We previously reported that low protein Z plasma levels are associated with acute coronary syndromes (ACS). Aim of the present study was to evaluate protein Z levels in different phases of activity of coronary atherosclerosis. Protein Z plasma levels were measured in 166 (131 male and 35 female) patients consecutively admitted to the University of Florence with a diagnosis of ACS (group A), 166 (131 male and 35 female) patients selected by age and gender in relation to group A from those with a clinical history of ACS who remained symptom- and/or event-free over the last year before the investigation (group B); and 332 (262 male and 70 female) controls comparable for age and gender with the other two groups. None had liver or renal dysfunction nor showed a positivity for antiphospholipid antibodies or for factor V Leiden mutation. Patients under warfarin therapy were excluded. Mean protein Z plasma levels were found to be significantly (P < 0.0001) lower in group A (1475 +/- 684.1 ng mL(-1)) and group B (1327.6 +/- 690.7 ng mL(-1)) as compared with control group (1650.1 +/- 634.5 ng mL(-1)), while no significant differences existed between the two groups of patients (P = 0.06). A logistic regression analysis, performed after the division of the study population into quartiles of protein Z levels and adjusted for all possible confounders, showed a significant increased risk of ACS for the lowest (<1213 ng mL(-1)) as compared with the highest quartile of protein Z in both groups of patients [group A odds ratio (OR): 2.7, 95% CI 1.3-5.5, P = 0.007; group B OR: 3.2, 95% CI 1.1-8.9, P = 0.02). In conclusion, these results strengthen our previous data on low protein Z plasma levels in ACS and indicate a possible dose-response effect of decreasing protein Z plasma levels on the coronary atherosclerotic disease.