Fifty-one pigs were fed a low-cholesterol basal diet, to which either 10% (by weight) of lard fat (group INORM,n=l), 2% cholesterol plus 8% lard fat (group II,n=33), or 2% cholesterol plus 4% lard fat plus 4% fish oil (group IIIPREVIn=H) was added. In all pigs, the left anterior descending coronary artery and the abdominal aorta were denuded at 1 month. In the first 24 hours thereafter, three animals in group II and two in group IIIPREv died suddenly. After 3 months, 0.5% bile acids was added to the diet in groups II and IIIPREV After 8 months the degree of atherosclerosis was evaluated in groups INORM and IIIPREV and in 14 animals from group II (HIND)- At 4 months, one animal from Group II died of pneumonia. For the next 4 months (postinduction period), the remaining 15 animals from group II received the basal diet, to which either 10% lard fat (group 11,$,n=6) or 5% lard fat plus 5% fish oil (group IIro,n=9) was added. The hypercholesterolemic diet increased plasma cholesterol from 2 to 9-12 mM after 8 months. Fish oil had no major effects on plasma lipids during both induction and postinduction. Superoxide production by granulocytes in response to the membrane receptor-dependent A'-forniyl-methionyl-leucyl-phenylalanine (fMLP) gave a higher response in group IIINDthan in group INORM-In group IIIPREV > the response to phorbol myristate acetate (PMA) and fMLP was lowered, while in groups IITOand 11,$ the responses to PMA and fMLP were not affected. The response to serum-treated zymosan was similar in all groups. Abrasion caused increases in free cholesterol (40%) and phospholipids (46%) in the abdominal aortas of group INORM animals. Hypercholesterolemia increased both free and esterified cholesterol in the entire aorta. Fish oil prevented accumulation of free cholesterol in the nonabraded ascending aorta during induction and further accumulation of free cholesterol and phospholipids in the abdominal aorta during postinduction. In the nonabraded ascending aorta, triglycerides were significantly (almost five times) lower in group IIF0than in group II,j. During both induction and postinduction, a large incorporation of n-3 polyunsaturated fatty acids (up to 20%) occurred in plasma and aortic cholesterol esters and phospholipids of groups IIroand IIIPREV In plasma lipids the major fatty acid was always 20: Sn-3, but in aortic lipids the incorporation of its elongated product 22: 5n-3 was generally equal to 20: 5n-3. Sudanophilia of the aorta was 0% for group INORM > 33% for group II)ND, 35% for group HIPREV, 9% for group IIu, and 7% for group IIFO. Luminal encroachment of the coronary arteries in groups INORM and IIINDwas similar. Fish oil (group IIIPREV) had no effect on the development of coronary atherosclerosis. Despite the normocholesterolemia during the postinduction period in group IILF, coronary atherosclerosis progressed to 12% in the right coronary artery and to 18% in the left anterior descending coronary artery, while in group IIF0these values were 5% and 10%, respectively. In conclusion, isocaloric administration of n-3 fatty acid ethyl esters did not prevent the development of coronary atherosclerosis (luminal encroachment) or aortic atherosclerosis (sudanophilia and aortic lipid content). However, fish oil feeding during the postinduction period attenuated the progression of coronary atherosclerosis and the accumulation of lipids in the damaged aortic wall, while aortic atherosclerosis as measured by sudanophilia was not affected. This study shows that the criteria used to assess experimental atherosclerosis are essential in the evaluation of the effects of n-3 fatty acids.