Do we need post-pregnancy screening with human chorionic gonadotrophin after previous hydatidiform mole to identify patients with recurrent gestational trophoblastic disease?