# Assessment of screening tools for diabetic sarcopenia in type 2 diabetes mellitus: evidence from a scoping review

## TL;DR

This review maps screening tools for diabetic sarcopenia in type 2 diabetes, comparing their accuracy and identifying gaps in validation.

## Contribution

The study systematically evaluates and categorizes screening tools for diabetic sarcopenia, highlighting their diagnostic performance and limitations.

## Key findings

- Muscle ultrasound showed high accuracy with sensitivity from 71.05% to 95.00%.
- Predictive models using Age, BMI, and HbA1c achieved AUC values between 0.800 and 0.932.
- SARC-F had moderate sensitivity but high specificity, while SARC-CalF improved diagnostic performance.

## Abstract

This study aimed to map and synthesize the available evidence on screening tools for diabetic sarcopenia in patients with type 2 diabetes mellitus (T2DM), highlighting their characteristics, application contexts, and research gaps.

A comprehensive search was conducted in PubMed, Web of Science, CNKI, and Wanfang Data to identify studies published from 2010 to Deccember 2025. Studies involving adults with T2DM that evaluated screening tools for sarcopenia against established diagnostic criteria (EWGSOP, AWGS, FNIH, or IWGS) were eligible. Two reviewers independently screened studies, extracted data, and assessed methodological quality using the QUADAS-2 tool. Findings were charted and synthesized narratively, with screening tools grouped into functional assessments, anthropometric measures, biomarker-based methods, imaging approaches, and predictive models.

A total of 24 studies with 9,469 participants were included. The most common screening tools were functional assessments, anthropometric measures, biomarkers, and muscle ultrasound. SARC-F showed moderate sensitivity (13.33%-62.63%) and high specificity (67.30%-91.67%), while SARC-CalF improved diagnostic performance. Muscle ultrasound demonstrated high accuracy, with sensitivity ranging from 71.05% to 95.00%. Predictive models with multiple variables (Age, BMI, HbA1c) showed AUC values between 0.800 and 0.932. Challenges included inconsistent cut-off values and limited validation across diverse populations.

Various screening approaches for diabetic sarcopenia have been explored, but no single tool is universally validated for T2DM. Combining functional questionnaires with objective assessments like ultrasound or biomarkers may offer a more practical solution. Future research should focus on standardizing thresholds and testing tools in diverse populations.