Abstract / Summary
Aim: Creatinine-based estimated glomerular filtration rate (eGFRcr) may overestimate kidney function in older adults with reduced muscle mass. Although current guidelines recommend cystatin C when creatinine-based estimates are unreliable, its routine use remains limited. We evaluated whether incorporating simple bedside muscle-related indicators into eGFRcr improves the identification of reduced kidney function defined by cystatin C-based estimated glomerular filtration rate (eGFRcys).
Methods: We conducted a multicenter cross-sectional study of hospitalized adults aged ≥ 60 years at four Japanese community hospitals. Three models were evaluated: eGFRcr only (Basic model), eGFRcr plus calf circumference (Model 1), and eGFRcr plus the finger-ring test (Model 2). The primary outcome was continuous eGFRcys, and eGFRcys < 45 mL/min/1.73 m2.
Results: Among 98 participants (median age 79 years), adding muscle-related indicators modestly improved model performance. Adjusted R2 increased from 0.641 (Basic model) to 0.668 (Model 1) and 0.680 (Model 2). The proportion of participants correctly classified with eGFRcys < 45 mL/min/1.73 m2 increased from 56.1% with the eGFRcr alone to 80.5% with both extended models. Discrimination also improved (area under the curve: 0.882 vs. 0.921 and 0.918, respectively), although the differences were not statistically significant.
Conclusion: Incorporating simple bedside muscle-related indicators to creatinine-based assessment improved the identification of reduced kidney function defined by eGFRcys in hospitalized older adults. These measures may complement eGFRcr interpretation when cystatin C measurement is unavailable or delayed. External validation is required before clinical implementation.