19y Male: 94 Male: 148–794
T1 Male: 20; female: 21 Male: <17.9; female: <19.3
T2 Male: 32; female: 21 Male: <24.5; female: <19.9
T3 Male: 22; female: 26 Male: <543; female: <41.8
T4 Male: 52; female: 62 Male: 8.65–636; female: 8.93–41.5
T5 Male: 17; female: 34 Male: 99.7–760; female: 3.75–49.6
Bohnetal,1 2019; CALIPER database12 0-<6 mo Male: 47; female: 49 Male: 5.8–548; female: <2.9–346 Cobas 8000 e602 (Roche)
6 mo-<11 y Male: 244; female: 244 <2.9
Male: 11-<15 y; female: 11-<19 y Male: 76; female: 150 Male: <2.9–563; female: <2.9–52
Male: 15-<19y Male: 75 Male: 49–769
Soldinet al.,15 2011; CALIPER database12 0-<1 y Male: 18; female: 29 Male: 2–10; female: 2–20 Cobas 6000 (Roche)
>1-<5 y Male: 97; female: 97 2.0–11.4
>5-<10 y Male: 96; female: 96 2.0–22.2
Male: >10-<15 y; female: >10-<20 y Male: 49; female: 80 Male: 2–830; female: 2–78
Male: >15-<20 y Male: 25 Male: 103–1011
Higgins et al.,26 2018; CALIPER database12 0-<6 mo Male: 68; female: 55 Male: 8.05–642; female: 5.54–401 VITROS 5600 (Ortho)
6 mo-<12 y Male: 388; female: 388 Male: <4.90–46.1; female: <4.90–46.1
Male: 12-<15 y; female: 12–19 y Male: 86; female: 153 Male: 10.4–717; female: 15.3–63.7
Male: 15-<19y Male: 83 Male: 117–782
Bogner 201939 14–17y Male: 63; female: 39 Male: 23.63–857.06; female: 12.68–80.98 VITROS EciQ (Ortho)
Karbasy et al.,27 2016; CALIPER database12 0-<1.5 y Male: 46; female: 48 Male: 0–284; female: 0–63.1 DxI and Access2 (Beckman Coulter)
1.5-<7 y Male: 169; female: 169 0–10.1
7–9 y Male: 58; female: 58 0–17.9
9-<12 y Male: 79; female: 79 0–47
Male: 12–19 y; female: 12-<15 y Male: 94; female: 43 Male: 11–566; female: 10.1–65.1
Female: 15-<19 y Female: 59 Female: 17.9–85.9
Wong et al.,2 2020 0-<15 d Male: 18; female: 13 Male: 7.2–183; female: 1.7–13.2 LC-MS/MS
15 d-<1 y Male: 34; female: 28 Male: 1.4–257; female: 0.6–6.3
1-<13 y Male: 138; female: 138 Male: 1.7–32.3; female: 1.7–32.3
13-<16 y Male: 18; female: 17 Male: 28–453; female: 14.4–39.2
16-<19 y Male: 30; female: 31 Male: 27.7–674; female: 3.2–66.6
Soldin et al.,15 2011 Male: 0–6 y; female: 0–5 y Male: 158; female: 127 Male: 4–31; female: 2–10 LC-MS/MS
Male: 7–9 y; female: 6–9 y Male: 148; female: 171 Male: 4–25; female: 5–13
Male: 10–12 y; female: 10–14y Male: 188; female: 190 Male: 5–418; female: 14–50
Male: 13–14y; female: 15–16 y Male: 245; female: 175 Male: 6–647; female: 12–53
Male: 15–16 y; female: 17–18 y Male: 209; female: 123 Male: 42–880; female: 16–50
Male: 17–19 y Male: 123 Male: 121–823
Mayo Clinic46 0–5 mo ND Male: 75–400; female: 20–80  LC-MS/MS
6 mo-9 y Male: <7–20; female: <7–20
10–11y Male: <7–130; female: <7–44
Male: 12–13 y; female: 12–16 y Male: <7–800; female: <7–75
Male: 14 y Male: <7–1200
Male: 15–16 y Male: 100–1200
17–18 y Male: 300–1200; female: 20–75
T1 Male: <7–20; female: <7–20
T2 Male: 8–66; female: <7–47
T3 Male: 26–800; female: 17–75
T4 Male: 85–1200; female: 20–75
T5 Male: 300–950; female: 12–60
LabCorp47 Newborn ND Male: 75.0–400.0; female: 20.0–64.0  LC-MS/MS
1–7 d Male: 20.0–50.0; female:<10.0
20–60d Male: 60.0–400.0
61 d-7 mo Male: <2.5–10.0
Male: 1–10 y; female: 1–9 y <2.5–10.0
>18 y Male: 264.0–916.0; female: 10.0–55.0
T1 Male: <2.5–10.0; female: <2.5–10.0
T2 Male: 18.0–150.0; female: 7.0–28.0
T3 Male: 100.0–320.0; female: 15.0–35.0
T4 Male: 200.0–620.0; female: 13.0–32.0
T5 Male: 350.0–970.0; female: 20.0–38.0
ARUP Laboratories31 Newborn ND Male: 75–400; female: 20–64  LC-MS/MS
1–5 mo Male: 14–363; female: <20
6–24 mo Male: <37; female: <9
2–3 y Male: <15; female: <20
4–5 y Male: <19; female: <30
6–7 y Male: <13; female: <7
8–9 y Male: 2–8; female: 1–11
10–11y Male: 2–165; female: 3–32
12–13 y Male: 3–619; female: 6–50
14–15y Male: 31–733; female: 6–52
16–17 y Male: 158–826; female: 9–58
T1 Male: 2–15; female: 2–17
T2 Male: 3–303; female: 5–40
T3 Male: 10–851; female: 10–63
T4, T5 Male: 162–847; female: 11–62
Quest Diagnostics48 1–10 d ND Male: <187; female: <24  Extraction RIA
11–30 d Not established
1–2 mo Male: 72–344; female:<17
3–4 mo Male: <201; female: <12
5–6 mo Male: <59; female: <13
7–11 mo Male: <16; female: <11
1–5 y Male: <5; female: <8
2–7 y Male: <25; female: <20
8–10 y Male: <42; female: <35
11 y Male: <260; female: <40
12–13 y Male: <420; female: <40
14–17.9 y Male: <1000; female: <40
Open in a new tab

Abbreviations: E2, estradiol; GC-MS/MS; gas chromatography–tandem mass spectrometry; LC-MS/MS, liquid chromatography-tandem mass spectrometry; LOD, limit of detection; ND, not disclosed; RIA, radioimmunoassays; T, Tanner stage; TBS, Tanner breast stage; TV, testicular volume.

SI conversion factors: To convert E2 to picomoles per liter, multiply by 3.671; testosterone to nanomoles per liter, multiply by 0.0347.

The use of Tanner stages instead of age groups may not always be appropriate to characterize E2 and total testosterone levels, especially in prepubertal children. For example, total testosterone concentrations for Tanner stage 1 are often slightly higher than those reported for male individuals who are about 5 years old because of inclusion of late prepubertal boys who may have early gonadarche or adrenarche. If a 5-year-old male individual presented in the clinic with early pubertal changes and total testosterone concentrations of 25 ng/dL (to convert to nanomoles per liter, multiply by 0.0347), which are high for his age and may indicate pathology, his total testosterone values would be flagged as high based on age-group PRIs but would fit into the normal range of a Tanner stage 1 group, causing the clinician to possibly miss the pathology.

PRIs for E2 and total testosterone do not consistently reflect the dynamic changes that occur early in life. Similar to observations for thyrotropin and FT4, the number of age groups for E2 and total testosterone PRIs are highly variable (eTable 1 in the Supplement). For total testosterone, the first age group ranged from 0 to 10 days to 0 to 1.5 years for male and female individuals, while the first age group for E2 ranged from 1 to 7 days to 0 to 19 years. The PRIs identified were generated with 13 different assays. Two PRIs for total testosterone are available for Cobas systems (Roche) using the CALIPER study, and each wasestablished using different age groups and numbers of individuals per age group (Table 2). Most commercial laboratories use mass spectrometry assays and have not adopted published PRIs. While mass spectrometry–based PRIs reflect low physiologic total testosterone concentrations expected in female individuals after birth, PRIs established by some clinical analyzers report female total testosterone concentrations approximately 8 times higher, suggesting that the use of accurate methodologies is central to establishing appropriate PRIs.

Growth Factor Biomarkers: IGF-1

The measurement of serum IGF-1 is often used in the diagnosis of disorders of growth hormone secretion, in the management of disorders that lead to nutritional insufficiency or catabolism, and for monitoring both growth hormone and IGF-1 replacement therapy.49 IGF-1 is used to assess growth hormone secretion and activity because IGF-1 is a major downstream effector of growth hormone.50,51 IGF-1 concentrations are highly dynamic from birth to adulthood, with large dynamic increases that occur with puberty before reaching a plateau around 15 years of age (Figure 2).52

We identified 8 PRIs for IGF-1 (eTable 3 in the Supplement). The PRI patterns differed considerably because of the number of age groups used to generate them (6–19; Figure 2 and eTable 1 in the Supplement). Three references in eTable 3 in the Supplement provided PRI data stratified by Tanner stage, and 6 PRIs provided data for each year of age. All PRIs were substratified by sex. Only 1 commercial laboratory used PRIs described in the literature.

Iron Status Indicators: Ferritin and Hemoglobin

Iron deficiency and iron deficiency anemia can have long-term effects on neurodevelopment; therefore, early diagnosis and treatment are critical.53 The American Academy of Pediatrics recommends universal screening for anemia at age 1 year by determination of hemoglobin concentrations. If hemoglobin concentrations are less than 11.0 g/dL (to convert to grams per liter, multiply by 10) at age 1 year, then further evaluation for iron deficiency anemia is required using tests such as serum ferritin.53 Serum ferritin concentrations decline in the first months and are lowest between age 15 and 24 months, after which they slightly increase.54 Hemoglobin concentrations do not change considerably between 12 and 18 months of age and increase slightly with age thereafter.55,56

For hemoglobin and ferritin, 14 and 19 PRIs were identified (eTable 5 in the Supplement), respectively. The number of individuals and age groups per PRI are reported in supplemental eTable 1 in the Supplement. Eleven hemoglobin and 16 ferritin PRIs were stratified by sex. Two hemoglobin PRIs from CALIPER used the same age groups but different numbers of individuals per group, and 5 ferritin PRIs from CALIPER used different age groups and numbers of individuals per group. The hemoglobin PRIs reported by commercial laboratories do not seem to be derived from PRIs published in the literature. Only 1 commercial laboratory for each analyte used PRIs published in the literature. While PRI patterns for hemoglobin are somewhat consistent, with values increasing with age, the PRI patterns for ferritin are highly inconsistent and do not seem to reflect the normal biological changes described in the literature (Figure 2).

Kidney Function Biomarker: Cystatin C

Cystatin C is a biomarker increasingly used for assessing kidney function. It is formed at a constant rate, freely filtered by the glomerulus, and reabsorbed and catabolized in the proximal tubules. Serum concentrations are highly dependent on glomerular function and increased concentrations indicate glomerular damage, whereas increased urinary concentrations indicate possible tubular injury.57 Cystatin C concentrations remain constant between 28 and 37 weeks of gestation and decline throughout the first 2 years of life in healthy children.58 When used in estimation equations, cystatin C shows a tighter correlation with criterion standard glomerular filtration rate (GFR) measures than creatinine, causing numerous organizations to recommend cystatin C use in routine clinical practice.59 In children, cystatin C–based estimated GFR equations better predict endstage kidney disease, cardiovascular morbidities, hospitalization, and death than creatinine-based equations because cystatin C blood concentrations are less dependent on muscle mass, nutritional status, sex, age, and developmental stage than creatinine, which is also influenced by mild kidney impairment.59–61 In pediatric patients, the creatinine-based Bedside Schwartz equation has predominantly been replaced with the creatinine–cystatin C–based Chronic Kidney Disease in Children Study equation, which has less bias and higher accuracy compared with a criterion standard method.62

Despite the increased use of cystatin C in clinical settings, very few PRIs exist for this biomarker. Six PRIs were identified (eTable 4 in the Supplement) and the number of individuals and age groups per PRI varied (eTable 1 in the Supplement). One study developed continuous PRIs, providing a comprehensive picture of cystatin C changes by age and sex during development.63 One additional PRI reflected the dynamic concentration changes occurring in early life. Four of the 6 PRIs were stratified by sex. Ziegelasch et al.64 reported significant differences between older pediatric male and female individuals and recommended using age- and sex-specific PRIs for cystatin C. Two commercial laboratories reported a single range for cystatin C and provided no information about the ages to which the reference interval applied. Additional studies are needed to establish the clinical utility of and PRIs for cystatin C outside of estimated GFR calculations.

Discussion

PRIs should accurately reflect the underlying biological and biochemical changes that occur over the course of healthy child development. We found many PRIs do not adequately reflect patterns of biomarker changes during child development. Some PRIs appear inappropriate for assessing a child’s health or monitoring treatment efficacy. PRI patterns and the extent and timing of changes should be similar across studies, even if absolute values vary across instruments. For IGF-1 and hemoglobin, PRIs were reasonably similar; however, a large degree of dissimilarity was observed for thyrotropin, ferritin, E2, total testosterone, and cystatin C. Inconsistencies were most notable in developmental stages during which biomarker concentrations change rapidly, as observed in the first year of life for thyroid function tests or during puberty for steroid hormones. These inconsistencies are, in part, attributable to high variability in the number of age groups used to represent certain developmental stages. Several reference intervals were reported from the CALIPER study and used the same study population and statistics but different analytical systems. Despite these commonalities, the number of age groups and PRI patterns are still different. Some age groups cover long periods and miss short periods of time with highly dynamic changes, such as those in the first year in life. One promising solution to overcome the challenges with age grouping is to transition to continuous reference intervals, as demonstrated in reference interval studies in children and adults and in an American Academy of Pediatrics–developed IGF-1 study.65–68

Biomarkers, such as thyrotropin and FT4 or hemoglobin and ferritin, are often used in combination to diagnose certain diseases. Therefore, PRI patterns should be consistent not only for individual biomarkers but also for the corresponding biomarkers within a physiologic pathway that are used to make clinical decisions. This is especially true for the tightly controlled thyrotropin and FT4 axis. Consistent relationships between thyrotropin and FT4, as well as hemoglobin and ferritin, were not reflected in the PRIs reviewed, including PRIs used in clinical laboratories. These findings also indicate that many PRIs do not reflect the underlying biochemistry of development, especially in very young children. When possible, PRI studies should measure all analytes used in the diagnostic workup in the same individual to better define their relationship in healthy children.

It is notable that despite the large number of PRIs published in the literature, very few are used by clinical laboratories. Only 4 of the 91 PubMed-derived and reference book–derived PRIs are used by the 13 commercial and clinical laboratories included in this study. Despite the large number of PRIs, it remains difficult for laboratories to identify the PRI that best reflects their patient population. This is complicated by the fact that some PRIs are based on data generated with analytical systems no longer available and that only limited information is available about the comparability of different PRIs and their transferability to different analytical systems.69,70 Additional well-designed studies from the US pediatric population may aid in determining PRIs for the national pediatric patient population and assessing comparability of PRIs from different regions. At the same time, facilitating the adoption of these PRIs by clinical laboratories could be achieved by using some of the approaches used in laboratory standardization and harmonization efforts.71 Continued communication and collaboration between clinicians and their laboratory colleagues ensures appropriate clinical test interpretation and patient assessment and remains essential to effective implementation of common PRIs.

Limitations

Published laboratory PRIs were obtained and used where possible; however, observations and assessments made about PRI implementation are, in part, anecdotal in nature because they are based on limited survey results. For this review, 9 clinical laboratories measuring pediatric patients samples were surveyed about PRIs used in practice; thus, points raised about how laboratories practice the adoption and application of PRIs may not be representative of all clinical laboratories.

Conclusions

Development of PRIs requires overcoming many technical and logistical challenges and is often hampered by the lack of availability of sufficiently large study populations that are representative of healthy children. Furthermore, the lack of assay standardization results in instrument-specific reference intervals that cannot easily be transferred across laboratories. As a result, researchers and laboratories continue to develop new PRIs in addition to the many PRIs already available. On the other hand, increasingly used biomarkers such as cystatin C still lack PRIs derived from appropriate and well-designed study cohorts. The US Centers for Disease Control and Prevention (CDC) has partnered with the American Association for Clinical Chemistry and other key stakeholders to improve pediatric patient care. The goals of this joint effort include (1) using data acquired from the National Health and Nutrition Examination Survey to generate continuous PRIs, (2) providing these PRIs to laboratories, clinicians, and researchers, and (3) facilitating their adoption by clinical laboratories through assay standardization. The National Health and Nutrition Examination Survey, conducted by CDC’s National Center for Health Statistics, is a well-established, continuous survey that collects nationally representative health data and information from adults and children across the United States’ population through physical examinations, extensive health questionnaires, and laboratory biomarker measurements.72 The CDC, in collaboration with clinical partners, has developed reference intervals in adults that can easily be adopted by clinical laboratories and is pursuing similar approaches through this partnership.

Supplementary Material

Supplemental Material

Footnotes

Concept and design: Lyle, Pokuah, Dietzen, Wong, Pyle-Eilola, Fuqua, Woodworth, Jones, Akinbami, Vesper.

Acquisition, analysis, or interpretation of data: Lyle, Dietzen, Wong, Pyle-Eilola, Fuqua, Woodworth, Jones, Akinbami, Garibaldi, Vesper.

Drafting of the manuscript: Lyle, Pokuah, Dietzen, Wong, Pyle-Eilola, Fuqua, Woodworth, Vesper. Critical revision of the manuscript for important intellectual content: Lyle, Dietzen, Wong, Fuqua, Woodworth, Jones, Akinbami, Garibaldi, Vesper. Statistical analysis: Dietzen, Wong.

Administrative, technical, or material support: Pokuah, Dietzen, Wong, Woodworth, Jones, Vesper.

Supervision: Lyle, Dietzen, Vesper.

Conflict of Interest Disclosures: Dr Dietzen reported personal fees from Danaher Diagnostics, Roche Diagnostics Point of Care, Waters Inc, and Becton Dickinson for serving on the scientific advisory board outside the submitted work. Dr Wong reported royalties from Elsevier and AACC Press. Dr Pyle-Eilola reported personal fees from Becton Dickinson for serving on a scientific advisory board outside the submitted work. No other disclosures were reported.

Disclaimer: The findings and conclusions in this report are those of the author(s) and do not necessarily represent the official position of the US Centers for Disease Control and Prevention/the Agency for Toxic Substances and Disease Registry. Use of trade names is for identification only and does not imply endorsement by the US Centers for Disease Control and Prevention, the Public Health Service, and the US Department of Health and Human Services.

REFERENCES

  • 1.Bohn MK, Higgins V, Asgari S, et al. Paediatric reference intervals for 17 Roche Cobas 8000 e602 immunoassays in the CALIPER cohort of healthy children and adolescents. Clin Chem Lab Med. 2019; 57(12):1968–1979. doi: 10.1515/cclm-2019-0707 [DOI] [PubMed] [Google Scholar]
  • 2.Wong E, Brungnara C, Straseski J, Kellogg M, Khosrow A. Pediatric Reference Intervals. 8th ed: Academic Press; 2020. [Google Scholar]
  • 3.Léger J, Olivieri A, Donaldson M, et al. ; ESPE-PES-SLEP-JSPE-APEG-APPES-ISPAE; Congenital Hypothyroidism Consensus Conference Group. European Society for Paediatric Endocrinology consensus guidelines on screening, diagnosis, and management of congenital hypothyroidism. J Clin Endocrinol Metab. 2014;99 (2):363–384. doi: 10.1210/jc.2013-1891 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Defining, Establishing, and Verifying Reference Intervals in the Clinical Laboratory: CLSI document EP28-A3c. Clinical and Laboratory Standards Institute; 2010. [Google Scholar]
  • 5.Katayev A, Fleming JK, Luo D, Fisher AH, Sharp TM. Reference intervals data mining: no longer a probability paper method. Am J Clin Pathol. 2015; 143(1):134–142. doi: 10.1309/AJCPQPRNIB54WFKJ [DOI] [PubMed] [Google Scholar]
  • 6.Ozarda Y, Sikaris K, Streichert T, Macri J; IFCC Committee on Reference intervals and Decision Limits (C-RIDL). Distinguishing reference intervals and clinical decision limits: a review by the IFCC Committee on Reference Intervals and Decision Limits. Crit Rev Clin Lab Sci. 2018;55(6):420–431. doi: 10.1080/10408363.2018.1482256 [DOI] [PubMed] [Google Scholar]
  • 7.Alnor AB, Vinholt PJ. Paediatric reference intervals are heterogeneous and differ considerably in the classification of healthy paediatric blood samples. Eur J Pediatr. 2019;178(7):963–971. doi: 10.1007/s00431-019-03377-w [DOI] [PubMed] [Google Scholar]
  • 8.Hoq M, Matthews S, Donath S, Carlin J, Ignjatovic V, Monagle P. Paediatric reference intervals: current status, gaps, challenges and future considerations. Clin Biochem Rev. 2020;41 (2):43–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Shaw JL, Binesh Marvasti T, Colantonio D, Adeli K. Pediatric reference intervals: challenges and recent initiatives. Crit Rev Clin Lab Sci. 2013;50(2): 37–50. doi: 10.3109/10408363.2013.786673 [DOI] [PubMed] [Google Scholar]
  • 10.Ichihara K, Ozarda Y, Barth JH, et al. ; Committee on Reference Intervals and Decision Limits, International Federation of Clinical Chemistry and Laboratory Medicine. A global multicenter study on reference values: 1: assessment of methods for derivation and comparison of reference intervals. Clin Chim Acta. 2017;467:70–82. doi: 10.1016/j.cca.2016.09.016 [DOI] [PubMed] [Google Scholar]
  • 11.Ichihara K, Ozarda Y, Barth JH, et al. ; Committee on Reference Intervals and Decision Limits, International Federation of Clinical Chemistry and Laboratory Medicine and Science Committee, Asia-Pacific Federation for Clinical Biochemistry. A global multicenter study on reference values: 2: exploration of sources of variation across the countries. Clin Chim Acta. 2017;467:83–97. doi: 10.1016/j.cca.2016.09.015 [DOI] [PubMed] [Google Scholar]
  • 12.Adeli K Canadian Laboratory Initiative on Pediatric Reference Intervals (CALIPER) database. Accessed August 13, 2021. https://caliper.research.sickkids.ca/#/ [DOI] [PubMed]
  • 13.Soghier LM, Fratantoni K, Reyes C, eds. Reference Range Values for Pediatric Care. 2nd ed. American Academy of Pediatrics; 2019. [Google Scholar]
  • 14.Wu A Tietz Clinical Guide to Laboratory Tests. 4th ed. Saunders; 2006. [Google Scholar]
  • 15.Soldin SJ, Wong EC, Brugnara C, Soldin OP. Pediatric Reference Intervals. 7th ed. American Association for Clinical Chemistry Press; 2011. [Google Scholar]
  • 16.Barth JH, Luvai A, Jassam N, et al. Comparison of method-related reference intervals for thyroid hormones: studies from a prospective reference population and a literature review. Ann Clin Biochem. 2018;55(1):107–112. doi: 10.1177/0004563217691549 [DOI] [PubMed] [Google Scholar]
  • 17.Adeli K, Higgins V, Trajcevski K, Palmert MR. Important considerations for interpreting biochemical tests in children. BMJ. 2018;361:k1950. doi: 10.1136/bmj.k1950 [DOI] [PubMed] [Google Scholar]
  • 18.Beasley CM Jr, Crowe B, Nilsson M, et al. Adaptation of the robust method to large distributions of reference values: program modifications and comparison of alternative computational methods. J Biopharm Stat. 2019;29 (3):516–528. doi: 10.1080/10543406.2019.1579223 [DOI] [PubMed] [Google Scholar]
  • 19.Higgins V, Asgari S, Adeli K. Choosing the best statistical method for reference interval estimation. Clin Biochem. 2019;71:14–16. doi: 10.1016/j.clinbiochem.2019.06.006 [DOI] [PubMed] [Google Scholar]
  • 20.Jones GRD, Haeckel R, Loh TP, et al. ; IFCC Committee on Reference Intervals and Decision Limits. Indirect methods for reference interval determination: review and recommendations. Clin Chem Lab Med. 2018;57(1):20–29. doi: 10.1515/cclm-2018-0073 [DOI] [PubMed] [Google Scholar]
  • 21.Tahmasebi H, Higgins V, Woroch A, Asgari S, Adeli K. Pediatric reference intervals for clinical chemistry assays on Siemens ADVIA XPT/1800 and Dimension EXL in the CALIPER cohort of healthy children and adolescents. Clin Chim Acta. 2019; 490:88–97. doi: 10.1016/j.cca.2018.12.011 [DOI] [PubMed] [Google Scholar]
  • 22.Hanley P, Lord K, Bauer AJ. Thyroid disorders in children and adolescents: a review. JAMA Pediatr. 2016;170(10):1008–1019. doi: 10.1001/jamapediatrics.2016.0486 [DOI] [PubMed] [Google Scholar]
  • 23.Hinton CF, Harris KB, Borgfeld L, et al. Trends in incidence rates of congenital hypothyroidism related to select demographic factors: data from the United States, California, Massachusetts, New York, and Texas. Pediatrics. 2010;125(suppl 2):S37–S47. doi: 10.1542/peds.2009-1975D [DOI] [PubMed] [Google Scholar]
  • 24.Fisher DA, Odell WD. Acute release of thyrotropin in the newborn. J Clin Invest. 1969;48 (9):1670–1677. doi: 10.1172/JCI106132 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Sack J, Fisher DA, Wang CC. Serum thyrotropin, prolactin, and growth hormone levels during the early neonatal period in the human infant. J Pediatr. 1976;89(2):298–300. doi: 10.1016/S0022-3476(76)80472-6 [DOI] [PubMed] [Google Scholar]
  • 26.Higgins V, Fung AWS, Chan MK, Macri J, Adeli K. Pediatric reference intervals for 29 Ortho VITROS 5600 immunoassays using the CALIPER cohort of healthy children and adolescents. Clin Chem Lab Med. 2018;56(2):327–340. doi: 10.1515/cclm-2017-0349 [DOI] [PubMed] [Google Scholar]
  • 27.Karbasy K, Lin DC, Stoianov A, et al. Pediatric reference value distributions and covariate-stratified reference intervals for 29 endocrine and special chemistry biomarkers on the Beckman Coulter Immunoassay Systems: a CALIPER study of healthy community children. Clin Chem Lab Med. 2016;54(4):643–657. doi: 10.1515/cclm-2015-0558 [DOI] [PubMed] [Google Scholar]
  • 28.Lim HH. Korean pediatric reference intervals for FT4, TSH, and TPO Ab and the prevalence of thyroid dysfunction: a population-based study. Clin Biochem. 2017;50(18):1256–1259. doi: 10.1016/j.clinbiochem.2017.09.024 [DOI] [PubMed] [Google Scholar]
  • 29.Gunapalasingham G, Frithioff-Bøjsøe C, Lund MAV, et al. Reference values for fasting serum concentrations of thyroid-stimulating hormone and thyroid hormones in healthy Danish/North-European white children and adolescents. Scand J Clin Lab Invest. 2019;79(1–2):129–135. doi: 10.1080/00365513.2019.1581945 [DOI] [PubMed] [Google Scholar]
  • 30.La’ulu SL, Rasmussen KJ, Straseski JA. Pediatric reference intervals for free thyroxine and free triiodothyronine by equilibrium dialysis-liquid chromatography-tandem mass spectrometry. J Clin Res Pediatr Endocrinol. 2016;8(1):26–31. doi: 10.4274/jcrpe.2152 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.ARUP Laboratories. Thyroxine, free by equilibrium dialysis/HPLC-tandem mass spectrometry. Accessed August 13, 2021. https://ltd.aruplab.com/Tests/Pub/0093244
  • 32.LabCorp. Thyroxine (T4), free, direct. Accessed August 13, 2021. https://www.labcorp.com/tests/001974/thyroxine-t-sub-4-sub-free-direct
  • 33.Mayo Clinic Laboratories. Thyroxine (T4), free, serum. Accessed August 13, 2021. https://endocrinology.testcatalog.org/show/FRT4
  • 34.Quest Diagnostics. T4 free (FT4). Accessed August 13, 2021. https://testdirectory.questdiagnostics.com/test/test-detail/866/t4-free-ft4?cc=MASTER
  • 35.Tanner J Growth at Adolescence. 2nd ed. Blackwell Scientific Publications; 1962. [Google Scholar]
  • 36.Parikh TP, Stolze B, Ozarda Y, et al. Diurnal variation of steroid hormones and their reference intervals using mass spectrometric analysis. Endocr Connect. 2018;7(12):1354–1361. doi: 10.1530/EC-18-0417 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Burger HG. Androgen production in women. Fertil Steril. 2002;77(suppl 4):S3–S5. doi: 10.1016/S0015-0282(02)02985-0 [DOI] [PubMed] [Google Scholar]
  • 38.Renault CH, Aksglaede L, Wøjdemann D, Hansen AB, Jensen RB, Juul A. Minipuberty of human infancy: a window of opportunity to evaluate hypogonadism and differences of sex development? Ann Pediatr Endocrinol Metab. 2020; 25(2):84–91. doi: 10.6065/apem.2040094.047 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Bogner B, Schwenoha K, Vogl M, et al. Evaluation of reference intervals of haematological and biochemical markers in an Austrian adolescent study cohort. Clin Chem Lab Med. 2019;57(6):891–900. doi: 10.1515/cclm-2018-0715 [DOI] [PubMed] [Google Scholar]
  • 40.Bae YJ, Zeidler R, Baber R, et al. Reference intervals of nine steroid hormones over the life-span analyzed by LC-MS/MS: Effect of age, gender, puberty, and oral contraceptives. J Steroid Biochem Mol Biol. 2019;193:105409. doi: 10.1016/j.jsbmb.2019.105409 [DOI] [PubMed] [Google Scholar]
  • 41.Ankarberg-Lindgren C, Dahlgren J, Andersson MX. High-sensitivity quantification of serum androstenedione, testosterone, dihydrotestosterone, estrone and estradiol by gas chromatography-tandem mass spectrometry with sex- and puberty-specific reference intervals. J Steroid Biochem Mol Biol. 2018;183:116–124. doi: 10.1016/j.jsbmb.2018.06.005 [DOI] [PubMed] [Google Scholar]
  • 42.Mayo Clinic Laboratories. Estradiol, serum. Accessed August 13, 2021. https://www.mayocliniclabs.com/test-catalog/overview/81816#Clinical-and-Interpretive
  • 43.LabCorp. Estradiol, sensitive, LC/MS. Accessed August 13, 2021. https://www.labcorp.com/tests/140244/estradiol-sensitive-lc-ms
  • 44.LabCorp. Estradiol. Accessed August 13, 2021. https://www.labcorp.com/tests/004515/estradiol
  • 45.Quest Diagnostics. Estradiol, ultrasensitive, LC/MS. Accessed August 13, 2021. https://testdirectory.questdiagnostics.com/test/test-detail/30289/estradiol-ultrasensitive-lcms?q=estradiol&cc=MASTER
  • 46.Mayo Clinic Laboratories. Testosterone, total, mass spectrometry, serum. Accessed August 13, 2021. https://pediatric.testcatalog.org/show/TTST
  • 47.LabCorp. Testosterone, total, women, children, and hypogonadal males, LC/MS-MS. Accessed August 13, 2021. https://www.labcorp.com/tests/070001/testosterone-total-women-children-and-hypogonadal-males-lc-ms-ms
  • 48.Quest Diagnostics. Testosterone, total, MS. Accessed August 13, 2021. https://testdirectory.questdiagnostics.com/test/test-detail/15983/testosterone-total-ms?q=Testosterone%20Total,%20MS&cc=MASTER
  • 49.Banerjee I, Clayton PE. Clinical utility of insulin-like growth factor-I (IGF-I) and IGF binding protein-3 measurements in paediatric practice. Pediatr Endocrinol Rev. 2006;3(4):393–402. [PubMed] [Google Scholar]
  • 50.Chinoy A, Murray PG. Diagnosis of growth hormone deficiency in the paediatric and transitional age. Best Pract Res Clin Endocrinol Metab. 2016;30(6):737–747. doi: 10.1016/j.beem.2016.11.002 [DOI] [PubMed] [Google Scholar]
  • 51.Frystyk J, Freda P, Clemmons DR. The current status of IGF-I assays: a 2009 update. Growth Horm IGF Res. 2010;20(1):8–18. doi: 10.1016/j.ghir.2009.09.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Clemmons DR. Consensus statement on the standardization and evaluation of growth hormone and insulin-like growth factor assays. Clin Chem. 2011;57(4):555–559. doi: 10.1373/clinchem.2010.150631 [DOI] [PubMed] [Google Scholar]
  • 53.Baker RD, Greer FR; Committee on Nutrition American Academy of Pediatrics. Diagnosis and prevention of iron deficiency and iron-deficiency anemia in infants and young children (0–3 years of age). Pediatrics. 2010;126(5):1040–1050. doi: 10.1542/peds.2010-2576 [DOI] [PubMed] [Google Scholar]
  • 54.Oatley H, Borkhoff CM, Chen S, et al. ; TARGet Kids! Collaboration. Screening for iron deficiency in early childhood using serum ferritin in the primary care setting. Pediatrics. 2018;142(6):e20182095. doi: 10.1542/peds.2018-2095 [DOI] [PubMed] [Google Scholar]
  • 55.Sherriff A, Emond A, Hawkins N, Golding J; ALSPAC Children in Focus Study Team. Haemoglobin and ferritin concentrations in children aged 12 and 18 months. Arch Dis Child. 1999;80(2): 153–157. doi: 10.1136/adc.80.2.153 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56.Perng W, Mora-Plazas M, Marin C, Villamor E. Iron status and linear growth: a prospective study in school-age children. Eur J Clin Nutr. 2013;67(6): 646–651. doi: 10.1038/ejcn.2013.56 [DOI] [PubMed] [Google Scholar]
  • 57.Herget-Rosenthal S, van Wijk JA, Bröcker-Preuss M, Bökenkamp A. Increased urinary cystatin C reflects structural and functional renal tubular impairment independent of glomerular filtration rate. Clin Biochem. 2007;40(13–14):946–951. doi: 10.1016/j.clinbiochem.2007.04.013 [DOI] [PubMed] [Google Scholar]
  • 58.Kandasamy Y, Rudd D. Cystatin C: a more reliable biomarker of renal function in young infants? a longitudinal cohort study. Acta Paediatr. 2021;110(4):1341–1345. doi: 10.1111/apa.15538 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 59.Grubb A Cystatin C is indispensable for evaluation of kidney disease. EJIFCC. 2017;28(4): 268–276. [PMC free article] [PubMed] [Google Scholar]
  • 60.Baxmann AC, Ahmed MS, Marques NC, et al. Influence of muscle mass and physical activity on serum and urinary creatinine and serum cystatin C. Clin J Am Soc Nephrol. 2008;3(2):348–354. doi: 10.2215/CJN.02870707 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 61.Shlipak MG, Mattes MD, Peralta CA. Update on cystatin C: incorporation into clinical practice. Am J Kidney Dis. 2013;62(3):595–603. doi: 10.1053/j.ajkd.2013.03.027 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 62.Salvador CL, Tøndel C, Rowe AD, et al. Estimating glomerular filtration rate in children: evaluation of creatinine- and cystatin C-based equations. Pediatr Nephrol. 2019;34(2):301–311. doi: 10.1007/s00467-018-4067-3 [DOI] [PubMed] [Google Scholar]
  • 63.Piepsz A, Tondeur M, Ham H. Revisiting normal (51)Cr-ethylenediaminetetraacetic acid clearance values in children. Eur J Nucl Med Mol Imaging. 2006;33(12):1477–1482. doi: 10.1007/s00259-006-0179-2 [DOI] [PubMed] [Google Scholar]
  • 64.Ziegelasch N, Vogel M, Müller E, et al. Cystatin C serum levels in healthy children are related to age, gender, and pubertal stage. Pediatr Nephrol. 2019; 34(3):449–457. doi: 10.1007/s00467-018-4087-z [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 65.Asgari S, Higgins V, McCudden C, Adeli K. Continuous reference intervals for 38 biochemical markers in healthy children and adolescents: comparisons to traditionally partitioned reference intervals. Clin Biochem. 2019;73:82–89. doi: 10.1016/j.clinbiochem.2019.08.010 [DOI] [PubMed] [Google Scholar]
  • 66.Hoq M, Matthews S, Karlaftis V, et al. Validation of the HAPPI kids continuous age-specific pediatric reference intervals. J Appl Lab Med. 2020;5(6): 1337–1344. doi: 10.1093/jalm/jfaa045 [DOI] [PubMed] [Google Scholar]
  • 67.Li K, Hu L, Peng Y, et al. Comparison of four algorithms on establishing continuous reference intervals for pediatric analytes with age-dependent trend. BMC Med Res Methodol. 2020;20(1):136. doi: 10.1186/s12874-020-01021-y [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 68.Mokhtar KM. TSH continuous reference intervals by indirect methods: a comparisons to partitioned reference intervals. Clin Biochem. 2020; 85:53–56. doi: 10.1016/j.clinbiochem.2020.08.003 [DOI] [PubMed] [Google Scholar]
  • 69.Southcott EK, Kerrigan JL, Potter JM, et al. Establishment of pediatric reference intervals on a large cohort of healthy children. Clin Chim Acta. 2010;411(19–20):1421–1427. doi: 10.1016/j.cca.2010.06.018 [DOI] [PubMed] [Google Scholar]
  • 70.Tahmasebi H, Trajcevski K, Higgins V, Adeli K. Influence of ethnicity on population reference values for biochemical markers. Crit Rev Clin Lab Sci. 2018;55(5):359–375. doi: 10.1080/10408363.2018.1476455 [DOI] [PubMed] [Google Scholar]
  • 71.Travison TG, Vesper HW, Orwoll E, et al. Harmonized reference ranges for circulating testosterone levels in men of four cohort studies in the United States and Europe. J Clin Endocrinol Metab. 2017;102(4):1161–1173. doi: 10.1210/jc.2016-2935 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 72.Centers for Disease Control and Prevention. National Health and Nutrition Examination Survey. Accessed October 6, 2021. https://www.cdc.gov/nchs/nhanes/index.htm

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental Material
NIHMS1847211-supplement-Supplemental_Material.pdf (450.9KB, pdf)

RESOURCES