Figure 1. Flowchart of the included studies for meta-analysis.

A total of 663 studies were identified, and 252 studies were excluded because of duplication. A total of 357 studies were excluded after reading the abstract because of not related. A total of 54 possible studies were full-text reviewed and 33 studies were excluded because of animal or cell studies (n=18), case report, meta-analysis and review (n=7), insufficient data (n=6). Finally, 21 studies were included for quantitative analysis.