Early identification of cardiac autonomic neuropathy using complexity analysis in children with type 1 diabetes.

Abstract

In children and adolescents, there are significant limitations to detecting cardiac autonomic neuropathy (CAN), an important contributor to morbidity and mortality in adults with type 1 diabetes (T1D). The analysis of heart rate variability (HRV) is one method available to detect CAN. Some evidence shows traditional linear HRV measures detect abnormalities in youth with T1D. In this study, we aimed to assess whether non-linear complexity analysis of HRV would assist identification of CAN in youth with T1D and to assess contributory factors. We studied 19 youth with T1D and 17 healthy controls. Each had an electrocardiogram recorded continuously for 10 min, at a sampling frequency of 1000 Hz. Using Labview software and an algorithm for complexity analysis, along with standard time-domain and spectral analysis, recordings of the electrocardiogram were analysed to detect differences in HRV between groups. Youth with T1D had significantly higher sample entropy than controls (P = 0.015) suggesting increased complexity in HRV, but similar detrended fluctuation analysis (P = 0.68). Youth with T1D also had increased % high frequency power (P = 0.017) and reduced mid-frequency power (P = 0.019) on spectral analysis. There were no differences in heart rate or blood pressure responses to standing, or time-domain analysis of HRV. Within the T1D group, sample entropy correlated strongly with triglycerides (r = 0.76, P = 0.001) and detrended fluctuation analysis correlated strongly with serum potassium (r = -0.86, P < 0.001). Complexity analysis of HRV, particularly using sample entropy, may aid detection of CAN in youth with T1D.