This study presents electrical measurements (both conductivity during the pulses and impedance spectroscopy before and after) performed in liver tissue of mice during electroporation with classical electrochemotherapy conditions (8 pulses of 100 µs duration). A four-needle electrode arrangement inserted in the tissue was used for the measurements. The undesirable effects of the four-electrode geometry, notably concerning its sensitivity, were quantified and discussed showing how the electrode geometry chosen for the measurements can impact the results. Numerical modelling was applied to the information collected during the pulse, and to the impedance spectra acquired before and after the pulses sequence. Our results show that the numerical results were not consistent, suggesting that other collateral phenomena not considered in the model are at work during electroporation in vivo. We show how the modification in the volume of the intra and extra cellular media, likely caused by the vascular lock effect, could at least partially explain the recorded impedance evolution. In the present study we demonstrate the significant impact that physiological effects have on impedance changes following electroporation at the tissue scale and the potential need of introducing them into the numerical models. The code for the numerical model is publicly available at gitlab.inria.fr/poignard/4-electrode-system.