TOPLINE

A higher intake of coffee was associated with dose-dependent reductions in the risk for cirrhosis, hepatocellular carcinoma (HCC), and liver-related mortality. Participants consuming at least five cups daily showed more than 30% lower risk for incident cirrhosis, nearly 50% lower risk for HCC, and over 40% lower risk for liver-related mortality than nondrinkers.

METHODOLOGY

• Coffee consumption has been associated with potential health benefits, but the specific compounds or mechanisms involved are unclear. Although studies on gene-coffee interactions related to mortality have been conducted recently, how these interactions affect liver outcomes remains unknown.
• Researchers analyzed data of 354,957 participants aged 40-69 years (mean age, 57.0 years; 49.5% men; 93.9% White) from the UK Biobank, recruited across 22 centers between 2006 and 2010; participants did not have cirrhosis or HCC at baseline.
• Coffee consumption was assessed via a self-reported questionnaire at baseline and during imaging visits, with participants reporting the number of cups they had per day (one to two cups, three to four cups, at least five cups, or none), the type of coffee consumed (caffeinated [instant or ground] or decaffeinated), and use of sugar and artificial sweeteners.
• A subcohort of participants underwent abdominal MRI approximately 10 years after baseline for the assessment of hepatic fat via proton density fat fraction (PDFF), fibroinflammation via iron-corrected T1 (with elevated values defined as ≥ 800 msec), and liver iron content.
• The primary outcomes were incident cirrhosis, HCC, and liver-related mortality occurring at least 1 year after enrollment, identified through linked hospital and death registry records over a median 13-year follow-up. Proteomic profiling using Olink proximity extension assays quantified 2941 proteins in approximately 50,000 participants using blood samples.

TAKEAWAY

• Participants who consumed at least five cups of coffee daily had lower risks than nondrinkers for incident cirrhosis (adjusted hazard ratio [aHR], 0.68; 95% CI, 0.58-0.79; P for trend < .001), HCC (aHR, 0.53; 95% CI, 0.34-0.83; P for trend = .002), and liver-related mortality (aHR, 0.58; 95% CI, 0.45-0.74; P for trend < .001).
• On MRI, participants having at least five cups of coffee daily had a lower hepatic fat content (ΔPDFF = -0.83%; 95% CI, -1.02 to -0.65; P for trend < .001), lower odds of elevated iron-corrected T1 (odds ratio [OR], 0.80; 95% CI, 0.64-0.99; P for trend = .02), and a lower liver iron concentration (Δ = -0.018; 95% CI, -0.0291 to -0.0074; P for trend = .004) than participants who did not consume coffee.
• Associations were comparable among those having caffeinated and decaffeinated coffee for changes assessed via MRI; however, adding sugar or artificial sweeteners to coffee was associated with higher odds of elevated iron-corrected T1 (OR, 1.36; 95% CI, 1.08-1.70).
• Proteomic analysis revealed 74 proteins to be significantly associated with both coffee intake and incident cirrhosis (false discovery rate < 0.01 for each association). Several proteins related to hepatocytes and complements were associated with a decreased risk for cirrhosis and higher levels with increased coffee consumption, whereas a group of markers related to fibrosis and macrophages was associated with an increased risk for cirrhosis and lower concentrations among those who consumed more coffee.

IN PRACTICE

“[The study] findings provide multidimensional evidence that coffee intake relates to improved liver health, even before clinical disease onset,” the authors of the study wrote. “Given coffee’s wide availability, safety, and affordability, promoting moderate unsweetened coffee consumption could represent a simple, scalable strategy for liver disease prevention.”

SOURCE

The study was led by Hyun-Seok Kim, MD, MPh, PhD, Karsh Division of Gastroenterology and Hepatology, Cedars-Sinai Medical Center, Los Angeles. It was published online in Clinical Gastroenterology and Hepatology.

LIMITATIONS

Coffee intake was self-reported by the participants via a questionnaire, which might have introduced recall bias. Residual confounding and reverse causation could not be excluded despite extensive adjustments. Most participants were of European ancestry and were health conscious, thereby limiting the applicability of the findings to other populations. The alcohol thresholds used for defining excessive alcohol consumption were based on UK government guidelines, which differed from the criteria for metabolic dysfunction-associated and alcohol-associated liver disease proposed in the updated nomenclature.

DISCLOSURES

The research was supported by grants from the National Institutes of Health or the National Institute on Alcohol Abuse and Alcoholism. One author disclosed providing consulting services to AstraZeneca, Eisai, Exact Sciences, and Fujifilm Medical Sciences.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.