https://orcid.org/0000-0002-1860-4764
Dear Editor,
We wish to thank H. Hemilä and E. Chalker for their Letter to the Editor, “Errors in a meta-analysis on vitamin C and COVID-19,” which provided valuable feedback and suggestions on our recent publication, which prompted us to thoroughly verify key data and re-evaluate relevant literature.1 Their insights have been critical in enhancing the rigor of our study.2
First, as Hemilä and Chalker suggested, a large randomized study published by Adhikari et al. in 2023 was not included in the assessment of mortality, an omission which could be detrimental to the findings. However, this article consisted of 2 independent trials. One trial, NCT02735707, which was registered in 2015, started in 2016 and was supplemented in 2020, did not explicitly distinguish that the patients treated with vitamin C were diagnosed with COVID-19.3,4 The inclusion criteria for patients in NCT02735707 were “community-acquired pneumonia, influenza, and COVID-19.”3,4 Although Adhikari et al. reported the outcomes of organ support-free days and hospital survival, the data on mortality from a definitive diagnosis of COVID-19 could not be identified and extracted in this article, and therefore this study was excluded from the analysis.3
Three trials have suggested that higher doses of vitamin C may induce a rebound effect, especially in critically ill patients.5 Adhikari's study found no benefit from 4 days of high-dose intravenous vitamin C on survival in patients with COVID-19, which indicated that the potential rebound effect may have been overlooked.5 Meta-analyses of large numbers of controlled trials have shown that vitamin C has beneficial effects on unspecified respiratory virus infections,5,6 and there is also evidence that the benefits extend to COVID-19.7 In addition, a recent study published in Turkish showed that high-dose vitamin C supplementation can reduce the mortality.8
Second, for the double counting of evidence in meta-analyses, we have now selected the most suitable outcome for studies presenting mortality, mechanical ventilation, and disease severity and hospitalization duration.9 After correction, our analysis showed that effects of vitamin C on alleviating clinical outcomes in patients with COVID-19 (odds ratio [OR] = 0.76, 95% CI, 0.62-0.94, I2 = 47%, P = .01), which included reducing the mortality risk (OR = 0.65, 95% CI, 0.49-0.86, I2 = 43%, P = .003) and the incidence of severity (OR = 0.63, 95% CI, 0.40-0.99, I2 = 31%, P = .04) in COVID-19 patients (Figure 1), but no shortening of the length of hospitalization (mean difference [MD] = 1.12, 95% CI, −0.32 to 2.57, I2 = 32%, P = .13) (Figure 2) compared with the control group. The findings suggest that vitamin C supplements may have a beneficial effect in reducing the disease severity and mortality of COVID-19 patients. But it is worth noting that these results are based on a limited number of participants, and randomized controlled trials are needed to confirm these findings.
Forest Plot of Treatment Effectiveness by Clinical Outcomes in the COVID-19 Vitamin C Group Compared with the Control Group. (A) Severity. (B) Hospitalization duration (readmission to the intensive care unit within hospitalization duration). (C) Mortality. (D) Mechanical ventilation. M-H indicates Mantel-Haenszel.
Forest Plot of Length of Hospital Stay and Length of ICU Stay in the COVID-19 Vitamin C Group Compared With the Control Group. (1) Length of hospital stay; (2) Length of ICU stay. ICU indicates intensive care unit.
Third, the results for each of the 4 outcomes are presented individually in Figure 1. In each of these clinically relevant outcomes, the administration of vitamin C to patients with COVID-19 may be tended toward a positive outcome or not. Although the clinical significance of combining 4 outcomes is not enough, the pooled effect size results may provide new insight into the overall propensity to develop clinical outcomes.10
Furthermore, we extend our sincerest apologies for the writing error in the total of patients and continuous outcomes. In the Abstract, “6831” should be replaced by “3429” and “OR = 1.16” should be replaced by “MD = 1.16.” In the Statistical Analysis section of Methods, “as were continuous variables” should be replaced by “continuous variables were used as MD and 95% CI.” In the results section on vitamin C supplements and the length of hospital and ICU stays, “OR = 1.16, OR = 1.08, and OR = 1.19” should be replaced by “MD = 1.16, MD = 1.08, and MD = 1.19,” respectively.
Author Contributions
Methodology, M.Q., K.X., Y.T., and Y.G.; validation and statistical analysis, M.Q., K.X., Z.C., and X.W.; writing–original draft preparation, M.Q. and K.X.; writing–review and editing, H.Z. and X.M.; supervision and project administration, X.M. All authors have read and approved the final manuscript.
Funding
This manuscript received no funding.
Conflicts of Interest
None declared.
