IDF21-0022 The non-invasive spectrum analysis of the radial pulse for renal function loss in patients with type 2 diabetes

Abstract

Background: Renal function loss is the sign of diabetic nephropathy and the risk for adverse renal events. Previous studies have shown some predictive power of Urine ACR (Albumin-Creatinine Ratio, eGFR (estimated glomerular filtration rate), and the radial pulse spectrum for the development or progression of diabetic nephropathy. However, the predicting power of those risk markers had not been surveyed completely. Aim: The purposes of this study was to examine if spectrum analysis of the radial pulse could be used to predict the development of adverse renal events, including 40% eGFR decrease, doubling of serum creatinine or end-stage renal disease (ESRD) in patients with type 2 diabetes. Method: The longitudinal study followed 2,193 patients with type 2 diabetes, 57.5% male, without prior renal failure or ESRD, through the division of Endocrinology and Metabolism of the Zhongxiao Branch of Taipei City Hospital (IRB: ISTCTN14306167) for over four-years. During the initial visit, a 12-second radial pulse was measured using a TD01C. Spectrum analysis of the radial pulse provided the first seven harmonics which were described by the index notation, Cn (n = 1–7). At each three-month follow up appointment kidney function blood tests were taken and correlated with adverse renal events. We identify the harmonics that are associated with the adverse renal event using logistic regression and Cox proportional hazard model. Results: In 2,193 patients, 106 (4.8%) had 40% decline of eGFR, 465 (21.2％) had doubling of serum creatinine, and 12 (<1%) had ESRD. The logistic regression identified C7 and C3 are important and independent predictors of the risk of renal adverse events after adjusting for age, sex, duration of diabetes, and smoking, (C3: HR, 0.288; 95% CI 0.140 to 0.594; P<0.001; C7: HR, 1.567; 95% CI 1.200 to 2.046; P<0.001). A Cox proportional hazard model with the RFLRisk index combining C3 and C7 with equal weight to identify patients who may be at high risk for adverse renal events. The patients with RFLRisk = High (C3<0.491 and C7>0.566) had 1.42 times cumulative incidence of adverse renal events, compared to patients with RFLRisk = Low (C3>0.491 and C7<0.566). Discussion: This study suggested that the specific harmonic profiles of radial pulse waves may be used as indicators to predict the risk of adverse renal events. High RFLRisk index leads to increasing risk of adverse renal events in patients with type 2 diabetes.