Folic acid’s potential to reduce diabetes risk in coronary patients faces conflicting evidence

A new Chinese cohort study suggests folic acid may lower diabetes risk in coronary heart disease patients, but conflicting evidence from trials questions its clinical benefit, prompting calls for dedicated randomised trials.

Folic acid may be linked to a lower risk of developing diabetes in people with coronary heart disease, according to a new Chinese multicentre cohort study, but the finding enters a field where randomised evidence has often been far less persuasive. That matters because folate has long appealed as a cheap, widely available way to lower homocysteine, yet major vascular trials have repeatedly struggled to turn that biochemical effect into a clear clinical gain.

The new study followed patients with coronary heart disease who did not have diabetes at the outset and compared those taking folic acid with those who were not. After exclusions and propensity-score matching, the final analysis covered 488 folic acid users and 1,952 non-users, with a median follow-up of 3.35 years and 510 new diabetes cases. The authors reported a lower adjusted risk among folic acid users, alongside a more favourable HbA1c trend over time. They also found a dose pattern: 0.8 mg a day looked stronger than 0.4 mg, and cumulative exposure above 140 mg was associated with a markedly lower risk. That dose is notable because 0.8 mg a day is also the regimen used in the China Stroke Primary Prevention Trial, one of the best-known folic acid studies in cardiovascular prevention.

Yet one of the closest randomised comparators points in a different direction. A post hoc diabetes analysis from the China Stroke Primary Prevention Trial, published in the Journal of Diabetes, studied 20,702 adults aged 45 to 75 with hypertension who were randomised to enalapril plus 0.8 mg folic acid daily or enalapril alone between May 2008 and August 2013. Among non-diabetic participants, that analysis found no significant reduction in new-onset diabetes over a median 4.5 years: 198 cases in the enalapril-folic acid group and 214 in the enalapril-only group, for a hazard ratio of 0.92. Restricting the analysis to those with baseline fasting glucose below 7 mmol/L did not materially change the result. That is an important point of tension with the new coronary heart disease cohort, suggesting that population, background treatment or study design may be shaping the signal.

Other work from the same Chinese trial network suggests the metabolic backdrop may matter as much as folate itself. A Scientific Reports analysis of 14,864 non-diabetic hypertensive participants from the same trial found that 1,615 developed new-onset diabetes by the end of follow-up. Higher triglycerides and a higher triglyceride-to-HDL ratio were linked to greater risk, while HDL appeared protective. Total cholesterol showed a protective association only in the enalapril-folic acid arm, with an odds ratio of 0.91 compared with 0.98 in the enalapril-only arm, and the interaction was statistically significant. Never-smokers also showed a more favourable pattern. None of that proves folic acid prevents diabetes, but it does suggest any benefit may depend on lipid status, smoking history and the wider cardiometabolic profile rather than applying evenly across patients.

Randomised evidence outside China has been less encouraging still. In the Women’s Antioxidant and Folic Acid Cardiovascular Study, 4,252 women at high cardiovascular risk who were free of diabetes at baseline were assigned either a daily combination pill containing 2.5 mg folic acid, 50 mg vitamin B6 and 1 mg vitamin B12 or placebo. After 7.3 years, investigators recorded 245 incident diabetes cases in the active-treatment group and 259 in the placebo group, equivalent to 11.5% versus 12.2%. The adjusted relative risk was 0.94, which was not statistically significant. The researchers also found no convincing sign that baseline dietary folate, vitamin B6 or vitamin B12 intake altered the overall result.

That fits a broader pattern from cardiovascular outcome trials. A smaller study in 593 patients with stable coronary artery disease found that 0.5 mg folic acid a day lowered plasma homocysteine from 12.0 to 9.4 µmol/l after three months and raised serum folate, but did not reduce the composite of death or vascular events over roughly two years. The larger HOPE-2 trial, which randomised 5,522 patients aged 55 or older with vascular disease or diabetes to folic acid plus vitamins B6 and B12 or placebo for a mean of five years, also found no significant reduction in its primary composite of cardiovascular death, myocardial infarction and stroke. Those events occurred in 18.8% of the active-treatment group and 19.8% of placebo. HOPE-2 did report fewer strokes, but not the broad vascular benefit once hoped for from homocysteine lowering.

The most recent attempt to pull this literature together also argues for restraint. A 2026 umbrella review in the Journal of Global Health examined 49 reviews covering 113 separate associations between folate and cardiovascular, cerebrovascular and metabolic outcomes. It concluded that no association reached a convincing or highly suggestive level of credibility. The strongest signals were only suggestive, including dose-response links between higher dietary folate intake and lower risks of coronary heart disease and stroke. The review also said the evidence base remains limited in both volume and scope, with relatively little work from low- and middle-income countries.

Taken together, the new coronary heart disease paper looks more like a serious hypothesis-generating study than a practice-changing case for routine folic acid prescribing to prevent diabetes. Its apparent dose gradient, particularly at 0.8 mg a day and above a cumulative 140 mg, gives future trials something specific to test. But the mixed record from the China Stroke Primary Prevention Trial diabetes analysis, the null WAFACS result and the disappointing outcome data from HOPE-2 mean the next step has to be a dedicated randomised trial in coronary patients without diabetes, with glycaemic end points set in advance and close attention paid to diet, baseline folate or homocysteine status and the reasons folic acid was prescribed in the first place.