Safety considerations of polyunsaturated fatty acids

¹ From the Department of Cardiology, Ullevål University Hospital, Oslo.

² Reprints not available. Address correspondence to J Eritsland, Department of Cardiology, Ullevål University Hospital, N-0407 Oslo, Norway.

ABSTRACT  
The n–6 and n–3 polyunsaturated fatty acids (PUFAs) are essential nutrients; intake of relatively small amounts of these fatty acids prevents nutritional deficiencies. Replacing dietary saturated fat with PUFAs may confer health gains. Experimental data support the notion that high intake of n–6 PUFAs may increase in vivo lipid peroxidation. This effect may be counteracted by dietary antioxidant supplementation. The influence of a high n–3 PUFA intake on measures of lipid peroxidation has been equivocal. In clinical trials, subjects who consumed diets rich in n–6 or n–3 PUFAs had fewer atherothrombotic endpoints than did control groups. In this report, data regarding the influence of PUFAs on lipid peroxidation as well as on cholesterol and glucose metabolism, hemostasis, and other aspects of interest are reviewed and discussed. Currently, daily intake of PUFAs as >10% of total energy is not recommended. Below this ceiling there is little evidence that high dietary intake of n–6 or n–3 PUFAs implies health risks.

Key Words: Polyunsaturated fatty acids • adverse effects • lipid peroxidation • cholesterol • blood glucose

INTRODUCTION  
Polyunsaturated fatty acids (PUFAs) of the n–6 and n–3 families are necessary for proper growth and body function (1). They cannot be synthesized by humans and must be obtained from the diet (1, 2). Deficiency syndromes resulting from insufficient dietary n–6 and n–3 PUFAs have been described (3, 4); however, such syndromes can be prevented with low intakes of n–6 and n–3 PUFAs, which should be achieved in a normal diet (2). The optimal daily intakes of PUFAs, and the optimal ratio of n–6 to n–3 PUFAs, remain unknown. When addressing these questions, other dietary constituents must also be taken into account (2). Very high intakes of PUFAs may carry a risk of adverse effects; thus, relevant experimental and clinical data on safety aspects of PUFAs will be reviewed.

LIPID PEROXIDATION  
When exposed to oxidant stress, PUFAs can be attacked by free radicals and oxidized into lipid peroxides (5). The peroxidative breakdown of PUFAs involves chain reactions that result in a variety of products such as aldehydes, ketones, and cyclic peroxides (6, 7). These reactions may propagate and modify lipids and proteins, eg, in cell membranes and lipoproteins that contain PUFAs (5, 6). Although still not fully elucidated, lipid peroxidation is thought to be one important mechanism involved in the pathogenesis of inflammation, cancer, and atherosclerosis (5–7).

Foods containing lipid peroxides are potentially toxic, and the more PUFAs are present in the diet, the more likely is peroxidation (5, 7). However, the rancid and unpleasant taste of foods rich in peroxidation products (5) usually prevents the intake of large amounts of lipid peroxides. It is unresolved whether chronic intake of smaller amounts of peroxidation products in food or supplements containing PUFAs stored or processed under conditions allowing oxidation presents a health hazard (5, 7).

Oxidative modification of LDL is the best-substantiated example of in vivo lipid peroxidation. Oxidatively modified LDL (ox-LDL) is assumed to play an important role in atherosclerosis (8, 9). Compared with native LDL, ox-LDL has been shown to have several properties thought to promote the development of atherosclerosis, including uptake by the macrophage scavenger receptor, leading to the formation of foam cells (8, 9). There is evidence that LDL is oxidatively modified in vivo (8, 9), although it has been difficult to develop reliable methods to quantify such modification (5, 10).

The susceptibility of LDL to oxidative modification under conditions of artificial oxidative stress can be measured in vitro (11). Although the clinical relevance of such tests can be questioned (10), an association between the susceptibility of LDLs to oxidation in vitro and estimates of coronary atherosclerosis has been described (12). The situation in vivo, however, is much more complex than that under controlled in vitro conditions. In the organism, various defense mechanisms including enzymes, other proteins, and water- and lipid-soluble antioxidants act protectively against lipid peroxidation in the circulation and in tissues, ie, artery wall (7). The ideal test for ox-LDL formation should reflect the presence and amounts of ox-LDL in the arterial intima, but so far this is possible only in specimens obtained from animal studies or in pathoanatomical samples. Several methods based on analyses of plasma have been used to assess ox-LDL formation, but none is considered ideal (5, 10). Measurement of thiobarbituric acid–reactive substances (TBARS), thought to reflect the formation of malondialdehyde, is applied commonly in clinical settings. This test is sensitive and suitable for application in large study populations; however, it is rather nonspecific and has other limitations as well (5, 10). The general lack of reliable methods for assessing lipoprotein oxidation in the circulation and in artery walls may help explain conflicting results of in vitro testing and clinical data from PUFA dietary studies.

In several clinical dietary trials it was shown that when PUFA intake is increased, the PUFA content of the LDL particles increases concordantly (13–15). The in vitro susceptibility of LDLs to undergo oxidative modification was reported to increase when the diet is rich in n–6 PUFAs compared with monounsaturated fats (13, 14, 16–18). Also, an increase in plasma TBARS concentrations during an n–6 PUFA–enriched dietary period was reported (19). In some but not all studies, supplementation with antioxidants increased the resistance of the LDL particles to oxidation (14, 15, 20). At present, whether a diet high in n–3 PUFAs increases the oxidative modification of LDLs is controversial (21). In the large Shunt Occlusion Trial, patients were supplemented with 4 g n–3 PUFA concentrate (which also supplied 15 mg vitamin E) daily; compared with the nonsupplemented control group, no significant differences in serum TBARS concentrations were seen after 9 mo (22).

Thus, on the basis of in vitro assessments and best substantiated in vivo for n–6 PUFAs, increased dietary intake of PUFAs may enhance the susceptibility of LDLs to undergo oxidative modification. However, when it comes to clinical endpoints there is little evidence to suggest that a high PUFA intake increases the risk of an adverse outcome. On the contrary, data from intervention trials show that when saturated fats are replaced by n–6 PUFAs, subjects are less prone to develop atherothrombotic complications (23–25). Epidemiologic data on large intakes of n–6 PUFAs are lacking.

Traditional Mediterranean diets, to which health-promoting effects are attributed, are rich in monounsaturated fatty acids (26). Only recently has the use of plant-derived n–6 PUFAs increased in various populations, and the long-term effects of such dietary changes are unknown (26). The results of epidemiologic studies suggest that high n–3 PUFA intake reduces the risk for cardiovascular disease (27–30), and in intervention trials an increased n–3 PUFA intake was reported to act favorably on the incidence of atherothrombotic endpoints (31, 32). The Shunt Occlusion Trial showed a significant linear trend toward fewer aortocoronary vein graft occlusions with increasing positive changes in serum phospholipid n–3 PUFA concentrations (33).

Thus, regardless of the influence on LDL oxidation, there seems to be a net beneficial effect on clinical outcomes by enrichment with dietary PUFAs. Several factors may account for these observations. First, serum LDL-cholesterol concentrations tend to decline when saturated fatty acids are replaced with PUFAs in the diet (1). Second, PUFAs—in particular n–3 PUFAs—may have antiatherothrombotic effects on growth factors, cytokines, and signal molecules (34–37). Third, PUFA-rich food sources are often rich in antioxidants. Fourth, the microenvironment of the artery wall is different from that of the circulation and, evidently, widely different from in vitro conditions. Finally, as mentioned previously, the assessment methods for lipid oxidation in vivo have inherent limitations in reflecting the processes in the arterial intima.

The dietary requirement of antioxidants with a diet rich in PUFAs has not been defined. However, along with a PUFA-rich diet, it seems reasonable to encourage a high intake of antioxidants, preferably incorporated in the habitual diet.

EFFECTS ON SERUM CHOLESTEROL  
Generally, when saturated fatty acids are replaced with PUFAs in the diet, total and LDL-cholesterol concentrations decrease (1). In some studies in which saturated fatty acids were replaced with n–6 PUFAs and the ratio of polyunsaturated to saturated fat increased, an HDL-cholesterol-lowering effect was shown (38–40), although not consistently (41). However, because LDL-cholesterol concentrations also decline with such n–6 PUFA–rich diets (42), the ratio of LDL to HDL cholesterol may not change significantly (39, 40). When n–3 PUFAs were given as supplements to the habitual diet, LDL-cholesterol-raising effects were noted by some (43–45), whereas HDL-cholesterol concentrations were unchanged or increased slightly (43). In the Shunt Occlusion Trial, daily supplementation of 4 g highly purified n–3 PUFA concentrate/d did not significantly influence serum concentrations of total, HDL, or LDL cholesterol compared with the unsupplemented group (22). Such factors as characteristics of the subjects (ie, their hyperlipemic phenotypes and background diet) and of the study (ie, the type of supplement given, time of follow-up, and design details) may affect study outcomes significantly.

EFFECTS ON GLUCOSE HOMEOSTASIS  
In some n–3 PUFA–supplementation studies, a deterioration of glycemic control was reported in subjects with type 2 diabetes (46, 47). Data from later reports, however, do not support this conclusion (48, 49). Again, factors associated with study individuals, their habitual diet, the n–3 PUFA sources, and study design can be of relevance to explain conflicting results. In controlled trials in hypertensive (50) and hypertriglyceridemic (49, 51) individuals, no influence on glucose homeostasis was ascribed to n–3 PUFA supplementation. In the aforementioned Shunt Occlusion Trial, no effects of n–3 PUFAs on plasma glucose, serum insulin, or serum C-peptide concentrations were seen when the supplemented group was compared with the control group after 9 mo of follow-up (22).

EFFECTS ON HEMOSTASIS  
The bleeding tendency (eg, from the nose and urinary tract and obstetric bleedings) of traditionally living Greenland Eskimos was described (52). Later, prolonged cutaneous bleeding time and reduced platelet in vitro aggregability, compared with Danish control subjects, were reported (53, 54). These findings were associated with the unique diet of Greenland Eskimos, whose estimated n–3 PUFA intake is 7–10 g/d (2, 55). This inhibition of platelet function has been explained by a shift in eicosanoid metabolism when arachidonic acid (20:4n–6) is replaced by eicosapentaenoic acid (20:5n–3) in platelet membranes. This shift results in the generation of thromboxanes and prostacyclins which, overall, leads to a more vasodilatory and antiaggregatory hemostatic profile (1, 53). Also, a small—or, less often, a large—decline in platelet count during periods of high n–3 PUFA supplementation has been described (1, 54). In periods of moderate n–3 PUFA supplemention (2–5 g/d), however, there was no clinical evidence of an increased bleeding tendency (30). In the Shunt Occlusion Trial, all patients were treated with either aspirin or warfarin and followed clinically for 1 y. There were no more clinically detected bleeding episodes in the n–3 PUFA–supplemented group than in the control group (56). Also, there were no significant changes in a broad panel of hemostatic variables that could be attributed to the supplement, which consisted of 3.4 g eicosapentaenoic and docosahexaenoic acids/d (56). Thus, from a clinical point of view, a moderate n–3 PUFA supplement does not seem to increase the risk of bleeding, with a possible caveat for individuals with inherited or acquired hemorrhagic diathesis.

IMMUNOSUPPRESSION  
There is evidence of favorable effects of n–3 PUFAs on some immunologic and inflammatory disorders (57–60), perhaps through an influence on cytokine and leukotriene generation (35, 37, 61). Hypothetically, large intakes of n–3 PUFAs could weaken defense mechanisms against infections or malignancies, but so far there are no definite clinical data on these issues.

CARCINOGENESIS  
In several populations, a positive correlation between fat intakes and mortality rates from certain cancers, particularly of the breast but also of the colon and prostate, has been described (62–64). This association applies to intakes of total as well as saturated fat (63, 64), but on the basis of experimental animal models a minimum requirement of n–6 PUFAs seems necessary for tumorigenesis enhancement (63). Several mechanisms that could contribute to this putative tumorigenesis have been proposed, among them effects mediated by oxidation products, by the generation of leukotrienes, or by cellular membrane changes (63, 65, 66). It has also been proposed that the association between fat intake and cancer could be attributed to the trans fatty acid component (67), but this hypothesis is controversial (66). In animal models, n–3 PUFAs tend to be neutral or to inhibit tumorigenesis (63, 66).

In one intervention study, a higher incidence of fatal carcinomas was reported in the group with high n–6 PUFA intake compared with the control group (68). This finding has not been confirmed in reports from other studies (69). Thus, a certain influence of n–6 PUFAs on carcinogenesis cannot at present be excluded. Most investigators recommend that intake of linoleic acid not exceed 10% of total energy (70).

LIVER FUNCTION  
A slight increase in the serum activity of liver enzymes during n–3 PUFA supplementation has been noted repeatedly (22, 71). The mechanisms remain unclear and these observations are presumed to be of no clinical significance. Autopsy reports have provided opposing data on the incidence of cholesterol gallstones in n–6 PUFA trials (72, 73). No prospective observations on this outcome have appeared.

OTHER CONCERNS  
An increase in PUFA intake should be at the expense of saturated fat intake because in the long run, a substantial addition of energy-dense PUFAs to the habitual diet could result in weight gain and accompanying metabolic disorders (70). Regarding toxicity, commercial and concentrated PUFA products should report and declare potentially toxic substances such as heavy metals, organic pesticides, fat-soluble vitamins, and lipid peroxides, as well as the amounts of other fatty acids and cholesterol (1).

CONCLUSION  
Evidence suggests that dietary PUFAs, when substituted for saturated fatty acids, will confer net health gains, most notably in cardiovascular disease. An upper limit of 10% of energy for n–6 PUFAs is presently recommended by experts, but a palatable and practicable diet will usually not exceed this limit. It also seems wise to ensure a high dietary intake of antioxidants. The risk of adverse effects of dietary PUFAs, whether on cholesterol or glucose metabolism or hemostatic function (except for large doses of n–3 PUFAs), seems small.

REFERENCES  

1. Goodnight SH, Harris WS, Connor WE, Illingworth DR. Polyunsaturated fatty acids, hyperlipidemia, and thrombosis. Arteriosclerosis 1982;2:87–113.
2. Nordøy A. Is there a rational use for n–3 fatty acids (fish oils) in clinical medicine? Drugs 1991;42:331–42.
3. Hansen AE. Serum lipid changes and therapeutic effects of various oils in infantile eczema. Proc Soc Exp Biol Med 1933;31:160–3.
4. Bjerve KS, Fischer S, Wammer F, Egeland T. -Linolenic acid and long-chain -3 fatty acid supplementation in three patients with -3 fatty acid deficiency: effect on lymphocyte function, plasma and red cell lipids, and prostanoid formation. Am J Clin Nutr 1989;49:290–300.
5. Halliwell B, Chirico S. Lipid peroxidation: its mechanism, measurement, and significance. Am J Clin Nutr 1993;57(suppl):715S–25S.
6. Dargel S. Lipid peroxidation—a common pathogenetic mechanism? Exp Toxicol Pathol 1992;44:169–81.
7. Esterbauer H. Cytotoxicity and genotoxicity of lipid-oxidation products. Am J Clin Nutr 1993;57(suppl):779S–86S.
8. Steinberg D, Parthasarathy S, Carew TE, Khoo JC, Witztum JL. Beyond cholesterol. Modifications of low-density lipoprotein that increase its atherogenicity. N Engl J Med 1989;320:915–24.
9. Parthasarathy S, Steinberg D, Witztum JL. The role of oxidized low-density lipoproteins in the pathogenesis of atherosclerosis. Annu Rev Med 1992;43:219–25.
10. Chait A. Methods for assessing lipid and lipoprotein oxidation. Curr Opin Lipidol 1992;3:389–94.
11. Esterbauer H, Striegl G, Puhl H, Rotheneder M. Continuous monitoring of in vitro oxidation of human LDL. Free Radic Res Commun 1989;6:67–75.
12. Regnström J, Nilsson J, Tornvall P, Landou C, Hamsten A. Susceptibility to low-density lipoprotein oxidation and coronary atherosclerosis in man. Lancet 1992;339:1183–6.
13. Reaven P, Parthasarathy S, Grasse BJ, et al. Feasibility of using an oleate-rich diet to reduce the susceptibility of low-density lipoprotein to oxidative modification in humans. Am J Clin Nutr 1991;54:701–6.
14. Reaven P, Parthasarathy S, Grasse BJ, Miller E, Steinberg D, Witztum JL. Effects of oleate-rich and linoleate-rich diets on the susceptibility of low-density lipoprotein to oxidative modification in mildly hypercholesterolemic subjects. J Clin Invest 1993;91:668–76.
15. Lussier-Cacan S, Dubreuil-Quidos S, Roederer G, et al. Influence of probucol on enhanced LDL oxidation after fish oil treatment of hypertriglyceridemic patients. Arterioscler Thromb 1993;13:1790–7.
16. Reaven P, Grasse BJ, Tribble DL. Effects of linoleate-enriched and oleate-enriched diets in combination with alpha-tocopherol on the susceptibility of LDL and LDL subfractions to oxidative modification in humans. Arterioscler Thromb 1994;14:557–66.
17. Bonanome A, Pagnan A, Biffanti S, et al. Effect of dietary monounsaturated and polyunsaturated fatty acids on the susceptibility of plasma low density lipoproteins to oxidative modification. Arterioscler Thromb 1992;12:529–33.
18. Abbey M, Belling GB, Noakes M, Hirata F, Nestel PJ. Oxidation of low-density lipoproteins: intraindividual variability and the effect of dietary linoleate supplementation. Am J Clin Nutr 1993;57:391–8.
19. Berry EM, Eisenberg S, Haratz D, et al. Effects of diets rich in monounsaturated fatty acids on plasma lipoproteins—the Jerusalem Nutrition Study: high MUFAs vs high PUFAs. Am J Clin Nutr 1991;53:899–907.
20. Oostenbrug GS, Mensink RP, Hornstra G. Effects of fish oil and vitamin E supplementation on copper-catalysed oxidation of human low density lipoprotein in vitro. Eur J Clin Nutr 1994;48:895–8.
21. Nenseter MS, Drevon CA. Dietary polyunsaturates and peroxidation of low density lipoprotein. Curr Opin Lipidol 1996;7:8–13.
22. Eritsland J, Arnesen H, Seljeflot I, Høstmark AT. Long-term metabolic effects of n–3 polyunsaturated fatty acids in patients with coronary artery disease. Am J Clin Nutr 1995;61:831–6.
23. Leren P. The effect of plasma cholesterol lowering diet in male survivors of myocardial infarction. A controlled clinical trial. Acta Med Scand 1966;466(suppl):1–92.
24. Dayton S, Pearce ML, Hashimoto S, Dixon WJ, Tomiyasu U. A controlled clinical trial of a diet high in unsaturated fat in preventing complications of atherosclerosis. Circulation 1969;40(suppl):II-1–63.
25. Miettinen M, Turpeinen O, Karvonen MJ, Elosuo R, Paavilainen E. Effect of cholesterol-lowering diet on mortality from coronary heart-disease and other causes. Lancet 1972;2:835–8.
26. Willett WC, Sacks F, Trichopoulou A, et al. Mediterranean diet pyramid: a cultural model for healthy eating. Am J Clin Nutr 1995; 61(suppl):1402S–6S.
27. Bjerregaard P, Dyerberg J. Mortality from ischaemic heart disease and cerebrovascular disease in Greenland. Int J Epidemiol 1988;17:514–9.
28. Bang HO, Dyerberg J, Brøndum Nielsen A. Plasma lipid and lipoprotein pattern in Greenlandic west-coast Eskimos. Lancet 1971; 1:1143–6.
29. Bang HO, Dyerberg J, Hjørne N. The composition of food consumed by Greenland Eskimos. Acta Med Scand 1976;200:69–73.
30. Simopoulos A. -3 Fatty acids in health and disease and in growth and development. Am J Clin Nutr 1991;54:438–63.
31. Burr ML, Fehily AM, Gilbert JF, et al. Effects of changes in fat, fish, and fibre intakes on death and myocardial reinfarction: Diet and Reinfarction Trial (DART). Lancet 1989;2:757–61.
32. de Lorgeril, Renand S, Mamelle N, et al. Mediterranean alpha-linolenic acid-rich diet in secondary prevention of coronary heart disease. Lancet 1994;343:1454–9.
33. Eritsland J, Arnesen H, Grønseth K, Fjeld NB, Abdelnoor M. Effect of dietary supplementation with n–3 fatty acids on coronary artery bypass graft patency. Am J Cardiol 1996;77:31–6.
34. Fox PL, DiCorletto PE. Fish oils inhibit endothelial cell production of platelet-derived growth factor-like protein. Science 1988;241:453–6.
35. Endres S, Ghorbani R, Kelley VE, et al. The effect of dietary supplementation with n–3 polyunsaturated fatty acids on the synthesis of interleukin-1 and tumor necrosis factor by mononuclear cells. N Engl J Med 1989;320:265–71.
36. Harker LA, Kelly AB, Hanson SR, et al. Interruption of vascular thrombus formation and vascular lesion formation by dietary n–3 fatty acids in fish oil in nonhuman primates. Circulation 1993; 87:1017–29.
37. De Caterina R, Cybulsky MI, Clinton SK, Gimbrone MA, Libby P. The omega-3 fatty acid docosahexaenoate reduces cytokine-induced expression of proatherogenic and proinflammatory proteins in human endothelial cells. Arterioscler Thromb 1994;14:1829–36.
38. Shepherd J, Packard CJ, Patsch JR, Gotto AM, Taunton OD. Effects of dietary polyunsaturated and saturated fat on the properties of high density lipoproteins and the metabolism of apolipoprotein A-1. J Clin Invest 1978;61:1582–92.
39. Schaefer EJ, Levy RI, Ernst ND, Van Sant FD, Brewer HB Jr. The effects of low cholesterol, high polyunsaturated fat, and low fat diets on plasma lipid and lipoprotein cholesterol levels in normal and hypercholesterolemic subjects. Am J Clin Nutr 1981;34:1758–63.
40. Jackson RL, Kashyap ML, Barnhart RL, Allen C, Hogg E, Glueck CJ. Influence of polyunsaturated and saturated fats on plasma lipids and lipoproteins in man. Am J Clin Nutr 1984;39:589–97.
41. Iacono JM, Dougherty RM. Lack of effect of linoleic acid on the high-density-lipoprotein-cholesterol fraction of plasma lipoproteins. Am J Clin Nutr 1991;53:660–4.
42. Katan MB, Zock PL, Mensink RP. Dietary oils, serum lipoproteins, and coronary heart disease. Am J Clin Nutr 1995;61(suppl): 1368S–73S.
43. Harris WS. Fish oils and plasma lipid and lipoprotein metabolism in humans: a critical review. J Lipid Res 1989;30:785–807.
44. Harris WS, Dujovne CA, Zucker M, Johnson B. Effects of a low saturated fat, low cholesterol fish oil supplement in hypertriglyceridemic patients. Ann Intern Med 1988;109:465–70.
45. Fumeron F, Brigant L, Ollivier V, et al. n–3 Polyunsaturated fatty acids raise low-density lipoproteins, high-density lipoprotein 2, and plasminogen-activator inhibitor in healthy young men. Am J Clin Nutr 1991;54:118–22.
46. Glauber H, Wallace P, Griver K, Brechtel G. Adverse metabolic effect of omega-3 fatty acids in non-insulin-dependent diabetes mellitus. Ann Intern Med 1988;108:663–8.
47. Friday KE, Childs MT, Tsunehara CH, Fujimoto WY, Bierman EL, Ensinck JW. Elevated plasma glucose and lowered triglyceride levels from omega-3 fatty acid supplementation in type II diabetes. Diabetes Care 1989;12:276–81.
48. Connor WE, Prince MJ, Ullmann D, et al. The hypotriglyceridemic effect of fish oil in adult-onset diabetes without adverse glucose control. Ann N Y Acad Sci 1993;683:337–40.
49. Sirtori CR, Paoletti R, Mancini M, et al. n–3 Fatty acids do not lead to an increased diabetic risk in patients with hyperlipemia and abnormal glucose tolerance. Italian Fish Oil Multicenter Study. Am J Clin Nutr 1997;65:1874–81.
50. Toft I, Bønaa KH, Ingebretsen OC, Nordøy A, Jenssen T. Effects of n–3 polyunsaturated fatty acids on glucose homeostasis and blood pressure in essential hypertension. Ann Intern Med 1995;123:911–8.
51. Eritsland J, Seljeflot I, Abdelnoor M, Arnesen H, Torjesen PA. Long-term effects of n–3 fatty acids on serum lipids and glycaemic control. Scand J Clin Lab Invest 1994;54:273–80.
52. Bang HO, Dyerberg J. The bleeding tendency in Greenland Eskimos. Dan Med Bull 1980;27:202–5.
53. Dyerberg J, Bang HO, Stoffersen E, Moncada S, Vane JR. Eicosapentaenoic acid and prevention of thrombosis and atherosclerosis? Lancet 1978;2:117–9.
54. Dyerberg J, Bang HO. Haemostatic function and platelet polyunsaturated fatty acids in Eskimos. Lancet 1979;2:433–5.
55. Bang HO, Dyerberg J. Fish consumption and mortality from coronary heart disease. N Engl J Med 1985;313:822–3 (letter).
56. Eritsland J, Arnesen H, Seljeflot I, Kierulf P. Long-term effects of n–3 polyunsaturated fatty acids on haemostatic variables and bleeding episodes in patients with coronary artery disease. Blood Coagul Fibrinolysis 1995;6:17–22.
57. van der Heide JJ, Bilo HJG, Donker JM, Wilmink JM, Tegzess AM. Effect of dietary fish oil on renal function and rejection in cyclosporine-treated recipients of renal transplants. N Engl J Med 1993;329:769–73.
58. Donadio JV, Bergstralh EJ, Offord KP, Spencer DC, Holley KE. A controlled trial of fish oil in IgA nephropathy. N Engl J Med 1994;331:1194–9.
59. Shahar E, Folsom AR, Melnick SL, et al. Dietary n–3 polyunsaturated fatty acids and smoking-related chronic obstructive pulmonary disease. N Engl J Med 1994;331:228–33.
60. Belluzzi A, Brignola C, Campieri M, Pera A, Boschi S, Miglioli M. Effect of an enteric-coated fish-oil preparation on relapses in Crohn's disease. N Engl J Med 1996;334:1557–60.
61. Lee TH, Hoover RL, Williams JD, et al. Effect of dietary enrichment with eicosapentaenoic and docosahexaenoic acids on in vitro neutrophil and monocyte leukotriene generation and neutrophil function. N Engl J Med 1985;312:1217–24.
62. Willett WC. Diet and health: what should we eat? Science 1994;264:532–7.
63. Carroll KK, Braden LM, Bell JA, Kalamegham R. Fat and cancer. Cancer 1986;58:1818–25.
64. Rose DP, Boyar AP, Wynder EL. International comparisons of mortality rates for cancer of the breast, ovary, prostate, and colon, and per capita food consumption. Cancer 1986;58:2363–71.
65. Carroll KK, Parenteau HI. A proposed mechanism for effects of diet on mammary cancer. Nutr Cancer 1991;16:79–83.
66. Carroll KK. The role of dietary fat in breast cancer. Curr Opin Lipidol 1997;8:53–6.
67. Enig MG, Munn RJ, Keeney M. Dietary fat and cancer trends—a critique. Fed Proc 1978;37:2215–20.
68. Pearce ML, Dayton S. Incidence of cancer in men on a diet high in polyunsaturated fat. Lancet 1971;1:464–7.
69. Ederer F, Leren P, Turpeinen O, Frantz ID. Cancer among men on cholesterol-lowering diets: experience from five clinical trials. Lancet 1971;2:203–6.
70. The Expert Panel. Report of the National Cholesterol Education Program Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults. Arch Intern Med 1988;148:36–69.
71. Schmidt EB, Møller JM, Svaneborg N, Dyerberg J. Safety aspects of fish oils. Drug Invest 1994;7:215–20.
72. Sturdevant RL, Pearce ML, Dayton S. Increased prevalence of cholelithiasis in men ingesting a serum-cholesterol-lowering diet. N Engl J Med 1973;288:24–7.
73. Miettinen M, Turpeinen O, Karvonen MJ, Paavilainen E, Elosuo R. Prevalence of cholelithiasis in men and women ingesting a serum-cholesterol-lowering diet. Ann Clin Res 1976;8:111–6.