Elbeltagi RM, Saeed NK, Bediwy AS, Al-Beltagi M. Ultrasound hepatic elastography: A non-invasive indicator of insulin resistance in the pediatric population: A systematic review. World J Clin Pediatr 2026; 15(2): 115963 [DOI: 10.5409/wjcp.v15.i2.115963]
Corresponding Author of This Article
Mohammed Al-Beltagi, MD, PhD, Professor, Senior Researcher, Department of Pediatrics, Faculty of Medicine, Tanta University, 1 Hassan Radwan Street, Tanta 31511, Algharbia, Egypt. mohamed.elbeltagi@med.tanta.edu.eg
Research Domain of This Article
Pediatrics
Article-Type of This Article
Systematic Reviews
Open-Access Policy of This Article
This article is an open-access article which was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/
Jun 9, 2026 (publication date) through May 16, 2026
Times Cited of This Article
Times Cited (0)
Journal Information of This Article
Publication Name
World Journal of Clinical Pediatrics
ISSN
2219-2808
Publisher of This Article
Baishideng Publishing Group Inc, 7041 Koll Center Parkway, Suite 160, Pleasanton, CA 94566, USA
Share the Article
Elbeltagi RM, Saeed NK, Bediwy AS, Al-Beltagi M. Ultrasound hepatic elastography: A non-invasive indicator of insulin resistance in the pediatric population: A systematic review. World J Clin Pediatr 2026; 15(2): 115963 [DOI: 10.5409/wjcp.v15.i2.115963]
Reem M Elbeltagi, Department of Medicine, Royal College of Surgeons in Ireland - Medical University of Bahrain, Busaiteen 15503, Muharraq, Bahrain
Nermin K Saeed, Medical Microbiology Section, Department of Pathology, Salmaniya Medical Complex, Governmental Hospitals, Ministry of Health, Manama 12, Bahrain
Nermin K Saeed, Medical Microbiology Section, Pathology Department, Royal College of Surgeons in Ireland - Medical University of Bahrain, Busaiteen 15503, Muharraq, Bahrain
Adel S Bediwy, Department of Pulmonology, Faculty of Medicine, Tanta University, Tanta 31527, Alghrabia, Egypt
Adel S Bediwy, Department of Pulmonology, University Hospital, Arabian Gulf University, Manama 26671, Bahrain
Mohammed Al-Beltagi, Department of Pediatrics, Faculty of Medicine, Tanta University, Tanta 31511, Algharbia, Egypt
Mohammed Al-Beltagi, Department of Pediatrics, University Hospital, Arabian Gulf University, Manama 26671, Bahrain
Co-first authors: Reem M Elbeltagi and Nermin K Saeed.
Author contributions: Elbeltagi RM was the primary investigator to conceptualized and designed the study, developed the search strategy and methodology, conducted the primary literature search and data extraction, performed the formal statistical analysis, and drafted the initial manuscript; Saeed NK contributed to refining the systematic review methodology and ensuring the rigor of data validation, critically reviewed the manuscript drafts, and provided intellectual input on data quality assessment; Elbeltagi RM and Saeed NK contributed equally to this manuscript as co-first authors; Bediwy AS provided specialized intellectual contributions during the design and interpretation phases, particularly advising on the clinical relevance of the findings and reviewing associated comorbidities, and critically revised the manuscript for important intellectual content; Al-Beltagi M served as the overall principal investigator and supervisor for the project, validated the study design and statistical analysis, provided expert pediatric and clinical interpretation of the final results, and conducted the final critical revision of the manuscript prior to submission. All authors reviewed and approved the final version of the manuscript and agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the study are appropriately investigated and resolved.
AI contribution statement: AI tools (Grammarly) were used solely for linguistic refinement and formatting assistance. No AI tool was involved in the generation of research data, interpretation of results, or formulation of conclusions. All AI-generated outputs were critically reviewed and revised by the authors.
Conflict-of-interest statement: All the authors report no relevant conflicts of interest for this article.
PRISMA 2009 Checklist statement: The authors have read the PRISMA 2009 Checklist, and the manuscript was prepared and revised according to the PRISMA 2009 Checklist.
Corresponding author: Mohammed Al-Beltagi, MD, PhD, Professor, Senior Researcher, Department of Pediatrics, Faculty of Medicine, Tanta University, 1 Hassan Radwan Street, Tanta 31511, Algharbia, Egypt. mohamed.elbeltagi@med.tanta.edu.eg
Received: October 30, 2025 Revised: December 3, 2025 Accepted: February 5, 2026 Published online: June 9, 2026 Processing time: 195 Days and 21.9 Hours
Abstract
BACKGROUND
Insulin resistance (IR) plays a pivotal role in the pathogenesis of metabolic dysfunction-associated steatotic liver disease. While non-invasive imaging methods are increasingly used in pediatrics, the extent to which hepatic elastography reflects IR in children remains unclear.
AIM
To evaluate the association between ultrasound-based hepatic elastography parameters and clinical indices of IR in the pediatric population.
METHODS
A systematic search of PubMed, Scopus, and Web of Science databases was conducted through October 2025. Studies assessing correlations between elastography parameters - controlled attenuation parameter (CAP) and liver stiffness measurement (LSM) - and IR indices were included. Data were pooled using random-effects meta-analysis with correlation coefficients (r) as the primary effect size. Subgroup and sensitivity analyses examined differences by IR index, cohort characteristics, and elastography modality.
RESULTS
Sixteen studies, encompassing 2,032 children and adolescents, were included. The pooled correlation between hepatic elastography (CAP/LSM) and IR indices was r = 0.44 (95% confidence interval: 0.38-0.50; I2 = 72%), indicating a moderate positive association. The strongest correlations were observed for adipose tissue IR (r = 0.65) and the metabolic score for IR (r = 0.49), surpassing simpler indices such as homeostatic model assessment of IR. CAP correlated moderately with early steatosis (r = 0.30-0.40), whereas LSM showed stronger associations with advanced fibrosis and systemic IR (r = 0.50-0.65). Heterogeneity was mainly attributed to differences in disease severity and measurement methods.
CONCLUSION
Ultrasound-based hepatic elastography provides a reliable, non-invasive surrogate for systemic metabolic dysfunction in pediatric metabolic dysfunction-associated steatotic liver disease. CAP reflects early, reversible hepatic fat accumulation, while LSM reflects more advanced fibrosis and systemic IR, and identifies fibrotic progression driven by chronic IR. The strongest associations with adipose tissue-IR and metabolic score-IR highlight the systemic, multisite nature of pediatric IR. Elastography thus holds promise as an integrated biomarker for IR severity, early risk stratification, and therapeutic monitoring in children and adolescents with metabolic risk factors.
Core Tip: This systematic review and meta-analysis highlight that ultrasound-based hepatic elastography, through liver stiffness measurement and controlled attenuation parameter, provides a reliable, non-invasive reflection of insulin resistance (IR) in children and adolescents with metabolic dysfunction-associated steatotic liver disease. Controlled attenuation parameter effectively identifies early hepatic fat accumulation, while liver stiffness measurement detects fibrotic progression linked to chronic IR. The strongest correlations with adipose tissue IR and metabolic score-IR emphasize that elastography captures systemic metabolic dysfunction beyond the liver. These findings support its integration into pediatric screening and monitoring to guide early intervention and prevention of IR and steatotic liver disease.
Citation: Elbeltagi RM, Saeed NK, Bediwy AS, Al-Beltagi M. Ultrasound hepatic elastography: A non-invasive indicator of insulin resistance in the pediatric population: A systematic review. World J Clin Pediatr 2026; 15(2): 115963
Childhood obesity has reached epidemic proportions and is now recognized as a significant, escalating global health concern. One of its most serious hepatic manifestations is metabolic dysfunction-associated steatotic liver disease (MASLD), currently the leading cause of chronic liver disease in children and adolescents worldwide[1]. Pediatric MASLD represents a continuum ranging from simple steatosis to steatohepatitis and fibrosis, ultimately predisposing affected individuals to early-onset cirrhosis and cardiometabolic morbidity. MASLD is intrinsically linked to metabolic syndrome, with insulin resistance (IR) serving as its core pathophysiological driver[2]. The presence of IR disrupts lipid and glucose metabolism, promoting hepatic steatosis, inflammation, and the progression to advanced fibrosis, making its early detection critical for preventing long-term complications[3].
However, the current diagnostic paradigm for both liver injury and IR in children is challenging. Liver biopsy, the gold standard for staging MASLD, is invasive and impractical for screening or serial monitoring. Concurrently, while the hyperinsulinemic-euglycemic clamp is the reference standard for quantifying IR, its complexity limits its use to research settings[4]. Commonly used clinical surrogates, such as homeostatic model assessment of IR (HOMA-IR), are readily available but can be influenced by multiple factors, underscoring the need for more integrated, non-invasive assessment tools[5].
This gap underscores an urgent need for reliable, non-invasive biomarkers that can simultaneously reflect hepatic and metabolic health in children. Ultrasound-based hepatic elastography has revolutionized the non-invasive assessment of liver disease. Techniques including transient elastography (TE) and shear wave elastography (SWE) provide rapid, reproducible, and radiation-free quantification of liver stiffness, a well-validated proxy for fibrosis[6]. Beyond its established role in the staging of liver damage, an emerging body of evidence suggests that liver stiffness may also reflect underlying metabolic dysfunction[7]. This suggests that elastography could serve as a real-time surrogate for IR in high-risk pediatric populations.
Despite this potential, the relationship between hepatic elastography values and validated measures of IR in children remains highly variable across individual studies. A comprehensive synthesis is required to clarify this association and determine the clinical utility of elastography in pediatric metabolic assessment. Therefore, the objective of this systematic review is to critically evaluate and synthesize the available evidence on the association between ultrasound hepatic elastography parameters and validated measures of IR in children and adolescents. By elucidating this relationship, we aim to determine whether elastography can serve not only as a tool for fibrosis staging but also as a non-invasive metabolic biomarker to guide early risk stratification and intervention in pediatric populations.
MATERIALS AND METHODS
This systematic review was conducted and reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines, which provide a standardized framework to ensure transparency and reproducibility in systematic reviews. A detailed protocol was prepared before commencing the review and registered with the International Prospective Register of Systematic Reviews (PROSPERO, Registration No. 1174651).
We included studies that focused on children and adolescents aged 18 years or younger who had obesity, MASLD, previously known as non-alcoholic fatty liver disease (NAFLD), or related metabolic conditions (Table 1). Eligible studies needed to measure liver stiffness using non-invasive ultrasound-based elastography techniques. These techniques included TE, which uses a vibration probe to measure liver stiffness; point SWE, which measures tissue stiffness at a specific location; and two-dimensional SWE, which generates a color-coded map of liver stiffness[8]. For comparison, studies also had to assess IR using validated indices, such as the HOMA-IR, fasting insulin and glucose measurements, or the hyperinsulinemic-euglycemic clamp test, which is considered the gold standard[9]. The primary outcomes of interest were the strength of correlation between elastography-derived liver stiffness and IR, the diagnostic accuracy of elastography (for example, area under the receiver operating curve, sensitivity, and specificity), and predictive results from regression analyses. Studies were excluded if they included only adult patients, if pediatric data were not analyzed separately, or if they were reviews, editorials, case reports, letters, or conference abstracts.
A comprehensive literature search was performed in three major electronic databases: PubMed/MEDLINE, EMBASE, and Scopus. The search period was set from January 2000 to October 2025 to capture the modern era of elastography use. We used both controlled vocabulary (e.g., Medical Subject Headings terms) and free-text keywords related to children, elastography, and IR. All search results were imported into EndNote X9, where duplicate records were removed. Screening was conducted in two stages: First, by reviewing titles and abstracts to assess relevance, and then by examining the full text of potentially eligible articles. Two reviewers performed this process independently to minimize bias, and disagreements were resolved either through discussion or with the input of a third reviewer. The study selection process is summarized using the PRISMA flow diagram (Figure 1).
Figure 1 The flow chart of the included studies.
BMI: Body mass index; IR: Insulin resistance; Adipo-IR: Adipose tissue insulin resistance; CAP: Controlled attenuation parameter; LSM: Liver stiffness measurement; SWV: Shear wave velocity; MASLD: Metabolic dysfunction-associated steatotic liver disease; QUS: Quantitative ultrasound.
Data extraction was also done independently by two reviewers using a standardized template, with a third reviewer verifying the accuracy. Information collected included study details (first author, publication year, and country), study design and sample size, participant characteristics [such as age, sex, and body mass index (BMI)], details of the elastography method used (type of device, probe, and reported liver stiffness values), measures of IR (HOMA-IR, fasting glucose, fasting insulin, or clamp-derived indices), and reported outcomes. These outcomes included correlation coefficients indicating the relationship between liver stiffness and IR, diagnostic accuracy measures such as area under the receiver operating curve, sensitivity, and specificity, as well as any cut-off thresholds used in the studies. If studies reported adjustments for confounding factors such as age, sex, or BMI, these were also noted.
The quality of the included studies was carefully assessed. Study selection followed PRISMA 2020 guidelines, and quality assessment used the Newcastle-Ottawa Scale for non-randomized studies, which evaluates the quality of study design based on how participants were selected, how comparable the study groups were, and how outcomes were measured. For studies that focused on diagnostic accuracy, we used the Quality Assessment of Diagnostic Accuracy Studies-2 tool, which is specifically designed to evaluate the risk of bias and applicability in diagnostic research. Both tools were applied independently by two reviewers, and any discrepancies in scoring were resolved through discussion until a consensus was reached.
For data synthesis, we first organized the findings into a descriptive summary (narrative synthesis), grouping them by elastography technique and by clinical subgroups such as obese vs non-obese children or those with vs without MASLD. Once sufficient studies provided similar data, we conducted a quantitative meta-analysis to statistically combine the results. In particular, we aimed to pool correlation coefficients between liver stiffness and IR indices. Because the included studies were expected to differ in terms of populations, techniques, and outcomes, we used a random-effects model to account for variability across studies. We assessed statistical heterogeneity using the statistic, where values above 50% indicated substantial differences between study results.
We performed sensitivity analyses to test the robustness of the pooled estimates. This involved conducting subgroup analyses by elastography type [TE vs SWE/acoustic radiation force impulse (ARFI)] and IR metric [HOMA-IR vs composite indices such as adipose tissue IR (Adipo-IR)/metabolic score for IR (METS-IR)]. Furthermore, we conducted sensitivity analyses by iteratively removing the highest-risk studies (e.g., those scoring low on the comparability domain of the Newcastle-Ottawa Scale) to confirm that no single study disproportionately influenced the overall pooled estimate. Finally, we evaluated the possibility of publication bias (i.e., the publication of only positive findings) by examining funnel plots and using Egger’s regression test.
RESULTS
Study selection and characteristics
Figure 1 shows the flowchart of the study. Of 54 articles, we included 16 studies (a total of 2051 pediatric participants) that evaluated the associations between hepatic elastography parameters and IR indices and provided sufficient data for the quantitative meta-analysis. These studies were conducted across multiple continents, reflecting a global interest in non-invasive liver assessment and metabolic health in youth. The cohorts were diverse, primarily focusing on children and adolescents with obesity/overweight (n = 10), NAFLD/MASLD (n = 5), or healthy controls (n = 1). Study characteristics are summarized in Table 2. Most were cross-sectional (81%), with mean participant ages ranging from 4 years to 20 years. Sample sizes ranged from 43 to 660 participants (median n = 68). TE was the dominant imaging modality, used in 12 studies to assess liver stiffness via liver stiffness measurement (LSM)[10], steatosis via controlled attenuation parameter (CAP)[11-15], or both[16-21]. Other techniques included SWE[11,22-25]; ARFI[23]; and quantitative ultrasound (QUS) attenuation[11]. The most common measure of IR was the HOMA-IR, used in 11 studies, followed by fasting insulin and other comprehensive metrics such as Adipo-IR and METS-IR.
Table 2 Characteristics of pediatric studies evaluating the correlation between elastography measures and insulin resistance metrics.
Quantitative synthesis: Association between liver stiffness and IR
A quantitative synthesis (meta-analysis) was conducted on 11 studies examining the relationship between liver stiffness [LSM/shear wave velocity (SWV)] and an IR metric. The pooled, weighted correlation demonstrated a moderate, statistically significant positive association between liver stiffness and IR in pediatric populations. The overall pooled effect size was r = 0.44 (95% confidence interval: 0.38-0.50; P < 0.001), confirming that greater liver stiffness is directly linked to higher systemic IR across diverse cohorts. Significant heterogeneity was detected across the included studies (I2 = 72%), indicating that differences in population characteristics, elastography modality, and IR metrics influence the magnitude of the observed effect. The results of the meta-analysis are visually represented in the forest plot (Figures 2 and 3).
Figure 2 Forest plot of study-level correlations between liver stiffness and insulin resistance in the pediatric population.
The forest plot visually represents the individual study correlations (liver stiffness measurement/shear wave velocity vs insulin resistance) and the overall pooled effect, including the subgroup analysis by patient population. Each line represents an individual study included in the meta-analysis (n = 11, total n = 1338). The pooled overall effect [r = 0.44 (0.39-0.49)] is shown as a gray diamond at the bottom, with horizontal lines indicating 95% confidence intervals under a random-effects model. NAFLD: Non-alcoholic fatty liver disease.
Figure 3 Subgroup comparison of pooled correlations between liver stiffness and insulin resistance.
This plot summarizes pooled effect sizes for two major subgroups: Children with obesity or non-alcoholic fatty liver disease [r = 0.53 (0.46-0.60), n = 399] and general/mixed pediatric populations [r = 0.33 (0.26-0.40), n = 939]. The grey diamond and dashed line represent the overall pooled correlation [r = 0.44 (0.39-0.49), n = 1338]. Blue diamonds denote subgroup estimates; horizontal bars indicate 95% confidence intervals under a random-effects model. NAFLD: Non-alcoholic fatty liver disease.
Correlation between liver stiffness (fibrosis) and IR
Eleven studies, involving 1239 participants, assessed the correlation between a measure of liver stiffness (LSM or SWV) and an IR metric. Pearson’s correlation coefficients (r) for liver stiffness ranged from weak to strong (r = 0.25 to r = 0.65). The strongest positive correlations were observed in cohorts with severe metabolic impairment: Heldens et al[19] in 2024 reported an r = 0.65 between LSM and Adipo-IR in adolescents with severe obesity, and Giannouli et al[10] in 2023 found an r = 0.61 between LSM and HOMA-IR in adolescents with polycystic ovary syndrome (PCOS). Moderate-to-strong associations were consistently observed in studies of children with obesity/overweight and confirmed NAFLD/MASLD, such as Karaman et al[24] in 2022 (r = 0.58, SWV and HOMA-IR) and Çakır and Acunaş[23] in 2025 (r = 0.55, ARFI SWV and HOMA-IR). The weakest correlation (r = 0.25) was reported by Rose et al[21] in 2023 between LSM and fasting insulin in a cohort of healthy, non-overweight controls, suggesting a stronger association in populations with pre-existing metabolic dysfunction.
Correlation between liver steatosis (fat) and IR
Five studies, totaling 663 participants, evaluated the correlation between a measure of liver fat (CAP or QUS attenuation) and an IR metric. Correlations in this group were consistently in the moderate range (r = 0.25-0.48). D'Hondt et al[11] in 2021 reported the highest correlation (r = 0.48) between QUS attenuation and HOMA-IR in obese youth, highlighting the utility of QUS. Most studies in this category used CAP, with results ranging from r = 0.30 to 0.39 (e.g., Ferraioli et al[12] in 2017: r = 0.39; Ata et al[17] in 2023: r = 0.32), confirming a robust, moderate relationship between liver fat content and systemic IR in pediatric cohorts.
Influence of study characteristics
The variability in the correlation magnitude is partly attributable to the differences in the patient population and study design. Regarding the population effect, the data suggest that the association between liver elastography and IR is strongest in clinical cohorts with advanced metabolic or hepatic disease (e.g., severe obesity, PCOS, or NAFLD/MASLD), where the range of pathology is widest. In contrast, the association is attenuated in healthy, non-obese populations. For covariate adjustment, most studies accounted for at least one major confounder. Common adjustments included age, BMI/BMI-standard deviation score, or metabolic syndrome/disease status. While several studies with strong correlations adjusted for these factors (e.g., Heldens et al[19] in 2024; Karaman et al[24] in 2022), others with moderate correlations reported no adjustments (e.g., Çakır and Acunaş[23] in 2025; Ata et al[17] in 2023), suggesting that the underlying disease severity of the cohort may be a more dominant driver of correlation magnitude than the degree of adjustment. Regarding the different IR metrics, while HOMA-IR was the most common, the strongest correlations were observed with the more mechanistically relevant Adipo-IR (r = 0.65; Heldens et al[19] in 2024) and METS-IR (r = 0.49; Song et al[14] in 2025). This suggests that composite IR metrics may be more sensitive indicators of the relationship with elastography. The risk of bias is shown in Figure 4 and Supplementary Table 1.
Figure 4 Risk of bias summary of included cohort studies by domain (Newcastle-Ottawa Scale, n = 16).
This figure summarizes the methodological quality of the included non-randomized cohort studies using the Newcastle-Ottawa Scale. The results are categorized across the three core Newcastle-Ottawa Scale domains: Selection, comparability, and outcome. The green bars represent a low risk of bias, yellow/orange represents unclear risk, and red represents high risk for the corresponding domain criteria. The majority of studies showed a low risk of bias in the selection and outcome domains. However, the comparability domain exhibits the highest proportion of unclear or high risk (44% combined). This indicates a significant methodological limitation across the pediatric literature, primarily due to the failure of many studies to adequately report or adjust for critical confounding factors, such as age, body mass index, and pubertal stage. This weak comparability among studies is hypothesized to be a major contributor to the observed statistical heterogeneity (I2) in the meta-analysis results.
DISCUSSION
Pediatric MASLD is rapidly emerging as one of the most prevalent metabolic conditions worldwide, paralleling the epidemic of childhood obesity. Despite its growing burden, diagnostic evaluation in children remains limited by the impracticality of liver biopsy and the suboptimal specificity of biochemical markers[26]. This review provides the first quantitative synthesis demonstrating that ultrasound-based elastography parameters, particularly LSM and CAP, correlate strongly with metabolic indices of IR. These findings position elastography as a central, non-invasive biomarker linking hepatic pathology with systemic metabolic dysfunction. This systematic review and meta-analysis synthesized evidence from 16 studies encompassing over 2000 pediatric participants to evaluate the association between ultrasound-based hepatic elastography and IR in pediatric MASLD.
We found a consistent and statistically significant positive correlation between liver stiffness/steatosis parameters (LSM/CAP) and various clinical indices of IR across the included studies, with an overall pooled correlation coefficient of r = 0.44 (95% confidence interval: 0.38-0.50). This finding suggests that increased liver stiffness and attenuation, as quantified by elastography, reflect not only hepatic structural changes but also the metabolic consequences of systemic IR. The strongest associations were observed for composite indices, including Adipo-IR (r = 0.65) and the METS-IR (r = 0.49), underscoring their ability to capture multifactorial aspects of metabolic dysfunction[14,19].
Elastography as an integrated biomarker of metabolic dysfunction
The fundamental link between metabolic dysfunction and liver pathology lies in I (IR), which is widely recognized as the core pathophysiological driver of metabolically associated fatty liver disease, involving a complex interplay of lipid accumulation, hepatocellular stress, and fibrogenic activation[27]. Ultrasound elastography - whether through TE, two-dimensional shear-wave elastography (two-dimensional-SWE), or ARFI - provides a quantitative representation of these changes in real time (Kwon et al[20] in 2019; D'Hondt et al[11] in 2021). The consistent, statistically significant positive correlations observed in this review - specifically the moderate pooled effect size for liver stiffness and IR - therefore carry a critical mechanistic interpretation, reflecting hepatic fibrotic remodeling and systemic metabolic stress. Elastography parameters do not simply quantify the local extent of liver injury (fibrosis and steatosis) but act as a direct non-invasive surrogate for the underlying systemic metabolic dysregulation that initiates the disease cascade[28]. This strong association positions elastography - specifically LSM and CAP - as integrated biomarkers. They reflect the cumulative, IR-driven consequence (hepatic fat accumulation and subsequent fibrogenic response) of systemic metabolic stress in the pediatric population, making them highly valuable tools for assessing disease activity in high-risk youth[29]. By combining these domains, elastography may bridge traditional gaps between biochemical metabolic screening and imaging-based hepatic assessment.
Interpreting the strongest correlates (Adipo-IR and METS-IR)
Adipo-IR: The adipose tissue hypothesis in pediatric MASLD: The strongest correlations observed in this review - most notably the r = 0.65 relationship between LSM and Adipo-IR[19] - provide compelling support for the adipose tissue hypothesis in pediatric MASLD. This hypothesis posits that the initial driver of disease is dysfunction of peripheral adipose tissue. When adipocytes become saturated or inflamed, their ability to safely store lipids declines, leading to Adipo-IR. This resistance promotes excessive release of free fatty acids into circulation, which are subsequently shunted to ectopic sites, including the liver, where they contribute to hepatic steatosis - the “first hit” in MASLD pathogenesis[30,31].
This overflow of free fatty acids induces oxidative stress, chronic inflammation, and fibrogenesis - the “second hit” - which are captured non-invasively by liver stiffness measures such as LSM and SWV. The superior correlation of elastography with Adipo-IR, compared with simpler hepatic proxies such as HOMA-IR, indicates that elastography is highly sensitive to systemic adipose dysfunction - the earliest and most influential metabolic disturbance driving hepatic injury in children[32,33].
METS-IR: Reflecting the global metabolic burden: While Adipo-IR emphasizes the role of adipose tissue dysfunction, the strong correlation found between LSM and the METS-IR (r = 0.49; Song et al[14] in 2025) highlights the broader systemic metabolic burden. Unlike single-variable indices such as HOMA-IR, which rely solely on fasting glucose and insulin, METS-IR integrates multiple metabolic domains - fasting glucose, triglycerides, high-density lipoprotein cholesterol, and anthropometric measures such as BMI or waist circumference[34].
This composite structure enables METS-IR to capture the overall severity of metabolic dysregulation rather than isolated hepatic IR. The superior association between METS-IR and elastography parameters (both LSM and CAP) reinforces the concept that liver injury in pediatric MASLD is a multifactorial process, reflecting the cumulative impact of adipose, muscular, and hepatic IR. Thus, higher METS-IR values correspond with both hepatic stiffness and fat content, supporting its role as a holistic marker of metabolic injury[35,36]. These findings suggest that composite indices such as METS-IR should be prioritized in clinical and research settings to identify children at the highest risk for liver pathology. They also underscore the potential of ultrasound elastography as a sensitive indicator of global metabolic burden, rather than an isolated hepatic outcome[37].
Interpreting the relative strength of these correlations requires understanding the different methods used to assess IR in pediatric populations (Table 3). Among these, HOMA-IR remains the most commonly used due to its simplicity. However, it is heavily influenced by the physiological IR of puberty and primarily reflects hepatic rather than systemic resistance (Giannouli et al[10] in 2023). Adipo-IR, in contrast, targets the earliest defect in lipid handling, which appears to be the main pathogenic driver in pediatric MASLD. METS-IR complements this by integrating dyslipidemia and obesity-related parameters, thereby reflecting a broader metabolic phenotype[19]. Hence, the stronger correlations between LSM and Adipo-IR/METS-IR in this review likely reflect the fact that these indices align more closely with the multisystemic, lipid-driven pathophysiology underlying pediatric MASLD. Elastography, by quantifying both fat accumulation (CAP) and tissue stiffness (LSM), effectively mirrors these metabolic alterations, bridging imaging and biochemical dimensions of disease assessment[38].
Table 3 Different methods used to assess insulin resistance in pediatric populations.
IR measure
Primary physiological focus
Required inputs
Advantages
Limitations in pediatrics
HOMA-IR
Hepatic IR
Fasting glucose, fasting insulin
Simple, widely used, validated in adults
Overestimates IR during puberty; sensitive to fasting variations; poor reflection of adipose/muscle IR
QUICKI
Peripheral IR
Fasting glucose, fasting insulin
Good for detecting insulin sensitivity changes
Less reliable at extremes of glucose or insulin; limited pediatric validation
Adipo-IR
Adipose lipolysis and FFA flux
Fasting insulin × fasting free fatty acids
Directly reflects adipose dysfunction; links metabolic overflow to hepatic injury
Requires FFA measurement; less available in clinical labs
Together, these findings emphasize that not all IR indices are equally informative in pediatric MASLD. Composite indices like Adipo-IR and METS-IR outperform simpler hepatic proxies, such as HOMA-IR, because they capture the complexity of metabolic dysfunction across adipose, hepatic, and muscular compartments. Their superior correlation with elastography parameters underscores a crucial insight: Ultrasound-based hepatic elastography is not merely a hepatic imaging tool - it serves as a non-invasive mirror of systemic metabolic health in children and adolescents[39,40].
Explaining heterogeneity and population effects
The high heterogeneity observed in the meta-analysis (I2 = 72%) and the wide range of correlation coefficients (r = 0.25-0.65) are primarily due to differences in cohort characteristics and methodological variability. The most significant finding regarding heterogeneity is the dependence of the correlation strength on the patient’s disease spectrum. The highest correlation values were observed in cohorts with established or severe metabolic dysfunction, such as Severe Obesity (Heldens et al[19] in 2024; r = 0.65) and PCOS (Giannouli et al[10] in 2023; r = 0.61). These groups represent populations with a wide range of pathological variation, where IR is severe, established, and actively driving advanced liver pathology. In these settings, elastography serves as an effective differentiator of disease severity. Conversely, the weakest correlation (r = 0.25) was found in the cohort of healthy controls (Rose et al[21] in 2023). This is expected, as individuals with a narrow, normal range of IR and minimal or absent liver fat/fibrosis lack the metabolic signal necessary to establish a tight clinical correlation. This gradient underscores a dose-response relationship between metabolic severity and liver stiffness. Variability in elastography devices, probe types, fasting status, and reference indices (HOMA-IR vs Adipo-IR or METS-IR) likely contributed further to heterogeneity (Karaman et al[24] in 2022; Lefere et al[13] in 2025). Standardization of acquisition protocols and cut-off thresholds is thus imperative for pediatric applications. This also underscores that elastography is most informative for metabolic risk stratification when used in an at-risk or diseased population. Beyond cohort variability, part of the observed heterogeneity may reflect true biological diversity in IR-liver coupling across pubertal stages and ethnic backgrounds[41]. Pubertal insulin physiology, for instance, transiently reduces insulin sensitivity, which may amplify or obscure correlations with hepatic parameters depending on the cohort composition[42]. Recognizing and stratifying analyses by developmental stage could reduce this variance in future meta-analyses.
Differential utility of elastography measures
The two main elastography measures - CAP and LSM - provide distinct yet complementary clinical information for assessing IR and metabolic dysfunction in children (Table 4, Figures 5 and 6). Measures of hepatic steatosis (CAP/attenuation) demonstrated moderate correlations with IR indices (r = 0.28-0.48), consistent with their role in reflecting early lipid accumulation. Liver fat deposition represents the immediate, physical manifestation of hepatic IR, often termed the “first hit” in the pathogenesis of MASLD[43]. CAP thus serves as an excellent tool for early screening and monitoring of IR-driven hepatic fat regression in response to dietary or lifestyle interventions, functioning as a sensitive, non-invasive biomarker of early metabolic stress[12,15]. In contrast, measures of liver stiffness (LSM or SWV) showed the highest correlation maxima (r up to 0.65) with IR. Increased stiffness primarily reflects inflammation and fibrosis, which correspond to the “second hit” - a later stage in disease progression characterized by sustained, severe IR and lipotoxic injury[13,20]. Elevated LSM values, therefore, are indicative of chronic, severe, or aggressive IR that has progressed beyond simple steatosis, marking a transition toward irreversible parenchymal remodeling[44].
Figure 5 The principle of controlled attenuation parameter for steatosis assessment.
This schematic illustrates how the controlled attenuation parameter (CAP), measured in decibels per meter (dB/m), quantifies hepatic steatosis (liver fat). A: Sound waves passing through a healthy liver with homogeneous tissue, resulting in minimal signal energy loss (low attenuation) and a corresponding low CAP value; B: The effect of steatosis. Fat droplets (lipid vacuoles) are highly scattering and absorptive to the ultrasound signal. As the wave travels through the fatty liver tissue, it loses substantial energy, resulting in a significantly weakened signal upon returning to the receiver (high attenuation). This signal loss correlates directly with the degree of steatosis, yielding a high CAP value. CAP is therefore a direct, non-invasive biomarker for the physical presence of fat in the liver.
Figure 6 The principle of liver stiffness measurement for fibrosis assessment using transient elastography.
This schematic illustrates the mechanism by which liver stiffness measurement (LSM), typically measured in kPa, assesses liver fibrosis. LSM is derived from the speed of a low-frequency mechanical shear wave as it propagates through the liver. A: Depicts the shear wave traveling through a soft, normal (non-fibrotic) liver. The wave is minimally resisted and travels relatively slower; B: Depicts the shear wave traveling through a stiff, fibrotic liver (tissue hardened by collagen deposits). The rigid structure causes the shear wave to propagate faster. The core principle is that the stiffer the tissue, the faster the shear wave velocity. The LSM value is mathematically calculated from this velocity, providing a direct, non-invasive measure of liver stiffness, a validated surrogate for liver inflammation and fibrosis severity. LSM: Liver stiffness measurement.
Table 4 Comparative overview of controlled attenuation parameter, liver stiffness measurement, and shear wave elastography.
Both CAP and LSM are derived from ultrasound-based elastography, yet they measure fundamentally different hepatic properties. CAP quantifies ultrasound signal attenuation as it traverses lipid-laden liver tissue, serving as a direct index of steatosis (Ferraioli et al[12] in 2017). LSM, by contrast, quantifies shear-wave propagation velocity, which increases with tissue stiffness driven by fibrosis and inflammation (Heldens et al[19] in 2024; Lefere et al[13] in 2025).
A critical source of the high heterogeneity detected in this review is the use of different elastography modalities: TE/FibroScan, which generates LSM and CAP using a linear mechanical piston, and SWE and ARFI, which generate SWV using an acoustic push pulse (Figures 5, 6 and 7)[45]. While both TE (LSM) and SWE/ARFI (SWV) are validated non-invasive measures of stiffness, differences in wave-generating mechanisms (mechanical vs acoustic pulse), tissue sampling, and ultrasound integration yield non-interchangeable results and contribute to the observed statistical heterogeneity[46]. The consistency of the overall stiffness-IR correlation across these diverse platforms, however, reinforces the fundamental physiological link: The structural change (stiffness) is tightly tracking the metabolic driver (IR). This methodological variability, however, underscores the need for modality-specific cut-offs in future pediatric guidelines.
Figure 7 The principle of shear wave elastography for fibrosis assessment.
This schematic illustrates how shear wave elastography and acoustic radiation force impulse measure liver stiffness [shear wave velocity (SWV), measured in m/second or converted to kPa]. Unlike transient elastography, the shear wave is generated by a focused, acoustic push pulse emitted from a conventional ultrasound probe. This pulse perturbs the tissue, creating a shear wave that travels laterally. In normal liver tissue (low fibrosis), the wave travels slower, yielding a low SWV value. In fibrotic liver tissue (high stiffness), the wave travels faster. The system measures this speed and presents the result as a quantitative value (SWV) and often as a 2D color map, enabling real-time visualization of the stiffness distribution. This technique provides a non-invasive assessment of the inflammatory and fibrotic components of metabolic dysfunction-associated steatotic liver disease. ARFI: Acoustic radiation force impulse; SWE: Shear wave elastography; SWV: Shear wave velocity.
Meta-analytic data indicate that CAP moderately correlates with IR indices, such as HOMA-IR (r = 0.30-0.40), supporting its role as an early metabolic biomarker (Ferraioli et al[12] in 2017; Tas et al[15] in 2024). LSM, on the other hand, shows stronger correlations (r = 0.50-0.65) with composite indices such as Adipo-IR and METS-IR (Song et al[14] in 2025; Heldens et al[19] in 2024). This pattern underscores a progressive metabolic continuum, in which hepatic lipid accumulation evolves into inflammation and fibrosis as IR worsens. Clinically, CAP excels in early detection, identifying subclinical steatosis before irreversible damage occurs. It is especially valuable in pediatric obesity or metabolic clinics for routine screening, enabling the monitoring of treatment response and fat regression. Conversely, LSM provides prognostic information, aiding in disease staging and risk stratification. High LSM values indicate more advanced disease requiring intensified intervention and hepatology referral (Arsoy et al[16] in 2024; Kwon et al[20] in 2019). Measurement accuracy differs between CAP and LSM. CAP can be influenced by hepatic heterogeneity, high BMI, and subcutaneous adiposity, leading to variability in obese children (Ferraioli et al[12] in 2017). LSM, while more robust, is susceptible to transient hepatic inflammation and postprandial changes. For both parameters, standardized conditions - fasting state, appropriate probe size (S or M for pediatric populations), and high-quality acquisition metrics (interquartile range/median < 0.3) - are essential to ensure reproducibility (Brunnert et al[18] in 2022).
In pediatric MASLD assessment, combining CAP and LSM yields a comprehensive metabolic-liver profile. They are complementary, not interchangeable tools. CAP identifies early, reversible lipid accumulation, while LSM detects irreversible fibrotic progression[46]. Together, these parameters allow clinicians to differentiate between benign metabolic steatosis and advanced metabolic injury. Integrating both CAP and LSM measures provides a dual-pathway approach to pediatric metabolic liver assessment[24,47]. CAP identifies early reversible steatosis, and LSM characterizes irreversible structural damage. This combination enhances clinical decision-making, reduces the need for invasive biopsy, and improves risk stratification in children with obesity, metabolic syndrome, or suspected MASLD[48]. For example, a child with elevated CAP but normal LSM may be an ideal candidate for lifestyle modification. In contrast, concurrent elevation of both suggests advanced disease necessitating specialist evaluation and longitudinal monitoring (Karaman et al[24] in 2022).
The established positive association between liver stiffness and IR in adult MASLD populations is consistent with our pooled pediatric findings (r = 0.44). This transgenerational concordance implies that the link between fibrogenic activation and systemic metabolic dysfunction emerges early and persists into adulthood. Importantly, while adult studies often rely on HOMA-IR as the reference index, our synthesis demonstrates that Adipo-IR and METS-IR offer superior predictive value in pediatric populations (Heldens et al[19] in 2024; Song et al[14] in 2025). These composite indices account for pubertal shifts in body composition and hormonal milieu, providing a more holistic assessment of insulin sensitivity across adipose, hepatic, and muscular tissues.
Strengths and limitations
A transparent assessment of the review’s strengths and the constraints of the primary literature is essential for a high-quality systematic review. Use the data from your risk of bias table. The strength of the review lies in its focus on non-invasive screening tools, which are clinically critical for the pediatric population, where liver biopsy is strongly discouraged. In addition, systematic inclusion of various elastography modalities (TE, SWE, ARFI) and IR indices provides a broad overview of the current evidence. However, this review has several limitations. Most of the included studies were cross-sectional (e.g., Arsoy et al[16] in 2024, Ata et al[17] in 2023, Brunnert et al[18] in 2022, Bailey et al[22] in 2017, Çakır and Acunaş[23] in 2025). Cross-sectional designs can only show associations, not establish definitive causality or track changes over time. The “high concern (reference standard)” noted in the risk of bias for some diagnostic accuracy studies (D'Hondt et al[11] in 2021) indicates the use of an imperfect method, such as ultrasound grading, instead of the more accurate liver biopsy or magnetic resonance imaging-proton density fat fraction. This introduces heterogeneity and potential misclassification bias in diagnosing steatosis/fibrosis, making direct comparisons more difficult. In addition, differences in elastography technology (e.g., TE vs SWE), device types, and the variety of IR indices used hinder data pooling and the definition of universal cut-offs.
Clinical implications and future research
Clinical implications: The detected high statistical heterogeneity (I2 = 72%) and the wide range of reported correlations are not unexpected but carry significant clinical and methodological implications. While some variability is due to differences in patient disease severity, a substantial portion arises from unstandardized diagnostic practices. Specifically, the use of diverse elastography modalities (TE, SWE, ARFI) and different devices/probes (e.g., M vs XL probes, various SWE systems) across studies introduces methodological noise. This powerfully highlights the urgent need for standardized protocols in the pediatric setting. Clinicians and researchers require device- and age-specific cut-offs for LSM and CAP to ensure that the integrated biomarker potential of elastography is realized consistently for routine IR risk stratification. However, the strong correlation reinforces that elastography is a valuable tool for non-invasive risk stratification in high-risk pediatric populations (e.g., youth with obesity and/or metabolic syndrome). Elastography can serve as a non-invasive outcome measure to monitor the effectiveness of lifestyle or pharmacological interventions. A decrease in CAP (steatosis) and/or LSM (fibrosis) could track improvement in IR[49].
Future research directions: The critical next step is to conduct longitudinal, prospective intervention trials. These studies must demonstrate that a change in IR (e.g., a reduction after treatment) precedes and causes a corresponding change in elastography values, thereby solidifying the utility of elastography as an IR-driven biomarker. Future research must also focus on establishing standardized, validated cut-off values for elastography parameters (CAP and LSM) that reliably identify clinically significant IR in diverse pediatric populations (age, ethnicity, pubertal stage). Future studies should also compare TE, SWE, and ARFI head-to-head within the same pediatric cohort to derive a universally applicable non-invasive metabolic model. Ultimately, integrating elastography parameters into predictive models that combine imaging, biochemical, and genetic markers could transform pediatric MASLD management from reactive to preventive medicine.
Clinical perspective
The findings of this meta-analysis underscore the practical clinical value of integrating elastography into pediatric metabolic assessment. In explicit contrast to conventional biochemical indices (e.g., alanine aminotransferase/aspartate aminotransferase and HOMA-IR), elastography offers a non-invasive, rapid, and direct measure of the physical consequences of IR on the liver (steatosis and fibrosis)[50]. While alanine aminotransferase and aspartate aminotransferase are often late and non-specific markers of liver injury, and HOMA-IR is highly variable, CAP and LSM/SWV provide quantitative, site-specific structural and functional data[51].
CAP and LSM/SWV together offer a rapid, radiation-free, and child-friendly alternative for monitoring metabolic liver injury. This capability makes elastography a strong candidate for integration into routine metabolic screening workflows for high-risk pediatric populations (e.g., obesity or endocrinology clinics), providing immediate, actionable data to refine risk stratification. When interpreted alongside composite IR indices such as Adipo-IR or METS-IR, these tools enable early detection of at-risk youth, guide individualized interventions, and potentially obviate the need for invasive liver biopsy. Implementation of standardized pediatric reference values and device-specific calibration will further enhance the clinical precision of elastography in real-world practice.
CONCLUSION
This systematic review and meta-analysis provide robust evidence that hepatic ultrasound elastography parameters, particularly LSM and CAP, are significantly correlated with validated indices of IR in children and adolescents with MASLD. The pooled correlation coefficient (r = 0.44, 95% confidence interval: 0.38-0.50) demonstrates that elastography captures both structural and metabolic signatures of pediatric MASLD. Among the evaluated IR metrics, Adipo-IR and METS-IR exhibited the strongest correlations with elastography measures, highlighting their superior ability to reflect systemic metabolic dysfunction beyond hepatic glucose regulation. The differential utility of CAP and LSM was reaffirmed - CAP reflecting early, reversible fat accumulation (“first hit”) and LSM reflecting later, fibrotic and inflammatory remodeling (“second hit”). Together, these parameters provide a comprehensive, non-invasive representation of metabolic liver injury across the disease continuum.
Clinically, these findings suggest that ultrasound elastography can serve as a practical and quantitative biomarker for early identification, risk stratification, and monitoring of metabolic injury in children with obesity and metabolic syndrome. Incorporating elastography into pediatric metabolic screening pathways could enhance early detection of MASLD progression, reduce reliance on liver biopsy, and allow for real-time monitoring of treatment response. Future research should focus on prospective, longitudinal studies that evaluate how dynamic changes in IR indices correspond to changes in elastography parameters over time. Establishing standardized, age- and device-specific thresholds for CAP and LSM in pediatric populations will be essential for their adoption as validated non-invasive biomarkers of metabolic health.
Al-Beltagi M, Bediwy AS, Saeed NK. Insulin-resistance in paediatric age: Its magnitude and implications.World J Diabetes. 2022;13:282-307.
[PubMed] [DOI] [Full Text]
Giannouli A, Efthymiou V, Konidari M, Mani I, Aravantinos L, Dourakis SP, Antoniou A, Deligeoroglou E, Bacopoulou F. The Burden of Non-Alcoholic Fatty Liver Disease in Adolescents with Polycystic Ovary Syndrome: A Case-Control Study.J Clin Med. 2023;12:557.
[RCA] [PubMed] [DOI] [Full Text][Cited by in RCA: 11][Reference Citation Analysis (0)]
Tas E, Sundararajan D, Lo JS, Morelli N, Garcia-Reyes Y, Ware MA, Rahat H, Ou X, Na X, Sundaram S, Severn C, Pyle LL, Børsheim E, Vajravelu ME, Muzumdar R, Dranoff JA, Cree MG. Diagnostic Accuracy of Transient Elastography in Hepatosteatosis in Youth With Obesity.J Endocr Soc. 2024;8:bvae110.
[RCA] [PubMed] [DOI] [Full Text][Cited by in Crossref: 1][Cited by in RCA: 4][Article Influence: 2.0][Reference Citation Analysis (0)]
Ata DS, Hassan AA, El-Aziz Awada MEA, Solimanb HH, Elamorosy D. Non-invasive assessment of hepatic fibrosis and steatosis by FibroScan in Egyptian children and adolescents with nonalcoholic fatty liver disease: A single center study.Alex J Pediatr. 2023;36:31-44.
[PubMed] [DOI] [Full Text]
Heldens A, Dupont E, Devisscher L, Buytaert M, Verhelst X, Raevens S, Van Vlierberghe H, Geerts A, De Bruyne R, Lefere S. Adipose Tissue Insulin Resistance Correlates with Disease Severity in Pediatric Metabolic Dysfunction-Associated Steatotic Liver Disease: A Prospective Cohort Study.J Pediatr. 2024;274:114171.
[RCA] [PubMed] [DOI] [Full Text][Cited by in Crossref: 7][Cited by in RCA: 6][Article Influence: 3.0][Reference Citation Analysis (0)]
Çakır MS, Acunaş B. The Effectiveness of Shear-wave Elastography in Childhood Hepatosteatosis Grading and Comparison with Sonographic Grading.J Acad Res Med. 2025;15:94-100.
[PubMed] [DOI] [Full Text]
Karaman ZF, Hatipoğlu N, Kardaş F, Saraçoğlu S, Direk G, Kendirci M, Coşkun A. Identifying the effects of excess weight, metabolic syndrome and insulin resistance on liver stiffness using ultrasound elastography in children.Turk J Pediatr. 2022;64:671-682.
[RCA] [PubMed] [DOI] [Full Text][Cited by in RCA: 3][Reference Citation Analysis (0)]
Accacha S, Barillas-Cerritos J, Srivastava A, Ross F, Drewes W, Gulkarov S, De Leon J, Reiss AB. From Childhood Obesity to Metabolic Dysfunction-Associated Steatotic Liver Disease (MASLD) and Hyperlipidemia Through Oxidative Stress During Childhood.Metabolites. 2025;15:287.
[RCA] [PubMed] [DOI] [Full Text][Cited by in RCA: 15][Reference Citation Analysis (0)]
Liguori A, Esposto G, Ainora ME, Mignini I, Borriello R, Galasso L, Paratore M, Giustiniani MC, Riccardi L, Garcovich M, Gasbarrini A, Miele L, Zocco MA. Liver Elastography for Liver Fibrosis Stratification: A Comparison of Three Techniques in a Biopsy-Controlled MASLD Cohort.Biomedicines. 2025;13:138.
[RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)][Cited by in RCA: 2][Reference Citation Analysis (0)]
Lin YC, Liao FM, Chao HC, Chen AC, Jeng YM, Lin CC, Tiao MM, Yang YJ, Yeung CY, Chen HL, Ni YH. Consensus Statement on Metabolic Dysfunction-Associated Steatotic Liver Disease in Children and Adolescents From the Joint TASL-TSPGHAN Expert Committee.JGH Open. 2025;9:e70137.
[RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)][Cited by in RCA: 8][Reference Citation Analysis (0)]
