Bioimpedance-based Authentication of Defrosted Versus Fresh Pork at the End of Refrigerated Shelf Life

Daniel E Osen; Sisay Mebre Abie; Ørjan G Martinsen; Bjørg Egelandsdal; Daniel Münch

doi:10.2478/joeb-2022-0017

Bioimpedance-based Authentication of Defrosted Versus Fresh Pork at the End of Refrigerated Shelf Life

J Electr Bioimpedance. 2023 Jan 14;13(1):125-131. doi: 10.2478/joeb-2022-0017. eCollection 2022 Jan.

Authors

Daniel E Osen¹, Sisay Mebre Abie^{1

2}, Ørjan G Martinsen^{1

3}, Bjørg Egelandsdal⁴, Daniel Münch^{5

2}

Affiliations

¹ Department of Physics, University of Oslo, 0316 Oslo, Norway.
² Faculty of Ecology and Natural Resource Management, Norwegian University of Life Sciences, 1432 Ås, Oslo, Norway.
³ Department of Clinical and Biomedical Engineering, Oslo University Hospital, 0372 Oslo, Norway.
⁴ Faculty of Chemistry, Biotechnology and Food Science, Norwegian University of Life Sciences, 1432 Ås, Oslo, Norway.
⁵ Animalia, Norwegian Meat and Poultry Research Center, 0513 Oslo, Norway.

Abstract

Correct food labeling is a legal requirement and helps consumers to make informed purchasing choices. Mislabeling defrosted meat as fresh is illegal in the EU. However, there are no standardized technologies to authenticate fresh versus defrosted meat. We address this by testing if bioimpedance-based measurements can separate defrosted meat from refrigerated-only meat at the end of shelf life, i.e., when also fresh meat shows deterioration. Pork sirloin samples from 20 pigs were first tested at 12 days postmortem ('fresh group'). This time point was chosen to represent a typical use-by date for refrigerated storage of fresh pork. Then, all samples were transferred to a -24°C freezer for 3 days and thawed for 2 days before final testing ('frozen-thawed group'). Bioimpedance analyses (BIA) were done in a frequency range of [10²-10⁶ Hz]. Weight, pH and electrode positioning were assessed to test for potential confounding effects. Statistics for treatment dependent differences were based on the established P_y parameter and phase angle, which were extracted from the BI spectra. We found that using bioimpedance testing with tetrapolar electrodes, P_y and phase angle allowed almost complete separation of fresh and previously frozen samples. However, within the whole sample population, there was some overlap between the spectra of fresh and frozen samples. Yet, based on P_y, only one fresh sample (5% of N_total=20) fell in the lowest P_y class with all the frozen samples. We used a multifactorial design that allowed to test the effects of potential confounding factors, such as electrode positioning and meat quality parameters. We found a relatively low explained variance for the P_y parameter, indicating that confounding effects from other factors or quality defects in fresh pork may affect the detection capacity of bioimpedance-based authentication of fresh pork. Our data, therefore, suggest that reliable fresh-label authentication with bioimpedance testing should be based on testing a small number of samples to represent a specific lot of pork that is to be inspected.

Keywords: Bioimpedance; Py-parameter; food labelling; freezing; meat; phase angle; thawing.