Researchers tested a new biopsy needle that can measure the electrical resistance (bioimpedance) of liver and tumor tissue during a liver biopsy. They found that liver and tumor tissue show different resistance at certain low and high frequencies, but the measurements varied a lot between patients and overlapped, so the method isn’t yet reliable for everyday use.

<i>Objective.</i>Liver biopsy is an essential procedure in cancer diagnostics but targeting the biopsy to the actual tumor tissue is challenging. Aim of this study was to evaluate the clinical feasibility of a novel bioimpedance biopsy needle system in liver biopsy and simultaneously to gather<i>in vivo</i>bioimpedance data from human liver and tumor tissues.<i>Approach.</i>We measured human liver and tumor impedance data<i>in vivo</i>from 26 patients who underwent diagnostic ultrasound-guided liver biopsy. Our novel 18 G core biopsy needle tip forms a bipolar electrode that was used to measure bioimpedance during the biopsy in real-time with frequencies from 1 kHz to 349 kHz. The needle tip location was determined by ultrasound. Also, the sampled tissue type was determined histologically.<i>Main results.</i>The bioimpedance values showed substantial variation between individual cases, and liver and tumor data overlapped each other. However, Mann-Whitney U test showed that the median bioimpedance values of liver and tumor tissue are significantly (p &lt; 0.05) different concerning the impedance magnitude at frequencies below 25 kHz and the phase angle at frequencies below 3 kHz and above 30 kHz.<i>Significance.</i>This study uniquely employed a real-time bioimpedance biopsy needle in clinical liver biopsies and reported the measured human<i>in vivo</i>liver and tumor impedance data. Impedance is always device-dependent and therefore not directly comparable to measurements with other devices. Although the variation in tumor types prevented coherent tumor identification, our study provides preliminary evidence that tumor tissue differs from liver tissue<i>in vivo,</i>and this association is frequency-dependent.