No beneficial effects of fish oil supplementation on vascular events or mortality in diabetic patients
ESC 2018 - Munich
***ASCEND: A randomized trial of omega-3 fatty acids (fish oil) versus placebo for primary cardiovascular prevention in 15,480 people with diabetes
Presented at the ESC congress 2018 by: Louise BOWMAN (Oxford, UK)
Introduction and methods
Increased fish oil (omega-3 fatty acids [FA]) intake has been associated with reduced cardiovascular (CV) risk and based on trials conducted in the 1980s and 1990s, omega-3 FA supplements are recommended for secondary prevention. For primary prevention, increased fish intake is recommended. However, recent meta-analyses of randomized trials have not shown benefits of omega-3 FA in primary or secondary prevention. The Study of Cardiovascular Events iN Diabetes (ASCEND) investigated whether omega-3 FA can reduce the risk of CV events and deaths in diabetic patients without known arterial disease.
The placebo-controlled ASCEND-trial randomized diabetic (T1DM and T2DM) patients, aged ≥40 years to 1g-capsule omega-3 FA daily vs placebo or daily 100mg aspirin vs placebo (olive oil capsule) in a 2x2 factorial design. Using streamlined methods with mail-based questionnaires, over 99% completion for morbidity and mortality was achieved over a mean follow-up of 7.4 years. Average adherence to omega-3 capsules was 77%. Individuals with previous history of myocardial infarction (MI), stroke or arterial revascularization procedure, or with currently prescribed aspirin, warfarin or any other blood thinning medication were excluded from the study, leaving a study population of 15,480 participants.
The primary outcome was a composite of serious vascular events (SVE), including non-fatal MI, non-hemorrhagic stroke or transient ischemic attack, and CV disease (excluding any intracranial hemorrhages). The secondary outcome was SVE or any revascularization, pre-specified for subgroup analyses.
- During follow-up, similar effects on SVE were observed with omega-3 FA supplements and placebo (8.9% vs. 9.2%, HR: 0.97, 95%CI: 0.87-1.08, P=0.55).
- Omega-3 FA supplements did not have different effects on SVE or revascularization in different subgroups of patients, based on sex, BMI or 5-year vascular risk, compared to placebo.
- Omega-3 FA supplements did not have different effects on cause-specific mortality, although a slight favorable effect on vascular death was noted (2.5% vs. 3.1%, HR: 0.82, 95%CI: 0.68-0.99).
- The HR for effect of omega-3 FA intake on all-cause mortality was HR: 0.95 (95%CI: 0.86-1.05).
- No safety concerns were observed during the trial.
ASCEND was the largest and longest duration placebo-controlled trial that studied omega-3 FA supplementation. The results show that omega-3 FA intake is safe, but does not result in beneficial effects on SVE, nor on all-cause or cause-specific mortality in diabetic patients.
The potential modest effect on vascular death should be interpreted with caution, in light of multiple comparisons, and considering that all-cause mortality was not significantly reduced. Dr. Bowman called it a ‘likely play of chance’. These data suggest that the current guidelines on omega-3 FA intake should be reconsidered; to date there does not seem to be any justification to take omega-3 FA supplements for CVD prevention.
During the press conference, it was discussed whether a high dose may potentially have shown an effect. The ASCEND trial studied the dose of 1 g daily, because this is the recommended dose. Trials are ongoing that evaluate 2-4 grams per day, and it remains to be seen whether that yields vascular benefit and whether these give tolerability issues.
Moreover, the question was raised what may explain the benefit seen in earlier observational studies. The discrepancy indicates the difficulties of conducting dietary studies; if a person eats fish, he or she is not eating a rib-eye steak, and that may be a good thing in itself. Considering the big market in supplements, and since there is no justification to take fish oil supplements, pharmacists also have a responsibility in educating patients.
Although this trial could be considered a primary prevention study, no signal of a benefit was seen. In the context of other studies, Bowman noted that it seems reasonable to extrapolate these results to secondary prevention populations.
Our reporting is based on the information provided at the ESC congress.