Role of diet in stroke incidence: an umbrella review of meta-analyses of prospective observational studies.

BACKGROUND: Stroke is one of the major challenges for the global healthcare system, which makes it necessary to explore the relationship between various modifiable factors and stroke risk. Recently, numerous meta-analyses of prospective observational studies have reported that dietary factors played a key role in the occurrence of stroke. However, the conclusions of previous studies have remained controversial and unclear. Accordingly, we conducted an umbrella review synthesizing and recalculating available evidence to assess the certainty of the associations between dietary factors and stroke. METHODS: Relevant meta-analyses examining the associations between dietary factors and stroke were searched in PubMed and Embase databases up to September 1, 2021. For each eligible meta-analysis, two independent reviewers appraised the methodologic quality using the AMSTAR 2 criteria and estimated the summary effect size, 95% confidence intervals, 95% prediction intervals, heterogeneity between studies, and small-study effects. Moreover, we further assessed the associations between dietary factors and ischemic stroke as well as hemorrhagic stroke. Lastly, a set of pre-specified criteria was applied to qualitatively evaluate the epidemiological credibility of each dietary factor. RESULTS: Overall, our umbrella review included 122 qualified meta-analyses for qualitative synthesis, involving 71 dietary factors related to food groups, foods, macronutrients, and micronutrients. Using the AMSTAR 2 criteria, 5 studies were assessed as high quality, 4 studies as moderate quality, and 113 studies as low or critically low quality. We identified 34 dietary factors associated with stroke occurrence, 25 dietary factors related to ischemic stroke, and 11 factors related to hemorrhagic stroke. Among them, high/moderate certainty epidemiological evidence demonstrated an inverse association between intake of fruits (RR: 0.90) and vegetables (RR: 0.92) and stroke incidence, but a detrimental association between red meat (RR: 1.12), especially processed red meat consumption (RR:1.17), and stroke incidence. Besides, the evidence of high/moderate certainty suggested that the intake of processed meat, fruits, coffee, tea, magnesium, and dietary fiber was associated with ischemic stroke risk, while consumption of tea, fruits, and vegetables was relevant to hemorrhagic stroke susceptibility. CONCLUSIONS: Our study has reported that several dietary factors have a significant impact on stroke risk and offered a new insight into the relationship between dietary modification and stroke occurrence. Our results may provide an effective strategy for stroke prevention.