Sample-Invariant Salivary Urea Quantification across the Clinically Relevant Range Using a Silicon Device-Based pH Sensor

Abstract

Urea is a clinically important biomarker for monitoring chronic kidney disease (CKD). Saliva provides a noninvasive alternative to blood for urea measurements, making it attractive for routine disease monitoring. Urease-based pH sensing has been extensively investigated for urea diagnostics due to its transduction simplicity and clinical relevance. However, this approach has two key challenges that are critical for quantitative accuracy but remain unexplored in existing literature. First, significant variations in salivary pH and buffering capacity across populations and within the same person lead to sample-dependent pH-urea calibration curves, thereby compromising quantitative accuracy. Second, pH sensor responses saturate at elevated urea concentrations, thus limiting quantitative detection across the full clinically relevant range (2–45 mM). To address these challenges, a saliva dilution strategy was investigated using bipolar junction transistor (BJT)-based sensors with enhanced pH sensitivity. The optimal conditions, comprising a 60-fold dilution of saliva in 4 mM MES buffer (pH 5.88), produced a saliva-independent linear pH–urea calibration curve spanning the entire clinically relevant urea concentration range. The BJT sensor-based urea measurements were validated against a commercial gold-standard colorimetric assay, with Bland–Altman analysis showing a mean difference of −0.858 mM. In summary, this work provides a pathway toward robust, scalable, and miniaturized urea sensors for portable CKD diagnostics.