Diagnostic performance of neck circumference for detecting steatotic liver disease in adults: a systematic review and meta-analysis protocol

Abstract

Introduction

Steatotic liver disease (SLD), particularly its metabolic forms (metabolic dysfunction-associated steatotic liver disease and metabolic dysfunction-associated steatohepatitis), is highly prevalent worldwide yet remains substantially underdiagnosed. Neck circumference (NC) is a simple anthropometric measure reflecting upper-body adiposity and cardiometabolic risk and has been increasingly investigated as a potential screening or risk-stratification tool for SLD. However, existing evidence is heterogeneous, and no comprehensive meta-analysis has synthesised the diagnostic performance of NC for detecting SLD. This systematic review and meta-analysis aim to evaluate the diagnostic accuracy and predictive value of NC in adults and to explore sources of heterogeneity across populations and study characteristics.

Methods and analysis

A systematic search of PubMed, Scopus, Web of Science and the Cochrane Library will be conducted for English-language studies published between January 2010 and September 2025. Observational studies in adults reporting NC measurements and confirmed steatotic liver disease diagnosed by imaging or biopsy will be included. Two reviewers will independently perform study selection and data extraction. Risk of bias and methodological quality will be assessed using a dual-framework approach: the Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool for diagnostic accuracy domains and the Joanna Briggs Institute (JBI) critical appraisal checklists for broader observational study design considerations. The primary outcomes are pooled sensitivity and specificity of neck circumference for detecting steatotic liver disease. Secondary outcomes include ORs, relative risks, correlation coefficients and diagnostic accuracy measures, as well as comparisons with other anthropometric indicators. Random-effects meta-analyses will be conducted where appropriate, with subgroup, sensitivity and publication bias analyses planned. Certainty of evidence for primary diagnostic accuracy outcomes (pooled sensitivity and specificity) and key secondary association outcomes will be assessed using the Grading of Recommendations Assessment, Development and Evaluation framework.

Ethics and dissemination

Ethical approval is not required, as this study synthesises published data. Results will be disseminated through peer-reviewed publication and scientific conferences to inform clinicians, researchers and policymakers on the potential role of neck circumference in screening and risk stratification for steatotic liver disease.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• The review is guided by a prospectively registered protocol and adheres to Preferred Reporting Items for Systematic Reviews and Meta-Analyses 2020 and Meta-analysis of Observational Studies in Epidemiology reporting standards. A completed Preferred Reporting Items for Systematic Reviews and Meta-Analyses Protocols checklist is provided as online supplemental file 1.

• Independent duplicate screening, data extraction and methodological appraisal using both Quality Assessment of Diagnostic Accuracy Studies-2 (for diagnostic accuracy domains) and Joanna Briggs Institute tools (for observational study quality) are implemented to enhance methodological rigour.

• Random-effects meta-analysis with predefined subgroup and sensitivity analyses is planned to address clinical and methodological heterogeneity.

• Differences in neck circumference measurement techniques and reference standards for steatotic liver disease may contribute to heterogeneity.

• Limiting inclusion to English-language observational studies may introduce language bias and residual confounding.

Introduction

The terminology used to describe fatty liver disease has undergone significant evolution in recent years. Historically, the condition was referred to as non-alcoholic fatty liver disease (NAFLD) and its progressive form non-alcoholic steatohepatitis (NASH). In 2023, an international multisociety consensus proposed a nomenclature update,¹ introducing steatotic liver disease (SLD) as the overarching umbrella term encompassing all causes of hepatic steatosis. Under this framework, SLD includes: metabolic dysfunction-associated steatotic liver disease (MASLD, formerly NAFLD) and its progressive form metabolic dysfunction-associated steatohepatitis (MASH, formerly NASH); metabolic dysfunction and alcohol-associated liver disease (MetALD), a newly introduced category describing individuals meeting MASLD criteria who also consume above-threshold levels of alcohol; alcohol-related liver disease (ALD); cryptogenic SLD; and SLD of specific aetiology (eg, drug-induced, genetic). MASLD represents a spectrum of liver disorders characterised by excessive hepatic fat accumulation in the presence of at least one cardiometabolic risk factor.¹,³ Globally, MASLD is the most prevalent chronic liver disease, affecting more than 30%–35% of adults worldwide.^{2 4} The prevalence of SLD is increasing together with metabolic disorders, including obesity, insulin resistance and type 2 diabetes.^{5 6} Despite its high prevalence and often asymptomatic course in early stages, SLD remains underdiagnosed in both clinical practice and population-based screening procedures.⁷ The disease progression is characterised by a transition from steatosis to chronic inflammatory processes in the liver, which can lead to scarring of hepatic tissue and development of fibrosis.^{8 9} Without timely intervention, this pathological condition can culminate in serious clinical complications, including liver cirrhosis and an increased risk of primary liver malignancy, specifically hepatocellular carcinoma.^{10 11} While the liver biopsy remains the gold standard for diagnosing SLD,¹² imaging modalities, including ultrasonography, CT and MRI, are widely used in clinical practice to determine the hepatic fat accumulation.^{13 14} These methods are expensive, invasive in the case of liver biopsy and impractical to use in a large number of people for routine screening. Consequently, there is an urgent need for simple, non-invasive and cost-effective screening tools in primary care and resource-limited settings that can facilitate the early detection of SLD and identify individuals at risk of disease progression.

More recent attention has focused on the neck circumference (NC) as a reliable anthropometric indicator of upper-body subcutaneous adipose tissue.¹⁵ NC is easy to measure, highly reproducible and relatively unaffected by diet, respiratory phase or clothing.¹⁶ Compared with traditional measuring methods, such as waist circumference (WC) and Body Mass Index (BMI), NC offers a unique assessment of ectopic fat distribution that may provide superior diagnostic utility for identifying individuals at risk of SLD.^{17 18}

Previous studies have suggested that NC can serve as a simple biomarker of obesity, insulin resistance and fatty liver.¹⁹,²³ However, to date, no comprehensive systematic review and meta-analysis has examined the diagnostic performance of NC for detecting SLD in adults or its applicability as a screening or risk-stratification tool in clinical and public health contexts. We aim to assess the diagnostic accuracy of NC for identifying steatotic liver disease in adults, as measured by pooled sensitivity and specificity, with the area under the receiver operating characteristic curve (AUC) as a secondary outcome.

This review will focus specifically on MASLD and MASH, the metabolic spectrum of SLD, as these are most directly relevant to neck circumference as an anthropometric marker of central adiposity and metabolic dysfunction. All other SLD subcategories, including MetALD, ALD, cryptogenic SLD and specific aetiology SLD, are excluded. Since a substantial body of existing literature predates the 2023 nomenclature change and continues to use legacy terms (NAFLD, NASH, metabolic dysfunction-associated fatty liver disease (MAFLD)), our search strategy is designed to capture all relevant studies regardless of terminology employed. The original terminology used by study authors will be retained as reported, with the corresponding current nomenclature equivalent noted alongside during data extraction.

Objectives

1. To evaluate the diagnostic performance of neck circumference for detecting steatotic liver disease in adults, by estimating pooled sensitivity and specificity using hierarchical diagnostic meta-analytic models.

2. To explore the association between NC and the presence or severity of steatotic liver disease, including steatosis grade and fibrosis stage, using effect estimates such as ORs, relative risks (RRs) and correlation/regression coefficients.

3. To compare the diagnostic and predictive performance of NC with other anthropometric indicators, such as BMI and WC, within studies where such comparisons are reported.

4. To investigate sources of heterogeneity in the diagnostic accuracy and associations of NC by population characteristics, sex, NC measurement method, disease definition, diagnostic modality and disease severity.

5. To identify methodological limitations and evidence gaps in the existing literature to inform future research and the potential clinical utility of NC as a screening or risk-stratification tool for steatotic liver disease.

By clarifying the predictive value and limitations of NC, this study seeks to inform pragmatic strategies for early identification of SLD.

Methods

This systematic review and meta-analysis will be conducted in accordance with the methodological guidance of the Joanna Briggs Institute²⁴ for evidence synthesis and will adhere to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Protocols (PRISMA-P)²⁵ to ensure transparency and completeness in reporting (online supplemental file 1: PRISMA-P checklist). The review will be reported in line with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA 2020)²⁶ and the Meta-analysis of Observational Studies in Epidemiology (MOOSE)²⁷ guidelines to maintain methodological rigour throughout the review process. The review process spans from September 2025 to January 2026.

Protocol registration

This systematic review and meta-analysis protocol was prospectively registered with the International Prospective Register of Systematic Reviews (PROSPERO; registration number: CRD420251243178).

Eligibility criteria

Table 1 summarises the inclusion and exclusion criteria applied to identify relevant studies for this systematic review and meta-analysis. The eligibility criteria were defined a priori and structured according to population, exposure, comparator, outcomes, study design, setting, language and publication period to ensure the relevance, methodological quality and interpretability of the included evidence on the diagnostic performance of NC for detecting steatotic liver disease in adults.

Table 1. Inclusion and exclusion criteria.

Category	Inclusion criteria	Exclusion criteria
Population	Adults aged ≥18 years from clinical or population-based settings, including individuals with confirmed SLD (including studies using historical terminology such as NAFLD, MAFLD, NAFL or NASH) and, where available, adults without steatotic liver disease	Participants aged <18 years; animal studies
Exposure	NC measured using any anatomical protocol, reported as a continuous or categorical variable	Studies without NC data or where NC could not be extracted
Comparator	Adults without steatotic liver disease when available; within-study comparisons across disease severity (eg, mild vs moderate/severe steatosis, fibrosis stages); studies reporting associations without a formal control group	Studies in which no comparative or association analysis related to NC is presented
Outcome	Presence or severity of steatotic liver disease diagnosed using ultrasound, CT, MRI, controlled attenuation parameter (CAP)/FibroScan or liver biopsy; diagnostic performance measures (AUC, sensitivity, specificity) or effect estimates (ORs, RRs, correlations)	Unconfirmed or unclear diagnosis of fatty liver disease; outcomes not related to steatosis or fibrosis
Context	Adult clinical, hospital-based or population-based settings in which both NC measurement and liver fat assessment were performed	Studies focused exclusively on paediatric populations or laboratory settings
Study type	Original, full-text, peer-reviewed observational studies with clearly reported sample sizes	Reviews, editorials, commentaries, conference abstracts without full data, case reports or case series
Study design	Cross-sectional, case–control and cohort studies (prospective or retrospective)	Randomised or non-randomised intervention studies; experimental laboratory studies
Language	English	Non-English publications
Publication date	January 2010 to September 2025	Published before January 2010
Access	Full-text articles available through open access or institutional access	Abstract-only publications; retracted studies

AUC, area under the curve; MAFLD, metabolic dysfunction-associated fatty liver disease; NAFL, non-alcoholic fatty liver; NAFLD, non-alcoholic fatty liver disease; NASH, non-alcoholic steatohepatitis; NC, neck circumference; RRs, relative risks; SLD, steatotic liver disease.

The search was limited to studies published from January 2010 onwards to reflect the period during which neck circumference began to be more widely investigated as a cardiometabolic and ectopic adiposity marker in adult populations.¹⁶ Around this time, emerging epidemiological evidence established NC as a reproducible anthropometric indicator associated with metabolic risk,^{16 28 29} prompting subsequent studies exploring its relationship with metabolic liver disease. Restricting the timeframe to 2010 onwards therefore ensures methodological relevance,¹⁶ consistency in diagnostic definitions³⁰ and alignment with contemporary imaging-based assessment of SLD.^{13 31}

To maintain methodological inclusiveness and avoid arbitrary exclusion, no fixed minimum sample size threshold will be applied. However, the potential influence of small-sample studies on pooled estimates will be explored through predefined sensitivity analyses.

Regarding disease definition, we adopt the current consensus terminology proposed by the multisociety nomenclature consensus.¹ The condition of interest is SLD, specifically restricted to its metabolic spectrum: MASLD and MASH. Studies reporting outcomes under legacy terminology (NAFLD, NASH, MAFLD) will be included if the study population is consistent with the current MASLD/MASH definition, specifically, the presence of hepatic steatosis of metabolic origin in the absence of significant alcohol consumption or other secondary causes of liver disease. Studies will be excluded if participants had: MetALD; ALD; viral hepatitis (hepatitis B or C); drug-induced liver injury; autoimmune liver disease; cryptogenic SLD and other specific aetiology SLD (genetic, endocrine or other secondary causes).

Information sources

A comprehensive search strategy across multiple databases will be employed in this systematic review and meta-analysis to identify relevant studies evaluating the diagnostic performance of NC for detecting steatotic liver disease in adults. The selected databases are PubMed, for comprehensive biomedical and clinical research; Scopus and Web of Science, for broad multidisciplinary coverage and citation tracking of epidemiological and diagnostic studies; and the Cochrane Library, to supplement these sources with records on diagnostic accuracy and observational research. Together, these databases are expected to provide comprehensive coverage of the existing literature on NC as an anthropometric marker for steatotic liver disease.

Search

Table 2 presents the search strategies developed for each database (PubMed, Scopus, Web of Science and the Cochrane Library). Each strategy combines relevant controlled vocabulary, where applicable, with free-text terms related to NC and steatotic liver disease, using Boolean operators and appropriate field tags to maximise sensitivity and ensure comprehensive retrieval of eligible studies. The search strategies were adapted to the indexing structure and syntax of each database. The full search strategies can be found in online supplemental file 2.

Table 2. Example search strategy.

Database	Example search strategy
PubMed/ MEDLINE	(‘neck circumference’) AND ((‘NAFLD’) OR (‘fatty liver’) OR (‘MAFLD’) OR (‘MASH’) OR (‘hepatic steatosis’) OR (‘liver fat’) OR (‘NASH’) OR (‘non-alcoholic fatty liver disease’) OR (‘Metabolic dysfunction-Associated Fatty Liver Disease’) OR (‘Metabolic dysfunction-Associated Steatohepatitis’) OR (‘Non-Alcoholic Steatohepatitis’) OR (‘NAFL’) OR (‘Non-Alcoholic Fatty Liver’) OR (‘MASLD’) OR (‘Metabolic Dysfunction-Associated Steatotic Liver Disease’)) Filters: English, 2010–2025
Scopus	(‘neck circumference’) AND ((‘NAFLD’) OR (‘fatty liver’) OR (‘MAFLD’) OR (‘MASH’) OR (‘hepatic steatosis’) OR (‘liver fat’) OR (‘NASH’) OR (‘non-alcoholic fatty liver disease’) OR (‘Metabolic dysfunction-Associated Fatty Liver Disease’) OR (‘Metabolic dysfunction-Associated Steatohepatitis’) OR (‘Non-Alcoholic Steatohepatitis’) OR (‘NAFL’) OR (‘Non-Alcoholic Fatty Liver’) OR (‘MASLD’) OR (‘Metabolic Dysfunction-Associated Steatotic Liver Disease’)) Limit: English, 2010–2025
Web of Science	(‘neck circumference’) AND ((‘NAFLD’) OR (‘fatty liver’) OR (‘MAFLD’) OR (‘MASH’) OR (‘hepatic steatosis’) OR (‘liver fat’) OR (‘NASH’) OR (‘non-alcoholic fatty liver disease’) OR (‘Metabolic dysfunction-Associated Fatty Liver Disease’) OR (‘Metabolic dysfunction-Associated Steatohepatitis’) OR (‘Non-Alcoholic Steatohepatitis’) OR (‘NAFL’) OR (‘Non-Alcoholic Fatty Liver’) OR (‘MASLD’) OR (‘Metabolic Dysfunction-Associated Steatotic Liver Disease’)) Filters: English, 2010–2025
Cochrane Library	(‘neck circumference’) AND ((‘NAFLD’) OR (‘fatty liver’) OR (‘MAFLD’) OR (‘MASH’) OR (‘hepatic steatosis’) OR (‘liver fat’) OR (‘NASH’) OR (‘non-alcoholic fatty liver disease’) OR (‘Metabolic dysfunction-Associated Fatty Liver Disease’) OR (‘Metabolic dysfunction-Associated Steatohepatitis’) OR (‘Non-Alcoholic Steatohepatitis’) OR (‘NAFL’) OR (‘Non-Alcoholic Fatty Liver’) OR (‘MASLD’) OR (‘Metabolic Dysfunction-Associated Steatotic Liver Disease’)) Limits: English, 2010–2025
Notes	Limits on language (English) and publication years (2010–2025) are consistently applied to capture relevant studies.

MASH, metabolic dysfunction-associated steatohepatitis; MASLD, metabolic dysfunction-associated steatotic liver disease; NAFLD, non-alcoholic fatty liver disease; NASH, non-alcoholic steatohepatitis.

See online supplemental file 1, Table 3.1 for the full PubMed/MEDLINE string, 3.2 for Web of Science string, 3.3 for Scopus string, 3.4 for Cochrane Library (CENTRAL) and Section 5 for the complete list of filters and limits applied.

Selection of sources of evidence

All records identified through database searching will be imported into Zotero³² for reference management and duplicate removal. Full-text articles will be retrieved from online sources and stored locally. Study screening will be conducted using structured Microsoft Excel spreadsheets to document inclusion decisions and reasons for exclusion at the full-text stage. Title and abstract screening will be performed independently by two reviewers experienced in systematic review methods, using predefined eligibility criteria. This strategy helps minimise the risk of excluding relevant studies and strengthens the reliability of study selection. Before full screening, the eligibility criteria will be pilot-tested on a small sample of studies (approximately 3–5) to ensure consistent interpretation between reviewers and to allow clarification or refinement of the criteria where necessary. A full-text assessment of the remaining studies will be conducted by the same reviewers. Any differences in decisions between the two reviewers will be addressed through discussion until agreement is reached. If disagreements persist, a third senior reviewer with substantial expertise in evidence synthesis and fatty liver disease research will make the final determination. This process is intended to maintain a fair, comprehensive and methodologically sound study selection. The study selection process will be reported using a PRISMA 2020²⁶ flow diagram, detailing the number of records identified, screened, excluded with reasons and included in the final review. Use of this standardised reporting framework promotes transparency, ensures the traceability of inclusion and exclusion decisions, and strengthens the reproducibility and credibility of the evidence synthesis. This multi-step selection process, incorporating independent assessment, consensus-based resolution of disagreements and adherence to established reporting standards, is designed to ensure the robustness, validity and reliability of study selection in this systematic review and meta-analysis evaluating neck circumference as a diagnostic indicator of steatotic liver disease.

Data charting process

Data extraction will be undertaken in a systematic manner using a predefined, standardised extraction form developed in advance in accordance with the review objectives and research questions. The form will be pilot tested on a small subset of included studies (approximately 3–5) and refined as necessary to ensure clarity, completeness and consistent application by all reviewers. Calibration exercises will be conducted to align the understanding of extraction items and minimise variability between reviewers.

Data will be extracted independently and in duplicate by two reviewers with expertise in review methodology to enhance accuracy and reduce the risk of error. Extracted information will encompass bibliographic details; study design and setting; sample size; population characteristics (including age, sex, BMI, comorbidities and ethnicity); diagnostic methods; NC assessment details (measurement protocol, anatomical location, units and cut-off values); characteristics of comparator groups; and diagnostic accuracy data, including AUC, sensitivity, specificity and, where available, raw 2×2 contingency table data (true positives, false positives, true negatives and false negatives), as well as effect estimates (ORs, RRs, correlations and regression coefficients) and comparisons with other anthropometric indices. Relevant methodological features, potential sources of bias and study limitations will also be documented. Following independent extraction, discrepancies will be reviewed and resolved through discussion and consensus. Any unresolved differences will be adjudicated by a third senior reviewer with extensive experience in systematic reviews and fatty liver disease research. Where data are missing or unclear, corresponding authors may be contacted for clarification, and all communication will be recorded.

All extracted data will be organised and managed using Microsoft Excel to support data integrity checks and facilitate subsequent quantitative and narrative synthesis. This structured and transparent approach to data extraction is intended to ensure the robustness and reliability of the review findings and to provide a solid foundation for evaluating the diagnostic performance of NC in steatotic liver disease.

Data items

The following data will be systematically extracted from eligible studies to enable a comprehensive synthesis of evidence on the diagnostic value of NC in identifying steatotic liver disease (Table 4 in online supplemental file 2). All variables are prespecified to promote consistent interpretation and support robust analysis:

⇒ Author(s) and publication year: names of the primary authors and year of publication to identify temporal trends and authorship patterns.

⇒ Country of study: geographical location(s) where the study was conducted, providing context regarding population characteristics and healthcare settings.

⇒ Study type and design: the methodological framework (eg, cross-sectional, case–control, cohort, prospective or retrospective) to evaluate the study’s nature and methodological quality.

⇒ Participant characteristics: description of the study population, including sample size, age, sex distribution, BMI, comorbidities (diabetes, obesity, metabolic syndrome) and ethnicity.

⇒ Disease definition used: given the recent nomenclature transition, a substantial body of existing literature continues to use legacy terminology including NAFLD, NASH and MAFLD. Therefore, studies will be included regardless of the terminology used by original authors, provided the study population is consistent with the current MASLD/MASH definition, specifically the presence of hepatic steatosis of metabolic origin in the absence of significant alcohol consumption or other secondary causes of liver disease. The original terminology used by study authors will be retained as reported, with the corresponding current nomenclature equivalent noted alongside during data extraction. Subgroup analyses may explore potential differences arising from diagnostic criteria or terminology used across studies. Studies will be analysed according to the definitions used by the authors, without reclassification. When metabolic criteria required for a MASLD diagnosis are reported, subgroup analyses may be conducted to explore potential differences. The potential impact of this nomenclature transition will be considered in the interpretation of results.

⇒ NC measurement: detailed information on NC measurement protocols will be extracted, including the anatomical landmark used (eg, below the laryngeal prominence, mid-neck or other specified sites) to address heterogeneity in exposure assessment. When reported, this information will be used to explore potential sources of heterogeneity through subgroup or sensitivity analyses.

⇒ Comparator: characteristics of control groups where available (adults without steatotic liver disease) or within-study comparisons across disease severity (eg, mild vs moderate/severe steatosis or fibrosis stages).

⇒ Diagnostic method and outcomes: modality used to assess steatotic liver disease (ultrasound, CT, MRI, CAP/FibroScan or biopsy); diagnostic performance metrics (AUC, sensitivity, specificity); and, where available, raw diagnostic accuracy data enabling reconstruction of 2×2 contingency tables (true positives, false positives, true negatives and false negatives), in addition to effect estimates (ORs, RRs, correlations, regression coefficients) and comparisons with other anthropometric indicators (BMI, WC, waist-to-hip ratio). Given the known variability in sensitivity and specificity across different diagnostic modalities, diagnostic modality will be considered a source of clinical heterogeneity. Where feasible, subgroup and sensitivity analyses will be conducted according to the reference standard used for disease diagnosis.

⇒ Adjustment and confounding factors: covariates considered in analyses, such as age, sex, BMI, WC or metabolic comorbidities, to account for potential confounding.

⇒ Subgroup or stratified analyses: data reported by sex, ethnicity, population type, NC measurement method, disease severity or diagnostic modality, if available.

⇒ Key study findings and limitations: main results regarding the association between NC and presence or severity of steatotic liver disease, including predictive performance metrics and any observed thresholds. Limitations noted by authors will also be extracted.

These included data items will support a robust quantitative and qualitative synthesis of the diagnostic utility of NC for steatotic liver disease.

Critical appraisal of individual sources of evidence

A systematic and structured appraisal of the included studies will be undertaken to enhance interpretability and contextualise the strength and reliability of the evidence base. Given anticipated heterogeneity in study populations, NC measurement protocols and diagnostic reference standards, both methodological quality and diagnostic risk of bias will be carefully evaluated.

The methodological quality of included observational studies (cross-sectional, case–control and cohort) will be independently assessed by two reviewers using the appropriate Joanna Briggs Institute (JBI) critical appraisal checklists.³³,³⁵ These tools enable evaluation of sampling strategies, exposure measurement validity, confounding control and statistical analysis.

For diagnostic accuracy outcomes, risk of bias and applicability concerns will additionally be assessed using the Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool.³⁶ This framework evaluates four domains: patient selection, index test (NC measurement), reference standard (diagnostic modality) and flow and timing. Each domain will be judged for risk of bias, and the first three domains will also be evaluated for applicability concerns.

Assessments will be conducted independently and in duplicate. Disagreements will be resolved through discussion and, when necessary, consultation with a third senior reviewer. Appraisal findings will not be used as exclusion criteria but will be presented descriptively and incorporated into sensitivity analyses and interpretations of the results where appropriate. By integrating design-specific and diagnostic-specific appraisal frameworks, this review aims to ensure a comprehensive and methodologically robust evaluation of the evidence base.

Synthesis of results

Results will be synthesised systematically, transparently and reproducibly to comprehensively evaluate the diagnostic performance of NC for detecting SLD in adults. The data obtained during the data extraction phase will be collected, structured and presented in narrative descriptions and tables to facilitate interpretation and identify key patterns and trends.

The quantitative data, including study characteristics, population demographics, NC measurement details, diagnostic modalities and outcome metrics (AUC, sensitivity, specificity, ORs, RRs, correlations), will be tabulated descriptively using summary statistics such as means, medians, SD, ranges and proportions to provide a comprehensive summary of the included studies. Forest plots, summary tables, graphs and other visual representations may be used to illustrate distributions across variables such as population type, NC measurement protocols, disease severity and geographic regions.

For diagnostic accuracy outcomes, pooled sensitivity and specificity will be estimated using a bivariate random-effects model, which jointly models sensitivity and specificity while accounting for within-study correlation and between-study heterogeneity. Because different studies may apply varying neck circumference cut-off thresholds, potential threshold effects will be formally evaluated. Where substantial threshold variability is identified, a hierarchical summary receiver operating characteristic (HSROC) model will be implemented to account for differences in diagnostic thresholds across studies. Summary ROC (SROC) curves will be constructed to visually summarise overall diagnostic performance, and threshold effects will be explored through inspection of ROC space and meta-regression where appropriate.

The AUC will be derived from the hierarchical model (bivariate or HSROC, as appropriate) rather than pooled independently, ensuring that summary discrimination reflects the joint modelling of sensitivity and specificity while accounting for threshold effects and between-study heterogeneity.

For non-diagnostic association outcomes (eg, ORs, relative risks, correlations), random-effects models will be applied using restricted maximum likelihood (REML) estimation to quantify between-study variance (τ²), as REML provides more robust variance estimation compared with the traditional DerSimonian–Laird method, which tends to underestimate between-study variance particularly in settings with small studies and high heterogeneity.³⁷,³⁹ Heterogeneity will be assessed using the I² statistic and Cochran’s Q test, and potential sources of heterogeneity will be explored through subgroup analyses and meta-regression according to sex, ethnicity, population type, NC measurement site, disease definition, diagnostic modality and disease severity. All diagnostic meta-analyses will be conducted using appropriate statistical software capable of fitting bivariate or HSROC models.

Sensitivity analyses will be performed to evaluate the robustness of the findings, including exclusion of high-risk-of-bias studies and small-sample studies (eg, <50 participants), and by comparing pooled estimates across different sample size strata.

Qualitative or descriptive findings, including methodological considerations, limitations and author-reported recommendations, will be synthesised narratively to contextualise quantitative results. Critical appraisal outcomes will be integrated descriptively to highlight the quality of evidence supporting key findings and to inform interpretive commentary.

Publication bias will be assessed according to outcome type. For diagnostic accuracy outcomes, Deeks’ funnel plot asymmetry test will be used where sufficient studies (generally ≥10) are available. For non-diagnostic association outcomes (eg, pooled ORs or relative risks), conventional funnel plots and Egger’s or Begg’s tests will be applied where appropriate.

This multilevel synthesis will provide a comprehensive, nuanced and methodologically rigorous evaluation of the diagnostic utility of NC in detecting steatotic liver disease, accounting for study heterogeneity, methodological quality and contextual factors across included studies. Further, the certainty of the evidence will be assessed using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) framework.⁴⁰ GRADE assessment will be applied to the primary diagnostic accuracy outcomes (pooled sensitivity and pooled specificity derived from hierarchical models) and to key secondary outcomes, including pooled ORs and relative risks for the association between neck circumference and steatotic liver disease. Certainty ratings will consider risk of bias, inconsistency, indirectness, imprecision and publication bias, following guidance for diagnostic test accuracy and observational evidence. Summary of findings tables will be constructed for the primary diagnostic outcomes.

Ethics and dissemination

As this systematic review and meta-analysis synthesises existing published data without collecting primary data from human participants, formal ethics approval is not required. Ethical standards will be maintained through accurate reporting, proper citation of original sources and transparent description of methods and results.

Findings will be disseminated widely to maximise impact and support translation into clinical and research practice. Dissemination strategies will include submission to peer-reviewed journals, presentations at national and international conferences and sharing via professional and stakeholder networks in hepatology, epidemiology and public health. In addition, summary reports or policy briefs may be prepared to inform clinicians, researchers and policymakers about the potential of NC as a low-cost, accessible screening tool for steatotic liver disease.

Patient and public involvement

None.

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information.