Skip to main content

The use of the so‐called ‘superchilling’ technique for the transport of fresh fishery products

EFSA Journal logo
Wiley Online Library

Meta data

Abstract

Superchilling entails lowering the fish temperature to between the initial freezing point of the fish and about 1–2°C lower. The temperature of superchilled fresh fishery products (SFFP) in boxes without ice was compared to that of products subject to the currently authorised practice in boxes with ice (CFFP) under the same conditions of on‐land storage and/or transport. A heat transfer model was developed and made available as a tool to identify under which initial configurations of SFFP the fish temperature, at any time of storage/transport, is lower or equal to CFFP. A minimum degree of superchilling, corresponding to an ice fraction in the fish matrix of SFFP equal or higher than the proportion of ice added per mass of fish in CFFP, will ensure with 99–100% certainty (almost certain) that the fish temperature of SFFP and the consequent increase of relevant hazards will be lower or equal to that of CFFP. In practice, the degree of superchilling can be estimated using the fish temperature after superchilling and its initial freezing point, which are subject to uncertainties. The tool can be used as part of ‘safety‐by‐design’ approach, with the reliability of its outcome being dependent on the accuracy of the input data. An evaluation of methods capable of detecting whether a previously frozen fish is commercially presented as ‘superchilled’ was carried out based on, amongst others, their applicability for different fish species, ability to differentiate fresh fish from fish frozen at different temperatures, use as a stand‐alone method, ease of use and classification performance. The methods that were considered ‘fit for purpose’ are Hydroxyacyl‐coenzyme A dehydrogenase (HADH) test, α‐glucosidase test, histology, ultraviolet–visible–near–infrared (UV‐VIS/NIR) spectroscopy and hyperspectral imaging. These methods would benefit from standardisation, including the establishment of threshold values or classification algorithms to provide a practical routine test.