Can linoleic acid contribute to coronary artery disease?

Abstract

The adipose tissue concentration of linoleic acid was positively associated with the degree of coronary artery disease (CAD) in a cross-sectional study of 226 patients undergoing coronary angiography. Linoleic acid concentration in adipose tissue is known to reflect the intake of this fatty acid. These results are therefore indicative of a positive relationship between linoleic acid intake and CAD. The platelet linoleic acid concentration was also positively associated with CAD. After confounding factors were allowed for, the eicosapentaenoic acid concentration in platelets was inversely associated with CAD for men, and the docosapentaenoic acid concentration in platelets was inversely associated with CAD for women; results consistent with several other studies that suggest that fish, and omega-3 fatty acids derived from fish and fish oils, can beneficially influence macrovascular disease.

Full text

228
Am J C/in Nuir l993;58:228-34. Printed in USA. © 1993 American Society for Clinical Nutrition
Can linoleic acid contribute to coronary artery disease?13
Jonathan M Hodgson, Mark L Wahlqvist, John A Boxall, and Nicholas D Balazs
ABSTRACT The adipose tissue concentration of linoleic
acid was positively associated with the degree of coronary artery
disease (CAD) in a cross-sectional study of 226 patients under-
going coronary angiography. Linoleie acid concentration in adi-
pose tissue is known to reflect the intake ofthis fatty acid. These
results are therefore indicative ofa positive relationship between
linoleic acid intake and CAD. The platelet linoleic acid coneen-
tration was also positively associated with CAD. After eon-
founding factors were allowed for, the eieosapentaenoie acid
concentration in platelets was inversely associated with CAD
for men, and the docosapentaenoie acid concentration in plate-
lets was inversely associated with CAD for women: results eon-
sistent with several other studies that suggest that fish, and w-3
fatty acids derived from fish and fish oils, can beneficially influ-
enee macrovascular disease. Am J C/in Nutr l993;58:228-
34.
KEY WORDS Linoleic acid, w-3 fatty acids, polyunsatu-
rated fatty acids, coronary artery disease, atherosclerosis
Introduction
Many studies have examined the relationships between diet
and end points ofeoronary heart disease (CHD) such as angina,
myocardial infarction, sudden death, angiographically assessed
coronary artery disease (CAD), and coronary mortality. The
majority have focused on the lipid components of the diet. A
high intake of saturated fatty acids is now considered to be a
positive risk factor for CHD and an adequate intake ofw-3 fatty
acids is believed to be influential in preventing CHD. The role
of linoleic acid (1 8:2n-6), an w-6 essential fatty acid, however,
is less clear.
The measurement of linoleic acid in adipose tissue provides
a good estimate oflong-term intake oflinoleic acid (1, 2). Several
studies have demonstrated an inverse relationship between adi-
pose tissue linoleic acid content and CHD (3-6). Population
studies have shown that low concentrations of adipose tissue
linoleic acid are associated with increased rates of CHD (3, 4),
and that this inverse relationship exists both between and within
populations (4). A cross-sectional survey of Scottish men dem-
onstrated that those with previously unidentified CHD, as defined
by angina peetoris or myocardial infarction, had a significantly
lower concentration ofadipose tissue linoleic acid than did men
without CHD (5). The inverse relationship between adipose tissue
linoleic acid and angina pectoris or myocardial infarction has
also been demonstrated in a case-control study (6). These results
are supported by the observation that as the vegetable-animal
fat ratio increased in Australia and North America, there has
been an associated reduction in total mortality. In England and
Wales where minimal changes in this ratio occurred, there were
also minimal changes in coronary mortality (7). However, not
all studies have produced results that would indicate a protective
role for linoleic acid. In a study by Blankenhorn et al (8) it was
found that increased intake oflinoleic acid significantly increased
the risk ofnew atherosclerotic lesions in human coronary arteries.
Evidence for an inverse association between the long-chain
w-3 fatty acids or fish intake and CHD has been accumulating.
A reduction in total mortality was demonstrated in a secondary
prevention intervention study in which the intervention was fatty
fish (9). Prospective studies have found an inverse association
between fish intake and CHD incidence (10, 1 1), although an
inverse association has not been demonstrated in all prospective
studies (12, 13). Fish intake has also been associated with im-
proved arterial wall characteristics (14). In a study in which
platelet fatty acids were measured, eieosapentaenoie acid (20:
5n-3) was inversely associated with angina peetoris and doco-
sapentaenoie acid (22:5n-3) was inversely associated with risk
ofacute myocardial infarction. Also in this study, adipose tissue
docosahexaenoic acid (22:6n-3) was inversely associated with
acute myocardial infarction (6).
In our study the fatty acids in adipose tissue and platelets were
measured. The relationships between each ofthe fatty acids and
the degree of angiographically assessed CAD were examined.
Subjects and methods
Population sample
Some of the subject characteristics are given in Table 1 . All
patients underwent coronary angiography and were on the rou-
tine cardiac catheterization list for investigation of chest pain
thought to be due to either CAD (97%) or valvular heart disease
(3%). Consecutive patients (160 males and 66 females aged 16-
80 y) were enrolled over 10 mo. All patients were included. The
I From the Departments of Medicine, Cardiology, and Clinical
Biochemistry, Monash Medical Centre, Melbourne, Australia.
2 Supported in part by the Prince Henry’s Hospital Wholetime Medical
Specialists Private Practice Fund.
3 Address reprint requests to ML Wahlqvist, Professor and Chairman
of Medicine, Monash University Department of Medicine, Monash
Medical Centre, Clayton Road, Clayton, Melbourne, Victoria 3168,
Australia.
Received June 28, 1992.
Accepted for publication February 26, 1993.
by guest on July 13, 2011www.ajcn.orgDownloaded from

LINOLEIC ACID AND CAD
229
TABLE I
Patient characteristics
Characteristic
Men
(n = 160)
Women
(n = 66)
%
Obtained from medical records
Angina
98 95
Previous myocardial infarction 39
27
Valvular heart disease
2.0 8.2
Hypertension
41 63
Taking medication
Aspirin
67 49
Nitrates
58 44
Calcium antagonists 48 42
Beta-blockers
4 1 39
Angiotensin-converting-enzyme
inhibitors 16 29
Obtained from questionnaire
Smokers
14 11
Exsmokers 58
58
Never smoked 28 31
Diabetes mellitus 8.7 9.1
No dietary change 49 44
Dietary change
S I 56
project was presented to and approved by the Monash Medical
Centre Ethics and Research Committee.
Clinical details such as age, diabetic status, smoking history,
height, weight, and past dietary change were gathered by using
a questionnaire administered at the time of angiography. Self-
reported height and weight were used to calculate body mass
index (BMI), which was calculated by dividing weight (kg) by
the square of height (m). Information about dietary change was
obtained with the questionnaire. Each patient was asked whether
they had made a conscious decision to make dietary changes at
any time in the previous 10 y. For analysis the patients were
divided into two groups: no dietary change and dietary change.
To estimate the total number ofeigarettes smoked, the number
of cigarettes smoked per day was multiplied by the duration of
time that the patient smoked. The measurement of blood pres-
sure was not included in this study for several reasons. Almost
all hypertensive patients were being treated for hypertension.
However, the anti-hypertensive drug treatment differed and so
too did the duration oftreatment and the degree ofblood pressure
lowering. This would make the interpretation of results from a
single blood pressure measurement difficult. Information about
history of hypertension and use of particular medication was
collected from the medical records. The presence or absence of
hypertension and use of particular medications were included
in the analysis. Coronary angiography was performed according
to the Judkins technique (1 5) and recorded on 35-mm movie
film. Two different scoring systems were used to quantify the
degree of CAD: a myocardial score and an extent score.
Angiographic scores
Myocardia/ score. A myocardial scoring system that takes into
account the degree ofstenosis ofany number ofarterial branches
and their relative importance in terms of the amount of myo-
eardium supplied, has been developed (16). This scoring system
takes into account the severity as well as the location of the
coronary lesions. A score from 0 to 15 (best to worst condition)
can be given.
E.vtent score. An angiographic scoring system that has been
designed to reflect the proportion of the coronary endothelial
surface area affected by atheroma has been developed as an es-
timate of the extent of coronary atherosclerosis ( 1 7). The pro-
portion of the coronary arterial tree with detectable atheroma,
identified as luminal irregularity, was scored with a maximum
of 10: a score of 0 indicates that no coronary atheroma was
detected, and a score of 10 means that 100% of the coronary
arteries visualized showed detectable atheroma.
Serum lipid measurements
Fasting blood was drawn from the femoral artery immediately
before cardiac catheterization and placed into evacuated glass
tubes. The untreated blood was allowed to clot and the serum
was separated by using standard procedures. Total cholesterol,
triglycerides, and high-density-lipoprotein (HDL) cholesterol
were measured in fresh serum.
Total cholesterol and triglycerides were measured enzymati-
cally with commercial kits ( 13225 and 22203, respectively; Trace
Scientific Pty Ltd. Clayton, Victoria, Australia). HDL cholesterol
was measured enzymatieally as for total cholesterol, after the
precipitation of apolipoprotein B-containing lipoproteins by
using equal volumes of 20% polyethylene glycol 6000 (Merck-
Schuehardt, Munich, Germany) and serum. Low-density-lipo-
protein (LDL) cholesterol was derived by using the Friedewald
formula adapted to Syst#{232}meInternational (SI) units (1 8). Cho-
lesterol and triglyceride measurements were performed on a
KONE Progress selective chemistry analyzer (KONE Instru-
ments Corporation, Espoo, Finland).
Fatti’ acid anal vsis
The fatty acid compositions of subcutaneous adipose tissue
and platelets were measured by using gas chromatography. The
methods relating to these measurements are presented below.
For collection and preparation of adipose tissue,  1-2 mg
adipose tissue was taken from the site of catheter insertion im-
mediately after coronary angiography. The sample was frozen
at -70 #{176}Cfor up to 12 mo, until all samples had been collected.
For collection and preparation of platelets. 20 mL EDTA-
anticoagulated blood was drawn from the femoral artery im-
mediately before cardiac catheterization and was used for platelet
harvesting. The tubes were mixed immediately after blood col-
leetion and again before platelet harvesting. Tubes ofblood were
centrifuged at 1 10 X g for 1 5 mm and the platelet-rich plasma
was removed then recentrifuged at 2000 X g for 10 mm at room
temperature. The plasma was removed and the platelets were
washed twice with 0.9% NaCl containing 1 g EDTA/L. The
platelet pellet was then frozen at -70 #{176}Cuntil extraction and
methylation.
Extraction and methylation of the adipose tissue and platelet
fatty acids were performed by using a modification of the one-
step method described by Lepage and Roy (19). Samples were
placed into glass tubes with 2 mL of 4: 1 methanol:toluene and
200 L aeetyl chloride was added slowly to each tube with con-
tinuous mixing, then samples were placed into an oven at 100
#{176}Cfor I h. The tubes were cooled under running water, 2.5 mL
potassium carbonate was added, and the tubes were mixed and
then centrifuged at 2000 X g for 10 mm at room temperature.
by guest on July 13, 2011www.ajcn.orgDownloaded from

230 HODOSON ET AL
TABLE 3
S   SD: n in brackets.
5 P < 0.001.
t P < 0.0001.
tP<0.05.
TABLE 2
Descriptive statistics for the study population5
Characteristic
Men Women
Extent score (10)1 4.3 ± 1.9 [l59fl 3.4 ± 2.3 [66]
Myocardial score (lS) 7.7 ± 3.6 [159]t
6.4 ± 4.25 [66]
Age (y) 59 ± 10.7 [159] 61 ± 10.3 [66]
Body mass indexil
25.7 ± 2.59 [132] 25.2 ± 4.30 [57]
Total cholesterol (mmol/L) 5.8 ± 0.93 [159] 6.4 ± 1.1 [66]
LDL cholesterol (mmol/L) 4.0 ± 0.86 [155] 4.4 ± 1.0 [66]
HDL cholesterol (mmol/L) 0.98 ± 0.28 [155] 1.25 ± 0.42 [66]
Triglycerides (mmol/L) I .8 ± 0.96 [1 59] 1.7 ± 0.77 [66]
Cigarettes smoked #{182}
741 ± 760 [ 129] 344 ± 606 [57]
S j;  SD: n in brackets.
t Scoring system for extent ofcoronary atherosclerosis, from 0 to 10.
 Significantly greater than women P < 0.0001,
§ Scoring system for degree of arterial stenosis, from 0 to 15.
II In kg/m2.
#{182}Estimate of the total number of cigarettes smoked over a lifetime.
The upper toluene phase was removed and placed into small
glass tubes, then dried under nitrogen.
Immediately after extraction and methylation ofthe fauy acids
the methylated fatty acids were dissolved in 50 L chloroform
for injection. A Shimadzu GC-9A gas chromatograph was used
with a flame ionization detector and a Shimadzu Chromatopac
C-R3A integrator (Shimadzu Corporation, Kyoto, Japan). A 50-
m glass capillary column with an 80% cyanopropyl silicone polar
phase was used. The methyl ester peaks were identified with
standards obtained from Nu Chek Prep, Inc (Elysian, MN).
Temperature programming was used for determining fatty acids.
The starting temperature of 130 #{176}Cwas raised at 5 #{176}C/minuntil
190 #{176}Cwas reached, then increased more slowly at 1.2 #{176}C/min
until 200 #{176}Cwas reached. The temperature was then held eon-
stant at 200 #{176}Cfor 15 mm. The CV in determining the per-
centages of the individual fatty acids varied depending on the
concentration of the fatty acid in the sample. For major com-
ponents (> 5%) the CV ranged from 1 .2% to 4.7%. For trace
fatty acids (< 1%) the CV ranged from 7.4% to 14%.
Statistics
The data-analysis package used for all the statistical analyses
performed was SAS (20,2 1). At a univariate level, Spearman’s
rank correlation coefficient (r) was used to determine the degree
and direction of association between two variables. To control
for covariates, the PARTIAL option was used. The Wileoxon
rank-sum test was performed to test whether there were differ-
enees between two population means.
Results
Descriptive statistics
Some descriptive statistics for the study population are pre-
sented in Table 2. The extent score ranged from 0 to 8.5 and
was significantly higher for men than for women. The myocardial
score ranged from 0 to 15 and was also significantly higher for
men than for women. Nineteen patients (8.4%) had no detectable
CAD, with both extent and myocardial scores of 0. The degree
of correlation between the two scores was high (r = 0.72, P
Fatty acid composition of adi pose tissue and platelets5
Fatty acid Men Women
% of total/any acids
Adipose tissue fatty acids
Laurie (12:0)
Myristic ( 14:0)
Palmitic (16:0)
Stearic (18:0)
Palmitoleic (16:ln-7)
Oleic (18:ln-9)
Linoleic (18:2n-6)
Homo-y-linolenic (20:3n-6)
Arachidonic (20:4n-6)
Alpha-linolenic (l8:3n-3)
Eicosapentaenoic (20:5n-3)
Docosapentaenoic (22:5n-3)
Docosahexaenoic (22:6n-3)
Platelet fatty acids
Palmitic (16:0)
Stearic (18:0)
Palmitoleic (l6:ln-7)
Oleic (18:ln-9)
Linoleic (l8:2n-6)
Homo--y-Iinolenic (20:3n-6)
Arachidonic (20:4n-6)
Docosatetraenoic (22:4n-6)
Alpha-linolenic (18:3n-3)
Eicosapentaenoic (20:5n-3)
Docosapentaenoic (22:5n-3)
Docosahexaenoic (22:6n-3)
0.43 ± 0.28 11061 0.38 ± 0.17 [39]
2.95 ± 0.80 [ 106] 2.67 ± 0.67 [39]
23.1 ± .96 [106] 22.4 ± 2.47 [39]
4.47 ± 1.29 [104] 4.27 ± 1.39 [39]
5.87 ± 2.37 [106] 5.72 ± 2.25 [39]
42.6 ± 3.1 1 [104] 42.5 ± 2.62 [391
12.6 ± 3.82 [106] 13.7 ± 3.83 [39]
0.17 ± 0.05 [105] 0.25 ± 0.1 1 [39]
0.32 ± 0.1 1 1105] 0.41 ± 0.12 [39]
0.56 ± 0.17 [106] 0.61 ± 0.23 [39]
0.10 ± 0.05 [68] 0.1 1 ± 0.04 [26]
0.15 ± 0.06 [104] 0.20 ± 0.08 [39]
0.1 1 ± 0.06 [101] 0.18 ± 0.10 [39]
17.9 ± 1.23 [1351 18.2 ± 1.52 [55]
17.9 ± I.! I [135] 7.7 ± 1.18 [55]
1.58 ± 0.37 [134] l.6l ± 0.36 [55]
15.7 ± .41 [135] 15.8 ± 1.22 [55]
5.46 ± 1.56 [135] 5.42 ± 1.07 [55]
0.98 ± 0.23 [134] 1.01 ± 0.21 [55]
16.9 ± 1.50 1135] 16.7 ± 1.14 [55]
1.49 ± 0.36 [135] 1.51 ± 0.3l [55]
0.39 ± 0.16 [129] 0.37 ± 0.13 [49]
2.14 ± 0.31 [135] 1.04 ± 0.35 [55]
0.86 ± 0.20 [134] 0.80 ± 0.25 [55]
2.13 ± 0.41 [1351 2.08 ± 0.33 [55]
< 0.000 1). The mean concentration. as a percentage of the total
fatty acids in adipose tissue and platelets are presented in Ta-
ble 3.
Risk factors. medication, andjmi’ acids
The correlations between several CAD risk factors and the
two CAD scores are shown in Table 4. Age showed the strongest
association with the two CAD scores. Cigarette smoking was
also positively associated with CAD. Associations between the
established serum lipid risk factors and the CAD scores were
TABLE 4
Spearman’s rank correlation coefficients describing the associations
between coronary artery disease (CAD) risk factors and CAD scores
Risk factors
Exten
Men
t score
Women
Myocard
Men
ial score
Women
Age 0.29w 0.39 0.42t
0.4l
Cigarette smoking 0.09 0.29 0.2 1
0.18
Body mass index -0. 16 0.09 -0.0 1
0.08
Totalcholesterol 0.03 0.13 0.05 0.10
LDLcholesterol 0.10 0.15 0.11 0.13
HDL cholesterol -0.05
-0. 14 -0.05 -0.24
Triglycerides -0.03 0.19 0.03 0.27
by guest on July 13, 2011www.ajcn.orgDownloaded from

LINOLEIC ACID AND CAD 231
generally weak and rarely significant, particularly for men. For platelet linoleic acid (P < 0.01) and a-linoleie acid (P < 0.05)
men, none ofthe associations was significant. For women, I-IDL for women. Nitrates were associated with lower adipose tissue
cholesterol was inversely associated with the myocardial score laurie acid (P < 0.05) for men and lower platelet stearie acid (P
and triglycerides were positively associated with the myocardial < 0.05) for women. Beta-blockers were associated with higher
score. For the total population, HDL cholesterol was signifi- adipose tissue oleic acid for women (P < 0.05). Calcium antag-
cantly inversely associated with both the extent score (r = -0. 14, onists were associated with higher concentrations ofadipose tis-
P < 0.05) and the myocardial score (r = -0. 16, P < 0.05). sue eieosapentaenioe acid (P < 0.05) and platelet docosapen-
Both men and women classed as hypertensive had a signifi- taenoie acid (P < 0.05) for women. ACE inhibitors were asso-
cantly higher extent score (P < 0.05) and myocardial score ciated with higher adipose tissue docosapentaenoie acid for men
(P < 0.05) than those without hypertension as a risk factor (P < 0.01).
for CAD. The relationships between particular fatty acids in adipose
The correlations between several of the CAD risk factors tissue with the same fatty acid measured in platelets are presented
and fatty acids in adipose tissue and platelets are presented in Table 6. In general, the associations between the same fatty
in Table 5. The other risk factor that was related to several acids in the different tissues were not strong. The fatty acid with
of the fatty acid measurements was hypertensive status. Men the highest concordance between adipose tissue and platelets
with hypertension had significantly higher adipose tissue was linoleic acid.
linoleie acid (P < 0.05) concentrations than men without
hypertension. Women with hypertension had significantly Adipose tissuejatty acids and CAD
lower platelet linoleic acid than women without hypertension The relationships between adipose tissue fatty acids and the
(P < 0.05). Dietary change by men was associated with a sig- two CAD scores are presented in Table 7.
nificantly higher concentration ofadipose tissue linoleic acid Men. Myristie acid was inversely associated with the myo-
(P < 0.05). Diabetic status was not associated with any ofthe cardial score. This association is not significant after age was
fatty acids. adjusted for (r = -0. 16, P > 0.05). Palmitoleic acid was inversely
Treatment ofthe patients with aspirin, nitrates, beta-blockers, associated with the extent score. After age, triglycerides, and
calcium antagonists, or angiotensin-converting-enzyme (ACE) cigarette smoking were adjusted for. the association remained
inhibitors were also related to concentrations of several of the significant (r = -0.24, P < 0.05).
fatty acids. Aspirin was associated with significantly higher Linoleic acid was positively associated with both the extent
platelet eieosapentaenoie acid for men (P < 0.05) and higher score and the myocardial score. After age was adjusted for, the
TABLE 5
Spearman’s rank correlation coefficients describing the associations between fatty acids in adipose tissue and platelets and risk factors
for coronary artery disease
Men Women
Fatty acid Age Smoking BMI LDL-C HDL-C TG Age Smoking BMI LDL-C HDL-C TG
Adipose tissue fatty acids
Lauric(l2:0) -0.15 -0.04 -0.17 0.03 0.09 -0.04 -0.17 -0.03 -0.30 -0.16 0.15 -0.18
Myristic (14:0) -0.13 -0.02 -0.17 0.02 -0.02 0.1 1 -0.21 0.00 _0.38* -0.22 0.05 0.11
Palmitic (16:0) -0.08 0.01 -0.04 0.02 -0.10 0.3lt -0.1 1 0.l8 0.04 -0.12 0.07 0.25
Stearic (18:0) 0.07 -0.2l -0.23 -0.20 0.05 -0.13 0.1 1 0.09 -0.50t -0.14 0.27 -0.17
Palmitoleic (l6:ln-7) -0.21 0.24 0.06 0.00 -0.01 0.l9 -0.14 -0.09 0.15 0.08 -0.05 0.08
Oleic (18:ln-9) -0.05 0.06 0.07 0.05 -0.01 0.01 -0. 10 0.03 0.06 -0.05 -0.25 0.02
Linoleic (l8:2n-6) 0.18 -0.09 0.02 0.07 0.04 -0.25t 0.29 -0.01 0.12 0.18 0.02 -0.09
Homo--y-linolenic (20:3n-6) 0.13 0.09 0.32t 0.07 -0.07 -0.02 0.32 0.06 0.45 0.03 0.03 0.28
Arachidonic (20:4n-6) -0.06 0.1 I 0.19 -0.12 0.02 0.08 0.07 -0.06 0.52t 0.07 0.06 0.17
Alpha-linolenic (18:3n-3) -0.03 -0. 16 0.05 0.05 0.07 -0.08 0. 10 -0.04 -0. 12 -0.01 0.27 -0.21
Eicosapentaenoic (20:5n-3) -0.1 1 0.15 -0.02 -0.14 0.25* -0.19 -0.21 -0.19 -0.34 -0.04 0.19 -0.54t
Docosapentaenoic (22:5n-3) 0.16 -0.02 0.04 -0.20 -0.00 -0.00 0.42t -0.19 0.22 -0.04 -0.00 0.28
Docosahexaenoic (22:6n-3) -0.03 -0.02 0.12 -0.15 0.09 -0.07 0.34 -0.05 0.17 -0.1 1 0.26 -0.01
Platelet fatty acids
Palmitic (16:0) -0.01 -0.10 -0.00 -0.14 0.07 0.1 1 0.44t 0.04 0.01 0.23 0.09 0.23
Stearic (18:0) -0.07 -0.13 -0.l8 0.08 0.13 _0.19* 0.01 -0.06 -0.44t -0.01 -0.10 -0.02
Palmitoleic (16:ln-7) 0.22 0.07 -0.03 -0.09 0.20 -0.09 0.1 I 0.14 -0.20 -0.05 -0.05 0.06
Oleic(l8:ln-9) -0.11 0.08 0.03 -0.14 -0.15 0.23t -0.04 -0.12 0.27 -0.06 0.10 0.13
Linoleic (18:2n-6) 0.1 1 -0.07 0.01 -0.09 -0.14 0.1 1 0.09 -0.1 1 -0.06 -0.21 0.09 -0.14
Homo--y-linolenic (20:3n-6) -0.06 0.07 0.12 0.04 -0.19 0.l8 #{216}3#{216}5 0.04 0.02 0.14 0.05 0.20
Arachidonic (20:4n-6) -0.20 -0.04 0.14 0.16 0.03 -0.03 -0.25 -0.04 0.10 0.12 0.05 0.12
Docosatetraenoic (22:4n-6) 0.01 0.04 0.09 0.04 0.05 -0.06 -0.36t -0.06 0.14 -0.24 -0.09 -0.12
Alpha-linolenic (l8:3n-3) 0.07 -0.07 -0.08 0.06 -0.06 -0.01 -0.04 -0. 12 0.20 -0.06 -0.10 0.07
Eicosapentaenoic (20:5n-3) -0.14 -0.02 0.14 0.l9 -0.07 0.07 -0.15 0.19 0.01 -0.12 0.04 -0.10
Docosapentaenoic (22:5n-3) -0.06 0.04 -0.04 0.17 0.02 -0.09 -0.17 0.00 -0.14 -0.15 0.26 -0.30
Docosahexaenoic (22:6n-3) -0.03 0.1 1 -0.05 0.08 0.04 -0.08 -0.18 0.00 0.13 0.09 -0.12 0.07
U p < o.os.
t P < 0.01.
by guest on July 13, 2011www.ajcn.orgDownloaded from

TABLE 7
*P<0.05.
t P < 0.01.
5P<0.0l.
tP<0.05.
232 HODGSON FT AL
(even partially) this inverse association.
TABLE 6
Spearman’s rank correlation coefficients describing the associations
between the adipose tissue fatty acids and the same fatty acids
measured in platelets
Fatty acid Men Women
Palmitic ( 16:0)
0.07 -0.08
Stearic(18:0) 0.3l 0.24
Palmitoleic ( 16: 1n-7) 0.22f -0.01
Oleic ( 18: 1n-9) 0. 16 0.27
Linoleic ( I 8:2n-6) 0.46t 0.49
Homo--y-linolenic (20:3n-6)
0. 18 0455
Arachidonic (20:4n-6) -0.05 -0.22
Alpha-linolenic ( 18:3n-3)
0.06 0.23
Eicosapentaenoic (20:5n-3) -0.33t 0.21
Docosapentaenoic (22:Sn-3) -0. 1 1 0.15
Docosahexaenoic (22:6n-3) -0.03 0.33t
5P<0.0l.
tP<0.0S.
 P < 0.0001.
§ P < 0.001.
association between linoleic acid and the extent score was not
significant (r = 0. 14, P > 0.05). The association with the myo-
eardial score remained (r = 0. 19, P < 0.05). After all covariates
that were related to adipose tissue linoleic acid were adjusted
for, namely age, triglycerides, hypertensive status, and dietary
change, linoleic acid was not significantly associated with the
extent score (r = 0. 1 3, P > 0.05), but the association with the
myocardial score remained (r = 0.20, P < 0.05). Adipose tissue
homo--y-linolenic acid was also positively associated with both
CAD scores. After age and BMI were adjusted for, homo--y-
linolenie acid was significantly associated with the extent score
(r = 0.27, P < 0.05) but not with the myocardial score (r = 0.09,
P > 0.05).
Women. Docosahexaenoic acid was positively associated with
both CAD scores. However, after age and BMI were adjusted
for, the associations between both the extent score (r = 0. 13, P
> 0.05) and the myocardial score (r - 0. 14, P > 0.05) were not
significant.
Platelet fatty acids and CAD
The relationships between the platelet fatty acids and the two
CAD scores are presented in Table 8.
Men. A positive association between linoleic acid and both
the CAD scores was found. These associations remained after
age was adjusted for [r = 0.25, P < 0.01 (extent score), r = 0.24,
P < 0.01 (myocardial score)]. Araehidonic acid was inversely
associated with both CAD scores. After age was adjusted for,
the associations between arachidonic acid and the CAD scores
were not significant [r = -0. 14, P > 0.05 (extent score), r
= -0. 12, P > 0.05 (myocardial score)]. Platelet arachidonie acid
was also inversely associated with platelet linoleic acid concen-
tration (r = -0.34, P < 0.0001). To determine whether the in-
verse association between linoleic acid and arachidonic acid for
men could be explained by age, cigarette smoking, diabetic status,
dietary change, plasma cholesterol concentration, or any other
serum lipid measurement, the association for each ofthese factors
was adjusted for. None of these variables were able to explain
Spearman’s rank correlation coefficients describing the associations
between adipose tissue fatty acids and coronary artery disease scores
Extent score Myocardial score
Fatty acid Men Women Men Women
Laurie (12:0) -0.09 0.01 -0. 18 -0.01
Myristic (14:0) -0.12 -0.15 -0.20 -0.25
Palmitic (16:0) -0.04 -0.12 -0.09 -0.16
Stearic ( 18:0) 0.00 0.08 -0.09 -0.0 I
Palmitoleic ( 16: I n-7) -0.2 1 -0. 1 8 -0. 10 -0.08
Oleic (18:ln-9) -0.15 -0.03 -0.18 0.05
Linoleic (18:2n-6) O.20 0.24 0.25t 0.21
Homo--y-linolenic (20:3n-6) 0.24 0.04 0.21 0.08
Arachidonic (20:4n-6) -0.03 -0. I I 0. I 2 -0.03
Alpha-linolenic (l8:3n-3) 0.08 0.14 0.05 0.04
Eicosapentaenoic (20:5n-3) 0.22 -0.15 0.10 -0.24
Docosapentaenoic (22:Sn-3) 0.10 0.1 1 -0.00 0.19
Docosahexaenoic (22:6n-3) 0.02 0.33 -0.01 0335
The long-chain w-3 fatty acids were not significantly inversely
associated with either of the CAD scores. However, after LDL
cholesterol concentration and aspirin use were adjusted for, both
of which were significantly related to platelet eicosapentaenoic
acid as well as age, eieosapentaenoic acid was inversely associated
with the myocardial score (r = -0.18, P < 0.05).
Women. Arachidonic acid was inversely associated with the
myocardial score. After age was adjusted for, this association
was not significant (r = -0.22, P > 0.05). The associations be-
tween the long-chain fatty acids and the CAD scores were not
significant. After calcium antagonist use was adjusted for, which
was related to the concentration of docosapentaenoic acid, do-
cosapentaenoic acid was significantly inversely associated with
both the extent score (r = -0.28, P < 0.05) and the myocardial
score (r = -0.37, P < 0.01). After age was further adjusted for,
the association with the extent score was not significant (r
= -0.24. P > 0.05) but remained significant for the myocardial
score (r = -0.33, P < 0.05).
TABLE 8
Spearman’s rank correlation coefficients describing the associations
between platelet fatty acids and coronary artery disease scores
Fatty acid
Extent
Men
score
Women
Myocard
Men
ial score
Women
Palmitic (16:0) -0.06 0.23
-0.04 0.19
Stearic ( 18:0) -0.03 0.05 -0. 16 -0. 1 1
Palmitoleic ( I 6: I n-7) -0.02 0.06 0.04 0. 1 I
Oleic(18:ln-9) -0.03
0.01 -0.05 0.14
Linoleic (18:2n-6) 0.27
0.00 0.26 -0.08
Homo-y-linolenic (20:3n-6) 0.09
0.02 -0.01 0.10
Arachidonic (20:4n-6) -0.19t -0.24
-0.19t -0.28t
Docosatetraenoic (22:4n-6) -0.08 -0.25 0.04 -0.20
Alpha-linolenic (18:3n-3) 0.17 0.01 0.12
-0.05
Eicosapentaenoic (20:5n-3) -0.07 -0.01 -0. 12 -0.10
Docosapentaenoic (22:Sn-3) -0.16 -0.19 -0.09 -0.24
Docosahexaenoic (22:6n-3) -0.06 -0.12 -0.05
-0.05
by guest on July 13, 2011www.ajcn.orgDownloaded from

LINOLEIC ACID AND CAD 233
Relations/up outcomes
A summary of the major relationship outcomes observed for
adipose tissue and platelet fatty acids after confounding factors
were adjusted for, is presented in Table 9.
Discussion
The myocardial score has been used routinely as a means for
quantitatively presenting the information obtained in a coronary
angiogram. This score takes into account both severity and lo-
cation of coronary lesions. Because severity is measured. the
score may relate to factors that complicate atherosclerotic lesions
in addition to the atherosclerosis itself. It has one major disad-
vantage if the relationships with risk factors for CAD are being
examined. The score is biased by lesion location. The extent
score is an estimate of the aggregate degree of atherosclerosis in
the coronary arteries. This score is not biased by lesion location.
it is quantitative, and it has been shown to be correlated more
strongly with CAD risk factors than with a vessel score or a
stenosis score ( 1 7). The extent score and the myocardial score
were significantly associated, but the degree of association sug-
gests that different aspects ofCAD are being assessed. However,
in general, those patients with more of the myocardium threat-
ened by CAD also had a higher percentage of their coronary
arteries affected by atherosclerosis.
The results for men from both adipose tissue and platelets
show a positive association between linoleic acid and CAD. Adi-
pose tissue is a good indicator of long-term intake of linoleic
acid (1, 2), whereas platelet fatty acids reflect shorter-term intake.
A correlation of 0.46 (Table 6) between adipose tissue linoleic
acid and platelet linoleic acid for men is consistent with the
different time frame ofthe measurements. However, factors other
than diet can also influence the concentration of linoleic acid
in both adipose tissue and platelets. Those factors that affect
platelet linoleie acid might not affect adipose tissue linoleic acid.
Although the degree ofassociation between linoleic acid in adi-
pose tissue or platelets and the CAD scores for men was not
high, it was stronger than many of the other recognized risk
factors for CAD. Together the results from platelets, and adipose
tissue in particular. are indicative of a positive association be-
tween linoleic acid intake and CAD.
Much of the evidence now available suggests an inverse re-
lationship between linoleic acid and CHD (3-6). However. one
study found that an increased intake oflinoleic acid significantly
increased the risk of new atherosclerotic lesions in human cor-
onary arteries (8). Interestingly, the end point in this study was
also angiographic. Studies finding an inverse relationship for
adipose tissue Iinoleic acid have used either rates of mortality
from CHD (3, 4), angina pectoris, or acute myocardial infarction
(5, 6) as an end point.
Ifthe positive associations observed between linoleic acid and
CAD in this study are due to a positive association between
linoleic acid intake and CAD, as suggested by our results, then
one possible explanation for the inconsistent findings may be
that high linoleic acid intake is a risk factor for coronary ath-
erosclerosis, just as low linoleic acid intake would appear to be
a risk factor for coronary events (3-6). The mean concentration
oflinoleic acid in the adipose tissue ofthe study population was
12.6% for men and I 3.7% for women, which is considerably
higher than the mean concentration oflinoleic acid in the adipose
tissue of European populations studied previously (3-6).
TABLE 9
Major relationship outcomes after confounding factors
were adjusted fore
Men
I) Adipose tissue palmitoleic acid (l6:ln-7) was inversely
associated with the extent score (r = -0.24, P < 0.05).
2) Adipose tissue linoleic acid (18:2n-6) was positively associated
with the myocardial score (r = 0.20. P < 0.05).
3) Platelet linoleic acid ( I 8:2n-6) was positively associated with
both the extent score (r = 0.25. P < 0.01) and the myocardial
score(r = 0.24. P < 0.01).
4) Adipose tissue homo--y-linolenic acid (20:3n-6) was positively
associated with the extent score (r = 0.27, P < 0.05).
5) Platelet eicosapentaenoic acid (20:Sn-3) was inversely
associated with the myocardial score (r = -0. 18. P < 0.05).
Women
I) Platelet docosapentaenoic acid (22:5n-3) was inversely
associated with the myocardial score (r = -0.33. P < 0.05).
S Possible confounding factors were age and other variables associated
with the particular fatty acids. r is Spearman’s rank correlation coefficient.
Several mechanisms that might account for the positive as-
sociations between linoleic acid and the CAD scores can be pro-
posed. Higher concentrations of linoleic acid may increase the
risk of oxidative modification of lipoproteins. The oxidative
modification ofLDL in particular is understood to be a pathway
to atherosclerosis (22). Another possibility is that linoleic acid
was a marker for the intake ofother factors in food that increase
the risk of CAD. A further alternative is that the observed as-
sociations for linoleic acid do not relate to intake, but to factors
that influence the metabolism of fatty acids in vivo. The results
from this study. however, provide insufficient evidence to support
any proposed mechanisms.
Adipose tissue palmitoleic acid was inversely associated with
the extent score. This is indicative ofan inverse association with
intake. However, the relationship between intake of saturated
or monounsaturated fatty acids and adipose tissue concentration
is not as strong as that for linoleic acid. Associations between
saturated or monounsaturated fatty acids measured in adipose
tissue and CAD therefore do not provide strong evidence for a
relationship with intake. Adipose tissue homo--y-linolenic acid
was positively associated with the extent score. This association
may be related to intake or to the metabolism of linoleic acid
in vivo. Several other significant associations between adipose
tissue fatty acids and the CAD scores were observed. All of these
associations could be explained by age and other covariates.
At a univariate level, not one ofthe long-chain w-3 fatty acids
was significantly inversely associated with the CAD scores. After
confounding factors were adjusted for, eicosapentaenoie acid
was inversely associated with the myocardial score for men and
docosapentaenoie acid was inversely associated with the myo-
cardial score for women. These results are supported by findings
from Wood et al (6), who found that platelet docosapentaenoic
acid was inversely associated with acute myocardial infarction,
and eieosapentaenoic acid was inversely associated with angina
peetoris (6). The results are also consistent with several other
studies that suggest that fish intake may be beneficial for CHD
risk (9-I I ). One of the most important mechanisms for the in-
verse relationship between the fish oils, or the .a-3 fatty acids,
and CHD is likely to be the effects ofw-3 fatty acids on eicosanoid
metabolism such that a more vasodilatory state with reduced
by guest on July 13, 2011www.ajcn.orgDownloaded from

234
HODGSON El AL
platelet aggregation results (23, 24). However, w-3 fatty acids
have been shown to reduce plasma triglyceride concentration
(25-27) and blood pressure (28. 29), both of which may reduce
the rate of progression of coronary atherosclerosis. The inverse
associations between platelet eieosapentaenoie acid for men and
docosapentaenoie acid for women, and the myocardial score
provide further evidence for a relationship between the intake
of this fatty acid and reduced risk of CAD. This is consistent
with the proposition that an adequate intake ofthese fatty acids
is beneficial to CAD. 13
References
I . Dayton S. Hashimoto S. Dixon W, Pearce ML. Composition of
lipids in human serum and adipose tissue during prolonged feeding
ofa diet high in unsaturated fat. I Lipid Res 1966:7:103-I I.
2. van Staveren WA, Deurenberg P. Katan MB, Berema I, de Groot
LC, Hoffmans MD. Validity of the fatty acid composition of sub-
cutaneous fat tissue microbiopsies as an estimate of the long term
average fatty acids composition of the diet in separate individuals.
Am I Epidemiol 1986:123:455-63.
3. Logan RL, Thomson M, Riemersma RA, et al. Risk factors for
ischaemic heart disease in normal men aged 40. Lancet 1978:1:949-
55.
4. Riemersma RA, Wood DA, Butler S. et al. Linoleic acid in adipose
tissue and coronary heart disease. Br Med I 1986:292:1423-7.
5. Wood DA, Butler S. Riemersma RA, Thomson M, Oliver MF. Adi-
pose tissue and platelet fatty acids and coronary heart diseases in
Scottish men. Lancet 1984:2:117-21.
6. Wood DA, Riemersma RA, Butler S. et al. Linoleic and eicosapen-
taenoie acids in adipose tissue and platelets and risk of coronary
heartdisease. Lancet 1987:1:177-83.
7. Dwyer T, Hetzel B. A comparison oftrends ofcoronary heart disease
mortality in Australia, USA, and England and Wales with reference
to three major risk factors-hypertension. cigarette smoking, and
diet. Int I Epidemiol 1980:9:6-71.
8. Blankenhorn DH, Johnston RL, Mack WI, Hafez A, El Zein MD,
Vailas LI. The influence of diet on the appearance of new lesions
in human coronary arteries. JAMA 1990:263:1646-52.
9. Burr ML, Fehily AM, Gilbert JF, et al. Effects ofehanges in fat, fish
and fibre intakes on death and myocardial reinfarction: Diet and
reinfarction trial. Lancet 1989:2:757-61.
10. Kromhout D. Bosschieter EB, deLezenne Coulander C. The inverse
relation between fish consumption and 20-year mortality from cor-
onary heart disease. N Engl I Med 1985:312:1205-9.
1 1. Shekelle RB, Missell L, Paul 0, Shryock AM, Stamler I. Fish con-
sumption and mortality from coronary heart disease. N EngI I Med
1985:3 13:820(letter).
12. Curb ID. Reed DM. Fish consumption and mortality from coronary
heart disease. N Engl I Med 1985:313:82l(letter).
13. Vollset SE. Heuch I. Bjelke E. Fish consumption and mortality from
coronary heart disease. N Engl I Med 1985:3 13:820- l(letter).
14. Wahlqvist ML, Lo CS, Myers KA. Fish intake and arterial wall
characteristics in healthy people and diabetic patients. Laneet 1989:2:
944-6.
15. Iudkins MP. Selective coronary arteriography: a percutaneous
transfemoral technique. Radiology 1967:69:815.
16. Brandt PWT, Partridge lB. Wattie WI. Coronary angiography:
method of presentation of the arteriogram report and a scoring sys-
tern. Clin Radiol 1977:28:361-5.
17. Sullivan DR. Marwick TH, Freedman SB. A new method of scoring
coronary angiograms to reflect extent of coronary atherosclerosis
and improve correlation with major risk factors. Am Heart I
l990;l 19:1262-7.
18. Friedwald WT, Levy RI, Fredrickson DS. Estimation of the con-
centration of low-density lipoprotein cholesterol in plasma without
the use ofthe preparative ultracentrifuge. Clin Chem 1972:18:499-
502.
19. Lepage G, Roy CC. Direct transesterification of all classes of lipids
in a one-step reaction. I Lipid Res 1986:27:1 14-20.
20. SAS Institute Inc. SAS introductory guide for personal computers:
statistics. version 6 ed. Cary, NC: SAS Institute mc, 1985.
21 . SAS Institute Inc. SAS users guide: statistics. version 5 ed. Cary,
NC: SAS Institute Inc. 1985.
22. Steinberg D, Parthasarathy S. Carew TE. Khoo IC. Witztum JL.
Beyond cholesterol: modifications of low density lipoprotein that
increase its atherogenicity. N Engl I Med 1989:320:915-24.
23. Cannon P1. Eicosanoids and the blood vessel wall. Circulation
1984;70:523-8.
24. Moncada S. Vane JR. Arachidonic acid metabolites and the inter-
actions between platelets and blood vessel walls. N EngI I Med
1979:300:1142-7.
25. Harris WS, Connor WE, McMurray MP. The comparative reductions
of the plasma lipids and lipoproteins by dietary polyunsaturated
fats: salmon oil versus vegetable oils. Metabolism 1983:32:179-84.
26. Illingworth DR. Harris WS, Connor WE. Inhibition oflow density
lipoprotein synthesis by dietary omega-3 fatty acids in humans. Ar-
teriosclerosis 1984;4:270-5.
27. Nestel P1, Connor WE, Reardon MF, Connor 5, Wong S. Boston
R. Suppression by diets rich in fish oil ofvery low density lipoprotein
production in man. I Clin Invest 1984:74:82-9.
28. Kestin M, Clifton P, Belling GB, Nestel P1. n-3 fatty acids of marine
origin lower systolic blood pressure and triglycerides but raise LDL
cholesterol compared with n-3 and n-6 fatty acids from plants.
Am I Clin Nutr 1990:1:1028-34.
29. Singer P, Berger I, Luck K, Taube C, Naumann E, Godicke W.
Long-term effect of mackerel diet on blood pressure, serum lipids
and thromboxane formation in patients with mild essential hyper-
tension. Atherosclerosis 1986:62:259-65.
by guest on July 13, 2011www.ajcn.orgDownloaded from