Dyslipidemia and an Unfavorable Fatty Acid Profile Predict

Left Ventricular Hypertrophy 20 Years Later

Johan Sundström, MD; Lars Lind, MD, PhD; Bengt Vessby, MD, PhD; Bertil Andrén, MD, PhD;
Antti Aro, MD, PhD; Hans O. Lithell, MD, PhD

Background—Left ventricular hypertrophy (LVH) is a common risk factor for cardiovascular mortality. Causes other than
hypertension have not previously been investigated longitudinally. The aim of the present study was to determine
hemodynamic, metabolic, and psychosocial predictors at 50 years of age for the prevalence of echocardiographic LVH
and geometric subtypes at age 70 by use of a large sample of men from the general population followed up for 20 years.
Methods and Results—In 1970 to 1973, all men born from 1920 to 1924 and residing in Uppsala County, Sweden, were
invited to participate in a health survey aimed at identifying risk factors for cardiovascular disease. At a reinvestigation
20 years later, echocardiographic left ventricular mass index was determined in 475 subjects. A 1-SD increase in body
mass index, systolic or diastolic blood pressure, fasting LDL/HDL cholesterol, serum triglycerides, or the serum
cholesterol ester proportion of several saturated fatty acids or oleic acid at age 50 significantly increased the odds of
having LVH at age 70 by 27% to 41%, whereas an increase in linoleic acid proportion was protective. Almost all
metabolic predictors were independent of ischemic heart disease, valvular disease, and use of antihypertensive
medication at age 70.
Conclusions—Dyslipidemia and indices of a low dietary intake of linoleic acid and high intake of saturated and
monounsaturated fats, as well as hypertension and obesity, at age 50 predicted the prevalence of LVH 20 years later in
this prospective longitudinal cohort study, thereby suggesting that lipids may be important in the origin of LVH.
(Circulation. 2001;103:836-841.)
Key Words: hypertrophy 䡲 lipids 䡲 fatty acids 䡲 insulin 䡲 epidemiology

E chocardiographically determined left ventricular hyper- patients even more than pharmacological antihypertensive
Downloaded from http://ahajournals.org by on March 8, 2019

trophy (LVH) is an important risk factor for cardiovas- treatment.14

cular and all-cause mortality and morbidity,1,2 and an in-
creased left ventricular relative wall thickness (RWT; wall
thickness divided by ventricular diameter) also has an adverse
prognosis.1,3
The cause of LVH is largely unknown. Cross-sectional
studies4 – 6 have shown male sex, age, hypertension, obesity,
certain valvular diseases, and previous myocardial infarction
to be related to LVH or left ventricular mass index (LVMI).
Heredity7 and alcohol use8 might also explain some of the
variance in left ventricular mass. In a prospective study,9
average blood pressure over 30 years was associated with
LVMI at follow-up. Hypertension is generally regarded as the
worst culprit, as indicated by the large number of clinical
trials on the subject.10 However, the variation in 24-hour
blood pressure explains only 25% to 30% of the variation in
left ventricular mass.11 Body size is a powerful determinant of
left ventricular mass12 and may explain part of the sex
difference in left ventricular mass.13 Weight reduction has
been shown to decrease LVMI in overweight hypertensive
Associations have recently been found15–17 between LVH
and the insulin resistance syndrome.18 The left ventricular
geometric correlates of insulin resistance are not clear; some
studies16,17,19,20 have found insulin resistance or impaired
glucose tolerance to be more closely related to thick left
ventricular walls or increased RWT than to LVH.
The fatty acid composition of serum cholesterol esters
(CEs) mainly reflects dietary fat quality over the previous 2
weeks21 and has been shown to predict myocardial infarc-
tion,22 but its relationship to LVH is not known. Furthermore,
the relationships between smoking or other psychosocial
factors and later LVH are not known.
Thus, there is a need for prospective studies of modifiable
predictors of LVH. The aim of the present study was to
determine hemodynamic, metabolic, dietary, and psychoso-
cial predictors at age 50 years for the prevalence of echocar-
diographic LVH and left ventricular geometric subtypes at
age 70 years by use of a large, regionally determined sample
of men from the general population followed up for 20 years.

Received July 20, 2000; revision received October 6, 2000; accepted October 10, 2000.
From the Departments of Public Health and Caring Sciences (J.S., B.V., H.O.L.) and Medical Sciences (L.L., B.A.), Uppsala University, Sweden, and
Department of Nutrition (A.A.), KTL (National Public Health Institute), Helsinki, Finland.
Correspondence to Johan Sundström, Department of Public Health and Caring Sciences/Geriatrics, PO Box 609, SE-75125 Uppsala, Sweden
(Kålsängsgränd 10D, SE-75319 Uppsala, Sweden). E-mail johan.sundstrom@geriatrik.uu.se
© 2001 American Heart Association, Inc.
Circulation is available at http://www.circulationaha.org

836
 Sundström et al
Methods
In 1970 to 1973, all men born from 1920 to 1924 and residing in
Uppsala County were invited to participate in a health survey23
aimed at identifying risk factors for cardiovascular disease. Of the
invited subjects, 2322 (82%) participated. At a reinvestigation of the
cohort 20 years later, an echocardiographic examination was made in
the first 583 consecutive subjects of 1221 available. The population
of the present study consisted of 475 men in whom determination of
left ventricular geometry was possible.24 All subjects gave informed
consent, and the study was approved by the Ethics Committee of
Uppsala University. Procedures followed were in accordance with
department guidelines.
Investigations at Age 50
These investigations have been described previously.23 Blood pres-
sure in the recumbent position was measured with a mercury
manometer, and radial pulse rate was counted. All blood samples
were drawn in the morning after an overnight fast. Blood glucose,
serum insulin, cholesterol, triglycerides, and HDL cholesterol were
measured, and LDL cholesterol was calculated with Friedewald’s
formula. The CE proportions of fatty acids (14:0 to 22:6 ␻3) were
determined by gas chromatography.22 No dietary records were
obtained, but in other population studies in middle-aged men in the
1970s in Sweden, intake of fat corresponded to ⬇40%, of carbohy-
drates to 45% to 50%, and of protein to 13% to 15% of energy intake.
The estimated intake of saturated fats corresponded to 17% to 18%
of energy intake. At age 70, 7-day dietary records in 444 of the 475
subjects showed an intake of fat corresponding to 35%, of carbohy-
drates to 48%, and of protein to 16% of energy intake, which was
comparable to other contemporary Swedish populations of the same
age.
A questionnaire covered level of physical activity (4 categories)
and education (5 categories). Coding of smoking (smoker, non-
smoker, ex-smoker), civil status, and socioeconomic status (3 social
classes, Central Bureau of Statistics) was based on interview
reports.23
Downloaded from http://ahajournals.org by on March 8, 2019
Investigations at Age 70
At age 70, 167 subjects were regularly using antihypertensive
medication, and 38 used lipid-lowering drugs, of whom 16 used
statins, 15 fibrates, 6 resins, and 2 nicotinic acid, as monotherapy or
in combination. Fifty-four subjects had been hospitalized owing to
ischemic heart disease (ICD-9 codes 410 to 414) between the
investigations at 50 and 70 years of age. Seventeen subjects had
significant echocardiographic valvular disease (aortic or mitral
stenosis or regurgitation grades 3 or 4).
A comprehensive 2D and Doppler echocardiography examination
was performed as described previously.24 Left ventricular mass was
divided by body surface area to obtain LVMI. LVH was defined as
an LVMI ⱖ150 g/m2 according to data from the Framingham Heart
Study,25 and a partition value of 0.4426 was used for RWT [(inter-
ventricular septum⫹posterior wall thickness)/left ventricular end-
diastolic diameter]. Thus, left ventricular geometry was considered
normal if RWT was ⬍0.44 and LVMI was ⬍150 g/m2. A normal
LVMI with increased RWT was denoted concentric remodeling,26
and a hypertrophic left ventricle was denoted eccentric if the RWT
was normal and concentric if the RWT was increased.
Statistical Analysis
Variables with a skewed distribution (fasting serum insulin, triglyc-
erides, LDL/HDL cholesterol, and CE proportion of palmitoleic,
stearic, ␥-linolenic, ␣-linolenic, eicosapentaenoic, and docosahexae-
noic acids) were logarithmically transformed to achieve normal
distribution, and these transformed variables were used in all
analyses. Logistic regression was used with LVH or increased RWT
at age 70 as outcome variables and standardized continuous variables
(mean⫽0, SD⫽1) or indicator variables for multilevel nominal
variables at age 50 as explanatory variables. Multiple logistic
regression was used to adjust for possible confounders, which were
treated as dichotomous variables. Squared terms and interaction
Dyslipidemia and Saturated Fats Predict LVH 837
terms were tested in all regressions. ANOVA was used to calculate
differences in means between left ventricular geometric subgroups
and factorial ANOVA to adjust for possible confounders. Post hoc
Bonferroni-adjusted comparisons were only performed if overall
ANOVA was significant. A ␹2 test was used to evaluate differences
in nominal variables between geometric subgroups. Two-tailed
significance values were given with P⬍0.05 regarded as significant.
Stata 6.0 software was used (Stata Corporation).
Results
Prediction of LVH
The prevalence of LVH at age 70 was 28% in the present
cohort. Among the variables associated with the insulin
resistance syndrome, a 1-SD increase in body mass index,
systolic or diastolic blood pressure, fasting LDL/HDL cho-
lesterol, or serum triglycerides at age 50 resulted in 27% to
41% increased odds of having LVH at age 70 (Table 1). A
1-SD increase in the CE proportion of any of myristic (14:0),
palmitic (16:0), stearic (18:0), oleic (18:1), or eicosapenta-
enoic acid (20:5 ␻3) at age 50 increased the odds of having
LVH at age 70 by 29% to 37%, whereas a 1-SD increase in
the proportion of linoleic acid (18:2 ␻6) at age 50 reduced the
odds of having LVH at age 70 by 24% (Table 1). Levels of
physical activity, education, smoking, civil status, and socio-
economic status at age 50, crude or adjusted, were not
associated with later LVH (data not shown). With simulta-
neous control for ischemic heart disease during follow-up,
valvular disease, and use of antihypertensive medication at
age 70, the adjusted odds ratios were similar to the crude odds
ratios, but systolic and diastolic blood pressure and palmitic
acid were no longer significantly associated with LVH (Table
1). With control also for body mass index at age 50,
LDL/HDL cholesterol and proportions of stearic, oleic, lino-
leic, and eicosapentaenoic acids remained significant predic-
tors of LVH.
Prediction of Increased Left Ventricular RWT
The prevalence of increased RWT at age 70 was 25% in the
present cohort. A 1-SD increase in CE proportion of oleic
acid (18:1), ␥-linolenic acid (18:3 ␻6), or dihomo-␥-linolenic
acid (20:3 ␻6) at age 50 increased the odds of having
increased RWT at age 70 by 28% to 32%, whereas a 1-SD
increase in the proportion of linoleic acid (18:2 ␻6) at age 50
reduced the odds of having increased RWT at age 70 by 26%
(Table 2). The studied psychosocial issues were not associ-
ated with increased RWT (data not shown). Controlling for
ischemic heart disease during follow-up, valvular disease,
and use of antihypertensive medication at age 70 gave results
similar to the unadjusted analysis (Table 2). When body mass
index at age 50 was also controlled for, proportions of oleic
and linoleic acid remained significantly associated with
increased RWT.
We reperformed all the above analyses in a subsample of
421 subjects without ischemic heart disease during follow-up.
In logistic regression analysis with LVH at age 70 as the
dependent variable, adjusted for valvular disease and use of
antihypertensive medication at age 70, significant predictors
were the same as when adjusted for ischemic heart disease
during follow-up in the total sample, with the addition of total
cholesterol and the exception of myristic acid (14:0). When
 838 Circulation February 13, 2001

TABLE 1. ORs for Having LVH at Age 70 for a 1-SD Increase in a Variable at Age 50
Crude OR Adjusted OR
n Characteristics at Age 50 (95% CI) (95% CI)*
Body mass index 475 24.8⫾2.9 kg/m2 1.39 (1.14–1.70)§ 1.29 (1.04–1.59)†
Systolic blood pressure 475 131⫾17 mm Hg 1.27 (1.05–1.55)† 1.07 (0.86–1.33)
Diastolic blood pressure 475 83⫾11 mm Hg 1.41 (1.16–1.72)§ 1.18 (0.95–1.48)
Pulse rate 475 68⫾11 bpm 0.97 (0.79–1.19) 0.90 (0.73–1.11)
fS-cholesterol 475 6.7⫾1.2 mmol/L 1.21 (0.99–1.48) 1.16 (0.95–1.43)
f-LDL/HDL cholesterol ratio 371 3.8⫾1.7 1.34 (1.06–1.69)† 1.31 (1.02–1.67)†
fS-triglycerides 475 1.7⫾0.8 mmol/L 1.36 (1.11–1.66)‡ 1.27 (1.03–1.57)†
fB-glucose 472 4.9⫾0.6 mmol/L 1.09 (0.89–1.33) 1.08 (0.88–1.32)
fS-insulin 323 11.4⫾6.9 ␮U/mL 1.23 (0.97–1.57) 1.20 (0.94–1.54)
CE–myristic acid 14:0 431 1.2⫾0.3% 1.29 (1.04–1.59)† 1.26 (1.02–1.56)†
CE–palmitic acid 16:0 431 11.7⫾0.9% 1.29 (1.04–1.61)† 1.24 (0.99–1.55)
CE–palmitoleic acid 16:1 431 3.5⫾1.1% 1.02 (0.82–1.26) 0.97 (0.78–1.21)
CE–stearic acid 18:0 431 1.2⫾0.3% 1.30 (1.05–1.60)† 1.33 (1.06–1.66)†
CE–oleic acid 18:1 431 19.1⫾2.4% 1.29 (1.05–1.60)† 1.27 (1.03–1.58)†
CE–linoleic acid 18:2 ␻6 431 54.6⫾4.6% 0.76 (0.61–0.93)‡ 0.77 (0.63–0.96)†
CE–␥-linolenic acid 18:3 ␻6 431 0.6⫾0.3% 1.02 (0.83–1.27) 1.01 (0.82–1.26)
CE–␣-linolenic acid 18:3 ␻3 431 0.6⫾0.2% 1.13 (0.91–1.40) 1.13 (0.91–1.41)
CE–dihomo-␥-linolenic acid 20:3 ␻6 431 0.6⫾0.1% 1.02 (0.83–1.26) 0.95 (0.76–1.18)
CE–arachidonic acid 20:4 ␻6 431 4.7⫾0.9% 1.13 (0.91–1.39) 1.15 (0.92–1.43)
CE–eicosapentaenoic acid 20:5 ␻3 431 1.2⫾0.6% 1.37 (1.10–1.70)‡ 1.38 (1.10–1.73)‡
CE–docosahexaenoic acid 22:6 ␻3 431 0.7⫾0.2% 1.23 (0.99–1.52) 1.24 (1.00–1.55)
Data are means or geometric means ⫾ SDs and odds ratios (95% CI). n indicates number of subjects; f, fasting; S, serum; and
B, blood.
*Adjusted for ischemic heart disease, valvular disease, and use of antihypertensive medication at age 70.
†P⬍0.05; ‡P⬍0.01; §P⬍0.001 for equal odds.
Downloaded from http://ahajournals.org by on March 8, 2019

increased RWT was used as dependent variable, only
dihomo-␥-linolenic acid (20:3 ␻6) was significant in the
ischemic heart disease–free subsample.
In the total sample, when also adjusting for use of lipid-
lowering drugs at age 70 (adjusting for antihypertensive
drugs, lipid-lowering drugs, and valvular disease at age 70
and ischemic heart disease during follow-up), the effect of
triglycerides (P⫽0.055) and LDL/HDL cholesterol
(P⫽0.063) on later LVH became of borderline significance,
but body mass index and CE proportions of fatty acids 14:0,
18:0, 18:1, 18:2 ␻6, and 20:5 ␻3 remained significant, and
the CE proportion of docosahexaenoic acid (22:6 ␻3) became
significantly directly related to later LVH. Regarding predic-
tion of later increased RWT, the same adjustment made oleic
acid lose significance, but other predictors remained
significant.
Characteristics at Age 50 According to Left
Ventricular Geometric Patterns at Age 70
The levels of several variables associated with the insulin
resistance syndrome and CE proportions of several fatty acids
determined at age 50 varied significantly between the 4 left
ventricular geometry groups determined at age 70 (Table 3).
In the concentric remodeling group, CE proportions of oleic
(18:1) and ␥-linolenic acid (18:3 ␻6) were significantly
higher and the proportion of linoleic acid (18:2 ␻6) was
significantly lower than in the normal geometry group. In the
concentric LVH group, diastolic blood pressure and CE
proportions of myristic (14:0), palmitic (16:0), oleic (18:1),
and eicosapentaenoic acid (20:5 ␻3) were significantly higher
and linoleic acid (18:2 ␻6) was significantly lower than in the
normal geometry group. In the eccentric LVH group, body
mass index, systolic and diastolic blood pressure, fasting
serum triglycerides, and CE proportions of oleic (18:1) and
eicosapentaenoic acid (20:5 ␻3) were significantly higher,
and linoleic acid (18:2 ␻6) was significantly lower than in the
normal geometry group. The studied psychosocial factors did
not vary between the left ventricular geometry groups (data
not shown). With simultaneous control for ischemic heart
disease during follow-up and for valvular disease and use of
antihypertensive medication at age 70, differences between
groups regarding proportions of oleic, linoleic, ␥-linolenic,
and eicosapentaenoic acids were still significant, whereas
other differences lost significance. When body mass index at
50 was also controlled for, differences in oleic, linoleic, and
eicosapentaenoic acids remained significant.
Discussion
In this 20-year follow-up of a large, regionally determined
sample of men from the general population, dyslipidemia and
a CE fatty acid composition indicating a high dietary intake
of saturated and monounsaturated fats and low intake of
linoleic acid at age 50 predicted the prevalence of LVH at age
70 to a similar degree as hypertension and obesity. The
 Sundström et al Dyslipidemia and Saturated Fats Predict LVH 839

TABLE 2. ORs for Having Increased Left Ventricular RWT at
Age 70 for a 1-SD Increase in a Variable at Age 50
Body mass index
Systolic blood pressure
Diastolic blood pressure
Pulse rate
fS-cholesterol
f-LDL/HDL cholesterol ratio
fS-triglycerides
fB-glucose
fS-insulin
CE–myristic acid 14:0
CE–palmitic acid 16:0
CE–palmitoleic acid 16:1
CE–stearic acid 18:0
CE–oleic acid 18:1
CE–linoleic acid 18:2 ␻6
CE–␥-linolenic acid 18:3
␻6
CE–␣-linolenic acid 18:3
␻3
CE–dihomo-␥-linolenic acid
20:3 ␻6
CE–arachidonic acid 20:4
␻6
CE–eicosapentaenoic acid
Downloaded from http://ahajournals.org by on March 8, 2019
reason for these differences is not clear, but the metabolic
associations may be affected by the normally observed
Crude OR Adjusted OR* age-related increase in RWT.27
n (95% CI) (95% CI) The relationship between an unfavorable CE fatty acid
475 1.09 (0.88–1.33) 1.10 (0.89–1.36) composition and later LVH is important, because the fatty
acid composition reflects dietary fat quality over the past
475 1.14 (0.93–1.39) 1.12 (0.90–1.41)
couple of weeks21 and thus is a modifiable risk factor.
475 1.18 (0.96–1.44) 1.18 (0.93–1.49)
Previous studies of the present cohort have shown that a fatty
475 1.12 (0.92–1.37) 1.11 (0.90–1.37)
acid profile indicating a high dietary intake of saturated fats
475 1.17 (0.95–1.44) 1.18 (0.96–1.45) and low intake of linoleic acid is related to insulin resistance28
371 1.18 (0.93–1.50) 1.20 (0.94–1.55) and predicts myocardial infarction over 19 years.22 Because
475 1.04 (0.85–1.28) 1.06 (0.86–1.32) the impact of other traditional risk factors on myocardial
472 0.88 (0.71–1.10) 0.88 (0.71–1.10) infarction22 was similar to their impact on LVH in the present
323 1.00 (0.77–1.29) 1.02 (0.78–1.32) study, LVH might be regarded as an intermediary risk factor
431 1.07 (0.87–1.33) 1.08 (0.87–1.34) on the pathway between the described risk factor profile and
431 1.24 (0.99–1.55) 1.26 (1.00–1.58)
myocardial infarction. The apparently reciprocal effects of
linoleic acid (which makes up ⬎50% of all CE fatty acids)
431 1.24 (1.00–1.53) 1.24 (1.00–1.54)
and the saturated fatty acids may be due to passive redistri-
431 1.04 (0.84–1.30) 1.06 (0.85–1.32)
bution of the less abundant acids after changes in linoleic acid
431 1.32 (1.06–1.64)† 1.33 (1.07–1.66)† proportion occur but may also reflect true pathophysiological
431 0.74 (0.60–0.92)‡ 0.73 (0.58–0.91)‡ effects of the saturated fatty acids. In these men, the dietary
431 1.28 (1.02–1.60)† 1.29 (1.03–1.62)† source of oleic acid was not olive oil (which was not used by
middle-aged men in Sweden in the early 1970s) but food
431 1.12 (0.90–1.39) 1.12 (0.90–1.40) groups containing a high proportion of saturated fatty acids,
such as dairy products, solid margarines, and meat products.
431 1.28 (1.03–1.58)† 1.28 (1.03–1.59)† The low impact of ␻3 polyunsaturated fatty acids, which are
generally regarded as favorable, and the initially surprising
431 1.03 (0.83–1.28) 1.04 (0.84–1.30) relationship between eicosapentaenoic acid (20:5 ␻3) levels
and later LVH may be because a high intake of linoleic
431 1.15 (0.92–1.43) 1.16 (0.93–1.45) acid–rich vegetable fats reduces ␻3-acid levels in serum21 by

20:5 ␻3
CE–docosahexaenoic acid 431 0.99 (0.79–1.22) 1.00 (0.80–1.24)
22:6 ␻3
Abbreviations as in Table 1.
*Adjusted for ischemic heart disease, valvular disease, and use of antihy-
pertensive medication at age 70.
†P⬍0.05; ‡P⬍0.01 for equal odds.
impact of obesity, dyslipidemia, and an unfavorable fatty acid
profile on LVH was independent of history of ischemic heart
disease, valvular disease, and use of antihypertensive medi-
cation at age 70. The relationships between dyslipidemia and
an unfavorable fatty acid profile and later LVH were also
independent of obesity at age 50. Previously, hypertension
has been related to the development of LVH in a prospective
study,9 but the predictive value of obesity, dyslipidemia, and
indices of an unfavorable diet for LVH has never been shown.
Focus has recently turned to the association between LVH
and the insulin resistance syndrome.15–17,19,20 Some cross-
sectional population studies on the subject exist,15,19,20 but
most studies have been made in small groups of hypertensive
patients, which would not be ideal for the search for corre-
lates to LVH other than hypertension. In the present study,
lipid derangements, unfavorable fatty acid profile, and obe-
sity seemed to be powerful longitudinal predictors of LVH
development, in contrast to a cross-sectional analysis of the
same cohort at age 70 in which several measures of glucose
intolerance and insulin resistance were related to an increased
RWT and concentric remodeling but less to LVH.20 The
a decreased conversion of ␣-linolenic acid (18:3 ␻3) to
eicosapentaenoic acid (20:5 ␻3) through competition for the
same enzyme systems. Other factors than dietary fat intake
may affect CE fatty acid profile, but the relationships between
LDL/HDL cholesterol and triglycerides and later LVH sup-
port the adverse role of a diet rich in saturated fats.
No psychosocial variables were associated with develop-
ment of LVH in the present study, in accordance with a
cross-sectional population study15 in which exercise physical
activity and smoking were not related to LVH. The present
cohort has been closely monitored for 20 years and may
therefore be healthier than average Swedish 70-year-old men.
Thus, associations in the present study may be weaker than in
the general population.
One limitation of the study is the lack of dietary records at
baseline and the resulting lack of knowledge about any
potential dietary certainties in the studied population, which
might influence the generalizability of the findings of the
present study. Other limitations of the study include absence
of echocardiographic data at baseline, possible underestima-
tion of the impact of variables with substantial measurement
error or missing data, and bias due to loss to follow-up. Of
course, confounding factors other than the ones for which we
adjusted may also exist. Because many analyses were per-
formed, some chance associations may have been found. This
study has a limited generalizability to women and other age
and ethnic groups, and further studies are needed for confir-
mation of these findings.
 840 Circulation February 13, 2001

TABLE 3. Characteristics at Age 50 According to Left Ventricular Geometric Patterns at Age 70

Concentric Concentric Eccentric Crude Adjusted
Normal Remodeling LVH LVH ANOVA P ANOVA P *
Number of subjects 262 79 40 94
2
Body mass index, kg/m 24.5⫾2.8 24.8⫾3.0 25.3⫾2.8 25.6⫾3.2‡ 0.008 0.08
Systolic blood pressure, mm Hg 129⫾17 132⫾15 135⫾17 135⫾17† 0.046 0.46
Diastolic blood pressure, mm Hg 81⫾10 83⫾10 86⫾12‡ 85⫾11‡ 0.003 0.25
Pulse rate, bpm 68⫾11 69⫾9 69⫾12 68⫾10 0.72 0.56
fS-cholesterol, mmol/L 6.6⫾1.2 6.7⫾1.1 7.0⫾1.1 6.8⫾1.3 0.12 0.17
f-LDL/HDL cholesterol ratio 3.6⫾1.8 3.8⫾1.9 4.2⫾1.2 4.0⫾1.5 0.06 0.09
fS-triglycerides, mmol/L 1.6⫾0.7 1.6⫾0.7 1.8⫾0.6 1.8⫾1.1‡ 0.03 0.16
fB-glucose, mmol/L 4.9⫾0.6 4.9⫾0.6 4.8⫾0.4 5.0⫾0.6 0.27 0.27
fS-insulin, ␮U/mL 11.0⫾6.7 11.6⫾6.5 11.0⫾5.1 12.8⫾8.1 0.15 0.24
CE–myristic acid 14:0, % 1.1⫾0.2 1.1⫾0.2 1.2⫾0.3† 1.2⫾0.3 0.04 0.06
CE–palmitic acid 16:0, % 11.6⫾0.8 11.7⫾0.9 11.9⫾0.9† 11.8⫾1.0 0.04 0.07
CE–palmitoleic acid 16:1, % 3.4⫾1.0 3.7⫾1.2 3.6⫾1.1 3.4⫾0.9 0.27 0.26
CE–stearic acid 18:0, % 1.1⫾0.2 1.2⫾0.3 1.2⫾0.3 1.2⫾0.3 0.08 0.07
CE–oleic acid 18:1, % 18.8⫾2.3 19.4⫾2.2† 19.9⫾2.7‡ 19.4⫾2.5† 0.009 0.01
CE–linoleic acid 18:2 ␻6, % 55.3⫾4.3 53.8⫾4.6† 53.2⫾5.0‡ 53.9⫾4.8† 0.003 0.005
CE–␥-linolenic acid 18:3 ␻6, % 0.6⫾0.3 0.7⫾0.4‡ 0.6⫾0.2 0.6⫾0.3 0.04 0.03
CE–␣-linolenic acid 18:3 ␻3, % 0.6⫾0.2 0.6⫾0.2 0.7⫾0.2 0.6⫾0.2 0.40 0.39
CE–dihomo-␥-linolenic acid 20:3 ␻6, % 0.5⫾0.1 0.6⫾0.2 0.6⫾0.1 0.6⫾0.1 0.08 0.08
CE–arachidonic acid 20:4 ␻6, % 4.7⫾0.9 4.8⫾1.0 4.7⫾0.8 4.9⫾1.0 0.36 0.38
CE–eicosapentaenoic acid 20:5 ␻3, % 1.2⫾0.5 1.2⫾0.7 1.4⫾0.5† 1.3⫾0.5† 0.03 0.03
CE–docosahexaenoic acid 22:6 ␻3, % 0.7⫾0.2 0.7⫾0.2 0.7⫾0.2 0.7⫾0.2 0.11 0.12
Data are means or geometric means ⫾ SDs. Abbreviations as in Table 1.
*Adjusted for ischemic heart disease, valvular disease, and use of antihypertensive medication at age 70.
Downloaded from http://ahajournals.org by on March 8, 2019

†P⬍0.05; ‡P⬍0.01 for no difference from group with normal left ventricular geometry, crude analysis.

In conclusion, dyslipidemia and indices of a low dietary
intake of linoleic acid and high intake of saturated and
monounsaturated fats, as well as obesity and hypertension, at
age 50 predicted the prevalence of LVH at age 70 in this
20-year follow-up of a large, regionally determined sample of
men derived from the general population. The impact of
obesity, dyslipidemia, and evidence of high dietary intake of
saturated fats on LVH was independent of history of ischemic
heart disease, valvular disease, and use of antihypertensive
medication, indicating that lipids may be important in the
origin of LVH.
Acknowledgments
This study was supported by the Swedish Medical Research Council
No. 5446, Trygg Hansa, Thuréus Foundation, Foundation for Geri-
atric Research (Stiftelsen för Geriatrisk Forskning), Geriatric Fund
(Geriatriska Fonden), and Uppsala University.
References
1. Koren MJ, Devereux RB, Casale PN, et al. Relation of left ventricular
mass and geometry to morbidity and mortality in uncomplicated essential
hypertension. Ann Intern Med. 1991;114:345–352.
2. Levy D, Garrison RJ, Savage DD, et al. Prognostic implications of
echocardiographically determined left ventricular mass in the Fra-
mingham Heart Study. N Engl J Med. 1990;322:1561–1566.
3. Verdecchia P, Schillaci G, Borgioni C, et al. Adverse prognostic signif-
icance of concentric remodeling of the left ventricle in hypertensive
patients with normal left ventricular mass. J Am Coll Cardiol. 1995;25:
871– 878.
4. Levy D, Anderson KM, Savage DD, et al. Echocardiographically detected
left ventricular hypertrophy: prevalence and risk factors: the Framingham
Heart Study. Ann Intern Med. 1988;108:7–13.
5. Hammond IW, Devereux RB, Alderman MH, et al. Relation of blood
pressure and body build to left ventricular mass in normotensive and
hypertensive employed adults. J Am Coll Cardiol. 1988;12:996 –1004.
6. Sundstrom J, Lind L, Andren B, et al. Left ventricular geometry and
function are related to electrocardiographic characteristics and diagnoses.
Clin Physiol. 1998;18:463– 470.
7. Post WS, Larson MG, Myers RH, et al. Heritability of left ventricular
mass: the Framingham Heart Study. Hypertension. 1997;30:1025–1028.
8. Manolio TA, Levy D, Garrison RJ, et al. Relation of alcohol intake to left
ventricular mass: the Framingham Study. J Am Coll Cardiol. 1991;17:
717–721.
9. Lauer MS, Anderson KM, Levy D. Influence of contemporary versus
30-year blood pressure levels on left ventricular mass and geometry: the
Framingham Heart Study. J Am Coll Cardiol. 1991;18:1287–1294.
10. Dahlof B, Pennert K, Hansson L. Reversal of left ventricular hypertrophy
in hypertensive patients: a meta-analysis of 109 treatment studies. Am J
Hypertens. 1992;5:95–110.
11. Majahalme S, Turjanmaa V, Weder A, et al. Blood pressure levels and
variability, smoking, and left ventricular structure in normotension and in
borderline and mild hypertension. Am J Hypertens. 1996;9:1110 –1118.
12. Lauer MS, Anderson KM, Kannel WB, et al. The impact of obesity on left
ventricular mass and geometry: the Framingham Heart Study. JAMA.
1991;266:231–236.
13. de Simone G, Devereux RB, Daniels SR, et al. Gender differences in left
ventricular growth. Hypertension. 1995;26:979 –983.
14. MacMahon SW, Wilcken DE, Macdonald GJ. The effect of weight
reduction on left ventricular mass: a randomized controlled trial in young,
overweight hypertensive patients. N Engl J Med. 1986;314:334 –339.
15. Marcus R, Krause L, Weder AB, et al. Sex-specific determinants of
increased left ventricular mass in the Tecumseh Blood Pressure Study.
Circulation. 1994;90:928 –936.
 Sundström et al
16. Lind L, Andersson PE, Andren B, et al. Left ventricular hypertrophy in
hypertension is associated with the insulin resistance metabolic
syndrome. J Hypertens. 1995;13:433– 438.
17. Paolisso G, Galderisi M, Tagliamonte MR, et al. Myocardial wall
thickness and left ventricular geometry in hypertensives: relationship with
insulin. Am J Hypertens. 1997;10:1250 –1256.
18. Reaven GM. Banting lecture 1988: role of insulin resistance in human
disease. Diabetes. 1988;37:1595–1607.
19. Ohya Y, Abe I, Fujii K, et al. Hyperinsulinemia and left ventricular
geometry in a work-site population in Japan. Hypertension. 1996;27:
729 –734.
20. Sundstrom J, Lind L, Nystrom N, et al. Left ventricular concentric
remodeling rather than left ventricular hypertrophy is related to the
insulin resistance syndrome in elderly men. Circulation. 2000;101:
2595–2600.
21. Nikkari T. Serum fatty acids and coronary heart disease in Finnish
populations. Prog Lipid Res. 1986;25:437– 450.
22. Ohrvall M, Berglund L, Salminen I, et al. The serum cholesterol ester
fatty acid composition but not the serum concentration of alpha
Downloaded from http://ahajournals.org by on March 8, 2019
Dyslipidemia and Saturated Fats Predict LVH 841
tocopherol predicts the development of myocardial infarction in
50-year-old men: 19 years follow-up. Atherosclerosis. 1996;127:65–71.
23. Hedstrand H. A study of middle-aged men with particular reference to
risk factors for cardiovascular disease. Ups J Med Sci Suppl. 1975;
19:1– 61.
24. Andren B, Lind L, Hedenstierna G, et al. Left ventricular hypertrophy and
geometry in a population sample of elderly males. Eur Heart J. 1996;17:
1800 –1807.
25. Levy D, Savage DD, Garrison RJ, et al. Echocardiographic criteria for left
ventricular hypertrophy: the Framingham Heart Study. Am J Cardiol.
1987;59:956 –960.
26. Ganau A, Devereux RB, Roman MJ, et al. Patterns of left ventricular
hypertrophy and geometric remodeling in essential hypertension. J Am
Coll Cardiol. 1992;19:1550 –1558.
27. Ganau A, Saba PS, Roman MJ, et al. Ageing induces left ventricular
concentric remodelling in normotensive subjects. J Hypertens. 1995;13:
1818 –1822.
28. Vessby B, Tengblad S, Lithell H. Insulin sensitivity is related to the fatty
acid composition of serum lipids and skeletal muscle phospholipids in
70-year-old men. Diabetologia. 1994;37:1044 –1050.