RESUMO
PURPOSE: Results from epidemiologic studies on coffee consumption and the risk of cutaneous melanoma are inconsistent. We conducted a meta-analysis to assess the associations between the consumption of total coffee, caffeinated coffee and decaffeinated coffee and the risk of cutaneous melanoma, respectively. METHODS: A literature search was performed in PubMed, Web of Science and EMBASE for relevant articles published up to August 2015. Pooled relative risks (RRs) with 95 % confidence intervals (CIs) were calculated with a random-effects model. Dose-response relationship was assessed by restricted cubic spline. RESULTS: Twelve studies involving 832,956 participants for total coffee consumption, 5 studies involving 717,151 participants for caffeinated coffee consumption and 6 studies involving 718,231 participants for decaffeinated coffee consumption were included in this meta-analysis. Compared with the lowest level of consumption, the pooled RRs were 0.80 (95 % CI 0.69-0.93, I (2) = 53.5 %), 0.85 (95 % CI 0.71-1.01, I (2) = 65.0 %) and 0.92 (95 % CI 0.81-1.05, I (2) = 0.0 %) for the consumption of total coffee, caffeinated coffee and decaffeinated coffee, respectively. In subgroup analysis by study design, the pooled RRs in cohort studies and case-control studies were 0.83 (95 % CI 0.72-0.97) and 0.74 (95 % CI 0.51-1.07) for total coffee consumption, respectively. Dose-response analysis suggested cutaneous melanoma risk decreased by 3 % [0.97 (0.93-1.00)] and 4 % [0.96 (0.92-1.01)] for 1 cup/day increment of total coffee and caffeinated coffee consumption, respectively. CONCLUSIONS: This meta-analysis suggests that coffee consumption may reduce the risk of cutaneous melanoma.