Post-mortem Studies of Fish Using Magnetic Resonance Imaging

Abstract

Although being less sensitive than many other analytical methods, magnetic resonance imaging (MRI) does have significant advantages such as being non-invasive and non-destructive. Furthermore, no sample preparation is needed prior to analysis and the method is not hampered by limitations regarding signal penetration depth. However, due to the high investment costs, the sheer size of the instrument and the necessary related infrastructure, MRI cannot presently be considered as a standard analytical tool in aquaculture or fish processing industry. As a research tool, however, taking advantage of the unique features of the method, we can obtain basic insight into a number of issues related to anatomical studies, composition and structure of tissues, distribution maps of fat, water, and salt as well as temperature profiles (mapping). Moreover, theoretical transport models can in turn be used to interpret the images. For the aquaculture industry, MRI studies may for instance be helpful to study the effect of feed composition (fat) and different feeding regimes on fat contents and distribution in tissues. In fish processing, MRI can be used as a tool for the optimization of various unit operations such as salting, freezing, and thawing. For a more detailed treatment of various MRI applications, see review by Hills [1]. MRI is a technique that offers a unique opportunity to “look inside” intact whole organisms, i.e. to produce high-quality cross-section images. Depending on the particular task, MRI instruments can produce several types of contrast images. This is achieved by programming and running specific MR sequences to make it possible to differentiate between protons in molecules having different mobilities or chemical environments. For example, it is possible to obtain MR images of “water” and “fat” [2], “diffusion weighted” images where only molecules with low mobility are visible [3] or high resolution images of connective tissue [4]. Contrast agents—administered to the circulatory system—added to affect relaxation times are often used in connection with medical in vivo examinations to obtain a specific MRI contrast [5]. MR images of water and fat in burbot (Lota lota) liver have been acquired. The MR signal intensity in the reconstructed images showed good correlation with the liver fat content in live fish. Significant correlations were also obtained for water and protein contents [2]. Collewet et al. [6] proposed to use MRI as a tool to assess fat distribution in fish. It was shown that higher contrast between muscle and adipose tissues can be obtained when the MR images are highly longitudinal relaxation time (T1) weighted. Moreover, based on comparison with chemical analysis of fat, better correlations with quantitative MR data were obtained by accounting for coil sensitivity and radio frequency (RF) field inhomogeneity. Compared with fresh cod (Gadus morhua), Howell et al. [7] reported that MR images of fish stored for 6 weeks at−30◦C were not different. In contrast, images obtained from fish stored at −8◦C exhibited dense lines indicative of tissue gaps. Fresh rainbow trout (Salmo gairdneri) was subjected to freezing– thawing before MRI analysis to visualize various organs, and to identify spatial distributions of lipidand collagenrich tissues. T1-weighed images were clearly shown to highlight the lipids relative to water containing tissues. On the other hand, transverse relaxation time (T2) and magnetization transfer (MT) contrasted images showed the clearest distinction between muscle tissues [8]. The effect of freezing and thawing of a lean species (cod) and a fatty species (mackerel, Scomber scombrus) was studied by Nott et al. [9]. In particular, the T sat 1 (measured during saturation transfer) and MT rate were shown to be sensitive parameters to assess the effects of frozen storage. Compared with fresh fish, the freezing–thawing cycle of both species induced increased MT rate and T1, whereas T sat 1 decreased. The same effect was consistently observed with increasing frozen storage time (2–12 weeks). Foucat et al. [10] used a cylindrical sample carrier in the MRI magnet allowing six frozen–thawed trout (frozen storage for 0–41 days) and reference tubes to be analyzed simultaneously. An image containing information about anatomical features, MT ratio, T2, and diffusion constants (parallel and perpendicular to muscle fiber orientation, D⊥) were acquired in 5 min. Compared with fresh fish, the freezing– thawing cycle significantly affected only the mean T2 and the D⊥-values, which changed from 42.3 to 47.1 ms and