Dietary intake of different carbohydrate types and asthma risk: a systematic review and meta-analysis.
Yijing Bao, Jingyan Zeng, Lijie Wang, Yongjia Sun, Yuncheng Wang, Ye Chen, Yang Xia, Yuhong Zhao, Lishen Shan, Hehua Zhang
Abstract
Open AccessBACKGROUND: Serving as the predominant energy substrate, dietary carbohydrate consumption patterns significantly influence human health. However, the association between carbohydrate intake and asthma risk remains unclear. This study aims to systematically review and meta-analyze existing evidence on associations between various carbohydrate types (including free sugars [FS], non-free sugars [NFS], refined grains [RG], whole grains [WG], and dietary fiber [DF]) and risk of asthma. METHODS: From inception to May 2025, we systematically searched major databases (PubMed, Embase, Web of Science, and Cochrane Library) for epidemiological research on the associations between varied dietary carbohydrate types and risk of asthma or wheeze. Random-effect models were employed to calculate pooled odds ratios (pORs) with 95% confidence intervals (CIs). The I-squared (I2) statistic was used to assess heterogeneity. RESULTS: Ninety-four investigations met inclusion criteria (26 cohort, 8 case-control, 60 cross-sectional). Intake of DF (asthma: pOR = 0.879, 95% CI: 0.781-0.989; wheeze: pOR = 0.844, 95% CI: 0.737-0.967), NFS (asthma: pOR = 0.922, 95% CI: 0.870-0.976; wheeze: pOR = 0.813, 95% CI: 0.745-0.887) and WG-rich foods (asthma: pOR = 0.851, 95% CI: 0.749-0.966) was associated with lower risks of asthma and wheeze. In contrast, consumption of FS-rich foods was associated with increased asthma risk (pOR = 1.176, 95% CI: 1.093-1.265). Despite significant heterogeneity (I2 = 53.1%-82.5%, all P < 0.05), sensitivity analyses (using trim-and-fill and leave-one-out methods) validated the stability of the findings. CONCLUSIONS: This study underscores the importance of promoting high-quality carbohydrate intake (characterized by low FS and high NFS, DF, and WG), to support asthma prevention at the population level.