Objective To determine whether intrapartum mean platelet volume (MPV) can predict new-onset delayed postpartum pre-eclampsia. Methods In a retrospective study, data were obtained for women with delayed postpartum pre-eclampsia (≤ 48 hours after delivery) who attended the Detroit Medical Center, Detroit, MI, USA, between January 2002 and July 2010. Patients were divided into two subgroups: new-onset delayed postpartum pre-eclampsia (no history of hypertensive disorder before 48 hours after delivery) and other late-onset/delayed postpartum pre-eclampsia (history of hypertensive disorders). Cases with no hypertensive disorders paired by delivery date were selected as controls. Receiver operating characteristic (ROC) curves were used to determine whether MPV could distinguish between cases and controls. Results Data for 130 cases of new-onset delayed postpartum pre-eclampsia, 71 cases of other late-onset/delayed postpartum pre-eclampsia, and 405 controls were collected. MPV was significantly different between the pre-eclampsia subgroups (P = 0.007). ROC curves showed that MPV significantly distinguished new-onset delayed postpartum pre-eclampsia cases (area under the curve 0.6, 95% confidence interval 0.5-0.6; P = 0.02), with a sensitivity and specificity of 58.1% and 55.4%, respectively. Conclusion Although MPV was significantly lower in women who went on to develop new-onset delayed postpartum pre-eclampsia than among controls, it is not a useful predictor. The low MPV suggests a different mechanism for the development of new-onset delayed postpartum pre-eclampsia.