Clinical Utility of the AITIS Model for Test-Free Identification of Sarcopenia in Patients with Stage IV–V Non-Dialysis-Dependent Chronic Kidney Disease

Why It Matters

Early, low‑cost detection of sarcopenia in advanced CKD enables timely nutrition and exercise interventions, potentially reducing hospitalizations and mortality without requiring specialized equipment.

Key Takeaways

• AITIS predicts sarcopenia with AUC 0.79 in stage IV–V CKD patients
• Model uses only age, sex, anthropometrics, and 20 functional questions
• Sensitivity 60% and specificity 82% outperform many resource‑intensive tests
• Accuracy improves to AUC 0.84 in low‑activity and stage IV CKD subgroups
• SHAP analysis highlights age, walking 1 km, and lifting 5 kg as top predictors

Pulse Analysis

Chronic kidney disease affects roughly one in ten adults worldwide, and patients in its advanced stages are especially prone to sarcopenia—a loss of muscle mass and function that drives falls, hospital stays, and cardiovascular events. Traditional sarcopenia screening relies on hand‑grip dynamometers, bioelectrical impedance analysis, or dual‑energy X‑ray absorptiometry, all of which demand trained staff and costly devices. In resource‑constrained clinics or home‑based care, these requirements create a diagnostic gap, leaving many high‑risk CKD patients undetected until complications arise.

The AITIS model sidesteps these barriers by leveraging a lightweight questionnaire that captures age, sex, body metrics and responses to 20 daily‑activity tasks. In a recent single‑center cohort of 236 stage IV–V CKD patients, AITIS delivered an AUC of 0.792, with overall accuracy of 75.8% and a specificity exceeding 80%. Notably, the model’s discriminative power rose to 0.843 among patients reporting minimal physical activity and to 0.824 for those in stage IV disease, suggesting heightened relevance where muscle decline is most pronounced. Explainability tools such as SHAP confirmed that age, the ability to walk 1 km, and the capacity to lift 5 kg drive the risk scores, aligning with clinical intuition about lower‑limb function.

For clinicians, AITIS offers a scalable, patient‑driven screening pathway that can flag sarcopenia risk before formal diagnostic thresholds are met. Early identification opens the door to targeted nutrition counseling, resistance‑training programs, and multidisciplinary monitoring, potentially curbing the cascade of adverse outcomes associated with muscle wasting. While the study’s single‑center, Asian cohort limits universal generalizability, the methodology sets a precedent for broader, multicenter validation and integration into telehealth platforms. As health systems seek cost‑effective ways to manage the growing CKD burden, AI‑enabled, test‑free tools like AITIS could become a cornerstone of preventive nephrology.