点击显示 收起
1 From the Hugh Sinclair Unit of Human Nutrition, School of Food Biosciences, The University of Reading, Reading, United Kingdom.
2 Supported by the Food Standards Agency, United Kingdom. 3 Reprints not available. Address correspondence to JA Lovegrove, School of Food Biosciences, Hugh Sinclair Unit of Human Nutrition, The University of Reading, PO Box 226, Whiteknights, Reading RG6 6AP, United Kingdom. E-mail: j.a.lovegrove{at}reading.ac.uk.
ABSTRACT
Background: The mechanisms involved in the increased mortality from coronary artery disease in British Indo-Asians are not well understood.
Objectives: This study aimed to investigate whether British Indo-Asian Sikhs have higher plasma triacylglycerol concentrations, lower platelet phospholipid levels, and lower dietary intakes of long-chain n–3 polyunsaturated fatty acids (PUFAs) than do age- and weight-matched Europeans and whether moderate dietary fish-oil intake can reverse these differences.
Design: A randomized, double-blind, placebo-controlled, parallel, fish-oil intervention study was performed. After a 2-wk run-in period, 44 Europeans and 40 Indo-Asian Sikhs were randomly assigned to receive either 4.0 g fish oil [1.5 g eicosapentaenoic acid (EPA) and 1.0 g docosahexaenoic acid (DHA)] or 4.0 g olive oil (control) daily for 12 wk.
Results: At baseline, the Indo-Asians had significantly higher plasma triacylglycerol, small dense LDL, apolipoprotein B, and dietary and platelet phospholipid n–6 PUFA values and significantly lower long-chain n–3 PUFAs (EPA and DHA) than did the Europeans. A significant decrease in plasma triacylglycerol, plasma apolipoprotein B-48, and platelet phospholipid arachidonic acid concentrations and a significant increase in plasma HDL concentrations and platelet phospholipid EPA and DHA levels were observed after fish-oil supplementation. No significant effect of ethnicity on the responses to fish-oil supplementation was observed.
Conclusions: Moderate fish-oil supplementation contributes to a reversal of lipid abnormalities and low n–3 PUFA levels in Indo-Asians and should be considered as an important, yet simple, dietary manipulation to reduce CAD risk in Indo-Asians with an atherogenic lipoprotein phenotype.
Key Words: Long-chain n–3 polyunsaturated fatty acids • LC n–3 PUFAs • eicosapentaenoic acid • EPA • docosahexaenoic acid • DHA • Indo-Asians • Sikhs • triacylglycerols • apolipoprotein B-48 • fish oil • nutrient intake
INTRODUCTION
In the United Kingdom, mortality from coronary artery disease (CAD) is 55% and 41% higher in Indo-Asian men and women, respectively, than in matched whites (1). The mechanisms underlying increased CAD mortality in British Indo-Asians are not well understood. Conventional risk factors [smoking, high blood pressure, and high plasma total cholesterol (TC)] do not account for the high CAD mortality in this population (2). The prevalence of diabetes in British Indo-Asians is almost 4-fold that of European whites (3, 4), and insulin resistance and its associated dyslipidemia (elevated plasma triacylglycerol and small dense LDL3 particles and low HDL-cholesterol concentrations, commonly known as the atherogenic lipoprotein phenotype) and metabolic abnormalities (central adiposity, glucose intolerance, elevated plasminogen activator inhibitor-1, and hyperinsulinemia) are more common in Indo-Asian groups (2–6).
The predisposition of the Indo-Asian groups to insulin resistance, associated lipid abnormalities and CAD may, in part, be determined by genetic factors, but diet may also play a significant role. Dietary data for Indo-Asian groups living in the UK is limited (6–11) with ethnic diversity in the UK Indo-Asian population associated with different dietary habits (12). Notable findings from the national food survey in 1985 (7) were the lower total fat intake, higher total polyunsaturated fatty acid (PUFA) intake, and higher ratio of PUFAs to saturated fatty acids in Indo-Asians compared with Europeans. However few studies have compared dietary intakes of long chain n–3 polyunsaturated fatty acids (LC n–3 PUFAs) in Indo-Asians and Europeans (8–10). Despite the limited dietary data on Indo-Asian groups, it is frequently proposed that dietary LC n–3 PUFA intakes and whole-body LC n–3 PUFA levels are lower in Indo-Asians living in Britain than in the indigenous European population (11–14). Although data support lower circulating and tissue levels of LC n–3 PUFAs in some Indo-Asian groups (7, 9, 10, 13–15), the situation with respect to their dietary LC n–3 PUFA intake is much less clear. For this reason it is uncertain whether the lower LC n–3 PUFA level in Indo-Asians is due to a lack of these fatty acids in the diet, to a dietary excess of n–6 PUFAs, or to a metabolic defect in the incorporation of LC n–3 PUFAs.
Numerous studies have reported the responses of European subjects to dietary supplementation with EPA and DHA (16–19), but there is limited data for Indo-Asians (20) or migrant Indo-Asians (15). Although reductions in fasting and postprandial plasma triacylglycerol concentrations are the most widely reported biological effect of LC n–3 PUFAs (19, 21–23), the response of other components of the atherogenic lipoprotein phenotype to LC n–3 PUFA supplementation remains to be clarified, especially in Indo-Asians groups.
The aim of this study was to determine whether Indo-Asian Sikhs living in Britain have lower platelet phospholipid contents and lower dietary intakes of LC n–3 PUFAs than do age- and weight-matched Europeans. A further aim was to investigate the metabolic responsiveness to fish oil in this group compared with that in Europeans.
SUBJECTS AND METHODS
Subjects
Eighty-four volunteers successfully completed the study. Forty-four Europeans (24 men, 20 women) and 40 Indo-Asian Sikhs (23 men, 17 women) were recruited from Reading and Slough, respectively, through a combination of methods, including posters (in Punjabi and English) displayed in shops, in Sikh temples, and around the University of Reading campus; leaflets (Punjabi and English) in the local newspapers and posted door to door; advertisements in Punjabi and local newspapers; letters sent to possible subjects who were identified from an electoral register; interviews on radio and television; and word of mouth. The subjects were recruited on the basis of a medical and lifestyle questionnaire followed by a screening blood sample.
For inclusion in the study, volunteers were required to have been resident in the United Kingdom for =" BORDER="0">2 y and to consume at least one traditional Asian meal per day. Exclusion criteria included smoking, hypertension, a body mass index (BMI; in kg/m2) <20 or >37, an age <25 or >70 y, plasma TC >8 mmol/L, plasma triacylglycerol >4 mmol/L, plasma glucose >8 mmol/L, glutamyl transferase >80 IU/L, hemoglobin <12.5 g/dL, and a diagnosis of cardiovascular disease, diabetes, or liver disease. In addition, persons were excluded if they participated in more than two 20-min sessions of aerobic exercise weekly, consumed fatty acid supplements, were following a weight-reducing diet or any other special diet, were receiving hypolipidemic therapy, or were taking any medications known to affect lipid metabolism. The age, anthropometric characteristics, and dietary macronutrient intakes of the study volunteers are shown in Table 1. The study received ethical approval from the University of Reading Ethics Committee, East Berkshire Research Ethics Committee, West Berkshire Local Research Ethics Committee, and the Hounslow District Research Ethics Committee. All subjects gave informed consent before participating in the study.
View this table:
TABLE 1. Anthropometric characteristics and dietary intakes of macronutrients and fatty acids estimated from 5-d diet diaries for the European and Indo-Asian subjects1
Power calculation
A total number of 80 subjects (40 Europeans, 40 Indo-Asian Sikhs) were calculated on the basis of a physiologic significant enrichment of platelet phospholipids with EPA of 1.5 mol%, with a power of 90%, and a significance of P < 0.01. These numbers allowed for a 10% dropout rate due to illness or other changes in circumstances.
Study design
This study was a double-blind, randomized, placebo-controlled, parallel dietary intervention. The volunteers were assigned to receive either fish oil or olive oil according to a stratified randomization procedure with the following strata: sex, BMI (18–24, 25–27, and >28), age (30–49 and 50–70 y), and triacylglycerol concentrations (0–1.4, 1.5–2.4, and >2.5 mmol/L). The fish-oil group was given four 1-g capsules of fish oil (Pikasol; Pronova Biocare, Aaslund, Norway) containing 60% LC n–3 PUFAs (each capsule provided 367 mg EPA and 225 mg DHA), and the olive oil group was given four 1-g capsules of olive oil (placebo; Pronova) daily. This dosage of fish oil corresponds to a portion (160 g) of mackerel eaten daily. The fish-oil and olive oil capsules provided 3.5 mg total tocopherol/capsule (1.9 mg -tocopherol/capsule). Olive oil was used as a placebo, because low intakes of olive oil (4 g/d) are considered to have little biochemical action on the variables evaluated in this study (19).
Study protocol
Height, weight, blood pressure, and waist circumference were measured on the first visit. All subjects were given olive oil (control) supplements and instructed to take 4 capsules/d (1 g olive oil/capsule), 2 capsules with breakfast, and 2 capsules with dinner for 2 wk (run-in period). The volunteers recorded all food and drink consumed for 5 d (3 weekdays and 2 weekend days) during the run-in period.
After the run-in period, the subjects and investigators were blinded and the participants were randomly assigned to consume either 4 g fish oil or 4 g olive oil (control) for 12 wk. Fasting blood samples, blood pressure, and weight were measured at 0 (baseline), 6, 12 (end of supplementation period), and 16 (4 wk after supplementation) wk. The subjects were asked to refrain from alcohol intake and strenuous exercise on the day before each clinical visit.
For all of the fasting blood samples collected, TC, HDL cholesterol, triacylglycerol, nonesterified fatty acids, glucose, and insulin were measured, and LDL-cholesterol concentrations were calculated by using the Friedewald formula (24). The following variables were measured only in the blood samples collected before (0 wk) and after (12 wk) supplementation: platelet phospholipid fatty acids (as a marker of tissue LC n–3 PUFA status), total apolipoprotein (apo) B, apo B-48, glycated hemoglobin (as a marker of long-term glycemic control), and -tocopherol (as a marker of antioxidant status). The LDL-density profile was assessed only at baseline.
Measurement of compliance
The compliance of the volunteers with the study protocol was monitored on the basis of self-reports of capsules consumed each day and on the number of remaining capsules returned at the end of the study. The LC n–3 PUFA content of the platelet phospholipids was measured before and after the intervention period.
Anthropometric measurements
A single trained investigator performed all of the anthropometric measurements. The subjects were weighed to the nearest 0.5 kg, and the subjects’ heights were measured to the nearest 1 cm with an upright stadiometer. Waist circumferences were measured 3 times midway between the lowest rib margin and the iliac crest to the nearest 1 mm in a standing position, and the mean was calculated. Blood pressure was measured on 3 occasions, separated by 2 min, and the mean was calculated.
Measurement of biochemical variables
Blood samples were collected into potassium EDTA-containing evacuated tubes and were centrifuged immediately at 1700 x g (3000 rpm) for 15 min at room temperature in a bench-top centrifuge. Plasma was portioned into flat-bottom, 3-mL plastic tubes (Thermo Lifesciences, Basingstoke, United Kingdom) and stored at –20 °C (or –80 °C for -tocopherol concentrations) until analyzed. Plasma TC, HDL cholesterol, triacylglycerol, nonesterified fatty acids, glucose, C-reactive protein (with the use of a high sensitivity assay), and apo B were analyzed with the use of an IL Monarch centrifugal analyser (Instrumentation Laboratory, Warrington, United Kingdom) and enzymatic colorometric kits (Instrumentation Laboratory). HDL-cholesterol concentrations were measured after precipitation of the fresh plasma with dextran-magnesium chloride reagent (25). Plasma insulin was measured with the use of a specific enzyme-linked immunosorbent assay (Dako, Cambridge, United Kingdom). The mean intra- and inter-assay CVs for TC, triacylglycerol, glucose, C-reactive protein, nonesterified fatty acids, apo B, and insulin were 2.1%, 1.9%, 2.0%, 2.5%, 2.4%, 2.9%, and 4.0% and 4.0%, 3.1%, 4.7%, 4.1%, 4.0%, 5.0%, and 5.5%, respectively.
Plasma apo B-48 was determined by using a specific competitive enzyme-linked immunosorbent assay (26) with a few modifications. Briefly, a heptapeptide-thyroglobulin conjugate consisting of the terminal residues of the apo B-48 molecule was used as the coating material in the enzyme-linked immunosorbent assay format. Samples were diluted 1:5 with phosphate-buffered saline and incubated with a specific polyclonal anti-apo B-48 antibody that recognized the carboxyl terminus of the protein on the surface of the lipoprotein particles and did not cross-react with apo B-100 (27). A standard curve was prepared by serial dilution of the apo B-48 heptapeptide in phosphate-buffered saline containing 10 g human serum albumin/L. The intraassay CV was 5.0% at 640 ng/mL. Glycated hemoglobin was measured in the Clinical Biochemistry Laboratory (Royal Berkshire Hospital) by HPLC, and -tocopherol was also measured with HPLC (28). Platelet phospholipid fatty acids were measured after platelets were isolated from whole blood according to the method of Indu and Ghafoorunissa (29). After the addition of butylated hydroxytoluene to all solvents to minimize auto-oxidation, the lipids were extracted from the isolated platelets with chloroform:methanol (2:1, by vol) followed by chloroform according to the method of Folch et al (30). The phospholipid fraction was isolated from the crude lipid by using a Sep-Pac C18 column (Waters Associates, Milford, MA) (31). The phospholipids were transmethylated by using sulfuric acid (15 mL/L methanol), and the fatty acids were quantified by gas chromatography with a CPSil 88 column (Chrompak; Walton-on-Thames, Surrey, United Kingdom) (28). The percentage of LDL3 was determined in fresh, unfrozen plasma. The LDL subclasses were separated by density-gradient ultracentrifugation and the relative percentage of small dense LDL3 was calculated by integrating the respective area under the LDL subclass profile (32).
Dietary analysis
The 5-d diet diaries were recorded on 3 weekdays and 2 weekend days during the 2-wk run-in period. Nutrient intakes were determined by using the nutritional database FOODBASE (version 2.0; Institute of Brain Chemistry and Human Nutrition, London). To improve the accuracy of the dietary data, the FOODBASE data were supplemented with >30 common Punjabi recipes found in books and collected from volunteers. For the purpose of the latter, 10 women were provided with scales and a booklet for recording the recipes of 12 common Punjabi dishes. The weight of the ingredients and the cooked weight were recorded to calculate cooking losses. In an attempt to determine the accuracy of the dietary data, the extent of underreporting and undereating was estimated. Underreporting was assessed by comparing reported energy intakes with total energy expenditure. The subjects were classified as underreporters when the reported energy intake was less than 1.2 x basal metabolic rate, which was determined by using the Bodystat 1500 (Bodystat Ltd, Isle of Man, United Kingdom). To estimate the possibility of undereating during the dietary assessment period, the subjects’ weights before and after the dietary assessment period were recorded and activity levels were estimated. The differences were assessed statistically.
Calculation of results
Insulin resistance in relation to glucose metabolism was assessed by surrogate measures, including measurements of fasting insulin concentrations and the use of the homeostasis model of insulin resistance (HOMA-IR; 33), which was calculated according to the following equation with the use of the mean (of 2 samples) fasting insulin and glucose concentrations:
RESULTS
Subjects
Two hundred eight subjects responded to the study advertisements (81 Europeans, 127 Indo-Asian Sikhs), of whom102 (45 European, 57 Indo-Asian Sikhs) started the study and 84 completed (44 Europeans, 40 Indo-Asian Sikhs) the study—dropout rates of 2% Europeans and 30% Indo-Asians. At baseline, the 2 ethnic groups were matched for age and BMI and had similar blood pressure and waist circumference (a surrogate measure of central adiposity) measurements as shown in Table 1. However the Indo-Asian volunteers displayed blood lipid profiles that were typical of an atherogenic lipoprotein phenotype (Table 2). The Indo-Asian group had significantly higher plasma triacylglycerol concentrations (P < 0.01), percentages of LDL3 (P < 0.01), and apo B concentrations (P < 0.05). In addition, Indo-Asian participants tended to have higher plasma insulin (P = 0.09) and HOMA-IR (P = 0.09) values and lower plasma HDL-cholesterol concentrations (P = 0.08), although these differences were not statistically significant.
View this table:
TABLE 2. Baseline values and changes from baseline for plasma variables in the European and Indo-Asian subjects after fish-oil (FO) and olive oil (OO) supplementation1
Nutrient intake
There were significant differences in macronutrient intakes between the 2 ethnic groups (Table 1). The Indo-Asian subjects had significantly lower intakes (expressed as % of total energy) of protein (P < 0.01), alcohol (P < 0.001), and cholesterol (P < 0.05) but significantly higher intakes of total fat (P < 0.05). Fatty acid intakes were also significantly different between the ethnic groups. Notably, the Indo-Asian group had significantly higher dietary intakes of total PUFAs (P = 0.001), n–6 PUFAs (P = 0.001), and monounsaturated fatty acids (P = 0.01) and significantly lower intakes of EPA (P = 0.001) and DHA (P = 0.01). The Indo-Asians had a significantly higher ratio of n–6 to n–3 PUFAs than did the European subjects (P = 0.001) (Table 3).
View this table:
TABLE 3. Baseline values and changes from baseline for platelet phospholipid fatty acid composition in the European and Indo-Asian subjects after fish-oil (FO) and olive oil (OO) supplementation1
The estimated degree of underreporting in the 2 ethnic groups during the dietary assessment period was 9.5% (P = 0.01) in the European group and 1.5% (NS) in the Indo-Asian group. To determine whether this was due to undereating or to incorrect recording of foods consumed during the assessment period, body weight before and after dietary assessment was measured, and an estimate of activity was recorded. There was a significant reduction in weight of 0.6% (P < 0.003) in the European group but no significant change in reported activity levels (data not shown). There was a nonsignificant reduction of 0.4% in the Indo-Asian group and no significant change in reported activity levels (data not shown). Undereating in the European group contributed significantly to the underreporting observed.
Platelet phospholipid fatty acid composition at baseline
The fatty acid composition of platelet phospholipids is shown in Table 3. There was a significantly higher membrane arachidonic acid content in the Indo-Asian group than in the European group (P = 0.05) and significantly lower EPA (P = 0.001) and DHA contents (P = 0.0001). A The ratio of n–6 to n–3 PUFAs was also significantly higher in the Indo-Asian than in the European group (P = 0.05).
Protocol compliance
The subjects’ compliance with the study protocol was acceptable by all methods of assessment. The number of missed supplements, assessed by capsule counts, was not significantly different between the 2 ethnic groups (Europeans: 4 ± 7; Indo-Asians: 4 ± 8) or between the 2 intervention groups (fish-oil group: 4 ± 7; olive oil group: 4 ± 7); these findings were supported by self-reports. The platelet phospholipid fatty acid data (details given below) also confirmed that the subjects in the fish-oil group had a significant enrichment of their platelets with both EPA and DHA, whereas the olive oil group did not.
Biochemical outcomes
Mean changes from baseline in the fasting plasma variables are shown in Table 2 for the fish-oil and olive oil groups. When all subjects were combined, there was a significant interaction between time and treatment in plasma triacylglycerol (fish-oil group, –9 ± 31%; olive oil group, 10 ± 40%; P = 0.002), plasma apo B-48 (fish-oil group, –16 ± 41%; olive oil group, 3 ± 38%; P = 0.01), plasma HDL cholesterol (fish-oil group, 4 ± 17%; olive oil group, –3 ± 14%; P = 0.03), and plasma -tocopherol (fish-oil group, 5 ± 21%; olive oil group, –7 ± 28%; P = 0.02). Significant reductions in plasma triacylglycerol (P = 0.01) and apo B-48 (P = 0.05) were also observed in the fish-oil group over time. Although there was no interaction between ethnic groups and treatment groups, there was a tendency for the Indo-Asian fish-oil group to have a greater reduction in plasma triacylglycerol (Indo-Asians: –19 ± 27%, P = 0.01; Europeans: 0 ± 32%), apo B (Indo-Asians: –7 ± 14%, P = 0.04; Europeans: 1 ± 12%), and apo B-48 (Indo-Asians: –26 ± 32%, P = 0.01; Europeans: –6 ± 46%) than the European fish-oil group. Concentrations of triacylglycerol and apo B-48 returned to baseline values 4 wk after supplementation. A time-by-ethnicity interaction was observed for HOMA-IR (P = 0.04) and fasting insulin (P = 0.05); the European fish-oil and olive oil groups had nonsignificantly different changes in these measures of insulin resistance, whereas the Indo-Asian fish-oil group had a greater increase in HOMA-IR and fasting insulin than did the Indo-Asian olive oil group. No significant difference was observed in any of the other variables measured, including body weight (data not shown).
When all subjects were combined, there was a significant time-by-treatment interaction for the platelet phospholipid fatty acid content of EPA (fish-oil group, 200 ± 113%; olive oil group, 1 ± 36%; P = 0.0001), DHA (fish-oil group, 55 ± 39%; olive oil group, 1 ± 16; P = 0.0001), arachidonic acid (fish-oil group, –11 ± 10%; olive oil group, 0 ± 9%; P = 0.0001), and oleic acid (fish-oil group, 6 ± 10%; olive oil group, 1 ± 5%; P = 0.01) (Table 3). There was no treatment-by-ethnic group interaction.
DISCUSSION
The hypothesis that, compared with Europeans, British Indo-Asians have a lower LC n–3 PUFA body status associated with lower dietary intakes of these fatty acids was supported by the data from this study. The Indo-Asian subjects had a significantly lower platelet phospholipid composition (used as a surrogate marker for tissue fatty acid status) of the LC n–3 PUFAs EPA and DHA and a significantly higher platelet phospholipid composition of arachidonic acid than did the European subjects, which supports the findings of previous studies (9, 10, 13, 14). In the current study, the platelet phospholipid fatty acid composition in the European group was similar to values previously reported in European subjects (34–37). The findings of the present study support the view that Indo-Asians have a lower LC n–3 PUFA status than do Europeans and has established clear differences in fatty acid status between carefully matched Indo-Asians and Europeans. A key question is whether these differences reflect different dietary exposures or arise as a result of differences in fatty acid metabolism and tissue incorporation.
At baseline, the Indo-Asians had a higher dietary fat intake (% of energy) than did the Europeans, which was mainly due to the higher n–6 PUFA intake in the Indo-Asians; this finding supports previous reports in various Indo-Asian groups (8–10, 38). Although our data support data reported for other ethnic groups, the dietary habits of Indo-Asian Sikhs may differ from other Indo-Asian groups, and nutrient intakes need to be confirmed in other UK Indo-Asian populations. However, the mean dietary fat intake of the European group was lower than that previously reported (39, 40), although it was similar to the lower dietary intake of 35% energy as fat in the recently published National Diet and Nutrition Survey of adults aged 19–64 y (41). In the current study, significant underreporting—due in part to undereating—was observed in the Europeans, which could explain, in part, the lower reported fat intake in this ethnic group. The macronutrient intakes of the Indo-Asian and European groups were similar to values reported by previous authors (6–8, 38–42). The total n–3 PUFA intakes were similar in the 2 ethnic groups, although EPA and DHA intakes were significantly lower in the Indo-Asians than in the Europeans. The finding in the current study of a low dietary LC n–3 PUFA intake in Indo-Asians is important because the accepted view that LC n–3 PUFA intakes are lower in this ethnic group is based on limited data.
The daily dose of 2.5 g EPA and DHA resulted in a highly significant increase in the platelet phospholipid fatty acid enrichment of EPA and DHA and a decrease in the n–6:n–3 PUFA ratio in both groups, although there was a tendency to a more marked increase in the Indo-Asian fish-oil group. Greater increases in platelet phospholipid levels relative to intakes were evident for EPA than for DHA, which supports previous data (34, 37, 43, 44). This finding may be attributable to many factors, namely higher baseline DHA levels, selective release of DHA from membranes, retroconversion of DHA to EPA, or reduced conversion of EPA to DHA because of the known inhibitory effect of EPA on 6-desaturase (16, 45, 46). The increase in the LC n–3 PUFA composition of platelets in the Indo-Asians resulted in similar tissue concentrations of these fatty acids in the 2 groups at the end of fish-oil supplementation, compared with significantly lower EPA and DHA platelet phospholipid levels in the Indo-Asians than in the Europeans at baseline. These data support the conclusion that there was no resistance to incorporation of EPA and DHA in the Indo-Asian or European group when dietary intakes of EPA and DHA were high. Enrichment of the LC n–3 PUFAs occurred in the Indo-Asians after fish-oil supplementation despite the significantly higher ratio of dietary n–6 to n–3 PUFAs in this group than in the Europeans (11.2 compared with 6.7, respectively, before supplementation and 4.5 compared with 2.7, respectively, after supplementation). Therefore, the hypothesis that greater competition for incorporation into membranes from the high dietary n–6 PUFA attenuating effects of LC n–3 PUFA supplementation (47–50) in Indo-Asians is not supported by this study.
The data also show that the Indo-Asian group was not resistant to the hypotriacylglycerolemic effects of the fish-oil supplementation, because there were significant reductions in mean fasting plasma triacylglycerol and apo B-48 concentrations and no significant interaction of ethnicity on the effects of treatment. The reduction in plasma triacylglycerol observed in the combined group supports previous studies (19, 21). The reduction in plasma triacylglycerol observed in the Indo-Asian subjects in the current study was consistent with that observed in a study of native Indo-Asians given 1.4g LC n–3 PUFAs/d (29). The fact that apo B-48 concentrations were reduced to an extent similar to that for total plasma triacylglycerol supports the view that LC n–3 PUFAs have an important effect on the metabolism of dietary-derived and hepatic triacylglycerol-rich lipoproteins.
The Indo-Asian population in the current study had characteristics typical of the metabolic syndrome, which supports previous reports (11, 51–58). A lower dietary and body status of LC n–3 PUFAs in the context of a high dietary n–6 PUFA intake has been linked with insensitivity to the actions of insulin (47, 59). Insulin sensitivity has been shown to be positively associated with the proportion of LC n–3 PUFAs in membranes (48, 60, 61). However, direct evidence for an effect of the n–6:n–3 PUFA ratio of membranes on insulin action in humans is extremely limited, and the dramatic beneficial effects on glucose tolerance observed in animals have not been observed in human studies. Therefore, it was thought prudent to investigate the effect of LC n–3 PUFA supplementation on surrogate markers of insulin resistance (HOMA-IR and fasted insulin). Protective effects of fish consumption on glucose tolerance, progression of type 2 diabetes (62), and insulin sensitivity (63) have been reported. However, the data in the current study and the data from another study by our group (56) showed no beneficial effect of higher dietary intakes or higher membrane contents of LC n–3 PUFA on measures of insulin resistance (HOMA-IR, fasting insulin). However, the current study was not powered on the insulin-resistant measures; therefore, these observations may reflect a low power. More detailed studies are required to specifically address the effect of dietary fatty acids on insulin resistance.
A total of 14 mg total tocopherol/d (7.6 mg -tocopherol/d) was provided by the olive oil and fish-oil capsules. There was a significant difference in plasma -tocopherol concentrations between the 2 treatment groups over time: a 5% increase in the fish-oil group and a 7% reduction in the olive oil group. This finding suggests that plasma antioxidant status was not compromised after a moderate dose of LC n–3 PUFAs.
The results of the current study confirm that Indo-Asians have a greater susceptibility to an atherogenic lipoprotein phenotype and metabolic syndrome and a lower tissue status (measured by platelet phospholipid fatty acids) of EPA and DHA than do Europeans. The Indo-Asians also had a significantly lower dietary intake of LC n–3 PUFAs and a higher intake of n–6 PUFAs than did the Europeans. After fish-oil supplementation, there was significant incorporation of LC n–3 PUFAs into platelet phospholipids in both the Indo-Asians and the Europeans, which resulted in significant reductions in plasma triacylglycerol and apo B-48 concentrations. These data suggest that the Indo-Asian population studied was highly responsive to the incorporation and triacylglycerol-lowering effects of dietary LC n–3 PUFAs. Fish-oil supplementation contributed to a reversal in the lipid abnormalities typical of a middle-aged Indo-Asian population and should be considered as an important, yet simple, dietary manipulation to reduce the risk of CAD in Indo-Asians with an atherogenic lipoprotein phenotype.
ACKNOWLEDGMENTS
We thank Bruce Griffin for analyzing the LDL subclasses, Chris Armah for analyzing the plasma -tocopherol concentrations, and the volunteers for their time and enthusiasm.
JAL, AMM, and CMW designed the study. SVML, SSL, and NS collected the data. SSL, SVML, and LMB analyzed the data. JAL wrote the manuscript. The authors had no conflicts of interest.
REFERENCES