To determine the size of reduction in homocysteine concentrations produced by dietary supplementation with folic acid and with vitamins B12 or B6.

English and non-English language studies were sought in MEDLINE. The reference lists were scanned and personal contact was made with relevant investigators.

Study designs of evaluations included in the review

Published and unpublished randomised controlled trials (RCTs) were included if they had assessed the effects on blood homocysteine concentrations of folic acid supplements with or without the addition of vitamins B6 or B12. Excluded were studies which did not include an untreated control group, and studies assessing either treatment after methionine loading or patients treated for fewer than 3 weeks.

Specific interventions included in the review

Daily dietary supplementation included folic acid in doses ranging from 0.4 mg to 10 mg, vitamin B12 in doses ranging from 0.02 mg to 1.0 mg (mean 0.5 mg) and vitamin B6 in doses ranging from 2 mg to 50 mg (mean 16.5 mg). Supplementation included the following; folic acid alone, folic acid plus either vitamin B6 or B12, folic acid plus both and placebo.

Participants included in the review

The participants had a mean age of 52 years (range of trial means 23 to 75 years). The median pre-treatment blood concentration of homocysteine was 11.8 micromoles/L and of folate was 11.6 nmol/L.

Outcomes assessed in the review

The main outcome assessed was the reduction in blood homocysteine in the treated group compared with the controls.

How were decisions on the relevance of primary studies made?

The authors do not state how the papers were selected for the review, or how many of the authors performed the selection.

The authors do not state that they assessed validity.

It is not stated how the following data were extracted from the primary studies: age, gender, smoking habits, history of vascular disease or hypertension, vitamin use before randomisation, allocated treatment regime (daily dose of folic acid, vitamin B6 or B12 and duration), blood concentrations of homocysteine, folate, vitamin B12 and B6 before and at the end of the treatment schedule. The proportional reductions in blood homocysteine in the treatment compared to the control groups was calculated.

How were the studies combined?

Multivariate regression analysis was used to determine the effects on homocysteine concentration of different doses of folic acid and the addition of vitamin B12 or B6. The proportionate reductions in blood homocysteine among treatment groups were determined by extending an analysis of covariance estimating the differences in post-treatment, log transformed homocysteine values after adjustment for baseline values of homocysteine. The model selected allowed the extent of the adjustment to vary between studies and to take account of folic acid dose, concomitant vitamin B6 or B12, age, sex and duration of treatment.

How were differences between studies investigated?

Heterogeneity across studies was assessed by multivariate regression analysis that allowed the effect of folic acid supplementation to differ in individual studies.

Twelve RCTs were used to evaluate the reduction in blood homocysteine with dietary supplementation (1,114 participants).

A correlation coefficient of 0.87 was found for homocysteine in pre-treatment blood samples collected on two occasions (664 patients). Blood homocysteine levels were available for 98% of those randomised.

The effect of folic acid on blood homocysteine differed among the studies. This heterogeneity was not explained by differences in age, gender or the duration of treatment. After adjusting for differences in the folic acid regime, the homocysteine-lowering effect of folic acid ranged from a proportional reduction of 16% (95% CI: 11, 20) among patients in the bottom one-fifth of pre-treatment blood homocysteine levels, to a reduction of 39% (95% CI: 36, 43) among the top fifth (P<0.001). The blood homocysteine-lowering effect of folic acid was greater at lower pre-treatment blood folate with a reduction in homocysteine of 37% (95% CI: 33, 40) for those in the lower fifth of pre-treatment folate levels, compared with an 18% reduction (95% CI: 14, 22) for those in the upper fifth (P<0.001). The associations remained significant in a model including pre-treatment homocysteine and folate levels (P<0.001). A sensitivity analysis was conducted by repeating the analysis both after excluding two studies with very high pre-treatment homocysteine levels, and after including two small unpublished trials. Neither of these analyses materially altered the findings.

Heterogeneity between the separate blood homocysteine-lowering effects in the different trials: after adjusting for pre-treatment blood homocysteine and folate levels, the P-values for heterogeneity were 0.15, 0.05 and 0.69 for folic acid less than 1 mg/day, 1 to 3 mg/day, and greater than 3 mg/day, respectively. For a standardised pre-treatment homocysteine level of 12 micro moles/L with folate level 12 nmol/L, folic acid doses of less than 1 mg, between 1 and 3 mg, and greater than 3 mg were each associated with a reduction in blood homocysteine of about one quarter.

Adding vitamin B12 to folic acid was associated with a reduction in blood homocysteine by a further 7%.

Adding vitamin B 6 to folic acid resulted in no further lowering of blood homocysteine.

Typically in Western populations, daily supplementation with both 0.5 mg to 5 mg folic acid and about 0.5 mg vitamin B12 would be expected to reduce blood homocysteine concentrations by about a quarter to a third. Large-scale randomised trials of folic acid and vitamin B12 supplementation in high-risk populations are needed to determine whether lowering blood homocysteine concentrations reduces the risk of vascular disease.

Published and unpublished studies in more than one language and meeting defined inclusion criteria were sought. Analysis included adjusting for potential confounding factors, investigation of heterogeneity and sensitivity analysis. Within-patient variability of blood homocysteine was calculated and the overall withdrawal rate mentioned. The variation in reduction of homocysteine level associated with pre-treatment blood homocysteine and folate levels and with varying doses of folic acid is illustrated graphically.

Limiting the literature search to one database may have resulted in the omission of some relevant studies, though contact with relevant investigators should have reduced the possibility of omissions. Details were omitted of the methods used to select articles for inclusion and to extract data. There is no evaluation of the validity of the included articles and the methods used to statistically assess heterogeneity are not stated. Fuller details of the included trials would have been helpful in determining the characteristics of the populations studied. There was no discussion on the limitations of the review, such as mention of other factors which may affect blood homocysteine levels and the limitations of results from studies that have a maximum duration of 12 weeks.

The authors suggest that large-scale randomised trials of folic acid and vitamin B12 supplementation in high-risk populations are needed to determine whether lowering blood homocysteine concentrations reduces the risk of vascular disease.

British Heart Foundation; Medical Research Council.

Homocysteine Lowering Trialists' Collaboration. Lowering blood homocysteine with folic acid based supplements: meta-analysis of randomised trials. Indian Heart J 2000;52 Suppl:S59-64.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.