Early identification of the pre-disease state is critical for effective intervention and improved outcomes in chronic kidney disease. However, current diagnostic approaches mainly rely on serum-based biomarkers, which are limited in capturing dynamic physiological changes or enabling real-time monitoring. Here, we present a smart contact lens (SCL) platform capable of real-time, non-invasive monitoring of tear uric acid (TUA) levels. The system integrates an electrochemical biosensor and wireless communication module, offering high sensitivity, stability, and selectivity. Using a hyperuricemic nephropathy (HN) rabbit model, we continuously tracked TUA levels and applied dynamic network biomarker (DNB) theory to identify the critical transition point preceding irreversible disease onset. The DNB-derived composite index enabled personalized identification of critical transition periods and detection of the reversible pre-disease state before progression to HN. Furthermore, longitudinal analysis revealed progressive shortening of the blood-to-tear diffusion lag time during disease progression, introducing transport kinetics as an additional dynamic biomarker associated with pathological changes. Continuous tear-derived dynamic signatures enabled more sensitive detection of early disease transitions than static serum biomarkers, highlighting the value of waveform-based analysis for pre-disease monitoring. Together, this work provides a framework for wearable, tear-based early warning signal detection and personalized monitoring of kidney dysfunction.