Effects of culinary treatments on the lipid nutritional quality of fish and shellfish

Highlights • High temperature and long heat treatment generally impairs EPA + DHA, n-3/n-6.• High temperature and long heat treatment cause hydrolysis of pro-atherogenic SFA.• Effects of frying on lipid nutritional quality depend on frying medium used.• The worse culinary treatment for fish and shellfish is frying using margarine.


Introduction
Coronary heart disease (CHD), also known as coronary artery disease, is a phenomenon where cholesterol deposits on the artery wall, causing the coronary arteries to become too narrow and unable to obtain the optimum oxygen supply.Since 1970, CHD has been the leading cause of global death and disability (Arroyo-Quiroz et al., 2020).Although the mortality rate of CHD in developed countries has continued to decline due to advances in prevention and treatment since the 1970s, it remains the largest contributor to mortality in most developed countries (GBD, 2018).In fact, the mortality of CHD has increased in most developing and undeveloped countries (Arroyo-Quiroz et al., 2020).To date, CHD is the third leading cause of mortality worldwide, with 17.8 million deaths annually (Brown et al., 2022).Early epidemiological studies found that Alaskan Natives and Greenland Eskimos who ate large amounts of seafood, had a lower mortality rate of CHD (Kromann and Green, 1980;Newman et al., 1993).Since then, many studies have been conducted to investigate the relationship between seafood intake and the incidence rate of CHD.Although some findings do not show a clear relationship between seafood consumption and the incidence rate of CHD (e.g.Bechthold et al., 2017;Hengeveld et al., 2018), many evidence have shown that high consumption of seafood contributes to the prevention of CHD (Kromhout et al., 2002;Zheng et al., 2012;He et al., 2004;Whelton et al., 2004;Zhang et al., 2020).
Seafood, especially fish and shellfish, is high quality animal protein that rich in polysaccharides (e.g.Tan et al., 2023a), carotenoids (e.g.Tan et al., 2022a) and omega-3 long-chain polyunsaturated fatty acids (n-3 LC-PUFA), especially eicosapentaenoic acid (EPA; C20:5n-3) and docosahexaenoic acid (DHA; C22:6n-3), have well-established beneficial properties for human health (Ajith and Jayakumar, 2018;Tan et al., 2021).Dietary guidelines recommend consuming two servings of seafood per week to prevent CHD (e.g.Arnett et al., 2019;Mohan et al., 2021).However, very low fish intake (<1 per week) is common in many regions, including America, Europe, Middle East, Southeast Asia and Africa (Stark et al., 2016;Hengeveld et al., 2018).In fact, a global dietary n-3 LC-PUFA survey based on EPA + DHA intake data revealed that>80% of the global population consumes<250 mg of n-3 LC-PUFA per day (minimum recommended value), while the global average intake of n-3 LC-PUFA is only 163 mg/day (Micha et al., 2014).Based on a meta-analysis, Stark et al. (2016) revealed that populations with access to adequate seafood (EPA + DHA in blood of > 8%) were limited to a few regions, including the Sea of Japan, Scandinavia, and areas with indigenous populations.
Culinary preparation is an important process for eliminating pathogenic microorganisms in food (e.g. Lee et al., 2008).However, at the same time, culinary treatment may lead to a decrease in the lipid nutritional quality of seafood, mainly due to heat-induced oxidation of n-3 PUFAs and the addition of exogenous oils that can modify the original fatty acid profile (e.g.Sampels, 2015;Biandolino et al., 2021).In addition, due to the limited accessibility for the world population in many regions to achieve optimal levels of fish and shellfish (Tan et al., 2020), monitoring the lipid nutritional quality of fish and shellfish during culinary preparation can serve as an alternative to monitoring the lipid nutritional quality of fish and shellfish for maximum CHD prevention effect.To date, most studies evaluating seafood consumption and CHD have focused on its relationship with the risk of CHD, and it is recommended to consume it once a week (e.g.Yang et al., 2016;Jayedi et al., 2018;Zhao et al., 2019;Zhang et al., 2020).However, the effects of culinary preparation of seafood on lipid nutritional quality is still unclear.
In this study, lipid profiles of fish and shellfish prepared with different culinary treatments were retrieved from the published literature to calculate various lipid nutritional quality indicators related to promoting or preventing CHD, including EPA + DHA, n-3:n-6, (MUFA + PUFA)/SFA-C18:0, atherogenic index (AI), thrombogenic index (TI) and hypocholesterolemic/hypercholesterolemic index (H/H).The changes (%) of these indicators in fish and shellfish prepared with different culinary preparations (relative to raw fish or shellfish) were then calculated and analyzed.To the best of our knowledge, this article represents the first comprehensive analysis of the effects of culinary treatments on the lipid nutritional quality of fish and shellfish in relation to the risk of CHD.The information in this article is very useful for understanding the effects of culinary preparation on the lipid nutritional quality of fish and shellfish.Such information will aid to provide guidance to consumers in choosing better culinary preparations to maximize the lipid nutritional quality of fish and shellfish, and to maximize the effectiveness of preventing CHD.

Literature search and data collection
Articles were obtained from Web of Science and Google Scholar (up to July 2022) using keywords such as "fatty acid profile of fish", "fatty acid profile of shellfish", "culinary preparation" and/or "cooking".In order to obtain any articles that may be missed in the online search, relevant articles in the reference list of each article were downloaded.This procedure was repeated until no new articles were found.As a result, a total of 3678 articles were obtained from the literature search.
All articles (n = 3678) were further screened for complete fatty acid profile table of raw and cooked fish or shellfish.After screening, only 76 articles met the requirements, and this comprehensive literature search and screening provided us with a reasonable number of fatty acid profiles of fish (n = 235) and shellfish (n = 69) to examine the effects of culinary treatment on the lipid nutritional quality of fish and shellfish in relation to the risk of CHD.
In order to verify the effect of culinary preparations on fatty acid content, the results were calculated on a dry weight basis.Moreover, in order to enable a comparison of the results from different studies, all data were expressed as a percentage of fatty acids to total fatty acids.

Data extraction and statistical analysis
From each fatty acid profile, we collected general information of the common name and scientific name of each analyzed species, the culinary treatment used, and the types of oil used if frying is involved.We also calculated various lipid nutritional quality indicators associated with prevention/promotion of CHD, including EPA + DHA (%), n-3:n-6, PUFA/SFA, (MUFA + PUFA)/SFA-C18:0, atherogenic index (AI), thrombogenic index (TI) and hypocholesterolemic/hypercholesterolemic index (H/H).
Omega-3 long-chain unsaturated fatty acids (n-3 LC-PUFAs), especially highly unsaturated ones (EPA and DHA) display several properties such as antithrombotic, antiinflammatory, antiarrhythmic and vasodilatory (Lombardo and Chicco, 2006).In addition, it is generally believed that consumption of high proportions of pro-inflammatory n-6 PUFA can lead to various adverse health effects (e.g. higher risk of cancer, ulcerative colitis), and equilibrate ratio of n-3 PUFA and n-6 PUFA can help in the prevention and treatment of many diseases (e.g hypertension and coronary artery problems) (Kinsella et al., 1990).Therefore, food rich in EPA and DHA with an n-3: n-6 ratio of > 0.45 (preferably > 1) are important indicators of healthy food to prevent CHD (Bhardwaj et al., 2016).
Saturated fatty acids (SFA) consumption, especially palmitic acid (C16:0) and myristic acid (C14:0), is associated with CHD by increasing low-density lipoprotein (LDL) cholesterol.As a result, since 1970, a reduction in SFA consumption has been recommended to reduce the risk of CHD (Mozaffarian et al., 2010).Therefore, PUFA/SFA has become an important indicator for assessing the risk of CHD and lipid nutritional quality of food.According to Health and Social Security, the recommended minimum value for the PUFA/SFA ratio is > 0.45, and ratios below this value are detrimental to human health (HMSO, 1994).Recently, many studies have shown that a diet rich in monounsaturated fatty acids (MUFA) has many health benefits, especially decreasing the risk factors of CHS (total cholesterol, LDL cholesterol and triglycerides) and increasing the high density lipoprotein cholesterol (Mozaffarian et al., 2010, Siri-Tarino et al., 2010).Considering the health benefits of MUFA and the fact that some SFAs (especially stearic acid (C18:0)) do not increase the risk of CHD, (MUFA + PUFA)/(SFA-C18:0) is another important lipid nutritional quality indicators assess the risk of CHD.
The atherogenic (AI) and thrombogenic indices (TI) were proposed by Ulbricht and Southgate (1991) as indicators of the stimulus potential of platelet aggregation, of which is used to evaluate the potential of food to influence the incidence of CHD.In particular, the low values of AI and TI suggest the higher levels of anti-atherogenic and anti-thrombogenic fatty acids and the greater potential to prevent the onset of CHD (Weber et al., 2008).As for the hypocholesterolemic/hypercholesterolemic (H/H) fatty acid ratio, the specific effects of fatty acids on the metabolism of lipoproteins transporting plasma cholesterol were considered, where lower values of HH are considered detrimental to human health.Therefore, a high HH value is desirable food with health benefits.AI, TI and H/H were calculated according to equations (1), ( 2) and (3), respectively (Ulbricht and Southgate, 1991;Santos-Silva et al., 2002): TI = (14:0 + 16:0 + 18:0)/((0.5 H/H = (18:1n-9 + 18:2n-6 + 20:4n-6 + 18:3n-3 + 20:5n-3 + 22:5n-3 + 22:6n-3)/(14:0 + 16:0) The changes (%) in these indicators in fish and shellfish prepared    with different culinary treatments (relative to raw fish and shellfish) were then calculated and analyzed.Prior to analysis, all data was translated (X + 1-min(X)) to ensure that the smallest number is 1, and then log 10 transformed.All statistical analyses were done on SPSS Windows Statistical Package TM (version 21), and significance for all analyses was set to P < 0.05 unless noted otherwise.Prior to analyses, all variables were tested for normality and variance homogeneity by Kolmogorov-Smirnov and Levene's tests, respectively.One-way ANOVA was performed followed by Turkey multiple comparison tests (Turkey HSD) to test for significant differences in EPA + DHA, n-3/n-6, PUFA/ SFA, (MUFA + PUFA)/(SFA-C18:0), AI, TI and H/H among raw and cooked fish and shellfish.

Results
Effects of culinary treatments on the EPA + DHA and n-3/n-6 ratio of fish and shellfish The effects of culinary treatment on EPA + DHA and n-3/n-6 of fish and shellfish are illustrated in Fig. 1.According to the study on the influence of culinary treatments on the EPA + DHA content (n = 177) and n-3/n-6 (n = 134) of fish, some culinary treatments (especially frying and braising) significantly impaired (P < 0.05) the EPA + DHA (Fig. 1A) and n-3/n-6 (Fig. 1B) of fish.The reduction of EPA + DHA and n-3/n-6 highly depends on the oil used (Fig. 2), among which margarine caused the greatest reduction (P < 0.05) in both EPA + DHA and n-3/n-6.In addition, frying fish with corn oil, soybean oil, rapeseed oil, canola oil and hydrogenated vegetable oil also significantly reduced (P < 0.05) the EPA + DHA content of fried fish (Fig. 2A).As for n-3/n-6, frying fish with sunflower oil, corn oil, palm oil, soybean oil and hydrogenated vegetable oil also cause significant reduction (P < 0.05) in n-3/n-6, but frying fish with canola oil caused significant increment (P < 0.05) in n-3/n-6 (Fig. 2B).
Similar to fish, the decreased of EPA + DHA and n-3/n-6 in fried shellfish is highly depends on the frying medium used, in which frying with sunflower oil caused the greatest reduction (P < 0.05) in EPA + DHA and n-3/n-6, followed by margarine, corn oil, olive oil and batter (P < 0.05) (Fig. 2C and 2D).

Effects of culinary treatments on the atherogenicity, thrombogenicity and hypocholesterolaemic/ hypercholesterolaemic indices of fish and shellfish
A total of 37, 99 and 99 studies associated with the effects of culinary treatment on the AI, TI and H/H of fish and shellfish are summarized in Fig. 5.In general, braising led to significant reduction (P < 0.05) in AI index, steaming, braising and graving caused significant reduction (P < 0.05) in TI index, and graving caused significant reduction (P < 0.05) in HH index.However, frying, braising and curry cooking significantly increased (P < 0.05) the HH index of fish.As for frying, frying with margarine and rapeseed caused significant increase (P < 0.05) and decrease (P < 0.05) of AI index, respectively (Fig. 6A).Frying with margarine, corn oil and hydrogenated vegetable oil significantly increased (P < 0.05) TI index, but frying with soybean oil, rapeseed oil and canola oil significantly decreased (P < 0.05) TI index (Fig. 6B).Frying with olive oil, soybean oil, rapeseed oil and canola oil significantly increased (P < 0.05) HH index, but frying with margarine significantly decreased (P < 0.05) HH index (Fig. 6C).
In shellfish, a total of 20, 31 and 31 studies addressed the effects of culinary treatments on AI, TI and H/H.In general, frying, microwave cooking and oven cooking significantly decreased (P < 0.05) the AI index of shellfish (Fig. 5D), grating significantly increased (P < 0.05) the TI index of shellfish (Fig. 5E), and microwave cooking, oven cooking and grating significantly increased (P < 0.05) the HH index of shellfish (Fig. 5F).As for frying, frying with sunflower oil and olive oil caused significant reduction (P < 0.05) in the AI index of shellfish (Fig. 6D).Frying with sunflower oil caused significant reduction (P < 0.05), but frying with olive oil, corn oil and margarine caused significant increment (P < 0.05) in the TI index of shellfish (Fig. 6E).Frying with corn oil and margarine caused significant reduction (P < 0.05), but frying with sunflower oil and olive oil caused significant increment (P < 0.05) in the HH index of shellfish (Fig. 6F).

Discussion
In general, studies have shown that high temperatures (e.g frying) and long period of heat treatment (e.g braising, roasting, smoking and curry cooking) during culinary preparation caused significant reduction in EPA + DHA and n-3/n-6 levels in fish and shellfish.EPA and DHA are very sensitive to oxidation, mainly due to the high degree of unsaturation (Weber et al., 2008;Tan et al., 2022bTan et al., , 2023b)).Oxidative degradation of LC-PUFA caused production of various radicals, -CHO, -CO and -CH3 that subsequently produce methylglyoxal, 2,3-butanedione, α,β-unsaturated aldehydes, α-dicarbonyl compounds and etc. (Ma et al., 2019).Therefore, it is not surprising that high temperature or prolong heat treatment during culinary preparation will lead to the oxidation of unsaturated fatty acids.Since EPA and DHA are the major fatty acids in fish and shellfish, the reduction in EPA and DHA also caused a significant reduction in the n-3/n-6 ratio.Interestingly, graving significantly increased the EPA + DHA content in fish.For example, the EPA + DHA content of (11.49%)Indian mackerel Bastrilliger kanagurta in gravy was significantly higher than that of raw mackerel (6.42%) (Marichamy et al., 2009), which may be associated with condensation of fish tissues and the release of EPA and DHA during cooking (Sampels, 2015).

Table 1
Effects of culinary treatments on the lipid nutritional quality of fish.It is worth noting that the reduction of EPA + DHA in fried fish and shellfish depends largely on the type of frying medium used, with the greatest reduction when frying with margarine and sunflower oil for fish and shellfish, respectively.Since different cooking oils contain different fatty acid compositions, it is not surprising that the fatty acid composition of fried fish and shellfish using different cooking oils change differently.This is mainly attributed to the absorption of oil and fatty acids in fish and shellfish during the frying process (Agren and Hamminen, 1993;Echarte et al., 2001;Gladyshev et al., 2006;Sioen et al., 2006;Weber et al., 2008;Bejaoui et al., 2019).For example, Sioen et al. (2006) compared the fatty acid profile of cod Gadus morhua submitted to frying using margarine and olive oil.It was found that both frying medium significantly reduce the EPA + DHA (from 49.61% to 9.83-12.27%)and n-3/n-6 (from 10.00 to 1.37 to 2.50) of Gadus morhua, with margarine caused significantly greater reduction than that of olive oil.Similarly, Weber et al. (2008) studied the changes in fatty acid profile of silver catfish Rhamdia quelen submitted to frying using different cooking oils (soybean oil, canola oil and hydrogenated vegetable oil).The results revealed that compared to Rhamdia quelen fried with canola oil, the reduction of EPA + DHA and n-3/n-6 in fish fried with hydrogenated vegetable oil and soybean oil was greater.Bejaoui et al. (2019) investigated the effect of three frying oils (corn, olive and margarine) on the lipid profile of clams (Venerupis decussata).The results revealed that frying with margarine caused the highest reduction in EPA + DHA (from 17.41 to 4.06%) and n-3/n-6 (from 5.30 to 0.56), followed by frying with corn oil (EPA + DHA = 10.27%;n-3/n-6 = 1.29) and olive oil (EPA + DHA = 12.02%; n-3/n-6 = 1.62).These observations indicate that frying fish and shellfish with cooking oils low in unsaturated fatty acids (e.g.margarine) results in significantly higher reduction of EPA + DHA and n-3/n-6 than using cooking oils rich in unsaturated fatty acids (e.g.olive oil).It is worth noting that although the culinary treatments reduced the n-3:n-6 ratio, except for some fried fish (Candela et al., 1998;Hosseini et al., 2014) and shellfish samples (Ghribi et al., 2017;Bejaoui et al., 2019;Biandolino et al., 2021), the n-3/n-6 of most cooked marine fish and marine shellfish is still higher than the recommended standard, indicating that marine fish and shellfish are still good for consumption after culinary preparation.
In general, none of the culinary treatment can significantly increase the AI index of fish and shellfish, indicating that cooking does not increase the risk of CHD.In fact, some culinary treatments decrease the AI (improve health benefits) of fish and shellfish.This is mainly attributed to high level of heat induced hydrolysis of pro-atherogenic SFA (C12:0, C14:0 and C16:0) and an increase of MUFA (e.g.Marichamy et al., 2009;Domiszewski et al., 2011;Merdzhanova et al., 2018;Brito et al., 2019;Biandolino et al., 2021).As for frying, frying with margarine and rapeseed oil caused a significant increase and decrease in AI in fried fish, mainly due to the high and low saturation of the frying medium, respectively (Türkkan et al., 2008;Marichamy et al., 2009;Alexi et al., 2019).For bivalves, frying with sunflower oil and olive oil significantly reduced AI index.Caution through, the AI index of most studies cannot be calculated because the content or composition of C12:0 was not provided.Therefore, more data are needed to further confirm the effects of frying medium on AI of seafood.
Similar to AI, some culinary treatments (e.g.steaming, braising and graving) reduce the TI of fish, indicating that these culinary treatments are more healthy through increasing the content of anti-thrombogenic fatty acids (e.g. de Castro et al., 2007;Marichamy et al., 2009;Koubaa et al., 2012;Brito et al., 2019).In terms of frying, most cooking oils can increase the TI of fish and shellfish, while frying with soybean oil, rapeseed oil and canola oil reduces the TI of fish, as well as frying with sunflower oil reduced the TI of shellfish.This is due to the fact that the absorption rate of cooking oil by food is highly dependent on the viscosity of cooking oil (Yang et al., 2020), of which the viscosity of soybean oil and canola oil is lower than other vegetable oils (Sahasrabudhe et al., 2017).In fact, Zotos et al. (2013) have shown that sunflower oil can penetrate into the flesh of anchovy Engraulis encrasicholus even within 3 min of frying.Therefore, the high absorption of cooking oils (rich in n-6) consequent in lower n-3/n-6 ratio, which helps to reduce TI in cooked seafood.As regards to H/H, many culinary treatments can improve the H/H of fish (frying, braising and curry cooking) and shellfish (microwave cooking, oven cooking and grating), indicating a lower risk of cardiovascular diseases.The increased in H/H is mainly attributed to the decrease in C14:0 and C16:0 (e.g.Hosseini et al., 2014;Ghribi et al., 2017;Alexi et al., 2019;Felici et al., 2019;Brito et al., 2019;Bejaoui et al., 2019;Golgolipour et al., 2019;Biandolino et al., 2021).
Taken together, braising, curry cooking, graving, canola oil frying, rapeseed oil frying and soybean oil frying are recommended culinary treatments for fish (Table 1).For shellfish, microwave cooking, oven cooking, olive oil frying and sunflower oil frying are recommended.It is worth noting that although most recommended culinary treatments caused a reduction of EPA + DHA and n-3/n-6, marine fish and shellfish are rich in EPA + DHA and have a high n-3/n-6 ratio, in which cooked fish and shellfish still maintain a high levels of EPA + DHA and the n-3/ n-6 ratio higher than the recommended value of > 0.45 (Bhardwaj et al., 2016).Since different culinary treatments have their own advantages and disadvantages, consumers can choose culinary treatments based on their personal interest in specific fatty acids/ indices.On the other hand, the worse way to cook fish and shellfish is frying using margarine.This is due to the fact that it results in the greatest reduction in EPA + DHA, n-3/n-6, PUFA/SFA, (MUFA + PUFA)/SFA-C18:0, AI, TI and H/H of fish and shellfish, indicating an increased risk of CHD.In addition, it is also not recommended to fry shellfish with corn oil, as this will increase the pro-thrombogenic saturated fatty acids and cause a major negative impact on the lipid nutritional quality of shellfish (Table 2).

Conclusion
In a nutshell, high temperature and long-term heat treatment during culinary preparation reduced EPA + DHA and n-3/n-6 of fish and shellfish, but might increase PUFA/SFA and (MUFA + PUFA)/SFA-C18:0 through induced hydrolysis of pro-atherogenic SFA.Frying has the most dynamic impact on the lipid nutritional quality of fish and shellfish, which largely depends on the frying medium used, mainly attributed to exchange of fatty acids between cooking oil and seafood.Among culinary treatments, braising, curry cooking, graving, canola oil frying, rapeseed oil frying and soybean oil frying are highly recommended in cooking fish, whereas microwave cooking, oven cooking, olive oil frying and sunflower oil frying are recommended for cooking shellfish.The worse culinary treatment for fish and shellfish is frying using margarine.
K. Tan et al.

Fig. 1 .
Fig. 1.Effects of culinary treatments on EPA + DHA and n-3/n-6 of fish (A and B) and shellfish (C and D).Different letters indicate significant difference (P < 0.05).

Fig. 2 .
Fig. 2. Effects of cooking mediums on EPA + DHA and n-3/n-6 of fried fish (A and B) and shellfish (C and D).HV = hydrogenated vegetable oil and different letters indicate significant difference (P < 0.05).
Fig. 6.Effects of cooking mediums on the atherogenicity (AI), thrombogenicity (TI) and hypocholesterolaemic/ hypercholesterolaemic (HH) indices of fried fish (A, B and C) and shellfish (D, E and F).HV = hydrogenated vegetable oil and different letters indicate significant difference (P < 0.05).

Table 2
Effects of culinary treatments on the lipid nutritional quality of shellfish.