The role of fatty acids from fish in the prevention of stroke

BMJ 2012; 345 doi: http://dx.doi.org/10.1136/bmj.e7219 (Published 30 October 2012) Cite this as: BMJ 2012;345:e7219
  1. Janette de Goede, postdoctoral researcher,
  2. Johanna M Geleijnse, associate professor
  1. 1Division of Human Nutrition, Wageningen University, 6700 EV Wageningen, Netherlands
  1. janette.degoede{at}wur.nl

Fish oil supplements may not be protective in at risk patients who are optimally managed

Fish consumption once or twice a week is widely recommended for cardiovascular health. Fish is the main dietary source of the long chain omega 3 fatty acids eicosapentaenoic acid and docosahexaenoic acid. Low doses of these fatty acids (about 250 mg/day) have been suggested to protect against death from coronary heart disease (CHD).1 Fewer data are available on the part that fish intake plays in preventing stroke. In a linked systematic review and meta-analysis of prospective studies and randomised controlled trials (doi:10.1136/bmj.e6698), Chowdhury and colleagues evaluate the role of fish and omega 3 fatty acid intake in the primary and secondary prevention of stroke.2

Several meta-analyses on fish and incident stroke have been published previously.3 4 5 A 2004 meta-analysis of eight population based prospective cohort studies found that eating fish at least once a week was significantly associated with a 13-31% reduction in the risk of stroke when compared with eating fish less than once per month. The association was most pronounced for ischaemic stroke. A recently published update of this meta-analysis, which analysed 16 prospective cohort studies, came to a similar conclusion, although the effect sizes were smaller (9-14% lower risk).5 Another meta-analysis, published in 2011, which was based on 15 prospective cohort studies, found that eating three extra portions of fish per week was significantly associated with a 6% reduction in the risk of stroke. The results were similar for ischaemic and haemorrhagic stroke.4 The authors assumed a dose-response effect, with a linear association between fish intake and reduced risk of stroke.

The present meta-analysis by Chowdhury and colleagues included data from 12 randomised controlled trials (RCTs) that tested the effect of an increased intake of long chain omega 3 fatty acids, as well as 26 prospective studies, 21 of which had data on fish intake, 10 on long chain omega 3 fatty acid intake, and four on circulating concentrations of omega 3 fatty acids.2 The analysis examined data from 794 000 participants, among whom there were 34 817 stroke events. On the basis of the cohort studies, consumption of fish two to four times a week compared with once a week or less was significantly associated with a 6% reduction in the risk of stroke. When the top third of baseline long chain omega 3 fatty acid consumption (as measured by self reported dietary exposure) was compared with the bottom third, the relative risk of stroke was 0.90 (95% confidence interval 0.80 to 1.01). Similar results were seen for the top compared with bottom third of baseline fish consumption (0.91, 0.86 to 0.97). Results for ischaemic and haemorrhagic stroke were broadly similar.

Analysis of observational data showed that biomarkers of omega 3 fatty acids in blood were not associated with the risk of stroke. In addition, meta-analysis of data from the RCTs—in which those in treatment groups consumed on average 1.8 g of long chain omega 3 fatty acids a day (about 10-20 times the dietary dose in Western countries) over three years—showed that supplementation did not reduce stroke. Overall, the pooled relative risk for supplementation was 1.03 (0.94 to 1.12). Ten of the 12 randomised controlled trials included patients with previous cardiovascular disease. In these secondary prevention trials, the risk of stroke was increased by 17% in the group supplemented with long chain omega 3 fatty acids, although this finding was not statistically significant, and possibly merits further study.

Chowdhury and colleagues conclude that the potential beneficial effect of fish intake on stroke probably results from the interplay of a wide range of nutrients in fish and cannot primarily be attributed to long chain omega 3 fatty acids. Although this hypothesis seems reasonable, the effects of fish and long chain omega 3 fatty acids cannot be separated in cohort studies because the two are highly correlated. The effect sizes for fish and for long chain omega 3 fatty acid intakes were similar in the current meta-analysis. However, for long chain omega 3 fatty acids, the 95% confidence interval around the estimate was wider—probably because of the smaller number of stroke events—and statistical significance was not reached.

Fish consumption is low in most European countries and the United States. For example, in the Netherlands about 40% of the population eat fish less than once a month.6 There is strong evidence that fish consumption only once a week compared with less than once a month or none at all protects against fatal CHD.7 In the present meta-analysis of stroke, however, people who ate fish once a week were included in the reference group. Beneficial associations within the very low range of intake, as is common in Western countries,5 were not captured in this analysis.

In nutritional cohort studies, residual confounding from other dietary or lifestyle habits is always a concern. Such confounding can be avoided in RCTs, but inverse associations between long chain omega 3 fatty acid intake and incident stroke have not been supported by RCTs.8 9 However, trials included in the present meta-analysis were not primarily designed to detect an effect on stroke. Furthermore, most participants in RCTs have been patients with CHD, who would have received gold standard medical treatment (particularly those in later trials). In well treated patients, the absolute risk of stroke is reduced and beneficial effects of omega 3 fatty acids on top of treatment will be difficult to detect. The current findings are in line with disappointing results from RCTs of supplementation with long chain omega 3 fatty acids for the prevention of CHD.10 11 It seems that the additional benefit of supplementation in patients who are optimally managed may be small.12

On the basis of available evidence is it reasonable to advise people that eating one or two portions of fish a week could reduce the risk of CHD and stroke. Any benefit of long chain omega 3 fatty acid supplementation for the secondary prevention of CHD and stroke is likely to be small. However, it is possible that patients who are less than optimally medically treated or who have additional risk factors (for example, as a result of comorbidities such as diabetes) may benefit.13 14


Cite this as: BMJ 2012;345:e7219


  • Research, doi:10.1136/bmj.e6698
  • Competing interests: Both authors have completed the ICMJE uniform disclosure form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare: no support from any organisation for the submitted work; no financial relationships with any organisations that might have an interest in the submitted work in the previous three years; no other relationships or activities that could appear to have influenced the submitted work.

  • Provenance and peer review: Commissioned; not externally peer reviewed.