Phytosterols, Red Yeast Rice, and Lifestyle Changes Instead of Statins

A Randomized, Double-Blinded, Placebo-Controlled Trial

David J. Becker, MD; Benjamin French, PhD; Patti B. Morris, RD; Erin Silvent, BA; Ram Y. Gordon, MD


Am Heart J. 2013;166(1):187-196. 

In This Article

Subjects and Methods

Design Overview

We designed a randomized, double-blinded, placebo-controlled trial to evaluate the combined lipid-lowering effects of RYR, phytosterols, and lifestyle change. We selected a 2 × 2 factorial design to simultaneously evaluate the impact of the 2 interventions (phytosterols vs placebo and lifestyle change vs usual care). All participants took RYR. Participants were recruited from July to October 2009 from 2 study centers: Abington Memorial Hospital, Abington, PA, and Chestnut Hill Hospital, Philadelphia, PA. The institutional review boards of both hospitals approved the trial. Written informed consent was obtained from all participants.

Setting and Participants

Participants 21 to 80 years of age with an LDL-C >100 mg/dL and <210 mg/dL were eligible if they had discontinued ≥1 statin because of myalgias or if they had refused statin therapy as prescribed by a physician.

Exclusion criteria included the following: history of statin-associated myositis or rhabdomyolysis; chronic pain or inability to exercise; myocardial infarction, stent, or coronary artery bypass grafting; use of lipid-lowering medications or supplements; use of dietary supplements that might mitigate SAM (ie, coenzyme Q10) or weight loss medication; a history of elevated transaminases; prior intolerance to a study drug; and abnormal baseline laboratory values: LDL-C <100 or >210 mg/dL, triglycerides (TGs) ≥400 mg/dL, creatine phosphokinase (CPK) >500 U/L, transaminases >1.5 times normal, or an abnormal thyroid-stimulating hormone.

Randomization and Interventions

The flow of participants through the trial is shown in Figure 1. Two hundred twenty eligible participants underwent screening blood work after stopping lipid-lowering medications or supplements for ≥30 days. Participants were randomized in blocks of 4, stratified by a history of statin intolerance. The randomization list was generated using the blockrand library of the R programming environment with fixed blocks.[12]

Figure 1.

Participant flow diagram.

Red yeast rice capsules (Sylvan Bioproducts, Kittanning, PA) and phytosterol tablets (Endurance Products Co, Tigard, OR) were purchased from the manufacturer. All supplements and placebo were prepared, bottled, and labeled by the Investigation Drug Service at the University of Pennsylvania, and the blinding code was known only to them. Placebo tablets were made by Endurance Products and were identical to the active product in size, shape, color, and smell.

All participants took RYR (three 600-mg capsules, twice daily) and randomized to take phytosterols (two 450 mg tablets, twice daily) or an identical-appearing placebo (2 tablets, twice daily). Participants were also randomly assigned to a 12-week lifestyle change program or usual care.[7] Participants in the lifestyle change group (LCG) attended weekly 3½-hour meetings led by a board-certified cardiologist. A dietitian educated the group about nutrition and encouraged participants to follow a modified Mediterranean diet. An exercise physiologist described the health benefits of exercise, stretching, and light strength training. Each session concluded with exposure to different relaxation techniques, including Tai Chi, meditation, and yoga. After the lifestyle program ended, participants were instructed to follow the program's recommendations and take study medication for an additional 40 weeks. Meetings were held monthly to provide study product. Treatment adherence was assessed by self-report of the average number of missed doses per week and pill counts. Attendance at the 12-week program and at all subsequent monthly meetings was 92%.

Participants randomized to the usual care group (UCG) met with the study coordinator at the beginning of the study and every 3 months thereafter to receive study materials, complete questionnaires, and medication adherence logs and receive laboratory requisitions.

All participants completed laboratory testing, measurements, and questionnaires at baseline and at weeks 12, 24, and 52. The trial was conducted November 2009 to November 2010.

Outcomes and Follow-up

The primary outcome was LDL-C. Secondary outcomes included total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), TGs, high-sensitivity cardiac C-reactive protein (hs-CRP), weight, body mass index (BMI), and development of myalgia.


Safety was assessed by measuring CPK and transaminase levels. Results of laboratory tests were reviewed independently at weeks 12 and 24 to monitor safety.


A fasting blood sample was obtained at each period for a lipid panel, hs-CRP, complete metabolic profile, and CPK. Thyroid-stimulating hormone level was obtained at baseline and week 52. Laboratory analyses were performed by Quest Diagnostics (Madison, NJ).

Analyses of RYR and phytosterols/placebo (online Supplementary Table I, Supplementary Table II) were performed by (White Plains, NY). Red yeast rice was tested for monacolin levels and the presence of citrinin, a potentially toxic byproduct, by high-performance liquid chromatography. Analyses of phytosterols and placebo were performed by gas chromatography.

Sample Size

For the purposes of sample size calculation, we assumed no interaction between the 2 interventions. At a power level of 80% and a type 1 error rate of 5%, a sample size of 200 would be required to detect a difference of 10 mg/dL in the change in LDL-C between groups, assuming an SD of 25 mg/dL.

Statistical Analysis

Baseline and demographic characteristics were summarized across treatment groups. Linear mixed-effects models with a treatment-by-time interaction term were used to contrast expected outcomes between treatment groups over time; a logistic model was used for the binary outcome of LDL-C <100 mg/dL. Subject-specific random intercepts were used to account for the correlation due to repeated measurements. Because randomization was stratified by history of statin intolerance, this factor was adjusted for in all models. Outcomes that exhibited a positively skewed distribution were ln transformed so that exponentiated regression coefficients quantified ratios in expected outcomes; ratios were converted to percent differences. Model assumptions were evaluated using residuals and normal quantile-quantile plots. Consistent with an intent-to-treat approach, all participants were included in the primary analysis regardless of adherence to their randomized therapy. Because the mixed-effects model accommodates data in which the presence of missing data depends on observed outcomes, no effort was made to impute missing data. We performed 2 sensitivity analyses: an as-treated analysis that only included participants who were adherent to their assigned therapy and an analysis that only included participants who completed the study protocol. For the primary outcome, P < .05 was considered statistically significant. For all secondary outcomes, P < .05 was considered to suggest statistical significance. All statistical tests were 2 sided. All analyses were completed using R 2.12.0 (R Development Core Team, Vienna, Austria).

All authors had complete access to the primary data and were solely responsible for the design and conduct of the study, all study analyses, the drafting and editing of the manuscript, and its final content. The trial was funded by unrestricted grants from the Commonwealth of Pennsylvania, Abington Memorial Hospital, and the Edna G. Kynett Memorial Foundation and registered at (NCT00897975).