Intended for healthcare professionals


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

BMJ 2006; 332 doi: (Published 30 March 2006) Cite this as: BMJ 2006;332:752

On meta-analyses

Hooper et al. recently published in the journal a Cochrane Database
Systematic Review on risk and benefits of omega 3 fats (1). They concluded
that omega 3 fats do not have a clear effect on total mortality,
cardiovascular events, or cancer.

The two main trials included in the meta
-analysis were: a) GISSI-Prevenzione (2,3) testing the effect of a daily
administration of 850 mg omega 3 fats given with 1 capsule of a highly
purified and concentrated form in 11323 post-MI patients, 1031 deaths,
duration 3.5 years, relative risk 0.86, 95% confidence interval 0.77 to
0.97); b) DART II (4) testing the effect of eating two portions of oily
fish each week (or to take up to 3 g of fish oil as a partial or total
substitute of fatty fish meals) in 3,114 patients with angina, 525 deaths,
duration about 8 years, relative risk 1.15, 95% confidence interval 0.98
to 1.34) trials. Therefore, 1,556 out of 1,995 total deaths occurred in
two trials with opposite results. From a conceptual (and substantial)
point of view, Hooper’s et al article is closer to a head-to-head
comparison of two trials than to a meta-analysis aimed at summarizing
properly the evidence on this issue.

While there are some concerns about
the use of qualitative assessments of trials on the base of appropriate
concealment of allocation to the study arms when applied to meta-analyses
including not concealable therapies like dietary interventions, most of
our concern for this meta-analysis pertains the inclusion of DART II
results. As the authors acknowledge in their paper, the
inclusion/exclusion of DART II can change profoundly the results of the
meta-analysis. To highlight this issue, it is worth considering that the
course of DART II was “rugged” by many difficulties.

Its recruitment phase
lasted from 1990 to 1996 and was hampered by an interruption of funding in
1992-93 that made recruitment of patients to be ceased and then restarted
one-year later. This “second phase” of the trial allowed the recruitment
of 2,003 additional patients, but major changes to the research protocol
were required to curtail costs (e.g., contacts between dieticians and
patients, collection of dietary and pharmacologicical data during follow-
up, ascertainment of incident myocardial infarction by examining hospital
notes). In addition a subrandomization was added among the patients
allocated “fish advice” who could receive omega 3 fats through either
dietary or pharmacological intervention (5).
The shortage of funds causing the lack of proper follow-up, the length of
the study follow-up, the temporary stop of the trial, the “sensitive”
nature of dietary intervention requiring constant reinforcement of the
message to be delivered to the study participants, the added randomization
to “three large capsules daily” in the second phase of the study, all
could have hampered the compliance of patients to the experimental
intervention as well as to pharmacological therapies known to be effective
in the prevention of cardiovascular disease, thus causing a higher than
expected rate of events in the experimental arm and eventually an
unexpectedly higher mortality rate in patients allocated omega 3 fats

GISSI-Prevenzione was a pragmatic trial that adopted a PROBE format
(Prospective Randomized Open Blinded Endpoint adjudication) like the
Hypertension Optimal Treatment study (6). Because of its design, GISSI-
Prevenzione was (wrongly) considered as a weak study and this explains the
results of the “sensitivity analysis” that was carried out after excluding
“studies at moderate or high risk of bias”.

Last but not least, it is somewhat strange to have a meta-analysis
published in 2006 dealing with trials published up to February 2002. A
number of new, relevant clinical trials a) are currently available in the
literature, b) have been recently presented in medical meetings, and c)
are going to be presented shortly. Was this a truly systematic overview
including all available evidence in this field? Did we need it ?

1) Hooper L, Thompson RL, Harrison RA, Summerbell CD, Ness AR, Moore
H, Worthington HV, Durrington, PN, Higgins JPT, Capps NE, Riemersma RA,
Ebrahim SBJ, Davey Smith G. Risk and benefits of omega 3 fats for
mortality, cardiovascular disease, and cancer: a systematic review. BMJ
2006. [Epub ahead of print; doi = 10.1136/bmj.38755.366331.2F]
2) GISSI Prevenzione Investigators. Gruppo Italiano per lo Studio della
Sopravvivenza nell’Infarto miocardico. Dietary supplementation with n-3
polyunsaturated fatty acids and vitamin E after myocardial infarction:
results of the GISSI-Prevenzione trial. Lancet 1999;354:447–455.
3) Marchioli R, Barzi F, Bomba E, Chieffo C, Di Gregorio DDMR, Franzosi
MG, Geraci E, Levantesi G, Maggioni AP, Mantini L, Marfisi RM,
Mastrogiussepe G, Minnini N, Nicolosi GL, Santini M, Schweiger C, Tavazzi
L, Tognoni G, Tucci C, Valagussa F. Early protection against sudden death
by n-3 polyunsaturated fatty acids after myocardial infarction: time
course analysis of the results of the Gruppo Italiano per lo Studio della
Sopravvivenza nell'Infarto Miocardico (GISSI)-Prevenzione. Circulation
4) Burr ML, Ashfield-Watt PAL, Dunstan FDJ, Fehily AM, Breay P, Ashton T,
Zotos PC, Haboubi NAA, Elwood PC. Lack of benefit of dietary advice to men
with angina: results of a controlled trial. Eur J Clin Nutr
5) Burr ML, Dunstan FDJ, George CH. Is Fish Oil Good or Bad for Heart
Disease? Two Trials with Apparently Conflicting Results. J Membrane Biol
6) Hansson L, Zanchetti A, Carruthers SG, Dahlof B, Elmfeldt D, Julius S,
Menard J, Rahn KH, Wedel H, Westerling S. Effects of intensive blood-
pressure lowering and low-dose aspirin in patients with hypertension:
principal results of the Hypertension Optimal Treatment (HOT) randomised
trial. HOT Study Group. Lancet 1998;351:1755-62.

Competing interests:
GISSI-Prevenzione co-ordinator

Competing interests: No competing interests

14 April 2006
Roberto Marchioli
Head, Laboratory of Clinical Epidemiology of Cardiovascular Disease
Luigi Tavazzi, Gianni Tognoni.
66030 Santa Maria Imbaro, Italy