Mini: Phytonutrient supplements for cardiovascular disease markers

A recent umbrella review summarized the effects of plant-based supplements for cardiovascular disease markers. Be careful with directly comparing the numbers they report though!

You are reading a free Examine Deep Dive.
Become an Examine+ member to get full access to the Examine Database, Study Summaries, and Supplement Guides.
Try for free for 7 days and stay on top of the latest research.
Summary

There’s a lot of research out there on supplements for cardiovascular disease (CVD) risk markers like LDL cholesterol (LDL-C) and blood pressure. All this research is for good reason: CVD is the cause of death[1] for around 50% of everyone in the world, and systolic blood pressure, LDL-C, and fasting plasma glucose levels are among the most predictive markers for dying from CVD. Presuming these relationships are causal (which is a pretty[2] good[3] assumption[4]), any intervention that reduces these markers can lower the risk of getting cardiovascular disease and ultimately dying from it.

But there’s an issue with having so much research on this important topic: it can be hard to navigate without losing sight of the big picture. That’s where umbrella reviews come in. Umbrella reviews are designed to summarize all the systematic reviews and meta-analyses on a given topic. One such umbrella review examining the effect of plant-based supplements on major cardiovascular risk markers was recently published[5]. The authors found 50 systematic reviews and meta-analyses looking at clinical trial evidence for a broad range of phytonutrient supplements. Since this is still quite a lot to process, this Mini narrows down their findings to the top five supplements by effect size for three major CVD risk markers in three separate tables below, along with some dosing information.

Figure 1: Plant-based supplements with the biggest effect sizes for lowering LDL-C

SupplementLDL-C reduction (mg/dL)Effects on other blood lipids (mg/dL)Usual dosage
Red yeast rice-33.64 (95% CI: -39.83 to -27.46)- Total cholesterol: -37.51 - Triglycerides: -20.37 - No clear effect on HDL-C200–3,600 mg daily, with most studies dosing around 500–1,200 mg daily
Fenugreek-26.05 (95% CI: -50.39 to -1.70)Total cholesterol: -25.12 - Triglycerides: -32.14 - HDL-C: 3.7310–60 grams of seed powder, or a variety of extracts whose specific dose depended on the extraction method used
Spirulina-19.02 (95% CI: -36.27 to -1.78)- Total cholesterol: -20.5 - No clear effect on triglycerides - HDL-C: 1.42800 mg to 8 grams daily, with 2 grams being the most frequent daily dose
Nigella sativa-18.45 (95% CI: -22.44 to -14.45)- Total cholesterol: -16.80 - Triglycerides: -15.73 - HDL-C: 1.93A wide range of oils, extracts, and powders were used. Roughly, extract doses were around 200–400 mg daily, powders and encapsulated solids were around 500– 2,000 mg daily, and oils were around 5 mL daily or 2,000–3,000 mg of encapsulated oil daily.
Silymarin-17.20 (95% CI: -28.50 to -5.91)- Total cholesterol: -17.22 - Triglycerides: -22.55 - HDL-C: 4.82Usually 280–600 mg daily of extract

Figure 2: Plant-based supplements with the biggest effect sizes on systolic blood pressure

SupplementSystolic blood pressure reduction (mmHg)Effects on diastolic blood pressure (mmHg)Usual dosage
Sour tea (Hibiscus sabdariffa)-7.58 (95% CI: -9.69 to -5.46)-3.53Usually brewed tea using bags containing 1.25–3 grams sour tea 2–3 times daily, or 100 mg daily extract
Ginger−6.36 (95% CI: −11.27 to −1.46)-2.120.5–3 grams ginger powder daily
Garlic−5.07 (95% CI: −7.30 to −2.85)-2.48Supplements standardized to either 7.8–31.2 mg daily of alliin or 0.6–14.7 mg daily of S-allylcysteine
Nigella sativa−3.26 (95% CI: −5.10 to −1.42)-2.800.5–1.6 grams of seed powder daily, or 0.2–3 grams of oil daily
Pycnogenol−3.22 (95% CI: −5.52 to −0.92)-1.9160–200 mg daily

Figure 3: Plant-based supplements with the biggest effect sizes on fasting blood glucose

SupplementFasting glucose reduction (mg/dL)Effects on HbA1c (%)Usual dosage
Aloe vera-30.05 (95% CI: -54.87 to -5.23)n/a (studies that explored fasting glucose did not measure A1c)Usually 1–2.8 grams of daily capsules, with 2 tablespoons juice and 0.2 grams of daily powder also used
Blueberry or cranberry-22.62 (95% CI: -37.72 to -7.52)-0.54Blueberry: about 0.5–1 grams of hydroalcoholic extract or 1 gram of dried powder Cranberry: Either 240 mL of cranberry juice daily, 40 grams of dried cranberry, or 55 grams of raw cranberry
Nettle (Urtica dioica)-18.01 (95% CI: -30.04 to -5.97)-0.771.5–10 grams daily
Nigella sativa15.18 (95% CI: -19.82 to -10.55)-0.45A wide range of oils, extracts, and powders were used. Roughly, extract doses were around 200–400 mg daily, powders and encapsulated solids were around 500–2,000 mg daily, and oils were around 5 mL daily or 2,000–3,000 mg of encapsulated oil daily
Propolis (bee glue)-13.51 (95% CI: -24.98 to -2.04)-0.52226.8–1,500 mg daily

While the tables above are a decent starting point for evaluating the efficacy of these plant-based supplements, there are quite a few problems associated with direct comparisons.

The first is clear from the above table: some of the confidence intervals are very wide (e.g., Aloe vera for fasting glucose, ginger for blood pressure), indicating that the evidence is compatible with both large effects and close to no effect. Second, the heterogeneity of the evidence was very large for most of these values, indicating that studies with large differences in their populations, dosage forms, methodologies, and more were combined, raising the question of how meaningful these averages are. This problem is especially apparent in the meta-analysis that combined blueberry and cranberry supplements. Third, although the authors of this umbrella review rated the meta-analyses as mostly moderate or high quality, the evidence that went into them wasn’t always the best. Sometimes the studies were at high risk of bias, and some of the meta-analyses looked at studies that didn’t have control groups or good randomization. Thus, these values and rankings should be interpreted with caution. Finally, the populations were often quite different from one another. For example, some meta-analyses looking at blood glucose effects involved only people with type 2 diabetes, while others included broader populations.