purpose: to evaluate the differences between the maternal and perinatal outcomes of pregnancies complicated by preeclampsia, according to the classification as the severe/mild form, and the early/late onset form. methods: a retrospective study with 211 pregnancies complicated by preeclampsia, assessed at a university reference center from 2000 to 2010. the diagnosis and disease severity were based on the values of blood pressure, proteinuria, and clinical and laboratory findings. the pregnant's age, skin color, parity, blood pressure, urine protein semiquantitative values, presence of bilateral notch in the uterine artery dopplervelocimetry and birth conditions were compared between patients with mild and severe disease, as well as between those of early/late onset. the disease was considered to be of early onset when diagnosed at less than 34 weeks of gestational age. results: most patients had the severe form of preeclampsia (82.8%), and the onset of the condition was early in 50.7%. blood pressure values (133.6±14.8 versus 115.4 mmhg, p=0.0004 and 132.2±16.5 versus 125.7 mmhg, p=0.0004) and semiquantitative proteinuria (p=0.0003 and p=0.0005) were higher in the early and severe forms compared to mild and late forms. infant birth weight (1,435.4±521.6 versus 2,710±605.0 g, 1,923.7±807.9 versus 2,415.0±925.0 g, p<0.0001 for both) and apgar score (p=0.01 for both) were smaller for severe and early preeclampsia compared to mild and late preeclampsia. on the other hand, the presence of a bilateral notch in the uterine arteries was linked to the forms of early onset (69.2 versus 47.9%, p=0.02), whereas fetal growth restriction was more frequent in the severe forms of preeclampsia (30 versus 4.4%, p=0.008). conclusion: the preeclampsia classification based on maternal clinical parameters better reflected the conditions of fetal nutrition, while the early onset of the condition was associated with placental vasculopathy detected by dopplervelocimetry.