And do they help with weight loss?
By the mid-2000s, fish oil supplements rich in omega-3 fatty acids were looking like lifesavers. The 2-year UK DART trial found that, in males recovering from a recent heart attack, 200-400 grams of fatty fish per week or 500 mg fish oil daily reduced the relative risk of overall and CHD mortality by 27% and 32%, respectively.
Over a decade later, the 3.5-year Italian GISSI trial reported 20% and 14% relative risk reductions in CHD mortality and overall mortality, respectively, among recent victims of heart attack taking 900 mg of EPA + DHA daily (the equivalent of 3 standard 1,000 mg fish oil capsules).
Even the intensely pro-pharma, omega-6-promoting American Heart Association took a shining to n3 supplements. A 2002 AHA review stated "RCTs have demonstrated that omega-3 fatty acid supplements can reduce cardiac events (eg, death, nonfatal MI, nonfatal stroke) and decrease progression of atherosclerosis in coronary patients."
But then the wheels started falling off.
The content below was originally paywalled.
The follow-up to the original DART trial was published in 2003. DART-2, involving 3,114 men under 70 years with angina, did not replicate the original positive results. In fact, the group assigned to fish/fish oil consumption not only failed to lower their cardiovascular and all-cause mortality, they actually experienced increased cardiac mortality.
The adjusted hazard ratio for cardiac death was 1.26 (equating to a relative risk increase of 26%), and even greater for sudden cardiac death (1.54). The excess risk was largely confined to the subgroup given fish oil capsules.
The results were concerning, but there were problems with the trial. Due to funding issues, the researchers were forced to perform DART-2 in two phases, with a twelve-month gap while recruiting was still underway. To determine compliance, the researchers measured blood levels of eicosapentaenoic acid (one of two main n3 fatty acids found in fish oil) among a small sample of fish oil and control subjects. Although mean EPA levels were higher after six months in the tested fish/fish oil patients, individual blood EPA concentrations varied widely, indicating many of them had not complied with the fish/fish oil recommendations.
The DART-2 results looked to be an anomaly, and sales of fish oil supplements continued to skyrocket. By the late 2000s, the question was not if you were taking fish oil, but how much.
So What Does More Recent Research Show?
Since then, many more clinical trials have examined the efficacy of fish oil on cardiovascular outcomes. We now have a clearer picture of n3 efficacy.
The good news is that the negative findings of DART-2 do indeed seem to be an anomaly.
The bad news is that, overall, fish oil supplements don’t appear very effective for preventing cardiovascular events, cardiovascular mortality, or all-cause mortality.
In 2020, the Cochrane Database of Systematic Reviews published their most recent assessment of the effects of fish- and plant-based omega-3 fats on all-cause mortality, cardiovascular events, and adiposity (a nice way of saying “overweight”).
They included randomized controlled trials (RCTs) that lasted at least 12 months and compared supplementation or advice to increase intake of long-chain n3s, alpha-linolenic acid, or both.
"Long-chain" omega-3 fatty acids (LCn3) are only found in animal foods, the most common being eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA). These are the two fatty acids that predominate in commercially available fish oil supplements.
Alpha-linolenic acid (ALA) is the main omega-3 fatty acid found in plant foods like flaxseed, chia seeds and walnuts. ALA is partially converted to LCn3 fatty acids within our bodies, and there has been some debate over the years regarding the efficiency of this conversion and the respective health benefits of LCn3 versus ALA.
Long-Chain n3 Results
The Cochrane analyses suggested “little or no effect” of increasing LCn3 intake on all-cause mortality. The risk ratio was 0.97, indicating a 3% relative risk reduction, which is miniscule.
This figure was obtained from a pooled analysis of 45 RCTs featuring 143,693 participants and 11,297 deaths. The researchers rated the data for the overall mortality figure as “high-certainty evidence.”
The corresponding risk ratio for cardiovascular mortality was 0.92 (117,837 participants; 5,658 deaths in 29 RCTs; moderate-certainty evidence).
The risk ratios for cardiovascular events, stroke and arrhythmia, were 0.96, 1.02, 0.99, respectively.
The most favorable risk ratio was seen for coronary heart disease mortality (0.90), but again, it must be remembered this equates to a measly 10% relative risk reduction. To place that figure in perspective, the number needed to treat for an additional beneficial outcome (NNTB) was 334. In other words, you’d need to treat 334 people to save a single person from CHD death.
The researchers analysed separate NNTBs for primary and secondary prevention.
Primary prevention occurs when you are trying to curtail the risk of an adverse outcome in healthy people.
Secondary prevention is when you set out to prevent that outcome in people who already have the illness in question (in this case, coronary disease). Many of the fish oil RCTs involved individuals who had already suffered an adverse cardiovascular event.
The CHD death NNTB for secondary prevention was 200, while in primary prevention it was 1,000.
Needless to say, those aren’t great odds.
Overall, the effects did not differ by trial duration or dose.
There was little evidence for the effects of eating fish.
Alpha-Linolenic Acid Results
As for increasing ALA intake, the Cochrane researchers concluded it probably makes little or no difference to all-cause mortality (RR 1.01, 19,327 participants; 459 deaths in 5 RCTs, moderate-certainty evidence). The RRs for cardiovascular mortality, coronary heart disease mortality and and coronary heart disease events were 0.96, 0.95 and 1.00, respectively.
Weight Loss
Fourteen trials reported on the effect of increasing LCn3 on body weight and were included in meta-analysis. They indicated little or no effect of LCn3 for weight loss in 17,000 participants (mean weight difference of 0.00 kg).
Few trials reported on more direct measures of adiposity (percentage body fat, percentage visceral fat, waist circumference, waist/hip ratio, abdominal circumference and hip circumference), with some suggesting higher adiposity and some lower adiposity in groups with more LCn3.
Other Effects
Increasing LCn3 and ALA had little or no effect on serious adverse events, lipids and blood pressure, except increasing LCn3 reduced triglycerides by 15% in a dose-dependent manner.
Overall, the Cochrane review included 86 trials involving 162,796 people up to February 2019. Twenty-eight trials were deemed very trustworthy (well-designed so as not to give biased results). Participants were adults, some with existing illness and some healthy, living in North America, Europe, Australia and Asia.
Doses of LCn3 ranged from 0.5 g/d of EPA and DHA to more than 5 g/d (19 RCTs had a dose of LCn3 < 1 g/d, 27 a dose of 1 to < 2 g/d, 12 of 2 to < 3 g/d, 19 RCTs had a dose of 3 or more g/d LCn3. One trial did not clearly state the dose).
Other systematic reviews have returned similar results, suggesting a lack of effect for omega-3 fats on all-cause mortality or a variety of cardiovascular diseases (Campbell 2013; Chowdhury 2012; Khoueiry 2013; Kotwal 2012; Kwak 2012; Mariani 2013; Rizos 2012; Zheng 2014).
The Cochrane researchers concluded: “We found high-certainty evidence that long-chain omega-3 fats (LCn3) do not have important positive or negative effects on mortality or cardiovascular events and moderate-certainty evidence that they have little or no effect on cardiovascular disease mortality, stroke or arrhythmia in primary or secondary prevention.”
Why Weren’t the Early Positive Results for LCn3 Replicated?
With the exception of the original DART and GISSI studies, all the studies reporting substantial reductions in overall mortality have been small studies, with under 400 participants in each group. In such small studies, a small difference in the death count between groups can drastically sway the risk reduction figures.
The GISSI-P study created a lot of buzz, but in retrospect we can see why the results gave false hope. While many of the n3 studies were double-blind, the GISSI-P study was an open-label study, meaning both the researchers and participants knew who was receiving fish oil. While deaths occurring during the trial were reportedly reanalyzed by an ad hoc and blinded committee of cardiologists and neurologists, the unblinded nature of the study could still have affected the behaviors, attitudes and reporting of both the subjects and the trial staff.
In 2008, the results of the double-blind GISSI-HF trial were published. It is revealing that this higher quality trial found far smaller relative reductions in overall and cardiovascular mortality (6% and 8%, respectively) from 1 gram daily of fish oil in heart failure patients. These figures are far more in line with those documented in the Cochrane analysis.
As for the positive findings in the single-blind DART trial, they may have simply been a chance occurrence, as no large trial has since replicated the findings. In the fish advice group, 136 subjects (14%) took fish oil capsules at 6 months as a partial or total substitute for fatty fish, and 191 out of 883 (22%) used fish oil at 2 years. So only a minority of the fish group actually used fish oil. Increased fish intake may have led to consumption of increased protein and other nutrients (such as iodine and selenium), and may have been accompanied by other dietary changes. I can only speculate about this, as the DART authors provided limited dietary information.
Again, it bears reiterating that more recent trials have not replicated the positive results of DART. The hypothetical benefits of increased fish intake must also be weighed against the very real risks of increased intake of PCB and heavy metals such as methylmercury.
The Bottom Line
In terms of reducing morbidity and mortality from cardiovascular causes, increased n3 intake has not lived up to its initial promise.
Overall, clinical trials find a small but negligible reduction in cardiovascular and overall mortality from fish oil supplementation.
Clinical trial evidence also shows n3 supplementation has no effect on weight loss.
Fish oil and other n3 supplements have been most widely hyped for their alleged ‘heart healthy’ benefits, and have also been the subject of some exuberant weight loss claims, so these are the outcomes I’ve focused on in this article.
The role of n3 supplementation for depression, skin conditions, pregnancy outcomes and other health conditions is beyond the scope of this article. Time permitting, I’ll address them in a future installment.
In the meantime, if your goal is to minimize your risk of heart disease, then keeping a lid on your bodily iron stores, maintaining an above-average level of physical fitness (lung capacity is an extremely important and woefully ignored predictor of health and longevity) and striving for good glycemic control will likely prove far more productive than fish oil supplementation.
Ciao,
Anthony.
Leave a Reply