Background: As an immune modulator, vitamin D has been implicated in the coronavirus disease-2019 (COVID-19) outcome. We aim to systematically explore the association of vitamin D serum levels with COVID-19 severity and prognosis.
Methods: The standardized mean difference (SMD) or odds ratio and 95% confidence interval (CI) were applied to estimate pooled results from six studies. The prognostic performance of vitamin D serum levels for predicting adverse outcomes with detection of the best cutoff threshold was determined by receiver operating characteristic curve analysis. Decision tree analysis by combining vitamin D levels and clinical features was applied to predict severity in COVID-19 patients.
Results: Mean vitamin D serum level of 376 patients, was 21.9 nmol/L (95% CI = 15.36-28.45). Significant heterogeneity was found (I2 = 99.1%, p < .001). Patients with poor prognosis (N = 150) had significantly lower serum levels of vitamin D compared with those with good prognosis (N = 161), representing an adjusted standardized mean difference of -0.58 (95% Cl = -0.83 to -0.34, p < .001).
Conclusion: Serum vitamin D levels could be implicated in the COVID-19 prognosis. Diagnosis of vitamin D deficiency could be a helpful adjunct in assessing patients' potential of developing severe COVID-19. Appropriate preventative and/or therapeutic intervention may improve COVID-19 outcomes.
Keywords: COVID-19; SARS-CoV-2; meta-analysis; outcome; vitamin D.
© 2020 Wiley Periodicals LLC.