BackgroundUnderstanding different patterns of fish consumption is an important component for risk assessment of contaminants in fish. A few studies on food consumption had been conducted in Malaysia, but none of them focused specifically on fish consumption. The objectives of this study were to document the meal pattern among three major ethnics in Malaysia with respect to fish/seafood consumption, identify most frequently consumed fish and cooking method, and examine the influence of demographic factors on pattern of fish consumption among study subjects.MethodsA cross-sectional survey was conducted between February 2008 and May 2009 to investigate patterns of fish consumption among Malaysian adults in Peninsular Malaysia. Adults aged 18 years and above were randomly selected and fish consumption data were collected using a 3-day prospective food diary.ResultsA total of 2,675 subjects, comprising male (44.2%) and female (55.7%) participants from major ethnics (Malays, 76.9%; Chinese, 14.7%; Indians, 8.3%) with a mean age of 43.4±16.2 years, were involved in this study. The results revealed 10 most frequently consumed marine fish in descending order: Indian mackerel, anchovy, yellowtail and yellow-stripe scads, tuna, sardines, torpedo scad, Indian and short-fin scads, pomfret, red snapper, and king mackerel. Prawn and squid were also among the most preferred seafood by study subjects. The most frequently consumed freshwater fish were freshwater catfish and snakehead. The most preferred cooking style by Malaysians was deep-fried fish, followed by fish cooked in thick and/or thin chili gravy, fish curry, and fish cooked with coconut milk mixed with other spices and flavorings. Overall, Malaysians consumed 168 g/day fish, with Malay ethnics’ (175±143 g/day) consumption of fish significantly (p<0.001) higher compared with the other two ethnic groups (Chinese=152±133 g/day, Indians=136±141 g/day).ConclusionFish consumption was significantly associated with ethnicity, age, marital status, residential area, and years of education of adults in Peninsular Malaysia, and the data collected are beneficial for the purpose of health risk assessment on the intake of contaminants through fish/seafood consumption.

Positional distribution of fatty acids in hoki and tuna oils was carried out in the present study via pancreatic lipase hydrolysis and 13 Carbon nuclear magnetic resonance ( 13 C NMR) spectroscopy. The two methods gave consistent results for hoki but conflicting values for tuna oil. Saturated fatty acids (SFA) were generally enriched at sn-2 while MUFA were preferentially located at sn-1,3 for hoki by pancreatic lipase treatment. Of the PUFA, DHA in hoki oil was significantly located at sn-2 position, whereas eicosapentaenoic acid (EPA) was randomly distributed via both methods. DHA and EPA in tuna oil were both preferentially located at sn-2 position by NMR spectroscopy. The NMR results for n-3 PUFA were in general agreement with other fish oils studied where DHA showed preference for sn-2 but EPA displayed preferential differences among species. The ratio of DHA in hoki oil at sn-2 position was higher than tuna oil, thus enhancing its bioavailability. The opposite was the case for EPA. For pancreatic lipase hydrolysis of tuna oil, results showed that both DHA and EPA were incompletely recovered along with other PUFA confirming that pancreatic lipase hydrolysis is an unreliable method for positional distribution determination in tuna oil compared to 13 C NMR analysis. The ratio of long chain PUFA to triglycerides exceeds 1:1 in tuna oil, causing steric hindrance for hydrolysis of the long chain PUFA residue at sn-1,3 position.Practical applications: Fish oil contains bioactive omega-3 fatty acids (DHA and EPA), however bioavailability and oxidative stability of these omega-3 fatty acids depends on their positional distribution. Pancreatic lipase has been used in determination of positional distribution of fatty acids in fish oil, but this method involves heat and is time consuming and not suitable for fish oil with more than one molecule of PUFA in a single triglyceride. NMR has gained popularity in positional distribution determination of n-3 fatty acids in fish oil as it is a rapid and non-destructive method compared to the conventional chemical and enzymatic methods.

Enrichment of the omega-3 (n-3) fatty acids of refined hoki oil (RHO) intact triglycerides (TG) and via free fatty acids (FFA), was carried out in the present study using established methods of dry fractionation (DF), low temperature solvent crystallization (LTSC) and urea complexation (UC) and positional distribution of fatty acids in the intact TG was determined by Nuclear Magnetic Resonance (NMR) analysis. Results showed that n-3 fatty acids were enriched in liquid fractions of all methods except DF, where the highest concentration was obtained via the UC method (83.00 %). The FFA form of the oil produced a higher concentration (40.81 %) of n-3 fatty acids via the LTSC method compared to the TG form (31.50 %). The percentages of the total saturated fatty acid (SFA) in the liquid fractions in all methods were lower, ranging from 1.60 % (UC) to 21.44 % (DF) compared to the RHO parent oil (24.05 %). The percentages of total monounsaturated fatty acids (MUFA) in the liquid fractions were similar to the solid fractions except for the UC method where total MUFA was six times higher in the solid fraction. In LTSC-FFA and UC methods, the enrichment factor for EPA was lower, ranging from 1.61 (LTSC-FFA) to 2.83 (UC), than DHA which ranged from 1.64 (LTSC-FFA) to 3.88 (UC). EPA was preferentially located at the sn-1,3 position and DHA was significantly located at the sn-2 position which is the favoured location for intestinal digestion.

Physicochemical characterisation and oxidative stability of refined hoki oil, unrefined hoki oil and unrefined tuna oil were carried out in the present study. Tuna oil contains a higher percentage of polyunsaturated fatty acids (42.57%) than the hoki oils (28.79-30.13%), which have higher percentages of monounsaturated fatty acids (45.02-47.16%). All oils showed a good ratio of n-3 to n-6 fatty acid (7.01-8.10). Cholesterol contents in the unrefined hoki (5149.40 lg g À1 ) and tuna (2045.48 lg g À1 ) oils were higher than the refined hoki oil (1411.27 lg g À1 ). Tuna has a higher concentration of natural a-tocopherol (752.49 lg g À1 ) but lower concentration of vitamin A (110.99 lg g À1 ) than unrefined hoki oil (151.44 lg g À1 and 997.60 lg g À1 , respectively). Higher percentages of unsaponifiable matter were found in the hoki oils (4.90-7.24%) compared with the tuna oil (0.56%). The hoki oils appear more yellow than the tuna oil, which is darker by comparison. Moisture, p-anisidine value and free fatty acid contents in the hoki oils were lower than the tuna oil. Other indicators of oxidative stability showed that the hoki oils were more stable than the tuna oil. International Journal of Food Science and Technology 2013Characterisation of hoki and tuna oils T. M. Tengku-Rozaina and E. J. Birch Mean AE SD; nd, not detected (<0.05%). Statistical analysis was carried out using mutivariate analysis, Tukey's test and paired-sample t-test abcd Values with different superscript letters within a row are significantly different at P < 0.05. xyz Values with different superscript letters within a column are significantly different at P < 0.05.