Atherosclerosis and Lipoproteins |
From the Department of Medicine, University of Helsinki, Helsinki, Finland.
Correspondence to Tatu A. Miettinen, MD, Department of Medicine, University of Helsinki, PO Box 340, FIN-00029 HYKS, Helsinki, Finland. E-mail tatu.a.miettinen{at}helsinki.fi
| Abstract |
|---|
|
|
|---|
Key Words: simvastatin cholesterol lowering cholestanol lathosterol sitosterol
| Introduction |
|---|
|
|
|---|
| Methods |
|---|
|
|
|---|
Analytical Methods
Serum squalene and sterols were measured by
GLC.3 16 Thus, total cholesterol, squalene,
and noncholesterol sterols, available for the entire
present study, were determined from the same GLC run. For GLC,
serum was saponified after addition of an internal standard
(5
-cholestane) followed by extraction of nonsaponified material and
conversion to trimethylsilyl derivatives. GLC analysis was
performed on a 50-m-long SE-30 column with use of an automated Hewlett
Packard instrument equipped with an automated peak calculator. Each run
quantified squalene, cholesterol, cholestanol, cholestenol,
desmosterol, lathosterol, campesterol, sitosterol, and avenasterol, as
noted in increasing order of GLC retention times. The avenasterol peak
contained trace amounts of sitostanol and a methylated precursor
sterol.
The noncholesterol sterols are transported in serum by lipoproteins, mainly by LDL.10 Thus, the decrease in concentration of LDL cholesterol by simvastatin also changes the concentrations of noncholesterol sterols. Therefore, we have expressed the noncholesterol sterols in terms of millimoles per mole of cholesterol, ie, as ratios of noncholesterol sterols to cholesterol, and expressed them in the text as noncholesterol sterol ratios. Ratios of precursor sterols to plant sterols (expressed only for lathosterol/campesterol) were also calculated because this variable reflects synthesis and absorption simultaneously.
Experimental Design
Two serum samples were analyzed from each patient before
randomization, and the mean value was calculated for the baseline
ratio. Other analyses were made at 6 weeks, 1 year, and 5 years
(indicates the end of the 5.4-year study) after baseline assessment
from a single serum sample each. The cholesterol value of
this method was 7.2% lower than the value of the central laboratory
that was used for the entire 4S study.
The patients were ranked by baseline cholestanol ratios into quartiles
(Table 1
). This division into
quartiles was used throughout the study for evaluation of the 6-week,
1-year, and 5-year cholesterol and other
noncholesterol sterol values and respective changes from
the baseline.
|
Statistical Analysis
Mean±SD or mean±SE values were calculated for each quartile.
The baseline cholestanol ratios were correlated with the ratios of
other variables at the baseline and at 1 and 5 years and with the
corresponding changes from the baseline values. The differences between
the quartiles, the changes from the baseline, and the differences
between the placebo and simvastatin groups in the 4
cholestanol quartiles were analyzed by ANOVA.
| Results |
|---|
|
|
|---|
Baseline ratios of plant sterols to cholesterol were insignificantly related to cholesterol concentrations, but those of cholesterol precursors exhibited weak negative respective correlations (eg, squalene r=-0.235, lathosterol r=-0.140; P<0.001 for both). Also, the ratios of precursor sterols to plant sterols were weakly negatively related to basal cholesterol levels (eg, lathosterol/campesterol r=-0.099; P<0.01); these findings suggest associations of the noncholesterol sterols with basal cholesterol concentration.
Simvastatin and Placebo Treatments
Body Weight
Body weight was increased by 0.34±0.03 kg (P<0.001)
in the whole population at 1 year and was similar in all cholestanol
quartiles. At 5 years, body weight increased in the survivors by
0.70±0.06 kg (P<0.001) from the baseline value, and it
increased gradually from the first to fourth cholestanol quartile by
0.50±0.12, 0.65±0.11, 0.68±0.14, and 0.97±0.13 kg, respectively,
without specific simvastatin effect.
Cholesterol Precursors
Correlations of the baseline ratios of cholestanol to
cholesterol with different lipid variables and their
changes during the study period are shown in Tables 2
and 3
,
respectively, for the simvastatin group. The correlations
remained significantly negative in the placebo group also (not shown)
throughout the study for each precursor sterol but not for squalene,
which was not changed by simvastatin. The higher the basal
ratios of cholestanol to cholesterol, the smaller were the
reductions of the ratios of precursor sterol to cholesterol
over the 5-year period (Table 3
); this finding was significant
in the placebo group only for lathosterol at 5 years
(r=-0.100). The absolute 6-week, 1-year, and 5-year changes
in the ratios of lathosterol to cholesterol are illustrated
in Figure 1
by the first and fourth basal
cholestanol quartiles (similar curves for the other precursors are not
shown). The second and third quartiles were mostly located between the
2 illustrated quartiles, as also shown by the correlations in Table 2
. The ratios in the placebo group tended to increase gradually
from the baseline with time, especially in the fourth quartile
(significantly so at 5 years).
|
|
|
The mean 6-week reductions of cholestenol, desmosterol, and lathosterol
by simvastatin were 33%, 21%, and 34% from the baseline,
but the respective reductions were markedly higher in the first (42%,
26%, and 40%) than in the fourth (24%, 18%, and 30%) quartile. The
reductions tended to decrease with treatment time, especially after the
first year (Figure 1
). However, the mean absolute reductions in
the ratios of the precursor sterols to cholesterol from the
respective placebo values were
2 times higher in the first than in
the fourth quartile over the 5-year treatment period (Figure 2
). The time-dependent increase of the
precursor sterol curves in Figure 1
was explainable by the
respective body mass index adjustment only in the placebo group for the
first quartile (P<0.05), but it was still significant for
other curves (P<0.01).
|
Cholestanol and Plant Sterols
The basal ratios of cholestanol to cholesterol
were positively related to those of plant sterols and cholestanol also
in the placebo group and to changes of plant sterols throughout the
study but were negatively related to their own changes (Tables 2
and 3
) in the placebo group throughout the study. In the placebo
group, the ratios of sitosterol and especially of cholestanol to
cholesterol were reduced from the basal values with time
significantly more in the fourth than in the first quartile, whereas
those of campesterol were similarly increased in all quartiles (Figure 3
).
|
Simvastatin increased the cholestanol ratios from the
baseline values significantly more in the first than in the fourth
quartile, whereas the highest plant sterol ratios and levels occurred
in the fourth quartile (Figures 3
and 4
). The increments from
the respective placebo values were higher for cholestanol in the first
than in the fourth quartile, whereas for the plant sterols, the
increments were higher in the fourth than in the first quartile. The
differences were virtually similar between the 2 quartiles during the
entire treatment period, and the difference even increased with time
for campesterol in the fourth quartile.
|
Serum Cholesterol
In the placebo group, cholesterol did not change
significantly during the 5-year period, and it was not related to the
baseline cholestanol ratios at any time point (Figure 5
).
|
Serum cholesterol concentration of the survivors, defined
by the first and fourth quartiles of basal cholestanol ratios, was
reduced by simvastatin more in the first than in the fourth
quartiles (Figure 5
) despite the significantly lower number of
patients (21 versus 36) on high doses of simvastatin. The
difference between the 2 quartiles was slight but statistically
significant by ANOVA (F=8.62, P=0.003); at 6 weeks, the
difference between the 2 quartiles was significant
(P<0.001), and the decrease of cholesterol was
higher in the first than in the fourth quartile (29.4±0.9% versus
25.6±0.9%, P<0.001). In addition, the baseline
cholestanol ratios were positively related to the
cholesterol concentrations (Table 2
) in the
simvastatin but not in the placebo group at 6 weeks
(r=0.195, P<0.001) and 5 years
(r=0.099, P=0.048) and to cholesterol
reduction at 6 weeks (Table 3
; r=0.127,
P<0.001).
| Discussion |
|---|
|
|
|---|
The baseline cholestanol ratio was highly significantly related to other noncholesterol sterols (positively to the plant sterols and negatively to the precursor sterols) in the coronary subjects during the entire study period of 5 years. Similar findings were observed earlier in a small cross-sectional group of randomly selected male subjects on a normal western diet.8 Thus, the subjects in the highest cholestanol quartile have the highest absorption efficiency and the lowest synthesis of cholesterol also among the coronary patients on a cholesterol-lowering diet.
The marked fall in the ratios of the cholesterol precursor
sterols can be ascribed to simvastatin-induced inhibition
of cholesterol synthesis, a finding reported earlier for
different statins in small groups of short-term
studies.7 18 19 20 21 22 The baseline cholestanol ratios and
quartiles predicted throughout the study the precursor sterol
reductions. The correlations in the simvastatin-treated
patients are most likely reflecting gradually decreasing
cholesterol synthesis in the patients from the lowest to
the highest cholestanol quartile. Thus, the markedly higher reduction
of the precursor sterol ratios in the first than in the fourth quartile
(Figure 1
) indicates lower inhibitory action of
simvastatin on cholesterol synthesis in the
highest than in the lowest cholestanol quartile.
Increased body weight increases cholesterol
synthesis.23 Thus, the slight gradual increase of body
weight, in proportion to the basal cholestanol quartiles, would be
expected to stimulate cholesterol synthesis during the
5-year period more in the highest than in the lowest quartiles of the
placebo and simvastatin groups. Accordingly, the
differences between the simvastatin and placebo curves of
the respective quartiles remained virtually unchanged throughout the
treatment period (Figure 2
).
Statin treatment reduces cholesterol absorption in patients
with familial
hypercholesterolemia,20 in
experimental animals on a high but not on a low cholesterol
diet,24 25 and in patients with no familial
hypercholesterolemia.18 The
increased ratios of cholestanol and plant sterols during the
simvastatin treatment have been interpreted in our earlier
statin studies18 19 20 26 to indicate reduced turnover of
cholesterol by decreased synthesis, resulting in reduced
biliary elimination of sterols, especially in the fourth quartile, with
highest sterol absorption. Reasons are not known for the different
responses of cholestanol ratios, especially in relation to campesterol
(Figure 3
), and for the reduction of cholestanol and sitosterol
in the placebo group during the 5-year period. It should be borne in
mind that the increase of body weight reduces the absorption efficiency
of cholesterol and serum plant sterol
ratios.27 28
The increased ratios of the plant sterols during the simvastatin treatment in the fourth quartile raise a question about their atherogenicity. Namely, phytosterolemia is strongly atherogenic,29 30 and high plant sterol levels have been suggested to provoke coronary heart disease even in nonsitosterolemic subjects.31 Not only the ratios but also the final 5-year campesterol and sitosterol concentrations were increased in the survivors despite reduced LDL cholesterol concentration, from the usual basal values of <1 mg/dL up to 2.8 and 1.5 mg/dL, respectively. Exchange of this type of sterol mixture with cholesterol of cell membranes or transfer, for instance in LDL, in the arterial walls of the patients may be functionally harmful, as evidenced by phytosterolemia. The atherogenic limit of plant sterol concentrations or of the ratios of plant sterols to cholesterol is not known, but in phytosterolemia, sitosterol concentrations, usually 10 to 30 mg/dL,32 are associated with coronary artery disease at a young age.
It is tempting to suggest that the high baseline absorption and low
synthesis of cholesterol results in a less favorable
cholesterol lowering in the highest cholestanol quartile.
In fact, the higher the basal cholestanol quartile, the lesser was the
fall in the serum cholesterol concentration and the higher
was the final treatment level of cholesterol even in
long-term simvastatin treatment of coronary
patients (Figure 5
). Despite markedly higher reduction of
cholesterol synthesis (lowered precursor ratios) by
simvastatin in the patients in the lowest versus the
highest cholestanol quartile, the respective difference in the serum
cholesterol level was markedly lesser. This
cholesterol difference alone could not be responsible for
the different clinical response13 ; something else, eg,
serum plant sterols, would have to be contributing. The change in
cholesterol synthesis is weakly related to serum
cholesterol lowering, as illustrated by sterol balance
studies showing virtually no change or slight reduction in the overall
synthesis of body cholesterol and a marked decrease in
serum cholesterol by statins.18 26 33 However,
in contrast to sterol balance data, the precursor sterol ratios usually
show marked decreases during statin treatment. On the other hand,
cholestyramine and stanol esters markedly increase
cholesterol synthesis, but they do not consistently
lower serum cholesterol.34 35
A highly significant negative correlation of the basal
cholestanol ratio with ratios of the cholesterol precursor
sterols and a positive correlation with ratios of the plant sterols for
up to 5 years indicated that the high basal cholestanol quartile still
predicted high absorption and low synthesis of cholesterol
even for the 5-year survivors in the simvastatin-treated
group. Thus, these patients with the high basal cholestanol ratios
should be treated with combined stimulation and inhibition of
cholesterol synthesis from the very beginning of
hypolipidemic treatment. For instance, starting treatment with resins
or plant stanol esters stimulates synthesis by respective malabsorption
of bile acids or cholesterol, followed by statin-induced
inhibition of synthesis in resistant cases.13 36
Replacement of normal dietary fat intake by the same amount of plant
stanol ester margarine during a chronic simvastatin
treatment actually revealed a more consistent (
10%) fall in
the serum cholesterol level of coronary patients
with a high versus a low basal cholestanol ratio.37
| Acknowledgments |
|---|
Received August 11, 1999; accepted November 4, 1999.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
S. Helske, T. Miettinen, H. Gylling, M. Mayranpaa, J. Lommi, H. Turto, K. Werkkala, M. Kupari, and P. T. Kovanen Accumulation of cholesterol precursors and plant sterols in human stenotic aortic valves J. Lipid Res., July 1, 2008; 49(7): 1511 - 1518. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. J. H. Jones Ingestion of Phytosterols is Not Potentially Hazardous J. Nutr., November 1, 2007; 137(11): 2485 - 2485. [Full Text] [PDF] |
||||
![]() |
H. P. Fransen, H. Verhagen, J. Plat, and R. P. Mensink Reply to Dr. Jones J. Nutr., November 1, 2007; 137(11): 2486 - 2486. [Full Text] [PDF] |
||||
![]() |
A. L. Garcia-Otin, M. Cofan, M. Junyent, D. Recalde, A. Cenarro, M. Pocovi, E. Ros, and F. Civeira Increased Intestinal Cholesterol Absorption in Autosomal Dominant Hypercholesterolemia and No Mutations in the Low-Density Lipoprotein Receptor or Apolipoprotein B Genes J. Clin. Endocrinol. Metab., September 1, 2007; 92(9): 3667 - 3673. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. P. Fransen, N. de Jong, M. Wolfs, H. Verhagen, W.M. M. Verschuren, D. Lutjohann, K. von Bergmann, J. Plat, and R. P. Mensink Customary Use of Plant Sterol and Plant Stanol Enriched Margarine Is Associated with Changes in Serum Plant Sterol and Stanol Concentrations in Humans J. Nutr., May 1, 2007; 137(5): 1301 - 1306. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Plat, I. Beugels, M. J. J. Gijbels, M. P. J. de Winther, and R. P. Mensink Plant sterol or stanol esters retard lesion formation in LDL receptor-deficient mice independent of changes in serum plant sterols J. Lipid Res., December 1, 2006; 47(12): 2762 - 2771. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. C. Ness, R. C. Holland, and D. Lopez Selective Compensatory Induction of Hepatic HMG-CoA Reductase in Response to Inhibition of Cholesterol Absorption. Experimental Biology and Medicine, May 1, 2006; 231(5): 559 - 565. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. P. Radermecker and A. J. Scheen Serum Plant Sterols and Atherosclerosis: Is There a Place for Statin-Ezetimibe Combination? J. Am. Coll. Cardiol., April 4, 2006; 47(7): 1496 - 1497. [Full Text] [PDF] |
||||
![]() |
T. A. Miettinen, M. Railo, M. Lepantalo, and H. Gylling Plant Sterols in Serum and in Atherosclerotic Plaques of Patients Undergoing Carotid Endarterectomy J. Am. Coll. Cardiol., June 7, 2005; 45(11): 1794 - 1801. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Plat, M. C. E. Bragt, and R. P. Mensink Common sequence variations in ABCG8 are related to plant sterol metabolism in healthy volunteers J. Lipid Res., January 1, 2005; 46(1): 68 - 75. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. R. Wilund, L. Yu, F. Xu, G. L. Vega, S. M. Grundy, J. C. Cohen, and H. H. Hobbs No Association Between Plasma Levels of Plant Sterols and Atherosclerosis in Mice and Men Arterioscler. Thromb. Vasc. Biol., December 1, 2004; 24(12): 2326 - 2332. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. R Thompson Management of dyslipidaemia Heart, August 1, 2004; 90(8): 949 - 955. [Full Text] [PDF] |
||||
![]() |
A. Rees Review: Ezetimibe -- clinical and pharmacological profile The British Journal of Diabetes & Vascular Disease, May 1, 2004; 4(3): 163 - 171. [Abstract] [PDF] |
||||
![]() |
G. C. Ness and K. R. Gertz Increased Sensitivity to Dietary Cholesterol in Diabetic and Hypothyroid Rats Associated with Low Levels of Hepatic HMG-CoA Reductase Expression Experimental Biology and Medicine, May 1, 2004; 229(5): 407 - 411. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. C. Ness and K. R. Gertz Hepatic HMG-CoA Reductase Expression and Resistance to Dietary Cholesterol Experimental Biology and Medicine, May 1, 2004; 229(5): 412 - 416. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. R. Matthan, A. Giovanni, E. J. Schaefer, B. G. Brown, and A. H. Lichtenstein Impact of simvastatin, niacin, and/or antioxidants on cholesterol metabolism in CAD patients with low HDL J. Lipid Res., April 1, 2003; 44(4): 800 - 806. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Nieminen, A.-M. Mustonen, P. Lindstrom-Seppa, V. Karkkainen, H. Mussalo-Rauhamaa, and J. V.K. Kukkonen Phytosterols Affect Endocrinology and Metabolism of the Field Vole (Microtus agrestis) Experimental Biology and Medicine, February 1, 2003; 228(2): 188 - 193. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. Gylling and T. A. Miettinen Inheritance of cholesterol metabolism of probands with high or low cholesterol absorption J. Lipid Res., September 1, 2002; 43(9): 1472 - 1476. [Abstract] [Full Text] [PDF] |
||||
![]() |
G.R. Thompson, F. O'Neill, and M. Seed Why some patients respond poorly to statins and how this might be remedied Eur. Heart J., February 1, 2002; 23(3): 200 - 206. [Full Text] [PDF] |
||||
![]() |
C. M. Ballantyne, A. G. Olsson, T. J. Cook, M. F. Mercuri, T. R. Pedersen, and J. Kjekshus Influence of Low High-Density Lipoprotein Cholesterol and Elevated Triglyceride on Coronary Heart Disease Events and Response to Simvastatin Therapy in 4S Circulation, December 18, 2001; 104(25): 3046 - 3051. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. H. O'Neill, D. D. Patel, B. L. Knight, C. K. Y. Neuwirth, M. Bourbon, A. K. Soutar, G. W. Taylor, G. R. Thompson, and R. P. Naoumova Determinants of Variable Response to Statin Treatment in Patients With Refractory Familial Hypercholesterolemia Arterioscler. Thromb. Vasc. Biol., May 1, 2001; 21(5): 832 - 837. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||