MO380: Bio-Conductivity Measurement for Electrical Impedance Tomography (EIT) in Chronic Kidney Disease—A Non-Invasive and Portable Monitoring Approach

Abstract

Abstract BACKGROUND AND AIMS Chronic kidney disease (CKD) is an escalating global health concern, and non-invasive means for early CKD detection is eagerly awaited. Kidneys are water- and electrolytes-rich organs with good bio-conductivity. Kidney fibrosis and tubular atrophy are characteristic pathological changes in CKD [1], and is associated with restricted blood perfusion, fluid diffusion [2,3] and hence reduced bio-conductivity. Electrical impedance tomography (EIT) emerges as a non-invasive, ionizing radiation-free imaging technique measuring dielectric properties of the body parts [4]. This study investigated the relationship of EIT indices and renal function in patients with various stages of CKD. METHOD We prospectively recruited 98 patients with different stages of CKD (Stage 1, n = 16; Stage 2, n = 15; Stage 3, n = 20, Stage 4, n = 36 and Stage 5, n = 11). EIT measurements were performed at two frequencies (33.6 and 100 kHz) using a portable EIT device with an electrode belt consisting of 16 gel electrodes. EIT data was pre-processed and reconstructed as mean conductivity maps. Frequency differencing was then performed between conductivity maps at two frequencies. Kidney-related dielectric parameters were extracted around the kidney region located at lower one-third of the frequency differencing conductivity map. Conductivity changes between frequencies were then computed and further regressed with individual biometrics. Predicted eGFR values was then computed and compared with the standard eGFR scores to evaluate its effectiveness on classifying CDK stages. RESULTS The detected mean bio-conductivity changes were −3.9, −67.5, −133.5, −212.6 and −294.7 (a.u.) in patients with CKD Stage 1, 2, 3, 4 and 5, respectively (Fig. 1A). Patients with more advanced stages of CKD tend to show more reduction in bio-conductivity, and eGFR correlated positively with reduction in bio-conductivity (R2 = 0.37, P < 0.001) (Fig. 1B). EIT-predicted eGFR (eGFREIT) was further computed by the captured conductivity changes and individual biometrics (Fig. 2). eGFREIT showed significant correlation with eGFR measured by blood tests (R2 = 0.54, P < 0.001). CONCLUSION EIT showed good correlation with renal function changes in CKD patients, and thus holds promise to be developed into a non-invasive and portable device for early CKD detection in both clinical and community settings.