To assess the association between exposure to prenatal outdoor air pollution and preterm birth to provide evidence in setting up programs for prevention on premature birth.

Case-control studies regarding association between pregnant exposure to outdoor air pollution and preterm birth were collected. Data were analyzed with Stata 12.0.

Ten articles including 48 556 cases and 548 495 controls were qualified for inclusion. Results from Meta-analyses showed pooled ORs for exposure to NO₂, PM₁₀, CO, PM_2.5, and NO during the entire pregnancy were 0.960 (95%CI: 0.935-0.985), 1.068 (95%CI: 1.035-1.103), 1.122 (95%CI: 1.078-1.168), 1.110 (95% CI: 1.043-1.181) and 0.994 (95% CI: 0.973-1.016). Association between air pollution and preterm birth varied with the periods of exposure. Pooled ORs for NO₂, PM₁₀ and SO₂ exposure during the first trimester were 1.117 (95%CI: 1.052-1.186), 0.968 (95%CI: 0.812-1.153) and 1.258 (95%CI: 0.758-2.089). Pooled ORs for NO₂, PM₁₀, SO₂ exposure during the second trimester were 1.000 (95% CI: 0.982-1.019), 1.127 (95% CI: 0.896-1.416) and 0.977 (95% CI: 0.711-1.342). Pooled ORs for NO₂, PM₁₀, SO₂ exposure during the third trimester were 1.006 (95%CI: 1.002-1.010), 1.053 (95% CI: 0.973-1.139) and 1.003 (95% CI: 1.000-1.006).

Exposures to PM₁₀, CO and PM_2.5 during the entire pregnancy, to NO₂ during the first trimester, or to NO₂ and SO₂ during the third trimester were associated with preterm births.

Air pollution; Preterm birth; Periods of exposure; Meta-analysis