European Journal of Clinical Nutrition (2005) 59, 13741378

& 2005 Nature Publishing Group All rights reserved 0954-3007/05 $30.00
www.nature.com/ejcn

ORIGINAL COMMUNICATION
Rapeseed oil, olive oil, plant sterols, and cholesterol metabolism: an ileostomy study
L Ellegard1*, H Andersson1 and I Bosaeus1
1

teborg, Sahlgrenska University Hospital, Go teborg, Sweden Department of Clinical Nutrition, University of Go

Objective: To study whether olive oil and rapeseed oil have different effects on cholesterol metabolism. Design: Short-term experimental study, with controlled diets. Setting: Outpatients at a metabolic-ward kitchen. Subjects: A total of nine volunteers with conventional ileostomies. Interventions: Two 3-day diet periods; controlled diet including 75 g of rapeseed oil or olive oil. Main outcome measures: Cholesterol absorption, ileal excretion of cholesterol, and bile acids. Serum levels of cholesterol and bile acid metabolites. Differences between diets evaluated with Wilcoxons signed rank sum test. Results: Rapeseed oil diet contained 326 mg more plant sterols than the olive oil diet. Rapeseed oil tended to decrease cholesterol absorption by 11% (P 0.050), and increased excretion of cholesterol, bile acids, and their sum as sterols by 9% (P 0.021), 32% (P 0.038), and 51% (P 0.011) compared to olive oil. A serum marker for bile acid synthesis (7a-hydroxy-4cholesten-3-one) increased by 28% (P 0.038) within 10 h of consumption, and serum cholesterol levels decreased by 7% (P 0.024), whereas a serum marker for cholesterol synthesis (lathosterol) as well as serum levels of plant sterols remained unchanged. Conclusions: Rapeseed oil and olive oil have different effects on cholesterol metabolism. Rapeseed oil, tends to decrease cholesterol absorption, increases excretion of cholesterol and bile acids, increases serum marker of bile acid synthesis, and decreases serum levels of cholesterol compared to olive oil. This could in part be explained by different concentrations of natural plant sterols. Sponsorship: Supported by the Goteborg Medical Society, the Swedish Medical Society, the Swedish Board for Agricultural Research (SJFR) grant 50.0444/98 and by University of Goteborg.

European Journal of Clinical Nutrition (2005) 59, 13741378. doi:10.1038/sj.ejcn.1602249; published online 10 August 2005
Keywords: cholesterol absorption; sterol excretion; dietary fibre; dietary fat; plant sterols; ileostomy

Introduction
Cholesterol can only be eliminated from the body by faecal excretion, except for minor amounts of metabolites of steroid hormones in urine. Sterol balance studies have been performed to evaluate the dietary effects on sterol excretion (Connor et al, 1969; Grundy & Ahrens, 1969; Nestel et al, 1975). These metabolic ward investigations were elaborate,

*Correspondence: L Ellegard, Department of Clinical Nutrition, University of Goteborg, Sahlgrenska University Hospital, S-413 45 Goteborg, Sweden. E-mail: Lasse.Ellegard@sahlgrenska.se Guarantor: L Ellegard. Contributors: LE was the principal investigator, IB and HA contributed to the study design, and all contributed to writing this paper. Received 30 December 2004; revised 1 May 2005; accepted 14 June 2005; published online 10 August 2005

but due to bacterial degradation and varying transit time, precision was insufficient to settle the question of dietary effects on sterol excretion (Andersson & Bosaeus, 1993). We have readressed the question of dietary effects on sterol excretion by studying ileostomy subjects without resection of the small intestine (Sandberg et al, 1981). In ileostomy subjects, transit time is both shorter and less variable, and with far less bacterial degradation than in normal subjects (Sandberg et al, 1981). Thus, even short time studies in ileostomy subjects are feasible, to examine dietary effects on cholesterol metabolism. (Andersson & Bosaeus, 1993). With this approach, diets with the potential to reduce serum cholesterol levels have been shown to increase sterol excretion from the small bowel, either as bile acids or as cholesterol, within 24 days, in carefully controlled dietary studies. This has been shown with diets low in fat, or high

Monounsaturated oils and cholesterol metabolism L Ellegard et al

1375
in polyunsaturated fat (Bosaeus & Andersson, 1987), high in monounsaturated fat (Bosaeus et al, 1992), pectin (Bosaeus et al, 1986), and with oat-bran (Lia et al, 1995). Reducing cholesterol intake increases net cholesterol excretion as shown previously (Ellegard & Bosaeus, 1994). We have also reported increased cholesterol excretion from a prudent diet, that is, a diet in accordance with contemporary dietary guidelines of less saturated fat and more fibre-rich foods rd & Bosaeus, 1991). The combination of both (Ellega modifications reduces cholesterol absorption, and increases cholesterol and sterol excretion from the small intestine (Ellegard et al, 2000b). The decrease in cholesterol absorption was correlated to plant sterol excretion, implying dietary plant sterol content to be an important determinant of cholesterol metabolism, together with cholesterol itself (Ellegard et al, 2000b). Thus, we speculated whether plant sterol content may be as important as fatty acid composition and dietary fat content in the regulation of sterol balance in humans. In order to investigate this relationship, we compared the effects of two monounsaturated oils, rapeseed oil (high in plant sterols) and olive oil (low in plant sterols), of similar although not identical fatty acid composition, on cholesterol metabolism in subjects with conventional ileostomies. serum thyroxine levels. One subject was overweight, and had elevated serum glucose twice, and thus was diagnosed as diabetic during the study. His results did not differ from those of the other subjects. No subjects had any signs of anaemia, inflammation, hepatic, or renal disease as judged by history, hospital records, and standard laboratory tests, except the oldest subject who had slightly elevated CRPlevels because of arthritis. All subjects had stable bowel function at the time of the study. Informed consent was obtained from each participant. The study protocol was approved by the Ethics Committee of Sahlgrenska University Hospital.

Materials
Diets The diets were composed of common food items with identical dishes in the two diets. The compositions of the diets, given in Table 2, were calculated from Swedish food composition tables (Swedish National Food Administration, 1996) and from analyses of energy, plant sterols, and

Table 2

Composition of study diets Rapeseed Olive 8.770.05 38 7 24 5 35 20275 258724

Subjects, materials and methods

Subjects Nine persons (three female, six male) volunteered for the study. Clinical data on the subjects are presented in Table 1. All subjects had previously undergone proctocolectomy for ulcerative colitis (n 7) or Crohns disease (n 2) and had well-functioning ileostomies for at least 2 y. No additional small-bowel resections had been performed in any of the subjects. Seven of the nine subjects used some medication (listed in Table 1), but these medications were kept constant during the study. Subjects using thyroxine had normal
Energy* (MJ) Total fat (E%) Saturated fat (E%) Monounsaturated fat (E%) Polyunsaturated fat (E%) Dietary fibre (g) Cholesterol* (mg) Plant sterols* (mg)

9.070.1 38 5 20 11 35 19876 58477

Calculated from food tables (Swedish National Food Administration, 1996). *Analysed values, mean7s.e.

Table 1 Clinical data on ileostomy subjects Subject number 1 2 3 4 5 6 7 8 BMI (mmol/l) 24.6 22.5 23.4 25.1 21.7 27.9 29.3 21.3 Serum cholesterol (mmol/l) 6 4.9 4.2 5.1 5.7 6.2 5.7 4 Serum HDL (mmol/l) 1.2 1.4 1 1.5 2.6 1.9 1.1 1.7 Serum LDL (mmol/l) 4.4 2.9 2.7 3.1 2.8 3.7 3.8 2.0 Serum triglycerides (mmol/l) 1.0 1.4 1.1 1.1 0.7 1.3 1.7 0.6 ApoE phenotype 3/3 3/3 4/3 3/3 3/3 3/3 4/3 3/3 Serum CRP (mg/l) 8 o5 o5 o5 o5 o5 o5 o5 o5

Sex M F M M F M M F

Age (y) 80 59 44 55 63 63 47 37

Medication Ibuprofen, ACEI Thyroxine Metoprolol Thyroxine, spironolactone, PPI

9 Median Range

34 55 3480

27.5 24.6 21.329.3

5.5 5.5 4.47.8

1 1.4 12.6

3.6 3.1 24.4

2.0 1.1 0.62.0

4/3

Atropiumbromide, allopurinol Antacids, oestrogen, alimemazin thyroxine, DHEA

European Journal of Clinical Nutrition

Monounsaturated oils and cholesterol metabolism L Ellegard et al

1376
cholesterol. To meet energy demands, one subject (#7) had 50% larger than normal portions of all items on the menu. 7a-hydroxy-4-cholesten-3-one in serum was determined by HPLC (Axelson et al, 1988). The variation coefficient for duplicate samples was 7.4%.

Methods
Protocol Each subject was studied during two 3-day periods, on the same weekday on each diet. The first day of each period was used for adaptation and to minimise any carryover effects from the habitual diets of the participants. There was a washout period of at least 4 days between the dietary periods. The order of the diets was randomised for each subject, and the diets were blinded for the laboratory staff, the investigators and the subjects. On the second day, 8.7 kBq of beta-4-[14C]-sitosterol and 17.4 kBq of 1a, 2a-n-[3H]-cholesterol were added at breakfast, lunch, and dinner, in three divided doses, to measure cholesterol absorption. Isotopes were purchased from Amersham International, Buckinghamshire, UK. Isotope analyses Isotope activities were assayed by liquid scintillation counting in a Beckman Tricarb 1900TR as we have described rd previously (Ellega & Bosaeus, 1994). The variation coefficients for duplicate analyses were 4.5% for [3H]-cholesterol and 3.4% for [14C]-b -sitosterol.

Collection of ileostomy contents Ileostomy contents were collected during the last 2 days of each period. Ileostomy bags were changed every second hour during the daytime and directly after awakening. The bags were immediately frozen on dry ice in Dewar vessels which the subjects kept at home. Each morning, the bags were delivered to the metabolic ward and freeze dried before chemical analysis.

Calculations and statistical methods Values of wet weight, dry weight, cholesterol, and bile acids are reported as medians with ranges for each dietary period. Cholesterol excretion minus dietary cholesterol intake is reported as net cholesterol excretion. Total bile acid plus net cholesterol excretion is termed net sterol excretion, which would correspond to cholesterol synthesis under steady-state conditions. Fractional cholesterol absorption was calculated from isotope ratios in ileostomy effluent, as described previously rd (Ellega & Bosaeus, 1994). All descriptive values are given as medians for 24 h with ranges, unless otherwise stated. Differences between diets were assessed by Wilcoxons signed rank test, using the computer package SYSTATs version 7.0.1 for Windowss (SPSS Inc., 1997).

Chemical analyses Wet weight and dry weight of ileostomy effluent were determined as previously described (Sandberg et al, 1981). Energy of the diets was determined by combustion in a Gallenkamp bomb calorimeter (Loughborough, Leicestershire, UK). Determinations of cholesterol and plant sterols in the diets, as well as cholesterol, bile acids, and plant sterols in the ileostomy contents were made by GLC, as previously described (Bosaeus & Andersson, 1987). Variation coefficients in duplicate samples from food and ileostomy excreta were 3.6% for cholesterol and 3.4% for bile acids. Before starting the study, standard laboratory tests including total serum cholesterol, triglycerides, and apo E phenotype, were determined by current methods at the Central Laboratory for Clinical Chemistry, Sahlgrenska University Hospital. Serum samples for analyses specific to the study were collected on the second day of the diet period after overnight fast, before breakfast at 0800 hours and before supper at 1800 hours. Serum concentrations of lathosterol, cholesterol, and plant sterols (beta-sitosterol, stigmasterol and campesterol) were determined by GLC (Miettinen, 1982). Lathosterol was expressed in `mol/100 mmol choles terol as simultaneously derived during analysis. Variation coefficients in duplicate samples of sera were 4.4% for lathosterol, 3.6% for cholesterol, and 4.9% for plant sterols. European Journal of Clinical Nutrition

Results
Both diets were well tolerated and consumed without leftovers. Excretion data are summarised in Table 3. Rapeseed oil tended to decrease cholesterol absorption by median 11% (P 0.050) compared to olive oil. Consequently, net excretion of cholesterol rose by median 9% (P 0.021), bile acid excretion rose by 32% (P 0.038), and total net sterol excretion increased by median 51% (P 0.011) on rapeseed oil compared to olive oil.

Table 3 Sterol excretion and cholesterol absorption in nine ileostomy subjects on rapeseed oil and olive oil diets Rapeseed oil diet Wet weight (g) Dry weight (g) Plant sterols (mg) Net cholesterol (mg) Bile acids (mg) Net sterols (mg) Cholesterol absorption (%) 667 (575962) 83.0 (70.7104.3) 631 681 634 1460 50 (500937) (5401229) (174970) (8961863) (3159) Olive oil diet 656 (5891016) 81.3 (67.7105.4) 243 623 482 964 56 (206337) (346883) (111692) (7021575) (2769)

Values are medians and ranges for 24 h. Bold values indicate Po0.05 by Wilcoxons signed rank sum test.

Monounsaturated oils and cholesterol metabolism L Ellegard et al

1377
Table 4 Serum levels of phytosterols, cholesterol, lathosterol, lathosterol/cholesterol ratio (L/C), and 7a-hydroxy-4-cholesten-3-one (7a-HC) in nine ileostomy subjects after 10 h with rapeseed oil vs olive oil diet Rapeseed oil Plant sterols (mmol/l) Cholesterol (mmol/l) Lathosterol (mmol/l) L/C ratio (mmol/l/100 mmol) cholesterol 7a-HC (ng/ml) 32.1 (1058) 5.5 (4.26.6) 11.8 (6.522) Olive oil 23.7 (858) 5.9 (4.66.8) 12.3 (6.822.9)

225 (127352) 29 (1771)

229 (162340) 27 (1858)

Values are medians and ranges. Bold values indicate Po0.05 by Wilcoxons signed rank sum test.

Plant sterol excretion was 160% higher on rapeseed oil than on olive oil, reflecting the higher plant sterol content of the rapeseed oil. Data on serum analyses are presented in Table 4. Serum cholesterol concentration as determined by GLC during analysis for lathosterol, decreased by 7% (P 0.024) within 10 h of consumption of rapeseed oil compared to olive oil diet. 7a-hydroxy-4-cholesten-3-one, as a marker for bile acid synthesis, increased by median 8% (P 0.038) within 10 h on rapeseed oil diet compared to olive oil. The lathosterol/ cholesterol ratio, as well as the absolute lathosterol concentration in serum, as markers for cholesterol synthesis, remained unchanged after 10 h of consumption of the rapeseed oil compared to olive oil, as did the phytosterol concentration in serum.

Discussion
Rapeseed oil and olive oil are both good sources of monounsaturated oleic acid and are often, from a dietetic perspective, considered equal with respect to cholesterol metabolism (Wood et al, 1998). However, as has been recently reviewed in this journal, oils high in monounsaturated fatty acids do not have the same effect on plasma cholesterol (Truswell & Choudhury, 1998). This small but controlled study shows that rapeseed oil might have more impact on cholesterol metabolism than olive oil. Thus, rapeseed oil tended to decrease cholesterol absorption, increased cholesterol excretion, and actually decreased serum cholesterol levels within 10 h of consumption of the rapeseed oil diet. This is supported by findings of Pedersen et al (2000)who reported rapeseed oil to be more effective than olive oil in reducing LDL-cholesterol, and attributed the differences in part to the squalene and plant sterol differences between the oils. The small but rapid 7% decrease in serum cholesterol by rapeseed oil with 326 mg more plant sterols is higher than predicted from trials with added plant sterols and stanols (Katan et al, 2003). However, this estimate is uncertain for doses below 1000 mg, and the reduction could be described as curvilinear, similar to the

relation between dietary cholesterol and serum cholesterol levels. In this study, bile acid excretion rose by 32% on rapeseed oil compared to olive oil, which was mirrored by a 8% increase in serum 7a-hydroxy-4-cholesten-3-one, a marker for bile acid synthesis. This increase could be detected within 10 h after consumption. This rapid response was first described by us in a controlled experiment in healthy subjects (Andersson et al, 2002). Plant sterols in high doses have been shown to decrease cholesterol absorption (Nor men et al, 2000), as well as to increase excretion of cholesterol (Miettinen et al, 2000), but whether plant sterols naturally occurring in normal food would have these effects have not been reported until recently. The plant sterol content of ordinary foods seems to influence cholesterol absorption, as found in single-meal experiments (Ostlund et al, 2002, 2003). We have earlier found a negative correlation between natural plant sterols in food and cholesterol absorption in ileostomy subjects (Ellegard et al, 2000b). Interestingly, the decline in cholesterol absorption although of borderline statistical significanceis close to that predicted (Ellegard et al, 2000b). We also found a positive correlation between changes in net sterol excretion of cholesterol and plant sterol excretion in a retrospective analysis of 14 different dietary interventions in ileostomy rd subjects (Ellega et al, 2000a). This supports the fact that part of the effect of cholesterol-lowering dietary interventions on sterol metabolism could be attributed to the plant sterol content. Dietary intake of natural plant sterols has been shown to be inversely related to serum cholesterol levels in men and women in the EPIC-Norfolk population (Andersson et al, 2004). Thus, the present study, together with previous reported experiences, indicates that plant sterols intrinsic in vegetables, fatty foods, and cereals are biologically active in cholesterol metabolism, to increase sterol excretion, and to a lesser extent, decrease cholesterol absorption.

Acknowledgements The expertise and technical assistance provided by Vibeke Malmros is greatly appreciated. Supported by the Gothenburg Medical Society, Grant number: 91/00, the Swedish Board for Agricultural Research (SJFR) Grant 50.0444/98, the Swedish Medical Society, Grant number 2002-405 and from the Sahlgrenska University Hospital under the LUA-agreement.

References
Andersson H & Bosaeus I (1993): Sterol balance studies in man. A critical review. Eur. J. Clin. Nutr. 47, 153159. rd Andersson M, Ellega L & Andersson H (2002): Oat bran stimulates bile acid synthesis in 8 h as measured by 7a-hydroxy-4-cholestene3-one. Am. J. Clin. Nutr. 76, 11111116. rd Andersson S, Skinner J, Ellega L, Welch A, Bingham S, Mulligan A, Andersson H & Khaw K (2004): Intake of plant sterols is inversely

European Journal of Clinical Nutrition

Monounsaturated oils and cholesterol metabolism L Ellegard et al

1378
related to serum cholesterol concentration in men and women in the EPIC Norfolk population: a cross sectional study. Eur. J. Clin. Nutr. 58, 13781385. Axelson M, Aly A & Sjovall J (1988): Levels of 7alfa-hydroxy-4cholesten-3-one in plasma reflect rates of bile acid synthesis in man. FEBS Lett. 239, 324328. Bosaeus I, Belfrage L, Lindgren C & Andersson H (1992): Olive oil instead of butter increases net cholesterol excretion from the small bowel. Eur. J. Clin. Nutr. 46, 111115. Bosaeus IG & Andersson HBO (1987): Short-term effect of two cholesterol-lowering diets on sterol excretion in ileostomy patients. Am. J. Clin. Nutr. 45, 5459. Bosaeus IG, Carlsson NG, Sandberg AS & Andersson H (1986): Effect of wheat bran and pectin on bile acid and cholesterol excretion in ileostomy patients. Hum. Nutr. Clin. Nutr. 40C, 429440. Connor WE, Witiak DT, Stone DB & Armstrong ML (1969): Cholesterol balance and fecal neutral steroid and bile acid excretion in normal men fed dietary fats of different fatty acid composition. J. Clin. Invest. 48, 13631375. rd Ellega L, Andersson H, Bosaeus I, Langkilde A-M & Lia A (2000a): Bioactive micronutrients in mediterranean diet and health. In COST 916 Action: Bioactive Plant Cell Wall Components in Nutrition and Health eds Amado R, Lairon D, Gerber M, Maiani G & Abt B, p 35, Luxembourg: Office for the official publications of the European communities. rd L & Bosaeus I (1991): Sterol and nutrient excretion in Ellega ileostomists on prudent diets. Eur. J. Clin. Nutr. 45, 451457. rd Ellega L & Bosaeus I (1994): Cholesterol absorption and excretion in ileostomy subjects on high- and low-dietary cholesterol intakes. Am. J. Clin. Nutr. 59, 4852. Ellegard L, Bosaeus I & Andersson H (2000b): Will recommended changes in fat and fibre intake affect cholesterol absorption and sterol excretion? An ileostomy study. Eur. J. Clin. Nutr. 54, 306313. Grundy SM & Ahrens EH (1969): Measurements of cholesterol turnover, synthesis and absorption in man, carried out by isotope kinetic and balance methods. J. Lipid Res. 10, 91107. Katan MB, Grundy SM, Jones P, Law M, Miettinen T & Paoletti R (2003): Efficacy and safety of plant stanols and sterols in the management of blood cholesterol levels. Mayo. Clin. Proc. 78, 965978. Lia A, Hallmans G, Sandberg A, Sundberg B & Andersson H (1995): Oat beta-glucan increases bile acid excretion and a fiber-rich barley fraction increases cholesterol excretion in ileostomy subjects. Am. J. Clin. Nutr. 62, 12451251. Miettinen TA (1982): Gas liquid chromatographic determination of fecal neutral sterols using a capillary column. Clin. Chim. Acta 124, 245248. Miettinen TA, Vuoristo M, Nissinen M, Jarvinen HJ & Gylling H (2000): Serum, biliary, and fecal cholesterol and plant sterols in colectomized patients before and during consumption of stanol ester margarine. Am. J. Clin. Nutr. 71, 10951102. Nestel PJ, Havenstein N, Homma Y, Scott TW & Cook LJ (1975): Increased sterol excretion with polyunsaturated-fat high-cholesterol diets. Metabolism 24, 189198. Normen L, Dutta P, Lia A & Andersson H (2000): Soy sterol esters and beta-sitostanol ester as inhibitors of cholesterol absorption in human small bowel. Am. J. Clin. Nutr. 71, 908913. Ostlund RE, Racette SB, Okeke A & Stenson WF (2002): Phytosterols that are naturally present in commercial corn oil significantly reduce cholesterol absorption in humans. Am. J. Clin. Nutr. 75, 10001004. Ostlund RE, Racette SB & Stenson WF (2003): Inhibition of cholesterol absorption by phytosterol-replete wheat germ compared with phytosterol-depleted wheat germ. Am. J. Clin. Nutr. 77, 13851389. Pedersen A, Baumstark MW, Marckman P, Gylling H & Sandstrom BM (2000): An olive oil-rich diet results in higher concentrations of LDL cholesterol and a higher number of LDL subfraction particles than rapeseed oil and sunflower oil diets. J. Lipid Res. 41, 19011911. Sandberg AS, Andersson H, Hallgren B, Hasselblad K & Isaksson B (1981): Experimental model for in vivo determination of dietary fibre and its effects on the absorption of nutrients in the small intestine. Br. J. Nutr. 45, 283294. Swedish National Food Administration (1996): Food Composition TableEnergy and Nutrients. Uppsala, Sweden: Livsmedelsverket. Truswell AS & Choudhury N (1998): Monounsaturated oils do not all have the same effect on plasma cholesterol. Eur. J. Clin. Nutr. 52, 312315. Wood D, De Backer G, Fargeman O, Graham I, Mancia G & Pyorala K (1998): Prevention of coronary heart disease in clinical practice. Summary of recommendations of the second joint task force of European and other societies on coronary prevention. J. Hypertens. 10, 14071414.

European Journal of Clinical Nutrition