The effect of alcohol intake on cardiovascular mortality has been described as a U- or J-shaped relationship; however, the exact mechanisms of reduction in mortality in moderate alcohol consumption are unknown. Albert and colleagues studied the association of alcohol and C-reactive protein (CRP) through a cross-sectional survey of patients participating in the Pravastatin Inflammation/CRP Evaluation (PRINCE) Study, which was a multi-center community-based trial that investigated the effect of pravastatin on CRP levels.
The study by Albert and colleagues included 1,679 patients without a history of myocardial infarction, stroke, or coronary revascularization (primary prevention arm) and 1,154 patients with a history of cardiovascular disease (secondary prevention arm). Participants were asked to define their alcohol intake in one of the following ways: no alcohol intake or less than one drink monthly; one to three drinks per month; one to four drinks per week; five to seven drinks per week; or at least two drinks daily. Baseline levels and mean change in CRP and lipid levels over six months were calculated for each alcohol category. The data were analyzed to compare baseline cardiovascular risk factors and alcohol consumption; to assess the relationship of alcohol intake and CRP overall and among various subgroups; to evaluate the correlation of alcohol and change in lipid parameters over time; and to assess for interactions between alcohol intake, pravastatin allocation, and change in CRP or lipid levels over time. CRP levels were higher in persons with a history of cardiovascular disease compared with those without such a history, and were higher in women than in men. In nearly all of the subgroups, moderate alcohol consumption (five to seven drinks weekly) and heavy alcohol consumption were significantly associated with the lowest baseline CRP levels. The relationship was less significant in women taking hormone therapy.
The authors acknowledge the following weaknesses in the study: the cross-sectional design limits direct determination of causality; self-reporting of alcohol consumption levels may have been inaccurate and may have resulted in underestimated true consumption; the types of alcoholic beverages were not described and may have theoretic limitations; and CRP levels were similar in moderate and heavy drinkers, possibly because of the smaller sample sizes in the heavy alcohol intake groups.
The authors conclude that alcohol use may lower cardiovascular mortality in part through an anti-inflammatory mechanism. The study further confirms the benefit of moderate alcohol use on mortality, but caution is advised when discussing the benefits with patients in light of the well-known health consequences associated with alcohol abuse.