Low-Carbohydrate-Diet Score and the Risk of Stroke in Older People: Guangzhou Biobank Cohort Study and Meta-Analysis of Cohort Studies

Abstract

Background: Evidence regarding the association of low-carbohydrate diet (LCD) with the risk of stroke is limited. We examined the associations of LCD with the risk of stroke in a population with relatively high carbohydrate intake and risk of stroke, and updated the evidence using meta-analysis. Methods: 19,850 participants aged 50+ years without stroke history at baseline (2003-6) in the Guangzhou Biobank Cohort Study were included. Low-carbohydrate-diet score was calculated based on the percentage of energy from carbohydrate, protein and fat, with higher score indicating a lower intake of carbohydrate and a higher intake of fat and protein. Three categories: total, animal-based, and vegetable-based LCD score including seven subtypes of LCD scores were created. Findings: During an average follow-up of 13·1 (standard deviation=2·5) years, 1,661 stroke events and deaths occurred, including 1,255 ischaemic strokes. Increasing “total LCD” and three “animal-based LCDs” scores were marginally non-significantly associated with a lower risk of stroke, while increasing three “vegetable-based LCDs” scores were not significantly associated with a lower risk of all stroke. The adjusted HR (95% CI) of all stroke and ischaemic stroke was 0·91 (0·79-1.05) and 0·90 (0·77-1·06) for the highest versus the lowest quartile of “total LCD” scores, and 0·89 (0·78-1·03) and 0·89 (0·76-1·04) for the highest versus the lowest quartile of “animal LCD” scores, respectively. Moreover, inverse associations with all stoke and ischaemic stroke in the second versus the lowest quartile of “animal LCD” scores were found, with the HR (95% CI) being 0·87, (0·77-0·997) and 0·83 (0·72-0·97), respectively. In addition, inverse associations with all stoke and ischaemic stroke both in the second and third versus the lowest quartile of “animal low-bad-carbohydrate-diet” scores were also found, with the HR (95% CI) being 0·83 (0·73-0·95) and 0·77 (0·66-0·90) in the second quartile , and 0·87 (0·76-0·99) and 0·83 (0·71-0·96) in the third quartile, respectively. A meta-analysis including 149,493 participants and 5,639 strokes found marginally non-significant association of the highest versus the lowest category of carbohydrate intake with higher risks of all stroke (pooled HRs 1·10, 95% CI 0·96-1·26). Interpretation: In a population which consumed a relatively high level of carbohydrate and low level of protein and fat, diets lower in carbohydrate, higher in animal protein and low-to-moderate in saturated fat, but not higher in vegetable protein and unsaturated fat, were associated with a lower risk of stroke. Funding Statement: The University of Hong Kong Foundation for Educational Development and Research, the Health Medical Research Fund in Hong Kong; the Guangzhou Public Health Bureau and the Guangzhou Science and Technology Bureau; Natural Science Foundation of Guangdong, Natural Science Foundation of China and the University of Birmingham, UK. Declaration of Interests: The authors declare that they have no conflict of interest.