ObjectiveTo determine if microalbumin excretion can predict the development of premature delivery.MethodsThe possibility of predicting, early in pregnancy, the development of a preterm delivery using urinary albumin was investigated in 1422 nulliparous women recruited prospectively. A first morning urine sample was collected at three occasions during pregnancy (8–14, 15–24, 25–24 weeks' gestation) for the determination of urinary albumin excretion.ResultsNo significant correlation was found between gestational age and urinary albumin (absolute concentration, albumin-creatinine ratio, or relative clearance of albumin) at either visit (rranging from −0.043 to 0.036;P> .1). The incidence of preterm birth was similar for the first and fourth quartiles of the urinary albumin-creatinine ratio for the second visit (5.8 and 5.7%, respectively).ConclusionUrinary albumin is not a useful marker of preterm birth in a low-risk general population.