Background: Bioelectrical impedance analysis (BIA) is a good alternative for assessing total body water in newborns. However, the lack of standardization of the method for its use in the neonatal population raises doubts about its reliability and external validation. Objective: To compare resistance (R) and reactance (Xc) values measured by BIA with electrodes placed in the anatomical position and with electrodes placed at 5-cm intervals.Method: Prospective, randomized, crossover study, in which R and Xc values were measured using monofrequency BIA. The choice of the first position of the electrodes to be analyzed in each NB was performed by a simple randomization, and subsequent measurements were performed in rapid succession. The calculated sample size was 203 evaluations. The paired t test was used to compare the means of R and Xc, and the Bland-Altman plot and linear regression were used to analyze how much the measured values were influenced by the electrode position.Results: Significant differences were found in the R (855. 25 ± 147.79 vs 618.95 ± 93.28 Ω) and Xc (50.94 ± 17.34 vs 39.76 ± 18.35 Ω) values measured with electrodes in the anatomical position and at 5-cm intervals, respectively.The electrodes in the anatomical position produced R and Xc values 37.7% and 16.6%, respectively, higher on average when compared with 5-cm intervals. Conclusions:The distance between the electrodes influences the measurement of R and Xc by BIA. These results highlight the importance of defining a standard for the placement of electrodes in newborns for the use of these measures in neonatal clinical practice.