Recent BMJ fish oil study

Response to the recent study published in the British Journal of Medicine regarding fish oil and major effects on cardiovascular markers.? This is just the type of study to throw confusion for many clinicians who are not familiar with clinical nutrition or research, which can result in many busy clinicians having ‘knee-jerk responses’ when evaluating whether to recommend fish consumption of fish oil supplement consumption in their patient's for the prevention of cardiovascular diseases.

This was a large study? cohort from the UK Biobank of over 400,000 patient's between the ages of 40 and 69, and provided an 11-year follow-up.? They were screened for the absence of cardiovascular findings when enrolled. Outcome measures were to measure incidence of atrial fibrillation, major cardiovascular adverse effects such as strokes & myocardial infarction and death over the time of the study.

This was a prospective observational study, and not a double-blind placebo controlled study, so there was no randomization on the placebo used i.e. what would have happened to these 400,000 cohort who did not eat fish or take fish oil?

The Intervention of whether fish oil was taken throughout this time was only recorded via survey or interview, hence it is unclear as to what type of supplements where taken. i.e. take it once or twice every week, or they have long periods where he did not take it , etc.? The study also does not mention the amount of fish oil supplementation, nor the quality of the fish oil supplementation, i.e. rancidity? good quality?

The baseline characteristics of the participants in the study was interesting; of the 400,000 plus cohort, almost 24% were obese, 42% were overweight, 33% being characterized as "a normal weight".

It is well-known that Body Mass Index/weight and obesity affects cardiovascular health adversely and contribute to many metabolic syndromes, involving high blood pressure, heart disease and also longevity.

Would have been good to have a placebo on this.

Other baseline characteristics of the participants was that of vegetable consumption; 29% of them ate 2-3 vegetables per week; 64.5% of participants had vegetables equal to, or greater than, 4-week.? Did they exclude potatoes as a vegetable? 6.8% had less than 2 servings/wk - Truly astounding. We know that deep leafy green vegetables, yellow-orange and red vegetables can provide significant phytonutrients that can enhance the health of the cardiovascular system and enhance longevity.

Physical activity levels baseline characteristics were not defined- just classified as low, moderate or high levels of physical activity. 18.5% had low physical activity, 40.8% had moderate).It did not say how they were determined. Physical activity is a good benchmark for cardiovascular ; - if you are not using the cardiovascular system well, it is not going to be as robust for you.

Alcohol consumption, which can affect inflammatory potential in the cardiovascular system, was analysed. They ruled out men drinking more than 8 units a day, females drinking more than 6 units a day from the program as this classified as binge drinking, but if you did less than that you are included in the program. Alcohol consumption was involved around 90% of the participants, obviously some heavier than others, which would have an effect on the cardiovascular health.

The above ‘co-variants’ above were added into the study regression models.

With better definitions, design and instruction the study could have been more useful.

The 11-year conclusions were ‘fairly neutral’. There is a slight increase in the hazard ratios (HR’s) from healthy to atrial fibrillation - seen in 4% of the cohort. There were major adverse cardiovascular effects in 5%, as well as death 5%.? Given the above, variants the observational nature of the study, confounding factors of compliance and the unknown quality and amount of fish (those eating less than (82%), or greater than 2 servings of fish per week (17%), and fish oil products that were used these findings do not really give us any good guidance as clinicians.

Other conclusions was that if, once you had Atrial fibrillation, fish oil showed a benefit in halting the progression of cardiovascular disease: decreasing the progression of atrial fibrillation patient's into major cardiovascular events such as stroke or myocardial infarction, and showed fish oils being, beneficial in decreasing the progression of heart failure towards death albeit with similar hazard ratios (around 0.92,0.85).

The conclusions were that fish oil supplementation, even with this fairly flawed study, "may be a risk factor for atrial fibrillation, but could be beneficial for preventing the progression of patients with atrial fibrillation from progressing to strokes, heart attacks and death from heart failure. Further studies are needed."

What could have improved this study?

A placebo would have been handy. Would we have 4% suffer from Atrial Fibrillation over the course of 11 years in a population that over 65% were overweight, 35% ate less than 3 vegetables a week, and 18.5% had low physical activity?

We could have had the patient's perform an omega index at least to look at the types of fats in the red blood cells What were baseline long chain fatty acids like eicosopentanoic acid (EPA) and docosuhexanoic acid (DHA) levels respectively??

What was the difference in omega 6? to omega-3 ratios? [omega 6 predominance is inflammatory and will outcompete omega-3's for absorption].?

Were participants given instructions on what constitutes fish oil?; there may have been candidates taking other oils such flaxseed oil, because it is an omega-3 acid. Many clinicians, and especially patients, are unaware of the difficulty in converting Flax seed ALA’s into long chain EPA.

As clinicians how do we deal with weak studies like this?

Be prepared to have this study touted in your clinical & lay realms; be prepared to discuss it with less educated clinicians, practitioners, GPs and other patients - its ‘neutral’ results and flawed design will drive confusion.? I wish this study had some tweaks in it to make it more useful for us as clinicians; Including a placebo would have been a ‘no brainer’. Maybe it was too expensive to add omega testing? Maybe it was designed & funded to obscure nutrition/diet information? Maybe it should have not been selected for publication?? Unfortunately the outcomes will sew continued confusion on diet/nutrition to both practitioners and patient's.

The bottom line with long chain ESSENTIAL fatty acids as found in fish oils is:

1) Our body's need them to properly function, that’s why they are ESSENTIAL.

2) We should try and get them from our diet as much as possible

3) In the absence of obtaining a strong omega-3 index between 8 and 12% in our red blood cells, we should augment our diets with quality, non-rancid fish oil ?supplements, such as cod liver oil, EPA/DHA supplements (sometimes up to 2 g a day).

4) Always include sufficient deep leafy green vegetables and low glycemic fruits to provide antioxidant support to prevent rancidity.

Cheers for listening!

Chen G, Qian Z(, Zhang J, et al

Regular use of fish oil supplements and course of cardiovascular diseases: prospective cohort study

BMJ Medicine 2024;3:e000451. doi: 10.1136/bmjmed-2022-000451