Diagnostic performance of point shear wave elastography and vibration-controlled transient elastography in paediatric metabolic dysfunction-associated steatotic liver disease: a systematic review and meta-analysis

Abstract

Objective

This study aimed to comprehensively assess the diagnostic accuracy of point shear wave elastography (pSWE) and vibration-controlled transient elastography (VCTE) in paediatric metabolic dysfunction-associated steatotic liver disease (MASLD).

Design

Systematic review and meta-analysis of diagnostic test accuracy using the Grading of Recommendations Assessment, Development and Evaluation approach with random-effects models.

Data sources

PubMed, Embase, Web of Science, Ovid (Medline), Cochrane, China National Knowledge Infrastructure, Wan Fang and OpenGrey were searched for publications from April 1989 to July 2024.

Eligibility criteria

The study included relevant records on the application of pSWE and VCTE in diagnosing MASLD in children (<19 years). These studies evaluated the diagnostic performance (sensitivity, specificity, area under the receiver operating characteristic curve) of two techniques for paediatric MASLD in the context of a relative gold standard with no language barriers. Studies of children with concurrent liver diseases were excluded.

Data extraction and synthesis

Two independent reviewers extracted data and assessed the risk of bias. Articles were assessed using Quality Assessment of Diagnostic Accuracy Studies 2 for diagnostic accuracy and potential biases, and bias risks were visually represented using the Risk of Bias Visualisation tool. Heterogeneity was assessed (Cochran Q statistic) and quantified (I² statistic). The analysis of the likelihood ratio graph (Fagan plot) indicates that both pSWE and VCTE provide valuable diagnostic support for MASLD.

Results

The comprehensive literature search yielded four pSWE studies encompassing 968 children and seven VCTE studies encompassing 1934 children. In our meta-analysis, VCTE had a sensitivity of 0.89 (95% CI, 0.79 to 0.94) and a specificity of 0.90 (95% CI, 0.83 to 0.95), which showed superior diagnostic accuracy compared with pSWE (sensitivity, 0.89 (95% CI, 0.81 to 0.94); specificity, 0.85 (95% CI, 0.81 to 0.89)).

Conclusion

While both pSWE and VCTE demonstrated appreciable diagnostic efficacy in paediatric MASLD, VCTE showed similar sensitivity but superior specificity, emerging as the more effective technique particularly in non-obese children. However, further investigation is warranted to fully elucidate the influence of probe selection and racial prevalence on these diagnostic modalities.

PROSPERO registration number

CRD42024514246.

Introduction

As living standards and economic conditions continue to evolve and improve, nonalcoholic fatty liver disease (NAFLD) has emerged as a significant health concern, affecting an increasing proportion of the paediatric population.¹ This condition is characterised by the accumulation of fat in the liver without excessive alcohol intake and is closely associated with metabolic syndrome, which encompasses several comorbidities, including obesity, type 2 diabetes, dyslipidaemia and hypertension.² Extensive research has demonstrated that the long-term detrimental effects of NAFLD on children are significantly more pronounced than those on adults, highlighting the urgent need for intervention and management strategies to address this growing epidemic. The early diagnosis of NAFLD is pivotal in mitigating disease progression, reducing the risk of liver cancer, enhancing patient prognosis and ultimately reducing mortality rates among children. Liver biopsy, although historically regarded as the gold standard for clinical diagnosis and staging of NAFLD and liver fibrosis, is not always feasible in paediatric populations due to its invasive nature and potential complications. These complications range from bleeding and biliary peritonitis to pneumothorax, posing particularly significant risks to paediatric patients and limiting its routine clinical application.³ Additionally, histological examination through biopsy may still be affected by sampling errors or inter-observer variability.⁴ Hence, it is crucial to combine serological markers with various imaging examinations (such as ultrasound (US) and MRI) for the diagnosis and staging of NAFLD. The MRI-estimated liver proton density fat fraction (PDFF) is highly correlated with histological steatosis grading, and it offers greater accuracy and reliability compared with traditional histological grading. PDFF accurately quantifies liver fat content with significant sensitivity and does not rely on physicians’ subjective judgement.⁵ It is widely regarded as the most reliable non-invasive alternative to biopsy.⁶ However, due to its high cost and the need for substantial patient compliance, it has not been widely accepted as the diagnostic gold standard for NAFLD in clinical practice, particularly in paediatric settings where children are often restless and uncooperative during the extended imaging procedure.

US, as a non-invasive, cost-effective, efficient and widely used diagnostic tool, has great prospects for large-scale screening and initial assessment of NAFLD.⁶ However, the diagnostic accuracy of conventional US in detecting various levels of hepatic steatosis remains inconsistent. It offers high diagnostic accuracy for moderate to severe hepatic steatosis (including ≥30% liver fat content), yet its sensitivity for mild steatosis (<30% liver fat) is still relatively low.7,9 Therefore, additional diagnostic examinations are crucial for the early and accurate diagnosis of NAFLD, particularly in its milder forms. US elastography (UE), with mechanical property assessment, indirectly or directly reflects differences in tissue elasticity and stiffness, demonstrating clinical utility for non-invasive diagnosis of various diseases of the liver, breast, thyroid, prostate and kidneys.¹⁰ Among these elastography techniques, vibration-controlled transient elastography (VCTE) has been globally applied as a pioneering approach. New methods in UE, including point shear wave elastography (pSWE) and two-dimensional shear wave elastography (2D-SWE), are increasingly being integrated into clinical practice for comprehensive liver assessment.

VCTE utilises the FibroScan device (Echosens, Paris, France) to generate shear waves that propagate through the liver, emitting pulse echoes to capture, receive and measure the shear wave speed, which directly reflects liver tissue stiffness. In contrast, pSWE is seamlessly integrated into traditional US systems, quantifying liver stiffness by utilising the power of transmitted acoustic pulses.¹¹ The flexible manipulation of the region of interest (ROI) allows easy assessment of any liver region, which makes it particularly suitable for children who tend to exhibit high activity levels, thereby reducing examination failure rates. As liver fibrosis represents a critical stage in the progression of liver disease, once established, its reversibility poses a significant challenge. Therefore, from the perspective of preventing liver fibrosis and safeguarding liver function, the early diagnosis of paediatric non-alcoholic fatty liver disease becomes crucial. Although many studies have directly or indirectly compared the diagnostic abilities of VCTE and pSWE in the non-invasive assessment of adult NAFLD and liver fibrosis,12,16analyzing early interventions in children with strong liver regeneration and repair capabilities offers profound prognostic advantages and may better inform clinical management strategies.

Notably, since 2023, three major international liver associations have proposed that metabolic dysfunction-associated steatotic liver disease (MASLD) should replace the term NAFLD.¹⁷ Emerging epidemiological evidence indicates a high consistency between the definitions of NAFLD and MASLD, with about 99% of NAFLD patients meeting the MASLD criteria.¹⁸ Therefore, this systematic review and meta-analysis aimed to comprehensively assess and directly compare the diagnostic performance and clinical utility of two non-invasive elastographic techniques—VCTE and pSWE—in paediatric MASLD, to establish which modality offers superior diagnostic reference value for clinical implementation.

Methods

Search strategy and selection criteria

A systematic search was conducted in seven biomedical databases: PubMed, Embase, Web of Science, Ovid (Medline), Cochrane, China National Knowledge Infrastructure (CNKI) and Wan Fang, covering publications from April 1989 to July 2024. This search was done to collect literature regarding the application of two UE techniques in the diagnosis and assessment of paediatric MASLD (NASLD). The literature search was systematically performed by using combinations of free text and Medical Subject Headings. Simultaneously, manual secondary searches were conducted on the references within the included studies and other similar literature to ensure thoroughness (online supplemental tables S1–S5online supplemental file 1, 1). Additionally, manual searches were conducted in OpenGrey for grey literature and non-English language publications to enhance the comprehensiveness of the literature inclusion.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• This review rigorously evaluated point shear wave elastography and vibration-controlled transient elastography for paediatric metabolic dysfunction-associated steatotic liver disease detection.

• Fagan’s nomogram was used to analyse performance across different prevalence settings.

• Subgroup analyses thoroughly addressed sources of heterogeneity in reference standards.

• A key limitation was the small number of the included studies.

• Not all studies used pathological confirmation as the gold standard due to population limitations.

Two independent researchers carefully screened the titles and abstracts of the retrieved articles to determine their eligibility for inclusion. If an article was considered potentially relevant, it underwent a comprehensive review of the full text. Cross-sectional, prospective or descriptive studies specifically involving children aged 19 years or younger were included. These studies focused on MASLD and NAFLD (in its prior nomenclature). Given the constraints of the study population, both techniques were evaluated against liver biopsy histopathology or MRI-derived parameters (such as MRI-PDFF) as reference standards.

The inclusion criteria were as follows: (1) clear diagnostic criteria, confirmed through biopsy, laboratory tests or relevant imaging, with data available for at least 50 patients; (2) examinations performed within 6 months before biopsy; and (3) ability to fully extract four-fold table data from the population (true-positive, false-positive, false-negative and true negative values).

The exclusion criteria were as follows: (1) children with concomitant liver diseases (viral hepatitis or benign and malignant liver tumours) or a history of related liver surgeries; (2) studies that did not include pSWE or VCTE diagnostic techniques; (3) insufficient data to calculate the necessary diagnostic indicators; (4) article types such as reviews, conference abstracts and case studies; and (5) data duplication.

Data extraction and quality assessment

Researchers Xu Zelin and Wang Sirui independently extracted data using standardised extraction tables to ensure precision and reproducibility. Any discrepancies encountered during this process were resolved through consensus achieved via consultation. If consensus could not be reached, these issues were resolved through rigorous group discussions or consultation with Li Jun (third researcher) to ensure data integrity and reliability. A comprehensive set of data was collected on various aspects of the included studies, including their characteristics (country, first author, year of publication and study type), patient demographics (cut-off, age and sex), as well as detailed information on the index text and its performance indicators (including the number of those with liver disease correctly detected, those with liver disease not detected, those with no liver disease incorrectly flagged as having liver disease and those with no liver disease correctly classified as not having liver disease, number of failures, sensitivity, specificity and area under the curve (AUC)). This study employed standardised diagnostic accuracy parameters to directly compare the performance of VCTE and pSWE, comprehensively characterising their diagnostic capabilities through sensitivity, specificity, AUC, positive likelihood ratio (PLR) and negative likelihood ratio (NLR).

Subsequently, the selected articles underwent rigorous quality assessment using the Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2) tool, with a focus on diagnostic accuracy. This evaluation allowed us to assess the methodological quality and potential biases in the included studies. Furthermore, an analysis chart was generated using the Risk of Bias Visualisation (ROBVIS) tool,¹⁹ providing a visual representation of the risks of bias across all included studies.

Statistical analysis

Relevant data were compiled from the included studies using Excel 2019 and analysed using STATA (version 15.1, USA) with the midas and metandi modules for diagnostic test accuracy meta-analysis. First, statistical heterogeneity was assessed using Cochran’s Q test and the I² statistic. The I² statistic quantifies the proportion of total variation attributable to between-study heterogeneity rather than chance, and clinical and methodological diversity was anticipated across paediatric elastography studies. Therefore, the plan was to use random-effects models regardless of heterogeneity metrics, which was ultimately confirmed appropriate by our heterogeneity assessment. For meta-analysis of diagnostic test accuracy studies, a bivariate random-effects model was employed to jointly pool sensitivity and specificity across studies, appropriately accounting for the correlation between these parameters. This model simultaneously analysed paired sensitivity and specificity data from each study, considering both within-study variability and between-study heterogeneity. From this model, pooled estimates of sensitivity, specificity, PLR, NLR, and diagnostic odds ratio (DOR) were calculated with their respective 95% CIs.

Several pre-planned analyses were conducted to explore heterogeneity sources:

1. Sensitivity analyses using the leave-one-out method to identify influential studies that affect heterogeneity. Each study was sequentially removed from the analysis to observe changes in the pooled estimates and heterogeneity metrics.

2. Subgroup analyses based on reference standard categorisation (imaging or liver histology, biopsy and MRI-PDFF), probe type for VCTE studies (S, M and XL) and geographic region (Europe, Asia, and the Americas).

3. Meta-regression examining the effect of study-level covariates including sample size, mean age, body mass index, obesity, gold standard method and technology type on diagnostic accuracy metrics. The meta-regression models were constructed using the metareg command in STATA, with log DOR as the outcome variable.

Summary receiver operating characteristic (SROC) curves were generated to visualise the overall diagnostic performance, and the AUC was calculated to quantify discriminative ability. A Fagan plot was used to illustrate the relationship between pre-test and post-test probabilities for clinically relevant prevalence scenarios (25%, 50% and 75%). Publication bias was assessed using the Deeks funnel plot, with p<0.05 indicating potential publication bias. All statistical tests were two-sided, and p<0.05 was considered statistically significant unless otherwise specified.

Reporting

This study strictly adhered to Preferred Reporting Items for Systematic Reviews and Meta-Analyses of Diagnostic Test Accuracy Studies guidelines for design and reporting, ensuring comprehensive coverage of specific reporting requirements for diagnostic accuracy studies.

Role of the funding source

The funding entities were not involved in the study’s design, data collection, data analysis, data interpretation, report writing or decision to publish.

Patient and public involvement

None.

Results

Literature search and study characteristics

This meta-analysis was registered on the PROSPERO website (CRD42024514246). A total of 2032 studies were retrieved, and after using EndNote to remove 577 duplicates, 46 studies were preliminarily included based on their titles and abstracts. Additionally, four studies were manually identified from the reference lists as suitable for inclusion in the meta-analysis. After excluding 39 studies due to the unavailability of the full text (12 studies), incomplete data (three studies), non-child populations (18 studies), unclear gold standards (five studies) and duplicated data (one study), a final total of 11 studies were included, as detailed in figure 1.

Figure 1. Preferred Reporting Items for Systematic Review and Meta-analyses 2020 flow chart.

A total of 8 years of studies published from 2016 to 2024 were included. The detailed features of VCTE20,26 and pSWE27,30 for diagnosing children in each study are summarised in table 1. Among these, three studies^{20 22 23} using the VCTE technique reported measurement success rates ranging from 47.59% to 96.77%, with detailed reports of the specific reasons for data collection failures (online supplemental table S6). The study drew from different regions worldwide, including Europe (33%), Asia (50%) and the Americas (17%). QUADAS-2 results show that no significant publication bias was observed across the studies (p=0.77, online supplemental figure S10), with a detailed quality assessment shown in figure 2.

Table 1. Characteristics of the included studies.

Author (year)	Country	Technology	N	Age(years)	Study/clinical setting	Gold standard	Cut-off	Failure
Ye XR et al27 (2020)	China	ARFI	212	8.7±1.0	Prospective, single-centre	Imaging or liver histology	1.09 m/s	0
Desai et al22 (2016)	France	FibroScan	69	16.0±2.9	Prospective, single-centre	Biopsy	225 dB/m	76
Ferraioli et al23 (2017)	Italy	FibroScan 502 Touch	289	11.5±2.7	Prospective, single-centre	Imaging or liver histology	251 dB/m	16
Zheng et al30 (2016)	China	ARFI	106	8.6±1.1	Prospective, single-centre	Imaging or liver histology	1.08 m/s	0
Xin et al28 (2020)	China	pSWE	431	9.7±1.8	Prospective, single-centre	Imaging or liver histology	1.35 m/s	0
He MQ et al37 (2020)	China	pSWE	219	9.21±1.20	Prospective, single-centre	Imaging or liver histology	4.202 kpa	0
Shin et al24 (2019)	USA	FibroScan	86	13.1±2.7	Prospective, single-centre	MRI-PDFF	241 dB/m	0
Runge et al20 (2021)	Netherlands	FibroScan 502 Touch	60	14.1±2.0	Prospective, double-centre	MRI-PDFF	277 dB/m	2
Chaidez et al25 (2022)	USA	FibroScan 502 Touch	206	13.7±3.7	Prospective, single-centre	Biopsy	259 dB/m	0
Kara et al26 (2024)	Turkey	FibroScan	211	14.30±2.49	Prospective, single-centre	Imaging	242 dB/m	0
Jia et al21 (2024)	China	FibroScan 502 Touch	105	12.3±3.9	Prospective, single-centre	MRI-PDFF	237 dB/m	0

ARFI, acoustic radiation force impulse; MRI-PDFF, magnetic resonance imaging proton density fat fraction.

Figure 2. Bias risk of the included studies (Quality Assessment of Diagnostic Accuracy Studies 2) criteria with the Cochrane Risk of Bias tool.

Heterogeneity analysis and performance of each technique in children with metabolic dysfunction-associated steatotic liver disease

The heterogeneity test revealed Q=31.06, p=0.0006, I²=70.84%, indicating significant statistical heterogeneity. The heterogeneity test using the leave-one-out method revealed that two studies exerted a substantial influence on the overall heterogeneity, and after their exclusion, I²=59.99% (online supplemental table S7). The combined sensitivity of the included studies was 0.89 (95% CI, 0.83 to 0.93), with specificity of 0.88 (95% CI, 0.84 to 0.92) and area under the receiver operating characteristic curve (AUROC) of 0.94 (95% CI, 0.92 to 0.96), which is close to 1, indicating a high accuracy of the result (online supplemental figure S8). These results indicate that UE has a good capability for diagnosing MASLD. Among them, the overall sensitivity and specificity of pSWE were 0.89 (95% CI, 0.81 to 0.94) and 0.85 (95% CI, 0.81 to 0.89), with an AUROC of 0.90 (95% CI, 0.87 to 0.92) (online supplemental figure S8 and figure 3). VCTE had an overall sensitivity of 0.89 (95% CI, 0.79 to 0.94), specificity of 0.90 (95% CI, 0.83 to 0.95) and AUROC of 0.95 (95% CI, 0.93 to 0.97), indicating that it has superior diagnostic performance compared with pSWE, and its specificity is superior with a lower likelihood of a false-positive rate. The subgroup analysis based on the diagnostic gold criteria included in each literature did not reveal significant statistical significance, indicating that the diagnostic gold criteria were not the main cause of heterogeneity (online supplemental figure S9). The inclusion of four articles^{20 21 25 30} resulted in a high overall heterogeneity; thus, the reasons must be further explored to ensure the reliability of the research results.

Figure 3. Sensitivity and specificity of the diagnostic performance of pSWE and VCTE in paediatric MASLD. Abbreviations: MASLD, metabolic dysfunction-associated steatotic liver disease; psWE, point shear wave elastography; VCTE, vibration-controlled transient elastography.

Fagan plot analysis

The analysis of the likelihood ratio graph (Fagan plot) demonstrates the diagnostic value of both pSWE and VCTE for MASLD (figure 4). When examining different disease prevalence scenarios with pre-test probabilities of 25%, 50% and 75%, VCTE showed significantly better performance in modifying diagnostic probabilities compared with pSWE. Specifically, with pSWE, the positive test results yielded post-test probabilities of 28%, 53% and 77%, whereas the negative results corresponded to post-test probabilities of 22%, 46% and 72%, respectively. In contrast, VCTE demonstrated much stronger diagnostic capability, with positive test results yielding substantially higher post-test probabilities of 76%, 90% and 97%, whereas negative results corresponded to much lower post-test probabilities of 4%, 11% and 27% respectively. These findings clearly indicate that VCTE demonstrates superior diagnostic performance for paediatric MASLD, particularly in its ability to significantly increase confidence in positive diagnoses and rule out disease with negative results.

Figure 4. Fagan plot analysis for pSWE and VCTE offers valuable diagnostic support for paediatric MASLD: (A–C) Fagan plot analysis for pSWE. (D–F) Fagan plot analysis for VCTE. (A and D) Pre-test probability at 25%, (B and E) pre-test probability at 50% and (C and F) pre-test probability at 75%.

Discussion

A systematic search was conducted in Eight common Chinese and English databases and grey literature databases, retrieving a total of 2032 articles. Based on strict inclusion and exclusion criteria, the identified articles were screened and evaluated, ultimately including 11 articles, of which six were journals belonging to the second quartile and higher. Using the QUADAS-2 standard, a detailed assessment of the literature quality was conducted, finding no significant publication bias (p=0.77). The diagnostic performance of pSWE and VCTE for paediatric MASLD was further analysed and compared, which revealed that VCTE had a sensitivity of 0.89 (95% CI, 0.79 to 0.94) and a specificity of 0.90 (95% CI, 0.83 to 0.95), showing superior diagnostic accuracy compared with pSWE, which demonstrated a sensitivity of 0.89 (95% CI, 0.81 to 0.94) and a specificity of 0.85 (95% CI, 0.81 to 0.89).

The heterogeneity test results for this study showed I²=70.84%, indicating considerable heterogeneity. The sensitivity analysis and leave-one-out method revealed that the studies by Runge JH²⁰ (sensitivity=0.75, specificity=0.75) and Zheng N³⁰ (sensitivity=0.80, specificity=0.82) were the main sources of heterogeneity, likely because the sensitivity and specificity of VCTE and pSWE in diagnosing MASLD were lower than those of other included studies with sensitivity or specificity ≥0.85. To further investigate the sources of heterogeneity, subgroup analyses were conducted based on the gold standard methods, revealing no significant heterogeneity between subgroups (p=0.78), although heterogeneity was present within the imaging or liver histology and MRI-PDFF subgroups (p≤0.01). Beyond the studies by Runge JH and Zheng N, heterogeneity may also arise from differences in US equipment, variations between the operating physicians, whether external force was applied during US elasticity data collection and whether strict fasting was implemented before measuring liver stiffness.

Amidst the global surge in childhood obesity, the emergence of insulin resistance and adipose tissue dysfunction has become increasingly evident.³¹ A recent prospective study underscores a robust correlation between childhood obesity and deleterious liver outcomes during adolescence. MASLD, which is a persistent condition from childhood into adulthood, poses profound risks to children’s health. The consideration of new diagnostic algorithms and renaming in 2023 make hepatic steatosis and cardiac metabolic risk factors the key determinants of MASLD, increasing the focus on metabolic damage.³² This is evidenced by the significant extrahepatic incidence of MASLD-associated pain, affecting both children and adults.^{31 33 34} However, due to the significant economic burden, the high levels of physical activity among children and challenges in patient cooperation during examinations, MRI, despite its excellent diagnostic capabilities for nonalcoholic fatty liver, often fall short as the preferred noninvasive modality. Conversely, UE emerges as an optimal diagnostic tool for nonalcoholic fatty liver in children, attributed to its low cost, procedural brevity, minimal cooperation requirements and high diagnostic reproducibility.

This meticulous study revealed that VCTE surpasses pSWE in diagnosing MASLD among children. Despite regional and ethnic disparities that shape the prevalence of this condition among paediatric populations, a perusal of the Fagan plot manifestly demonstrates VCTE’s superior diagnostic accuracy compared with pSWE across various incidence rates (prior probabilities).³⁵ However, a meta-analysis study pointed out that VCTE has limitations in fully grading hepatic steatosis in patients with MASLD.³⁶ This observation underscores VCTE’s superiority in the paediatric context, irrespective of epidemiological variations. In VCTE studies reporting measurement failures, all four articles employed M-type probes,^{22 23 36} whereas two additional articles^{22 23} also incorporated XL-type probes. The study²² exhibiting the highest proportion of failures revealed that among the 76 invalid measurements, 67 were attributed to S-probe scanning. A previous study ¹⁶ consistently showed that the frequency of common faults detected by VCTE using M probes was significantly higher than with pSWE, exceeding a tenfold difference. However, the findings of our meta-analysis indicate a higher fault rate associated with S probes. This disparity may stem from variations in the baseline characteristics of the studied populations; however, conclusive evidence requires further in-depth analysis with a larger sample size.

This meta-analysis represents a groundbreaking study, being the first to assess the diagnostic accuracy of pSWE and VCTE technologies specifically in paediatric MASLD patients. The study not only validates the feasibility of extending diagnostic gold standards beyond the traditional pathological tissue biopsy but also establishes the diagnostic value of both methods for MASLD across various prevalence and prior probabilities. Furthermore, in contrast to previous studies, this study provides a preliminary exploration of the relationship between different VCTE probe selections and the corresponding measurement failure rates.

However, several limitations should be acknowledged. First, among the 11 studies included, 5 featured small sample sizes (n≤120), 10 were conducted as single-centre studies and five originated from China, collectively affecting the generalisability and reliability of the findings. Additionally, there is a notable absence of intra-centre comparisons, particularly those encompassing both pSWE and TE within the same patient population. Moreover, given the limited availability of research publications, the I2 statistic remained at a moderate level during heterogeneity analysis, even after excluding two outlier studies. Lastly, there is a dearth of in-depth analysis and validation regarding the grading of MASLD steatosis by these two technologies. In light of these limitations, in clinical practice, the limitations of both technologies must be considered, tailoring their use to the specific characteristics of individual patients and patient groups. Future research efforts should aim to incorporate larger sample sizes and adopt more rigorous study designs, such as multicentre trials with standardised protocols, to further illuminate the diagnostic capabilities of these technologies in paediatric MASLD patients.

Conclusion

In the context of non-invasive diagnostic methods for paediatric MASLD, VCTE has emerged as a more effective modality compared with pSWE. Particularly, its diagnostic precision is heightened in obese children with a large waist circumference. This finding not only confirms the clinical effectiveness of VCTE but also highlights its capacity to enhance diagnostic accuracy in paediatric patients with MASLD.

Data availability statement

Data are available upon reasonable request.