Background: Evidence suggests controversial results based on the antibacterial and anti-inflammatory effects of azithromycin (AZI) in the treatment of childhood asthma. This study was to further evaluate the efficacy and safety of AZI in childhood asthma. Methods: We searched PubMed, Embase (via Ovid), Cochrane Library, China National Knowledge Infrastructure, Chinese Scientific Journals database, WANFANG, and Chinese Biomedical Literature database from inception to November 11, 2020. Randomized controlled trials (RCTs) of AZI versus placebo or one positive control drug, AZI plus anti-asthma drugs (AADs) versus the same AADs, and AZI plus AADs versus placebo or one positive control drug plus the same AADs were included. Primary outcomes were number of exacerbations (NoE); score of clinical tools to assess asthma control after treatment; number of days to relieve symptoms with β2 agonist (DBA); post-treatment lung function indicators, including FEV1% of predicted value (pFEV1%), FVC% of predicted value (pFVC%), FEV1/FVC% of predicted value (pFEV1/FVC%), and PEF% of predicted value (pPEF%). Secondary outcomes were post-treatment fractional exhaled nitric oxide (FENO); post-treatment eosinophil counts in sputum (sEOS) or blood (bEOS); author self-reported outcomes related to asthma (AROs); and adverse events (AEs). Results: 61 eligible reports from 59 studies were finally included. AZI plus AADs shows no statistically significant difference in NoE (RR = 0.49; 95% CI, 0.07 – 3.26; P = 0.05) and sEOS (MD = -1.13%; 95% CI, -3.54% – 1.29%; P = 0.36) compared to AADs alone. The post-treatment C-ACT score was improved after AZI plus salmeterol and fluticasone (SF) treatment compared to SF alone (MD = 2.97; 95% CI, 2.39 – 3.54; P < 0.001). Results from three studies which could not be meta-analyzed showed that AZI may reduce DBA compared to placebo. AZI combined with AADs could improve post-treatment pFEV1% (AZI + glucocorticoid (GC) vs GC: MD = 6.92%; 95% CI, 1.47% – 12.37%; P = 0.01. AZI + leukotriene receptor antagonist (LTRA) vs LTRA: MD = 24.88%; 95% CI, 21.47% – 28.29%; P < 0.001. AZI + GC + BA vs GC + BA: MD = 12.40%; 95% CI, 9.72% – 15.08%; P < 0.001), pFEV1/FVC% (AZI + GC vs GC: MD = 10.24%; 95% CI, 6.44% – 14.03%; P < 0.001. AZI + GC + BA vs GC + BA: MD = 9.05%; 95% CI, 5.66% – 12.44%; P < 0.001. AZI + BA vs LTRA + BA: MD = 14.48%; 95% CI, 11.84% – 17.12%; P < 0.001), and pPEF% (MD = 7.00%, 95% CI, 2.53% – 11.47%; P = 0.002), but not improve pFVC% (MD = -10.37; 95% CI, -20.86% – 0.12%; P = 0.05), compared to AADs alone. Post-treatment bEOS was significantly higher in the AZI group than in the traditional Chinese medicine compound granules group (MD = 0.07×109/L; 95% CI, 0.05×109 – 0.09×109; P < 0.001). No statistically significant difference in bEOS after treatment with AZI plus montelukast (MON) and loratadine (LOR) compared to MON and LOR (MD = 0.03×109/L; 95% CI, -0.06×109 – 0.12×109; P = 0.50). Meanwhile, AZI combined with AADs did not increase AEs (RR = 0.76; 95% CI, 0.51 – 1.13; P = 0.17). Conclusions: AZI was beneficial in improving some clinical symptoms and lung functions in childhood asthma. AZI did not increase AEs when combined with AADs.