The homocysteine trials nobody quotes

If you read only the observational literature, the case looks settled: higher homocysteine, more strokes and heart attacks. Lower homocysteine, less risk. Supplement companies stop there.

The intervention trials are where the story gets uncomfortable, and they are the part usually left out.

HOPE-2 (2006)

Folic acid, B6 and B12 for five years in people with vascular disease. Homocysteine fell. The primary composite of cardiovascular death, myocardial infarction and stroke was not significantly reduced. Stroke alone was lower, which was a secondary endpoint — promising, but not the pre-specified question.

VITATOPS (2010)

B vitamins in 8,164 people with a recent stroke or transient ischaemic attack. Homocysteine fell. The primary composite outcome — stroke, myocardial infarction or vascular death — did not differ.

SEARCH (2010)

Folic acid plus B12 in 12,064 survivors of myocardial infarction. Homocysteine fell substantially. Major vascular events did not fall.

The 2017 Cochrane review

Pooling the randomised evidence, homocysteine-lowering interventions did not reduce cardiovascular events. The authors were careful: not "proven ineffective", but "no demonstrated benefit", with wide uncertainty.

So what do we do with a high homocysteine?

Two conclusions can coexist.

First, correct genuine B12 deficiency. Not for the heart — for the nervous system, for the blood count, and because deficiency is a real clinical problem with real consequences. That part of the story is not in dispute.

Second, do not sell vitamins as cardiovascular prevention. The marker is associated with risk; changing the marker has not reliably changed outcomes. Anyone who says otherwise is quoting half the literature.

There is a plausible explanation for the gap: homocysteine may be a marker of vascular damage rather than a cause, the way a fever is a marker of infection. Lowering the thermometer does not treat the infection. Mendelian randomisation studies, which are less vulnerable to reverse causation, have generally found weaker effects than the observational cohorts — consistent with that reading.

What this means for your own decision

If you take the check and land in a high band, the actionable step is still a blood test, and still correcting a confirmed deficiency. What you should not expect is that B12 supplementation lowers your risk of a heart attack. It might. The trials say we do not know yet — and that honest uncertainty is more useful than a promise.

Educational material, not medical advice.