This study investigates the association of wartime stress variables and coronary artery disease as determined by coronary angiography in Lebanon in 1986, a country with an ongoing civil war for over a decade. A total of 127 patients who underwent coronary angiography at the American University of Beirut Medical Center were individually matched on age and sex with visitor controls free from any evidence of clinical coronary artery disease. Arteriographic cases (greater than or equal to 70% maximal stenosis) were compared with two control groups: arteriographic controls (entirely normal coronaries) and visitor controls. Findings suggest that there is a relation between exposure to both acute and chronic war events and coronary artery disease in this patient population. The reporting of exposure to acute war events was significantly higher in cases compared with both visitor controls (odds ratio (OR) = 2.4, 95% confidence interval (Cl) 1.17-4.90) and arteriographic controls (OR = 2.8, 95% Cl 0.93-8.47). Crossing the "green-lines" that separate two belligerent sides, considered as an attribute of war-related chronic stress, was more frequent in cases compared with visitor controls (OR = 3.25, 95% Cl 1.54-6.89) and arteriographic controls (OR = 5.38, 95% Cl 1.65-17.6). The relation observed between wartime stress and coronary artery disease could not be explained by possible overreporting of stressful events in patients with suspected coronary artery disease or by an increase in clinical awareness for the disease for those under continuous stress. Adjusting for the effect of the well-established coronary artery disease risk factors did not alter the above findings.