European Journal of Clinical Nutrition (2001) 55, 748754

2001 Nature Publishing Group All rights reserved 09543007/01 $15.00

www.nature.com/ejcn

Original Communication
Effects of palm oil and dietary cholesterol on plasma lipop
roteins:
results from a dietary crossover trial in free-living
subjects
1,2 2
L E Bautista *, OF Herran and C Serrano
1
Instituto
Colombiano de Investigacion Biomedica, Universidad Industrial de Santander, Bucanamanga, Columbia; and
2
Fundacion
Cardiovascular del Oriente Colombiano, Bucanamanga, Columbia

Objective: To study the effects of palm oil (PO) and egg consumption (E) on plasma lipoproteins.
Design: Randomized crossover trial.
Setting: Free-living subjects.
Subjects: Twenty-eight healthy male students aged 20 34 y.
Interventions: Four typical Colombian diets (10 878 kJ=day; 57% energy in carbohydrates, 12% energ
y in
proteins and 31% energy in fats) were consumed for 4 weeks. The HPOLC diet was high in PO (8.8% energy
as
palmitic acid, PA) and low in eggs (181.2 mg=kJ of dietary cholesterol, DC); the HPOHC diet was high in PO a
nd
high in eggs (866.1 mg=kJ of DC); the MPOMC diet was moderate in PO (6.3% energy as PA) and moderate i
n
eggs (581.6 mg=kJ of DC); and the LOPOMC diet had no PO and was moderate in eggs (543.9 mg=kJ of DC)
.
Main outcome: Total (TC), low density (LDL-c), and high density lipoprotein cholesterol (HDL-c), a
nd
triacylglycerols (TAG) were measured on a pool of three fasting blood samples collected in consecutive
days
the last week of each diet.
Results: Comparison of the HPOHC and HPOLC diets showed increases in TC and LDL-c of 0.21 (P 0.01),
and
0.16 mmol=l (P 0.05). Comparison of LOPOMC and MPOMC diets showed increases in TC and LDL-c of 0.39
(P < 0.001), and 0.38 mmol=l (P < 0.001), respectively. No signicant changes in HDL-c or TAG were
observed.
Conclusions: Our ndings suggest that non-extreme short-term changes in PO and DC consumption lead
to
signicant elevations in plasma TC and LDL-c.
Sponsorship: CENIPALMA, Fundacion Cardiovascular del Oriente Colombiano, Universidad Industrial
de
Santander.
Descriptors: dietary fats; dietary cholesterol; palm oil; lipoproteins; cross-over studies
European Journal of Clinical Nutrition (2001) 55, 748754

Introduction may be partially due to changes in popul

ation dietary
During the last three decades, death rates from co
ronary *Correspondence: L E Bautista, Uniformed Services Univers
heart disease (CHD) have tripled and cardiovascula
ity of the
r dis- Health Sciences (USUHS), PMB Division of Epi and Biost
eases have become the main cause of death in the C at, Room
olom- A1039, 4301 Jones Bridge Road, Bethesda, MD 20814-
bian population (Pabon Rodriguez, 1993). These ch 4799, USA.
anges E-mail: lbautista@usuhs.mil
Guarantor: L E Bautista.
 Contributors: L EB was responsible for study design, supervision of patterns. In this period, palm oil has become th
the e main
running of the study, statistical analysis, and writing of the paper. source of edible fats contributing up to 80% of
OFH all fat
designed the diets used in the study, supervised food preparation a
nd consumed by this population. Also, it is well know
consumption, and participated in the writing of the paper. CS was i n that
n elevated plasma total cholesterol (TC), low density
charge of laboratory analysis and data quality, and participated in t lipo-
he
writing of the paper.
protein cholesterol (LDL-c) and triacylglycerols
Received 24 April 2000; revised 5 February 2001; (TAG), as
accepted 5 February 2001 well as low levels of high density lipoprotein chole
sterol
(HDL-c) are risk factors for CHD and that th
ey are
inuenced and can be modied by the type and
amount
of dietary fats (Ascherio & Willett, 1995).
Palm oil is the most important edible source of pal
mitic
acid (C16:0), one of the main saturated fatty acids (
SFAs)
in the human diet. Several studies have shown that
dietary
palmitic acid (DPA) has a hypercholesterolemic
effect,
particularly when compared with polyunsaturated
fatty
acids (PUFAs; Keys et al, 1957; Keys et al, 1
965a,c;
Laine et al, 1982; Mattson & Grundy, 1985; Anony
mous,
1990; Ng et al, 1991; Denke & Grundy, 1992).
Never-
theless, these studies have been questioned (Reiser,
1973;
Anonymous, 1987, 1990) because they were too shor
t, had
1983), while it is not important in low-
cholesterol
(100 mg=day) diets (Wolf & Grundy, 1983). The fact t
small sample size (Laine et al, 1982; Mattson & G hat
rundy, the hypercholesterolemic effects of SFAs could depend on
1985; Ng et al, 1991; Denke & Grundy, 1992), diet composition makes it difcult to extrapolate stu
and the dy
diets used had an unusually high fat content (Keys results among populations with different types of die
et al, ts.
1965b; Laine et al, 1982; Mattson & Grundy, 1985; Moreover, other studies (Baudet et al, 1984; Marzuki et a
Denke l,
& Grundy, 1992) and cholesterol (Laine et al, 1991; Hornstra et al, 1991; Sundram et al, 1992)
1982; have
Mattson & Grundy, 1985; Ng et al, 1991; De failed to show a hypercholesterolemic effect of DPA when
nke & compared to PUFAs such as linoleic acid. Due to t
Grundy, 1992). Also, there is evidence that the hype he
rchol- uncertainty regarding the hypercholesterolemic effects
esterolemic effect of SFAs depends on the amount of
of fat DPA, and the fact that those effects may depend on
and cholesterol in the diet (Laine et al, 1982; diet
Wolf & composition, we carried out an experimental study
Grundy, 1983; Baudet et al, 1984; Schonfeld et al, to
1982). ascertain the short-term effects of increased dietary
Some studies have shown that the effect of the de choles-
gree of terol (DC) and DPA consumption among normal subjects,
saturation of dietary fats is increased in high chol following their daily activities and taking their typical die
esterol t.
diets (National Diet Heart Study Research Group, We were particularly interested in the effect of high-DC
1968; as
Durrington et al, 1977; Laine et al, 1982; Wolf & Gr compared to a low-DC in the context of high-PA diet,
undy, as
well as in the effect of moderate-PA as compared to Effects of palm oil and dietary cholesterol
low- LE Bautista et al
PA in the context of a moderate-DC diet, because 749
these included fresh fruits, beans, green salad, rice, red me
patterns are typical in the studied population. at,
poultry, sh, yucca root, and plantain. Four different diets
were used, all of them with 2600 calories=day, 57% energ
y
Methods from carbohydrates, 12% energy from proteins, and 31%
energy from fats. The high palm oil low cholesterol diet
We carried out a double-blind 44 dietary crossover (HPOLC) was high in DPA (8.8% energy) and low in DC
trial. (181.2 mg=kJ). The high palm oil high cholesterol die
Twenty-eight male university students (20 to 34- t
y-old) (HPOHC) was high in DPA (8.8% energy) and high in DC
were randomly assigned to the four diet sequences (866.1 mg=kJ). The moderate palm oil moderate choles
(seven -
students per diet group) of a randomly selecte terol diet (MPOMC) had a moderate content of DPA (6.3%
d 44 energy) and DC (581.6 mg=kJ). Finally, the low palm oil
standard Latin square. A standard square was moderate cholesterol diet (LPOMC) had virtually no DPA
deemed and a moderate amount of DC (543.9 mg=kJ). Changes i
appropriate because the duration of each diet made n
carry- DC were mostly due to the addition of one egg=day to the
over effects and diet sequence irrelevant. All the stdiet and changes in DPA were due to replacement
udents of
were living in a university residence and consume sunower oil (5.1% DPA; 3.5% other SFAs; 57.7% oleic
d their acid; 15.6% linoleic acid; and 0.9% other non-SFAs) b
three meals per day at the university cafeteria, includi y
ng the palm oil (43.8% DPA; 6.2% other SFAs, 38.9% oleic acid;
weekend. None had any evidence of physical or psyc 10.6% linoleic acid; and 0.4% other non-SFAs).
hiatric The four groups of students received the diets in th
disease and all had normal blood pressure, fasting gly e
cemia following sequence. Group 1, HPOHC, MPOMC, LPOMC,
and blood coagulation tests. The study was approved HPOLC; Group 2, MPOMC, LPOMC, HPOLC, HPOHC;
by the Group 3, LPOMC, HPOLC, HPOHC, MPOMC; and Group
Research Ethics Committee of the Universidad Ind 4, HPOLC, HPOHC, MPOMC, LPOMC. Each diet was
ustrial consumed for 4 weeks and foods consumed outside th
de Santander, School of Medicine, and informed c e
onsent study were recorded by the participants and its nutrie
was obtained from all participants. nt
Diets were composed of solid food, prepared i content was estimated using Colombian food composition
n the tables (Instituto Colombiano de Bienestar Familiar (ICBF),
university kitchen by trained personnel, and consume 1992; Quintero et al, 1990). A daily diet compliance record
d ad- was kept by the staff of the study. All participants wer
libidum on site. Breakfast included fresh fruits, whole e
milk instructed to continue with their regular activities through-
drinks, cereal, eggs and cheese, while lunch and out the trial and they had no knowledge of the specic diet
dinner being consumed at any study period.
During the last week of each diet period each subjec
t
was weighted and his blood pressure was measured. Also,
three 12 h fasting venous blood samples were drawn fro
m
each subject in consecutive days of the last week of eac
h
diet period. Blood samples were processed and plasma was
stored at 720 C. Plasma lipids measurements were carried
out in a pool of the three plasma samples, in order to reduc
e
day-to-day variability in cholesterol levels. Plasma lipo-
protein measurements followed the standard protocol of the
Lipid Research Clinics Program (Lipid Research Clinics
Program, 1974). To reduce laboratory-related variability,
blood samples were analyzed in batches and measurements
were repeated in a 10% random sample of all participants,
for quality control. Lipid measurements were carried out
without knowledge of the type of diet the subject wa
s
eating. evaluate their statistical signicance. Multiple linea
The effects of the experimental diets on lipid concen r
tra- regression was used to explore possible association
tion were estimated by using the methodology of s
basic between participants baseline characteristics and response
contrasts suggested by Senn (1993). Mean average cha to each experimental diet. To assess the effects of t
nges he
in plasma lipids concentration were calculated and pair
ed t- European Journal of Clinical Nutrition
test and analysis of variance (ANOVA) were us
ed to
Effects of palm oil and dietary cholesterol
LE Bautista et al
750
controlled diets on the variability of the change in missing. Plasma lipid measurements were carried ou
lipids t in a
concentration, we calculated the coefcient of variati pool of three blood samples 80% of the time, in a p
on for ool of
each basic contrast. Pearson correlation coefcients two samples 17% of the time, and in one sample 3%
were of the
used to quantify the potential association betwe time.
en the Changes in plasma lipoproteins due to changes i
response to DPA and the response to DC. n DC
were estimated by contrasting lipoprotein levels at th
e end
Results of the HPOHC diet and the HPOLC diet. Sim
ilarly,
Twenty-eight male subjects were enrolled in the changes in lipoprotein prole associated with DPA
study were
(mean age 24.6 y; 95% CI, 23.5, 25.8). Only one estimated by comparing plasma lipoprotein levels
of them at the
had total cholesterol levels over 6.21 mmol=l, bu end of the LPOMC diet and the MPOMC diet. There w
t eight as a
(28.6%) had cholesterol levels between 5.17 and non signicant average reduction of 1.1 kg in body w
6.21 mmol=l. Mean TC, LDL-c, HDL-c and TAG eight
were during the study period (P 0.09).
4.80 mmol=l (95% CI, 4.50, 5.11), 3.35 (95% C The increase in DC resulted in a signicant rise
I, 3.09, in TC
3.62), 0.93 (95% CI, 0.86, 1.01), and 1.12 (95% CI and LDL-c: 0.21 mmol=l (P 0.009) and 0.16
, 0.96, mmol=l
1.29) mmol=l, respectively. After randomization, ba (P 0.047), respectively (Table 2). However, the
seline high-
TC, LDL-c and TAG were higher in Groups 1 and 2 cholesterol diet did not result in signicant increa
than in ses in
Groups 3 and 4 (Table 1). However, these differences HDL-c nor in TAG: 0.03 mmol=l (P
were 0.095) and
not statistically signicant and did not compromi 0.04 mmol=l (P 0.390), respectively. Overall, chan
se the ges in
validity of our results because each subject acted as h TC associated with DC intake showed a large varia
is own bility,
control. There were no drop-outs and participants ranging from 70.41 to 1.19 mmol=l (Figure 1).
con- Nine
sumed 90% of all meals provided in the study. (33.3%) subjects showed no increase or even a d
Foods ecrease
consumed most frequently outside the study diets in TC, while 10 (37.0%) showed increases greate
were r than
bottled drinks, bread and sweets. Diet complianc 0.26 mmol=l. The standard deviation of the me
e was an TC
similar in the four diet-sequence groups and for all change was 0.39 mmol=l and the coefcient of v
diets. ariation
Only one out of 140 blood lipids measurement was 184.1%. The variability in LDL-c change was
s was similar
to that in TC. LDL-c change varied from
70.62 to
1.24 mmol=l, the standard deviation of the mean
change
 Table 1 Baseline participants characteristics after random assignment to the four diet sequences
a
Diet sequence groups

Characteristic All 1 2 3 4 P-value

Age (y) 24.6 25.3 25.6 23.4 24.3 0.55

Weight (kg) 62.0 63.4 63.3 58.9 63.0 0.60
Total cholesterol (mmol=l) 4.80 5.04 5.07 4.57 4.53 0.56
HDL cholesterol (mmol=l) 0.93 0.93 1.01 0.92 0.87 0.61
LDL cholesterol (mmol=l) 3.35 3.59 3.45 3.15 3.22 0.63
Triacylglycerols (mmol=l) 1.12 1.13 1.31 1.10 0.94 0.53
a
Group 1: (high palm oil high cholesterol), (moderate palm oil moderate cholesterol), (low palm oil moderate
cholesterol), and (high palm oil low cholesterol).
Group 2: (moderate palm oil moderate cholesterol), (low palm oil moderate cholesterol), (high palm oil low
cholesterol), and (high palm oil high cholesterol).
Group 3: (low palm oil moderate cholesterol), (high palm oil low cholesterol), (high palm oil high cholesterol),
and (moderate palm oil moderate cholesterol).
Group 4: (high palm oil low cholesterol), (moderate palm oil high cholesterol), (moderate palm oil moderate
cholesterol), and (low palm oil moderate cholesterol).

Table 2 Changes in plasma lipids associated with changes in dietary cholesterol and palmitic acid content in the diet
Dietary contrast Plasma lipid Change (mmol=l) 95% condence interval P-value

High vs low cholesterol Total cholesterol 0.21 (0.07, 0.36) 0.009

LDL cholesterol 0.16 (0.002, 0.32) 0.047
HDL cholesterol 0.03 (70.006, 0.08) 0.095
Triacylglycerols 0.04 (70.06, 0.15) 0.390
High vs low palm oil Total cholesterol 0.39 (0.19, 0.59) < 0.001
LDL cholesterol 0.38 (0.21, 0.55) < 0.001
HDL cholesterol 0.03 (70.006, 0.07) 0.097
Triacylglycerols 70.05 (70.002, 0.07) 0.405

European Journal of Clinical Nutrition

Effects of palm oil and dietary cholesterol
LE Bautista et al
751

Figure 1 Changes in total cholesterol associated with a high col

esterol was 0.40 mmol=l and the coefcient of variation
diet.
was
249.7%.
 Consumption of the high-DPA diet resulted in Figure 2 Changes in total cholesterol associated wity a high palm oil
diet.
increases
of 0.39 mmol=l in TC (P < 0.001) and 0.38 m
mol=l in TC had no signicant inuence (P 0.11) on the effect of
LDL-c (P < 0.001). Neither HDL-c nor TAG wer DC on TC (Figure 1).
e sig-
nicantly affected by the high DPA diet (Table 2)
. The Discussion
response to DPA was very variable, ranging from 70.
49 to Our results show that non-extreme short term changes i
1.50 mmol=l. Two participants (7.1%) lowered their T n
C by DPA and DC consumption lead to signicant elevations in
0.21 mmol=l or more, 11 (39.3%) had changes plasma TC and LDL-c. On the other hand, plasma HDL-c
between and TAG changed very little in response to DPA or DC.
70.23 and 0.23 mmol=l and 15 (53.6%) had in These results are consistent with results from previou
crease s
! 0.26 mmol=l (Figure 2). The standard deviation studies (Keys et al 1957, 1965a c; Hegsted et al, 19
of TC 65;
changes was 0.51 mmol=l and the coefcient of v Roberts et al, 1981; Tan et al, 1980; Sacks et al, 19
ariation 84;
was 129.7%. Results for LDL-c were similar as McMurry et al, 1982; Mattson & Grundy, 1985; Lain
those e
for TC. et al, 1982; Anonymous, 1990; Ng et al, 1991; Denke
DC and DPA effects on blood lipoprotein prole &
were Grundy, 1992), even though our study was conducted with
independent of diet sequence (P 0.24). Howev typical diets, among free-living subjects who continued
er, the their regular daily activities.
subjects response to DC tend to be correlated We should keep in mind that the observed effect
to the s
response to DPA, as shown by a Pearson corr occurred in young males taking a diet with 31% energ
elation y
coefcient of 0.46 (P 0.02). Similar results were coming from fat. The rise in TC associated to increased DC
found intake was similar to the expected change predicted using
for LDL-c (correlation coefcient 0.41, P Keys formula (Keys et al, 1965a). Similarly, TC changes
0.03). in response to DPA were consistent with the expecte
Finally, age, sex, and initial weight did not mod d
ify the change according to Keys studies (Keys et al, 1957).
effect of a higher consumption of DC or DPA on pl Not all study subjects experienced an increase in plasma
asma lipids when challenged with DC or DPA. The existence of
lipids. Nevertheless, subjects with higher baseline c
holes- European Journal of Clinical Nutrition
terol levels were more responsive to the high DP
A diet
(Figure 2). Each increment of one mg=dl of baseli
ne TC
resulted in an additional change of 0.006 mmol=l
in the
effect of DPA on TC (P 0.048). On the contrary, ba
seline
Effects of palm oil and dietary cholesterol not a xed characteristic of the subject (Kata
LE Bautista et al n & Beynen,
752 1992). Although the accurate identication
hypo- and hyper-responders to DC seems to be of hypo- and
individually hyper-responders to DC and DPA is of practical
determined, but the classication of subjects in either interest for
group dietary advice, no feasible test is currently
compromises the diet-independent within-person available to
variability accomplish this task.
of serum cholesterol (Katan & Beynen, 1992). On the Moreover, we found that subjects who hav
other e an accen-
hand, it is known that hypo- and hyper- tuated response to DC tend to be also hyper-
responders to responsive to
dietary fats do exist, but total insensitivity of DPA. The correlation between DC and DP
serum A responses
cholesterol is usually due to random uctuations an observed in our study (r 0.46, P 0.02)
d it is was similar to
 that reported in previous experimental studies (r compared with polyunsaturated fatty acids.
0.50, More recently published studies have faile
n 23; P < 0.05, Katan et al, 1988; Katan & d to nd a
Beynen, hypercholesterolemic effect of PO (de Bosh
1992). Although Katan & Beynen (1987) have re et al, 1996;
ported Choudhury et al, 1995; Ghafoorunissa et al, 1
that subjects who were hyper-responsive to D 995; Mutalib
C had et al, 1999). Results from de Bosch et al
higher levels of TC than subjects who were (1996) can be
hypo- questioned because her experiments were not
responders, we did not nd such a relationship randomized,
in our no other measure for control of potential confo
study. However, TC response to DPA was related to unders was
base-
line TC levels. This suggests that subjects with hyp European Journal of Clinical Nutrition
ercho-
lesterolemia are even more sensible to DPA than thos
e who
are normocholesterolemic.
Changes due to DC and DPA were very variable, b
ut the
response to DPA was less so than the response to DC.
This is
consistent with the fact that the consumption of fatty
acids
explains most of the variability in plasma cholesterol l
evels
(Hegsted et al, 1965). The exact mechanism by which
SFAs
increment plasma cholesterol has not been establish, b
ut there
is some evidence that this effect is mediated by suppre
ssion of
the LDL receptor (Grundy & Vega, 1990).
Our results contradict those of other studies (B
audet
et al, 1984; Marzuki et al, 1991; Hornstra et al
, 1991;
Sundram et al, 1992) that reported a neutral DPA
effect.
The absence of a DPA effect in Marzukis study (Mar
zuki
et al, 1991) could be due to random error because on
ly one
blood sample was used to measure plasma chol
esterol.
Also, the palm oil used by Baudet et al (1984) had
a low
concentration of palmitic acid (22%) and a high conc
entra-
tion of oleic acid (58%), which could result in lower pl
asma
cholesterol levels due to the effect of the monounsatu
rated
fatty acid.
Moreover, in the studies of Hornstra et al (199
1) and
Sundram et al (1992), the control diet was the usual
diet of
the participants and was very similar to the experi
mental
diet in fatty acid proportional composition. Therefore,
their
results could not be considered as valid evidenc
e of a
hypocholesterolemic or even neutral effect of DPA
when
 questionable their conclusion. Their data actually s
howed
used, and no information regarding compliance nor data an increase in TC and LDL-c of 0.41 mmol=l in the
quality control are provided to the reader. Thus, her PO diet
conclusion that there are no adverse inuences of PO on group, and a pooled increase of 0.25 mmol=l in the c
plasma lipoprotein prole seems to be unfounded. ontrol
Similarly, in a crossover trial among free-living young group (calculated from Table 2 of Mutalib et al,
adults, Choudhury et al (1995) did not nd a hypercholes- 1999).
terolemic effect of palmolein when compared with olive Moreover, due to the large variability in the respo
oil. However, they replaced only half of the usual indivi- nse to
dualized dietary fat intake by palmolein or olive oil, dietary PO and the small sample size used, it is likel
corresponding to 17% of total dietary fat intake. Moreover, y that
study subjects prepared their own food and were instructed both groups were not comparable regarding factors rel
to exclude from their diets fat sources such as butter ated
, to the response to dietary PO. Furthermore, they had n
margarine, eggs and cooking oils, making the study diets o real
unrealistic. On the contrary, only one type of cooking oil data on the nutrient content of the control diet. Ass
(palm or sunower oil) was used in our study, and food uming
preparation was strictly standardized. The low percentage that the content of PO in the control diet was 3% of e
of total dietary fat intake change in the Choudhury study nergy
and the lack of direct control over other fat sources may depended on the unlikely fact that 10 non-
explain why palmolein consumption did not result in randomly
increased cholesterol levels. selected subjects were a representative sample
In another dietary crossover trial comparing groundnut of the
oil with PO, Ghafoorunissa et al (1995) found no increase Scottish population regarding characteristics of the di
in TC, LDL-c or HDL-c associated with consumption of the et.
PO diet. However, their study included only 12 subjects Since sources of potential error were tightly contro
and plasma lipoprotein measurements were based on a lled,
single blood sample taken at the end of each diet period. it is unlikely that our study results were biased. B
The small sample size and the use of a single blood sample ecause
resulted in large variability in the measurement of plasma diet sequences were randomly assigned and each
lipids and low power to detect between-diet differences. subject
For example, mean TC for the group consuming the PO acted as his own control, selection bias can be rul
diet was 155 mg=dl, but the 95% condence interval was ed out.
very wide, from 137.9 to 172.1 mg=dl (our calculation). Measurement errors were also unlikely because labor
Finally, based on the results of a parallel experimental atory
design, Mutalib et al (1999) reported that changes in TC tests were carried out without knowledge of the type o
and LDL-c associated to a diet with 26% energy from PO f diet
(15 subjects) were similar to those in a wild control group being consumed and were based on a pool of
of 10 subjects receiving their usual diet, with 3% energy multiple
from PO. However, design and interpretation issues make
have shown that a 1% reduction in TC yields a 2% reducti
on
in the long term risk of coronary heart disease (Lip
blood samples to reduce individual day-to-day ids
variability in Research Clinics Program, 1984; Stamler et al, 1986; La
plasma cholesterol levels. w
Although it is well known that changes in et al, 1994), the hypercholesterolemic effect of DPA (8
plasma %
cholesterol associated with changes in the diet are change) and DC (4% change) observed in our study coul
stable d
after the third week in the same diet, extrapolation result in a signicantly higher risk of atherosclerosis a
of our nd
results to long-term effects is compromised because cardiovascular diseases in this population. Since no suitabl
each e
diet lasted only one month. Another study limitation i tests to distinguish between responders and non-
s the responders
lack of measurements of other plasma lipid fractions to PA or DC are currently available, restriction in
such the
as lipoprotein(a) and HDL-c subfractions. consumption of palm oil and eggs in subjects at high ri
The effects of DPA and DC observed in this study sk
may of coronary heart disease may be advisable
have important implications in dietary counseling. Co
nsid- Acknowledgements We thank Dr Patricio Lopez for reviewing a
ering that experimental and observational cohort nd
studies helping to edit the nal manuscript.
 Effects of palm oil and dietary cholesterol
References LE Bautista et al

Anonymous (1987): New ndings on palm oil. Nutr. Rev. 45, 205 753
207. Instituto Colombiano de Bienestar Familiar (ICBF) (1992): Tabla de
Anonymous (1990): Saturated fatty acids in vegetable oils. Co composicion de alimentos colombianos. Santa Fe de Bogota: Ministerio
uncil on de Salud.
Scientic Affairs. J.A.M.A. 263, 693 695. Katan MB, Berns MAM, Glatz JFC, Knuiman JT, Nobels A & De Vries
Ascherio A & Willett W (1995): New directions in dietary stu JHM (1988): Congruence of individual responsiveness to dietary
dies of cholesterol and to saturated fats in humans. J. Lipid Res. 29, 883.
coronary heart disease. J. Nutr. 125, 647S 655S. Katan MB & Beynen AC (1987): Characteristics of human hypo- and
Baudet MF, Dachet C, Lasserre M, Esteva O & Jacotot B hyperresponders to dietary cholesterol. Am. J. Epidemiol. 125, 387 399.
(1984): Katan MB & Beynen AC (1992): Hyporesponders and hyperresponders to
Modication in the composition and metabolic properties of changes in diet. In Cholesterol and Coronary Heart Disease. The Great
human Debate. Proceedings of an International Conference, ed. P Gold,
low density and high density lipoproteins by different dieta Grover S, DAK Roncari, pp 177 187. Park Ridge, NJ: Parthenon.
ry fats. Publishing Group.
J. Lipid Res. 25, 456 468. Keys A, Anderson JT & Grande F (1957): Prediction of serum cholesterol
Choudhury N, Truswell AS & McNeil Y (1995): Comparison of pl responses of men to changes in fats in the diet. Lancet ii, 959 966.
asma Keys A, Anderson JT & Grande F (1965a): Serum cholesterol response to
lipids and vitamin E in young and middle-aged subjects on potato changes in the diet. I. Iodine values of dietary fat versus 2S-
crisps P.
fried in palmolein and highly oleic sunower oil. Am. J. Clin. Nut Metabolism 14, 747 758.
r. 61, Keys A, Anderson JT & Grande F (1965b): Serum cholesterol response to
1043 1051. changes in the diet. II. The effect of cholesterol in the diet. Metabolism
de Bosch NB, Bosch V & Apitz R (1996): Dietary fatty acids in 14, 759 765.
athero- Keys A, Anderson JT & Grande F (1965c): Serum cholesterol response to
thrombogenesis: inuence of palm oil ingestion. Haemostasis 26, changes in the diet. IV. Particular saturated fatty acids in the die
46 t.
54. Metabolism 14, 776 787.
Denke MA & Grundy SM (1992): Comparison of effects of lauric aci Laine DC, Snodgrass CM, Dawson EA, Ener MA, Kuba K, & Frantz ID Jr
d and (1982): Lightly hydrogenated soy oil versus other vegetable oils as
palmitic acid on plasma lipids and lipoproteins. Am. J. Clin. Nut a
r. 56, lipid-lowering dietary constituent. Am. J. Clin. Nutr. 35, 683 690.
895 898. Law MR, Wald NJ & Thompson SG (1994). By how much and how
Durrington PN, Bolton CH, Hartog M, Hartog M, Angelinetta R, Emm quickly does reduction in serum cholesterol concentration lower risk of
ett P ischemic heart disease? Br. Med. J. 308, 367 373.
& Furniss S (1977): The effect of a low-cholesterol, high- Lipids Research Clinics Program (1974): Lipid Research Clinics Program.
polyunsatu- Manual of Laboratory Operations. Bethesda, MD: National Institutes of
rate diet on serum lipids levels, apolipoprotein B levels and trygli Health.
ceride Lipids Research Clinics Program (1984): The Lipid Research Clinics
fatty acid composition. Atherosclerosis 27, 465 475. Coronary Primary Prevention Trial Results reduction in incidence
Ghafoorunissa, Reddy V & Sesikaran B (1995) Palmolein and grou of coronary heart disease. J.A.M.A. 251, 351 364.
ndnut Marzuki A, Arshad F, Razak TA & Jaarin K (1991): Inuence of dietary
oil have comparable effects on blood lipids and platelet aggregat fat on plasma lipid proles of Malaysian adolescents. Am. J. Clin. Nutr.
ion in 53, 1010S 1014S.
healthy Indian subjects. Lipids 30, 1163 1169. Mattson FH & Grundy SM (1985): Comparison of effects of dietary
Grundy SM & Vega GL (1990): Causes of high blood chol saturated, monounsaturated, and polyunsaturated fatty acids on plasma
esterol. lipids and lipoproteins in man. J. Lipid Res. 26, 194 202.
Circulation 81, 412 427. McMurry MP, Connor WE & Cerqueira MT (1982): Dietary cholesterol
Hegsted DM, McGandy RB, Myers ML & Stare FJ (1965): Quantit and the plasma lipids and lipoproteins in the Tarahumara Indians:
ative a
effects of dietary fat on serum cholesterol in man. Am. J. Clin. N people habituated to a low cholesterol diet after weaning. Am. J. Clin.
utr. 17, Nutr. 35, 741 744.
281 295. Mutalib MSA, Wahle KWJ, Duthie GG, Whiting P, Peace H & Jenkinson
Hornstra G, Van Houwelingen AC, Kester AD & Sundram K (1991): A (1999): The effect of dietary palm oil, hydrogenated rape and soya oil
A plam on indices of coronary heart disease risk in healthy Scottish volunteers.
oil-enriched diet lowers serum lipoprotein(a) in normocholestero Nutr. Res. 19, 335 348.
lemic National Diet Heart Study Research Group (1968): The national diet
volunteers. [Letter]. Atherosclerosis 90, 91 93. -
heart study nal report. Circulation 38(Suppl I), 1 428.
Ng TK, Hassan K, Lim JB, Lye MS & Ishak R (1991): Nonhypercholes-
terolemic effects of a palm-oil diet in Malaysian volunteers. Am. J. Clin.
Nutr. 53, 1015S 1020S.
Pabon Rodrguez A (1993): La mortalidad en Colombia 1953
1991:
Bogota: Instituto Nacional de Salud.
Quintero D, Alzate MC & Moreno S (1990): Tabla de composicion de
alimentos. Medellin: Centro de Atencion Nutricional.
Reiser R (1973): Saturated fat in the diet and serum cholesterol concentrati
on:
a critical examination of the literature. Am. J. Clin. Nutr. 26, 524 555.
Roberts SL, McMurry MP & Connor WE (1981): Does egg feeding (ie
dietary cholesterol) affects plasma cholesterol in man? The results of a
double blind study. Am. J. Clin. Nutr. 34, 2092 2099.
Sacks FM, Salazar J, Miller L, Foster JM, Sutherland M, Samonds KW,
Albers JJ & Kass EH (1984): Ingestion of egg raises plasma low density
lipoproteins in free-living subjects. Lancet i, 647 649.
Schonfeld G, Patsch W, Rudel LL , Nelson C, Epstein M & Olson RE
(1982): Effects of dietary cholesterol and fatty acids on plas
ma
lipoproteins. J. Clin. Invest. 69, 1072 1080.
 European Journal of Clinical Nutrition
Effects of palm oil and dietary cholesterol
LE Bautista et al
754
SennS(1993):Cross- Tan MH, Dickinson MA, Albers JJ, Havel RJ, Cheung MC & Vigne
overTrialsinClinicalResearch.NewYork:JohnWiley. JL
Stamler J, Wentworth D, Neaton JD (1986): Is the relationship betwe (1980): The effect of a high cholesterol and saturated fat
en serum diet on
cholesterol and risk of premature death from coronary heart serum high-density lipoprotein-cholesterol, apoprotein A-I, and
disease apopro-
continuous and graded? Findings in 356,222 primary screnees tein E levels in normolipidemic humans. Am. J. Clin. Nutr. 33, 2
of the 559
Multiple Risk Factor Intervention Trial (MRFIT). J.A.M.A. 256, 28 2565.
23 Wolf RN & Grundy SM (1983): Inuence of exchanging carbohydrat
2828. e for
Sundram K, Hornstra G, von Houwelingen AC & Kester AD ( saturated fatty acids on plasma lipids and lipoproteins in man. J.
1992): Nutr.
Replacement of dietary fat with palm oil: effect on human serum 113, 1521 1528.
lipids,
lipoproteins and apolipoproteins. Br. J. Nutr. 68, 677 692.
European Journal of Clinical Nutrition
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.