Influence of the type of electrodes in the assessment of body composition by bioelectrical impedance analysis in the supine position

Dupertuis, Yves M.; Pereira, Amanda Gomes [UNESP]; Karsegard, Véronique L.; Hemmer, Alexandra; Biolley, Emma; Collet, Tinh-Hai; Genton, Laurence

Influence of the type of electrodes in the assessment of body composition by bioelectrical impedance analysis in the supine position

Data

2022-11-01

Autores

Dupertuis, Yves M.

Pereira, Amanda Gomes

Karsegard, Véronique L.

Tipo

Artigo

Resumo

Background & aims: The main source of error in body composition assessment of bedridden patients by bioelectrical impedance analysis (BIA) is the electrode inadequacy and placement. As electrocardiogram (ECG) electrodes are often used for BIA measurements, this study aimed to compare three of them with a reference BIA electrode. Methods: BIA was performed sequentially on 24 healthy subjects in the supine position, using 3 different ECG electrodes (3M® Red Dot® 2330; Ambu® BlueSensor 2300; Ambu® BlueSensor SU-00-C) and the reference electrode (Bianostic AT®) for the BIA device (Nutriguard-M®, Data Input, Germany). Resistance (R), reactance (Xc), phase angle (PhA), appendicular skeletal muscle index (ASMI), fat-free mass index (FFMI) and fat mass percentage (FM%) obtained with the different electrodes were compared using Bland–Altman plots, repeated measures one-way ANOVA and paired t-test. Patient characteristics potentially involved in BIA measurement differences were assessed using linear regression analysis. Results: The study population consisted of 9 men and 15 women, 33% and 47% of whom were overweight, respectively. The measured R was within the physiological range for all men (428–561 Ω) and women (472–678 Ω), regardless of the type of electrodes used. Compared to the reference electrode, the 3M® Red Dot® 2330 and Ambu® BlueSensor SU-00-C electrodes gave significantly different Xc and PhA values, but only the Ambu® BlueSensor SU-00-C gave significantly different ASMI, FFMI and FM% at 50 kHz, with biases of −0.2 kg/m2, −0.3 kg/m2 and +1.4%, respectively. The higher the current frequency, the lower was the Xc and PhA measured by the Ambu® BlueSensor SU-00-C compared to the reference electrode. These measurement differences seemed mainly due to the too small gel area of the Ambu® BlueSensor SU-00-C (154 mm2) compared to the reference electrode (1311 mm2). Conclusions: The use of electrodes with small gel area affects BIA measurement in the supine position, especially when PhA is used as an indicator of the nutritional status. Therefore, it is essential to specify the type of electrodes and carry out comparative tests before changing consumables for body composition assessment, to ensure BIA measurement reliability in clinical and research settings.