<p>Chronic kidney disease (CKD) requires regular biochemical monitoring of renal function and mineral metabolism; however, the invasiveness of serum-based testing limits frequent assessment, prompting interest in saliva as a non-invasive alternative despite ongoing challenges related to collection methods and biomarker specific limitations. In this cross-sectional study, paired saliva and serum samples were collected from 47 healthy adults and 47 CKD patients. Saliva was obtained via passive drool, Salivette^®, and cotton roll. Creatinine, urea, calcium, and PTH were measured using standardized assays. Inter-method agreement was assessed with intraclass correlation coefficients (ICC) and Bland-Altman analysis, correlations salivary and serum biomarkers were assessed using correlation analyses evaluated salivary-serum associations, and linear regression examined predictive performance. ICC analysis showed moderate agreement for salivary creatinine, urea, and PTH across collection methods, with average-measure ICCs consistently higher than single-measure, while calcium had poor agreement. Bland-Altman plots revealed narrower limits of agreement for creatinine and urea, especially using cotton roll and passive drool, versus wider limits for calcium and PTH. Salivary creatinine and urea correlated moderately to strongly with serum levels, highest with cotton roll (creatinine ρ = 0.584; urea ρ = 0.648), and predicted serum concentrations in CKD patients (R² = 0.214, 0.339) better than in healthy controls, whereas calcium and PTH showed weak correlations and minimal predictive value. Salivary creatinine and urea show reliable agreement with serum levels and modest predictive value, supporting their potential utility as non-invasive biomarkers for renal function monitoring in CKD, although further validation is required before clinical implementation, whereas calcium and PTH are less valid, with cotton roll collection performing comparably to passive drool.</p>