Abstract
Background
Phytosterols are structurally similar to cholesterol and partially inhibit intestinal absorption of cholesterol, although their impact on coronary artery disease (CAD) risk remains to be elucidated.
Objectives
This study aimed to prospectively assess the associations between total and individual phytosterol intake and CAD risk in United States health professionals.
Methods
The analysis included 213,992 participants from 3 prospective cohorts—the Nurses’ Health Study (NHS), NHSII, and Health Professionals Follow-Up Study—without cardiovascular disease or cancer at baseline. Diet was assessed using a validated food frequency questionnaire every 2–4 y since baseline. Associations between phytosterol intake and the risk of CAD, such as nonfatal myocardial infarction and fatal CAD, were evaluated using Cox proportional hazards regression models.
Results
More than 5,517,993 person-years, 8725 cases with CAD were documented. Comparing extreme quintiles, pooled hazard ratios (95% CIs) of CAD were 0.93 (0.86, 1.01; P-trend = 0.16) for total phytosterols, 0.89 (0.82, 0.96; P-trend = 0.05) for campesterol, 0.95 (0.88, 1.02; P-trend = 0.10) for stigmasterol, and 0.92 (0.85, 1.00; P-trend = 0.09) for β-sitosterol. Nonlinear associations were observed for total phytosterols, campesterol, and β-sitosterol: the risk reduction plateaued at intakes above ∼180, 30, and 130 mg/d, respectively (P-nonlinearity < 0.001). In a subset of participants (N range between 11,983 and 22,039), phytosterol intake was inversely associated with plasma concentrations of total cholesterol, triglycerides, high-density lipoprotein cholesterol, and IL-6 and positively associated with adiponectin, whereas no significant associations were observed for low-density lipoprotein cholesterol or C-reactive protein concentrations.
Conclusions
Higher long-term intake of total and major subtypes of phytosterols may be associated with a modest reduction in CAD risk, displaying a nonlinear relationship that plateau at moderate intake levels. The role of phytosterols in preventing CAD warrants further investigation.
Keywords: phytosterol, campesterol, stigmasterol, β-sitosterol, coronary artery disease, lipids, inflammation
Introduction
Phytosterols, comprising plant sterols and stanols, occur naturally in all foods of plant origins and are particularly abundant in vegetable oils, vegetables, fruits, nuts, seeds, and legumes [1]. Natural dietary intake of phytosterols varies between 200 and 400 mg/d in Western diets and higher (600+ mg/d) in vegetarian or vegan diets [2]. Phytosterols are structurally similar to cholesterol and may partially inhibit intestinal absorption of cholesterol [1]. A meta-analysis of 124 randomized controlled studies with a mean phytosterol dose of ∼2,000 mg/d found significant reduction in low-density lipoprotein (LDL) cholesterol [3]. Certain randomized clinical trials also suggested that phytosterols may have anti-inflammatory properties [[4], [5], [6]]. Because lowering LDL cholesterol and inflammation levels is associated with reduced risk of coronary artery disease (CAD) [7,8], clinical guidelines for the management of dyslipidemia and/or prevention of cardiovascular disease recommend ∼2,000 mg/d intake of phytosterols [9,10].
However, evidence regarding dietary phytosterols in relation to the risk of CAD remains inconsistent and limited. A few cross-sectional and prospective studies have mainly focused on the blood concentrations of phytosterols in association with CAD risk [[11], [12], [13], [14], [15]], but results are equivocal. Only 2 prospective studies have examined the association between dietary phytosterols and CAD risk but showed inconsistent findings. One study conducted in the European Prospective Investigation into Cancer and Nutrition-the Netherlands (EPIC-NL) cohort found that total dietary phytosterols were not significantly associated with the risk of CAD [16]. By contrast, an analysis in the Women’s Health Initiative (WHI) cohort showed an inverse association between higher dietary phytosterol intake and CAD risk [17]. In addition, given the varied structures and intake amounts of individual phytosterols (eg, campesterol, stigmasterol, and β-sitosterol), it is likely that they may have different associations with CAD risk. Nevertheless, evidence is lacking from large population-based cohort studies regarding the association between individual phytosterols and CAD risk.
Therefore, we aimed to provide a comprehensive assessment on the associations of long-term intake of total phytosterols and individual phytosterols in 3 large prospective cohorts in the United States with dietary phytosterols measured repeatedly over 30 y of follow-up. We also examined the associations between dietary phytosterols and blood lipid concentrations [total cholesterol, LDL cholesterol, high-density lipoprotein (HDL) cholesterol, and triglycerides], in addition to other cardiovascular disease (CVD) risk markers, such as high-sensitivity C-reactive protein (hs-CRP), IL-6, and adiponectin, in a subset of participants.
Methods
Study population
The Health Professionals Follow-Up Study (HPFS) was started in 1986 and recruited 51,529 United States male health professionals aged 40–75 y. The Nurses’ Health Study (NHS) was established in 1976 and included 121,700 female registered nurses aged 30–55 y. The Nurses’ Health Study II (NHSII) was initiated in 1989 and recruited 116,340 female nurses aged 25–42 y. In all 3 cohorts, participants responded to a mailed questionnaire on their lifestyle factors and medical history at baseline and similar questionnaires were sent biennially to update participant demographic and lifestyle information and identify incident diseases. The cumulative response rate in 30 cohorts exceeded 90% [18,19].
The study baseline was 1986 for the HPFS study, 1984 for the NHS study, and 1991 for the NHSII study when diet was first comprehensively assessed using a validated semiquantitative food frequency questionnaire (FFQ) [[20], [21], [22]]. We excluded participants with prevalent CVD or cancer at baseline, had unusual total energy intake (<500 or >3,500 kcal/d for females and <800 or >4200 kcal/d for males), completed baseline FFQ only, or had missing data on the dietary phytosterol intake. A flowchart that summarizes the participant selection process is shown in Supplemental Figure 1. In the final analysis, the sample size for the HPFS was 43,250, and it was 77,245 for the NHS and 93,497 for the NHSII.
The study protocol was approved by the Human Research Committee of Brigham and Women’s Hospital and the Harvard T.H. Chan School of Public Health. Completion and return of study questionnaires implied informed consent of the participants.
Assessment of dietary phytosterol intake, demographic, lifestyle, and CVD risk markers
In all 3 cohorts, diet was assessed using a validated FFQ at baseline and updated every 2–4 y during the follow-up. For each food item listed in the FFQ, participants were asked about their average consumption frequency of a prespecified portion size during the previous year. The average daily intake of individual phytosterols was calculated by multiplying the frequency of consumption of each phytosterol-containing food item by phytosterol content and then summing across all foods. Total phytosterol intake was the sum of all 3 major individual phytosterols (campesterol, stigmasterol, and β-sitosterol). Because the intake and the major contributors of individual phytosterols changed during the follow-up (Supplemental Figure 2), we listed the top 10 food contributors of each individual phytosterol intake at baseline, the middle of follow-up, and the end of follow-up. Details about assessment of demographic and lifestyle factors and CVD risk markers can be found in Supplemental Methods.
Disease outcome assessment
The primary disease outcome for this analysis was total CAD, which included nonfatal myocardial infarction (MI) and fatal CAD. Both definite and probable cases were included in the analysis. In the 3 cohorts, permission was sought to review medical records of participants who reported having a nonfatal MI on a follow-up questionnaire. Study physicians who were blinded to the exposure status confirmed or refuted a reported MI according to the presence of symptoms and either typical electrocardiographic changes or elevated cardiac enzyme concentrations [23,24]. Deaths were identified through reports from the next of kin, the postal authorities, or searching the National Death Index. Fatal CAD was confirmed by hospital records, by autopsy, or if CAD was listed as the cause of death on the certificate with evidence of previous CAD. Sudden deaths without cardiac causes were not considered as fatal CAD in this analysis.
Statistical analysis
Given the increasing trend of phytosterol intake during the follow-up (Supplemental Figure 3), participant characteristics at the median of the follow-up (2000 for HPFS and NHS studies and 2003 for the NHSII study) were presented for the main analysis. We also presented participant characteristics at baseline. The intakes of total phytosterols and individual phytosterol were cumulatively averaged to reflect long-term usual intake. Person-years were calculated from the return of the baseline questionnaires to the CAD diagnosis date, death date, date of last return of a valid follow-up questionnaire, or the end of follow-up (31 January 2016, for the HPFS, 30 June 2014, for the NHS, and 30 June 2017, for the NHSII), whichever occurred first. To decrease the impact from the potential reverse causality that participants with existing diseases might change their usual diet intake, we stopped updating diet for participants once they reported diabetes, stroke, coronary artery bypass graft, or cancer during follow-up [25]. These participants’ last-updated dietary intake was used for their subsequent follow-ups. If the values are missing, we replaced them with the values in the questionnaire from the preceding follow-up cycle.
An age-stratified (months) and calendar time–stratified multivariable-adjusted Cox proportional hazards model was used to estimate the hazard ratios (HRs) and 95% CIs for the association between cumulatively averaged total phytosterols and individual phytosterol and risk of CAD. The proportional hazards assumption was evaluated by including an interaction term between categorical total/individual phytosterol variable and the duration of follow-up, and we did not detect violations in the main analyses. Intakes of total phytosterols and individual phytosterols were time-varying exposures and categorized into quintiles in this analysis. Multivariable models adjusted for potential factors associated with CAD risk, such as ethnicity (White, African American, Asian, and others), smoking status (never smoked, past smoker, currently smoke 1–14 cigarettes per day, 15–24 cigarettes per day, or ≥25 cigarettes per day), body mass index (BMI; <21.0, 21.0–22.9, 23.0–24.9, 25.0–26.9, 27.0–29.9, 30.0–32.9, 33.0–34.9, or ≥35.0 kg/m2), alcohol intake (0, 0.1–4.9, 5.0–9.9, 10.0–14.9, 15.0–29.9, and ≥30.0 g/d), multivitamin use (yes/no), use of cholesterol-lowering medication (yes/no), physical activity (quintiles), total energy intake (quintiles), dietary transfat (quintiles), saturated fatty acids (quintiles), red meat and processed meat (quintiles), total dairy intake (quintiles), refined grains (quintiles), fish (quintiles), dietary cholesterol (quintiles), a family history of MI (yes/no), postmenopausal hormone use (females only; premenopausal, never, former, current, or missing), and oral contraceptive use (NHSII only; ever used/never used). Except for ethnicity and a family history of MI, all other covariates were modeled as time-varying covariates in the multivariable models. The median value of total phytosterols and each individual phytosterol by quintiles was modeled as continuous variables to calculate P values for trend. Data from each cohort were analyzed separately and then combined using a fixed-effect model. We further conducted a sensitivity analysis using simple updated phytosterol intakes as exposure to test the robustness of our results. To explore the dose–response relationship between total phytosterols and individual phytosterol intake and CAD risk, we applied cubic spline analysis (number of knots = 4) adjusting for the abovementioned covariates and data origin (NHS, NHSII, and HPFS). We excluded the highest and lowest 1 percentile of dietary phytosterol intake to limit of the impact of extreme values. After excluding extreme values, we used the minimum intake of total and individual phytosterols as the reference concentration of this spline regression analysis (114.46 mg for total phytosterols, 12.19 mg for stigmasterol, 16.79 mg for campesterol, and 82.55 mg for β-sitosterol). We further repeated the analysis using the intake amount at the first knot as the reference level (141.0 mg for total phytosterols, 15.13 mg for stigmasterol, 20.95 mg for campesterol, and 102.45 mg for β-sitosterol), which are approximately corresponding to the 10th percentile.
In addition, we estimated the associations of substituting 100 mg/d of total dietary phytosterols and β-sitosterol or 30 mg/d of stigmasterol and campesterol for cholesterol intake by including both variables in continuous form in the same multivariable model. The difference in their coefficients plus their covariance was used to estimate the relative risk and 95% CI for the substitution [26]. We presented results of 100-mg/d substitution for total dietary phytosterols and β-sitosterol and 30-mg/d substitution for stigmasterol and campesterol because this intake amount was more realistic for this study population.
A stratified analysis was conducted among participants who developed hypercholesterolemia during the follow-up, stratified by the use of cholesterol-lowering medications (yes/no). In this analysis, we further stopped updating diet once they developed hypercholesterolemia during the follow-up. The P value for interaction was calculated by evaluating the significance of product terms between cholesterol-lowering medications and quintiles of total phytosterols on CAD risk through a likelihood ratio test.
Linear regression analyses were conducted to assess the associations of total and individual phytosterols with lipids and CVD risk markers. The average intake of total phytosterols and individual phytosterols was calculated from the FFQs administered before the date of blood collection (1986, 1990, and 1994 in HPFS; 1984, 1986, and 1990 in NHS; and 1991, 1995, and 1999 in NHSII). Multivariable models were adjusted for the same covariates included in the Cox proportional hazards model, with additional manual adjustment in statistical analyses for study cohort, fasting and case–control status, LDL cholesterol, HDL cholesterol, and triglycerides. The β-coefficient from the linear regression represents the change in concentrations of lipids and CVD risk markers with per 1-mg/d increment in phytosterols. To be consistent with the previous study [16], we multiplied our β-coefficient by 50 to show the change in concentrations of lipids and CVD risk markers with per 50-mg/d increment in phytosterols.
Sensitivity analyses were conducted to test the robustness of the results. First, instead of adjusting for individual foods and nutrients, we adjusted for healthful plant-based diet index to account for potential impact by other components in plant-based foods. Second, to examine whether the association between individual phytosterol and CAD risk was independent of other phytosterols, we mutually adjusted for 3 individual phytosterols in the model. Third, we repeated the main analysis but we further stopped updating diet once participants developed hypercholesterolemia during the follow-up.
Analyses were conducted with the SAS software for UNIX (version 9.4; SAS Institute) and RStudio Server Pro (version 1.4.1094-2; RStudio, PBC). The 2-sided P values of <0.05 were considered statistical significance.
Results
Among the 3 cohorts, major food contributors of campesterol included dark bread, breakfast cereal, pasta, pizza, broccoli, beans, orange, and English muffin. Top stigmasterol-containing food included beans, dark chocolate, lettuce, and banana. Major food sources for β-sitosterol contained orange, peas, apple, nuts, broccoli, and pasta (Supplemental Figure 2). At the median follow-up, the median intake of phytosterols ranged from 183.6 (NHS) to 226.9 mg/d (HPFS), and the averaged median intake over the entire follow-up period was 193 mg/d (range: 7–820 mg/d) across 3 cohorts. Of the 3 individual phytosterols (campesterol, stigmasterol, and β-sitosterol), β-sitosterol was the most abundant and accounted for 74.5% of dietary phytosterol intake. Participants with higher intake of total phytosterols were older and had more favorable health and lifestyle profiles, such as lower BMI, lower prevalence of hypertension, lower likelihood to be current smokers, higher levels of physical activity, and better diet quality indicated by lower intake of saturated fatty acids, dietary cholesterol, transfat, red and processed meat, total dairy intake, and higher intake of whole grains and fish (Table 1). When using the baseline data instead of the median follow-up, similar characteristics were observed when comparing highest compared with lowest quintiles of total phytosterol intake (Supplemental Table 1). Similar to the trend of phytosterol intake over time (Supplemental Figure 3), we also observed an increasing trend of diet quality (measured by alternative healthy eating index) and healthy food intake such as whole grains, fruits, and vegetables over the years (Supplemental Figure 4).
TABLE 1.
Age-standardized characteristics of study participants in Health Professionals Follow-up Study (1986–2014), Nurses’ Health Study (1984–2014), and Nurses’ Health Study II (1991–2017) at median follow-up (N = 201,244)∗.
| Health Professionals Follow-up Study (2000) |
Nurses’ Health Study (2000) |
Nurses’ Health Study II (2003) |
||||
|---|---|---|---|---|---|---|
| Q1 (n = 7578) | Q5 (n = 7577) | Q1 (n = 14,145) | Q5 (n = 14,143) | Q1 (n = 18,525) | Q5 (n = 18,525) | |
| Total phytosterols, mg | 174.4 (160.4–183.7) | 285.4 (271.9–308.1) | 146.4 (135.8–153.2) | 223.6 (214.9–238.0) | 152.9 (141.8–160.2) | 239.3 (228.5–257.8) |
| Campesterol | 25.9 (23.3–28.2) | 42 (38.9–46.1) | 21.9 (20.0–23.5) | 32.7 (30.6–35.3) | 23.7 (21.7–25.5) | 35.2 (32.7–38.2) |
| Stigmasterol | 19.4 (17.2–21.5) | 29.4 (26.6–32.8) | 16.5 (14.8–18.0) | 23.6 (21.7–26.0) | 17.0 (15.3–18.7) | 25.0 (22.6–28.0) |
| β-sitosterol | 128.0 (117.4–135.2) | 214.6 (203.9–233.0) | 107.2 (99.1–112.5) | 167.9 (160.9–179.2) | 111.0 (102.6–116.8) | 179.9 (171.4–194.7) |
| Age, y | 64.0 ± 8.6 | 68.0 ± 9.3 | 64.5 ± 6.9 | 68.6 ± 6.9 | 48.0 ± 4.7 | 49.7 ± 4.4 |
| Body mass index, kg/m2 | 26.8 ± 4.2 | 25.4 ± 3.6 | 25.3 ± 5.1 | 24.3 ± 4.3 | 25.4 ± 6.0 | 23.7 ± 4.6 |
| Ethnicity, % | ||||||
| White | 96.0 | 93.3 | 98.3 | 96.5 | 96.6 | 94.4 |
| African American | 1.9 | 2.5 | 1.2 | 1.8 | 0.4 | 0.7 |
| Asian | 1.0 | 2.9 | 0.2 | 0.4 | 0.8 | 3.1 |
| Others | 1.0 | 1.2 | 0.3 | 1.3 | 2.1 | 1.7 |
| Physical activity, MET-h/wk | 13.9 (3.2–32.8) | 27.5 (10.7–51.8) | 6.5 (1.6–16.2) | 15.2 (5.2–30.5) | 13.3 (3.4–33.2) | 27.0 (10.5–70.9) |
| Hypertension | 43.6 | 38.5 | 50.1 | 43.8 | 24.5 | 17.0 |
| High cholesterol | 46.9 | 48.3 | 57.8 | 57.5 | 35.1 | 29.9 |
| Use of cholesterol-lowering medication | 18.7 | 21.8 | 22.7 | 23.5 | 5.5 | 3.8 |
| Smoking status, % | ||||||
| Never smokers | 38.9 | 53.4 | 39.6 | 48.8 | 62.2 | 66.3 |
| Past smokers | 50.9 | 44.4 | 44.1 | 46.1 | 23.9 | 28.6 |
| Current smokers | 10.3 | 2.2 | 16.3 | 5.1 | 13.9 | 5.2 |
| Family history of MI, % | 11.2 | 12.5 | 19.1 | 19.6 | 48.7 | 47.6 |
| Multivitamin use, % | 42.5 | 56.9 | 54.7 | 63.7 | 48.9 | 60.4 |
| Oral contraceptive use, % | — | — | 49.8 | 49.6 | 74.7 | 71.5 |
| Hormone use, % | ||||||
| Premenopausal | — | — | 2.2 | 2.3 | 66.5 | 68.2 |
| Postmenopausal-never | — | — | 27.8 | 23.2 | 11.1 | 10.6 |
| Postmenopausal-current | — | — | 41.3 | 45.9 | 12.6 | 11.5 |
| Postmenopausal-past | — | — | 28.7 | 28.6 | 9.7 | 9.8 |
| Alcohol consumption (g/d) | 9.5 (1.8–28.7) | 4.8 (0.9–12.5) | 2.0 (0–10.3) | 1.8 (0.2–6.0) | 0.9 (0–3.5) | 1.8 (0–5.4) |
| Total energy intake (kcal/d) | 1958 ± 556 | 1967 ± 530 | 1712 ± 458 | 1744 ± 450 | 1780 ± 492 | 1813 ± 474 |
| Saturated fatty acid (g/d)1 | 25.2 ± 9.3 | 17.2 ± 5.8 | 22.4 ± 4.1 | 16.7 ± 3.2 | 24.2 ± 4.5 | 18.0 ± 3.7 |
| Dietary cholesterol (mg/d)1 | 329.1 ± 108.8 | 226.7 ± 79.1 | 263.1 ± 77.1 | 212.0 ± 61.7 | 256.1 ± 64.5 | 201.1 ± 59.6 |
| Trans fat (g/d)1 | 1.4 ± 0.6 | 1.0 ± 0.5 | 1.8 ± 0.4 | 1.4 ± 0.5 | 1.6 ± 0.4 | 1.1 ± 0.4 |
| Red and processed meat (servings/d)1 | 2.7 ± 5.8 | 2.5 ± 10.5 | 1.2 ± 0.6 | 0.7 ± 0.4 | 1.2 ± 0.6 | 0.6 ± 0.4 |
| Total dairy intake (servings/d)1 | 1.7 (1.0–3.1) | 1.1 (0.6–1.7) | 1.4 (0.9–2.2) | 1.2 (0.8–1.8) | 1.7 (1.0–2.7) | 1.4 (0.9–2.1) |
| Refined grain (servings/d)1 | 1.1 ± 0.9 | 1.1 ± 0.8 | 0.7 ± 0.5 | 0.9 ± 0.5 | 1.5 ± 0.8 | 1.5 ± 0.8 |
| Whole grain (servings/d)1 | 1.2 ± 0.9 | 2.5 ± 1.3 | 0.2 ± 0.3 | 0.6 ± 0.6 | 1.1 ± 0.7 | 2.1 ± 1.0 |
| Fish (servings/d)1 | 0.3 ± 0.2 | 0.5 ± 0.4 | 0.2 ± 0.2 | 0.4 ± 0.3 | 0.2 ± 0.2 | 0.3 ± 0.3 |
Abbreviations: MET-h/wk, metabolic equivalent–hours per week; Q, quintiles for total phytosterols; MI, myocardial infarction.
Values are n, median (interquartile range), mean ± SD, or %, unless otherwise indicated.
Values are energy adjusted.
Intercorrelations between phytosterols and other dietary factors are shown in Supplemental Figure 5. In all 3 cohorts, the intake amounts of individual phytosterols were highly correlated with each other.
During 5,517,993 person-years of follow-up, 8,725 cases with CAD were documented, of which 4,800 cases were nonfatal MI and 3,903 cases were fatal CAD. Higher intake of campesterol and β-sitosterol were associated with marginally lower risks of total CAD. In a multivariable-adjusted model, comparing the highest compared with lowest quintile, the HRs (95% CIs) were 0.93 (0.86, 1.01) for total phytosterol (P-trend = 0.16), 0.89 (0.82, 0.96) for campesterol (P-trend = 0.05), 0.95 (0.88, 1.02) for stigmasterol (P-trend = 0.10), and 0.92 (0.85, 1.00) (P-trend = 0.09) for β-sitosterol (Table 2). For campesterol, the association was observed for fatal CAD but not nonfatal MI (Table 2). Associations of total phytosterol and individual phytosterol with CAD risk were similar when using simply updated diet but attenuated toward null as expected (Supplemental Table 2) [27]. The results were largely the same after adjusting for healthful plant-based diet index, whereas the associations for fatal CAD were slightly strengthened (Supplemental Table 3). In particular, comparing extreme quintiles, the HRs (95% CIs) were 0.85 (0.76, 0.96) for total phytosterol (P-trend = 0.03) and 0.87 (0.77, 0.98) for β-sitosterol (P-trend = 0.05) (Supplemental Table 3).
TABLE 2.
Pooled hazard ratios (95% CIs) between phytosterol intake and CAD risk in Health Professionals Follow-Up Study (1986–2014), Nurses’ Health Study (1984–2014), and Nurses’ Health Study II (1991–2017) (N = 213,992)∗
| Q1 | Q2 | Q3 | Q4 | Q5 | P-trend1 | |
|---|---|---|---|---|---|---|
| Total phytosterol | ||||||
| Median levels (IQR), mg2 | 141.5 (128.4–150.3) | 168.7 (163.2–173.8) | 188.6 (183.7–193.7) | 210.7 (204.6–217.8) | 249.9 (235.9–273.7) | |
| Total CAD | ||||||
| Case/person-year | 2005/1,098,973 | 1658/1,104,659 | 1652/1,104,815 | 1688/1,105,679 | 1722/1,103,867 | |
| Age-adjusted | Reference | 0.77 (0.72, 0.83) | 0.72 (0.67, 0.77) | 0.67 (0.63, 0.72) | 0.63 (0.59, 0.67) | <0.001 |
| Multivariable-adjusted3 | Reference | 0.93 (0.87, 0.99) | 0.93 (0.87, 1.00) | 0.93 (0.86, 1.00) | 0.93 (0.86, 1.01) | 0.16 |
| Nonfatal MI | ||||||
| Case/person-year | 1,068/1,098,986 | 951/1,104,671 | 944/1,104,822 | 944/1,105,685 | 893/1,103,875 | |
| Age-adjusted | Reference | 0.85 (0.77, 0.92) | 0.80 (0.74, 0.88) | 0.76 (0.70, 0.84) | 0.68 (0.62, 0.75) | <0.001 |
| Multivariable-adjusted3 | Reference | 0.94 (0.86, 1.03) | 0.95 (0.87, 1.05) | 0.96 (0.86, 1.06) | 0.95 (0.85, 1.06) | 0.49 |
| Fatal CAD | ||||||
| Case/person-year | 932/1,099,955 | 700/1,105,575 | 704/1,105,725 | 742/1,106,574 | 825/1,104,701 | |
| Age-adjusted | Reference | 0.69 (0.62, 0.76) | 0.62 (0.56, 0.69) | 0.57 (0.52, 0.63) | 0.56 (0.51, 0.62) | <0.001 |
| Multivariable-adjusted3 | Reference | 0.89 (0.80, 0.99) | 0.89 (0.80, 0.99) | 0.88 (0.79, 0.99) | 0.90 (0.80, 1.01) | 0.13 |
| Campesterol | ||||||
| Median levels (IQR), mg2 | 20.9 (18.8–22.2) | 25.1 (24.3–25.9) | 28.3 (27.5–29.0) | 31.7 (30.8–32.8) | 37.9 (35.7–41.8) | |
| Total CAD | ||||||
| Case/person-year | 2047/1,098,230 | 1659/1,104,277 | 1537/1,105,771 | 1753/1,105,723 | 1729/1,103,992 | |
| Age-adjusted | Reference | 0.79 (0.74, 0.84) | 0.70 (0.65, 0.74) | 0.74 (0.70, 0.79) | 0.66 (0.62, 0.70) | <0.001 |
| Multivariable-adjusted3 | Reference | 0.91 (0.85, 0.98) | 0.86 (0.80, 0.93) | 0.97 (0.90, 1.04) | 0.89 (0.82, 0.96) | 0.05 |
| Nonfatal MI | ||||||
| Case/person-year | 1057/1,098,242 | 936/1,104,286 | 879/1,105,781 | 1028/1,105,729 | 900/1,104,002 | |
| Age-adjusted | Reference | 0.87 (0.80, 0.95) | 0.79 (0.72, 0.87) | 0.89 (0.82, 0.97) | 0.74 (0.68, 0.81) | <0.001 |
| Multivariable-adjusted3 | Reference | 0.94 (0.86, 1.03) | 0.89 (0.81, 0.98) | 1.04 (0.95, 1.15) | 0.92 (0.83, 1.03) | 0.58 |
| Fatal CAD | ||||||
| Case/person-year | 983/1,099,215 | 720/1,105,172 | 653/1,106,593 | 722/1,106,704 | 825/1,104,844 | |
| Age-adjusted | Reference | 0.71 (0.64, 0.78) | 0.60 (0.54, 0.66) | 0.60 (0.54, 0.66) | 0.57 (0.52, 0.63) | <0.001 |
| Multivariable-adjusted3 | Reference | 0.87 (0.78, 0.96) | 0.81 (0.73, 0.90) | 0.87 (0.78, 0.97) | 0.84 (0.75, 0.94) | 0.010 |
| Stigmasterol | ||||||
| Median levels (IQR), mg2 | 15.6 (14.1–16.7) | 18.9 (18.3–19.6) | 21.3 (20.7–21.9) | 23.9 (23.2–24.7) | 28.6 (26.9–31.4) | |
| Total CAD | ||||||
| Case/person-year | 1859/1,101,518 | 1681/1,104,825 | 1636/1,105,549 | 1722/1,104,602 | 1827/1,101,499 | |
| Age-adjusted | Reference | 0.88 (0.83, 0.94) | 0.82 (0.77, 0.88) | 0.81 (0.76, 0.87) | 0.81 (0.76, 0.87) | <0.001 |
| Multivariable-adjusted3 | Reference | 0.99 (0.92, 1.06) | 0.95 (0.88, 1.02) | 0.95 (0.89, 1.02) | 0.95 (0.88, 1.02) | 0.10 |
| Nonfatal MI | ||||||
| Case/person-year | 1002/1,101,528 | 943/1,104,832 | 939/1,105,562 | 962/1,104,612 | 954/1,101,506 | |
| Age-adjusted | Reference | 0.92 (0.84, 1.00) | 0.89 (0.82, 0.98) | 0.89 (0.81, 0.97) | 0.85 (0.77, 0.93) | 0.001 |
| Multivariable-adjusted3 | Reference | 0.96 (0.87, 1.05) | 0.95 (0.86, 1.04) | 0.94 (0.85, 1.03) | 0.93 (0.84, 1.03) | 0.19 |
| Fatal CAD | ||||||
| Case/person-year | 854/1,102,436 | 737/1,105,732 | 689/1,106,408 | 755/1,105,545 | 868/1,102,409 | |
| Age-adjusted | Reference | 0.84 (0.76, 0.93) | 0.73 (0.66, 0.81) | 0.73 (0.66, 0.81) | 0.77 (0.70, 0.85) | <0.001 |
| Multivariable-adjusted3 | Reference | 1.02 (0.92, 1.13) | 0.93 (0.84, 1.04) | 0.95 (0.85, 1.05) | 0.95 (0.85, 1.06) | 0.19 |
| β-Sitosterol | ||||||
| Median levels (IQR), mg2 | 102.5 (92.9–109.1) | 123.1 (134.7–142.5) | 138.6 (134.7–142.5) | 155.8 (151.1–161.3) | 186.4 (175.5–204.8) | |
| Total CAD | ||||||
| Case/person-year | 1987/1,098,940 | 1712/1,104,113 | 1671/1,105,108 | 1668/1,105,104 | 1687/1,104,728 | |
| Age-adjusted | Reference | 0.80 (0.75, 0.85) | 0.72 (0.67, 0.77) | 0.66 (0.62, 0.70) | 0.61 (0.57, 0.65) | <0.001 |
| Multivariable-adjusted3 | Reference | 0.96 (0.89, 1.02) | 0.94 (0.88, 1.01) | 0.93 (0.86, 1.00) | 0.92 (0.85, 1.00) | 0.09 |
| Nonfatal MI | ||||||
| Case/person-year | 1076/1,098,955 | 992/1,104,124 | 929/1,105,115 | 933/1,105,109 | 870/1,104,736 | |
| Age-adjusted | Reference | 0.87 (0.80, 0.95) | 0.78 (0.71, 0.85) | 0.74 (0.68, 0.81) | 0.65 (0.60, 0.71) | <0.001 |
| Multivariable-adjusted3 | Reference | 0.97 (0.89, 1.06) | 0.92 (0.84, 1.02) | 0.94 (0.85, 1.04) | 0.92 (0.82, 1.03) | 0.16 |
| Fatal CAD | ||||||
| Case/person-year | 905/1,099,940 | 714/1,105,066 | 738/1,105,988 | 733/1,106,002 | 813/1,105,534 | |
| Age-adjusted | Reference | 0.71 (0.64, 0.78) | 0.65 (0.59, 0.72) | 0.57 (0.52, 0.63) | 0.56 (0.51, 0.62) | <0.001 |
| Multivariable-adjusted3 | Reference | 0.92 (0.83, 1.02) | 0.94 (0.85, 1.05) | 0.90 (0.81, 1.01) | 0.92 (0.82, 1.04) | 0.25 |
Abbreviations: CAD, coronary artery disease; IQR, interquartile range; MI, myocardial infarction.
Values are hazard ratio (95% CI) unless otherwise indicated. Hazard ratios were pooled using fixed-effect models.
Median value in each quintile category was used to calculate P value for trend.
Data were combined from 3 cohorts to calculate the median (IQR).
Models were age-stratified (months) and calendar time–stratified and adjusted for ethnicity (White, African American, Asian, and others), smoking status (never smoked, past smoker, currently smoke 1–14 cigarettes per day, 15–24 cigarettes per day, or ≥25 cigarettes per day), time-varying body mass index (<21.0, 21.0–22.9, 23.0–24.9, 25.0–26.9, 27.0–29.9, 30.0–32.9, 33.0–34.9, or ≥35.0 kg/m2), alcohol intake (0, 0.1–4.9, 5.0–9.9, 10.0–14.9, 15.0–29.9, and ≥30.0 g/d), multivitamin use (yes/no), use of cholesterol-lowering medication (yes/no), physical activity (quintiles), total energy intake (quintiles), dietary transfat (quintiles), saturated fatty acid (quintiles), red meat and processed meat (quintiles), total dairy intake (quintiles), refined grains (quintiles), fish (quintiles), dietary cholesterol intake (quintiles), and family history of myocardial infarction. For female, postmenopausal hormone use (premenopausal, never, former, current, or missing), and oral contraceptive use (NHSII only; ever used, never used) were further adjusted.
In the restricted cubic spline analysis using the minimum value as the reference point, the risk reduction in total CAD plateaued after ∼180 mg/d of total phytosterol intake, ∼30 mg/d of campesterol intake, and ∼130 mg/d of β-sitosterol intake (all P for nonlinearity <0.05) (Figure 1). When using the first knot as the reference intake, we observed similar results (Supplemental Figure 6). When stratified by cohorts, the inverse associations among total phytosterols, campesterol, and β-sitosterol and total CAD risk were stronger among the NHS cohort than those with HPFS and NHSII cohorts (Supplemental Table 4). The risk reduction in total CAD comparing extreme quintiles of total phytosterol intake was significant in NHS [0.83 (0.74, 0.93)], but not in NHSII [1.05 (0.81, 1.35)] or HPFS [0.96 (0.86, 1.07)] (Supplemental Table 4).
FIGURE 1.
Dose–response relationships between phytosterol intake and total CAD risk (N = 213,992). The graph was the multivariate analysis of phytosterol levels and CAD risk. The highest and lowest 1 percentile of dietary phytosterol intake was excluded. Minimum values after exclusion were used as the reference point (114.5 mg for total phytosterol, 12.2 mg for stigmasterol, 16.8 mg for campesterol, 82.6 mg for β-sitosterol). The analysis was age-stratified (months) and calendar time–stratified and adjusted for ethnicity (White, African American, Asian, and others), smoking status (never smoked, past smoker, currently smoke 1–14 cigarettes per day, 15–24 cigarettes per day, or ≥25 cigarettes per day), time-varying body mass index (<21.0, 21.0–22.9, 23.0–24.9, 25.0–26.9, 27.0–29.9, 30.0–32.9, 33.0–34.9, or ≥35.0 kg/m2), alcohol intake (0, 0.1–4.9, 5.0–9.9, 10.0–14.9, 15.0–29.9, and ≥30.0 g/d), multivitamin use (yes/no), use of cholesterol-lowering medication (yes/no), physical activity (quintiles), total energy intake (quintiles), dietary transfat (quintiles), saturated fatty acid (quintiles), red meat and processed meat (quintiles), total dairy intake (quintiles), refined grains (quintiles), fish (quintiles), family history of myocardial infarction, and data origin (NHS, NHSII, or HPFS). For female, postmenopausal hormone use (premenopausal, never, former, current, or missing), and oral contraceptive use (NHSII only; ever used/never used) were further adjusted.
Abbreviations: CAD, coronary artery disease; NHS, Nurses’ Health Study; HPFS, Health Professional Follow-up Study.
In the substitution analysis, replacing 100-mg dietary cholesterol with total phytosterols was associated with a 10% reduction in the total CAD risk (HR: 0.90; 95% CI: 0.84, 0.95). For individual phytosterols, replacing 100-mg dietary cholesterol with β-sitosterol was associated with 12% (HR: 0.88; 95% CI: 0.81, 0.95) lower risks of total CAD, and replacing 30-mg dietary cholesterol with dietary campesterol was associated with 10% (HR: 0.90; 95% CI: 0.80, 1.00) lower risks of total CAD. However, replacing 30-mg dietary cholesterol with stigmasterol was not associated with CAD risk (HR: 0.91; 95% CI: 0.79, 1.04) (Table 3).
TABLE 3.
Change in hazard ratios for the associations of the substitution of 100 mg dietary cholesterol with dietary phytosterol with risks of developing CAD (N = 213,992)∗
| Total CAD risk | Hazard ratios (95% CI) |
|---|---|
| Substitute 100-mg dietary cholesterol with total dietary phytosterol1 | 0.90 (0.84, 0.95) |
| Substitute 100-mg dietary cholesterol with dietary β-sitosterol1 | 0.88 (0.81, 0.95) |
| Substitute 30-mg dietary cholesterol with dietary campesterol1 | 0.90 (0.80, 1.00) |
| Substitute 30-mg dietary cholesterol with dietary stigmasterol1 | 0.91 (0.79, 1.04) |
Abbreviations: CAD, coronary artery disease; NHS, Nurses’ Health Study; HPFS, Health Professional Follow-up Study.
1Hazard ratios were pooled using fixed-effect models.
Models were age-stratified (months) and calendar time–stratified and adjusted for ethnicity (White, African American, Asian, and others), smoking status (never smoked, past smoker, currently smoke 1–14 cigarettes per day, 15–24 cigarettes per day, or ≥25 cigarettes per day), time-varying body mass index (<21.0, 21.0–22.9, 23.0–24.9, 25.0–26.9, 27.0–29.9, 30.0–32.9, 33.0–34.9, or ≥35.0 kg/m2), alcohol intake (0, 0.1–4.9, 5.0–9.9, 10.0–14.9, 15.0–29.9, and ≥30.0 g/d), multivitamin use (yes/no), use of cholesterol-lowering medication (yes/no), physical activity (quintiles), total energy intake (quintiles), dietary transfat (quintiles), saturated fatty acid (quintiles), red meat and processed meat (quintiles), total dairy intake (quintiles), refined grains (quintiles), fish (quintiles), and family history of myocardial infarction. For female, postmenopausal hormone use (premenopausal, never, former, current, or missing) and oral contraceptive use (NHSII only; ever used/never used) were further adjusted.
Among participants with hypercholesterolemia, the associations between total phytosterols and individual phytosterol and total CAD risk were similar among those with and without taking cholesterol-lowering medications (all P-interactions > 0.05) (Table 4).
TABLE 4.
Hazard ratios (95% confidence intervals) for the associations of dietary phytosterol with risks of developing CAD and potential interaction with intakes of cholesterol-lowering medications among people who had hypercholesterolemia (N = 128,881)∗
| Total CAD risk | Q1 | Q2 | Q3 | Q4 | Q5 | P-interaction1 |
|---|---|---|---|---|---|---|
| Total phytosterol | ||||||
| Not taking cholesterol-lowering medications | Reference | 0.91 (0.79, 1.05) | 0.81 (0.70, 0.94) | 0.87 (0.76, 1.01) | 0.91 (0.78, 1.06) | 0.50 |
| Taking cholesterol-lowering medications | Reference | 0.99 (0.85, 1.15) | 0.98 (0.84, 1.15) | 0.91 (0.78, 1.07) | 0.91 (0.77, 1.08) | |
| Campesterol | ||||||
| Not taking cholesterol-lowering medications | Reference | 0.96 (0.83, 1.09) | 0.90 (0.78, 1.03) | 0.84 (0.73, 0.97) | 0.93 (0.81, 1.08) | 0.37 |
| Taking cholesterol-lowering medications | Reference | 0.89 (0.76, 1.03) | 0.92 (0.80, 1.07) | 0.83 (0.71, 0.96) | 0.84 (0.72, 0.98) | |
| Stigmasterol | ||||||
| Not taking cholesterol-lowering medications | Reference | 1.00 (0.87, 1.15) | 1.04 (0.90, 1.20) | 0.99 (0.86, 1.14) | 0.98 (0.85, 1.13) | 0.57 |
| Taking cholesterol-lowering medications | Reference | 1.02 (0.87, 1.19) | 0.91 (0.77, 1.06) | 1.02 (0.87, 1.18) | 1.07 (0.92, 1.25) | |
| β-Sitosterol | ||||||
| Not taking cholesterol-lowering medications | Reference | 0.86 (0.75, 0.99) | 0.79 (0.68, 0.91) | 0.85 (0.73, 0.98) | 0.91 (0.78, 1.06) | 0.43 |
| Taking cholesterol-lowering medications | Reference | 1.05 (0.90, 1.22) | 1.00 (0.86, 1.17) | 0.92 (0.80, 1.10) | 0.92 (0.78, 1.10) | |
Abbreviations: CAD, coronary artery disease; NHS, Nurses’ Health Study; HPFS, Health Professional Follow-up Study.
Values are hazard ratio (95% CI) unless otherwise indicated. Models were age-stratified (months) and calendar time–stratified and adjusted for ethnicity (White, African American, Asian, and others), smoking status (never smoked, past smoker, currently smoke 1–14 cigarettes per day, 15–24 cigarettes per day, or ≥25 cigarettes per day), time-varying body mass index (<21.0, 21.0–22.9, 23.0–24.9, 25.0–26.9, 27.0–29.9, 30.0–32.9, 33.0–34.9, or ≥35.0 kg/m2), alcohol intake (0, 0.1–4.9, 5.0–9.9, 10.0–14.9, 15.0–29.9, and ≥30.0 g/d), multivitamin use (yes/no), physical activity (quintiles), total energy intake (quintiles), dietary transfat (quintiles), saturated fatty acid (quintiles), red meat and processed meat (quintiles), total dairy intake (quintiles), refined grains (quintiles), fish (quintiles), family history of myocardial infarction, and data origin (NHS, NHSII, or HPFS). For female, postmenopausal hormone use (premenopausal, never, former, current, or missing), and oral contraceptive use (NHSII only; ever used/never used) were further adjusted.
Median value in each quintile category was used to calculate P value for trend.
In the sensitivity analysis that mutually adjusted for 3 individual phytosterols, the significant associations of campesterol with total CAD and fatal CAD did not change, whereas the association between β-sitosterol and total CAD were attenuated to null (Supplemental Table 5). In the analysis that further stopped updating diet after the diagnosis of hypercholesterolemia, associations between total phytosterols and individual phytosterol and CAD risk were similar (Supplemental Table 6 and Supplemental Figure 7).
Results of associations between total phytosterol and blood lipids and other CVD risk markers are summarized in Table 5. Higher total phytosterol intake was inversely associated with total cholesterol (−0.023 mmol/L per 50 mg/d; P = 0.02), triglycerides (−0.031 mmol/L; P = 0.01) and HDL cholesterol (−0.012 mmol/L per 50 mg/d; P = 0.02). Total phytosterols were not significantly associated with lower concentrations of LDL cholesterol (−0.004 mmol/L per 50 mg/d; P = 0.74). In addition, higher total phytosterol intake was inversely associated with concentrations of IL-6 (−0.081 pg/mL per 50 mg/d; P < 0.001) and was positively associated with adiponectin (0.302 μg/mL per 50 mg/d; P < 0.001). Total phytosterols were not significantly associated with lower concentrations of hs-CRP (−0.085 mg/L per 50 mg/d; P = 0.08). The association of individual phytosterols with these markers were similar for campesterol and β-sitosterol (Supplemental Table 7).
TABLE 5.
Multiple linear regression analysis of associations between dietary phytosterols and blood levels of lipids, inflammation markers, and anti-inflammation markers∗
| Sample size | β per 50-mg/d phytosterol | P | |
|---|---|---|---|
| Total cholesterol (mmol/L) | 22,039 | −0.023 (−0.042, −0.005) | 0.02 |
| LDL cholesterol (mmol/L) | 11,983 | −0.004 (−0.030, 0.022) | 0.74 |
| HDL cholesterol (mmol/L) | 15,030 | −0.012 (−0.022, −0.002) | 0.02 |
| Triglycerides (mmol/L) | 16,374 | −0.031 (−0.052, −0.009) | 0.01 |
| hs-CRP (mg/L) | 21,550 | −0.085 (−0.181, 0.011) | 0.08 |
| IL-6 (pg/mL) | 14,375 | −0.081 (−0.128, −0.033) | <0.001 |
| Adiponection (μg/mL) | 18,547 | 0.302 (0.164, 0.441) | <0.001 |
Abbreviations: hs-CRP, high-sensitivity C-reactive protein; NHS, Nurses’ Health Study; HPFS, Health Professional Follow-up Study.
Values are β coefficient (95% CI) unless otherwise indicated. Models were age-stratified (months) and calendar time–stratified and adjusted for case–control study design, ethnicity (White, African American, Asian, and others), smoking status (never smoked, past smoker, currently smoke 1–14 cigarettes per day, 15–24 cigarettes per day, or ≥25 cigarettes per day), time-varying body mass index (<21.0, 21.0–22.9, 23.0–24.9, 25.0–26.9, 27.0–29.9, 30.0–32.9, 33.0–34.9, or ≥35.0 kg/m2), alcohol intake (0, 0.1–4.9, 5.0–9.9, 10.0–14.9, 15.0–29.9, and ≥30.0 g/d), multivitamin use (yes/no), use of cholesterol-lowering medication (yes/no), physical activity (quintiles), total energy intake (quintiles), dietary transfat (quintiles), saturated fatty acid (quintiles), red meat and processed meat (quintiles), total dairy intake (quintiles), refined grains (quintiles), fish (quintiles), family history of myocardial infarction, and data origin (NHS, NHSII, or HPFS). For female, postmenopausal hormone use (premenopausal, never, former, current, or missing), and oral contraceptive use (NHSII only; ever used/never used) were further adjusted.
Discussion
In 3 prospective cohorts of 213,992 participants, higher long-term intake of total phytosterols and 2 individual phytosterols (campesterol and β-sitosterol) was associated with moderate CAD risk reduction in a nonlinear fashion. The risk reduction in CAD plateaued at intakes above ∼180, 30, and 130-mg/d, respectively. In addition, phytosterols intake was inversely associated with plasma concentrations of total cholesterol, triglycerides, HDL cholesterol, and IL-6 and were positively associated with adiponectin concentrations. Phytosterols were not significantly associated with LDL cholesterol or hs-CRP.
Existing guidelines recommend 2000 mg/d of phytosterols for dyslipidemia management [28,29]. This intake amount may lead to a 10% decrease in circulating LDL cholesterol concentrations [28], which is associated with an estimated 9% potential reduction in CAD risk according to data from statin trials [30]. However, the diet usually only provides 200–600 mg/d of phytosterols [2]. Observational studies, such as EPIC-NL study [16], WHI study [17], and our research, all addressed phytosterol intake at this modest amounts with the average intake of 296 mg/d (range: 83–966 mg/d), 313 mg/d (range: NA), and 193 mg/d (range: 7–820 mg/d), respectively. In the EPIC-NL, the authors observed a nonsignificant inverse association for total phytosterol intake [16]. However, the HRs for CAD comparing extreme phytosterol categories in the EPIC-NL [16], WHI [17], and this study were similar, at 0.84 (95% CI: 0.70, 1.01), 0.83 (95% CI: 0.73, 0.95), and 0.93 (95% CI: 0.86, 1.01), respectively. In comparison with the EPIC-NL and WHI studies, our research offered a dynamic view of phytosterol intake trends over time, such as major individual phytosterols, and their association with CAD risk. Interestingly, we found that the inverse associations of total phytosterols, campesterol, and β-sitosterol with CAD risk plateaued at higher intake amounts, suggesting that even modest phytosterol consumption may confer minor cardiovascular health benefits. Further studies are needed to confirm these nonlinear associations.
A recent meta-analysis of 124 human studies has found that intakes of phytosterols between 600–1100 mg/d and 3,300 mg/d could lower LDL cholesterol concentrations by 5%–12.4% [3], suggesting a potential dose–response relationship. The EPIC-NL study observed an inverse association of higher intake of total phytosterols with lower total cholesterol (−0.06 mmol/L per 50 mg/d; P = 0.038) and lower LDL cholesterol (−0.07 mmol/L; P = 0.007) [16]. In comparison, this study found a weaker association of phytosterol with total cholesterol (−0.023 mmol/L per 50 mg/d; P = 0.02), and the association with LDL cholesterol did not reach statistical significance (−0.004 mmol/L per 50 mg/d; P = 0.74). Interestingly, previous studies also found that the intake of phytosterol exceeding 3000 mg/d has little additional benefit for lowering LDL concentrations [3]. Previous studies also showed that combining phytosterol and cholesterol-lowering medication (eg, statins) may lead to an additive LDL-lowering effect [28,31]. Taken together, the existing evidence suggests that phytosterol intake may have only a weak effect on lowering blood cholesterol concentrations unless the intake reaches a much higher amount than typical intake amounts in Western diets.
The key mechanism for the cardioprotective function of phytosterols is postulated to be the partial inhibition of intestinal absorption of dietary and biliary cholesterol [32]. It has been estimated that ∼2000 mg/d phytosterols could reduce cholesterol absorption by 30%–40% [28]. However, whether this effect can be translated into CAD risk reduction is uncertain because the association between dietary cholesterol and CAD risk has not been established in epidemiologic studies [33,34]. It is worth mentioning that the effects of phytosterol intake on blood lipid concentrations might be variable and inconsistent, ranging from LDL cholesterol reduction to elevated concentrations LDL cholesterol or triglycerides at 2000 mg/d intake amount [35,36]. In this study, we did not observe a significant association between phytosterol intake and LDL cholesterol. Meanwhile, in vitro studies and experimental animal models have demonstrated the anti-inflammatory properties of phytosterols [37], although short-term randomized double-blind trials ranging between 3 and 16 wks among healthy and hypercholesterolemic participants generated mixed results [4,[38], [39], [40], [41]]. Our results found that dietary phytosterols were inversely associated with proinflammatory markers (hs-CRP and IL-6) and positively associated with the anti-inflammatory marker (adiponectin). Finally, individuals with sitosterolemia, a rare autosomal recessive inherited disorder, have elevated circulating phytosterol concentrations (50-fold to 100-fold) on phytosterol consumption and may display premature coronary artery disease and severe cardiovascular disease [42,43]. Collectively, the current evidence suggests that the inter-relationship between phytosterol intake, blood lipid concentrations, and CAD risk can be complex, and future studies are warranted to further elucidate the mechanisms through which phytosterol intake may modulate CAD risk.
Study strengths and limitations
The main strengths of this study are the repeated measurements of 3 individual phytosterols, dietary intake, and lifestyle factors over a long follow-up of >30 y and the large sample size and the comprehensive adjustment of potential and established risk factors. Several limitations merit consideration. First, measurement errors of dietary phytosterol assessment are inevitable. However, the measurement errors are independent of CAD ascertainment and are, thus, more likely to be nondifferential and attenuate the study estimates toward the null. In addition, the use of cumulatively averaged intake to capture the long-term intake of phytosterol would decrease the random measurement errors. Second, the intake of phytosterol in the current population is much lower than the 2000-mg/d recommended dose by dietary guidelines, and the FFQ did not collect information on the intake of phytosterol supplementation. These limitations prevented us from evaluating potential further CAD risk reduction at higher phytosterol intake. Third, phytosterols comprised plant sterols and stanols. This study only examined major types of sterols because we did not have data of stanols or other sterols. Fourth, our study is observational in nature, thus we cannot exclude the role of residual and unmeasured confounding in the observed associations, especially given the inconsistent associations of phytosterol intake with LDL and HDL cholesterol concentrations and CAD risk, respectively. Finally, our population included primarily non-Hispanic White health professionals; therefore, the generalizability of our findings to other racial/ethnic groups may be limited.
In conclusion, observations from 3 large prospective cohorts in the United States suggest that higher long-term intakes of total and individual phytosterols, such as campesterol and β-sitosterol, may be associated with a modest and nonlinear risk reduction in total CAD, with the potential benefits appearing to plateau at certain intake amounts. Our findings align with the recommendation of adhering to healthy plant-based dietary patterns that emphasize higher consumption of healthy phytosterol-containing foods such as vegetables, fruits, nuts, seeds, and legumes. The role of phytosterols from natural sources and enriched food sources and phytosterol supplementation, in preventing CAD risk needs further investigation among population-based cohort studies with higher phytosterol intake and in randomized clinical trials.
Author contributions
The authors’ responsibilities were as follows – YW, QS: designed the research; JEM, EBR, QS: involved in acquisition of data; YW, BL: conducted research and analyzed data; YW: drafted the manuscript; all authors: provided interpretation of data, critically revised the manuscript for important intellectual content, and have approved the final manuscript.
Data availability
Data described in this article may be made available on request pending on application to and approval by the Channing Division of Network Medicine. Further information including the procedures to obtain and access data from the Nurses’ Health Studies and Health Professionals Follow-Up Study is described at https://www.nurseshealthstudy.org/researchers (contact e-mail: nhsaccess@channing.harvard.edu) and https://sites.sph.harvard.edu/hpfs/for-collaborators/.
Conflict of interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Footnotes
Supplementary data to this article can be found online at https://doi.org/10.1016/j.ajcnut.2023.11.020.
Appendix A. Supplementary data
The following are the Supplementary data to this article:
References
- 1.Genser B., Silbernagel G., De Backer G., Bruckert E., Carmena R., Chapman M.J., et al. Plant sterols and cardiovascular disease: a systematic review and meta-analysis. Eur. Heart J. 2012;33:444–451. doi: 10.1093/eurheartj/ehr441. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Trautwein E.A., McKay S. The role of specific components of a plant-based diet in management of dyslipidemia and the impact on cardiovascular risk. Nutrients. 2020;12:2671. doi: 10.3390/nu12092671. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Ras R.T., Geleijnse J.M., Trautwein E.A. LDL-cholesterol-lowering effect of plant sterols and stanols across different dose ranges: a meta-analysis of randomised controlled studies. Br. J. Nutr. 2014;112:214–219. doi: 10.1017/S0007114514000750. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Jenkins D.J., Kendall C.W., Marchie A., Faulkner D.A., Wong J.M., de Souza R., et al. Effects of a dietary portfolio of cholesterol-lowering foods vs lovastatin on serum lipids and C-reactive protein. JAMA. 2003;290:502–510. doi: 10.1001/jama.290.4.502. [DOI] [PubMed] [Google Scholar]
- 5.Micallef M.A., Garg M.L. Anti-inflammatory and cardioprotective effects of n-3 polyunsaturated fatty acids and plant sterols in hyperlipidemic individuals. Atherosclerosis. 2009;204:476–482. doi: 10.1016/j.atherosclerosis.2008.09.020. [DOI] [PubMed] [Google Scholar]
- 6.Clifton P.M., Mano M., Duchateau G.S., van der Knaap H.C., Trautwein E.A. Dose-response effects of different plant sterol sources in fat spreads on serum lipids and C-reactive protein and on the kinetic behavior of serum plant sterols. Eur. J. Clin. Nutr. 2008;62:968–977. doi: 10.1038/sj.ejcn.1602814. [DOI] [PubMed] [Google Scholar]
- 7.Pasternak R.C., Smith S.C., Jr., Bairey-Merz C.N., Grundy S.M., Cleeman J.I., Lenfant C. ACC/AHA/NHLBI clinical advisory on the use and safety of statins. Circulation. 2002;106:1024–1028. doi: 10.1161/01.cir.0000032466.44170.44. [DOI] [PubMed] [Google Scholar]
- 8.Koenig W. Inflammation and coronary heart disease: an overview. Cardiol. Rev. 2001;9:31–35. doi: 10.1097/00045415-200101000-00007. [DOI] [PubMed] [Google Scholar]
- 9.Piepoli M.F., Hoes A.W., Agewall S., Albus C., Brotons C., Catapano A.L., et al. 2016 European Guidelines on cardiovascular disease prevention in clinical practice: the Sixth Joint Task Force of the European Society of Cardiology and Other Societies on Cardiovascular Disease Prevention in Clinical Practice (constituted by representatives of 10 societies and by invited experts) Developed with the special contribution of the European Association for Cardiovascular Prevention & Rehabilitation (EACPR) Eur. Heart J. 2016;37:2315–2381. doi: 10.1093/eurheartj/ehw106. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Jellinger P.S., Handelsman Y., Rosenblit P.D., Bloomgarden Z.T., Fonseca V.A., Garber A.J., et al. American Association of Clinical Endocrinologists and American College of Endocrinology Guidelines for Management of Dyslipidemia and Prevention of Cardiovascular Disease. Endocr. Pract. 2017;23:1–87. doi: 10.4158/EP171764.APPGL. [DOI] [PubMed] [Google Scholar]
- 11.Escurriol V., Cofán M., Moreno-Iribas C., Larrañaga N., Martínez C., Navarro C., et al. Phytosterol plasma concentrations and coronary heart disease in the prospective Spanish EPIC cohort. J. Lipid Res. 2010;51:618–624. doi: 10.1194/jlr.P000471. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Assmann G., Cullen P., Erbey J., Ramey D.R., Kannenberg F., Schulte H. Plasma sitosterol elevations are associated with an increased incidence of coronary events in men: results of a nested case-control analysis of the Prospective Cardiovascular Münster (PROCAM) study. Nutr. Metab. Cardiovasc. Dis. 2006;16:13–21. doi: 10.1016/j.numecd.2005.04.001. [DOI] [PubMed] [Google Scholar]
- 13.Matthan N.R., Pencina M., LaRocque J.M., Jacques P.F., D’Agostino R.B., Schaefer E.J., et al. Alterations in cholesterol absorption/synthesis markers characterize Framingham offspring study participants with CHD. J. Lipid Res. 2009;50:1927–1935. doi: 10.1194/jlr.P900039-JLR200. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Pinedo S., Vissers M.N., von Bergmann K., Elharchaoui K., Lütjohann D., Luben R., et al. Plasma levels of plant sterols and the risk of coronary artery disease: the prospective EPIC-Norfolk Population Study. J. Lipid Res. 2007;48:139–144. doi: 10.1194/jlr.M600371-JLR200. [DOI] [PubMed] [Google Scholar]
- 15.Rajaratnam R.A., Gylling H., Miettinen T.A. Independent association of serum squalene and noncholesterol sterols with coronary artery disease in postmenopausal women. J. Am. Coll. Cardiol. 2000;35:1185–1191. doi: 10.1016/s0735-1097(00)00527-1. [DOI] [PubMed] [Google Scholar]
- 16.Ras R.T., van der Schouw Y.T., Trautwein E.A., Sioen I., Dalmeijer G.W., et al. Intake of phytosterols from natural sources and risk of cardiovascular disease in the European Prospective Investigation into Cancer and Nutrition-the Netherlands (EPIC-NL) population. Eur. J. Prev. Cardiol. 2015;22:1067–1075. doi: 10.1177/2047487314554864. [DOI] [PubMed] [Google Scholar]
- 17.Glenn A.J., Lo K., Jenkins D.J.A., Boucher B.A., Hanley A.J., Kendall C.W.C., et al. Relationship between a plant-based dietary portfolio and risk of cardiovascular disease: findings from the Women’s Health Initiative Prospective Cohort Study. J. Am. Heart Assoc. 2021;10 doi: 10.1161/JAHA.121.021515. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Fung T.T., Malik V., Rexrode K.M., Manson J.E., Willett W.C., Hu F.B. Sweetened beverage consumption and risk of coronary heart disease in women. Am. J. Clin. Nutr. 2009;89:1037–1042. doi: 10.3945/ajcn.2008.27140. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Schulze M.B., Manson J.E., Ludwig D.S., Colditz G.A., Stampfer M.J., Willett W.C., et al. Sugar-sweetened beverages, weight gain, and incidence of type 2 diabetes in young and middle-aged women. JAMA. 2004;292:927–934. doi: 10.1001/jama.292.8.927. [DOI] [PubMed] [Google Scholar]
- 20.Willett W.C., Sampson L., Stampfer M.J., Rosner B., Bain C., Witschi J., et al. Reproducibility and validity of a semiquantitative food frequency questionnaire. Am. J. Epidemiol. 1985;122:51–65. doi: 10.1093/oxfordjournals.aje.a114086. [DOI] [PubMed] [Google Scholar]
- 21.Salvini S., Hunter D.J., Sampson L., Stampfer M.J., Colditz G.A., Rosner B., et al. Food-based validation of a dietary questionnaire: the effects of week-to-week variation in food consumption. Int. J. Epidemiol. 1989;18:858–867. doi: 10.1093/ije/18.4.858. [DOI] [PubMed] [Google Scholar]
- 22.Rimm E.B., Giovannucci E.L., Stampfer M.J., Colditz G.A., Litin L.B., Willett W.C. Reproducibility and validity of an expanded self-administered semiquantitative food frequency questionnaire among male health professionals. Am. J. Epidemiol. 1992;135:1114–1126. doi: 10.1093/oxfordjournals.aje.a116211. discussion 1127–1136. [DOI] [PubMed] [Google Scholar]
- 23.Curb J.D., McTiernan A., Heckbert S.R., Kooperberg C., Stanford J., Nevitt M., et al. Outcomes ascertainment and adjudication methods in the Women’s Health Initiative. Ann. Epidemiol. 2003;13:S122–S128. doi: 10.1016/s1047-2797(03)00048-6. [DOI] [PubMed] [Google Scholar]
- 24.Rose G.A., Blackburn H. World Health Organization; Geneva: 1982. Cardiovascular survey methods. [PubMed] [Google Scholar]
- 25.Bernstein A.M., Rosner B.A., Willett W.C. Cereal fiber and coronary heart disease: a comparison of modeling approaches for repeated dietary measurements, intermediate outcomes, and long follow-up. Eur. J. Epidemiol. 2011;26:877–886. doi: 10.1007/s10654-011-9626-x. [DOI] [PubMed] [Google Scholar]
- 26.Bernstein A.M., Sun Q., Hu F.B., Stampfer M.J., Manson J.E., Willett W.C. Major dietary protein sources and risk of coronary heart disease in women. Circulation. 2010;122:876–883. doi: 10.1161/CIRCULATIONAHA.109.915165. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Hu F.B., Stampfer M.J., Rimm E., Ascherio A., Rosner B.A., Spiegelman D., et al. Dietary fat and coronary heart disease: a comparison of approaches for adjusting for total energy intake and modeling repeated dietary measurements. Am. J. Epidemiol. 1999;149:531–540. doi: 10.1093/oxfordjournals.aje.a009849. [DOI] [PubMed] [Google Scholar]
- 28.Gylling H., Plat J., Turley S., Ginsberg H.N., Ellegård L., Jessup W., et al. Plant sterols and plant stanols in the management of dyslipidaemia and prevention of cardiovascular disease. Atherosclerosis. 2014;232:346–360. doi: 10.1016/j.atherosclerosis.2013.11.043. [DOI] [PubMed] [Google Scholar]
- 29.National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III), Third report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III) final report. Circulation. 2002;106:3143–3421. [PubMed] [Google Scholar]
- 30.Mihaylova B., Emberson J., Blackwell L., Keech A., Simes J., Barnes E.H., et al. The effects of lowering LDL cholesterol with statin therapy in people at low risk of vascular disease: meta-analysis of individual data from 27 randomised trials. Lancet. 2012;380:581–590. doi: 10.1016/S0140-6736(12)60367-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Han S., Jiao J., Xu J., Zimmermann D., Actis-Goretta L., Guan L., et al. Effects of plant stanol or sterol-enriched diets on lipid profiles in patients treated with statins: systematic review and meta-analysis. Sci. Rep. 2016;6 doi: 10.1038/srep31337. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.De Smet E., Mensink R.P., Plat J. Effects of plant sterols and stanols on intestinal cholesterol metabolism: suggested mechanisms from past to present. Mol. Nutr. Food Res. 2012;56:1058–1072. doi: 10.1002/mnfr.201100722. [DOI] [PubMed] [Google Scholar]
- 33.Berger S., Raman G., Vishwanathan R., Jacques P.F., Johnson E.J. Dietary cholesterol and cardiovascular disease: a systematic review and meta-analysis. Am. J. Clin. Nutr. 2015;102:276–294. doi: 10.3945/ajcn.114.100305. [DOI] [PubMed] [Google Scholar]
- 34.Carson J.A.S., Lichtenstein A.H., Anderson C.A.M., Appel L.J., Kris-Etherton P.M., Meyer K.A., et al. Dietary cholesterol and cardiovascular risk: a science advisory from the American Heart Association. Circulation. 2020;141:e39–e53. doi: 10.1161/CIR.0000000000000743. [DOI] [PubMed] [Google Scholar]
- 35.Alphonse P.A.S., Ramprasath V., Jones P.J.H. Effect of dietary cholesterol and plant sterol consumption on plasma lipid responsiveness and cholesterol trafficking in healthy individuals. Br. J. Nutr. 2017;117:56–66. doi: 10.1017/S0007114516004530. [DOI] [PubMed] [Google Scholar]
- 36.Chen Y., She Y., Kaur R., Guo N., Zhang X., Zhang R., et al. Is plant sterols a good strategy to lower cholesterol? J. Oleo Sci. 2019;68:811–816. doi: 10.5650/jos.ess19116. [DOI] [PubMed] [Google Scholar]
- 37.Vilahur G., Ben-Aicha S., Diaz-Riera E., Badimon L., Padró T. Phytosterols and inflammation. Curr. Med. Chem. 2019;26:6724–6734. doi: 10.2174/0929867325666180622151438. [DOI] [PubMed] [Google Scholar]
- 38.Devaraj S., Autret B.C., Jialal I. Reduced-calorie orange juice beverage with plant sterols lowers C-reactive protein concentrations and improves the lipid profile in human volunteers. Am. J. Clin. Nutr. 2006;84:756–761. doi: 10.1093/ajcn/84.4.756. [DOI] [PubMed] [Google Scholar]
- 39.Jenkins D.J., Kendall C.W., Marchie A., Faulkner D.A., Wong J.M., de Souza R., et al. Direct comparison of a dietary portfolio of cholesterol-lowering foods with a statin in hypercholesterolemic participants. Am. J. Clin. Nutr. 2005;81:380–387. doi: 10.1093/ajcn.81.2.380. [DOI] [PubMed] [Google Scholar]
- 40.Madsen M.B., Jensen A.M., Schmidt E.B. The effect of a combination of plant sterol-enriched foods in mildly hypercholesterolemic subjects. Clin. Nutr. 2007;26:792–798. doi: 10.1016/j.clnu.2007.05.008. [DOI] [PubMed] [Google Scholar]
- 41.Hansel B., Nicolle C., Lalanne F., Tondu F., Lassel T., Donazzolo Y., et al. Effect of low-fat, fermented milk enriched with plant sterols on serum lipid profile and oxidative stress in moderate hypercholesterolemia. Am. J. Clin. Nutr. 2007;86:790–796. doi: 10.1093/ajcn/86.3.790. [DOI] [PubMed] [Google Scholar]
- 42.Salen G., Horak I., Rothkopf M., Cohen J.L., Speck J., Tint G.S., et al. Lethal atherosclerosis associated with abnormal plasma and tissue sterol composition in sitosterolemia with xanthomatosis. J. Lipid Res. 1985;26:1126–1133. [PubMed] [Google Scholar]
- 43.Payne M.Y. Too young to be having a heart attack. Lancet. 2001;358:S64. doi: 10.1016/S0140-6736(01)07076-3. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data described in this article may be made available on request pending on application to and approval by the Channing Division of Network Medicine. Further information including the procedures to obtain and access data from the Nurses’ Health Studies and Health Professionals Follow-Up Study is described at https://www.nurseshealthstudy.org/researchers (contact e-mail: nhsaccess@channing.harvard.edu) and https://sites.sph.harvard.edu/hpfs/for-collaborators/.