Cocoa and Blood Pressure: What the Cochrane Review Found

If you have read that dark chocolate lowers blood pressure, the claim almost certainly traces back to a set of small trials that a Cochrane group pulled together in 2017. Cochrane reviews are the closest thing nutrition has to a final word, so it is worth knowing what this one found, and how much weight it can bear.

What was reviewed

The review, Ried, Fakler and Stocks, Cochrane Database of Systematic Reviews 2017, gathered randomized trials that gave adults a flavanol-rich cocoa product (dark chocolate, cocoa powder or a cocoa drink) for at least two weeks and compared it with a flavanol-free or low-flavanol control. It ended up with 35 trials and 1,804 participants, most of them healthy. The trials ran from two to 18 weeks, nine weeks on average.

What it found

Cocoa lowered blood pressure by a small amount: 1.76 mmHg systolic and 1.76 mmHg diastolic on average. That is a real, statistically detectable effect. It is also small, and the authors describe it as small themselves.

Three details matter more than the headline number.

First, the effect depended on where people started. In participants who already had high blood pressure, systolic pressure fell by about 4 mmHg. In participants with normal blood pressure, there was no significant change. The difference between those groups was only borderline significant, so the authors flag it as something to confirm rather than a settled finding.

Second, the doses were high. Trials gave between 30 and 1,218 mg of flavanols a day, 670 mg on average, in anywhere from 1.4 to 105 grams of cocoa product. A typical bar of commercial dark chocolate sits toward the low end of that range, and milk chocolate supplies very little.

Third, when the authors excluded trials with authors employed by the sponsoring industry, the effect got slightly smaller. They read that as a sign of some reporting bias. Trials where participants knew what they were getting also trended toward a bigger effect than blinded ones, which raises the question of expectation.

What it does not show

The trials were short. None can tell you whether the effect holds up over years, or whether it changes the rate of strokes or heart attacks. The authors say long-term trials on clinical outcomes are still needed. Because of unexplained variation between trials, they downgraded the quality of the evidence from high to moderate.

Cocoa was well tolerated. About 1 percent of people on cocoa reported stomach complaints or nausea, against 0.4 percent on the control.

What this means for what you eat

A small reduction in blood pressure is worth having, and dark chocolate with a high cocoa percentage is a reasonable food to include if you like it. It is not a treatment for anything, and it does not replace a conversation with your doctor about a blood pressure reading you are worried about. Cocoa flavanols are interesting. The effect is around two points, the evidence is moderate, and it stops at 18 weeks.