Objective To compare the short and mid‐term outcomes of preterm twins by chorionicity of pregnancy. Design Prospective nation‐wide population‐based EPIPAGE‐2 cohort study. Setting 546 maternity units in France, between March and December 2011. Population A total of 1700 twin neonates born between 24‐34 weeks of gestation. Methods The association of chorionicity with outcomes was analyzed using multivariate regression models. Main Outcome Measures First, survival at 2‐years corrected age with or without neurosensory impairment and secondarily perinatal, short‐ and mid‐term outcomes (survival at discharge, survival at discharge without severe morbidity) were described and compared by chorionicity. Results In the EPIPAGE 2 cohort, 1700 preterm births were included (850 twin pregnancies). 1220 (71.8%) were from Dichorionic (DC) pregnancies and 480 from monochorionic (MC) pregnancies. MC pregnancies had three times more medical terminations than DC pregnancies (1.67% vs 0.51%, p <0.001) while there were three times more stillbirths in MC than in DC pregnancies (10.09% vs 3.78%, p <0.001). Both twins were alive at birth in 86.6% of DC pregnancies compared to 80.0% among MC pregnancies (p = 0.008). No significant difference according to chorionicity was found regarding neonatal deaths and morbidities. Likewise, for children born earlier than 32 weeks, the two‐year follow‐up neurodevelopmental results were not significantly different between DC and MC twins. Conclusions This study confirms that MC pregnancies have a higher risk of adverse outcomes. However, the outcomes among preterm twins admitted to neonatal intensive care units are similar irrespective of chorionicity.