LAUSR

Larger mean platelet volume (MPV) has been associated with adverse health outcomes in high-risk populations or patients with cardiovascular disease (CVD). We tested the association of MPV with mortality in a prospective cohort study including 17,402 subjects randomly recruited from an adult general population within the Moli-sani study (2005-2010). Two distinct subgroups (with or without CVD at baseline) were subsequently analysed. Hazard ratios (HR) were calculated using multivariable Cox-proportional hazard models. Over a median follow up of eight years (137,547 person-years), 925 all-cause deaths occurred (330 vascular, 351 cancer and 244 other deaths). In a multivariable model, the highest MPV quintile (mean MPV=10.0 fL), as compared to the lowest one, was associated with reduced risk of overall mortality (HR=0.79; 95 % confidence interval 0.64-0.98), cancer death (HR=0.70; 0.49-1.00) and death from other non- vascular/non cancer causes (HR=0.55; 0.36-0.84) but not with vascular mortality. The inverse association with overall death appeared even stronger in the subgroup without CVD at baseline (HR=0.64; 0.50-0.81). In contrast, within 920 subjects reporting a previous CVD event, larger MPV was associated with higher risk of total mortality (HR=1.69; 1.05-2.72; p for interaction=0.048) and with a trend of risk for other cause-specific deaths. In conclusion, larger MPV is associated with lower risk of overall and non-vascular death in subjects apparently free from CVD, but appears to be a predictive marker of death in patients with CVD history. The latter is a likely effect modifier of the association between MPV and death.