Comparing the accuracy of ES-BC, EIS-GS, and ES Oxi on body composition, autonomic nervous system activity, and cardiac output to standardized assessments

John E Lewis¹, Stacey L Tannenbaum¹, Jinrun Gao³, Angelica B Melillo¹, Evan G Long¹, Yaima Alonso², Janet Konefal¹, Judi M Woolger², Susanna Leonard¹, Prabjot K Singh¹, Lawrence Chen¹, Eduard Tiozzo^1
1Department of Psychiatry and Behavioral Sciences, ²Department of Medicine, University of Miami Miller School of Medicine, Miami, FL, ³State Farm Insurance, Bloomington, IL, USA

Background and purpose: The Electro Sensor Complex (ESC) is software that combines three devices using bioelectrical impedance, galvanic skin response, and spectrophotometry: (1) ES-BC (Electro Sensor-Body Composition; LD Technology, Miami, FL) to assess body composition, (2) EIS-GS (Electro Interstitial Scan-Galvanic Skin; LD Technology) to predict autonomic nervous system activity, and (3) ES Oxi (Electro Sensor Oxi; LD Technology) to assess cardiac output. The objective of this study was to compare each to a standardized assessment: ES-BC to dual-energy X-ray absorptiometry (DXA), EIS-GS to heart rate variability, and ES Oxi to BioZ Dx Diagnostic System (BioZ Dx; SonoSite Inc, Bothell, WA).
Patients and methods: The study was conducted in two waves. Fifty subjects were assessed for body composition and autonomic nervous system activity. Fifty-one subjects were assessed for cardiac output.
Results: We found adequate relative and absolute agreement between ES-BC and DXA for fat mass (r = 0.97, P < 0.001) with ES-BC overestimating fat mass by 0.1 kg and for body fat percentage (r = 0.92, P < 0.001) with overestimation of fat percentage by 0.4%. For autonomic nervous system activity, we found marginal relative agreement between EIS-GS and heart rate variability by using EIS-GS as the predictor in a linear regression equation (adjusted R² = 0.56, P = 0.03). For cardiac output, adequate relative and absolute agreement was found between ES Oxi and BioZ Dx at baseline (r = 0.60, P < 0.001), after the first exercise stage (r = 0.79, P < 0.001), and after the second exercise stage (r = 0.86, P < 0.001). Absolute agreement was found at baseline and after both bouts of exercise; ES Oxi overestimated baseline and stage 1 exercise cardiac output by 0.3 L/minute and 0.1 L/minute, respectively, but exactly estimated stage 2 exercise cardiac output.
Conclusion: ES-BC and ES Oxi accurately assessed body composition and cardiac output compared to standardized instruments, whereas EIS-GS showed marginal predictive ability for autonomic nervous system activity. The ESC software managing the three devices would be useful to help detect complications related to metabolic syndrome, diabetes, and cardiovascular disease and to noninvasively and rapidly manage treatment follow-up.

Keywords: fat mass, autonomic nervous system activity, Electro Sensor Complex, dual-energy X-ray absorptiometry, heart rate variability, and bioimpedance cardiography