BMJ  2006;332:915 (15 April), doi:10.1136/bmj.332.7546.915-a

Letter

Risks and benefits of omega 3 fats

A few thoughts on systematic review

EDITOR—Hooper et al's conclusions that omega 3 fats have no effect on total mortality, combined cardiovascular events, or cancer are somewhat misleading.¹

Firstly, their null findings could be partly explained by the use of the composite end points. They ignore strong biological evidence for the potentially disease specific effects of omega 3 fat. The underlying hypotheses for this study are not clearly stated for the main analyses. Systematic reviews have been published of the effect on mortality from coronary heart disease of long chain omega 3 polyunsaturated fatty acids in randomised controlled trials and fish in prospective cohort studies.^{2 3}

Secondly, the authors focus primarily on the summary estimates from heterogeneous studies.

Thirdly, Hooper et al exclude 108 potential cohorts that have no omega 3 assessment. Dietary intake of long chain omega 3 fatty acids assessed by dietary instruments is more likely to be a surrogate marker of fish consumption. The effects of dietary long chain omega 3 fatty acid intake cannot be isolated from fish intake. In observational studies, the focus should therefore be on fish consumption rather than intake of long chain omega 3 fatty acid. Hooper et al also compare the most exposed quantile with the least exposed quantile. Since the amounts of omega 3 intake substantially varied between these two extreme groups across individual studies, the combined results may differ depending on the range of omega 3 intake and number of exposure groups in the primary studies.

Fourthly, the authors do not provide data for exploring any dose-response relation or possible threshold for the effects of omega 3 fats on different end points of interest. Neither do they provide further evidence to justify the sufficiency and robustness of their results.

Finally, their study does not distinguish between primary prevention and secondary prevention by omega 3 fatty acids. Mixing them together could lead to misinterpretation of the results.

Department of Preventive Medicine, Feinberg School of Medicine, Northwestern University, 680 North Lake Shore Drive, Suite 1102, Chicago, IL 60611, USA kahe{at}northwestern.edu

Division of Preventive Medicine, Brigham and Women's Hospital, 900 Commonwealth Avenue East, Boston, MA 02215, USA

---

Competing interests: None declared.

References

1. Hooper L, Thompson RL, Harrison RA, Summerbell CD, Ness AR, Moore HL, et al. Risks and benefits of omega 3 for mortality, cardiovascular disease, and cancer: systematic review. BMJ 2006;332: 752-60. (1 April.)[Abstract/Free Full Text]
2. Bucher HC, Hengstler P, Schindler C, Meier G. N-3 polyunsaturated fatty acids in coronary heart disease: a meta-analysis of randomized controlled trials. Am J Med 2002;112: 298-304.[CrossRef][ISI][Medline]
3. He K, Song Y, Daviglus ML, Liu K, Van Horn L, Dyer AR, et al. Accumulated evidence on fish consumption and coronary heart disease mortality: a meta-analysis of cohort studies. Circulation 2004;109: 2705-11.[Abstract/Free Full Text]

CiteULike    Complore    Connotea    Del.icio.us    Digg    Reddit    Technorati    What's this?

Related Article

Risks and benefits of omega 3 fats for mortality, cardiovascular disease, and cancer: systematic review

Lee Hooper, Rachel L Thompson, Roger A Harrison, Carolyn D Summerbell, Andy R Ness, Helen J Moore, Helen V Worthington, Paul N Durrington, Julian P T Higgins, Nigel E Capps, Rudolph A Riemersma, Shah B J Ebrahim, and George Davey Smith
BMJ 2006 332: 752-760. [Abstract] [Full Text] [PDF]

This article has been cited by other articles:

• Theodoratou, E., McNeill, G., Cetnarskyj, R., Farrington, S. M., Tenesa, A., Barnetson, R., Porteous, M., Dunlop, M., Campbell, H. (2007). Dietary Fatty Acids and Colorectal Cancer: A Case-Control Study. Am J Epidemiol 166: 181-195 [Abstract] [Full text]  
• von Schacky, C., Harris, W. S. (2007). Cardiovascular benefits of omega-3 fatty acids. Cardiovasc Res 73: 310-315 [Abstract] [Full text]