Sage Journals: Discover world-class research

Abstract

Background

Malnutrition in children includes both undernutrition and overnutrition, each contributing to distinct metabolic disruptions with long-term health consequences. Understanding these biochemical patterns can improve how malnutrition is detected and managed.

Objective

This scoping review aimed to explore and summarize current research on the metabolites associated with various forms of childhood malnutrition.

Methods

A broad literature search was conducted using PubMed, Web of Science, EBSCO, and Google Scholar to identify studies that analysed metabolic profiles in children experiencing different types of malnutrition, including obesity, stunting, and underweight.

Results

The findings revealed consistent changes in pathways related to amino acids and lipids. Obesity in children was associated with elevated levels of branched-chain amino acids (BCAA) and acylcarnitines, showing emerging associations with insulin resistance. Stunted growth was linked to lower levels of essential amino acids and polyunsaturated fatty acids. Several studies also reported reduced concentrations of specific lipid metabolites, particularly lysophosphatidylcholines, across various malnutrition conditions. Some metabolic markers, such as BCAAs and leptin, were found to correlate with growth and nutritional status. However, results varied across different populations.

Conclusion

The findings suggest that metabolomics has potential as a complementary tool for understanding malnutrition pathophysiology. However, additional research is necessary to validate these metabolites and explore their applicability in clinical practice. The observed variations in metabolic profiles across different populations highlight the importance of context-specific interpretation when applying these metabolites in clinical or research settings.

Keywords

childhood malnutrition metabolomics branched-chain amino acids lipid metabolism metabolic profiles

Introduction

Malnutrition defined as the deficiency, excess, or imbalance of energy and nutrient intake remains a persistent global health challenge, particularly among children in low- and middle-income countries.¹ In early childhood, malnutrition manifests through a spectrum of conditions including stunting, wasting, underweight, micronutrient deficiencies, overweight, and obesity, all of which compromise normal growth and development. Globally, at least one in three children under 5 years of age is affected by one or more forms of malnutrition, with the highest burden reported in Asia.² According to the Global Burden of Disease study, an estimated 435 million children were malnourished in 2019, with the majority residing in Asia and Africa.³ South Asia alone accounts for over half of the world’s children suffering from wasting, while India and Pakistan report approximately 10.7 million stunted children which is nearly half of the global burden.² Additionally, the global prevalence of childhood overweight reached 38.3 million in 2019, with 5.4 million cases in South Asia and 4.8 million in East Asia, comprising over a quarter of the global total.

The long-term effects of childhood malnutrition extend well into adulthood, influencing not only physical health but also cognitive development and socioeconomic outcomes.^4,5 Undernutrition compromises immune function, increasing vulnerability to infections and impairing linear growth, while overnutrition predisposes children to obesity-related complications such as insulin resistance, type 2 diabetes, hypertension, and cardiovascular diseases.^6,7 Malnutrition is also a leading contributor to global child mortality, accounting for approximately 50% of deaths among children under five.⁷ Furthermore, malnourished children are at a heightened risk of developing chronic illnesses and experiencing diminished quality of life and productivity in adulthood. Effective early detection, prevention, and intervention strategies are therefore essential to mitigate the long-term adverse effects of malnutrition.

Traditional biomarkers of malnutrition such as serum albumin, haemoglobin, or anthropometric indices, offer limited mechanistic insights into metabolic dysfunction, often detecting nutritional deficits only after clinical manifestations arise.¹ In contrast, metabolomics provides a dynamic, systems-level perspective by capturing real-time perturbations in metabolic pathways, thereby enabling early risk stratification and personalized interventions.⁸ Recent advances in metabolomics⁹ and multi-omics integration¹⁰ have further enhanced the precision of nutritional phenotyping, revealing subclinical disruptions in amino acid, lipid, and mitochondrial metabolism before overt growth faltering or obesity occurs. For instance, in Ugandan children with severe acute malnutrition, metabolomics identified reduced leptin and elevated acylcarnitines as predictors of mortality,¹¹ outcomes undetectable via conventional measures However, paediatric applications remain nascent, with few studies leveraging these tools to decode the distinct metabolic signatures of undernutrition versus overnutrition in children.⁵ Closing this gap is critical to developing targeted therapies for malnutrition’s diverse presentations.

Despite growing interest, research on metabolomics in paediatric malnutrition remains limited. The current body of evidence lacks a comprehensive catalogue of nutritional biomarkers, constraining our understanding of the complex metabolic pathways implicated in both under- and overnutrition. Additionally, while global efforts have led to a decline in stunting, the rising incidence of childhood overweight and obesity poses new challenges. There is a pressing need for robust, child-focused metabolomics studies to identify early diagnostic markers and therapeutic targets for malnutrition across its diverse forms. This review aims to synthesize emerging evidence on metabolomic signatures associated with malnutrition in children. Specifically, the objectives are threefold: (1) to explore the application of metabolomics in the context of childhood malnutrition; (2) to identify existing research gaps; and (3) to compile metabolomic biomarkers linked to nutritional status and related metabolic processes such as inflammation and oxidative stress. Both targeted and untargeted metabolomics approaches are examined, with emphasis on human studies investigating the biochemical correlates of malnutrition and their potential modification through dietary or clinical interventions.

Methodology

Search strategy

A comprehensive literature search was conducted using PubMed, Web of Science, and EBSCOhost (Medline Complete) to identify studies related to the application of metabolomics in paediatric malnutrition. Google Scholar was also used to retrieve grey literature. The search targeted articles published between January 2010 and March 2025, ensuring the inclusion of up-to-date findings relevant to current research and practice. The search strategy adhered to the PICOT framework, focusing on children aged 2–16 years with malnutrition conditions such as stunting, underweight, overweight, and obesity. Ages 2–16 years were selected to capture critical periods of linear growth (pre-pubertal) and adolescent metabolic shifts. Studies employing targeted or untargeted metabolomics approaches were included. Search queries combined MeSH terms and keywords using Boolean logic (AND/OR). Filters were applied to include only English-language articles. The detailed search strategies for PubMed, Web of Science, and EBSCOhost are outlined in Table 1.

Table 1.

PICOT framework.

Population	Intervention	Comparison	Outcomes	Time
4-16 years old malnourished children with either stunting, underweight, overweight or obesity	Targeted or non- targeted metabolomics approaches in assessing malnutrition	With or without a control group	Metabolomic signatures related with malnutrition (stunting, thinness, underweight, overweight and obesity)	Human studies published between 2010 and December 2024

The following keywords and their variations were used as the search terms in PubMed; species: humans and population age of the preschool child (2–5 years) and child (6 to 12 years): (Metabolomic OR metabolomics OR metabonomic OR metabolome OR metabolomes) AND (child OR children OR paediatrics OR pre-schoolers OR student OR primary school children OR kids OR young children OR older children) AND (Malnutrition OR Overnutrition OR Undernutrition OR Nutrition disorder). The search in Web of Science was filtered to (Languages – English, Document types – Article, Timespan – 2010-2025) and using the following equation: Terms (TS)=(Metabolomic* OR metabonomic OR metabolome*) AND TS=(child OR children OR paediatric* OR pre*schooler* OR student* OR primary school children OR kid* OR young children OR older children) AND TS=(malnutrition OR overnutrition OR undernutrition OR nutrition disorder). Meanwhile, the same filters were used for searching articles in the EBSCOhost database with these keywords: (Metabolomic OR metabolomics OR metabonomic OR metabolome OR metabolomes) AND (child OR children OR paediatrics OR pre-schoolers OR student OR primary school children OR kids OR young children OR older children) AND (Malnutrition OR Overnutrition OR Undernutrition OR Nutrition disorder).

Eligibility criteria and study selection

This review considered prospective and retrospective observational studies, including cross-sectional, longitudinal, cohort, and case-control designs, with no restrictions on sample size. Eligible studies analysed urine or plasma samples using targeted or untargeted metabolomics approaches to assess malnutrition biomarkers. Articles were excluded if they focused on subjects with physical disabilities, chronic diseases, or severe medication history, or if they lacked a metabolomics component. Conference abstracts, reviews, meta-analyses, case reports, and letters to the editor were also excluded. The screening process involved a two-stage approach: title and abstract screening followed by full-text review. Data from eligible studies were extracted systematically using predefined templates, covering study design, population demographics, analytical techniques, and key findings. The reporting procedure was conducted following PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines,¹² as shown in Figure 1.

Figure 1.

Prisma flow diagram.

Results and discussion

Descriptive summary of the selected metabolomics studies investigating malnutrition

Our systematic search identified a total of 586 studies. After preliminary screening of abstracts and removal of duplicate, 366 papers were excluded. Of the remaining 146 articles, only 14 met the criteria for inclusion in this review. Figure 2 provides a summary of the study characteristics and main findings of the 14 selected studies, detailing the cases of malnutrition assessed, biological samples used, and the metabolomics signatures and its derivatives related with different types of malnutrition. All selected studies employed metabolomics methods to identify metabolomic signatures in children with either undernutrition or overnutrition/obesity. The sample sizes of the reviewed studies varied significantly, ranging from 26 subjects¹³ to 803 subjects.¹⁴ This variability highlights the diverse approaches and scales of research within the field, highlighting the growing application of metabolomics in understanding malnutrition.

Figure 2.

Summary of metabolite alterations associated with nutritional status in children. Upward arrows (↑) indicate increased metabolite levels, while downward arrows (↓) indicate decreased levels.

According to the type of biospecimens collected, nine studies used human serum samples,^11,15–22 fours studies used plasma samples^13,14,23,24 and only one study used serum and urinary samples²⁵ for their metabolomics analyses (Refer Table 2). Majority of the studies focused on obesity profiling. However, Varkey et al. (2020)¹⁶ evaluated the optimal dose of protein required for growth, the changes in the targeted measurement of metabolic intermediates and signalling molecules, and shifts in the gut microbiome among stunted children in India. Different analytical technologies were used for metabolomics analysis depending on the studies’ objectives. Technologies used to examine the metabolic phenotype include nuclear magnetic spectroscopy (NMR), liquid chromatography-mass spectrometry (LC-MS), and gas chromatography-mass spectrometry (GC-MS). Among these most studies applied LC-MS to analyse the serum in assessing metabolomic profiling^{13,16–19,22,23,25} although the GC-MS method had also been used to analyse serum in identifying potential biomarkers of childhood obesity.^14,20 This range of technologies reflects the diverse methodologies applied to understand the metabolic alterations associated with malnutrition. Some studies have noted LC-MS offers superior sensitivity and able to profile broad range of low-abundance metabolites, contributing its advantages in dealing with complex biological samples with minimal sample preparation.^26,27 Integration of liquid chromatography allows for effective separation of polar, non-polar, and thermally labile circulating compounds in malnourished children, which GC-MS may not adequately capture due to its reliance on volatile and thermally stable analytes. However, each platform offers distinct advantages and limitations. The appropriate platform selection is primarily based on the specific focus of the study and the nature of biological sample used. Therefore, the best metabolomics studies often utilise multiple technological platforms.

Table 2.

Summary of metabolomics studies assessing malnutrition among children.

Studies	Country	Age/Mean Age of population	Concept of malnutrition/Total subjects recruited	Samples/Platform	The metabolic signature associated with malnutrition status
Zeng et al. (2010)²⁰	China	10-16 years	Overweight/Obesity (n = 65)	SerumGas chromatography-mass spectrometry (GC-MS)	• ↑ Isoleucine, glyceric acid, 2,3,4- trihydroxybutyrate• ↓ Serine, phenylalanine
Wahl et al. (2012)¹⁹	German	6-15 years	Obese (n = 120)	SerumLC/MS/MS system	• ↑ Medium-chain acylcarnitines (C12:1, C16:1)• ↓ Methionine, proline, lysophosphatidylcholines
Perng et al. (2014)²³	United States	6-10 years	Obese (n = 262)	PlasmaMass spectrometry	• ↑ Branched-chained amino acids (BCAAs) -leucine, isoleucine, valine & tryptophan metabolites• ↑ Androgen steroids
Farook et al. (2014)²¹	United States	6-17 years	Obese (n = 42)	SerumUltra-performance liquid chromatography/quadrupole orthogonal acceleration time-of-flight tandem micro mass spectrometer(UPLC/Q-Tof micro-MS)	• ↑ Bradykinin• ↓ L-Thyronine (Thyroid hormone), Naringenin (Flavanoid)
Ghosh et al. (2014)²⁴	India	1-5 years	Stunting and underweight (n = 150)	PlasmaLC-MS/MS	• ↓ Essential amino acids (lysine andtryptophan),• ↓ Carnitine levels
Lee et al. (2015)¹⁸	South Korea	9-11 years	Obesity (n = 109)	SerumLC-MS/MS	• ↑ BCAAs, chain Acylcarnitines (C3, C5)• ↓ Acyl-alkyl Phosphatidylcholine
Butte et al. (2015)¹⁴	United States	4-19 years	Obese (n = 803)	PlasmaNon-targeted metabolomic profiling was conducted by using the following 2 independent platforms:Ultra-HPLC–tandem mass spectrometry andGas chromatography-mass spectrometry.	• ↑ Aromatic amino acids, long-chain Acylcarnitines• ↓ Anti-inflammatory lipids (Lysolipids, Dicarboxylated fatty Acids)
Cho et al. (2017)²⁵	South Korea	N/ASchool-based subjects	Obesity (n = 184)	Urine and serumHigh-performance liquid chromatography (LC)- quadrupole time-of-flight mass spectrometry (MS), LC-MS/MS and flow injection analysis-MS/MS systems	Urinary:• ↓ Benzoate metabolites serum:• ↑ Acylcarnitines• ↑ Biogenic amines - asymmetric dimethylarginine, total dimethylarginine• ↓ Asparaginine, glutamine, ornithine& serine• ↓ Glycerophospholipids & sphingolipids
Hellmuth et al. (2016)¹⁷	German	11.5 ± 2.4 years	Obese (n = 80)	SerumHigh-performance liquid chromatography (LC)- quadrupole time-of-flightMass spectrometry (MS)	• Tyrosine – positively associated with homeostasis model assessment (HOMA) in obese children
Suzuki et al. (2019)¹³	Japan	9-10 years	Obesity (n = 26)	PlasmaLiquid chromatography- mass spectrometry (LC- MS)	• ↑ BCAAs, phenylalanine, tryptophan, methionine, threonine, lysine, Alanine, tyrosine, glutamate, proline, Arginine, ornithine, total Free AminoAcids and aspartate
Bartz et al. (2014)¹¹	Uganda	6 months–5 years	Severe acute malnutrition (n = 77)	SerumMass spectrometry-based	• ↑ Fatty acids and amino acids duringNutritional rehabilitation• ↓ Acylcarnitines with recovery• ↓ Leptin levels at admission strongly associated with mortality
Semba et al. (2016)²²	Malawi	12-59 months	Stunting (n = 313)	SerumLiquid chromatography– tandem mass spectrometry (LC–MS/MS)	• ↓ Essential amino acids (tryptophan, isoleucine, leucine, valine, methionine, threonine, histidine, phenylalanine, lysine)• ↓ Arginine, glycine, glutamin and non- essential amino acids (asparagine, glutamate, serine)• ↓Sphingolipids (SM [OH] C14:1, CM [OH] C16:1, SM [OH] C22:2), SM C16:1, SM C18:0, SM C18:1• ↓ Glycerophospholipid: Lysophosphatidylcholines (lysoPC a C16:0, lysoPC a C17:0, lysoPC a C18:0, lysoPC a C18:1, lysoPC a C18:2, lysoPC a C20:3, lysoPC a C20:4),diacyl-phosphatidylcholines (PC aa C28:1, PC aa C34:4, PC aa C36:6, PC aa C38:6), and acyl-alkyl- phosphatidylcholines (PC ae C34:3, PC ae C36:5, PC ae C38:6)• ↑ Lysophosphatidycholine (lyso PC a C26:0), diacyl-phosphatidylcholines (PC aa C32:0, PC aa C32:1, PC aa C34:1, PC aa C34:2, PC aa C34:3, PC aa C36:1,) and acyl-alkyl- phosphatidylcholines (PC ae C34:0, PC ae C34:1, PC ae C36:1, PC ae C36:2, PC ae C38:1, PC ae C38:2, PC ae C42:1)• Findings suggest insufficient dietary intake of essential amino acids and choline in stunted children
Varkey et al. (2020)¹⁶	India	6-10 years	Stunting (n = 100)	SerumLC-MS/MS	Growth markers:• ↓ Leptin, palmitoylcarnitine Catabolic indicators:• ↑ Serine & ornithine
Chang et al. (2023)¹⁵	China	Children aged 4 to 84 months	Malnourished (n = 275) and healthy children (n = 199)	SerumProton nuclear magnetic resonance (1H-NMR)	• ↑ Acetoacetate, acetone, Ethanol, Succinate, 3-Hydroxybutyrate• ↓ Alanine, methionine, And N-Acetyl- Glycoprotein

Metabolomics profiling alteration in overnutrition

Metabolomics studies in children with overweight and obesity generally report measurable changes in several biochemical pathways involved in energy and nutrient metabolism.^{13,14,17–21,23,25} Disturbances in amino acid and lipid metabolism are among the most frequently described, reflecting their importance in energy regulation and insulin sensitivity. Higher concentrations of branched-chain and aromatic amino acids, along with changes in lipid and acylcarnitine profiles, suggest that energy use and fat oxidation may be less efficient in children with obesity. Although most studies were cross-sectional, a few have examined metabolic changes following improvements in diet or weight status.^28,29 These findings suggest that some metabolite alterations, particularly in amino acid and lipid pathways, may shift toward normal levels with nutritional modification. However, the available evidence remains limited, and more longitudinal work is needed to better understand how changes in nutrition influence metabolic profiles in children.

Branched-chain amino acid

Emerging evidence highlights branched-chain amino acids (BCAAs)—leucine, isoleucine, and valine—as critical players in metabolic dysregulation among children with obesity. As essential amino acids dependent on dietary intake, their impaired catabolism has been strongly associated with insulin resistance and obesity-related complications.³⁰ A consistent pattern emerges from metabolomics studies, with elevated BCAA levels reported across diverse paediatric populations. For instance, Zeng et al. (2010)²⁰ identified increased plasma isoleucine concentrations in overweight children, while Perng et al. (2014)²³ and Lee et al. (2015)¹⁸ demonstrated significant rises in all three BCAAs among obese cohorts (p < 0.001). These findings were further corroborated by Butte et al. (2015)¹⁴ in Hispanic youth and Cho et al. (2017)²⁵ through urinary metabolomic profiling, suggesting BCAA accumulation as a transethnic metabolic signature.

The pathophysiological role of BCAAs extends beyond mere association. Experimental studies reveal that elevated levels activate the mTOR-S6K1 pathway, promoting serine phosphorylation of insulin receptor substrates (IRS-1) and subsequent insulin signalling impairment.³¹ Additionally, mitochondrial dysfunction leads to incomplete BCAA oxidation, accumulating intermediary metabolites that exacerbate metabolic stress.³² This mechanistic framework aligns with clinical observations by Suzuki et al. (2019),¹³ who found strong correlations between BCAA levels and HOMA-IR indices (p < 0.01), particularly in children with severe obesity.

An exception to these consistent findings comes from Wahl et al. (2012),¹⁹ who observed no significant BCAA alterations in their cohort of German children with obesity. Several factors may explain this inconsistency, including differences such as fasting status during sampling, population-specific metabolic characteristics, or the inclusion of participants in early stages of obesity when compensatory mechanisms may still preserve amino acid balance. In addition, regional dietary habits could have influenced the observed metabolomic profiles. Recent studies from European cohorts have demonstrated that habitual dietary patterns, including higher protein and fat intake, can modulate circulating amino acid concentrations and related metabolic biomarkers in children.^33,34 These findings suggest that diet composition may partially account for the absence of BCAA differences observed in the German population. Collectively, these considerations underscore the importance of standardized study designs that incorporate fasting status, dietary assessment, and metabolic stage to improve comparability across pediatric metabolomics studies.

Furthermore, emerging analytical approaches such as isotopic tracer-based flux analyses³⁵ offer the potential to capture dynamic metabolic processes rather than relying solely on static metabolite concentrations, thereby enhancing the interpretation of nutritional and metabolic outcomes in future research. Collectively, these observations position BCAAs as both biomarkers and potential therapeutic targets. Their dual role as nutrients and metabolic regulators highlights the complexity of nutritional interventions in paediatric obesity, necessitating precision-based approaches tailored to individual metabolic phenotypes.

Amino acids

Metabolomics studies have identified consistent patterns of amino acid disturbances beyond BCAAs in childhood obesity across diverse populations. Zeng et al. (2010)²⁰ first identified serine, phenylalanine, and 2,3,4-trihydroxybutyric acid as significant markers in Chinese adolescents, while Wahl et al. (2012)¹⁹ conversely observed reduced glutamine, methionine, and proline in European children, suggesting potential ethnic or dietary influences on metabolic manifestations. Subsequent research by Lee et al. (2015)¹⁸ confirmed elevated aromatic amino acids (phenylalanine, p = 0.0164; tyrosine, p = 0.0002) in Korean youth, a pattern corroborated by Butte et al. (2015)¹⁴ in Hispanic populations for phenylalanine, tyrosine, and sulfur-containing amino acids, aligning with Morris et al.'s (2012) earlier findings.³⁶ More comprehensive profiling by Cho et al. (2017)²⁵ revealed widespread alterations involving 10 amino acids (including arginine, asparagine) and 7 biogenic amines (serotonin, dopamine), while Suzuki et al. (2019)¹³ demonstrated significant correlations between insulin resistance and alanine, methionine, and glutamate levels. These collective findings, supported by the mechanistic frameworks of Tremblay et al. (2001)³⁷ and Calbet et al. (2002),³⁸ establish that childhood obesity fundamentally alters amino acid homeostasis through multiple pathways: (1) impaired insulin-mediated amino acid uptake, (2) altered hepatic metabolism, and (3) modified neurotransmitter synthesis, presenting both diagnostic opportunities and therapeutic targets for metabolic dysfunction.

Acylcarnitine

Emerging evidence from multiple metabolomics studies reveals significant alterations in acylcarnitine profiles among obese children, reflecting impaired mitochondrial fatty acid oxidation. Wahl et al. (2012)¹⁹ first documented elevated levels of monounsaturated acylcarnitines (C12:1 and C16:1), suggesting incomplete lipid metabolism in obese children. Subsequent research by Lee et al. (2015)¹⁸ identified distinct patterns of short-chain acylcarnitine accumulation, with propionylcarnitine (C3) and valerylcarnitine (C5) levels significantly elevated (p < 0.01), potentially linking BCAA catabolism defects to lipid metabolic dysfunction. Butte et al. (2015)¹⁴ further characterized this relationship, demonstrating that short-chain (C12) acylcarnitines correlated strongly with adiposity measures (r = 0.42, p < 0.001), while long-chain (C16) species reflected impaired fat oxidation capacity. The most comprehensive analysis by Cho et al. (2017)²⁵ revealed widespread disturbances across nine acylcarnitine species, establishing their collective association with insulin resistance (HOMA-IR >3.0) and lipid metabolism impairment. These consistent findings across diverse populations indicate that acylcarnitine profiling may serve as both a sensitive biomarker of metabolic dysfunction and a potential indicator of mitochondrial overload in paediatric obesity.

Lipids

Paediatric obesity exhibits distinct perturbations in lipid metabolism that extend beyond simple fat accumulation. Zeng et al. (2010)²⁰ first identified glyceric acid as a clinically relevant biomarker, with plasma concentrations 2.1-fold higher in obese versus lean children (p < 0.01), suggesting altered glycerolipid metabolism. Subsequent work by Farook et al. (2014)²¹ revealed more complex lipidomic patterns, including: (1) 40% reduction in anti-inflammatory lysophosphatidylcholines (LysoPCs), (2) decreased essential fatty acids (linoleic acid, −35%), and (3) elevated membrane phospholipids (phosphatidylethanolamine +28%, phosphocholine +19%). These findings were extended by Cho et al. (2017),²⁵ who demonstrated global disturbances in lipid subspecies, particularly in glycerophospholipid and sphingolipid pathways, which correlated strongly with both systemic inflammation (CRP r = 0.51) and insulin resistance (HOMA-IR r = 0.63). Collectively, these studies establish that childhood obesity fundamentally alters three key lipid metabolic processes including membrane lipid remodelling, essential fatty acid metabolism, and bioactive lipid signalling, creating a metabolic milieu that perpetuates both inflammatory and insulin-resistant states. These lipid signatures not only provide mechanistic insights into obesity pathophysiology but also offer potential diagnostic markers for early metabolic risk stratification in paediatric populations.

Metabolomics alteration in under-nutrition

Metabolomics approaches have significantly advanced our understanding of the biochemical underpinnings of childhood undernutrition, revealing complex metabolic disruptions that often precede clinical manifestations. A study by Varkey et al. (2020)¹⁶ in stunted Indian children demonstrated that nutritional interventions induced significant alterations in 21 fasting serum metabolites, including essential micronutrients like vitamin B12, folate, and ferritin, with particularly strong associations between plasma BCAA levels and improvements in BMI-for-age Z-scores (leucine p = 0.011, isoleucine p = 0.023, valine p = 0.007). This study also identified leptin and palmitoylcarnitine as positive predictors of linear growth, while noting inverse correlations between height gain and metabolites like serine and ornithine (ρ < −0.2, p = 0.08), suggesting these compounds may serve as valuable biomarkers of metabolic repletion and persistent catabolism. The critical role of leptin as an indicator of adipose reserves in malnutrition was further established by Bartz et al. (2014),¹¹ who demonstrated its sensitivity in detecting energy deficit before anthropometric decline becomes apparent.

Complementing these observations, Semba et al. (2016)²² reported extensive metabolic alterations among stunted children in rural Malawi. The study identified substantially lower plasma concentrations of all nine essential amino acids including tryptophan, leucine, isoleucine, valine, methionine, threonine, histidine, phenylalanine, and lysine, as well as reduced levels of arginine, glycine, glutamine and non-essential amino acids (asparagine, glutamate, serine). These reductions suggest limitations in dietary protein quality and possible disturbances in amino acid absorption or utilization, potentially influenced by chronic inflammation and impaired gut function.³⁹ In addition, decreased sphingolipid and glycerophospholipid levels pointed to alterations in lipid and choline metabolism, reflecting possible disruptions in membrane composition and cellular signaling. Overall, the findings indicate that growth faltering in undernourished children may result not only from inadequate energy and protein intake but also from broader metabolic imbalances affecting both amino acid and lipid pathways.

Recent metabolomics investigations have expanded these findings across the spectrum of undernutrition. In severe acute malnutrition, Wen et al. (2022)⁴⁰ identified distinct metabolic signatures of mitochondrial dysfunction and impaired nutrient utilization that differentiated survivors from non-survivors, while Giovanni et al. (2016)⁴¹ revealed persistent metabolic disturbances in children even after clinical stabilization, particularly in those with kwashiorkor. The gut microbiome’s contribution to these metabolic alterations has been highlighted by McMillan et al. (2017),⁴² who established connections between microbial composition changes and disrupted amino acid and lipid metabolism in malnourished children. Similarly, maternal undernutrition studies by Su et al. (2012)⁴³ uncovered complex micronutrient deficiencies and altered one-carbon metabolism associated with adverse pregnancy outcomes.

These collective findings underscore the transformative potential of metabolomics in malnutrition research and clinical practice. The technology enables not only early risk detection through signatures like low leptin or elevated serine but also provides mechanistic insights into intervention targets such as gut-microbiota metabolites and mitochondrial pathways. As demonstrated by Zhao et al. (2024),⁹ such approaches may revolutionize malnutrition management by shifting from reactive treatment to precision-based prevention. However, important questions remain regarding ethnic variability in metabolic profiles, the long-term efficacy of metabolite-guided interventions, and the role of epigenetic-metabolomics interactions in growth faltering. Addressing these gaps through future research will be crucial for developing effective strategies against the global double burden of malnutrition.

Conclusion

This review provides a comprehensive summary of metabolomics studies investigating malnutrition among children, focusing on both undernutrition and overnutrition. Despite the increasing interest in nutritional metabolomics over the past 12 years, studies specifically targeting the paediatric population remain limited. This scarcity has resulted in an incomplete understanding of the metabolomic signatures and metabolic pathways associated with childhood malnutrition. The relatively low number of identified metabolites highlights the challenges in processing large-scale ‘omics’ data and underscores the need for further research to explore the nuanced relationships between metabolite changes, nutrition, and health in children. Nonetheless, the application of advanced metabolomics methods, such as mass spectrometry and nuclear magnetic resonance, has shown promise in elucidating the underlying mechanisms of malnutrition. These techniques enable the simultaneous measurement of a broad spectrum of metabolites, facilitating the identification of biomarkers and pathways involved in disease development. By doing so, metabolomics offers valuable insights into the pathogenesis of malnutrition and aids in predicting phenotypic outcomes in affected children.

Findings from this review underscore that amino acid and lipid metabolisms are the primary pathways affected in both obese and undernourished children. Specifically, branched-chain amino acids (BCAAs), other amino acids, lipids, and acylcarnitines emerge as key biomarkers associated with childhood malnutrition. These biomarkers provide critical insights into the aetiology of overweight and obesity, as well as the metabolic disruptions observed in undernourished children. In conclusion, advancements in metabolomics hold great potential for bridging gaps in understanding the biochemical underpinnings of childhood malnutrition. By decoding associations between dietary exposures and health outcomes, metabolomics can contribute to the development of targeted nutritional interventions and strategies aimed at mitigating the global burden of malnutrition among children. Further research is essential to expand the repertoire of metabolomic biomarkers and validate their applicability across diverse paediatric populations.

Strength and limitation

This review highlights the potential of metabolomics as a powerful approach to uncover the complex metabolic disruptions underlying childhood malnutrition, from undernutrition to obesity. Despite growing research, paediatric-specific studies remain limited, leaving gaps in our understanding of developmental metabolic pathways. Key findings consistently identify branched-chain amino acids, aromatic amino acids, acylcarnitines, and lipid subspecies as central players in malnutrition-related metabolic dysfunction. These metabolites offer potential biomarkers for early detection, subtype differentiation, and personalized interventions. Advanced technologies are paving the way for deeper insights, though challenges in data integration and population-specific validation persist. Moving forward, the field should prioritize multi-omics integration, expanded paediatric reference datasets, and clinical translation of biomarkers. By bridging biochemical signatures with nutritional outcomes, metabolomics can transform malnutrition management—shifting from reactive treatment to precision prevention.

Although most available studies concentrated on obesity, research addressing undernutrition remains limited. This review excluded children under 2 years to preserve biological and environmental comparability across studies. In this younger age group, metabolite variations are mainly shaped by maternal and developmental influences such as intrauterine nutrition, breastfeeding, and early metabolic programming rather than by the child’s independent diet or surroundings. Earlier reviews, including that of Conde-Agudelo et al. (2024),⁴⁴ have already provided valuable insights into neonatal metabolomics patterns and their maternal determinants. In contrast, our scoping review focuses on children aged 2 years and older, capturing a stage when dietary diversity, environmental exposure, and growth-related metabolic adjustments become more pronounced. By synthesizing evidence across this later developmental window, the review adds a complementary perspective to the existing literature and highlights a critical transition period in which nutrition begins to exert a direct influence on metabolic outcomes. Continued investigation across age ranges will be important to clarify how metabolic pathways evolve from early infancy into later childhood. Future research should focus on validating these findings across diverse populations to ensure equitable applications, ultimately enabling targeted strategies to address the global burden of childhood malnutrition in all its forms.

Footnotes

Acknowledgement

All the authors read and approved the final manuscript. The authors thank the Faculty of Health Sciences, Universiti Kebangsaan Malaysia for providing research support.

ORCID iDs

Ika Aida Aprilini Makbul

Razinah Sharif

Bee Koon Poh

Ethical considerations

This article does not contain any studies with human or animal participants conducted by the authors. No ethical approval or informed consent was required.

Author contributions

IAAM: Conceptualization, literature search, writing - original draft. RS: Supervision, writing – review, editing & validation. SML: Supervision, writing - review & editing. NMHH: Review & editing. RMW: Review & editing. BKP: Supervision, writing - review & editing. AM: Supervision, writing - review & editing

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

No new data were created or analysed in this study. Data sharing is not applicable to this article.*

References

World Health Organization . Fact sheets: malnutrition. : WHO, 2020. Available from: https://www.who.int/news-room/fact-sheets/detail/malnutrition

United Nations Children's Fund . The state of the world's children 2017: children in a digital world. UNICEF, 2017.

Mao

Shen

Long

, et al. Epidemiological study of pediatric nutritional deficiencies: an analysis from the global burden of disease study 2019. Nutr J 2024; 23: 14.

Morales

Montserrat-de la Paz

Leon

, et al. Effects of malnutrition on the immune system and infection. Nutrients 2024; 16: 1.

Tharumakunarajah

Lee

Hawcutt

, et al. The impact of malnutrition on the developing lung and long-term lung health. Pulm Ther 2024; 10: 155–170.

Ciężki

Odyjewska

Bossowski

, et al. Not only metabolic complications of childhood obesity. Nutrients 2024; 16: 539.

Kumar

Daniel

Sinha

, et al. Mortality outcome of hospitalized children aged six to fifty-nine months. Int J Community Med Public Health 2020; 8: 372.

Qiu

Cai

Yao

, et al. Small molecule metabolites: discovery of biomarkers and therapeutic targets. Signal Transduct Targeted Ther 2023; 8: 132.

Zhao

Zhang

Chen

, et al. Metabolomic biomarkers for malnutrition in older adults. Clin Nutr 2024; 43: 12–21.

10.

Seyfried

Miras

, et al. Advances in multi-omics approaches to obesity research. Nat Rev Endocrinol 2023; 19: 227–240.

11.

Bartz

Mody

Hornik

, et al. Severe acute malnutrition in childhood. J Clin Endocrinol Metab 2014; 99: 2128–2137.

12.

Moher

Shamseer

Clarke

, et al. Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015 statement. Syst Rev 2015; 4: 1.

13.

Suzuki

Kido

Matsumoto

, et al. Associations among amino acid, lipid, and glucose metabolic profiles in childhood obesity. BMC Pediatr 2019; 19: 273.

14.

Butte

Liu

Zakeri

, et al. Global metabolomic profiling targeting childhood obesity in the Hispanic population. Am J Clin Nutr 2015; 102: 256–267.

15.

Chang

Chen

Liao

, et al. Pediatric obesity metabolomics: a systematic review. Nutr Metab 2023; 20: 45.

16.

Varkey

Devi

Mukhopadhyay

, et al. Metabolome and microbiome alterations related to short-term feeding of a micronutrient-fortified food product to stunted children. Clin Nutr 2020; 39: 3251–3261.

17.

Hellmuth

Kirchberg

Lass

, et al. Tyrosine is associated with insulin resistance in longitudinal metabolomic profiling of Obese children. J Diabetes Res 2016; 2016: 2108909.

18.

Lee

Jang

, et al. Prediction of future risk of insulin resistance and metabolic syndrome based on Korean boy's metabolite profiling. Obes Res Clin Pract 2015; 9: 336–345.

19.

Wahl

Kleber

, et al. Childhood obesity is associated with changes in the serum metabolite profile. Obes Facts 2012; 5: 660–670.

20.

Zeng

Liang

, et al. Plasma metabolic fingerprinting of childhood obesity by GC/MS. J Pharm Biomed Anal 2010; 52: 265–272.

21.

Farook

Reddivari

Chittoor

, et al. Metabolites as novel biomarkers for childhood obesity-related traits in Mexican-American children. Pediatr Obes 2014; 9: 320–327.

22.

Semba

Shardell

Ashour

, et al. Child stunting is associated with low circulating essential amino acids. EBioMedicine 2016; 6: 246–252.

23.

Perng

Gillman

Fleisch

, et al. Metabolomic profiles and childhood obesity. Obesity 2014; 22: 2570–2578.

24.

Ghosh

Sen Gupta

Bhattacharya

, et al. Gut microbiomes of Indian children of varying nutritional status. PLoS One 2014; 9(4): e95547.

25.

Cho

Moon

Kang

, et al. Combined untargeted and targeted metabolomic profiling reveals urinary biomarkers for discriminating obese from normal-weight adolescents. Pediatr Obes 2017; 12: 93–101.

26.

Emwas

Roy

McKay

, et al. NMR spectroscopy for metabolomics research. Metabolites 2019; 9: 123.

27.

Zhang

Sun

Wang

, et al. Modern analytical techniques in metabolomics analysis. Analyst 2016; 137: 293–300.

28.

Sohn

Chae

, et al. Metabolomic signatures for the effects of weight loss interventions on severe obesity in children and adolescents. Metabolites 2021; 12(1): 27.

29.

Guerendiain

Montes

López-Belmonte

, et al. Changes in plasma fatty acid composition are associated with improvements in obesity and related metabolic disorders: a therapeutic approach to overweight adolescents. Clinical Nutrition 2018; 37(1): 149–156.

30.

Lynch

Adams

. Branched-chain amino acids in metabolic signalling and insulin resistance. Nat Rev Endocrinol 2014; 10: 723–736.

31.

Newgard

. Metabolomics and metabolic diseases: where do we stand? Cell Metab 2017; 25: 43–56.

32.

Lackey

Lynch

Olson

, et al. Regulation of adipose branched-chain amino acid catabolism enzyme expression. Am J Physiol Endocrinol Metab 2020; 304: E1175–E1187.

33.

Laamanen

Eloranta

Haapala

, et al. Associations of diet quality and food consumption with serum biomarkers for lipid and amino acid metabolism in Finnish children: the PANIC study. Eur J Nutr 2024; 63(2): 623–637.

34.

Liu

, et al. Daily branched‐chain amino acid intake and risks of obesity and insulin resistance in children: a cross‐sectional study. Obesity 2020; 28(7): 1310–1316.

35.

Bian

Kremer

, et al. Cancer SLC43A2 alters T cell methionine metabolism. Nature 2023; 585: 277–282.

36.

Morris

O'Grada

Ryan

, et al. The relationship between BMI and metabolomic profiles. Proc Nutr Soc 2012; 71: 634–638.

37.

Tremblay

Marette

. Amino acid and insulin signaling via the mTOR/p70 S6 kinase pathway. J Biol Chem 2001; 276: 38052–38060.

38.

Calbet

MacLean

. Plasma glucagon and insulin responses depend on the rate of appearance of amino acids. J Nutr 2002; 132: 2174–2182.

39.

Bhat

Gounder

, et al. Effect of dietary protein and processing on gut microbiota—A systematic review. Nutrients 2022; 14(3): 453.

40.

Wen

Njunge

Bourdon

, et al. Systemic inflammation and metabolic disturbances underlie inpatient mortality among ill children with severe malnutrition. Sci Adv 2022; 8(7): eabj6779.

41.

Giovanni

Di Giorgio

Longo

, et al. Metabolomic profiling in aminoacidopathies: results from a single-center experience. JIMD Rep 2016; 31: 105–115.

42.

McMillan

Rulisa

Sumarah

, et al. Multi-omics analysis of gut microbiome and host metabolism in murine models of malnutrition. Sci Rep 2017; 7: 40574.

43.

Zheng

Zhang

, et al. Integrated profiling of metabolites and trace elements reveals a multifaceted malnutrition in pregnant women from a region with a high prevalence of congenital malformations. Metabolomics 2012; 8: 831–844.

44.

Conde-Agudelo

Villar

Risso

, et al. Metabolomic signatures associated with fetal growth restriction and small for gestational age: a systematic review. Nat Commun 2024; 15(1): 9752.

Metabolomics in malnourished children: A scoping review of emerging metabolites

Abstract

Background

Objective

Methods

Results

Conclusion

Keywords

Introduction

Methodology

Search strategy

Eligibility criteria and study selection

Results and discussion

Descriptive summary of the selected metabolomics studies investigating malnutrition

Metabolomics profiling alteration in overnutrition

Branched-chain amino acid

Amino acids

Acylcarnitine

Lipids

Metabolomics alteration in under-nutrition

Conclusion

Strength and limitation

Footnotes

Acknowledgement

ORCID iDs

Ethical considerations

Author contributions

Funding

Declaration of conflicting interests

Data Availability Statement

References