Raynaud's phenomenon and diet

What the research found.

• One double-blind trial tested fish oil against olive oil in people who had Raynaud's phenomenon. The median time before cold-induced symptoms began rose from 31.3 minutes at baseline to 46.5 minutes at six weeks on it. That difference came out at a P value of 0.04 in the trial.

  DiGiacomo and colleagues, American Journal of Medicine, 1989

• In that trial, 5 of 11 people with primary Raynaud's on fish oil couldn't be induced to develop symptoms at week 6 or 12. On olive oil it was 1 of 9. The benefit was seen in primary Raynaud's rather than in secondary Raynaud's, which is the detail most often dropped.

  DiGiacomo and colleagues, American Journal of Medicine, 1989

• In a randomized placebo-controlled trial, a plant extract reduced attacks per week from 13.2 to 5.8. That's a fall of 56 percent, against 27 percent on the placebo. There were no significant differences in blood flow measures between the two groups, so the mechanism the trial proposed wasn't demonstrated.

  Muir and colleagues, Vascular Medicine, 2002

• One systematic review looked at behavior change interventions for Raynaud's phenomenon, which is the literature sitting next door to this one. It covered eight studies and 495 participants in total. It concluded that there's no evidence either to support or to refute the claim that those interventions work.

  Daniels and colleagues, BMJ Open, 2018

The whole dietary literature is two trials

Everything tested about diet in Raynaud's phenomenon amounts to two randomized trials, one published in 1989 and one in 2002, together enrolling under a hundred people. Both of them reported a positive result of one kind or another. Neither one of them has been repeated at a size that would settle anything in either direction.

That's an unusual position for a page on this site to find itself in. On most subjects the problem is a mass of weak studies pulling in different directions. Here the problem is two isolated positive findings, decades apart, in a condition that affects a great many people indeed.

Both of them deserve reporting with their numbers and their limits, which is what follows below. What neither deserves is being called an established treatment. The distance between those two descriptions is the reason this page is written the way that it is.

The 1989 fish oil trial

A double-blind, controlled, prospective study tested fish oil against olive oil in people with Raynaud's phenomenon. The median time before cold-induced symptoms began rose from 31.3 minutes at baseline to 46.5 minutes at six weeks, at a P value of 0.04. Mean digital systolic pressures in a 10 degree Celsius water bath were 32 mmHg higher, at a P value of 0.02.

The most striking result was that 5 of 11 people with primary Raynaud's taking fish oil couldn't be induced to develop symptoms at all at week 6 or week 12, against 1 of 9 on olive oil, at a P value of 0.05. That's a substantial difference and it rests on twenty people. Twenty is a small enough number that two or three of them moving would change the answer.

The finding that counts most practically is where the benefit appeared. It was seen in primary Raynaud's and not in secondary Raynaud's, meaning the form that accompanies a connective tissue disease. Anyone whose Raynaud's comes with systemic sclerosis, lupus, or another rheumatic diagnosis falls outside the group where this trial found anything, and that detail is routinely dropped when the study is quoted.

The 2002 extract trial

A randomized, double-blind, placebo-controlled trial of a plant extract found the number of attacks per week falling from 13.2 to 5.8, a reduction of 56 percent, against a 27 percent reduction on placebo. The difference was significant at a P value below 0.00001, and attack frequency was the only outcome where treatment beat placebo significantly. That last clause is worth holding onto, because it narrows what the trial showed.

The trial also looked for a mechanism and didn't find one. There were no significant differences in blood rheology between the two groups, meaning the measures of how blood flows didn't separate. So something reduced attacks and the explanation the trial proposed wasn't demonstrated by it.

This page doesn't name the product tested. Putting a purchasable branded supplement on a medical site on the strength of one trial is a decision for the site's owner rather than for whoever writes the page, and the figures are given so you can weigh the evidence either way. If you meet the product elsewhere, you'll know what supports it.

How to read two positive small trials

Small positive trials are the most seductive kind of evidence and the least reliable. Publication favors positive results, small studies produce larger apparent effects than big ones, and a trial of thirty people can be moved substantially by two or three participants having a mild winter. That doesn't mean these results are wrong.

What it means is that the appropriate response is interest rather than confidence. If a supplement is cheap, safe for you, and you have a way of judging whether it's helping, trying one is a reasonable personal experiment. If it's expensive, or interacts with something you take, the evidence here doesn't have enough weight behind it to justify that.

Fish oil at supplement doses affects bleeding, which is the specific caution worth raising with whoever manages your care, particularly if you take anticoagulants. That's a conversation rather than a reason not to, and it's the sort of thing supplement labels don't emphasize. It belongs in the conversation before the bottle is opened rather than after.

What has the best evidence here

Keeping warm, by a wide margin. A randomized comparison of silver-lined against ordinary gloves found the Raynaud's condition score falling from 6.4 to 3.9 on either type, with no difference between them. That's a meaningful improvement from something inexpensive, and the ordinary gloves did as well as the expensive ones.

Stopping smoking has the clearest mechanism available to anyone who smokes, since smoking affects blood vessels directly. It isn't specific to Raynaud's. It is the change with the best-understood reasoning behind it in this whole area.

The behavioral literature, meanwhile, can't support a conclusion. A systematic review of eight studies and 495 participants concluded there's no evidence to support or refute whether behavior change interventions work. Set against that, two small positive supplement trials look better than they otherwise might, and it's worth remembering that's a comparison against a very low bar.

Common misconceptions.

Cautions specific to this condition.

• Tell your rheumatology team before starting fish oil, since at supplement doses it affects bleeding and interacts with anticoagulants.
• Note that the fish oil benefit appeared in primary Raynaud's and not in secondary Raynaud's, so it may not apply if you have a connective tissue disease.
• Attacks vary with the weather and the season, so anything tried without a set period and a way of judging it will look like it worked or failed for reasons of its own.
• Take new Raynaud's after the age of forty to a doctor before spending anything on supplements, because a secondary cause becomes more likely with later onset.
• Mention fingertip sores or ulcers promptly, since they need assessment.
• Keep warm regardless, since gloves of either type reduced the Raynaud's condition score from 6.4 to 3.9 in a randomized comparison.

Questions patients ask.