Erythrocyte membrane fatty acid composition as a marker of dietary compliance in hyperlipidaemic subjects

Erythrocyte membrane fatty acid composition as a marker of dietary compliance in hyperlipidaemic subjects

ATHEROSCLEROSIS Atherosclerosis I 17 (1995) 245-252 Erythrocyte membrane fatty acid composition as a marker of dietary compliance in hyperlipidaemic...

702KB Sizes 0 Downloads 60 Views

ATHEROSCLEROSIS Atherosclerosis

I 17 (1995) 245-252

Erythrocyte membrane fatty acid composition as a marker of dietary compliance in hyperlipidaemic subjects Maria B. Tynan”, D. Paul Nicholls*‘, Suzanne M. Maguire”, Ian C. Steele”, Cyril McMaster b, Raymond Moore b, Elisabeth R. Trimble’, Jack Pearce“ “Department of’ Medicine, Royal Victoria Hospital, Belfast BTl2 6BA. UK bDepartment of’ Child Health, Royal Victoria Hospital, Belfast BTI2 6BA, UK ‘Department of Clinical Biochemistry, The Queen’s University BeJfkst and Royal Victoria Hospital, Be/j&t BTl2 6BA. UK ‘Department of’ Food Science, The Queen’s Unirersity 0f’Belfir.v. Ne\$orge Lane. Be!first BT9 5PX. C’K

Received5 April 1994: revision received 27 February 1995; accepted 8 March 1995

Abstract

Dietary intervention is the first treatment step in the management of hyperlipidaemia, but there are few objective criteria of compliance. Whether intensive dietary intervention would produce a detectable change in erythrocyte membrane fatty acid composition which could be used as a marker of compliance was examined in 31 new hyperlipidaemic patients. Over a 6 month period, body mass index fell from 29.0 to 26.9 kg/m2 (P < 0.001) and total cholesterol by 19% from 8.16 to 6.58 mmol/l (P < 0.001). The energy derived from fat was reduced from 38.5% to 29.6X (P < O.OOl),and the ratio of dietary polyunsaturated to saturated (PS) fatty acids in the diet increased from 0.45 to 0.66 (P < 0.01). Small but significant changes were recorded in several red cell membrane fatty acids, and the P:S ratio increased from 0.91 to 1.13 (P < 0.001). It would appear, therefore, that red cell membrane changes parallel dietary changes and hence are a potential marker for compliance with dietary changes. Keywords:

Diet: Erythrocytes; Fatty acids; Hyperlipidaemia

1. Introduction Dietary intervention is the cornerstone of the treatment of hyperlipidaemia [I]. Current recommendations are that the percentage energy derived from fat should not exceed 30% and that the contribution of saturated fatty acids should not * Corresponding 002l-915Oi95j$O9.50

author.

exceed 10% [2], the remaining fat coming from polyunsaturated and monounsaturated sources. In practice, such dietary advice amounts to sensible eating within the normal range of foods, and so it is difficult to assess patient compliance. Detailed dietary histories are time consuming to take, require considerable skill, but remain largely subjective, and so there is a need for an objective test of dietary compliance.

0 1995 Elsevier Science Ireland Ltd. All rights reserved

SSDI 0021-9 150(95)05578-K.

246

M.B. Tynan et al. / Atherosclerosis

The membrane lipids of erythrocytes and platelets should equilibrate with the surrounding plasma, as both lack the enzymatic apparatus necessary to synthesize or esterify lipids [3]. Diet affects plasma fatty acids and hence erythrocyte and platelet membrane composition [4-81. Maximal changes occur about 4 weeks after the diet has been varied in lipid composition and it has, therefore, been suggested that erythrocyte fatty acid composition could be used as a marker of dietary compliance [9,10]. The aim of the present study was to determine whether qualitative changes in red cell membrane fatty acid composition could be detected following the changes in dietary composition made for therapeutic purposes in a lipid clinic. 2. Methods 2.1. Subjects In the first part of the study, the effects of hypercholesterolaemia on erythrocyte fatty acid composition were examined. Blood samples were taken in the fasting state from 12 patients presenting to the Lipid Clinic at the Royal Victoria Hospital for treatment of primary hyperlipidaemia (Frederickson type IIa in nine, IIb in three). None were taking any drugs and none had been given any previous dietary advice. Seven were male; the mean age was 55.9 years (range 44-64) body weight 77 kg (56-102) and body mass index (BMI) 28.3 kg/m2 (23.0-36.5). None had familial hypercholesterolaemia as defined by tendon xanthomata in the patient or a first degree relative. The samples were compared with those from 12 age- and sex-matched normocholesterolaemic control subjects, also in the fasting state. Seven were male and the mean age of the group was 57.1 years (range 40-70) body weight 76.7 kg (62-94) and body mass index 27.4 kg/m2 (21.8-41.2). In the second part of the study, 31 newly diagnosed subjects presenting to the Lipid Clinic for treatment of primary hyperlipidaemia (10 male, age 52 + 10 years, BMI 29 f 4 kg/m2) were recruited to assessthe response to dietary intervention. None had familial hypercholesterolaemia, as defined above; 17 were classified as

117 (1995) 245-252

Frederickson type IIa and 14 as type IIb. The subjects had not received any previous dietary counselling and were not taking hypolipidaemic drugs or any dietary supplements such as lecithin, garlic or fish oil capsules. Subjects were seen at weeks 0 (baseline), 12 and 24. 2.2. Dietary analysis Subjects were visited at home and asked to complete a weighed inventory of all food and drink consumed for the 7 days preceding the hospital visits at weeks 0, 12 and 24, and to record this in a food diary. After this initial home visit, subjects were asked not to change their dietary habit, so that the 7 day collection prior to week 0 would represent a typical week’s food intake. Instructions were given on the use of an electronic weighing scale (Salter Microtronic Electronic Scale, Model 2001; Salter Houseware, Tonbridge, UK). Subjects were asked to record brand names of foods and to provide a complete description of the method of preparation and cooking, and recipes for composite dishes. Food and drink items were coded for identification purposes and analysed by computer. Nutrient intakes were calculated using data from food composition tables [l l] and additional supplemental tables [12]. The fatty acid composition of the diet was also calculated using recommended methods [ 131 and data from food manufacturers. Body mass index (BMI) was calculated at weeks 0, 12 and 24. Weight was measured in light clothing without shoes using a Seca digitial electronic scale accurate to 200 g. Height was measured using a stadiometer. 2.3. Nutrition education At week 0 initial assessmentsof BMI and dietary patterns were made as above. Lipid lowering dietary advice was given based on the European Atherosclerosis Society guidelines [14]. The dietary goals were: (i) to attain ideal body weight for height; (ii) to decreasetotal fat intake to 30% of total energy intake; (iii) to reduce saturated fatty acid intake to less than 10% of total energy with the remaining fat coming from polyunsaturated and monounsaturated sources; (iv) to increase dietary fibre to 30-35 g daily; (v) to reduce

M.B.

Tynan

et al.

Atherosclerosis

dietary cholesterol intake to less than 300 mg daily. At weeks 12 and 24 subjects returned for follow-up consultations. 2.4. Blood sampling and analysis At weeks 0, 12 and 24 a fasting blood sample of 15 ml was drawn. Total serum cholesterol and triglyceride were measured by enzymatic methods on a Cobas Bio centrifugal analyser (Roche), HDL-cholesterol by heparin-manganese or phosphotungstic acid precipitation, and LDL-cholesterol computed from the Friedwald equation. For analysis of erythrocyte membrane fatty acid composition, 10 ml of whole blood was collected into a heparinised tube. Extraction of the erythrocyte membranes was performed according to the method of Dodge and Phillips [ 151as modified by Hanahn and Ekholm [16]. After centrifugation for 10 min at 4°C and removal of the plasma and the huffy coat, the remaining erythrocytes were washed twice with. normal saline and centrifuged at 1500 x g for 10 min between each washing. The third wash was given with Trolox (97% 6-hydroxy 5,7,8-tetramethylchroman-2-carboxylic acid (Aldrich Chemical Co.; Dorset, UK)) 5% in normal saline. This stabiliser solution was added to prevent oxidation of the membrane fatty acids [17,18] when it was not possible to deliver the samples immediately to the laboratory for analysis (these samples were stored at - 70°C for 4-6 weeks). Aliquots of erythrocytes (0.5 ml) were transferred into clean glass tubes and 6 ml chloroforrr-methanol (2:1, v/v) containing 10 mg% butylated hydroxytoluene (BHT) added, mixed and allowed to extract for 30 min with occasional mixing. Calcium chloride solution (0.2%, w/v; 2 ml) was added and the solution mixed and centrifuged at 1500 x g for 10 min. Methylation of the samples was carried out as described by Morrison and Smith [19]. The total fatty acid composition was determined by manually injecting 1 ~1 aliquots into a Perkin Elmer Sigma chromatograph equipped with a PE LCI 100 laboratory computing integrator and 25 m x 0.22 mm i.d. BP 20, fused silica capillary column, film thickness 0.25 pm (SGE, Milton Keynes, UK) for analysis. Helium gas

Ii 7 (1995)

245-252

241

flow rate was 1.5-2.0 mlimin, flame ionization temperature 300°C and injector temperature 240°C. The oven temperature was programmed from 70”-160°C at 25”C/min and then at 160”250°C at 3”C/min, isothermal for 20 min. The fatty acid methyl esters were identified according to their retention times by comparison with known standards (Sigma Chemical Company, Dorset, UK; Supelchem, Herts, UK; Field Analytical Co., Surrey, UK). Serum lipids and erythrocyte membrane fatty acid composition were re-evaluated at the week 12 and week 24 visits. 2.5. Stutisticul analysis Results from the two groups in the first study were compared using non-parametric statistics (Mann-Whitney U-test). Coefficient of variation (CV, O/o)was calculated from the mean and S.D. of five consecutive measurements of the same sample as (S.D./mean) x 100. In the second study, analysis of variance was used. 3. Results

In the first study, the mean total serum cholesterol concentration in the patient group was 9.04 mmol/l (range 8.0-10.3) with an HDL cholesterol level of 1.45 mmol/l (0.84-2.30), an LDL cholesterol concentration of 6.58 mmol/l (4.817.65) and a fasting triglyceride level of 2.35 mmol/ 1 (1.15-5.36). In the control group, the corresponding figures were: total cholesterol 5.33 mmol/l (4.12-6.49), HDL cholesterol 1.39 mmol/l (0.8442.21), LDL cholesterol 3.45 mmol/l (2.484.72) and triglyceride 1.06 mmol/l (0.53- 1.49). The fatty acid composition of the erythrocyte membranes in the patients and the controls showed only minor differences (Table 1). Five fatty acids (palmitic, 16:O;stearic, l&O:, oleic, 18:1; linoleic 18:2; arachidonic, 20:4) accounted for an average of 86.6% of the total in patients and 89.6% in controls. Not shown are the fatty acids that contributed < 1% of the total. In the second study, significant reductions in BMI, serum total cholesterol, LDL-cholesterol and triglyceride values were observed at weeks 12 and 24 (Table 2). HDL-cholesterol increased sig-

248

M.B. Tynan et al. 1 Atherosclerosis

Table 1 Erythrocyte membrane total fatty acid composition (%) in hyperlipidaemic patients (n = 12) and normal controls (n = 12) Fatty acid

Patients

Controls

cv (%)

14:o l6:O 16:ln7 18:0 l8:l 18:2n6 20:3n6 20:4n6 20:5n3 22:4n6 22:5n3 22:6n3

0.6 k 0.4 29.6 k 3.3 1.1 +0.6 12.4+ 2.3 17.7+ 2.0 14.1&- 1.7 1.4 + 0.6 12.8 + 2.1 1.350.4 1.1 *0.9* 1.8 + 1.0 4.3 f 1.9*

1.0 kO.5 32.2 f 2.4 0.9kO.3 12.1* 1.9 19.5+ 2.1 14.0f 1.2 l.OkO.4 11.8f 1.9 1.1 kO.9 2.0 f 0.7 1.8 +0.4 2.8 k 0.7

4.1 3.2 12.7 3.2 16.4 4.8 14.3 14.3 13.2

Mean f S.D. values are shown. CV, coefficient of variation of assay. *P < 0.05 (Mann-Whitney U-test).

nificantly at week 24. During this time, significant changes in dietary intake were also observed (Table 3). In particular, total energy intake was reduced and the percentage energy derived from fat fell from 38.5 to 29.6% (P < 0.001). The polyunsaturated to saturated fatty acids (PCS)ratio increased from 0.45 to 0.66 (P < O.Ol), and the linoleic to oleic acid (L:O) ratio from 0.43 to 0.61 (P < 0.01). BMI correlated with serum total cholesterol at week 24 (r= 0.45, P < 0.01) and with serum triglyceride levels at weeks 12 (r = 0.39, P < 0.05) and 24 (r = 0.60, P < O.OOl),but not with either at baseline (week 0). The changes in erythrocyte membrane fatty acids are given in Table 4. Although the changes in individual levels were small, there was a de-

117 (1995) 245-252

creasein the major saturated fatty acids (myristic, palmitic and stearic) and an increase in other fatty acids, so that the P:S ratio increased from 0.91 to 1.13 (P < 0.001). The L:O ratio in red cell membranes remained unchanged over the 6 month period (0.79 vs. 0.82). The individual changes in P:S ratio are shown in Fig. 1. In 27 patients, the ratio increased during the study period. In two there was no change and in two the ratio reduced. In addition to the correlations observed above, comparisons were also made between: dietary and red cell membrane saturated fatty acid content; dietary and membrane polyunsaturated and monounsaturated fat content; dietary and membrane content of each individual fatty acid; dietary and membrane L:O and P:S ratios; and changes in serum lipid fractions and changes in membrane fatty acids, L:O and P:S ratio. None were significant, except that the dietary and red cell membrane contents of eicosapentanoic acid (20:5; r = 0.67; P < 0.001) and docosa-hexanoic acid (22:6; r = 0.54; P < 0.01) were related. 4. Discussion Since the use of erythrocyte membrane fatty acid composition as a marker for dietary compliance was first suggested [4], it has been applied rather infrequently, although study of the fatty acid pattern in the red cell has several advantages. No fatty acid synthesis, chain elongation or desaturation occurs in the membrane [20], and hence it reflects dietary fatty acid composition. Maximal changes occur about 4 weeks after the fatty acid composition of the diet has been varied and remain unchanged thereafter, so that membrane

Table 2 Body mass index and serum lipid fractions in 31 hyperlipaemic subjects at weeks 0, 12 and 24 Week

Body mass index (kg/m2) Total cholesterol (mmol/l) Triglyceride (mmol/l) HDL-cholesterol (mmol/l) LDL-cholesterol (mmol/l)

0

12

24

S.E.M. (d.f. = 60)

F ratio (PI

29.0 8.16 2.63 1.35 5.64

28.1 7.32 2.36 1.37 4.89

26.9 6.58 1.95 1.50 4.19

0.171 0.111 0.097 0.038 0.112


M. B. Tynan er ul. I Atherosclerosis

249

117 (1995) 24.L25.2

Table 3 Results of 7 day food diary dietary analysis in 31 patients Week

S.E.M. (d.f. = 60)

P

24 12 0 -
changes reflect diet over the preceding few weeks or months [4,21]. In contrast, large changesin plasma or platelet fatty acid composition can occur over a few days, and adipose tissue fatty acid patterns change so slowly that they are only a valid index of the habitual diet over a period of 2-3 years [22]. The use of erythrocyte membrane fatty acid composition may therefore be considered analogous to the use of glycosylated haemoglobin levels as an index of glycaemic control in diabetes mellitus. The first part of the present study showed that the fatty acid composition of the red cell membrane is similar in hypercholesterolaemic patients and in a control group. This suggeststhat the diet of the two groups was also similar, and that the diet of the patients corresponded to that of the general population who had not received any specific dietary advice. Other studies have also indicated that erythrocyte membrane composition remains normal despite a hyperlipidaemic environment [23], although morphological changesin the red cell may occur [24-261.

In the second part of the study, the observed reduction in serum total cholesterol of about 19% over a 6 month period is similar to or higher than that in other studies of intensive dietary management [27,28]. However, it must be acknowledged that the present study did not include a control group for comparison, although it is unlikely that such a reduction would occur by chance alone. LDL-cholesterol decreasedby 26% over the same time period and HDL-cholesterol increased, in contrast to previous studies which have shown no change or even a decrease [29,30]. A possible explanation for the increase of HDL-cholesterol may be that patients were encouraged to increase daily exercise as part of overall lifestyle improvements. The changes observed in serum triglyceride concentrations may relate to a reduction in BMI, as has been previously documented [31,32]. Several interesting findings emerged over the study period when comparing the fatty acid composition of the diet with that of the erythrocyte membrane. Dietary intake of myristic (14:O) and

250

M.B. Tynan et al. / Atherosclerosis

1I7 (1995) 245-2.52

Table 4 Fatty acid composition of total erythrocyte membrane in 31 hyperlipidaemic subjects at weeks 0, 12 and 24 Week

14:Omyristic 16:Opalmitic 16:I palmitoleic 18:Ostearic 18:ln-9 oleic 18:2n-6 linoleic 18:3n-3 alpha linoleic 18:3n-6 gamma linoleic 18:4n-3 morotic 20:I n-9 eicosanoic 20:2n-6 eicosadienoic 20:3n-6 dihomogamma linoleic 20:3n-9 eicosatrienoic 20:4n-6 arachidonic 20:5n-3 eicosapentanoic 22:1n-9 docosanoic 22:4n-6 docosatetranoic 22:5n-3 alpha docosapentanoic 22:5n-6 gamma docosapentanoic 22:6n-3 docosahexanoic Saturated fatty acids, % of total Polyunsaturated fatty acids, ‘l/oof total Monounsaturated fatty acids, % of total P:S ratio

S.E.M. (d.f. = 60)

P

0

12

24

0.63 29.51 1.08 12.23 18.03 14.14 0.05 0.03 0.06 0.07 0.00 1.27 0.06 12.82 1.08 0.00 2.19 2.20 0.12 4.18

0.56 28.76 1.01 Il.53 18.49 14.18 0.04 0.02 0.05 0.04 0.00 1.29 0.10 13.56 1.08 0.00 2.39 I .97 0.25 4.24

0.46 26.13 0.86 11.85 17.69 14.43 0.05 0.02 0.10 0.06 0.00 1.68 0.16 14.75 1.09 0.00 2.61 2.48 0.20 5.13

0.041 0.480 0.051 0.325 0.301 0.229 0.014 0.014 0.017 0.023 0.000 0.104 0.046 0.215 0.090 0.00 0.096 0.107 0.041 0.202

< 0.05 < 0.001 < 0.01 N.S. N.S. N.S. N.S. N.S. N.S. N.S. N.S. < 0.001 N.S. < 0.001 N.S. N.S. < 0.05 < 0.01 N.S. < 0.01

42.52

41.03

38.55

0.479

< 0.001

38.26

39.35

42.81

0.514

< 0.001

19.22 0.91

19.63 0.98

18.65 1.13

0.322 0.026

N.S. < 0.001

Values are means (n = 31). Amounts are expressed as a percentage of total fatty acid composition unless otherwise indicated.

palmitic (16:O) acids decreased significantly, and this was reflected in the erythrocyte membrane. Intake of dietary stearic acid (18:0) decreasedand a small reduction in erythrocyte stearic acid content was also observed. The only other dietary fatty acid to show a significant change over the study period was linoleic acid (18:2), which increased, but this was not reflected in the linoleic acid content of the erythrocyte membrane. However, the amounts of dihomogamma linolenic (20:3n-6), arachidonic (20:4n-6) and docosatetranoic (22:4n-6) acids in the erythrocyte membrane did increase significantly. Each of these fatty acids is synthesised from linoleic acid. Neither linoleic (18:2n-6) nor alpha linolenic (18:3n - 3) acid can be synthesised by mammalian cells [33.34]. Both

undergo chain elongation and desaturation by the same enzymes and hence competition occurs. At the same relative concentration long chain polyunsaturated derivatives are formed more readily from 18:3n-3, but this preference can be reversed by increasing the relative concentration of 18:2n-6 so that when the diet has a high 18:2n-6/18:3n-3 ratio, chain elongation and desaturation of 18:3n-3 is strongly suppressed [35,36]. This may explain the increase in n-6 derivatives seen in the erythrocyte membrane, as the dietary intake of 18:2n-6 increased significantly but intake of 18:3n-3 remained constant. The significant increase in 20:4n-6 in the erythrocyte membrane is likely to be due to conversion of 18:2n-6 as no significant dietary increase of 20:4n-6 was ob-

I

1.2

*

3

:

1 0.6 1

4

_

2

3

pi 4

5

~~

,

~---~+

6

7

8

~~. 9

10

1

-c

~_

;

11

12

13

14

individual

~-~~~~15

.~~-~

18

17

~_ ~~_~_ 18

19

20

.-. 21

22

+ ~~_~~~+ 23

24

25

26

_~

~_~

27

28

1. .~_~_ 29

30

31

Patient Number

Fig. I. Individual values for polyunsaturateddsaturated (P:S) fatty acid ratio in erythrocyte laemic patients before and after 6 months of dietary intervention.

served. The increases in the concentrations of eicosapentanoic (20:5n-3) and docosahexanoic (22:6n-3) acids in the erythrocyte mambrane may be due to an increased dietary intake of fish [8], as subjects were encouraged to eat fish two or three times per week. Current measures of dietary compliance may be unsatisfactory for a variety of reasons. Non-laboratory measures such as dietary recall may be inaccurate because of poor memory, or the tendency of patients to eat differently when keeping a food record [37]. Laboratory analyses such as serum cholesterol or triglyceride concentrations may also lack validity because of other influences and the variation in response to diet [38]. It has been reported that the value of erythrocyte membrane fatty acid composition as a monitor of dietary fatty acid intake is limited [39], but the data in this study would suggest otherwise. In particular, an increase in membrane P:S ratio could prove to be a useful marker. In conclusion, although it is envisaged that changes in body weight. serum lipid levels and subjective assessment of the diet will remain the principal methods to assesslipid-lowering dietary compliance in a

~.

membranes from 31 hypercholestero-

clinical setting, the erythrocyte membrane fatty acid composition may be a useful objective marker for the future in research or larger epidemiological studies. It could perhaps be used to enhance dietary assessmentinterviews, serving as a control in investigating fatty acid intake. References [I] Betteridge DJ, Dodson PM, Durrington PN, Hughes. EA, Laker MF, Nicholls DP, Rees JAE. Seymour CA, Thompson CR, Winder AF, Winocour PH. Wray R. Management of hyperlipidaemia: guidelines of the British Association. Postgrad Hyperlipidaemia Med J 1993;63:359. [2] Committee on Medical Aspects of Food Policy. Diet and Cardiovascular Disease. DHSS Report on Health and Social Subjects, No. 28. London: HMSO, 1984. [3] Cooper RA. Abnormalities of cell-membrane fluidity in the pathogenesis of disease. New Engl J Med 1971;297:311. [4] Farquhar JW, Ahrens EH. Effect of dietary fats on human erythrocyte fatty acid patterns. J Clin Invest 1963;42:675. [S] Dougherty RM. Galli C, Ferro-Luzzi A. Iacono JM. Lipid and phospholipid fatty acid composition of plasma. red blood cells, and platelets and how they are affected by dietary lipids: a study of normal subjects from Italy. Finland and the USA. Am J Clin Nutr 1987:45:443.

252

M.B. Tynan et al. / Atherosclerosis

PI Sagher FA. The effects of variation in dietary fat in early

life, with particular reference to the small intestine. PhD Thesis, Queen’s University of Belfast, 1988. [71 Pagnan A, Corrocher R, Ambrosio GB, Ferrari S, Guarini P, Piccolo D, Opportuno A, Bassi A, Olivieri 0, Baggio G. Effects of an olive-oil-rich diet on erythrocyte membrane lipid composition and cation transport systems. Clin Sci 1989;76:87. 181Hessel E, Agren JJ, Paulitschke M, Hanninen 0, Hanninen A, Lerche D. Freshwater fish diet affects lipid composition, deformability and aggregation properties of erythrocytes. Athersclerosis 1990;82:37. 191 Angelic0 F, Arca M, Calvieri A, Cantafora A, Guccione P, Monini P, Montali A, Ricci G. Plasma and erythrocyte fatty acids: a methodology for evaluation of hypocholesterolaemic dietary interventions. Prev Med 1983;12:124. 1101Gloatz JFC, Soffers AEMF, Katan MB. Fatty acid composition of serum cholesteryl esters and erythrocyte membranes as indicators of linoleic acid intake in man. Am J Clin Nutr 1989;49:269. [Ill Paul AA, Southgate DAT. McCance and Widdowson’s “The Composition of Foods”, 4th Edn. London: HMSO, 1978. [121Paul AA, Southgate DAT, Russell J. First Supplement to McCance and Widdowson’s “The Composition of Foods”. London: HMSO, 1980. [I31 Broadhurst AI, Stockey L, Wharf SG, Faulks RM, Renson JM. Validity of calculating fatty acid intake from mixed diets. Human Nutr Appl Nutr 1987;41A:lOl. [I41 European Atherosclerosis Society Study Group. The recognition and management of hyperlipidaemia in adults: a policy statement of the European Atherosclerosis Society. Eur Heart J 1988;9:571. [I51 Dodge JT, Phillips GB. Composition of phospholipids and of phospholipid fatty acid and aldehydes in human red cells. J Lipid Res 1967;8:667. V61 Hanahn DJ, Ekholm J. Changes in erythrocyte membranes during preparation, as expressed by ATPase activity. Biochim Biophys Acta 1972;255:413. [I71 Scott JW, Cort WM, Harley JH, Parrish DR, Saucy G. 6-Hydroxychroman-2-carboxylic acids: novel anti-oxidants. J Am Oil Chem Sot 1974;51:200. 1181 Cort WM, Scott JW, Harley JH. Proposed antioxidant exhibits useful properties. Food Technol 1975;Nov:46. 1191 Morrison WR, Smith LM. Preparation of fatty acid methyl esters and dimethylacetals from lipids with boron fluoride-methanol. J Lipid Res 1964;5:600. PI Pittman JG, Martin DB. Fatty acid biosynthesis in human erythrocytes - evidence in mature erythrocytes for an incomplete long chain fatty acid synthesising system. J Clin Invest 1966;45:165. Pll Horwitt MK, Harvey C, Century B. Effect of dietary fats on fatty acid composition of human erythrocyte and chick cerebella. Science 1959;130:917. WI Beynen AC. Plasma and liver cholesterol concentrations in mice fed semipurified diets differing in the amount of

117 (1995) 245-252

cholesterol and type of fat. Int J Vit Nutr Res 1987;56:387. 1231 Neerhout RC. Erythrocyte stromal lipids in hyperlipemic states. J Lab Clin Med 1968;71:448. ~241 Hui DY, Harmony JAK. Interaction of plasma lipoproteins with erythrocytes. I. Alteration of erythrocyte morphology. Biochim Biophys Acta 1979;550:407. ~251 Atkinson JB, Stacpoole PW, Swift LL. Morphologic abnormalities of erythrocytes from patients with homozygous familial hypercholesterolaemia. Biochim Biophys Acta 1982;712:211. WI Muller S, Ziegler 0, Donner M, Drouin P, Stolz JF. Rheological properties and membrane fluidity of red blood cells and platelets in primary hyperlipoproteinemia. Atherosclerosis 1990;83:231. v71 LeCornu K. Audit of dietary input into a hospital lipid clinic. J Human Nutr Diet 1991;4:121. Lw Kris-Etherton PM, Krummel D, Dreon D, Mackey S, Wood P. The etrect of diet on plasma lipids, lipoproteins and coronary heart disease. J Am Diet Assoc 1988;88:1373. ~291 Mattson FH, Grundy SM. Comparison of effects of dietary saturated, monounsaturated and polyunsaturated fatty acids in plasma lipids and lipoproteins in man. J Lipid Res 1985;26:194. [301 Nestel PJ. Polyunsaturated fatty acids (n - 3, n-6). Am J Clin Nutr 1987;45:1161. [311 Kannel WB, Gordon T, Castelli WP. Obesity, lipids and glucose tolerance: the Framingham study. Am J Clin Nutr 1979;32:1238. 1321 Brennan PJ, Simpson JM, Blacket RB, McGilchran CA. Effects of body weight on serum cholesterol, serum triglycerids, serum urate and systolic blood pressure. J Med 1980;10:15. [331 Burr GO, Kass JP, Brown JB, Frankel J. On the fatty acids essential in nutrition. J Nutr 1938;15:15. [341 Holman RT. Essential fatty acid deficiency in humans. In: Galli C (ed). Dietary Lipids and Postnatal development. New York: Raven Press, 1973;27. [351 Eddy DE. Dietary fat influences on brain and liver fatty acid composition: importance of docosahexanoic acid (22:6n-3). PhD Thesis, University of Nebraska, 1973. 1361Brenner RR. The oxidative desaturation of unsaturated fatty acids in animals. Mol Cell Biochem 1974;3:41. [371 Livingstone MBE, Prentice AM, Strain JJ, Coward WA, Black AE, Barker ME, McKenna PG, Whitehead RG. Accuracy of weighed dietary records in studies of diet and health. Br Med J 1990;300:708. [381 Hyman MD, Instill W, Palmer RH, O’Brien J, Gordon L, Levine B. Assessing methods for measuring compliance with a fat controlled diet. Am J Pub1 Health 1982;72:152. [391 Berlin E, Bhathena SJ, Judd JT, Nair PB, Jones Y, Taylor PR. Dietary fat and hormonal effects on erythrocyte membrane fluidity and lipid composition in adult women. Metabolism 1989;38:790.