Background & aims: We performed a meta-analysis to illuminate and quantify the potential relationship between poultry intake and risk of stroke through summarizing available evidence using a dose-response meta-analysis of prospective cohort studies.

Methods: Pertinent prospective cohort studies attained using electronic searches through PubMed, and Scopus up to September 25th, 2017. Relative risks (RR) s with corresponding 95% confidence intervals (CIs) of the highest versus the lowest for cohort studies were evaluated using DerSimonian and Laird random-effects models to find combined RRs. We also assessed the dose-response effect of this relationship.

Results: A total of 7 studies involving 354 718 participants were met inclusion criteria. The pooled RR of total stroke risk was [RR = 0.92 95% CI, 0.82, 1.03, I² = 19.8%, P_{heterogeneity} = 0.28] for the highest versus lowest categories of poultry intake. Subgroup analysis showed an inverse associations for the US people [RR = 0.86 95% CI, 0.77, 0.95, I² = 0.0%, P_{heterogeneity} = 0.38] and women [RR = 0.83, 95% CI, 0.72, 0.93, I² = 0.0%, P_{heterogeneity} = 0.63]. We did not obtain any significant association in the subtypes of strokes with highest versus lowest poultry intake [Ischemic stroke (RR = 0.91; 95% CI: 0.79, 1.02, I² = 0.0%, p = 0.93), Hemorrhagic (RR = 0.82; 95% CI: 0.59, 1.04, I² = 20.5%, P_{heterogeneity} = 0.28)]. One serving per week increment in poultry intake was not associated with the risk of stroke (RR: 1.00, 95% CI: 0.96, 1.03, I² = 69.0%, P_{heterogeneity} = 0.004). Nonlinear dose-response meta-analysis showed lower risk of stroke at consumption ∼1 serving/week.

Conclusion: Our meta-analysis revealed that poultry intake is not associated with total stroke risk; furthermore, an inverse relationship in US population and females must be interpreted with caution.