Coffee consumption and C-reactive protein levels: A systematic review and meta-analysis

Aims: Considerable debate exists regarding the association between coffee consumption and C-reactive protein (CRP) levels and the shape of this association. We conducted a metaanalysis to assess the relationship between coffee consumption and CRP levels. Data synthesis: We searched PubMed and Web of Science databases and conducted a hand search as of June 27, 2023. Meta-analyses were conducted using standardized mean differences (SMDs) with random-effects models, based on the geometric mean of CRP from included studies. We identified 13 studies for our systematic review and included 11 cross-sectional studies, involving a total of 66,691 subjects, in our meta-analysis. We found a linear inverse association between coffee consumption and CRP levels (p-value = 0.002) and did not find evidence of a nonlinear association (p for non-linearity = 0.13). Compared with the lowest category of coffee consumption (median, non-drinkers), the SMDs of log-transformed CRP levels were -0.02 (95% confidence interval [CI]: -0.05 to 0.00) for the third highest (median, 0.5 cup/day), -0.09 (95% CI: -0.15 to -0.04) for the second highest (median, 2.5 cups/day), and -0.14 (95% CI: -0.25 to -0.04) for the highest category (median, 4.5 cups/day). The inverse association tended to be stronger in women, but the difference by gender was not significant. Compared to the limited number of studies not adjusting for smoking, those that adjusted showed a strong linear inverse association, although the difference was not significant. Conclusions: Our findings indicate that coffee consumption is inversely associated with CRP levels. These associations may vary with potential modifiers, including gender and smoking adjustment. (c) 2024 The Italian Diabetes Society, the Italian Society for the Study of Atherosclerosis, the Italian Society of Human Nutrition and the Department of Clinical Medicine and Surgery, Federico II University. Published by Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.