1657
Plasma Lipoprotein (a) Levels in Men and
Women Consuming Diets Enriched in Saturated,
Cis-, or 7>aws-Monounsaturated Fatty Acids
Beverly A. Clevidence, Joseph T. Judd, Ernst J. Schaefer, Jennifer L. Jenner,
Alice H. Lichtenstein, Richard A. Muesing, Janet Wittes, Matthew E. Sunkin
Abstract Studies that have shown adverse effects of trans-
unsaturated fatty acids on plasma lipoprotein (a) [Lp(a)] levels
have used levels of trans-fatty acid that are higher than those in
the average U.S. diet. This study was conducted to clarify the
effects on Lp(a) of trans-fatty acids levels commonly found in
U.S. diets. Lp(a) levels were measured in a double-blind study
of 29 men and 29 women who ate 4 controlled diets in random
order for 6 weeks each. Fatty acids represented 39% to 40% of
energy. The diets were: (1) Oleic (16.7% of energy as oleic
acid); (2) Moderate trans (3.8% of energy as frawj-monoenes,
approximately the trans content of the U.S. diet); (3) High trans
(6.6% of energy as trans-monoenes); (4) Saturated (16.2% of
energy as lauric plus myristic plus palmitic acids). The Satu-
rated diet lowered Lp(a) levels significantly (by 8% to 11%).
L ipoprotein (a), a low-density lipoprotein (LDL)
Downloaded from http://ahajournals.org by on January 20, 2023
particle linked to apoprotein (a) [apo(a)], may
be an independent risk factor for developing
coronary heart disease. 1 5 Although the role of Lp(a) in
promoting heart disease has not been definitively eluci-
dated, the particle has both atherogenic and thrombotic
potentials. Moreover, it has been reported to be consid-
erably more atherogenic than LDL. 5
Lipoprotein (a) levels are thought to be largely con-
trolled by genetic factors. 6 However, several reports
have indicated that the type of fat consumed can alter
Lp(a) levels. 711 The possibility of modulating Lp(a)
levels by dietary means is intriguing because effective
pharmacological treatments for lowering Lp(a) concen-
trations are limited. 12
7ra/w-unsaturated fatty acids, produced largely during
the partial hydrogenation of vegetable oils, have been
implicated in raising Lp(a) plasma levels.9-10 However,
this finding has not been consistent among studies. 1013
Received August 13, 1996; revision accepted February 19, 1997.
From the Diet and Human Performance Laboratory, Beltsville
Human Nutrition Research Center, Agricultural Research Service,
USDA (B.A.C., J.T.-J., M.E.S.), Beltsville, MD; Lipid Metabolism
Laboratory, the Jean Mayer USDA Human Nutrition Research
Center on Aging at Tufts University (E.J.S., J.L.-J., A.H.L.),
Boston, MA; The Lipid Research Clinic (R.A.M.), The George
Washington University Medical Center, Washington, DC; and
Statistics Collaborative (J.W.), Washington, DC.
Correspondence to Beverly A. Clevidence, PhD, Diet and
Human Performance Laboratory, Beltsville Human Nutrition
Research Center, Building 308, Room 115, BARC-East, 10300
Baltimore Avenue, Beltsville, MD 20705-2350. E-mail
Bev@bhnrc.arsusda.gov
© 1997 American Heart Association, Inc.
Compared to the Oleic diet, the trans diets had no adverse
effect on Lp(a) levels when all subjects were considered
collectively. A subset with initially high levels of Lp(a) (>30
mg/dL), however, responded to the High trans diet with a slight
(5%) increase in Lp(a) levels relative to the Oleic and Moder-
ate trans diets. Thus, in amounts commonly found in the typical
U.S. diet, saturated fatty acids consistently decrease Lp(a)
concentrations. The adverse effects of replacing cis- with
trans-fatty acids are only suggestive and are restricted to high
trans intakes in subjects with high Lp(a) levels. (Arterioscler
Thromb Vase Biol. 1997;17:1657-1661.)
Key Words • lipoprotein (a) • human diet • trans-fatty
acids • hydrogenated fat • saturated fat • dietary fatty
acids • monounsaturated fat • elaidic acid
These discrepancies may be due to variation in dietary
design, length of feeding, characteristics of study sub-
jects, or the amounts or isomeric compositions of the
trans-fatty acids used. Reports from recent human feed-
ing trials agree that trans- and c/s-fatty acids have
different metabolic effects on plasma LDL and possibly
on high-density lipoprotein (HDL), and that trans-
unsaturates adversely affect lipoprotein profiles. 1013 - 17 If
at levels of intake characteristic of the U.S. diet trans-
monounsaturates are consistently demonstrated to have
the additional adverse effect of elevating Lp(a) levels,
the need to reduce trans-fatty acids in the U.S. food
supply would be accentuated.
We measured Lp(a) levels from plasma samples col-
lected during a study that investigated the response of
blood lipids to trans-fatty acid consumption. 17 The ob-
jective of the present investigation was to compare
plasma Lp(a) levels that result from consuming diets
containing ?ra«.?-unsaturated fatty acids within the range
typically consumed in the U.S. diet on cw-monounsatu-
rated fatty acids (oleic) and "cholesterol-raising" satu-
rated fatty acids (lauric, myristic, and palmitic acids).
Methods
Subjects
The subjects, healthy adults (29 men, 29 women) with a
mean age of 43 years (range, 25 to 65 years), were free of
metabolic disorders, including obesity, diabetes, gout, endo-
crine disorders, or liver and kidney disease. Recruitment details
have been reported.17 Inclusion criteria required that volun-
teers have (1) plasma cholesterol levels between the 50th and
the 75th percentiles for the sex and age of the volunteers
screened; (2) HDL cholesterol levels above 0.91 mmol/L (35
 1658 Arteriosclerosis, Thrombosis, and Vascular Biology
Selected Abbreviations and Acronyms
ANOVA = analysis of variance
apo(a) = apoprotein (a)
HDL = high-density lipoprotein
LDL = low-density lipoprotein
Lp(a) = lipoprotein (a)
mg/dL) for men and 1.03 mmol/L (40 mg/dL) for women; and
(3) triglyceride levels under 3.39 mmol/L (300 mg/dL).
Subjects included 12 African Americans (7 men and 5
women), 7 smokers, and 7 women on steroidal hormones (2 on
oral contraceptives; 5 on hormone replacement). The type and
dosage of hormones did not change over the course of the study
as assessed from daily records of prescription and nonprescrip-
tion medication. Procedures for research with human volun-
teers were approved by the Institutional Review Board,
Georgetown University School of Medicine.
Experimental Diets and Design
Four diets enriched in fatty acids—oleic (Oleic), moderate
trans (Mod Trans), high trans (High Trans), and saturated
(Sat)—were fed in a Latin square design (6-week periods) so
that each subject consumed all 4 diets. Subjects were blind to
dietary treatments. A previous report describes the diets and
their formulation and analyzed nutrient values and procedures
for blinding.17
Each diet contained 39% to 40% of energy as fatty acids (Table
1). The Sat diet contained 16% of energy as cholesterol-raising
saturated fatty acids (lauric+myristic+palmitic) and 11% of en-
ergy as oleic acid. To formulate the Oleic, Mod Trans, and High
Trans diets, the Sat diet was modified by replacing 6% of energy
from saturated fat (lauric+myristic+palmitic) with an equal
Downloaded from http://ahajournals.org by on January 20, 2023
amount of either oleic acid (Oleic diet), a combination of oleic
and trans-fatty acids (Mod Trans diet), or trans-fatty acids (High
Trans diet). The Oleic, Mod Trans, and High Trans diets provided
17% to 18% of energy as monounsaturates (oleic or o\e\c+trans-
fatty acids). Naturally occurring &ww-isomers from meat, dairy
products, and other foods represented about 0.7% of energy in all
diets. The levels of stearic acid, which, unlike other common
saturated fatty acids is not hypercholesterolemic,1820 were held
constant across diets at 3% of energy. Cholesterol levels were held
constant at approximately 0.083 mmol/MJ (13.5 mg/100 kcal) or
about 400 mg/d at the average energy level. Dietary fiber intake
was 22 g/d to 25 g/d at an energy intake of 11.72 Ml.
Meals were prepared at the Human Studies Facility, Belts-
ville Human Nutrition Research Center, using a 14-day menu
cycle. Trans-fatty acids were incorporated into margarines,
baked goods, dips, gravies, and other foods using partially
hydrogenated fats typical of those found in the U.S. food
TABLE 1. Intakes of the 58 Participants on the 4
Experimental Diets (% of energy)
Mod High
Oleic Trans Trans Sat
Total fatty acids 39.0 39.0 39.0 40.0
Protein 15.0 15.0 15.0 15.0
9.5 16.2
L+M+P* 10.6 9.9
Stearic acid 2.8 3.1 3.2
Oleic acid 16.7 14.1 11.4 10.9
Trans-fatty acidsf 0.7 3.8 6.6
Linoleic acid 6.1 6.0 6.2
Other fatty acidsj 2.4 2.4 2.2
* L+M+P=Lauric+myristic+palmitic acids.
t Trans-fatty acids include naturally occurring frans-isomers from
meat, dairy products, and other foods.
t The category for other fatty acids includes short-chain fatty acids
(C<12), saturated fatty acids (C>18), and cis-isomers of mono- and
polyunsaturated fatty acids other than oleic and linoleic acids.
Vol 17, No 9 September 1997
supply. Coconut oil, high oleic sunflower oil, and high linoleic
safflower oil also were used to achieve the desired fatty acid
compositions of the diets. The test fats were provided by
member companies of the Institute of Shortening and Edible
Oils (Washington, DC). Foods were weighed in direct propor-
tion to caloric requirements. Subjects ate breakfast and dinner
meals under supervision, Monday through Friday, at the dining
facility. Lunch and weekend meals were packed for off-site
consumption. Subjects agreed to eat all of the food and only the
food provided by the study. Alcohol consumption was not
allowed. Body weights were recorded before breakfast, Mon-
day through Friday, and caloric intake was adjusted, as neces-
sary, to maintain body weights within a range of ±1 kg.
Each of the 4 diets was composited and analyzed 6 times at
the 11.72-MJ level and once at the 7.53-MJ level. Proximate
analysis of protein, fat and carbohydrate,21 and total dietary
fiber22 was performed by Hazelton Laboratories (Madison,
WI). Levels of trans-fatty acids were measured by infrared
spectroscopy;23 cis- and fra/w-positional isomers were deter-
mined by gas chromatography.24 Trans 18:1 monounsaturated
fatty acids comprised 98% or more of the total dietary trans-
fatty acids.
Lipoprotein Assays
Blood was collected from fasting subjects (>12-h fast) prior
to breakfast (between 6 AM and 8:30 AM) with replicates taken
on Monday and Wednesday, or Tuesday and Thursday, during
the 6th week of each feeding period. In addition, baseline
samples were collected twice the week before controlled feed-
ing started. Procedures used for blood sampling and processing
are those described in the protocol for the Lipid Research
Clinics program.25 Blood was collected in tubes containing 1.5
mg/mL Na2 EDTA (final concentration) promptly cooled on
wet ice, and plasma was separated by low-speed centrifugation
(1400 Xg) for 20 minutes at 4°C. Plasma samples were coded to
blind analysts to subject treatments. The method of Gidez et
al.2S was used to isolate HDL. The triglyceride and cholesterol
contents of plasma and the HDL fractions were determined
enzymatically (Sigma Chemical Co., St. Louis, MO). Concen-
trations of LDL cholesterol were calculated by the method of
Friedewald et al.27 A detailed description of the methodology
for determinations of lipoproteins other than Lp(a) has been
reported.17
Samples for Lp(a) analysis were frozen at -90°C. After the
study was completed, samples were shipped on dry ice to the
Lipid Metabolism Laboratory of the Jean Mayer USDA Hu-
man Nutrition Research Center on Aging at Tufts University
for analysis. Levels of Lp(a) were measured as described
previously3'1325 with a commercially available enzyme-linked
immunosorbent assay (Strategic Diagnostics, Newark, DE).
Briefly, diluted samples (1:201) were incubated for 1 hour in
microtiter strip wells coated with a monoclonal anti-Lp(a)
antibody that does not cross-react with plasminogen. After 4
washes, a 20-minute incubation with a polyclonal anti-Lp(a)
horseradish peroxidase conjugate was followed by another 4
washes. An incubation of substrate (hydrogen peroxide) and
chromogen (O-phenylenediamine) was stopped after 20 min-
utes by the addition of 2 N sulfuric acid. The absorbance was
read at 492 nm with a Dynatech MR 600 microtiter plate
reader (Dynatech Inc., Vienna, VA). The Lp(a) values were
calculated from the standard curve generated from the set of
3.1
standards provided with each plate. Intraassay and interassay
coefficients of variation were 3.2% and 5.2%, respectively.
0.7
All samples from an individual subject were included in a
6.1 single run.
2.9
Statistical Analysis
Values for Lp(a) were the average of the replicate determi-
nations taken on 2 days during the final week of the feeding
periods. Both logarithmic- and square-root transformations
were conducted to normalize Lp(a) data. Statistical analyses
 Clevidence et al Dietary Fatty Acids and Plasma Lp(a) Levels 1659

TABLE 2. Characteristics of the 58 Participants there were no statistically significant differences among
at Baseline* Lp(a) levels produced by consumption of the Oleic, Mod
Men Women Both Trans, or High Trans diets.
(n=29) (n=29) (n=58) The Lp(a) data were grouped into 3 strata according
Age (years) 41+2.0 44±1.9 43±1.4 to baseline Lp(a) levels: low (<5 mg/L), medium (>5
Body mass index 27.1+0.6 25.8±0.7 26.4±0.5 mg/dL but <30 mg/L), high (>30 mg/L). The cutoff of
Race (black/white) 7:22 5:24 12:46 30 mg/L was chosen because values at this level and
Steroid hormone 7 7 above28"30 have been associated with increased risk of
Smokers 2 5 7 coronary artery disease. The cutoff level of 5 mg/L was
Total cholesterol (mmol/L) 5.33+0.49 5.28+0.70 5.30±0.59 chosen arbitrarily. Data were analyzed by stratum to
LDL cholesterol (mmol/L) 3.44+0.44 3.21 ±0.62 3.31+0.54 avoid the possibility that the non-normal distribution of
HDL cholesterol (mmol/L) 1.29±0.36 1.45+0.31 1.37±0.34
Lp(a) would mask the effects of dietary fatty acids. In
Lp(a) (mg/dL) 24.9±3.8 24.7±5.3 24.8±3.2
particular, subjects with low levels of Lp(a) would not be
* Values are means±standard error or frequencies. expected to have important changes in response to diet.
We also attempted to stratify our data further to ascer-
were determined on both the original and transformed data. tain whether subjects with the very highest levels accord-
The reported data are arithmetic means. Data were analyzed ing to sextile ranking had increases in Lp(a) levels that
by analysis of variance (ANOVA) for main effects and inter- were clearly related to intake of trans-tatty acids. We
actions of gender, diet, and period using SAS version 6.06 (SAS performed analyses both within sextile and using the
Institute, Cary, NC). sextile as a stratifying variable. We report data for 3
strata (Table 3) because tests of the fit of the resulting
Results models showed no evidence that more than 3 strata
As commonly reported for this lipoprotein, Lp(a) would describe the data more accurately.
values were skewed; a high proportion of values fell at For each of the 3 strata, the Sat diet produced lower
the lower end of the range. The transformed data had levels of Lp(a) than did the other diets. Subjects who had
approximate normal distributions. Transformed and un- high levels of Lp(a) at baseline had slightly higher Lp(a)
transformed data produced essentially the same results,
levels when they ate the High Trans (56.0±0.6 mg/dL)
but the fit of the model was better for the raw than for
than when they consumed the Mod Trans (53.4±0.6
the transformed data. Thus Lp(a) values are given as
mg/dL) or Oleic (53.2±0.6 mg/dL) diets. These differ-
original, rather than transformed, data.
ences were nominally significant (i.e., statistically signif-
Baseline icant without correction for multiple comparisons,
Downloaded from http://ahajournals.org by on January 20, 2023

Baseline levels of Lp(a) and other plasma lipoproteins
are reported in Table 2. Levels of Lp(a) ranged from 0.5
mg/dL to 111 mg/dL. No difference in Lp(a) values due
to gender was detected at baseline. African Americans
had higher levels of Lp(a) (32.0±8.9 mg/dL) than whites
(22.9±3.4 mg/dL); however, the differences were not
statistically significant. It should be noted that the sta-
tistical test for difference by race has low power because
of the small number of African Americans. There were
no significant correlations of Lp(a) levels with total,
LDL, or HDL cholesterol levels.
Response to Diet
Because the Lp(a) values of men and women did not
differ in response to diet, the data were pooled across
gender. When the 58 subjects were considered collec-
tively, the Sat diet produced Lp(a) levels that were 8% to
11% lower than levels produced by the other diets; these
differences were statistically significant (Table 3). By
contrast, when all subjects were considered collectively,
P<.03). However, the effect of the High Traits diet was
not statistically significant after Bonferroni correction
for multiple comparisons, suggesting that the increase
could have been due to chance. Replacing cis- with
trans-tatty acids had no effect on Lp(a) levels of subjects
who had low or medium levels of Lp(a) at baseline and
no effect when all subjects were considered collectively.
The Lp(a) levels were not correlated with LDL choles-
terol levels (r<.02; P>A for all diets).
Discussion
We were particularly interested in the effect of trans-
fatty acids on Lp(a) given the recent evidence that the
ingestion of partially hydrogenated vegetable oils ad-
versely affects lipoprotein profiles.10'13-17 If trans-fatty
acids also increase Lp(a) levels, which may be an inde-
pendent risk factor for coronary heart disease,'-s then
public health concerns about the widespread use of food
products containing these fats might justifiably become
stronger.

TABLE 3. Levels of Lp(a) and LDL Cholesterol in Subjects Consuming 4 Experimental Diets*
Subjectst Oleic Mod Trans High Trans Sat
Lipoprotein(a) (mg/dL):): 58 23.8+0.4 a 23.8±0.4 a 24.7±0.4 a 21.9±0.4 b
Low 11 2.9+0.8 a 2.9+0.8" 3.0±0.8 a 2.6±0.8"
Medium 28 15.2±0.5a 15.1 ±0.5= 15.1±0.5a 12.6±0.5 b
High 19 53.2±0.6 a 53.4+0.6" 56.0+0.6" 50.4±0.6C
LDL cholesterol (mmol/L) 58 3.34+0.24" 3.54±0.24 b 3.60+0.24"° 3.64+0.24°
* Within a given row, values with different superscripts are nominally significantly different from one another (P<.05), that
is, no correction has been made for multiple comparisons. Values are means + standard error,
t Subjects are 29 men and 29 women.
t Low, 5 mg/dL or less; medium, greater than 5 mg/dL and less than 30 mg/dL; high, 30 mg/dL or greater.
 1660 Arteriosclerosis, Thrombosis, and Vascular Biology Vol 17, No 9 September 1997
Trans-fatty acids are produced from c/s-isomers dur-
ing the hydrogenation of vegetable oils. Partially hydro-
genated vegetable oils are widely used in margarines,
shortenings, commercial frying fats, and baked goods, in
large part because they provide the physical properties
and stability of saturated fats but contain less saturated
fatty acids than the traditional hard fats. Food compo-
sition data for trans-fatty acids of individual foods are
sparse, so data concerning the typical intake of trans-
fatty acids in the United States are less precise than for
most other types of fatty acids.31 Estimated intakes of
trans-fatty acids based on availability in the food supply
range from 8.1 g32 to 13.3 g33 per person per day, or
approximately 3% to 6% of energy for most adults in the
U.S. population. We chose to add trans-fatty acids from
partially hydrogenated vegetable oils at 3% of energy
intake because we estimate that this level approximates
the average level of U.S. consumption. Recognizing that
some individuals consume larger amounts, we also
added trans-fatty acids at a higher level, 6% of energy.
The Mod Trans diet provided 6.7 g of trans-fatty acids
for women at their average intake level of about 8.40 MJ
(2000 kcal) and 10.7 g for men at their average intake
level of about 13.39 MJ (3200 kcal) from partially
hydrogenated vegetable oils. The High Trans diets pro-
vided about twice these levels. These diets contained
additional (—2 g/d) trans-fatty acids from natural
sources (largely dairy products), as did the Oleic and Sat
diets.
Katan and coworkers9 found median levels of Lp(a) to
be 73% and 41% higher when 10% of energy as trans-
Downloaded from http://ahajournals.org by on January 20, 2023
monounsaturated fatty acids replaced an equivalent
amount of cholesterol-raising saturated fatty acids (lau-
ric, myristic, and palmitic acids) or oleic acid, respec-
tively. In another experiment from that laboratory, me-
dian levels of Lp(a) increased by 23% when 8% of
energy from trans-fatty acids replaced either linoleate or
stearate.9 Nestel et al.10 reported that trans-fatty acids at
7% of energy produced a 19% increase in Lp(a) levels
when elaidic acid was substituted for palmitic acid.
However, Lichtenstein et al.13 observed no increase in
Lp(a) levels when trans-fatty acids were fed at a more
moderate level, 4% of energy, by replacing corn oil with
corn oil margarine. Similarly, after adjusting for multiple
comparisons, we found no significant effects on Lp(a)
levels as a result of substituting either 3% or 6% of
energy as trans-fatty acids at the expense of oleic acid.
However, our data suggest that high levels of trans-fatty
acids may be associated with small increases in Lp(a)
levels in a subpopulation, those with high levels of Lp(a).
This conclusion is based on the observation that the 19
subjects with high baseline levels of Lp(a) (^30 mg/dL)
had, on average, a 5% increase in Lp(a) when they
consumed the diet with 6% of energy as trans-fatty acids
rather that the diets with oleic acid or with 3% of energy
as trans-fatty acids. This group also exhibited a 10%
increase in Lp(a) levels when 6% of energy as trans-fatty
acids replaced 6% of energy as saturated fatty acids. The
biological significance of these changes is unknown.
The dietary frans-monounsaturates in our study and
those from studies by Katan and coworkers91415 were
similar in that at least 98% of the total dietary trans-fatty
acids were 18:1 monounsaturates. They differed, how-
ever, in the distribution of trans-isomers. The predomi-
nant trans-isomers in the Mod Trans and High Trans
diets were 18:ln-8 (27.5%), 18:ln-7 (21.9%), 18:ln-9
(18.8%), 18:ln-6 (15.0%), and 18:ln-10 (14.2%). The
distribution of fram-isomers used by Katan and cowork-
ers9-14'15 was: 18:ln-9 (29.1%), 18:ln-8 (23.2%), and
18:ln-7 (20.6%). The relative distribution of trans-
monoenes in our study is typical of fats hydrogenated in
the United States. Thus, a possible explanation for the
lack of Lp(a) response in our study and that of Lichten-
stein et al.13 may be the differing pattern of trans-
monoenes produced by the different hydrogenation
methods used in the United States and for the Dutch
studies. The hydrogenation method used in the United
States produces less elaidic acid (18:ln-9), and elaidic
acid has been fed at higher levels in studies reporting
that trans-fatty acids affect Lp(a) levels.9'1415
Significant, although weak, correlations between lev-
els of Lp(a) and of apoprotein B or LDL have been
noted by some,5-34-35 but not all,14-36'37 investigators. We
found no relationships between Lp(a) levels and levels
of other lipoproteins; correlations were very small and
none was significantly different from zero. Our data are
consistent with the concept that Lp(a) is independently
regulated. Hepatic production may be the dominant
factor influencing levels of Lp(a).38
The diet enriched with saturated fatty acids produced
significantly lower levels of Lp(a) than did diets enriched
with c/s-monounsaturated and frans-monounsaturated
fatty acids. This finding is consistent with those of
previous reports that the cholesterol-raising saturated
fatty acids (lauric, myristic, and palmitic acids) produce
lower Lp(a) concentrations than oleic acid9 and that
butter produces lower levels of Lp(a) than either par-
tially hydrogenated safflower oil or partially hydroge-
nated fish oil.11
Assuming that the standard errors observed in this
study were the true underlying standard errors associ-
ated with diet, this study had an 80% power to detect
changes (in mg/dL) of 0.6, 2.1, and 4.4 for the low-,
medium-, and high-Lp(a) groups, respectively, and of 1.6
for the subjects collectively. These values were calcu-
lated with a Bonferroni adjustment for multiple compar-
isons (a=.05/3). Using a nominal a level of .05, the study
had an 80% power to detect changes (in mg/dL) of 0.5,
1.9, and 3.7 for the low-, medium-, and high-Lp(a)
groups, respectively, and 1.4 for the subjects collectively.
Thus, the study most likely could detect meaningful
differences in Lp(a) levels that occurred in response to
diet. We detected some differences that were smaller,
yet statistically significant, than those for which we had
80% power.
In summary, our data suggest that for individuals with
high levels of Lp(a) trans-fatty acids may adversely affect
Lp(a) levels. This finding, which was statistically signifi-
cant only prior to correcting for multiple comparisons,
must be considered suggestive rather than conclusive. If,
as generally believed, a high level of Lp(a) is an impor-
tant risk factor for cardiovascular disease, then the
biological significance of these changes will best be
evaluated by conducting dietary studies of individuals
with high levels of Lp(a). Perhaps more pertinent for the
general population is the impact of saturated fatty acids
on Lp(a) levels. Saturated fatty acids not only appear to
lower Lp(a) levels, but also do so consistently, regardless
of the initial Lp(a) level. Because of the well-known
ability of saturated fatty acids to raise LDL levels, it
 Clevidence et al Dietary Fatty Acids and Plasma Lp(a) Levels
remains important to clarify the impact of trans-fatty
acids on Lp(a) relative to viable dietary alternatives,
such as oleic acid.
Acknowledgments
We thank the Institute of Shortening and Edible Oils,
Washington, DC, and member companies for their financial
support and for supplying test fats and analytical data for these
fats. We thank Dr. Edward Emken, National Center for
Agricultural Utilization Research, Agricultural Research Ser-
vice, for analyzing cis- and trans-fatty acid positional isomers in
the test fats and experimental diets. We thank Priscilla Steele,
RD, and the staff of the Human Studies Facility, Beltsville
Human Nutrition Research Center, for preparing the con-
trolled diets. Jennifer Jenner was supported by National Insti-
tutes of Health training grant T32AG0029.
References
1. Hoefler G, Harnoncourt F, Paschke E, Mirtl W, Pfeiffer KH,
Kostner GM. Lipoprotein Lp(a): a risk factor for myocardial
infarction. Arteriosclerosis. 1988;8:398-401.
2. Rosengren A, Wilhelmsen L, Eriksson E, Risberg B, Wedel H.
Lipoprotein (a) and coronary heart disease: a prospective case-
control study in a general population sample of middle aged men.
BrMedJ. 1990;301:1248-1251.
3. Schaefer EJ, Lamon-Fava S, Jenner JL, McNamara JR, Ordovas
JM, Davis E, Abolafia JM, Lippel K, Levy RI. Lipoprotein (a)
levels and risk of coronary heart disease in men: the Lipid Research
Clinics Coronary Primary Prevention Trial. JAMA. 1994;27t:
999-1003.
4. Bostom AG, Gagnon DR, Cupples LA, Wilson PWF, Jenner JL,
Ordovas JM, Schaefer EJ, Castelli WP. A prospective investigation
of elevated lipoprotein (a) detected by electrophoresis and cardio-
vascular disease in women: the Framingham Heart Study. Circu-
lation. 1994;90:1688-1695.
5. Rhoads GG, Dahlen G, Berg K, Morton NE, Dannenberg AL.
Downloaded from http://ahajournals.org by on January 20, 2023
Lp(a) lipoprotein as a risk factor for myocardial infarction. JAMA.
1986;256:2540-2544.
6. Bloomsma DI, Kaptein A, Kempen HJM, Leuven JAG, Princen
HMG. Lipoprotein (a): relation to other risk factors and genetic
heritability. Results from a Dutch parent-twin study. Atherosclero-
sis. 1993;99:23-33.
7. Hornstra G, van Houwelingen AC, Kester ADM, Sundram K. A
palm oil-enriched diet lowers serum lipoprotein (a) in normocho-
lesterolemic volunteers. Atherosclerosis. 1991;90:91-93.
8. Beil FU, Terres W, Orgass M, Greten H. Dietary fish oil lowers
lipoprotein (a) in primary hypertriglyceridemia. Atherosclerosis.
1991;90:95-97.
9. Mensink RP, Zock PL, Katan MB, Hornstra G. Effect of dietary cis
and trans fatty acids on serum lipoprotein (a) levels in humans.
J Lipid Res. 1992;33:1493-1501.
10. Nestel P, Noakes M, Belling B, McArthur R, Clifton P, Janus E,
Abbey M. Plasma lipoprotein lipid and Lp(a) changes with substi-
tution of elaidic acid for oleic acid in the diet. / Lipid Res. 1992;
33:1029-1036.
11. Almendingen K, Jordal O, Kierulf P, Sandstad B, Penersen JI.
Effects of partially hydrogenated fish oil, partially hydrogenated
soybean oil, and butter on serum lipoproteins and Lp(a) in men.
J Lipid Res. 1995;36:1370-1384.
12. Carlson LA, Hamsten A, Asplund A. Pronounced lowering of
serum levels of lipoprotein Lp(a) in hyperlipidaemic subjects
treated with nicotinic acid. / Intern Med. 1989;226:271-276.
13. Lichtenstein AH, Ausman LM, Carrasco W, Jenner JL, Ordovas
JM, Schaefer EJ. Hydrogenation impairs the hypolipidemic effect
of corn oil in humans. Hydrogenation, trans fatty acids, and plasma
lipids. Arterioscler Thromb. 1993;13:154-161.
14. Mensink RP, Katan MB. Effect of dietary trans fatty acids on
high-density and low-density lipoprotein cholesterol levels in
healthy subjects. N Engl J Med. 1990;323:439-445.
15. Zock PL, Katan MB. Hydrogenation alternatives: effects of trans
fatty acids and stearic acid versus linoleic acid on serum lipids and
lipoproteins in humans. J Lipid Res. 1992;33:399-410.
1661
16. Wood R, Kubena K, O'Brien B, Tseng S, Martin G. Effect of butter,
mono- and polyunsaturated fatty acid-enriched butter, trans fatty
acid margarine, and zero trans fatty acid margarine on serum lipids
and lipoproteins in healthy men. J Lipid Res. 1993;34:1-11.
17. Judd JT, Clevidence BA, Muesing RA, Wittes J, Sunkin ME,
Podczasy JJ. Dietary trans fatty acids: effect on plasma lipids and
lipoproteins of healthy men and women. Am J Clin Nutr. 1994;59:
861-868.
18. Bonanome A, Grundy SM. Effect of dietary stearic acid on plasma
cholesterol and lipoprotein levels. N Engl J Med. 1988;318:
1244-1248.
19. Denke MA, Grundy SM. Effects of fats high in stearic acid on lipid
and lipoprotein concentrations in men. Am J Clin Nutr. 1991;54:
1036-1040.
20. Hegsted DM, McGandy RB, Myers ML, Stare FJ. Quantitative
effects of dietary fat on serum cholesterol in man. Am J Clin Nutr.
1965;17:281-295.
21. Helrich K, ed. Official Methods of Analysis of the Association of
Official Analytical Chemists, 15th ed. Methods 955.046, 979.09 (pro-
tein); 926.08, 925.09 (moisture); 922.06, 954.02 (fat); 923.03 (ash).
Arlington, VA: AOAC, 1990.
22. Helrich K, ed. Official Methods of Analysis of the Association of
Official Analytical Chemists, 15th ed. Method 985.29 (total dietary
fiber in foods). Arlington, VA: AOAC, 1990.
23. Madison BL, DePalma RA, D'Alonzo RP. Accurate determination
of trans isomers in shortenings and edible oils by infrared spectro-
photometry. / Am Oil Chem Soc. 1982;59:178-191.
24. Walker RC. Method Ce lc-89. In: Firestone D, ed. Official Methods
and Recommended Practices of the American Oil Chemists' Society,
4th ed. Champaign, IL: American Oil Chemists Society, 1989.
25. Hainline A, Karon J, Lippel K, eds. Manual of laboratory operations,
Lipids Research Clinics Program. Lipid and lipoprotein analysis, 2nd
ed. Bethesda, MD: National Heart Lung and Blood Institute, NIH,
DHHS, 1982.
26. Gidez LI, Miller GJ, Burstein M, Slagle S, Eder HA. Separation
and quantitation of subclasses of human plasma high density
lipoproteins by a simple precipitation procedure. J Lipid Res. 1982;
23:1206-1223.
27. Friedewald WT, Levy RI, Fredrickson DS. Estimation of the con-
centration of low-density lipoprotein cholesterol in plasma, without
use of the preparative ultracentrifuge. Clin Chem. 1972;18:499-502.
28. Kostner GM, Avogaro P, Cazzolato G, Marth E, Bittolo-Bon G,
Qunici GB. Lipoprotein Lp(a) and the risk for myocardial
infarction. Atherosclerosis. 1981;38:51-61.
29. Armstrong VW, Cremer P, Eberle E, Manke A, Schulze F, Wicland
H, Kreuzer H, Seidel D. The association between serum Lp(a)
concentrations and angiographically assessed coronary atheroscle-
rosis. Dependence on serum LDL levels. Atherosclerosis. 1986;62:
249-257.
30. Dahlen GH, Guyton JR, Attar M, Farmer JA, Kautz JA, Gotto
AM. Association of levels of lipoprotein Lp(a), plasma lipids, and
other lipoproteins with coronary artery disease documented by
angiography. Circulation. 1986;74:758-765.
31. Emken EA. Trans fatty acids and coronary heart disease risk:
physiochemical properties, intake, and metabolism. Am J Clin Nutr.
1995;62:659S-669S.
32. Hunter JE, Applewhite TH. Reassessment of trans fatty acid avail-
ability in the U.S. diet. Am J Clin Nutr. 1991;54:363-369.
33. Enig MG, Atal S, Keeney M, Sampugna J. Isomeric trans fatty acids
in the U.S. diet. J Am Coll Nutr. 1990;9:471-486.
34. Guyton JR, Dahlen GH, Patsch W, Kautz JA, Gotto AM. Rela-
tionship of plasma lipoprotein Lp(a) levels to race and to apoli-
poprotein B. Arteriosclerosis. 1985;5:265-272.
35. Schreiner PJ, Morrisett JD, Sharrett AR, Patsch W, Tyroler HA,
Wu K, Heiss G. Lipoprotein (a) as a risk factor for preclinical
atherosclerosis. Arterioscler Thromb. 1993;13:826-833.
36. De Pergola G, Giorgino F, Cospite MR, Giagulli VA, Cignarelli M,
Ferri G, Giorgino R. Relation between sex hormones and serum
lipoprotein and lipoprotein (a) concentrations in premenopausal
obese women. Arterioscler Thromb. 1993;13:675-679.
37. Krempler F, Kostner GM, Bolzano K, Sandhofer F. Turnover of
lipoprotein (a) in man. J Clin Invest. 1980;65:1483-1490.
38. Rader DJ, Cain W, Zech LA, Usher D, Brewer HB. Variation in
lipoprotein (a) concentrations among individuals with the same
apolipoprotein (a) isoform is determined by the rate of lipoprotein
(a) production. J Clin Invest. 1993;91:443-447.