From the results, it has clearly been shown that the concentrated rTG fish oil supplement, at the prescribed daily dose, produces a greater increase in whole blood ω-3FAs than the EE fish oil supplement, the krill oil PL supplement and the salmon oil TG supplement, at their prescribed doses. Both the EPA + DHA combination and DHA alone exhibited the greatest increase with the rTG supplement, and the greatest decrease in ω-6 FAs. For the increase in EPA alone, the concentrated TG fish oil supplement is similar to the EE fish oil supplement in this regard, but much higher that the krill oil PL and salmon oil TG supplements. The percentage change in EPA for the concentrated rTG fish oil and EE fish oil supplement is more than four times that of the krill oil and salmon oil supplements.
For DHA, this increase in the concentrated rTG fish oil supplement is illustrated effectively in Table 3, where the percentage increase in DHA for the concentrated rTG fish oil product is 44%, which is highly significant and much greater than for any of the other supplements. Only the other TG supplement, derived from salmon oil , showed any significant increase, but at a much lower level of significance than the concentrated rTG fish oil supplement, i.e. P = 0.0137 vs. p < 0.0001, respectively (data not shown). In order of best to worst, in terms of significant beneficial changes in ω-3FA biomarkers, in combinations and in calculated ratios, the ranking is concentrated rTG fish oil > EE fish oil > Salmon oil TG > Krill Oil PL. Both the krill oil and the salmon oil TG produced a smaller increase in blood levels of most of the ω-3FA biomarkers of interest, when compared to concentrated rTG fish oil and EE fish oil.
However, given the considerable variability in the supplement doses given to trial subjects, all of the above is to be expected, i.e. the higher the dose, the greater the rise in fatty acids of interest. In order to more accurately compare the four supplements, the results were extrapolated for an intake of 1,000 mg, but this was less than successful, given the considerable number of subjects in some of the groups whose blood levels of some omega-3 fatty acids decreased from Day 0 to Day 28. For example, the krill oil EPA extrapolated increase was 232% with all subjects’ data included, 354% with all negative values deleted, and 266% with all negative values replaced with a zero. However, neither of these latter two values is any more accurate that the value with all data included, as the fact remains that in 8 of 35 subjects in the PL group, blood levels of EPA actually decreased in value during the supplementation period. Because of the difficulties encountered in data extrapolation, i.e. the inaccuracies inherent in the methodology, given the negative results for some subjects, a statistical analysis of this data was not considered to be warranted.
Given the high supplement levels of EPA in the rTG and EE groups, relative to the PL and TG groups, (650 and 756 vs. 150 and 180 mg/day, respectively) one could assume that these groups’ blood EPA values should have increased, and this was the case; indeed, none of these subjects reduced EPA levels, and this tends to reinforce the belief of good compliance with the trial protocol. That being the case, there is no reason to believe that these same subjects were any less compliant when consuming the other two products in this crossover trial, particularly when subjects consumed the four products in different and random order, i.e. they were not suffering from “intake fatigue” towards the end of the study, given that at least some of them consumed the latter two products (PL and TG) at the beginning or middle of the study. Thus, one may conclude that the negative values obtained for the EE, PL and TG groups are an accurate reflection of what transpired in the trial.
One explanation of the negative results for some of the subjects in the latter two groups (PL and TG) is the possibility that these subjects were consumers of a significant amount of fish prior to the study, and that the amount of EPA and DHA in the supplements for these groups was insufficient to match their previous intake from food, prior to supplementation. In future designs, it would be useful to ask subjects to refrain from consuming any fish products for an appropriate amount of time prior to study onset, and to supplement all subjects with a nominal amount of EPA and DHA for a run-in period prior to the commencement of the trial.
Although only small differences in gender comparisons were noted, the number of subjects in each group was small, i.e. 18 males and 17 females, and this may have been inadequate for accurate statistical comparison, such that it was insufficiently powered to detect gender differences. In order to more fully examine possible gender differences, future recruitment of a larger number of subjects in each group would permit more in-depth analysis of possible gender differences following omega-3 supplementation.
Of greater interest to the general population, and particularly to the significant number of North Americans at risk for cardiovascular disease, are the effects of supplementation on CVD risk reduction. The results for the measured and derived biomarkers specifically related to aspects of cardiovascular disease risk reduction clearly illustrate that the consumer must give careful consideration to the type and dosage of supplement they choose. For example, fewer than half of the subjects who consumed the krill oil supplement, at the recommended dosage, reduced the risk for sudden death, heart disease, death from fatal ischemic heart disease and sudden myocardial infarction, according to the measured and calculated indices. The majority experienced no benefit, and a few actually experienced a possible increase in risk.
Conversely, the vast majority of subjects who consumed the concentrated rTG fish oil were projected to reduce their risk of the aforementioned CVD-related conditions, except for the risk of death from fatal ischemic heart disease, where most of the subjects consuming each of the supplements experienced no change in this risk. Results for the EE fish oil group were somewhat better than for the krill oil PL or TG fish oil group, but not as good as those of the concentrated fish oil rTG group.
The fatty acid analyses employed in this study did not include the isolation of serum or plasma phospholipids, or of red blood cell phospholipids. These analyses would more fully reflect the concentrations of ω-3 FAs available for further metabolism. It is at this level of metabolism that ω-3 FAs exhibit their greatest beneficial effects; for example, as anti-inflammatory omega-3 prostaglandin precursors, they compete metabolically with pro-inflammatory ω-6 precursors [36–38]. In any future study, it would be useful to examine the concentrations of ω-3 FAs in membrane phospholipids, vis-à-vis serum phospholipid fatty acid analyses. Also, levels of lipid peroxides in supplements and in subjects’ plasma samples would be useful as an indicator of any deterioration of long chain fatty acids in the supplements.
In addition, a head-to-head comparison of the supplements utilised in this trial, at equivalent doses of EPA and DHA, would be useful in determining their relative bioavailability and their efficacy in increasing blood levels of omega-3 fatty acids, and in reducing CVD risk. In many omega-3 clinical trials, compliance is measured both by pill count before and after supplementation, and by percentage increase in blood EPA levels post-consumption. In this study, only the former was possible, as EPA levels were highly reliant on the level of EPA supplementation. In a study with equivalent doses of EPA, percentage change in EPA post-supplementation could be utilised to further verify compliance. Furthermore, although compliance by pill count was excellent, there is no way to verify if subjects did indeed fully comply with the protocol, and some of the results may have been a result of lack of compliance, rather than the relative efficacy of the supplement per se.
A statistical assessment of the order of treatment showed no effect of treatment order on the results. Likewise, there were no statistical differences among baseline levels of EPA and DHA after each washout period. Nonetheless, it would be useful, in future studies, to measure other biomarkers in addition to levels of omega-3 fatty acids, for example levels of carrier proteins, receptors, etc., in order to ensure that absence of any carry-over effect after each washout period. In addition, diet records and/or food frequency questionnaires during washout periods would be useful in determining both background diet and omega-3 intake of subjects. Alternatively, subjects might be given a list of foods (e.g. fish and fish products) and supplements that would be forbidden from consumption during the washout periods, as this would facilitate equality of baseline levels of omega-3 fatty acids at the beginning of each test period.