- Email: [email protected]
Comparison of diets supplemented with fish oil or olive oil on plasma lipoproteins in insulin-dependent diabetics
Comparison
of Diets Supplemented With Fish Oil or Olive Oil on Plasma Lipoproteins in Insulin-Dependent Diabetics
Trevor A. Mori, Robert Vandongen, John R.L. Masarei, Ian L. Rouse, and Diana Dunbar This study was designed to compare changes in high-density lipoprotein (HDL)- and low-density lipoprotein (LDL)-cholesterol in normolipidaemic male insulin-dependent diabetics (IDD) following dietary supplementation with either the fish oil concentrate Max EPA or olive oil. The contribution of the small quantity of cholesterol in Max EPA to these changes was also examined. Twenty-seven subjects were matched in groups of three and randomly allocated to one of three treatment groups of nine subjects each. Subjects were given 15 l-g capsules of oil daily for 3 weeks, consisting of either Max EPA, olive oil, or olive oil to which was added the same amount of cholesterol as contained in Max EPA, respectively. There was a significant increase in eicosapentaenoic acid, and a decrease in arachidonic acid, in the platelet membrane phospholipids of subjects taking Max EPA. In this group, there was an approximately 30% increase in serum HDL,-cholesterol (0.59 2 0.07 to 0.77 ? 0.11 mmol/L, mean + SEM; P < .Ol) and a corresponding decrease in HDl+cholesterol (0.79 * 0.03 to 0.71 ? 0.03 mmol/L; P < .05). Although total and LDL-cholesterol concentrations were also higher after Max EPA, the changes were not significant. Triglycerides were significantly decreased by Max EPA. There were no significant changes in lipids in the groups given olive oil. These results show that compared with olive oil, dietary supplementation with Max EPA substantially increases HDL,cholesterol in insulin-dependent diabetics. This is most likely due to a selective effect of ~3 fatty acids. An increase of this proportion in the cardioprotective HDL,-cholesterol, if sustained over time, could have a major impact on vascular health in diabetics. Copyright 0 1991 by W.B. Saunders Company
D
IETS in insulin-dependent diabetics should aim to improve glucose tolerance and to reduce hyperlipidemia and therefore the risk of vascular complications frequently encountered in this disease. Epidemiological evidence suggests that eating fish, and in particular fish oils rich in w3 fatty acids, may lower the incidence of atherosclerotic vascular disease. This effect has been attributed to an improvement in the plasma lipid profile and in the indices of thrombosis and inflammation.’ Previous studies comparing fish oil with vegetable oil supplements in hyperlipidemic subjects have shown either
no change,‘-4 or an increase in high-density lipoprotein (HDL), and particularly HDL2-cholesterol5 Since vegetable oils containing 06 polyunsaturated fats, such as linoleic actd, have been reported to lower HDL-cholesterol,6 a comparison with the effect of fish oil may not be strictly valid. A more appropriate comparison would be with olive oil, which causes little if any reduction in HDL-cholesterol levels,’ although the effect on HDL subfractions appears not to have been studied in any detail. However, a recently published report8 found that diets enriched with either monounsaturated or polyunsaturated oil lowered HDL equally in healthy subjects, although the changes were not significant. Because of the continuing uncertainty about the effect of monunsaturated fats on HDL-cholesterol, a comparison with the apparently selective effect of fish oil in increasing HDL,-cholesterol was considered important. Diabetics were selected as subjects for this study, because their predisposition to hyperlipidemia and vascular disease makes them likely targets for dietary advice aimed at improving plasma lipoprotein levels. SUBJECTS AND METHODS Twenty-seven male insulin-dependent diabetics, aged 21 to 43 years, who were nonsmokers, were recruited from the Diabetic Clmic at Royal Perth Hospital. The protocol was approved by the Hospital’s Human Rights Committee. Seven patients had retinopathy. whereas no complications were clinically detectable in the remainder. The duration of diabetes ranged from 3 to 34 years. All Metabolism,
Vol40,
No 3 (March),
1991:
pp 241-246
participants were fully informed about the nature of the study and gave written consent. Nonfasting serum cholesterol and triglyceride concentrations in all patients were less than 6.5 and 3.0 mmol/L, respectively, and none were taking any lipid-lowering agents. As the purpose of the study was to examine the effect of fish oil as a dietary supplement, no attempt was made to change or standardize the composition of the individual‘s background diabetic diet. However, subjects were requested not to change their current eating patterns or food intake. Any changes in diet. alcohol intake, physical activity, insulin requirements. and medication taken were assessed by means of a weekly questionnaire. Baseline serum lipids, plasma creatinine. plasma glucose, and platelet phospholipid fatty acid concentrations were measured approximately 2 hours after a light standardized breakfast. Serum lipids were measured on two occasions, 1 week apart, and the results averaged to obtain baseline values. Subjects were deliberately not fasted to minimize interference with diabetic control. They were matched in groups of three. as closely as possible for age and body mass index (BMI) calculated as weight/height,’ and then randomly allocated to one of three treatment groups, each containing nine subjects. Diets were supplemented daily with 15 g of encapsulated oil (1.0-g capsules) for 3 weeks: group A received Max EPA (Reckitt &Colman Pharmaceuticals, Sydney, Australia), group B olive oil (Reckitt & Colman); and group C olive oil enriched with cholesterol to the same concentration as Max EPA (prepared and encapsulated by R.P. Scherer. Melbourne, Australia). This latter group was included to determine a possible contribution of the small amount of cholesterol in Max EPA on plasma lipoprotein concentrations. The study was carried out double-blind with the subjects unaware of the contents of their oil capsules, which were identical in appearance.
From the University Department ofMedicine. Royal Perth Hospital, Perth, Western Australia. Suppotied by grants from the Diabetes Research Foundatton of Western Australia and the National Heart Foundation ofAustralia. Address reprint requests to TrevorA. Mori, PhD, University Department of Medicine, Medical School, Rqyal Perth Hospital, 35 Victoria Square, Perth, Western Australia, 6000. Copyrikht 0 1991 by W.B. Saunders Compuny 0026-049519114003-0005$03.00/0 241
MORI ET AL
242
The fatty acid and cholesterol composition of the oil is shown in Table 1 and the daily amounts received in Table 2. The fatty acid composition and cholesterol content was verified by gas liquid chromatography (GLC) (Hewlett-Packard Model 5980A) and the vitamin A and E content (92 and 526 p&g, respectively) by high-performance liquid chromatography.’ These vitamins were not added to the olive oil preparations. The above measurements were repeated, at approximately the same time of day, at the end of the 3 weeks’ dietary treatment and again 6 weeks after this period.
Table 2. Daily Amounts of Saturated, Monounsaturated,
w6 and ~3
Fatty Acids, and Cholesterol Provided by the Dietary Oils Olive Oil + Max
EPA
Olive Oil
(Group Al
Saturated (g)
Cholesterol
(Group B)
(Group C)
4.2
1.9
1.9
4.3
12.3
12.3
~6 (g)
0.3
0.8
0.8
03 (9)
5.2 0.0
51.0
Monounsaturated
(g)
Cholesterol (mg)
50.0
Plasma Lipid Levels Lipids from serum frozen at -20°C were assayed at the completion of the study. Cholesterol and triglyceride levels were determined enzymatitally on an Abbott ABA-100 bichromatic analyser with reagents from Abbott (Pasadena, CA), and standardized with serum-based calibrators that were obtained from the Australian Lipid Standards Program with values that were traceable to the Centers for Disease Control (Atlanta, GA). The coefficients of variation (CV) were 1.5% and 2.0%, respectively. Serum HDL-cholesterol was assayed on a heparin-manganese chloride supernatant.‘O HDL-cholesterol subfractions HDL, and HDL, were measured by the double precipitation method.” Mean values and SDS were calculated on 42 separate runs with aliquots of serum kept at -70°C: mean total HDL-cholesterol level, 1.61 ‘_ 0.057 (SD) mmol/L (CV, 3.5%); HDL,-cholesterol level. 0.734 + 0.063 mmol/L (CV, 8.6%), and HDb-cholesterol level, 0.874 ? 0.057 mmol/L (CV, 6.5%). Low-density lipoprotein (LDL)cholesterol levels were calculated from the Friedewald formula.” Platelet Phospholipid Fatty Acids Platelets were prepared as described previously’ and total phospholipids separated from the crude lipid extracts by thin-layer chromatography. Fatty acid methyl esters were prepared by treatment of samples with 4% H,SO, in methanol at 90°C for 20 minutes, analyzed by gas chromatography, and subsequently identified by comparison with authentic standards. The amounts of individual fatty acids were calculated as relative percentages with the evaluated fatty acids set at 100%. Statistical Evaluation Data were analyzed on SAS using repeated measures ANOVA. When a significant treatment-time interaction was demonstrated, Table 1. Percentage Composition of the Major Fatty Acids and Cholesterol in the Dietary Oils Olive Oil + Max EPA
Olive Oil
Cholesterol
Fatty acid 14:o
6.90
16:0
15.78
10.02
10.04
16:l (07)
9.30
0.70
0.70
190
2.95
2.50
2.57
l&l
(09)
13.84
81.49
81.19
18:2 (06)
0.99
5.29
5.50
18:3 (03)
0.48
20:4 (06)
1.04
205 (w3)
18.96
22:5 (~3)
2.71
22:6 (~3)
12.23
Cholesterol
0.33
Duncan’s modified t test was used to compare between-group differences at each visit. RESULTS
From Table 3, it is seen that the groups were comparable
for age, BMI, plasma glucose, insulin dose, and renal function. All subjects complied with the protocol and there were no major changes in diet, alcohol intake, physical activity, or insulin requirements. There were no significant (P > .05) changes in weight in any of the groups between baseline and the end of the 3 weeks’ treatment (A, 70.0 ? 2.5 to 71.2 ? 2.6; B, 80.3 ? 2.6 to 79.5 f 2.7; C, 74.3 2 3.0 to 76.7 ? 2.6 kg). Apart from occasional reports of a feeling of fullness and mild epigastric discomfort, no other side effects were encountered. One subject in group B withdrew during the supplementation period for reasons unrelated to the study. Compliance with the protocol was 99% in group A, 98% in group B, and 99% in group C as ascertained by capsule count. This was confirmed in group A, taking Max EPA, by the incorporation of EPA (20:5) into platelet phospholipids (Table 4) and a corresponding reduction in arachidonic acid (20:4). Max EPA supplementation also resulted in significant decreases in platelet o-6 fatty acids 20:3 and 22:4, and significant increases in the 03 fatty acids 2215 and 22:6. The platelet fatty acid composition in groups B and C was not altered by the treatment, with the exception of a small and insignificant increase in oleic acid (18:l) (14.02 2 0.53 to 14.94 + 0.24 and 13.29 2 0.18 to 14.17 ? 0.18, respectively). Samples for fatty acid analysis were not obtained from one subject each in group A and B. Serum Cholesterol and Triglycerides
Samples were not collected at all visits from one subject in group B who was deleted from the analysis. HDL,- and HDL,-cholesterol were not isolated and the Friedewald formula could not be applied for the calculation of LDL-
Table 3. Clinical and Biochemical Characteristics of Subjects in Groups A (Max EPA), B (Olive Oil), and C (Olive Oil and Cholesterol)
0.34
Plasma Creatinine
(mmol/L)
(mmol/L)
Age
EIMI
A
34.4 f 1.9
24.1 2 0.4
12.0 +- 2.2
51 + 5
87.7 k 3.7
6
33.8 + 2.1
24.7 + 0.8
12.2 2 2.7
46?
4
86.4 lr 5.4
C
34.2 k 1.9
24.5 2 0.8
10.7 + 1.9
50?
6
87.2 + 4.2
Group
-
Plasma Glucose
NOTE. Values are means + SEM.
FISH OIL RAISES
HIGH-DENSITY
LIPOPROTEIN
CHOLESTEROL
243
GROUP
Table 4. Percentage Composition of Fatty Acids in Platelet Phospholipids in Group A (Max EPA) (n = 6) at Baseline (week 0).
C
6
A
After 3 Weeks of Treatment With Max EPA (week 3), and 6 Weeks After the Treatment
-
BaselIne htty 16:0
Acid
DMA
16:0 18:0 DMA
1.08 + 0.21
+ 0.35
12.45
2.38 + 0.44
r
(Week 9)
12.43
2.62 k 0.36
1.35 2 0.19
-t 0.40
17.88
+ 0.40
18.09
k 0.22
18:l
13.92
2 0.33
14.75
+ 0.34
13.56
+ 0.32
20:3
5.52 + 0.33
1.62 2 0.12
20:4
26.48
5.79 t 0.28
1.18 + 0.08*
-t 0.55
22.32
1.67 + 0.12
t 0.63*
26.34
20:5
4.70 +- 0.30
7.56 k 0.39*
5.33 + 0.30
2.97 + 0.20
1.64 -+ 0.18*
3.08 2 0.21
i2:5
5.09 + 0.39
6.32 k 0.29t
5.71 t 0.25
22:6
4.72
6.67 2 0.26*
5.94 k 0.29*
NOTE.
Values
Abbreviation: For significance sures ANOVA):
are means DMA,
5g
l.O-
$
0.8-
p % d
0.6-
J IN
0.4-
k 0.72
22~4
+ 0.39
p
2 0.47
18.47
6.08 -t 0.36
1.2 -
0.69 -t 0.12
2 0.66
18:0 18:2
-
(Week 3)
0.82 -t 0.26
1.4
Posttreatment
Treatment
(Week 0)
12.77
Period (week 9)
$
0.2-
t SEM.
”
dimethylacetal.
of difference
0
from
baseline
values
(repeated
3
9
0
mea-
3
9
0
3
‘P -< .Ol; tP < .05.
Fig 1. Individual serum HDL,-cholesterol in patients receiving Max EPA (group A) (n = 9) at baseline (week 0). after 3 weeks of treatment (week 3), and after the treatment period (week 9).
cholesterol in another subject because triglyceride levels exceeded 4.0 mmol/L. As shown in Table 5, there was a tendency for serum total and LDL-cholesterol concentrations to increase following Max EPA treatment (group A), similar to that previously reported when the changes reached statistical significance.’ However, triglyceride levels were significantly lowered by Max EPA (representing a mean decrease of 35%), there being no changes in the other groups (Table 5). Although HDL-cholesterol as a whole was not altered by the dietary oil supplements in any of the groups (Table S), there was a substantial, and approximately 30%, increase in the HDL, subfraction (P < .Ol) in the Max EPA group compared with insignificant increases of 7% and 2% in
groups B and C, respectively. The individual responses in the three groups are illustrated in Fig 1. The changes in triglycerides and HDLZ-C in group A were not significantly correlated. HDL,-cholesterol decreased, but not significantly, in group A, representing a mean decrease of 10%. There were no changes in groups B or C (Table 5). DISCUSSION
These results demonstrate a unique and selective effect of Max EPA fish oil in raising the level of HDLL-cholesterol
Table 5. Serum Lipid Levels (mmol/L) at Baseline (week 0). After 3 Weeks of Treatment
(week 3). and 6 Weeks After the Treatment
Period
(week 9) in the Three Groups: Group A (Max EPA), B (Olive Oil), and C (Olive 011 and Cholesterol) Week Group
0
3
9
0
3
Cholesterol
9
Triglycerides
A
4.93
2 0.19
5.20 k 0.30
4.98 2 0.23
1.39 + 0.12
0.19 i- 0.11*
1.36 -t 0.17
B
4.86 2 0.28
5.04 * 0.34
5.08 k 0.29
1.23 t 0.18
1.14 + 0.16
1.37 + 0.19
C
5.08 5 0.25
5.23 k 0.38
4.98 5 0.39
1.00 + 0.08
1.18 t 0.16
1.22 f 0.18
A
2.90
? 0.18
3.30 k 0.30
2.96 2 0.25
1.39 k 0.08
1.48 k 0.11
1.39 + 0.10
6
2.61
-t 0.16
2.83 2 0.25
2.85 _f 0.20
1.41 k 0.06
1.46 t 0.08
1.38 + 0.08
C
3.19
2 0.25
3.28 + 0.37
3.11 + 0.39
1.43 -t 0.05
1.41 + 0.06
1.30 + 0.08
A
0.59 k 0.07
0.77 t 0.11*
0.60 k 0.09
0.79
0.71 + 0.03t
0.70 % 0.04
B
0.59 k 0.03
0.63 5 0.05
0.61 + 0.04
0.85 r 0.04
0.84 + 0.05
0.81 ? 0.06
C
0.59 ? 0.04
0.58 2 0.06
0.51 + 0.05
0.84 + 0.03
0.82 2 0.03
0.79 + 0.04
LDL-Cholesterol
HDL-Cholesterol
HDLJholesterol
NOTE.
Values
For significance modified (P r .Ol).
t test):
are means
+ SEM. Group
of difference group
A differs
from
group
HDL,-Cholesterol + 0.03
A, n = 9; B, n = 8; C, n = 9.
baseline
values
(repeated
B, tP < .05. A significant
9
WEEKS
measures
ANOVA):
treatment-time
lP -c .Ol. Between-group interaction
was
present
differences for triglycerides
at each visit (Duncan’s and HDL,-cholesterol
244
in insulin-dependent diabetics. Furthermore, the findings support our previous observation in diabetic patients of a smaller increase in serum total and LDL-cholestero1,‘3.‘4 but the changes on this occasion failed to reach statistical significance. The decrease in plasma triglycerides is the most consistent effect of dietary fish oils in both normall and hyperlipidemic16 subjects, although it was not seen in another study in insulin-dependent diabetics following an almost identical Max EPA regimen.” However, as triglyceride concentrations were measured in nonfasting blood samples, our findings may not be strictly comparable to those of others. Impairment of very-low-density lipoprotein (VLDL) synthesis,” or increased remova1,‘9 through activation of lipoprotein lipase, are possible mechanisms for the decrease in triglycerides and increase in HDL-cholesterol induced by fish oils. Activation of activity of lipoprotein lipase, by increasing the breakdown of triglyceride-rich lipoproteins and lowering serum triglyceride concentration, could facilitate the transfer of surface lipids from these particles to HDL,” and the conversion of HDL, to HDL,. It was recently suggested that in situations where HDbcholesterol concentrations were elevated, the lipoprotein particles may be preferentially enriched with apolipoprotein E,” which appears to lower platelet sensitivity to aggregation. This could be an intriguing explanation for the reduction in platelet aggregation observed in several fish oil studies.’ Body weight remained constant and the changes in plasma lipids were not accompanied by major alterations in diet or in other factors influencing HDL-cholesterol, in particular exercise” and alcohoLZ3 which may independently increase the level of this lipoprotein. The choice of placebo or control preparation in dietary studies of this nature is limited by the availability of oils having no demonstrable effects on lipid metabolism. Nevertheless, the design of the study, comparing the changes in lipids following Max EPA with olive oil, and olive oil enriched with cholesterol, clearly identifies the w3 fatty acids as the components most likely responsible for the 30% increase in HDb-cholesterol. The addition of a small quantity of cholesterol to olive oil, which is otherwise free of cholesterol, had no further effect on plasma lipids. Therefore, it would be reasonable to conclude that the small amount of cholesterol present in Max EPA plays no part in increasing HDL,-cholesterol, or altering the concentration of other plasma lipids, as shown by the uniform increase in LDL-cholesterol in the three groups. In fact, w3 fatty acids have been shown to attenuate the hypercholesterolemic effect of dietary cholestero1.24 Similarly, linoleic acid and oleic acid are unlikely to be involved, except for the remote possibility of these fatty acids having a biphasic action with the low amounts present in Max EPA raising HDL,-cholesterol and the higher quantities in olive oil having no effect. It is of interest that no significant changes in either total or LDL-cholesterol were found in the olive oil-treated groups. This is in contrast to the reduction in cholesterol observed in mildly hypercholesterolemic patients given a
MORI ET AL
high monounsaturated fat diet, principally in the form of oleic acid.25These different responses may be related to the amount of oleic acid given in the latter study (28% of total daily calories), the pre-diet lipid levels, and the absence of diabetes. In addition to effects on plasma lipoproteins, the fatty acids also differed in their degree of incorporation into platelet membrane phospholipids. EPA was most avidly incorporated, displacing arachidonic acid and remaining higher than baseline levels even 6 weeks after Max EPA supplementation was ceased. Docosapentaenoic acid (225) and docosahexaenoic acid (22:6), both present in Max EPA as carbon chain elongation and desaturation products of EPA, were also significantly increased after 3 weeks of Max EPA. The latter (22:6) remained significantly higher than baseline 6 weeks after ceasing Max EPA. Similarly, the w6 fatty acids 20:3 and 22:4, which are the immediate precursor and metabolite, respectively, of arachidonic acid, were significantly decreased during Max EPA administration. There was only a small increase in the percentage of phospholipid oleic acid in the olive oil groups, indicating a high degree of selectivity of fatty acid incorporation into platelet membranes. The change in membrane composition resulting from a more selective uptake of 03 fatty acids may have important functional implications in relation to platelet aggregation and the release of secretory products.26 The reported effect of fish oil on plasma total cholesterol has been inconsistent, due largely to differences in the amount given and in the pretreatment cholesterol concentration. Large quantities of fish oil, equivalent to at least 10 g of EPA daily over several weeks, generally results in a decrease in cholesterol in normal subjects,27.Win patients with hyperlipidemia of the type IIb phenotype,16 and, in particular, in patients with gross hypertriglyceridemia as in the type V category.‘6.Z9 Lower amounts of fish oil, comparable to those used in this study, have not usually altered lipids in normolipidemic subjects,26 although a slight increase in HDL-cholesterol, due to the HDL, subfraction, has been reported.30 In insulin-dependent diabetics, significant increases in total and LDL-cholesterol were observed in an earlier study from our laboratory.13 Similar trends were presented in this report, but the changes were not significant, perhaps due to inadequate sample size and statistical power. Increases in LDL-cholesterol have also occurred in patients with hypertriglyceridemia’,‘6.29 and have been attributed to increased production of LDL particles, rather than higher cholesterol content per particle, from smaller VLDL particles, which are formed in preference to larger particles during fish oil feeding.z,3’,3’ It is uncertain if the effect of fish oil in raising HDL,cholesterol is confined to patients with diabetes, as suggested by our earlier study,14 or if it also occurs in other populations. A similar increase in HDL,-cholesterol was recently reported in a small uncontrolled trial where insulin-dependent women were given 6 g of o-3 fatty acids daily for 10 weeks.33 Of interest was the additional finding
FISH OIL RAISES HIGH-DENSITY
LIPOPROTEIN
245
CHOLESTEROL
that postheparin plasma lipolytic activity did not change, suggesting that enhanced VLDL removal does not explain the increase in HDLZ-cholesterol. In epidemiological terms, the 30% increase in HDL,-cholesterol (equivalent to an average of 0.15 mmol/L) is substantial and could be expected to have a marked impact on the incidence of cardiovascular disease if indeed this is inversely related to the plasma level of this subfraction3“ To date, the evidence I’OI the apparent benefit associated with higher HDLcholesterol, with a 0.026 mmol/L increment predicted to reduce coronary risk by 2% to 3%,35 is confined to HDL as a
whole, rather than to its subfractions. The potential benefit of the increase in HDL,-cholesterol may need to be considered carefully against the small increase in LDLcholesterol. However, it may be concluded that the plasma lipoprotein profile in insulin-dependent diabetics was more favorably affected by a fish oil-enriched than by an olive oil-enriched diet. ACKNOWLEDGMENT
We gratefully acknowledge the technical assistance of Lynette Kelly.
REFERENCES Vandongen R: Fish oils and cardiovascular disease. Med J Aust 146:236-237, 1987 2. Harris WS, Dujovne CA, Zucker M, et al: Effects of a low saturated fat, low cholesterol fish oil supplement in hypertriglyceridaemic patients. Ann Intern Med 109:465-470,1988 .T Harris WS, Zucker ML, Dujovne CA: Omega-3 fatty acids in hypertriglyceridemic patients: Triglycerides vs methyl esters. Am J Clin Nutr 48:992-Y97, 1988 L. Sanders TAB, Sullivan DR. Reeve J, et al: Triglyceridelowering effect of marine polyunsaturates in patients with hypertriglyceridemia. Arteriosclerosis 5:459-465, 1985 5. Demke DM, Peters GR, Linet 01, et al: Effects of a fish oil concentrate in patients with hypercholesterolemia. Atherosclerosis ‘0:73-SO,1988 (1. Mattson FH, Grundy SM: Comparison of effects of dietary saturated, monounsaturated and polyunsaturated fatty acids on plasma lipids and lipoproteins in man. J Lipid Res 26:194-202,1985 -. Grundy SM: Comparison of monounsaturated fatty acids and carbohydrates for lowering plasma cholesterol. N Engl J Med 314:745-748. 1986 8. Mensink RP, Katan MB: Effect of a diet enriched with monounsaturated or polyunsaturated fatty acids on levels of low-density and high-density lipoprotein cholesterol in healthy women and men. N Engl J Med 321:436-441, 1989 (J. Mori TA, Codde JP. Vandongen R, et al: New findings in the t‘at:y acid composition of individual platelet phospholipids in man after dietary fish oil supplementation. Lipids 22:744-750, 1987 0. Warnick GR, Albers JJ: A comprehensive evaluation of the heparin-manganese precipitation procedure for estimating high density lipoprotein cholesterol. J Lipid Res 19:65-76, 1978 1. Gidez LI, Miller GJ, Burstein M, et al: Separation and quantitation of subclasses of human plasma high density lipoproteins by single precipitation procedure. J Lipid Res 23:1206-1223. 1982 2. Friedewald WT. Levy RI, Fredrickson DS: Estimation of the concentration of low-density lipoprotein cholesterol in plasma without use of the preparative ultracentrifuge. Clin Chem 18:49950’. 1972 i 3. Vandongen R, Mori TA, Codde JP, et al: Hypercholesterolaemic effect of fish oil in insulin-dependent diabetic patients. Med J Aust 148:141-143,198s : 4. Mori TA. Vandongen R, Masarei JRL, et al: Dietary fish oils increase serum lipids in insulin-dependent diabetics compared with healthy controls. Metabolism 38:404-409, 1989 15. Sanders TAB, Roshanai F: The influence of different types of omega-3 fatty acids on blood lipids and platelet function in healthy volunteers. Clin Sci 64:91-99, 1983 26. Phillipson BE, Rothrock DW, Connor WE, et al: Reduction of plasma lipids. lipoproteins and apoproteins by dietary fish oil in
patients with hypertriglyceridaemia. N Engl J Med 312:1210-1216, 1985 17. Haines AP. Sanders TAB, Imeson JD. et al: Effects of a fish oil supplement on platelet function, haemostatic variables and albuminuria in insulin-dependent diabetics. Thromb Res 43:643655,1986 18. Nestel PJ, Connor WE, Reardon MF, et al: Suppression by diets rich in fish oil of very low density lipoprotein production in man. J Clin Invest 74:82-89, 1984 19. Harris WS, Connor WE, Inkeles SB, et al: Dietary omega-3 fatty acids prevent carbohydrate-induced hypertriglyceridemia. Metabolism 33:1016-1019. 1984 20. Patsch JR, Prasad S, Gotto AM, et al: Postprandial lipemia. A key for the conversion of high density lipoprotein 2 into high density lipoprotein 3 by hepatic lipase. J C’lin Invest 74:2017-2023, 1984 21. Desai K, Mistry P, Bagget C, et al: Inhibition of platelet aggregation by abnormal high density lipoprotein particles in plasma from patients with hepatic cirrhosis. Lancet 1:693-695. 1989 22. Thompson PD, Lazarus B, Cullinane E, et al: Exercise. diet or physical characteristics as determinants of HDL-levels in endurance athletes. Atherosclerosis 46:333-339, 1983 23. Masarei JRL, Puddey IB. Rouse IL. et al: Effects of alcohol consumption on serum lipoprotein-lipid and apoprotein concentrations. Atherosclerosis 60:79-87. 1986 24. Nestel PJ: Fish oil attenuates the cholesterol induced rise in lipoprotein cholesterol. Am J Clin Nutr 43:752-757, 1986 25. Grundy SM, Florentin L, Nix D, et al: Comparison of monounsaturated fatty acids and carbohydrates for reducing raised levels of plasma cholesterol in man. Am J Clin Nutr 47:965-969. 1988 26. Herold PM, Kinsella JE: Fish oil consumption and decreased risk of cardiovascular disease: A comparison of findings from animal and human feeding trials. Am J Clin Nutr 43:566-598, 1986 27. Harris WS, Connor WE. McMurry MP: The comparative reductions of the plasma lipids and lipoproteins by dietary polyunsaturated fats: Salmon oilversus vegetable oils. Metabolism 32:179184.1983 28. Illingworth DR, Harris WS, Connor WE: Inhibition of low density lipoprotein synthesis by dietary omega-3 fatty acids in humans. Arteriosclerosis 4:270-275, 1984 29. Simons LA, Hickie JB, Balasubramaniam S: On the effects of dietary n-3 fatty acids (Max EPA) on plasma lipids and lipoproteins in patients with hyperlipidaemia. Atherosclerosis 54:75-88, 1985 30. Sanders TAB, Mistry M: Controlled trials of fish oil supplements on plasma lipid concentrations. Br J Clin Pratt 38:78-81. 1984
246
31. Packard CJ, Munro A, Lorimer AR, et al: Metabolism of apolipoprotein B in large, triglyceride-rich very low density lipoproteins of normal and hypertriglyceridaemic subjects. J Clin Invest 74:2178-2192,1984 32. Sullivan DR, Sanders TAB, Trayner IM, et al: Paradoxical elevation of LDL apoprotein B levels in hypertriglyceridemic patients and normal subjects ingesting fish oil. Atherosclerosis 61:129-134,1986 33. Bagdale JD, Buchanan WE, Levy RA, et al: Effects of
MORIET
AL
omega-3 fish oils on plasma lipids, lipoprotein composition, and postheparin lipoprotein lipase in women with IDDM. Diabetes 39:426-431, 1990 34. Miller NE, Hammett F, Saltissi S, et al: Relation of angiographically defined coronary artery disease to plasma lipoprotein subfractions and apolipoproteins. Br Med J 282:1741-1744,198l 35. Gordon DJ, Probstfield JL, Garrison RJ, et al: High-density lipoprotein cholesterol and cardiovascular disease-Four prospective American studies. Circulation. 79:8-l& 1989
of Diets Supplemented With Fish Oil or Olive Oil on Plasma Lipoproteins in Insulin-Dependent Diabetics
Trevor A. Mori, Robert Vandongen, John R.L. Masarei, Ian L. Rouse, and Diana Dunbar This study was designed to compare changes in high-density lipoprotein (HDL)- and low-density lipoprotein (LDL)-cholesterol in normolipidaemic male insulin-dependent diabetics (IDD) following dietary supplementation with either the fish oil concentrate Max EPA or olive oil. The contribution of the small quantity of cholesterol in Max EPA to these changes was also examined. Twenty-seven subjects were matched in groups of three and randomly allocated to one of three treatment groups of nine subjects each. Subjects were given 15 l-g capsules of oil daily for 3 weeks, consisting of either Max EPA, olive oil, or olive oil to which was added the same amount of cholesterol as contained in Max EPA, respectively. There was a significant increase in eicosapentaenoic acid, and a decrease in arachidonic acid, in the platelet membrane phospholipids of subjects taking Max EPA. In this group, there was an approximately 30% increase in serum HDL,-cholesterol (0.59 2 0.07 to 0.77 ? 0.11 mmol/L, mean + SEM; P < .Ol) and a corresponding decrease in HDl+cholesterol (0.79 * 0.03 to 0.71 ? 0.03 mmol/L; P < .05). Although total and LDL-cholesterol concentrations were also higher after Max EPA, the changes were not significant. Triglycerides were significantly decreased by Max EPA. There were no significant changes in lipids in the groups given olive oil. These results show that compared with olive oil, dietary supplementation with Max EPA substantially increases HDL,cholesterol in insulin-dependent diabetics. This is most likely due to a selective effect of ~3 fatty acids. An increase of this proportion in the cardioprotective HDL,-cholesterol, if sustained over time, could have a major impact on vascular health in diabetics. Copyright 0 1991 by W.B. Saunders Company
D
IETS in insulin-dependent diabetics should aim to improve glucose tolerance and to reduce hyperlipidemia and therefore the risk of vascular complications frequently encountered in this disease. Epidemiological evidence suggests that eating fish, and in particular fish oils rich in w3 fatty acids, may lower the incidence of atherosclerotic vascular disease. This effect has been attributed to an improvement in the plasma lipid profile and in the indices of thrombosis and inflammation.’ Previous studies comparing fish oil with vegetable oil supplements in hyperlipidemic subjects have shown either
no change,‘-4 or an increase in high-density lipoprotein (HDL), and particularly HDL2-cholesterol5 Since vegetable oils containing 06 polyunsaturated fats, such as linoleic actd, have been reported to lower HDL-cholesterol,6 a comparison with the effect of fish oil may not be strictly valid. A more appropriate comparison would be with olive oil, which causes little if any reduction in HDL-cholesterol levels,’ although the effect on HDL subfractions appears not to have been studied in any detail. However, a recently published report8 found that diets enriched with either monounsaturated or polyunsaturated oil lowered HDL equally in healthy subjects, although the changes were not significant. Because of the continuing uncertainty about the effect of monunsaturated fats on HDL-cholesterol, a comparison with the apparently selective effect of fish oil in increasing HDL,-cholesterol was considered important. Diabetics were selected as subjects for this study, because their predisposition to hyperlipidemia and vascular disease makes them likely targets for dietary advice aimed at improving plasma lipoprotein levels. SUBJECTS AND METHODS Twenty-seven male insulin-dependent diabetics, aged 21 to 43 years, who were nonsmokers, were recruited from the Diabetic Clmic at Royal Perth Hospital. The protocol was approved by the Hospital’s Human Rights Committee. Seven patients had retinopathy. whereas no complications were clinically detectable in the remainder. The duration of diabetes ranged from 3 to 34 years. All Metabolism,
Vol40,
No 3 (March),
1991:
pp 241-246
participants were fully informed about the nature of the study and gave written consent. Nonfasting serum cholesterol and triglyceride concentrations in all patients were less than 6.5 and 3.0 mmol/L, respectively, and none were taking any lipid-lowering agents. As the purpose of the study was to examine the effect of fish oil as a dietary supplement, no attempt was made to change or standardize the composition of the individual‘s background diabetic diet. However, subjects were requested not to change their current eating patterns or food intake. Any changes in diet. alcohol intake, physical activity, insulin requirements. and medication taken were assessed by means of a weekly questionnaire. Baseline serum lipids, plasma creatinine. plasma glucose, and platelet phospholipid fatty acid concentrations were measured approximately 2 hours after a light standardized breakfast. Serum lipids were measured on two occasions, 1 week apart, and the results averaged to obtain baseline values. Subjects were deliberately not fasted to minimize interference with diabetic control. They were matched in groups of three. as closely as possible for age and body mass index (BMI) calculated as weight/height,’ and then randomly allocated to one of three treatment groups, each containing nine subjects. Diets were supplemented daily with 15 g of encapsulated oil (1.0-g capsules) for 3 weeks: group A received Max EPA (Reckitt &Colman Pharmaceuticals, Sydney, Australia), group B olive oil (Reckitt & Colman); and group C olive oil enriched with cholesterol to the same concentration as Max EPA (prepared and encapsulated by R.P. Scherer. Melbourne, Australia). This latter group was included to determine a possible contribution of the small amount of cholesterol in Max EPA on plasma lipoprotein concentrations. The study was carried out double-blind with the subjects unaware of the contents of their oil capsules, which were identical in appearance.
From the University Department ofMedicine. Royal Perth Hospital, Perth, Western Australia. Suppotied by grants from the Diabetes Research Foundatton of Western Australia and the National Heart Foundation ofAustralia. Address reprint requests to TrevorA. Mori, PhD, University Department of Medicine, Medical School, Rqyal Perth Hospital, 35 Victoria Square, Perth, Western Australia, 6000. Copyrikht 0 1991 by W.B. Saunders Compuny 0026-049519114003-0005$03.00/0 241
MORI ET AL
242
The fatty acid and cholesterol composition of the oil is shown in Table 1 and the daily amounts received in Table 2. The fatty acid composition and cholesterol content was verified by gas liquid chromatography (GLC) (Hewlett-Packard Model 5980A) and the vitamin A and E content (92 and 526 p&g, respectively) by high-performance liquid chromatography.’ These vitamins were not added to the olive oil preparations. The above measurements were repeated, at approximately the same time of day, at the end of the 3 weeks’ dietary treatment and again 6 weeks after this period.
Table 2. Daily Amounts of Saturated, Monounsaturated,
w6 and ~3
Fatty Acids, and Cholesterol Provided by the Dietary Oils Olive Oil + Max
EPA
Olive Oil
(Group Al
Saturated (g)
Cholesterol
(Group B)
(Group C)
4.2
1.9
1.9
4.3
12.3
12.3
~6 (g)
0.3
0.8
0.8
03 (9)
5.2 0.0
51.0
Monounsaturated
(g)
Cholesterol (mg)
50.0
Plasma Lipid Levels Lipids from serum frozen at -20°C were assayed at the completion of the study. Cholesterol and triglyceride levels were determined enzymatitally on an Abbott ABA-100 bichromatic analyser with reagents from Abbott (Pasadena, CA), and standardized with serum-based calibrators that were obtained from the Australian Lipid Standards Program with values that were traceable to the Centers for Disease Control (Atlanta, GA). The coefficients of variation (CV) were 1.5% and 2.0%, respectively. Serum HDL-cholesterol was assayed on a heparin-manganese chloride supernatant.‘O HDL-cholesterol subfractions HDL, and HDL, were measured by the double precipitation method.” Mean values and SDS were calculated on 42 separate runs with aliquots of serum kept at -70°C: mean total HDL-cholesterol level, 1.61 ‘_ 0.057 (SD) mmol/L (CV, 3.5%); HDL,-cholesterol level. 0.734 + 0.063 mmol/L (CV, 8.6%), and HDb-cholesterol level, 0.874 ? 0.057 mmol/L (CV, 6.5%). Low-density lipoprotein (LDL)cholesterol levels were calculated from the Friedewald formula.” Platelet Phospholipid Fatty Acids Platelets were prepared as described previously’ and total phospholipids separated from the crude lipid extracts by thin-layer chromatography. Fatty acid methyl esters were prepared by treatment of samples with 4% H,SO, in methanol at 90°C for 20 minutes, analyzed by gas chromatography, and subsequently identified by comparison with authentic standards. The amounts of individual fatty acids were calculated as relative percentages with the evaluated fatty acids set at 100%. Statistical Evaluation Data were analyzed on SAS using repeated measures ANOVA. When a significant treatment-time interaction was demonstrated, Table 1. Percentage Composition of the Major Fatty Acids and Cholesterol in the Dietary Oils Olive Oil + Max EPA
Olive Oil
Cholesterol
Fatty acid 14:o
6.90
16:0
15.78
10.02
10.04
16:l (07)
9.30
0.70
0.70
190
2.95
2.50
2.57
l&l
(09)
13.84
81.49
81.19
18:2 (06)
0.99
5.29
5.50
18:3 (03)
0.48
20:4 (06)
1.04
205 (w3)
18.96
22:5 (~3)
2.71
22:6 (~3)
12.23
Cholesterol
0.33
Duncan’s modified t test was used to compare between-group differences at each visit. RESULTS
From Table 3, it is seen that the groups were comparable
for age, BMI, plasma glucose, insulin dose, and renal function. All subjects complied with the protocol and there were no major changes in diet, alcohol intake, physical activity, or insulin requirements. There were no significant (P > .05) changes in weight in any of the groups between baseline and the end of the 3 weeks’ treatment (A, 70.0 ? 2.5 to 71.2 ? 2.6; B, 80.3 ? 2.6 to 79.5 f 2.7; C, 74.3 2 3.0 to 76.7 ? 2.6 kg). Apart from occasional reports of a feeling of fullness and mild epigastric discomfort, no other side effects were encountered. One subject in group B withdrew during the supplementation period for reasons unrelated to the study. Compliance with the protocol was 99% in group A, 98% in group B, and 99% in group C as ascertained by capsule count. This was confirmed in group A, taking Max EPA, by the incorporation of EPA (20:5) into platelet phospholipids (Table 4) and a corresponding reduction in arachidonic acid (20:4). Max EPA supplementation also resulted in significant decreases in platelet o-6 fatty acids 20:3 and 22:4, and significant increases in the 03 fatty acids 2215 and 22:6. The platelet fatty acid composition in groups B and C was not altered by the treatment, with the exception of a small and insignificant increase in oleic acid (18:l) (14.02 2 0.53 to 14.94 + 0.24 and 13.29 2 0.18 to 14.17 ? 0.18, respectively). Samples for fatty acid analysis were not obtained from one subject each in group A and B. Serum Cholesterol and Triglycerides
Samples were not collected at all visits from one subject in group B who was deleted from the analysis. HDL,- and HDL,-cholesterol were not isolated and the Friedewald formula could not be applied for the calculation of LDL-
Table 3. Clinical and Biochemical Characteristics of Subjects in Groups A (Max EPA), B (Olive Oil), and C (Olive Oil and Cholesterol)
0.34
Plasma Creatinine
(mmol/L)
(mmol/L)
Age
EIMI
A
34.4 f 1.9
24.1 2 0.4
12.0 +- 2.2
51 + 5
87.7 k 3.7
6
33.8 + 2.1
24.7 + 0.8
12.2 2 2.7
46?
4
86.4 lr 5.4
C
34.2 k 1.9
24.5 2 0.8
10.7 + 1.9
50?
6
87.2 + 4.2
Group
-
Plasma Glucose
NOTE. Values are means + SEM.
FISH OIL RAISES
HIGH-DENSITY
LIPOPROTEIN
CHOLESTEROL
243
GROUP
Table 4. Percentage Composition of Fatty Acids in Platelet Phospholipids in Group A (Max EPA) (n = 6) at Baseline (week 0).
C
6
A
After 3 Weeks of Treatment With Max EPA (week 3), and 6 Weeks After the Treatment
-
BaselIne htty 16:0
Acid
DMA
16:0 18:0 DMA
1.08 + 0.21
+ 0.35
12.45
2.38 + 0.44
r
(Week 9)
12.43
2.62 k 0.36
1.35 2 0.19
-t 0.40
17.88
+ 0.40
18.09
k 0.22
18:l
13.92
2 0.33
14.75
+ 0.34
13.56
+ 0.32
20:3
5.52 + 0.33
1.62 2 0.12
20:4
26.48
5.79 t 0.28
1.18 + 0.08*
-t 0.55
22.32
1.67 + 0.12
t 0.63*
26.34
20:5
4.70 +- 0.30
7.56 k 0.39*
5.33 + 0.30
2.97 + 0.20
1.64 -+ 0.18*
3.08 2 0.21
i2:5
5.09 + 0.39
6.32 k 0.29t
5.71 t 0.25
22:6
4.72
6.67 2 0.26*
5.94 k 0.29*
NOTE.
Values
Abbreviation: For significance sures ANOVA):
are means DMA,
5g
l.O-
$
0.8-
p % d
0.6-
J IN
0.4-
k 0.72
22~4
+ 0.39
p
2 0.47
18.47
6.08 -t 0.36
1.2 -
0.69 -t 0.12
2 0.66
18:0 18:2
-
(Week 3)
0.82 -t 0.26
1.4
Posttreatment
Treatment
(Week 0)
12.77
Period (week 9)
$
0.2-
t SEM.
”
dimethylacetal.
of difference
0
from
baseline
values
(repeated
3
9
0
mea-
3
9
0
3
‘P -< .Ol; tP < .05.
Fig 1. Individual serum HDL,-cholesterol in patients receiving Max EPA (group A) (n = 9) at baseline (week 0). after 3 weeks of treatment (week 3), and after the treatment period (week 9).
cholesterol in another subject because triglyceride levels exceeded 4.0 mmol/L. As shown in Table 5, there was a tendency for serum total and LDL-cholesterol concentrations to increase following Max EPA treatment (group A), similar to that previously reported when the changes reached statistical significance.’ However, triglyceride levels were significantly lowered by Max EPA (representing a mean decrease of 35%), there being no changes in the other groups (Table 5). Although HDL-cholesterol as a whole was not altered by the dietary oil supplements in any of the groups (Table S), there was a substantial, and approximately 30%, increase in the HDL, subfraction (P < .Ol) in the Max EPA group compared with insignificant increases of 7% and 2% in
groups B and C, respectively. The individual responses in the three groups are illustrated in Fig 1. The changes in triglycerides and HDLZ-C in group A were not significantly correlated. HDL,-cholesterol decreased, but not significantly, in group A, representing a mean decrease of 10%. There were no changes in groups B or C (Table 5). DISCUSSION
These results demonstrate a unique and selective effect of Max EPA fish oil in raising the level of HDLL-cholesterol
Table 5. Serum Lipid Levels (mmol/L) at Baseline (week 0). After 3 Weeks of Treatment
(week 3). and 6 Weeks After the Treatment
Period
(week 9) in the Three Groups: Group A (Max EPA), B (Olive Oil), and C (Olive 011 and Cholesterol) Week Group
0
3
9
0
3
Cholesterol
9
Triglycerides
A
4.93
2 0.19
5.20 k 0.30
4.98 2 0.23
1.39 + 0.12
0.19 i- 0.11*
1.36 -t 0.17
B
4.86 2 0.28
5.04 * 0.34
5.08 k 0.29
1.23 t 0.18
1.14 + 0.16
1.37 + 0.19
C
5.08 5 0.25
5.23 k 0.38
4.98 5 0.39
1.00 + 0.08
1.18 t 0.16
1.22 f 0.18
A
2.90
? 0.18
3.30 k 0.30
2.96 2 0.25
1.39 k 0.08
1.48 k 0.11
1.39 + 0.10
6
2.61
-t 0.16
2.83 2 0.25
2.85 _f 0.20
1.41 k 0.06
1.46 t 0.08
1.38 + 0.08
C
3.19
2 0.25
3.28 + 0.37
3.11 + 0.39
1.43 -t 0.05
1.41 + 0.06
1.30 + 0.08
A
0.59 k 0.07
0.77 t 0.11*
0.60 k 0.09
0.79
0.71 + 0.03t
0.70 % 0.04
B
0.59 k 0.03
0.63 5 0.05
0.61 + 0.04
0.85 r 0.04
0.84 + 0.05
0.81 ? 0.06
C
0.59 ? 0.04
0.58 2 0.06
0.51 + 0.05
0.84 + 0.03
0.82 2 0.03
0.79 + 0.04
LDL-Cholesterol
HDL-Cholesterol
HDLJholesterol
NOTE.
Values
For significance modified (P r .Ol).
t test):
are means
+ SEM. Group
of difference group
A differs
from
group
HDL,-Cholesterol + 0.03
A, n = 9; B, n = 8; C, n = 9.
baseline
values
(repeated
B, tP < .05. A significant
9
WEEKS
measures
ANOVA):
treatment-time
lP -c .Ol. Between-group interaction
was
present
differences for triglycerides
at each visit (Duncan’s and HDL,-cholesterol
244
in insulin-dependent diabetics. Furthermore, the findings support our previous observation in diabetic patients of a smaller increase in serum total and LDL-cholestero1,‘3.‘4 but the changes on this occasion failed to reach statistical significance. The decrease in plasma triglycerides is the most consistent effect of dietary fish oils in both normall and hyperlipidemic16 subjects, although it was not seen in another study in insulin-dependent diabetics following an almost identical Max EPA regimen.” However, as triglyceride concentrations were measured in nonfasting blood samples, our findings may not be strictly comparable to those of others. Impairment of very-low-density lipoprotein (VLDL) synthesis,” or increased remova1,‘9 through activation of lipoprotein lipase, are possible mechanisms for the decrease in triglycerides and increase in HDL-cholesterol induced by fish oils. Activation of activity of lipoprotein lipase, by increasing the breakdown of triglyceride-rich lipoproteins and lowering serum triglyceride concentration, could facilitate the transfer of surface lipids from these particles to HDL,” and the conversion of HDL, to HDL,. It was recently suggested that in situations where HDbcholesterol concentrations were elevated, the lipoprotein particles may be preferentially enriched with apolipoprotein E,” which appears to lower platelet sensitivity to aggregation. This could be an intriguing explanation for the reduction in platelet aggregation observed in several fish oil studies.’ Body weight remained constant and the changes in plasma lipids were not accompanied by major alterations in diet or in other factors influencing HDL-cholesterol, in particular exercise” and alcohoLZ3 which may independently increase the level of this lipoprotein. The choice of placebo or control preparation in dietary studies of this nature is limited by the availability of oils having no demonstrable effects on lipid metabolism. Nevertheless, the design of the study, comparing the changes in lipids following Max EPA with olive oil, and olive oil enriched with cholesterol, clearly identifies the w3 fatty acids as the components most likely responsible for the 30% increase in HDb-cholesterol. The addition of a small quantity of cholesterol to olive oil, which is otherwise free of cholesterol, had no further effect on plasma lipids. Therefore, it would be reasonable to conclude that the small amount of cholesterol present in Max EPA plays no part in increasing HDL,-cholesterol, or altering the concentration of other plasma lipids, as shown by the uniform increase in LDL-cholesterol in the three groups. In fact, w3 fatty acids have been shown to attenuate the hypercholesterolemic effect of dietary cholestero1.24 Similarly, linoleic acid and oleic acid are unlikely to be involved, except for the remote possibility of these fatty acids having a biphasic action with the low amounts present in Max EPA raising HDL,-cholesterol and the higher quantities in olive oil having no effect. It is of interest that no significant changes in either total or LDL-cholesterol were found in the olive oil-treated groups. This is in contrast to the reduction in cholesterol observed in mildly hypercholesterolemic patients given a
MORI ET AL
high monounsaturated fat diet, principally in the form of oleic acid.25These different responses may be related to the amount of oleic acid given in the latter study (28% of total daily calories), the pre-diet lipid levels, and the absence of diabetes. In addition to effects on plasma lipoproteins, the fatty acids also differed in their degree of incorporation into platelet membrane phospholipids. EPA was most avidly incorporated, displacing arachidonic acid and remaining higher than baseline levels even 6 weeks after Max EPA supplementation was ceased. Docosapentaenoic acid (225) and docosahexaenoic acid (22:6), both present in Max EPA as carbon chain elongation and desaturation products of EPA, were also significantly increased after 3 weeks of Max EPA. The latter (22:6) remained significantly higher than baseline 6 weeks after ceasing Max EPA. Similarly, the w6 fatty acids 20:3 and 22:4, which are the immediate precursor and metabolite, respectively, of arachidonic acid, were significantly decreased during Max EPA administration. There was only a small increase in the percentage of phospholipid oleic acid in the olive oil groups, indicating a high degree of selectivity of fatty acid incorporation into platelet membranes. The change in membrane composition resulting from a more selective uptake of 03 fatty acids may have important functional implications in relation to platelet aggregation and the release of secretory products.26 The reported effect of fish oil on plasma total cholesterol has been inconsistent, due largely to differences in the amount given and in the pretreatment cholesterol concentration. Large quantities of fish oil, equivalent to at least 10 g of EPA daily over several weeks, generally results in a decrease in cholesterol in normal subjects,27.Win patients with hyperlipidemia of the type IIb phenotype,16 and, in particular, in patients with gross hypertriglyceridemia as in the type V category.‘6.Z9 Lower amounts of fish oil, comparable to those used in this study, have not usually altered lipids in normolipidemic subjects,26 although a slight increase in HDL-cholesterol, due to the HDL, subfraction, has been reported.30 In insulin-dependent diabetics, significant increases in total and LDL-cholesterol were observed in an earlier study from our laboratory.13 Similar trends were presented in this report, but the changes were not significant, perhaps due to inadequate sample size and statistical power. Increases in LDL-cholesterol have also occurred in patients with hypertriglyceridemia’,‘6.29 and have been attributed to increased production of LDL particles, rather than higher cholesterol content per particle, from smaller VLDL particles, which are formed in preference to larger particles during fish oil feeding.z,3’,3’ It is uncertain if the effect of fish oil in raising HDL,cholesterol is confined to patients with diabetes, as suggested by our earlier study,14 or if it also occurs in other populations. A similar increase in HDL,-cholesterol was recently reported in a small uncontrolled trial where insulin-dependent women were given 6 g of o-3 fatty acids daily for 10 weeks.33 Of interest was the additional finding
FISH OIL RAISES HIGH-DENSITY
LIPOPROTEIN
245
CHOLESTEROL
that postheparin plasma lipolytic activity did not change, suggesting that enhanced VLDL removal does not explain the increase in HDLZ-cholesterol. In epidemiological terms, the 30% increase in HDL,-cholesterol (equivalent to an average of 0.15 mmol/L) is substantial and could be expected to have a marked impact on the incidence of cardiovascular disease if indeed this is inversely related to the plasma level of this subfraction3“ To date, the evidence I’OI the apparent benefit associated with higher HDLcholesterol, with a 0.026 mmol/L increment predicted to reduce coronary risk by 2% to 3%,35 is confined to HDL as a
whole, rather than to its subfractions. The potential benefit of the increase in HDL,-cholesterol may need to be considered carefully against the small increase in LDLcholesterol. However, it may be concluded that the plasma lipoprotein profile in insulin-dependent diabetics was more favorably affected by a fish oil-enriched than by an olive oil-enriched diet. ACKNOWLEDGMENT
We gratefully acknowledge the technical assistance of Lynette Kelly.
REFERENCES Vandongen R: Fish oils and cardiovascular disease. Med J Aust 146:236-237, 1987 2. Harris WS, Dujovne CA, Zucker M, et al: Effects of a low saturated fat, low cholesterol fish oil supplement in hypertriglyceridaemic patients. Ann Intern Med 109:465-470,1988 .T Harris WS, Zucker ML, Dujovne CA: Omega-3 fatty acids in hypertriglyceridemic patients: Triglycerides vs methyl esters. Am J Clin Nutr 48:992-Y97, 1988 L. Sanders TAB, Sullivan DR. Reeve J, et al: Triglyceridelowering effect of marine polyunsaturates in patients with hypertriglyceridemia. Arteriosclerosis 5:459-465, 1985 5. Demke DM, Peters GR, Linet 01, et al: Effects of a fish oil concentrate in patients with hypercholesterolemia. Atherosclerosis ‘0:73-SO,1988 (1. Mattson FH, Grundy SM: Comparison of effects of dietary saturated, monounsaturated and polyunsaturated fatty acids on plasma lipids and lipoproteins in man. J Lipid Res 26:194-202,1985 -. Grundy SM: Comparison of monounsaturated fatty acids and carbohydrates for lowering plasma cholesterol. N Engl J Med 314:745-748. 1986 8. Mensink RP, Katan MB: Effect of a diet enriched with monounsaturated or polyunsaturated fatty acids on levels of low-density and high-density lipoprotein cholesterol in healthy women and men. N Engl J Med 321:436-441, 1989 (J. Mori TA, Codde JP. Vandongen R, et al: New findings in the t‘at:y acid composition of individual platelet phospholipids in man after dietary fish oil supplementation. Lipids 22:744-750, 1987 0. Warnick GR, Albers JJ: A comprehensive evaluation of the heparin-manganese precipitation procedure for estimating high density lipoprotein cholesterol. J Lipid Res 19:65-76, 1978 1. Gidez LI, Miller GJ, Burstein M, et al: Separation and quantitation of subclasses of human plasma high density lipoproteins by single precipitation procedure. J Lipid Res 23:1206-1223. 1982 2. Friedewald WT. Levy RI, Fredrickson DS: Estimation of the concentration of low-density lipoprotein cholesterol in plasma without use of the preparative ultracentrifuge. Clin Chem 18:49950’. 1972 i 3. Vandongen R, Mori TA, Codde JP, et al: Hypercholesterolaemic effect of fish oil in insulin-dependent diabetic patients. Med J Aust 148:141-143,198s : 4. Mori TA. Vandongen R, Masarei JRL, et al: Dietary fish oils increase serum lipids in insulin-dependent diabetics compared with healthy controls. Metabolism 38:404-409, 1989 15. Sanders TAB, Roshanai F: The influence of different types of omega-3 fatty acids on blood lipids and platelet function in healthy volunteers. Clin Sci 64:91-99, 1983 26. Phillipson BE, Rothrock DW, Connor WE, et al: Reduction of plasma lipids. lipoproteins and apoproteins by dietary fish oil in
patients with hypertriglyceridaemia. N Engl J Med 312:1210-1216, 1985 17. Haines AP. Sanders TAB, Imeson JD. et al: Effects of a fish oil supplement on platelet function, haemostatic variables and albuminuria in insulin-dependent diabetics. Thromb Res 43:643655,1986 18. Nestel PJ, Connor WE, Reardon MF, et al: Suppression by diets rich in fish oil of very low density lipoprotein production in man. J Clin Invest 74:82-89, 1984 19. Harris WS, Connor WE, Inkeles SB, et al: Dietary omega-3 fatty acids prevent carbohydrate-induced hypertriglyceridemia. Metabolism 33:1016-1019. 1984 20. Patsch JR, Prasad S, Gotto AM, et al: Postprandial lipemia. A key for the conversion of high density lipoprotein 2 into high density lipoprotein 3 by hepatic lipase. J C’lin Invest 74:2017-2023, 1984 21. Desai K, Mistry P, Bagget C, et al: Inhibition of platelet aggregation by abnormal high density lipoprotein particles in plasma from patients with hepatic cirrhosis. Lancet 1:693-695. 1989 22. Thompson PD, Lazarus B, Cullinane E, et al: Exercise. diet or physical characteristics as determinants of HDL-levels in endurance athletes. Atherosclerosis 46:333-339, 1983 23. Masarei JRL, Puddey IB. Rouse IL. et al: Effects of alcohol consumption on serum lipoprotein-lipid and apoprotein concentrations. Atherosclerosis 60:79-87. 1986 24. Nestel PJ: Fish oil attenuates the cholesterol induced rise in lipoprotein cholesterol. Am J Clin Nutr 43:752-757, 1986 25. Grundy SM, Florentin L, Nix D, et al: Comparison of monounsaturated fatty acids and carbohydrates for reducing raised levels of plasma cholesterol in man. Am J Clin Nutr 47:965-969. 1988 26. Herold PM, Kinsella JE: Fish oil consumption and decreased risk of cardiovascular disease: A comparison of findings from animal and human feeding trials. Am J Clin Nutr 43:566-598, 1986 27. Harris WS, Connor WE. McMurry MP: The comparative reductions of the plasma lipids and lipoproteins by dietary polyunsaturated fats: Salmon oilversus vegetable oils. Metabolism 32:179184.1983 28. Illingworth DR, Harris WS, Connor WE: Inhibition of low density lipoprotein synthesis by dietary omega-3 fatty acids in humans. Arteriosclerosis 4:270-275, 1984 29. Simons LA, Hickie JB, Balasubramaniam S: On the effects of dietary n-3 fatty acids (Max EPA) on plasma lipids and lipoproteins in patients with hyperlipidaemia. Atherosclerosis 54:75-88, 1985 30. Sanders TAB, Mistry M: Controlled trials of fish oil supplements on plasma lipid concentrations. Br J Clin Pratt 38:78-81. 1984
246
31. Packard CJ, Munro A, Lorimer AR, et al: Metabolism of apolipoprotein B in large, triglyceride-rich very low density lipoproteins of normal and hypertriglyceridaemic subjects. J Clin Invest 74:2178-2192,1984 32. Sullivan DR, Sanders TAB, Trayner IM, et al: Paradoxical elevation of LDL apoprotein B levels in hypertriglyceridemic patients and normal subjects ingesting fish oil. Atherosclerosis 61:129-134,1986 33. Bagdale JD, Buchanan WE, Levy RA, et al: Effects of
MORIET
AL
omega-3 fish oils on plasma lipids, lipoprotein composition, and postheparin lipoprotein lipase in women with IDDM. Diabetes 39:426-431, 1990 34. Miller NE, Hammett F, Saltissi S, et al: Relation of angiographically defined coronary artery disease to plasma lipoprotein subfractions and apolipoproteins. Br Med J 282:1741-1744,198l 35. Gordon DJ, Probstfield JL, Garrison RJ, et al: High-density lipoprotein cholesterol and cardiovascular disease-Four prospective American studies. Circulation. 79:8-l& 1989