Risk Prediction Models for Sarcopenia in Dialysis Patients: A Systematic Review.

Area of Science:

• Nephrology
• Geriatrics
• Biostatistics

Background:

• Sarcopenia is a significant concern for dialysis patients, with numerous risk prediction models developed.
• The quality and performance of existing sarcopenia prediction models in this population are not well-integrated or evaluated.
• There is a need for a comprehensive overview to guide the development of high-quality prediction models.

Purpose of the Study:

• To provide a comprehensive overview of current risk prediction models for sarcopenia in dialysis patients.
• To assess the quality and performance of existing models.
• To offer a reference for developing future high-quality prediction models.

Main Methods:

• A systematic search of ten electronic databases was conducted up to March 8, 2024.
• Two researchers independently assessed the risk of bias and applicability of 12 included studies.
• A meta-analysis of common predictors was performed using Revman 5.4 software.

Main Results:

• Thirteen risk prediction models for sarcopenia in dialysis patients were identified, with prevalence ranging from 6.60% to 63.73%.
• Model performance varied, with Area Under the Curve (AUC) from 0.776 to 0.945; only six models underwent internal validation and two external validation.
• Most studies exhibited a high risk of bias, and common predictors included age, body mass index, muscle circumference, and C-reactive protein.

Conclusions:

• Developing robust prediction models for sarcopenia onset in dialysis patients necessitates a rigorous design.
• Future verification requires multicenter external validation to ensure generalizability and reliability.
• The current landscape of prediction models is characterized by significant bias and limited external validation.