Anthropometric and cardiometabolic effects of polyphenols in people with overweight and obesity: an umbrella review

Context: Polyphenols are plant-based compounds with potential anti-inflammatory, antioxidant, and anti-obesogenic properties. However, their effects on health outcomes remain unclear. Objective: To evaluate the effects of polyphenols on anthropometric and cardiometabolic markers. Data sources: Six electronic databases-namely, EMBASE, CINAHL, PubMed, Scopus, The Cochrane Library (reviews only), and Web of Science-were searched for relevant systematic reviews with meta-analyses (SRMAs). Data extraction: Three reviewers performed the data extraction via a data-extraction Microsoft Excel spreadsheet. Data analysis: An umbrella review and meta-analysis of existing SRMAs was conducted. Eighteen SRMAs published from 2015 to 2023, representing 445 primary studies and 838 unique effect sizes, were identified. Meta-analyses were conducted using random-effects models with general inverse variance. Polyphenol-containing foods were found to significantly improve weight (-0.36 kg; 95% confidence interval [CI]: -0.62, 0.77 kg; P < 0.01, I2 = 64.9%), body mass index (-0.25 kg/m2; 95% CI: -0.34, -0.17 kg/m2; P < 0.001, I2 = 82.4%), waist circumference (-0.74 cm; 95% CI: -1.34, -0.15 cm; P < 0.01, I2 = 99.3%), low-density-lipoprotein cholesterol (-1.75 mg/dL; 95% CI: -2.56, -0.94; P < 0.001, I2 = 98.6%), total cholesterol (-1.23 mg/dL; 95% CI: -2.00, -0.46; P = 0.002, I2 = 94.6%), systolic blood pressure (-1.77 mmHg; 95% CI: -1.77, -0.93 mmHg; P < 0.001, I2 = 72.4%), diastolic blood pressure (-1.45 mmHg; 95% CI: -2.09, -0.80 mmHg; P < 0.001, I2 = 61.0%), fat percentage (-0.70%; 95% CI: -1.03, -0.36%; P < 0.001, I2 = 52.6%), fasting blood glucose (-0.18 mg/dL; 95% CI: -0.35, -0.01 mg/dL; P = 0.04, I2 = 62.0%), and C-reactive protein (CRP; including high-sensitivity-CRP [hs-CRP]) (-0.2972 mg/dL; 95% CI: -0.52, -0.08 mg/dL; P = 0.01, I2 = 87.9%). No significant changes were found for high-density-lipoprotein cholesterol (-0.12 mg/dL; 95% CI: -1.44, 0.69; P = 0.67, I2 = 89.4%) and triglycerides (-1.29 mg/dL; 95% CI: -2.74, 0.16; P = 0.08, I2 = 85.4%). Between-study heterogeneity could be explained by polyphenol subclass differences. Conclusion: The findings of this umbrella review support the beneficial effects of polyphenols on anthropometric and metabolic markers, but discretion is warranted to determine the clinical significance of the magnitude of the biomarker improvements.