Vitamin C May Improve Left Ventricular Ejection Fraction: A Meta-Analysis

Abstract

Background: Vitamin C deprivation can lead to fatigue, dyspnea, oedema and chest pain, which are also symptoms of heart failure (HF). In animal studies vitamin C has improved contractility and mechanical efficiency of the heart. Compared with healthy people, patients with HF have lower vitamin C levels, which are not explained by differences in dietary intake levels, and more severe HF seems to be associated with lower plasma vitamin C levels. This meta-analysis looks at the effect of vitamin C on left ventricular ejection fraction (LVEF).

Methods: We searched for trials reporting the effects of vitamin C on LVEF. We assessed the quality of the trials, and pooled selected trials using the inverse variance, fixed effect options. We used meta-regression to examine the association between the effect of vitamin C on LVEF level and the baseline LVEF level.

Results: We identified 15 trials, three of which were excluded from our meta-analysis. In six cardiac trials with 246 patients, vitamin C increased LVEF on average by 12.0% (95% CI 8.1-15.9%; P < 0.001). In six non-cardiac trials including 177 participants, vitamin C increased LVEF on average by 5.3% (95% CI 2.0-8.5%; P = 0.001). In meta-regression analysis we found that the effect of vitamin C was larger in trials with the lowest baseline LVEF levels with P = 0.001 for the test of slope. The meta-regression line crossed the null effect level at a baseline LVEF level close to 70%, with progressively greater benefit from vitamin C with lower LVEF levels. Some of the included trials had methodological limitations. In a sensitivity analysis including only the four most methodologically sound cardiac trials, the effect of vitamin C was not substantially changed.

Conclusions: In this meta-analysis, vitamin C increased LVEF in both cardiac and non-cardiac patients, with a strong negative association between the size of the vitamin C effect and the baseline LVEF. Further research on vitamin C and HF should be carried out, particularly in patients who have low LVEF together with low vitamin C intake or low plasma levels. Different dosages and different routes of administration should be compared.

Keywords: antioxidant; coronary artery bypass graft surgery (CABG); heart failure; left ventricular function; oxidative stress; percutaneous coronary intervention (PCI); randomized trials; systematic review.

Figure 1

Flow diagram of the searches, which identified 16 publications reporting 15 trials on the effect of vitamin C on LVEF (–53). The search terms are listed in Supplementary File 1.

Figure 2

Risk of bias summary of the trials that reported on the effect of vitamin C on LVEF. Review authors' judgments are shown for each risk of bias item for each included trial. A green plus sign (+) indicates that there is no substantial concern for bias in the particular quality item. A question mark (?) indicates that conclusions are unable to be drawn regarding potential bias. The reference numbers to the trials are shown in Table 1. Justifications for the quality assessments are described in Supplementary Table S1.

Figure 3

Calculation of the effect of vitamin C on LVEF. The calculation is illustrated with figures from the Basili et al. trial during PCI (percutaneous coronary intervention) (43). In the placebo group, LVEF was increased from a baseline level of 53.7 to 54.1% after the intervention, while in the vitamin C group from 52.3 to 58.3%. These correspond to 0.4 and 6.0 percentage point (pp) increases in LVEF, respectively. Thus, subtracting the placebo group change from the vitamin C group change gives the net effect of vitamin C as 5.6 pp. On the relative scale this corresponds to a 10.7% (= 5.6/52.3) increase in LVEF due to vitamin C from the baseline level.

Figure 4

Effect of vitamin C on LVEF. The upper subgroup shows the cardiac trials and the lower subgroup shows the non-cardiac trials. The effect of vitamin C is shown as the percentage difference between the vitamin C and placebo groups from the baseline LVEF; see Figure 3. The horizontal lines indicate the 95% CI for the vitamin C effect and the blue squares in the middle of the horizontal lines indicate the point estimate of the effect in the particular trial. The size of the blue square reflects the weight of the trial in the meta-analysis. The red diamond shape indicates the pooled effect and 95% CI for the two subgroups and for all 12 trials. See Supplementary Files 1, 2 for the description of the trials and the calculations.

Figure 5

Effect of vitamin C on LVEF by the baseline LVEF. The effect of vitamin C is shown as the percentage differences between the vitamin C and control groups from the baseline LVEF. The vertical lines indicate the 95% CI for the vitamin C effect in each trial. The cardiac trials are indicated by filled circles, and the non-cardiac trials by open circles. The red horizontal dotted line indicates the null effect. The diagonal line shows the meta-regression line with P = 0.0008 for the test that the slope is null. There are no indications of residual heterogeneity over the regression line, P = 0.9. For the calculations, see Supplementary File 1.