[Preventive effect of Vitamin B supplementation on recurrent stroke: a Meta-analysis]

To evaluate whether Vitamin B supplementation could prevent ischemic stroke recurrence.

Randomized controlled trials (RCTs) observing Vitamin B supplementation in patients with stroke was performed in databases including ScienceDirect, PubMed/Medline, China National Knowledge Infrastructure, Chinese Biomedical Data-Base, Wanfang Database, and VIP Chinese Science and Technology Journal Database to find related studies in English or Chinese published before August 2016. The patients in control group received a placebo or basic therapy without Vitamin B, and those in experimental group was treated with Vitamin B alone or Vitamin B on the basis of conventional treatment. The data were collected by two researchers independently and the quality of studies was assessed by the modified Jadad Scale. The Meta-analysis was performed using Stata 12.0, funnel plot was drawn, and Egger and Begg regressions were used to evaluate the publication bias, and sensitivity was also analyzed.

Seven RCTs studies were enrolled to analyze with a total number of 9 846 stroke patients, 4 755 patients in control group, and 5 091 in experimental group, respectively. (1) Vitamin B supplementation for prevention of recurrent stroke: heterogeneity test results showed a heterogeneity in literatures enrolled (I² = 62.9%, P = 0.009), and a random effect model was used for Meta-analysis. It was shown that the incidence of recurrent stroke in the experimental group was significantly lower than that in the control group [pooled relative risk (RR) = 0.64, 95% confidence interval (95%CI) = 0.47-0.87], which indicated that the supplementation of Vitamin B could prevent the recurrence of stroke. Cumulative Meta-analysis showed that Vitamin B supplementation exhibited positive effects in the prevention stroke recurrence from 2012. The 95%CI tended to be stable while demonstrating good change trend as sample growing. The publication bias evaluation results showed that the funnel plot was not symmetrical by visual inspection, further quantitative analysis showed that P value from Egger regression was 0.008, while that from Begg regression was 0.035, both P < 0.05, suggesting there were some publication bias. The sensitivity analysis showed that the overall results were stable and reliable. (2) The effect of Vitamin B supplementation on plasma homocysteine (Hcy) levels in stroke patients: heterogeneity test results showed a heterogeneity in literatures enrolled (I² = 96.2%, P = 0.000), and a random effect model was used for Meta-analysis. It was shown that compared with control group, the plasma Hcy levels of patients after Vitamin B supplementation in experimental group were significantly decreased [pooled weighted mean difference (WMD) = -6.92, 95%CI = -9.11 to -4.73), indicating that Vitamin B could significantly reduce plasma Hcy levels in stroke patients. Cumulative meta analysis showed that, as time went on, the relevant research samples were increased, 95%CI tended to be stable and the variation tendency was better. The publication bias evaluation results showed that the funnel plot was symmetry by visual inspection, and further quantitative analysis showed that the P value from Egger regression was 0.345, and that from Begg regression was 0.764, both P > 0.05, which indicating that there was no evidence of publication bias in the study included.

Vitamin B supplementation was associated with a lower risk of recurrent stroke in stroke patients and could significantly improve the quality of secondary prevention of stroke. Furthermore, supplementation of Vitamin B could reduce plasma Hcy levels in stroke patients which might contribute to its effect in preventing stroke recurrence.