- Click here for explanation of classification scheme.

To evaluate the influence of esterified plant sterols on serum lipid concentrations in adults with mild-to-moderate primary hypercholesterolemia following the NCEP Step I diet.

• Men and women aged 21-75 years old
• Abstain from all hypolipidemic therapy (including lipid-lowering medications and supplements thought to alter lipid metabolism) for four weeks before attending clinic screening
• LDL-C between 3.4-5.2mmol/L (130-200mg/dL) and TG<4.0mmol/L (<350mg/dL).

• BMI≥35
• Cancer other than non-melanoma skin cancer or basal cell carcinoma within past five years
• Significant atherosclerotic, gastrointestinal, pulmonary, hepatic, renal, endocrine, immune or biliary disease in recent six months
• Poorly controlled hypertension (HTN) with systolic>160mmHg or diastolic>95mmHg
• Type 1 and poorly controlled or insulin-requiring type 2 diabetes
• Those on drugs for hemostasis (other than stable-dose aspirin or dipyridamole)
• Women of childbearing potential were required to be on contraception during trial.

This study is a randomized, double-blind, three-group parallel controlled study. Subjects were placed into groups that received: Either 8% (low-sterol group) sterol-ester enriched spreads, 16% (high-sterol group) sterol-ester enriched spreads, or the control spread with no plant sterol-esters. Plant sterols included B-sitosterol, campesterol and stigmasterol. Throughout the trial, subjects consumed 14g spread per day, separated into two 7-gram servings. Subjects were asked not to use the spreads for cooking or frying.

Subjects were recruited by phone in the Chicago area and underwent prescreening. Five weeks out, they were required to abstain from all hypolipidemic therapy for four weeks prior. Starting at the four week lead-in, subjects consumed 50% fat spread with no added plant sterol and educated by RD on following NCEP Step I diet (given booklet called Step by Step: Eating to Lower Your High Blood Cholesterol). At the end of the four-week lead in period, subjects were randomly assigned to one of the groups for a five-week intervention period.  Subjects and investigators were blinded to subject randomization status. 

Subjects returned to the clinic three times during the five-week intervention at weeks two, 3.5 and five. 

Screening

• Five weeks out: Eligibility further assessed by medical history review, fasting serum lipid profiles, apoA-I, apoB, physical exam, ECG, clinical lab tests (chemistry, hematology, urinalysis), three-day diet records (two weekdays, one weekend)
• Four weeks out: Vital signs, body weight, serum lipid profile
• Two weeks out: Vital signs, body weight and serum lipid profile measurements
• One week out: apoA-I and apoB, carotenoids, three-day diet records; vital signs, body weight and serum lipid profile measurements
• Zero weeks: apoA-I and apoB, carotenoids, fat-soluble vitamins and sterol; vital signs, body weight and serum lipids
• Two weeks: Vital signs, body weight, serum lipid profile
• 3.5 weeks: Vital signs, body weight, serum lipid profile, apoA-I, apoB, carotenoids, three-day diet records
• Five weeks: Vital signs, body weight, serum lipid profile, apoA-I, apoB, carotenoids, fat-soluble vitamin and sterols, physical exam, ECG, clinical lab tests (serum chemistry, hematology, urinalysis).

Measurements

• Three-day diet records were analyzed using National Data System for Research (University of Minnesota)
• Compliance was evaluated by interviewing subjects and counting the unopened product packages returned to the clinic at weeks -2, -1, zero, two, 3.5 and five. Non-compliance was defined as consumption<80% of scheduled servings during the study period.
• Serum total cholesterol, HDL-C and TGs were measured by Myers et al methods in accordance with CDC lipid measurement standardization program
• LDL-C was calculated using the Friedewald equation
• Vitamin, strol, and carotenoid profiles were measured using HPLC.

Statistical Analysis

• ANOVA, Kruskal-Wallis, and Chi-square as appropriate
• Compliant subjects were included in the analysis; an intent-to-treat analysis included all randomly assigned subjects was performed
• For lipid variables, screening equaled to average of values at weeks -5 and -4; baseline equal to average of values at -2, -1 and zero weeks, and end of the study equal to average of 3.5 and five weeks
• Pairwise comparisons of lipid responses between groups was conducted with Scheffe's test
  • TG responses were not normally distributed so Kruskal-Wallis and Mann-Whitney U tests were performed
• Statistical comparisons among groups for vitamin, carotenoid and sterol responses were conducted with the Kruskal-Wallis test, and pairwise comparisons were made with Scheffe's test on ranked values
• Differences in incidence of adverse events and lab value shifts was assessed using Fisher's exact test
• Dietary intakes were compared by using ANOVA and Kruskal-Wallis tests, with pairwise comparisons done with Scheffe's test.

Primary Outcome Variable

Percent change from baseline LDL-C

Of 388 subjects screened, 224 were randomly assigned to study groups. Subjects were randomized as follows:

• Control group (N=92)
• Low-sterol group (N=92)
• High-sterol group (N=40).

Five (2%) of the 224 dropped out of study before completing intervention period (two control, two low-sterol, one high-sterol). Reasons for withdraw included withdrawn consent (N=2), withdraw by investigator for non-compliance (N=2) and serious adverse event unrelated to study product (N=1). 

The groups did not differ in mean or median age, sex, BMI, race, smoking status, alcoholic drinks consumed per week, SBP or DBP. There were no significant differences among groups.

• Predominantly non-smokers with mean age of 60 years old and mean BMI of 27
• Men and women were equally distributed in low- and high-sterol groups but control was 61% female
• Average systolic BP was about 115mmHg, diastolic average was about 73mmHg
• About 80% were non-Hispanic whites.

 

Mean compliance with study product consumption exceeded 90% in all study groups, and there were no significant differences between the control and intervention groups.

In general, the nutrient intakes were not significantly different among the three groups at screening (week -4), at baseline (week zero) and at the end of the study (week five). After the intervention period, total and saturated fatty acid intakes were within NCEP Step 1 diet recommendations.

Subjects in the low- and high-sterol groups who consumed >80% of the scheduled servings (per protocol analyses) had TC values that were 5.2% and 6.6% lower, LDL-C values that were 7.6% and 8.1% lower, apoB values that were 6.2% and 8.4% lower and ratios of total to HDL-C that were 5.9% and 8.1% lower, respectively, than values for the control group (P<0.001 for all). Additionally, TG concentrations decreased by 10.4% in the high-sterol group, but were not significantly different between the low-sterol group and the control group. ApoA-I and HDL-C responses did not differ significantly among the three groups. In the intent-to-treat subjects, total and LDL-C, the ratio of total to HDL-C and apoB were a mean of 3.9% and 5.6%, 5.9% and 6.9%, 5.5% and 7.3% and 5.1% and 6.4% lower in the low-sterol and high-sterol groups, respectively, than in the control group (all P<0.01). There were no significant differences among groups in the number of shifts from low-normal to high or high-normal to low clinical chemistry, hematology or urinalysis values. Serum concentrations of fat-soluble vitamins and carotenoids were generally within reference ranges at baseline and post-intervention, except for trans-ß-carotene after correction for TC concentration (reductions of 16.8% and 23.5% in low- and high-sterol groups). Serum plant sterol concentrations increased from baseline (0.48% of total sterol by weight) to 0.64% and 0.71% by wt for the low- and high-sterol groups, respectively (P<0.05 compared with control).

After five weeks of 1.1g sterols:

• TC Baseline: 6.17±0.07mmol/L (238.5±2.9mg/dL)  
• TC Ending: Not reported     
• Percent Change: -2.6±0.9% (P<0.001)     
• LDL-C Baseline: 4.08±0.07mmol/L (157.8±2.7mg/dL)  
• LDL-C Ending: Not reported      
• Percent Change: -4.9±1.3% (P<0.001)
• Total: HDL-C Baseline: 4.72±0.12
• Total: HDL-C Ending: Not reported
• Percent Change: -4.3±1.1% (P<0.001)
• HDL-C and TG: No statistically significant changes.

After five weeks of 2.2g sterols:

• TC Baseline: 6.13±0.12mmol/L (237.0±4.5mg/dL)
• TC Ending: Not reported 
• Percent Change: -4.0±1.2% (P<0.001) 
• LDL-C Baseline: 4.03±0.10mmol/L (155.7±3.8mg/dL)
• LDL-C Ending: Not reported
• Percent Change: -5.4±1.6% (P<0.001)
• Total: HDL-C Baseline: 4.83±0.212
• Total: HDL-C Ending: Not reported
• Percent Change: -6.5±1.4% (P<0.001)
• HDL-C and TG: No statistically significant changes.

Results indicate that plant sterol intakes of 1.1 and 2.2 grams per day produce clinically important reductions in LDL-C when the sterols are delivered in reduced-fat spreads incorporated into an NCEP Step I diet. The results of this study, together with other recent investigations of sterol esters, outlined in the recently approved health claim for sterol and stanol products, provide evidence that plant sterol esters are a well-tolerated and efficacious adjunct to a lipid-lowering diet, capable of producing clinically significant reductions in LDL-C in persons with mild-to-moderate primary hypercholesterolemia.

Industry:	Lipton Inc, Chicago Center for Clinical Research Food Company: Other:
University/Hospital:	Tufts University, Boston VA Medical Center

Strengths

• Well-controlled study with high compliance and low drop-out rate
• Men and women were evenly distributed in both the low- and high-sterol groups
• Race distribution among the three groups was similar to that reported in the general US population
• Statistical analyses were clear and appropriate.

Weaknesses

Subjects limited to the Chicago area