R McArthur’s research while affiliated with University of North Carolina at Chapel Hill and other places

The aim was to establish whether interesterification of oils, an alternative to the use of trans fatty acids in margarine manufacture, adversely affects plasma lipids. Twenty-seven hypercholesterolemic men participated in a double-blind, crossover trial of three margarines: 1) high-linoleic acid, moderate trans fatty acids; 2) high-palm oil blend (predominantly lauric, myristic, palmitic, oleic, and linoleic acids); and 3) an interesterified form of the high-palm oil margarine. Both high-palm oil margarines led to similar low-density-lipoprotein (LDL) cholesterol concentrations (4.43 +/- 0.94 and 4.54 +/- 0.88 mmol/L, respectively), which were significantly higher than the LDL concentrations after the high-linoleic acid margarine (4.02 +/- 0.85 mmol/L, P < 0.001). Interesterification transferred substantial proportions of palmitic acid into the sn-2 position of glycerides and unsaturated fatty acids into the sn-1,3 positions, and this was reflected in plasma chylomicrons. This study shows that interesterification of oils used to harden margarines does not raise plasma cholesterol more than does the margarine's constituent fatty acids.

To determine the blood pressure responses in elderly normotensive men and women to dietary sodium and to the diunsaturated fatty acid dihomogammalinolenic acid (DGLA), which is derived from linoleic acid. Blood pressure responses were assessed in 66 subjects (36 male, 30 female; mean age 65 years) on two diets differing by approximately 70 mmol/day sodium, combined with daily supplements of either 1 g DGLA or 1 g safflower oil, giving a four-group parallel design. After a common period of salt restriction and salt supplementation, two sets of parallel groups continued with either salt or placebo tablets. The study was blinded, except for dietary adjustments based on 24-h urinary sodium excretion values measured once every 2 weeks. Blood pressures were also measured automatically once every 2 weeks. Urinary sodium excretion (sodium intake) correlated significantly with systolic and diastolic blood pressures. A strong interaction with sex (P < 0.001 for systolic blood pressure) reflected greater responsiveness in women to changing sodium intake. A second major determinant of blood pressure responsiveness was the waist: hip ratio, an index of central obesity; this correlation was independent of the initial sodium intake, initial blood pressure or body mass index. The waist: hip ratio was a powerful predictor of blood pressure changes with sodium intake in women only; women with android fat distribution were, similarly to men, less sensitive to dietary sodium. Daily supplements of 1 g DGLA doubled the concentration of DGLA in plasma but did not influence blood pressure. Among elderly normotensive subjects, women responded to changes in sodium intake with greater changes in blood pressure than men did. Furthermore, this response was strongly related to the gynaecoid distribution of body fat.

The effect of additional dietary trans fatty acids (7% energy) on plasma lipids was assessed in a double-blind comparison of four separate diets: 1, enriched with butter fat (lauric-myristic-palmitic); 2, oleic acid-rich; 3, elaidic acid-rich; 4, palmitic acid-rich. The total dietary period was 11 weeks and comprised normal foods plus specific fat supplements. In 27 mildly hypercholesterolemic men, total and LDL cholesterol were significantly lower during the 3-week oleic acid-rich diet, and were similar during the other three diets. For the four diets LDL cholesterol levels were in mg/dl: 1, 163; 2, 151; 3, 165; 4, 161. HDL cholesterol was significantly higher with the palmitic acid-rich diet, 42 mg/dl, compared with elaidic acid, 38 mg/dl, which in turn was not lower than with oleic acid, 38 mg/dl. Plasma elaidic acid concentration rose seven-fold with the trans fatty acid diet but did not increase the vulnerability of LDL to oxidative change. The elaidic acid-rich diet led to significant elevations in the level of Lp[a] compared to all the other test diets. The Lp[a] level increased to 296 +/- 220 U/l in the elaidic acid-rich period from 235 +/- 182 (mean +/- SD) in the first ("butter") period (P less than 0.001) compared with 249 +/- 204 in the palmitic acid period (P less than 0.001) and 236 +/- 201 in the oleic acid period (NS).(ABSTRACT TRUNCATED AT 250 WORDS)

The effects of oat bran and wheat bran on plasma lipid concentrations were compared in a crossover study. Each bran (123 g oat bran or 54 g wheat bran) added nearly 18 g of nonstarch polysaccharide to a background diet containing about 10 g nonstarch polysaccharide. Twenty-three men (average plasma cholesterol level = 5.84 mmol/L, and low-density-lipoprotein (LDL) cholesterol level = 4.11 mmol/L) were randomly assigned to either the oat or wheat bran diet for 4 weeks and then changed to the alternate bran diet for a similar period. The oat bran diet produced significantly lower levels of plasma total cholesterol and LDL cholesterol: 5.65 +/- 0.16 and 3.88 +/- 0.15 mmol/L (mean +/- standard error) for oat bran vs 5.89 +/- 0.16 and 4.11 +/- 0.16 mmol/L for wheat bran. Food intake diaries showed that average consumption of total fat and saturated fat was identical during the two test periods, which excluded displacement of fat as an explanation for lowering of plasma cholesterol by oat bran. Our results indicate that in mildly hypercholesterolemic men, a diet high in soluble oat fiber can significantly lower plasma total cholesterol and LDL cholesterol and thus potentially lower the risk of coronary heart disease.

We tested semihardened blends of edible oils, suitable for commercial food manufacture, with a lower-than-conventional saturated fatty acid content, for their effects on plasma cholesterol. Twenty-six mildly hypercholesterolemic men took part in a double-blind crossover experiment in which two test blends were compared with two control dietary periods [which resembled the Australian fat intake: proportions of polyunsaturated, monounsaturated, and saturated fatty acids (PMS) 0.4:0.9:1]. PMS in the test diets was approximately 0.8:1.3:1 and resulted in significantly lower LDL-cholesterol concentrations (reductions of less than or equal to 7.7%). HDL cholesterol and plasma triglyceride were unchanged. The trans fatty acid (mainly elaidic) content of the blends was 16%, raising its contribution to energy by 4% but without apparent effect on LDL and HDL concentrations. Provided the overall ratio of linoleic acid to palmitic acid in commercial edible-oil blends exceeds that in the prevailing national diet, partial hydrogenation will not negate the LDL-lowering potential.

Twenty-one mildly hypercholesterolemic men aged 30-59 y were provided with comparable barley and wheat foods for each of 4 wk in a crossover-designed experiment. The purpose of the study was to examine the influence of two sources of dietary fiber (nonstarch polysaccharides, NSP) on blood lipids and glucose concentrations. Barley contains beta-glucan as a source of soluble dietary fiber (DF) whereas wheat contains the largely insoluble cellulose and hemicellulose fiber. Total dietary fiber increased from a previous intake of 21-38 g/d during the period of study for the two groups. Consumption of barley relative to wheat foods was associated with a significant fall in both plasma total cholesterol (6%, P less than 0.05) and in low-density-lipoprotein cholesterol (7%, P less than 0.02) whereas triglyceride and glucose concentrations did not change significantly. It is concluded that barley dietary fiber is more effective than wheat dietary fiber at lowering blood cholesterol in hypercholesterolemic men.

Twenty-one mildly hypercholesterolemic men aged 30–59 y were provided with comparable barley and wheat foods for each of 4 wk in a crossover-designed experiment. The purpose of the study was to examine the influence of two sources of dietary fiber (nonstarch polysaccharides, NSP) on blood lipids and glucose concentrations. Barley contains β-glucan as a source of soluble dietary fiber (DF) whereas wheat contains the largely insoluble cellulose and hemicellulose fiber. Total dietary fiber increased from a previous intake of 21 –38 g/d during the period of study for the two groups. Consumption of barley relative to wheat foods was associated with a significant fall in both plasma total cholesterol (6%, P < 0.05) and in low-density-lipoprotein cholesterol (7%, P < 0.02) whereas triglyceride and glucose concentrations did not change significantly. It is concluded that barley dietary fiber is more effective than wheat dietary fiber at lowering blood cholesterol in hypercholesterolemic men.

The plasma cholesterol lowering potential of canned baked beans was examined in a cross-over comparison with canned spaghetti. The difference in total dietary non-starch polysaccharide (NSP) of 12 g daily (6.6 g difference in soluble NSP), was insufficient to alter the plasma cholesterol, HDL cholesterol, triglyceride and glucose concentrations in 20 mildly hyper-cholesterolaemic men. Thus, eating an average of six 440 g cans of this source of baked beans per week, large servings, does not lower the plasma cholesterol when dietary fat is not displaced.