Vascular endothelial growth factor levels in tuberculosis: A systematic review and meta-analysis

Background: Changes in endothelial function are implicated in the spread of tuberculosis (TB). Studies suggest a role for the vascular endothelial growth factor (VEGF) in TB-related endothelial function changes. However, the findings of studies investigating the VGEF profile in TB are not consistent, and no formal systematic review and meta-analysis exists summarizing these studies. Methods: We did a meta-analysis of studies assessing VEGF levels in patients with TB. A systematic search on June 25, 2021, was conducted for eligible studies that made VEGF measurements in an unstimulated sample, e.g., a blood fraction (plasma or serum), cerebrospinal fluid (CSF), pleural effusion (PE), or bronchoalveolar lavage fluid, and ascites or pericardial fluid for patients with TB and controls without TB. Also, studies that made simultaneous measurements of VEGF in blood and PE or CSF in the same patients with TB were included. Longitudinal studies that provided these data at baseline or compared pre-post anti-tuberculosis treatment (ATT) levels of VEGF were included. The primary outcome was the standardized mean difference (SMD) of VEGF levels between the comparison groups. Results: 52 studies were included in the meta-analysis. There were 1787 patients with TB and 3352 control subjects of eight categories: 107 patients with transudative pleural effusion, 228 patients with congestive heart failure (CHF)/chronic renal failure (CRF), 261 patients with empyema and parapneumonic effusion (PPE), 241 patients with cirrhosis, 694 healthy controls (with latent TB infection or uninfected individuals), 20 patients with inactive tuberculous meningitis (TBM), 123 patients with non-TBM, and 1678 patients with malignancy. The main findings are as follows: (1) serum levels of VEGF are higher in patients with active TB compared with healthy controls without other respiratory diseases, including those with latent TB infection or uninfected individuals; (2) both serum and pleural levels of VEGF are increased in patients with TPE compared with patients with transudative, CHF/CRF, or cirrhotic pleural effusion; (3) ascitic/pericardial fluid, serum, and pleural levels of VEGF are decreased in patients with TB compared with patients with malignancy; (4) pleural levels of VEGF are lower in patients with TPE compared with those with empyema and PPE, whereas serum levels of VEGF are not different between these patients; (5) both CSF and serum levels of VEGF are increased in patients with active TBM compared with controls, including patients with inactive TBM or non-TBM subjects; (6) post-ATT levels of VEGF are increased compared with pre-ATT levels of VEGF; and (7) the mean age and male percentage of the TB group explained large and total amount of heterogeneity for the meta-analysis of blood and pleural VEGF levels compared with healthy controls and patients with PPE, respectively, whereas these moderators did not show any significant interaction with the effect size for other analyses. Discussion: The important limitation of the study is that we could not address the high heterogeneity among studies. There might be unmeasured factors behind this heterogeneity that need to be explored in future research. Meta-analysis findings align with the hypothesis that TB may be associated with abnormal vascular function, and both local and systemic levels of VEGF can be used to trace this abnormality.