A meta-analysis of leptin reference ranges in the healthy paediatric prepubertal population

Abstract

Objective

The initial discovery of leptin (1994) has given rise to a substantial number of published studies. This study aimed at identifying the published data on the reference ranges of total, free and bound leptin concentration in the healthy prepubertal population.

Methods

A search was conducted on original English language studies published from 1994 to 2005 in the following databases: PubMed (n = 58), EMBASE (n = 4), Biological Abstracts (n = 2) and Science Finder Scholar (n = 66). A cited reference search was completed in Science Citation Index on studies with a leptin range. A meta-analysis was completed on included studies containing a dataset and a sample size for a leptin concentration range and/or mean±standard deviation for a healthy prepubertal population. Preanalytical and analytical variations were examined. Preanalytical variables included aspects such as fasting state and gender, while analytical variation comprised the type of leptin assay methodology.

Results

Twelve studies met the inclusion criteria. One study examined free leptin; 11 studies examined total concentration. No studies reported leptin reference ranges established by Clinical and Laboratory Standards Institute (CLSI) criteria, although four studies reported specific study leptin ranges. The methodology of enzyme-linked immunosorbent assay demonstrated a wider leptin range than radio immunoassay (0.56–36.35 vs. 1.01–12.21 ng/mL). Males had a significantly lower mean leptin concentration than females (P = 0.0006); obese children had a higher concentration than non-obese (P = 0.0001).

Conclusion

No studies have established CLSI-based leptin reference ranges in prepubertal healthy children and there is a wide variation in the published leptin concentrations. These differences suggest that caution should be used in the interpretation and comparison between studies.

Introduction

Leptin was first discovered as the ob gene protein product in 1994.¹ It is primarily an adipocyte-derived hormone and could be an important biomarker in childhood obesity. Obesity is prevalent in both developed and developing countries and global obesity rates have significantly risen in children over the past two to three decades.^2–5 The identification and application of useful biomarkers, such as leptin, may allow for the early identification of children at risk of obesity and may be useful in monitoring those on weight management programmes.

Leptin circulates in free and bound forms in humans.⁶ Free leptin is an active protein that can bind to circulating proteins and tissue-bound receptors. The major binding protein in circulating human blood is the soluble leptin receptor.⁷ The soluble leptin receptor could be a significant modulator determining the amount of total leptin in circulation by altering the amount of: (1) free leptin that is degraded; (2) free leptin that is bound and (3) free leptin that is able to bind other proteins and tissue-bound receptors.⁸

A major factor influencing leptin concentrations in humans is adipose tissue mass.⁹ Human leptin mRNA and proteins are regulated by changes in body fat and food intake.^9,10 More specifically, leptin concentrations have a positive correlation with body mass index (BMI) and fat mass, as well as a negative correlation with fat-free mass.^11–13 Gender differences in child and adolescent leptin concentrations have been established because fat mass is directly related to circulating leptin concentrations and fat-free mass is inversely related to circulating leptin concentrations.^11,14 Specifically, females have greater amounts of subcutaneous fat and males tend to have a higher percentage of fat-free mass, therefore females should have higher leptin concentrations than males.^15,16 However, it is important to highlight, that other biological factors affect leptin concentrations in children and adolescents, including oestradiol and testosterone concentrations.¹⁷ Subsequently, gender differences in leptin concentration are not merely due to body composition.

There are two immunoassay methods primarily used to analyse circulating total leptin concentrations: (1) radioimmunoassay (RIA) or (2) enzyme-linked immunosorbent assay (ELISA). These assays have been vital in supporting the discovery of the roles of leptin and how it is regulated in both the healthy or diseased states. Although both RIA and ELISA methods have been used to quantify total circulating leptin concentrations, these methods can differ with respect to antibody specificity for detecting leptin. RIA was the first type of immunoassay¹⁸ and early research relied on polyclonal antibodies that are less specific than the more recently developed monoclonal antibodies. Current RIA and ELISA kits, depending on the manufacturer, are based on either monoclonal or polyclonal antibodies. Polyclonal antibodies detect a variety of epitopes; this recognition can lead to unwanted cross-reactivity. Antibody specificity could, thereby, account for some concentration variations when measuring leptin and makes it difficult to compare results across studies. Furthermore, the lack of an internationally accepted standard reference material makes comparisons of study results an additional challenge. These limitations, and other preanalytical and analytical sources of variation, may not always be acknowledged when reporting leptin concentration values or reference ranges.

A reference range, or reference interval, is defined by the Clinical and Laboratory Standards Institute (CLSI) as the range between and including the upper and lower reference limit that is estimated to include 95% of the values for the population determined by the reference participants.¹⁹ The measured laboratory test result of an individual is compared with a reference interval for the purpose of making a medical diagnosis, therapeutic decisions or other physiological assessments.¹⁹

The objective of this meta-analysis is to determine the leptin reference ranges and mean concentration values that have been established and published for healthy, prepubertal pediatric population. In addition, it is also verified if potential sources of variation are taken into consideration when establishing the reference values.

Methods

Study design

This meta-analysis was conducted on scientific literature published from January 1994 to May 2005. Databases searched included: PubMed, EMBASE, Biological Abstracts and Science Finder Scholar. In addition, a cited reference search was completed in Science Citation Index on the resulting studies from the databases searched that provided a leptin concentration range in the healthy, pediatric prepubertal population.

Inclusion/exclusion criteria

The search strategy was limited to published English Language studies focusing on prepubertal and healthy children. Each study had to include a data set accompanied by a sample size for the given leptin concentration range and/or a given mean and standard deviation (SD). This information was required for the statistical analysis. In addition, each study had to include basic details of the analytical method used for determining leptin concentration.

Search yield

In total, 134 studies met the English language, prepubertal, healthy child criteria: PubMed = 58; EMBASE = 4; Biological Abstracts = 2; Science Finder Scholar = 66; Science Citation Index = 4. Only four studies gave a sample size and a leptin concentration range. Two of these studies, in addition to eight others, provided a data set with a mean and SD. This resulted in a total of 12 studies that met all inclusion criteria for the meta-analysis.

Analysis

Intercooled Stata 8.2 for Windows was used for all data analysis. The meta-analysis statistical approach followed the principles outlined by Winer.²⁰ Data from all included studies were analysed together and, wherever applicable they were also analysed by: (1) participant in the fasted or non-fasted state (FS or NFS, respectively); (2) males and females (M or F, respectively); (3) obese or non-obese and (4) RIA or ELISA method. When comparing males vs. females and obese vs. non-obese, pooled variances were calculated to determine the P values and any differences between-groups within each study. These P values were pooled and Fisher chi-squared tests were performed across studies. To determine reference ranges, the lowest concentration and highest concentration in the included studies were obtained. The leptin concentration distribution in most studies was unknown; however, two studies specifically noted a skewed distribution.^21,22 The direction of the skewed distribution was not noted, but the distribution become more normal following a log transformation. The weighted mean reference range was calculated for studies that used the same assay methodology. All results are represented as ng/mL. A P value <0.05 was considered significant.

Results

The 12 studies that met the inclusion criteria are summarized in Table 1. The studies are characterized by total sample size, gender sample size, participant ages (range or mean [SD], as available) and pubertal status determination. Zero studies identified genuine leptin reference ranges, according to CLSI recommendations. Four studies gave the complete range of leptin concentrations found in the healthy children (Table 2); two of these studies and an additional eight studies, therefore a total of 10 studies, provided a mean and SD (Table 3). Of the 12 studies, 11 examined total leptin concentration, while one study examined free leptin concentration.²¹ RIA (n = 10) was the more commonly used assay methodology relative to ELISA (n = 2).

Table 1

Basic characteristics of all included studies and leptin inclusion criteria met

	Sample size all		Age (years) all		Determination of prepubertal status	Inclusion criteria*
Study	M	F	M	F		Range	Mean (SD)
Arslanian et al.²⁶	22		10.5 (0.94)†		Physical exam for Tanner staging;⁵³ plasma oestradiol (F) testosterone (M) concentration		Х
	13	9	–
Argente et al.²⁵	Lean: 20		Tanner 1‡		Physical exam for Tanner staging⁵⁴		Х
	10	10	–	–
	Obese: 14		8.7 (1.7)‡
	–	–	–	–
Buyukgebiz et al.²⁸	51		4.2 (2.1)‡		–	Х	Х
	20	31	–	–
Bjarnason et al.³²	107		9.7 (−)‡		Tanner and Whitehouse⁵⁵ for breast/pubic hair; Zachman et al.⁵⁶ for testicular volume	Х
	–	–	–	–
Charmandari et al.²⁷	28		8.9 (1.59)‡		Physical exam for Tanner staging^57,58		Х
	16	12	–	–
Coutant et al.²²	Lean: 40		9.1 (1.90)‡		Serum oestradiol (F) or testosterone (M) concentration		Х
	24	16
	Obese: 42		9.6 (1.94)‡
	25	17
Garcia-Mayor et al.²⁹	515		–		Serum oestradiol (F) or testosterone (M) concentration		Х
	327	188	8.4 (2.53)‡	8.9 (1.65)‡
Horlick et al.²⁴*	28		–		Physical exam for Tanner staging;⁵⁴ oestradiol, testosterone, gonadotropins (LH, FSH) concentration	Х	Х
	13	15	8.7 (1.80)‡	8.5 (1.16)‡
Kavazarakis et al.³¹	294		9.3 (1.66)‡		Physical exam for Tanner staging^{(not referenced)}		Х
	154	140	–	–
L'Allemand et al.²¹	Lean: 26		2.7–11.4^§		Pubertal stage 1⁵⁹ (stage for breast, pubic hair, genitalia)	Х
	9	17	–	–
	Obese: 26		2.9–13.1^§
	10	16	–	–
Moore et al.²³	472		8 (0.68)‡		–		Х
	251	221	–	–
Roemmich et al.³⁰	18		–		Physical exam for Tanner staging⁵⁴		Х
	10	8	10.9 (1.3)‡	10.0 (1.4)‡

– Denotes not provided in study

*Study reported leptin range and/or mean (SD) to meet inclusion criteria

‡Mean (SD)

‡Actual age not reported for group

^§Age range

Table 2

Combined male and female leptin concentration ranges for healthy prepubertal children in included studies

Study	Sample size	Lower concentration (ng/mL)	Upper concentration (ng/mL)	Assay type
L'Allemand et al.²¹*	26	0.19‡	6.27‡	RIA⁶⁰
Buyukgebiz et al.²⁸	51	0.4‡	12.7‡	RIA (DSL)
Bjarnason et al.³²	107	1.5‡	13.3‡	RIA (Linco)
Horlick et al.²⁴	28	0.56‡	36.35‡	ELISA (Amgen)

*Free leptin concentration

‡Reported study mean

‡Weighted mean

Table 3

Combined male and female mean (SD) leptin concentration, fasting state and assay method for healthy prepubertal children in included studies

Study	Sample size	Mean (SD) (ng/mL)	Fasting state	Assay type
Argente et al. ²⁵	20	5.2 (3.96)	FS	RIA (Linco)
Arslanian et al. ²⁶	22	9.3 (9.38)	FS	RIA (Linco)
Buyukgebiz et al. ²⁸	51	4.9 (3.1)	FS	RIA (DSL)
Charmandari et al. ²⁷	28	2.5 (3.17)	FS	RIA (Esoterix Endocrinology)
Coutant et al. ²²	40	3.8 (2.53)	FS	RIA (BioMerieux)
Garcia-Mayor et al. ²⁹	515	4.78 (3.93)	NFS	RIA (Linco)
Horlick et al. ²⁴	28	6.3 (11.02)	FS	ELISA
Kavazarakis et al. ³¹	294	11.99 (11.79)	FS	ELISA
Moore et al. ²³	472	2.08 (0.71)	FS	RIA (Linco)
Roemmich et al. ³⁰	18	7.61 (7.12)	FS	RIA (Linco)

The nature of the leptin assay methods were not well described in the included papers. Little information beyond the type of assay and the manufacturer of the assay was provided. All studies, except one,²³ provided at least one the following assay characteristics: interassay variation, intra-assay variation, detection range or sensitivity.

The complete range of leptin concentrations for healthy, prepubertal children was specified in four studies (Table 2). The three RIA studies had a narrower study leptin range (1.01–12.21 ng/mL), although L'Allemande et al.²¹ specifically examined free leptin. There was a trend in the four studies demonstrating that ELISA had a higher upper value and a wider leptin concentration range compared with the RIA methodology (Tables 2 and 4).

Table 4

Leptin concentration ranges and mean (SD) by fasting state, gender and assay methodology for healthy prepubertal children

Number of studies	Studies	Sample size	Fasting state	Gender	Assay	Mean (SD) (ng/mL)	Leptin range (ng/mL)
3	L'Allemand et al.,²¹ Buyukgebiz et al.,²⁸ Bjarnason et al.³²	184	FS and NFS	M & F	RIA	–	1.01–12.21*
1	Buyukgebiz et al.²⁸	51	FS	M and F	RIA	–	0.4–12.7
1	Horlick et al.²⁴	28	FS	M and F	ELISA	0.56–36.35
2	L'Allemand et al.,²¹ Bjarnason et al.³²	133	NFS	M & F	RIA	–	1.24–11.93*
7	Arslanian et al.,²⁶ Argente et al.,²⁵ Buyukgebiz et al.,²⁸ Charmandari et al.,²⁷ Coutant et al.,²² Moore et al.,²³ Roemmich et al.³⁰	651	FS	M and F	RIA	2.92 (2.57)*	–
2	Horlick et al.²⁴* Kavazarakis et al.³¹	322	FS	M and F	ELISA	11.50 (11.73)*	–
1	Horlick et al.²⁴*	13	FS	M	ELISA	9.1 (15.14)	–
5	Arslanian et al.,²⁶ Argente et al.,²⁵ Coutant et al.,²² Moore et al.,²³ Roemmich et al.³⁰	308	FS	M	RIA	2.55 (2.62)*	–
1	Horlick et al.²⁴*	15	FS	F	ELISA	3.9 (5.4)	–
5	Arslanian et al.,²⁶ Argente et al.,²⁵ Coutant et al.,²² Moore et al.,²³ Roemmich et al.³⁰	264	FS	F	RIA	3.01 (2.05)*	–
1	Garcia-Mayor et al.²⁹	515	NFS	M and F	RIA	4.78 (3.93)	–
1	Garcia-Mayor et al.²⁹	327	NFS	M	RIA	4.38 (3.98)	–
1	Garcia-Mayor et al.²⁹	188	NFS	F	RIA	5.47 (3.84)	–

*Weighted mean

Of the 10 studies, 9 provided a mean and SD value for leptin concentrations had the participants in a fasted state. The duration of the fast was not necessarily specified, but most stated an overnight fast between 9 h and 12 h. Analytically, RIA was the method of choice for detecting leptin, whereas only one ELISA study monitored participants in a fasted state and this study provided the widest range for leptin concentrations (0.56–36.35 ng/mL).²⁴

The mean and SD provided in 10 studies were analysed by gender, obesity state and/or fasting state (Table 4). There was a total of five RIA studies that provided mean (SD) for males and females. A significant difference was found between male and female children (χ ²(10) = 30.7592, P = 0.0006), with males having lower concentrations than females. There was only one ELISA study that evaluated gender differences in leptin concentration, thus a meta-analyses was not possible.

The relationship between obesity and mean (SD) leptin concentrations was also examined. There were two RIA studies that provided data for a comparison between obese (n = 56, mean leptin = 23.95 [SD 10.12]) and non-obese (n = 60, mean leptin = 4.27 [SD = 3.07]) participants. Argente et al.²⁵ defined obese children as having a BMI-Standard Deviation Score (BMI-SDS) that was greater than 2.0 (male and female BMI-SDS in the study was 4.34 [0.48] and 4.09 [0.46], respectively) and Coutant et al.²² defined obesity as body weight being >120% of the ideal value for age and height (percentage of ideal body weight in the study was 151.6 [12.96]). The obese children had significantly higher leptin concentrations than non-obese children (χ²(4) = 23.0300, P = 0.0001). There were no ELISA studies that compared obese with non-obese populations.

Utilizing either the RIA or ELISA method of detection influenced the concentration of leptin in the non-fasted or fasted state (Table 4). There was a significant difference in mean (SD) leptin concentration between the ELISA (n = 2) and RIA (n = 7) methods (11.50 [11.73] and 2.92 [2.57], respectively).when participants were in a fasted state (P < 0.0001). There were no ELISA studies in the non-fasted state to allow for a similar comparison. Additional analysis showed a significant difference in mean (SD) in RIA studies when comparing the non-fasted (4.78 [3.93], n = 1) and fasted (2.92 [2.57], n = 7) states. Specifically, the non-fasted state gave a higher mean and a wider SD (P < 0.0001).

Studies reviewed in this analysis monitored both serum and plasma leptin concentration. There were three studies using plasma,^23,24,26 eight studies using serum^{21,25,27–32} and one study that used both.²² The plasma concentrations of hormones were generally higher than serum concentrations, as expected, partly due to the removal of fibrinogen in the serum.³³

Discussion

Preanalytical, analytical and postanalytical sources of variation can significantly affect the measurement of leptin concentrations in the blood. This meta-analysis focused on the preanalytical sources affecting the child (gender, puberty [prepubertal only inclusion], weight, specimen source [serum vs. plasma] and diet) and analytical sources involving the method of analysis. All these sources of variation made it difficult to accurately determine a prepubertal reference range in healthy children. Results primarily showed that: (1) No leptin reference ranges have been identified for healthy, prepubertal children according to the CLSI criteria; (2) ELISA methodology reported higher mean total leptin concentrations and greater SD than RIA; (3) females had higher mean leptin concentrations; (4) obese children had higher mean leptin concentrations and (5) participants in a fasted state gave a lower mean leptin concentration and less variation than those in the non-fasted state.

Sample collection timing is an important preanalytical variable to acknowledge.³⁴ Leptin has been reported to undergo circadian and ultradian cycling.^35–37 Serum leptin concentrations are highest between midnight and early morning. In addition, leptin can be influenced by factors, such as meal timing, appetite suppression while sleeping and relative total body fat.^38,39 The majority of the studies included in this meta-analysis involved blood collection in the early to late morning (approximately 06:00 h to 11:00 h).^22,24–31 However, two studies^21,23 did not specify the time of blood collection and one study³² gave a time frame spanning between 10:00 h and 16:00 h. This variation in specimen collection timing could significantly impact the leptin concentrations obtained in the study.

Fasting state and food intake are two other important preanalytical variables to acknowledge.³⁴ It is recognized that during a fasted state, glycogen stores become the primary energy source for the body through the process of glycogenolysis. Fasting longer than 12–15 h will result in the depletion of glycogen stores and a consequent increase in gluconeogenesis.⁴⁰ This meta-analysis found lower leptin concentrations in the fasting vs. non-fasting state. This is further supported in the study by Wagner et al.⁴¹ whereby a steady decrease in leptin concentrations were detected from 08:00 h to 08:00 h and over 24 h or 72 h of starvation. Wagner et al. ⁴¹ also found that food intake caused fluctuations in leptin concentrations that have been proposed to be mediated through insulin, as shown by the observed leptin and insulin temporal (6 h) coupling. Insulin and leptin have been extensively examined for their role in energy balance and food intake signalling. Insulin specifically increases leptin mRNA expression and increases leptin secretion by adipocytes.^42,43 In a study that examined isolated rat adipocytes, leptin was observed to specifically impair a number of insulin metabolic actions.⁴⁴ For instance: (1) leptin induced a concentration-dependent desensitization of the glucose transport system for activation by insulin; (2) leptin inhibited insulin from activating glycogen synthase in a concentration-dependent manner and (3) leptin inhibited insulin-stimulated lipogenesis in a concentration-dependent manner.⁴⁴ The interaction of leptin and insulin is necessary to identify, as fluctuations in insulin concentrations could lead to fluctuations in leptin concentrations. It is important for leptin to be measured at the same time for all study participants and, when relevant, across all study days. Moreover, three-day or 24-hour recall diet records should be utilized to assess diet intake.

In addition to sample collection concerns, there are a number of other preanalytical criteria that can affect proteins in blood sample. These are important to acknowledge when measuring leptin concentration. Puberty can be a significant confounder when analysing hormonal concentrations.¹¹ This meta-analysis focused on prepubertal children to circumvent the gender differences in leptin concentration that arise with and following puberty. It aimed at determining a specific reference range only for the prepubertal child population. Other factors, such as posture during the venipuncture, could also affect leptin concentration. The change from a supine to a sitting position can shift body water from the intravascular to the interstitial compartment and this leads to a higher concentration of the large molecules that are not filterable.³⁴ There is, hence, a decrease in the concentration of proteins in the supine position due to gravity effects pulling water away from the upper extremities into the lower body. Only one study specifically noted the patient posture and venipuncture technique utilizing a 30 min resting supine position prior to venipuncture.²⁷

Nagy and Gower⁴⁵ stated almost 10 yr ago that caution should be taken when comparing absolute leptin concentrations across studies and this is still true today. The acknowledgment of a decrease in analytical sources of variation will lead to greater accuracy in the standardization of leptin measurement. An internationally accepted leptin standard reference material does not currently exist and until this is achieved it will be difficult to make meaningful comparisons or draw meaningful conclusions between different populations and different methods of analysis.

The studies that met the inclusion criteria for this meta-analysis primarily examined only total leptin concentration in prepubertal children. Although this meta-analysis found a significant difference in leptin concentrations between obese and non-obese children, free leptin concentration was not analysed. Studies analysing total, free and bound leptin concentrations across populations or over time may find more significant shifts in free and bound leptin concentrations, vs. total leptin concentration. For example, Magni et al.⁴⁶ identified a significantly higher free leptin concentration in obese vs. lean subjects, and Reinehr et al.⁴⁷ reported a significant increase in bound leptin concentration of obese children that had substantial weight loss (BMI-SDS >0.5) following a 1-yr intervention programme. A free vs. bound leptin concentration shift could be clinically significant, as bound hormones are pharmacologically or physiologically inactive; only free hormones can act on target sites to elicit biological responses.⁴⁸ The physiological importance of a greater prevalence of free leptin concentration with obesity, however, remains unclear.⁴⁶ Future studies must not only focus on the standardization of total leptin concentration, but also free leptin concentration reference ranges.

Only two ELISA studies met the criteria for this analysis. Accordingly, general inferences can be drawn when comparing the RIA and ELISA methods; definitive conclusions are difficult. The ELISA methodology resulted in wider study leptin concentration ranges, higher means and greater deviations. This is in contrast to the study by Carlson et al.⁴⁹ that focused on healthy lean (31 [14.1] yr, BMI = 24 [2.8], n = 8), healthy obese (36 [15.8] yr, BMI = 31 [3.2], n = 10) and Prader-Willi syndrome (age = 12 [6] yr, BMI = 28 [4], n = 4) participants. Plasma leptin concentrations for all participants were not significantly different when comparing the RIA and ELISA methodologies (25.9 [15.9] ng/mL and 24.7 [16.9] ng/ml, respectively) and the two methodologies were highly correlated (r = 0.957, P < 0.0001). This study by Carlson et al.⁴⁹ had a small sample size, which could play a significant role when comparing the ELISA vs. RIA results relative to the current meta-analysis.⁴⁹ A larger sample size, as found in this meta-analysis, enables for narrower confidence intervals and provides a more powerful clinical applicability of the results. In addition, Carlson et al.⁴⁹ did not clarify if the EDTA-anticoagulant tubes were tested for any cross-reactivity with leptin or if the EDTA interfered with the specificity of the antibody. Although the present meta-analysis only had two ELISA studies, the overall participant sample size (n = 322) is large and should not be discounted.

The rise in childhood obesity rates is a concern for the physical, psychological and social wellbeing of overweight and obese children. Comorbidities in children are previously well-described⁵⁰ and can include metabolic, hepatic, psychosocial and cardiovascular conditions. In 1996, it was identified in a group of obese and non-obese children and young adults that leptin concentrations: (1) increase early in obesity development, (2) are elevated in obese children and (3) have a strong positive correlation with the subcutaneous fat depot in children and adults.⁵¹ This interaction between leptin and obesity was assessed in the studies that were included in this study. The significantly greater leptin concentrations in obese compared with non-obese populations in this meta-analysis may be important, although this is based solely on the two RIA studies that met the inclusion criteria. Leptin concentrations, as previously stated, have been found to have a positive correlation with BMI and fat mass and a negative correlation to fat-free mass.^11–13 Subcutaneous fat synthesizes more serum leptin than other fat depots.^15,52 Consequently, females are going to have higher leptin concentrations than males, as found in this meta-analysis. Since leptin concentration is influenced by gender and fat mass, it is important to control for these preanalytical sources of variation when conducting, comparing and interpreting data and when formulating conclusions with studies analysing leptin concentrations.

Appropriate pediatric reference ranges are imperative for accurate clinical diagnoses. The laboratory measured test result for a child is commonly compared with the reference range in order to subsequently determine their medical diagnosis or therapeutic assessments, as previously stated.¹⁹ According to the CLSI, reference ranges should be determined by a priori sampling with clear inclusion/exclusion criteria for the reference individuals.¹⁹ This will generate reference data on clearly defined characteristics, including: age, gender, health status, specimen collection and laboratory analysis.¹⁹ The development of this type of reference range is ideal and researchers should strive for it. This meta-analysis, however, was unable to identify any leptin reference ranges for the healthy, pediatric population that met these criteria. The included studies only reported the leptin concentration range for their study population, which comprised the lowest and highest concentration detected. The establishment of leptin reference ranges remains to be completed for healthy, prepubertal children. These ranges will need to be developed by gender and include further stratification by fat mass.

Conclusion

There are many published articles on leptin action in animal models and human adults, but only limited articles based on studies in children. Caution should be used when citing published leptin ranges and means within the prepubertal paediatric population. An assumption that there is a known healthy range for leptin cannot be made. This is partially due to the absence of an internationally recognized standard reference material. In addition, the preanalytical and analytical sources of variation within the published studies make it difficult to accurately interpret the values. Therefore, there is a need to establish total and free leptin concentration reference ranges in this population, while controlling for the known sources of variation of leptin concentrations. Specifically, total and free leptin concentrations should be determined in both healthy prepubertal males and females: (1) with either the RIA or ELISA methodology, (2) under a fasted state, (3) with a consistent venipuncture time and posture, (4) while controlling for participant's adiposity and (5) identifying specimen source. These results would subsequently be able to stand as a primary reference for future studies in the burgeoning area of leptin research, in both the healthy and diseased states.

References

Zhang

, Proenca

, Maffei

, Barone

, Leopold

, Friedman

. Positional cloning of the mouse obese gene and its human homologue. Nature 1994;372:425–32

World Health Organization. Obesity: preventing and managing the global epidemic. Report of a WHO Consultation on Obesity. Geneva: World Health Organization, 2000

Ogden

, Flegal

, Carroll

, Johnson

. Prevalence and trends in overweight among us children and adolescents, 1999–2000. JAMA 2002;288:1728–32

Magarey

, Daniels

, Boulton

. Prevalence of overweight and obesity in Australian children and adolescents: reassessment of 1985 and 1995 data against new standard international definitions. Med J Aust 2001;174:561–4

Tremblay

, Katzmarzyk

, Willms

. Temporal trends in overweight and obesity in Canada, 1981–1996. Int J Obes Relat Metab Disord 2002;26:538–43

Sinha

, Opentanova

, Ohannesian

, Evidence of free and bound leptin in human circulation. Studies in lean and obese subjects and during short-term fasting. J Clin Invest 1996;98:1277–82

Lammert

, Kiess

, Bottner

, Glasow

, Kratzsch

. Soluble leptin receptor represents the main leptin binding activity in human blood. Biochem Biophys Res Commun 2001;283:982–8

Huang

, Wang

, Li

. Modulation of circulating leptin levels by its soluble receptor. J Biol Chem 2001;276:6343–9

Maffei

, Halaas

, Ravussin

, Leptin levels in human and rodent: Measurement of plasma leptin and ob RNA in obese and weight-reduced subjects. Nat Med 1995;1:1155–61

10.

Considine

, Sinha

, Heiman

, Serum immunoreactive-leptin concentrations in normal-weight and obese humans. N Engl J Med 1996;334:292–5

11.

Ahmed

, Ong

, Morrell

, Longitudinal study of leptin concentrations during puberty: sex differences and relationship to changes in body composition. J Clin Endocrinol Metab 1999;84:899–905

12.

Brandao

, Lombardi

, Nishida

, Hauache

, Vieira

. Serum leptin concentration during puberty in healthy nonobese adolescents. Braz J Med Biol Res 2003;36:1293–6

13.

Ellis

, Nicolson

. Leptin levels and body fatness in children: effects of gender, ethnicity, and sexual development. Pediatr Res 1997;42:484–8

14.

Nagy

, Gower

, Trowbridge

, Dezenberg

, Shewchuk

, Goran

. Effects of gender, ethnicity, body composition, and fat distribution on serum leptin concentrations in children. J Clin Endocrinol Metab 1997;82:2148–52

15.

Montague

, Prins

, Sanders

, Digby

, O'Rahilly

. Depot- and sex-specific differences in human leptin mRNA expression: implications for the control of regional fat distribution. Diabetes 1997;46:342–7

16.

Masuzaki

, Ogawa

, Isse

, Human obese gene expression. Adipocyte-specific expression and regional differences in the adipose tissue. Diabetes 1995;44:855–8

17.

Ankarberg-Lindgren

, Dahlgren

, Carlsson

, Leptin levels show diurnal variation throughout puberty in healthy children, and follow a gender-specific pattern. Eur J Endocrinol 2001;145:43–51

18.

Yalow

, Berson

. Assay of plasma insulin in human subjects by immunological methods. Nature 1959;184:1648–9

19.

NCCLS. How to define and determine reference intervals in the clinical laboratory; approved guideline-second edition. Wayne, PA: NCCLS document C28-A2: 2000

20.

Winer

, Brown

, Michels

. Statistical Principles in Experimental Design. 3rd edn. New York: McGraw-Hill, 1991

21.

L'Allemand

, Schmidt

, Rousson

, Brabant

, Gasser

, Gruters

. Associations between body mass, leptin, IGF-I and circulating adrenal androgens in children with obesity and premature adrenarche. Eur J Endocrinol 2002;146:537–43

22.

Coutant

, Lahlou

, Bouvattier

, Bougneres

. Circulating leptin level and growth hormone response to stimulation tests in obese and normal children. Eur J Endocrinol 1998;139:591–7

23.

Moore

, Goldblatt

, Bates

, Prentice

. Impact of nutritional status on antibody responses to different vaccines in undernourished Gambian children. Acta Paediatr 2003;92:170–6

24.

Horlick

, Rosenbaum

, Nicolson

, Effect of puberty on the relationship between circulating leptin and body composition. J Clin Endocrinol Metab 2000;85:2509–18

25.

Argente

, Barrios

, Chowen

, Sinha

, Considine

. Leptin plasma levels in healthy Spanish children and adolescents, children with obesity, and adolescents with anorexia nervosa and bulimia nervosa. J Pediatr 1997;131:833–8

26.

Arslanian

, Suprasongsin

, Kalhan

, Drash

, Brna

, Janosky

. Plasma leptin in children: relationship to puberty, gender, body composition, insulin sensitivity, and energy expenditure. Metabolism 1998;47:309–12

27.

Charmandari

, Weise

, Bornstein

, Children with classic congenital adrenal hyperplasia have elevated serum leptin concentrations and insulin resistance: Potential clinical implications. J Clin Endocrinol Metab 2002;87:2114–20

28.

Buyukgebiz

, Ozturk

, Yilmaz

, Arslan

. Serum leptin concentrations in children with mild-to-moderate protein-energy malnutrition. Pediatr Int 2003;45:550–4

29.

Garcia-Mayor

, Andrade

, Rios

, Lage

, Dieguez

, Casanueva

. Serum leptin levels in normal children: Relationship to age, gender, body mass index, pituitary-gonadal hormones, and pubertal stage. J Clin Endocrinol Metab 1997;82:2849–55

30.

Roemmich

, Clark

, Mantzoros

, Gurgol

, Weltman

, Rogol

. Relationship of leptin to bone mineralization in children and adolescents. J Clin Endocrinol Metab 2003;88:599–604

31.

Kavazarakis

, Moustaki

, Gourgiotis

, Relation of serum leptin levels to lipid profile in healthy children. Metabolism 2001;50:1091–4

32.

Bjarnason

, Boguszewski

, Dahlgren

, Leptin levels are strongly correlated with those of GH-binding protein in prepubertal children. Eur J Endocrinol 1997;137:68–73

33.

Lundbald

. Considerations for the use of blood plasma and serum for proteomic analysis. Internet J Genome Proteo 2005:1 (2)

34.

Narayanan

. The preanalytic phase. An important component of laboratory medicine. Am J Clin Pathol 2000;113:429–52

35.

Licinio

, Negrao

, Mantzoros

, Sex differences in circulating human leptin pulse amplitude: clinical implications. J Clin Endocrinol Metab 1998;83:4140–7

36.

Saad

, Riad-Gabriel

, Khan

, Diurnal and ultradian rhythmicity of plasma leptin: effects of gender and adiposity. J Clin Endocrinol Metab 1998;83:453–9

37.

Sinha

, Sturis

, Ohannesian

, Ultradian oscillations of leptin secretion in humans. Biochem Biophys Res Commun 1996;228:733–8

38.

Schoeller

, Cella

, Sinha

, Caro

. Entrainment of the diurnal rhythm of plasma leptin to meal timing. J Clin Invest 1997;100:1882–7

39.

Sinha

, Ohannesian

, Heiman

, Nocturnal rise of leptin in lean, obese, and non-insulin-dependent diabetes mellitus subjects. J Clin Invest 1996;97:1344–7

40.

Adam-Perrot

, Clifton

, Brouns

. Low-carbohydrate diets: nutritional and physiological aspects. Obes Rev 2006;7:49–58

41.

Wagner

, Oberste-Berghaus

, Herpertz

, Time relationship between circadian variation of serum levels of leptin, insulin and cortisol in healthy subjects. Horm Res 2000;54:174–80

42.

Wabitsch

, Jensen

, Blum

, Insulin and cortisol promote leptin production in cultured human fat cells. Diabetes 1996;45:1435–8

43.

Russell

, Petersen

, Rao

, Leptin expression in adipose tissue from obese humans: depot-specific regulation by insulin and dexamethasone. Am J Physiol 1998;275:E507–15

44.

Muller

, Ertl

, Gerl

, Preibisch

. Leptin impairs metabolic actions of insulin in isolated rat adipocytes. J Biol Chem 1997;272:10585–93

45.

Nagy

, Gower

. Correlates of leptin concentration in the San Antonio heart study. Ann Epidemiol 1997;7:79–80

46.

Magni

, Liuzzi

, Ruscica

, Free and bound plasma leptin in normal weight and obese men and women: relationship with body composition, resting energy expenditure, insulin-sensitivity, lipid profile and macronutrient preference. Clin Endocrinol (Oxf) 2005;62:189–96

47.

Reinehr

, Kratzsch

, Kiess

, Andler

. Circulating soluble leptin receptor, leptin, and insulin resistance before and after weight loss in obese children. Int J Obes (Lond) 2005;29:1230–5

48.

Mycek

, Harvey

, Champe

. Pharmacology. In: Harvey

, Champe

, eds. Lippincott's Illustrated Reviews. 2nd edn. Philadelphia: Lippincott-Raven Publishers, 1997: p. 475

49.

Carlson

, Snead

, Oeser

, Butler

. Plasma leptin concentrations in lean and obese human subjects and Prader-Willi syndrome: comparison of RIA and ELISA methods. J Lab Clin Med 1999;133:75–80

50.

Daniels

, Arnett

, Eckel

, Overweight in children and adolescents: pathophysiology, consequences, prevention, and treatment. Circulation 2005;111:1999–2012

51.

Caprio

, Tamborlane

, Silver

, Hyperleptinemia: an early sign of juvenile obesity. Relations to body fat depots and insulin concentrations. Am J Physiol 1996;271:E626–30

52.

Masuzaki

, Ogawa

, Isse

, Human obese gene expression. Adipocyte-specific expression and regional differences in the adipose tissue. Diabetes 1995;44:855–8

53.

Tanner

. Growth and maturation during adolescence. Nutr Rev 1981;39:43–55

54.

Tanner

. Growth and Adolescence. 2nd edn. Oxford: Blackwell Scientific Publications, 1962

55.

Tanner

, Whitehouse

. Clinical longitudinal standards for height, weight, height velocity, weight velocity, and stages of puberty. Arch Dis Child 1976;51:170–9

56.

Zachmann

, Prader

, Kind

, Hafliger

, Budliger

. Testicular volume during adolescence. Cross-sectional and longitudinal studies. Helv Paediatr Acta 1974;29:61–72

57.

Marshall

, Tanner

. Variations in pattern of pubertal changes in girls. Arch Dis Child 1969;44:291–303

58.

Marshall

, Tanner

. Variations in the pattern of pubertal changes in boys. Arch Dis Child 1970;45:13–23

59.

Prader

, Largo

, Molinari

, Issler

. Physical growth of Swiss children from birth to 20 years of age. First Zurich longitudinal study of growth and development. Helv Paediatr Acta Suppl 1989;52:1–125

60.

Horn

, Geldszus

, Potter

, von zur Muhlen

, Brabant

. Radioimmunoassay for the detection of leptin in human serum. Exp Clin Endocrinol Diabetes 1996;104:454–8