Effects of ambient fine particulate matter (PM2.5) chemical constituents on preterm birth (PTB) are inconclusive. We performed a sibling-matched cohort study to explore the association between PM2.5 constituents and the risk of PTB. Changes in PM2.5 and the constituents (ΔPM2.5, ΔOC, ΔBC, ΔSO42−, ΔNH4+, ΔNO3−) were calculated as the value at a follow-up pregnancy minus exposures in prior pregnancy. A fixed-effect regression model was used to determine the association of the within-group variation in PM2.5 total mass and constituents with PTB risk. Gestational age differences within matched siblings were also treated as outcomes. A total of 62,334 sibling pairs were analyzed. We observed that each 10 μg/m3 increase in ΔPM2.5 in first, second, and third trimester were not statistically significant associated with the risk of PTB, but shortened gestational age was observed, with -0.009 week (95% CI: -0.013, -0.005), -0.005 (95% CI: -0.008, -0.002), -0.012 (95%CI: -0.015, -0.008) and -0.019 week (95% CI: -0.026, -0.012) decrease in first, second, third and whole pregnancy. Our findings indicate that increase in ambient PM2.5 constituents was associated with a shorten gestational age. This study added a novel cohort evidence for the association of changes of PM2.5 components with gestational age.