Determination of eicosapentaenoic acid and docosahexaenoic acid contents and the oxidation level of fish oil supplements from Bahrain market

Abstract Eicosapentaenoic acid and docosahexaenoic acid are two types of omega-3 fatty acids that provide numerous health benefits. It is thus important to gain the recommended daily intake of eicosapentaenoic acid and docosahexaenoic acid and compensate for omega-3 deficiency through supplements. However, supplements are very susceptible to oxidation, which affects their biological significance. This study aimed to examine the quality of fourteen different commercially available fish oil supplements in Bahrain markets by comparing their actual and labelled Eicosapentaenoic acid and docosahexaenoic acid contents, determination of their oxidation levels by measuring the peroxide value, anisidine value, and the total oxidation value. Fish oils were methylated, separated, and identified by Gas Chromatography. Measurement of peroxide value was conducted using titration, and the anisidine value using spectrophotometry. The total oxidation value was calculated based on anisidine value and peroxide value values. Results showed that most brands have lower actual total fish oil per capsule than the labelled values. Only two brands (14.29%) showed an approximately similar amount of Eicosapentaenoic acid and docosahexaenoic acid, whereas six brands (42.86%) have a lower actual amount of both Eicosapentaenoic acid and docosahexaenoic acid. In addition, five brands (35.71%) have higher actual EPA and lower DHA values. Only one brand has higher actual amounts of both Eicosapentaenoic acid and docosahexaenoic acid. The results also showed that 57.1% of the tested supplements (8 out of 14) exceeded the recommended levels of peroxide value, while only one brand (7.14%) exceeded the safe levels of anisidine value. Moreover, five brands (35.71%) recorded a total oxidation value level higher than the recommended. Compared to other studies in several countries, the oxidative levels of supplements in Bahrain are not very high; nonetheless, it must be improved by applying oxidation levels regulations and storage regulations regarding fish oil supplements.


Introduction
Eicosapentaenoic acid (EPA), and docosahexaenoic acid (DHA) are classified as one of the essential polyunsaturated fatty acids (PUFAs).Human intake of these fatty acids depends on the health status of the individual; as a result, the American Heart Association (AHA) recommends that people suffering from heart diseases should take 1000 mg of EPA and DHA per day from the consumption of oily fish or supplements (Covington, 2004;Ghasemi Fard et al., 2019;Kris-Etherton et al., 2000;Kris-Etherton, Harris, and Appel, 2002;McDonnell, French, Baggerly, & Harris, 2019).Whereas the 2015-2020 Dietary Guidelines for Americans by the United States Department of Agriculture (USDA) and Health and Human Services (HHS) recommended at least two servings of $8 oz of a variety of seafood per week for the general population for cardiovascular health (U.S.Department of Agriculture & U.S. Department of Health & Human Services, 2020).
Fish consumption has been shown to reduce the risk of cardiovascular disease and cancer due to its high content of omega-3 PUFAs specifically EPA and DHA.The levels of EPA and DHA differ between fish species and even within the same species due to several environmental factors (Freije, 2017;Musaiger & D'Souza, 2011;Simopoulos, 1999aSimopoulos, , 1999b)).
Given that consuming enough levels of omega-3 fatty acids through diet is not always achievable, demand for omega-3 fatty acid supplements has risen dramatically (Opperman, Marais, & Spinnler, 2011;Opperman & Benade, 2013).The study of Al-Arrayed, Al Maskati, and Abdullah (1999) has shown that the level of EPA þ DHA in local fish from Bahrain is much lower (0.03-0.239 g/3 oz) than that reported by the United States Department of Agriculture (USDA) Nutrient.Based on the study of Freije (2009) it was also concluded that it is difficult to attain the required daily amount of 1000 mg of EPA þ DHA through the consumption of local fish in Bahrain, therefore fish oil supplements can be considered as an alternative option.
Oxidation is the natural process that occurs in the unsaturated fatty acids detected in oils and fats.When the unsaturated fatty acids are exposed to oxygen from the atmosphere, oxygen interacts with the double bond in unsaturated fatty acids.Every lipid is susceptible to oxidation, whether fish oil capsules or cooking oils.Oxidation results in oil rancidity, causing an odor or a fishy taste in the fish oil supplements (Ismail, Bannenberg, Rice, Schutt, & Mackay, 2016;Kolanowski, 2010).
So far, oxidation has no disadvantageous health effects.Nonetheless, oxidative products can cause oxidation and inflammation in the body tissues, which makes researchers concerned that it might bring about cancer or cardiovascular problems.Although oxidized lipids are produced during inflammation, ingesting them from rancid oil will not have the same effect.Anyhow, human nature has an innate avoidance of rancid food, which limits its intake (Ismail et al., 2016).
The main parameters used to measure fish oil oxidation are the para-anisidine value (AV) and the peroxide value (PV).PV measures the primary oxidation level, while AV measures the secondary.An easy method for checking rancidity is by tasting or smelling the product.The measurement of the amount of peroxide found in oil is called PV. Peroxides are the primary products created when unsaturated fatty acids are oxidized, and as the oxidation process increases, the PV increases.However, when peroxides further react the PV starts to decrease.Therefore, reduced PV is not an indicator of the oil quality.Additional measurement of the secondary oxidation is needed.After peroxides are formed, they undergo further oxidation and produce secondary oxidation products.For measuring the secondary products, para-anisidine reacts with the oil sample, and then the absorbance is measured at a specific wavelength (350 nm).The test mainly calculates the existence of 2,4-alkadienals and 2-alkenals (Ismail et al., 2016).For a comprehensive analysis of the amount of oxidation, the total oxidation value (Totox) is calculated using the following equation (2PV þ AV).Authorities like the Global Organization for EPA and DHA Omega-3s (GOED) have set an upper limit of oxidation in fish oil.The maximum recommendation equals AV ¼ 20 meq/kg, PV ¼ 5 meq/kg, and Totox ¼ 26 meq/kg (Cameron-Smith, Albert, & Cutfield, 2015).
To our knowledge, the present study is the first attempt to provide information about the quality of fish oil supplements that contain 1000 mg of EPA and DHA regarding their content and oxidation level.Thus, the present study would fulfill the existing knowledge gap, and most probably assure the quality standard of fish oil supplements available in Bahrain available in Bahrain.
This study aimed to compare the actual EPA and DHA contents of fish oil supplements from the Bahrain market with those stated on the labels; to measure their peroxide values (PV), anisidine values (AV), and calculate the total oxidation values (Totox).In addition, investigate whether the oxidation values met or exceeded the internationally recommended indices of oxidation markers and thus evaluate the quality of fish oil supplements based on their EPA þ DHA contents and oxidation levels.

Material and methods
Fourteen brands of fish oil supplements were purchased from retail stores in Bahrain.Supplements were selected with an expiry date of at least 18 to 24 months from the time of purchase.Those brands were selected from the many brands available in Bahrain that contained only EPA þ DHA (1000 mg) from fish without additives or flavors.The brand codes, supplements price, price per capsule, country of origin, best-before-date, number of capsules, and labelled EPA þ DHA contents per capsule were documented.The capsules were stored at 4 C to minimize the capsules' oxidation level.
The fish oil contents were aspirated by syringes and needles immediately before analysis.The oil contents of 10-30 capsules were combined to obtain pooled fish oil sample (20 ml), which was used to measure EPA, DHA, PV, AV, and Totox.Four replicates were used for each measurement, and the intra-assay relative coefficient of variation (CV %) for all samples tested was calculated.Inter-assay CVs% of <15 are generally acceptable, whereas intra-assay % CVs should be <10 (Ramljak et al., 2013).

Fatty acids methylation
Fatty acids methylation was done following the AOAC method (1990), with some modifications demonstrated by Ozogul, Sims ¸ek, Balikc ¸i, and Kenar (2012).Twenty-five mg, equivalent to (twenty-five ml) of aspirated supplemental fish oil, were mixed with 1.5 ml of 0.5 M methanolic sodium hydroxide.The mixture was heated at 100 C for 7 min and cooled down at room temperature.Two ml of 14% Borontrifloride methanol (CH4BF3O) were added, and the mixture was heated at 100 C for 5 min and was cooled to 30-40 C. One ml of isooctane was dropped over the mixture, and the test tube was shaken for 30 s. Five ml of saturated sodium chloride was added to the mixture and shaken again for 30 s.The contents of the mixture were allowed to separate into two layers, in which the upper layer, which contains isooctane with fatty acid methyl esters (FAMEs), was removed carefully and placed in a clean test tube while the lower layer was extracted again with one ml of isooctane, shaken for the third time for 30 s.The two isooctane extracts were combined, dried using a tiny amount of anhydrous sodium sulphate, and finally concentrated to one ml under nitrogen gas.

Fatty acid analysis
The analysis was done by Perkin Elmer, Clarus 500 GC-FID equipped with a thermal conductivity detector (TCD) and flame ionization detector (FID) (Al-Asheeri, Freije, & Perna, 2020).The FAMEs were separated by fused carbon-silica column (crossbond, polyethylene glycol, carbowax, stabilwax), with 0.25 mm diameter and 30 m length, the temperature range between 40-260 C and 0.25 mm film thickness.The oven temperature was 200 C and held constant for 80 min for each sample, the FID temperature was 300 C, and TCD temperature was 150 C, and the injection temperature was constant at 250 C with a split ratio equal to 1:20.The carrier gas was nitrogen, and its flow rate was 0.76 ml/min.The injection volume was 5 ml with three pre and post-sample washes.The rate of sampling was 12.5 Hz, and the FAMEs were identified depending on their peak area against authentic standards (PUFA No. 1 and No.2), which SUPELCO USA provided.

PV and AV measurement
PV and AV measurements were done following the European Pharmacopoeia (Ph.Eur.) method ( 2004), as described by Albert et al. (2015).

PV measurement
Measurement of PV was carried out using the iodine-visual titration technique.Fish oil (2.5 g) was weighed, 50 ml of 3:2 (glacial acetic acid:trimethylpentane) and 500 ml of saturated potassium iodide solution were added to the flask.The mixture was shaken thoroughly for 60 s before 30 ml of water was added.Subsequently, the formed yellow solution was titrated with 0.01 N sodium thiosulphate solution.When the solution became almost colorless, 1% of starch solution (500 ml) was added.The titration with 0.01 N sodium thiosulphate solution continued until the solution became totally colorless.The same procedure, yet without adding fish oil, was used to measure Vcontrol.Using the formula "PV ¼ [10 Â (V À Vcontrol)]/m", the peroxide value was obtained in meq/kg, where m refers to the oil mass that was weighed within the volumetric flask and V and Vcontrol refers to the sodium thiosulphate volume that was titrated.

AV measurement
Fish oil (0.2 g) was weighed within a small vial and 10 ml of trimethylpentane were added.Against a trimethylpentane reference solution, the absorbance (A 1 ) was read using spectrophotometry at 350 nm.Into 5 ml of trimethylpentane and oil solution, 1 ml of 2.5 g/l of p-anisidine in acetic acid was added.After 10 min and against a reference of 1 ml of p-anisidine in acetic acid and 5 ml trimethylpentane solution, the absorbance (A 2 ) was read using spectrophotometry at 350 nm.Using the formula "AV ¼ [10 Â [1.2 Â (A 2 À A 1 )]/m", the anisidine value was obtained in meq/kg, where m refers to the oil mass that was weighted within the small vial.

Statistical analysis
The statistical analysis was performed using a statistical package from Microsoft Office Excel 2017.Descriptive analysis was done and presented as means, standard error (SE), and 95% confidence interval).Means were used for comparison.Association between variables of interest (EPA, DHA, AV, PV, and Totox) were assessed using simple linear correlations, general linear regression models, and Spearman's rank.p-Value of <0.05 was considered statistically significant.

Results
The intra-assay relative coefficient of variation (CV %) for all the samples analyzed was calculated.All CV% values were found acceptable.The CV% values of EPA ranged from 0.001 to 0.64, and DHA ranged from 0.008 to 0.80.However, the CV% of PV ranged from 0 to 0.17, while AV ranged from 0.05 to 0.24, and Totox ranged from 0.001 to 0.11.
Table 1 shows the 14 tested brands of fish oil supplements containing 1000 mg of EPA and DHA available in Bahrain markets.There was significant variation in price through different brands, with the calculated recommended retail price (RRP) per gram of fish oil ranging from $0.02 to $0.20.No significant relation was found between RRP price and actual EPA and DHA contents.
Labelled DHA ranged from 36 up to 380 mg/capsule.Only two brands (14.3%) showed an approximately similar amount.Those were brand 7, in which the actual DHA content was 33.97 ± 1.82 mg/capsule, whereas the labelled content was 36 mg/capsule, and brand eight, actual DHA content equal to120.72 ± 0.98 mg/capsule, whereas the labelled content was 120 mg/capsule.In comparison, only one brand, 5 (7.1%), showed a relatively higher actual amount of 154.93 ± 124.59 mg/capsule than the labelled content of 120 mg/capsule.All the remaining brands (1-4, 6, 9-14) (78.6%) have a lower actual DHA amount than the labelled.
Tables 3 and 4 show the mean values of EPA, DHA, AV, PV, and Totox of each tested brand in an estimated 95% confidence interval.
Table 5 shows that there is a significant correlation between EPA and DHA (p-value <0.01) and between Totox and AV (p-value <0.01).No significant correlation was found between Totox and PV.
A positive association was observed between missing EPA þ DHA (labelled minus actual content) and PV, whereas a negative association was found between missing EPA þ DHA (labelled minus actual content) with AV and Totox (Figures 1-3).
No significant relation was found between supplements price and EPA and DHA contents.

Discussion
Fatty acids are very important energy components within the body (Boelen, Dijk, Damste, Rijpstra, & Buma, 2013), where the most important group that has a crucial role in human health are omegas (Fialkow, 2016), especially omega-3 and omega-6 (Stice, 2019).EPA and DHA are the main omega-3 fatty acids needed to be consumed in a diet for their essential role (Lopes et al., 2017).However, diet might not be enough; thus, fish oil supplementation is essential to compensate for their deficit (Opperman et al., 2011).Therefore, testing fish oil supplements is crucial to ensure they contain the needed dosage.The overall results of this study are consistent with a similar previous study conducted in South Africa in which out of 45 tested commercially available fish oil supplements from South African markets, more than half of them failed to attain claimed contents of EPA and DHA or one of them, and 15% has more than what was claimed (Opperman et al., 2011).Albert et al. (2015) also tested 32 supplement brands from New Zealand markets and found that only three brands have EPA and DHA contents equal to or higher than what is claimed, 2/3 (29 supplement brands) contain <67% of the labelled amounts, and two supplements had only 1/3 of the labelled amount.
Furthermore, both Albert et al. (2015) and Opperman et al. (2011) indicated that the highest percentage of the commercially available fish oil supplements have lower EPA and DHA contents than the labelling, attributing these findings to two main reasons.The first is the poor-quality control by the manufacturing company, where measurement of the supplement's content after each manufacturing stage is missing, leading to a reduction in EPA and DHA contents due to oxidative damage and other factors resulting in low product efficacy.Secondly, there is no seasonable control which means they always depend on the same fish sources without considering the normal fluctuations that occur seasonally affecting natural omega-3 present in fish, they always rely on the same amounts without measurements (Albert et al., 2015;Opperman et al., 2011).
All these factors have indicated misleading information with incorrect supplement contents in New Zealand and South African markets (Albert et al., 2015;Opperman et al., 2011).Supplying customers with misleading information results in consumers wasting their money without having their daily needs, while excess omega-3 can lead to bleeding, high risk of haemorrhagic stroke, high lipid peroxidation, and suppression of the immune system leading to oxidative damage (Albert et al., 2015;Opperman et al., 2011).
As mentioned previously, fish oils are sensitive to oxidation, thus the benefit and quantity of these oils are reduced.Oxidation of fish oils results in the liberation of primary and secondary oxidation products.In this sense, the peroxide value (PV), the anisidine value (AV), and the total oxidation value (Totox) are calculated to find out the degree of oxidation (Albert et al., 2015;Barnes, Bloom, and Nahin, 2008;Barnes, McFann, Powell-Griner, & Nahin, 2004).
In the present study, most of the tested supplements exceeded the recommended levels of PV which is less than the levels found in the New Zealand study (83%) (Albert et al., 2015) and in the South Africa study (84%) (Opperman & Benade, 2013), but higher than the level of the United Arab Emirates (40.95%) (UAE) study (Jairoun, Shahwan, & Zyoud, 2020) and the Canadian study (17%) (Jackowski et al., 2015).
In regards to AV, only a small number of the tested supplements exceeded the safe levels of AV.Which was below the New Zealand study (25%) (Albert et al., 2015) and the Canadian study (41%) (Jackowski et al., 2015), but slightly higher than the UAE study (6.8%) (Jairoun et al., 2020).The AV level is an indication of further oxidation and the production of harmful secondary metabolites (Albert, Cameron-Smith, Hofman, & Cutfield, 2013;Shahidi & Zhong, 2010).
Comprehensively, three brands were found to be unsafe to consume, and three were in the early stages of oxidation because only PV is higher than the recommended (Table 2).However, it is enough to have one parameter above the safe line for the brand to be considered unsafe for consumption.In the present study, most of the tested brands are oxidized and some have all the parameters of PV, AV, and Totax within the recommend values.Therefore, oxidation could be related to the low levels of EPA and DHA found in some fish oil supplements tested as there was an association between AV, PV, and Totox, and the missing EPA and DHA suggesting reduced efficacy.
Generally, the findings of this study are compatible with previous studies.As for Jackowski et al. (2015), 171 North American over-the-counter omega-3 PUFAs supplements were tested to evaluate their oxidation safety, in which 50% of the tested products exceeded the recommended values.In addition, 18% of the tested supplements are nearly unsafe, with a 1-3 year prior expiry date.It was observed that flavor-free supplements recorded lower AV and Totox in comparison with supplements with flavor additives.However, children supplements had the highest PV, AV, and Totox degrees.Their findings propose that flavor additives are linked to increased AV levels, which is very concerning since most kid's supplements contain flavor additives (Jackowski et al., 2015).However, the present study tested flavor-free supplements only, so the AV levels were not affected by this factor.
Furthermore, Kolanowski (2010) tested 19 brands of Warsaw over-the-counter fish oil supplements and found that the PVs of fish body oils and fish liver oils were ranging from 1.0 to 9.8 meq/kg.After increasing the storage temperature, PV increased in every supplement up to 2.2-12.5 meq/kg.Nonetheless, at 20 C storage condition, the PV was stable during the entire time for all the products (Kolanowski, 2010).The tested fish oil supplements in the current study were stored at 4 C once opened and were analyzed immediately, however, they were stored at room temperature in the Bahrain market.Some of the tested supplements contain antioxidant additives like Vitamin E found in brand 2, and tocopherols found in brands 4, 6, and 7.However, the presence of antioxidants did not prevent oxidation which has been seen in brand 7.These results agreed with Jackowski et al. (2015) findings in which most of the tested supplements were highly oxidized, despite the addition of antioxidants (Jackowski et al., 2015).Therefore, even if antioxidants were added, oxidation might be decreased but not averted (Lange, Nakamura, Gosslau, & Li, 2019;Zuta, Simpson, Zhao, & Leclerc, 2007).
Fish oil supplements that meet the specifications of quality (including oxidative status) and are consumed within the recommended daily intake cause no passive health effects.Moreover, long-term exposure to oxidized fish oil products is likely to cause oxidative stress, inflammation, and lipid metabolism disorders (Turner, McLean, & Silvers, 2006).However, clinical trials never reported any of the studies related to fish oil oxidation which makes it difficult to understand the effects of consuming the oxidized supplements in the current study (Albert, Cameron-Smith, Hofman, & Cutfield, 2013).
Compared to other countries, the oxidative levels of supplements in Bahrain are not very high.Nonetheless, it must be improved by implying standard procedures for fish oil extracting and processing (Bannenberg et al., 2017;Jairoun et al., 2020) and having regulations considering the storage of the fish oil supplements.When fish oil oxidizes, free radicals and peroxides are released.The peroxides may cause peroxidation of membrane lipids, oxidative stress, and cell impairment.Lipids peroxidation promotes inflammation-linked diseases like neurodegenerative diseases, Alzheimer's disease, and carcinogenesis.These harmful effects were proved by animal studies (Bartsch & Nair, 2006;Grimm et al., 2016;Lange et al., 2019;Maruyama et al., 2014;Pamplona et al., 2005;Sayre et al., 1997;Yakubenko & Byzova, 2017).Animal studies indicate that many oxidized lipids will affect the body.Thus, the consequences of fish oil oxidation are unknown (Albert et al., 2015;Esterbauer, 1993).

Conclusion
The results of the present study concluded that 8 out of 14 analyzed fish oil supplements available in Bahrain have at least one of these values (PV, AV, and Totox) above the recommended guideline.Interestingly, about 42.9% of the supplements were safe to consume.There was a positive association between PV and the missing EPA þ DHA, and a negative association was found between EPA þ DHA and AV and Totox, which suggests that primary oxidation products are linked to the EPA and DHA levels.Therefore, it is recommended to place omega-3 supplements in a controlled environment to minimize oxidation as much as possible.In addition, it is recommended to analyze flavored fish oil supplements, especially children's supplements.

Study limitation
Fish oil supplements with RRP per gram of fish oil of more than 0.2 US$ were not analyzed because of their high price.However, such supplements will not be consumed as much as the selected supplements with a reasonable RRP per gram of fish oil (US$ 0.02-0.20).

Table 1 .
List of the tested fish oil supplement brands comparing between their labelled and actual fish oil contents (EPA þ DHA).
EPA: eicosapentaenoic acid; DHA: docosahexaenoic acid; RRP: retail price per gram of fish oil; UK: United Kingdom; USA: United States of America; US$: United States dollar.ÃResults are expressed as mean ± SE.

Table 2 .
List of the tested fish oil supplement brands and their PV, AV, and Totox values.

Table 3 .
Estimated marginal means and confidence interval 95% for the tested brands EPA and DHA (mg/capsule).

Table 4 .
Estimated marginal means and confidence of interval 95% for the tested brands PV, AV, and Totox (Meq/kg).

Table 5 .
Correlation between actual EPA and DHA values with PV, AV, and Totox values.