Objective The aims of this study were to establish prospectively the prevalence of objective bladder dysfunction before and after delivery by means of urodynamic investigations and to assess the effect of obstetric variables on bladder function.

Design Prospective longitudinal study. Twin channel subtracted cystometry was performed in the standing and sitting position, with a cough stress test at the end of filling. The investigations were repeated three months postpartum.

Participants Two hundred and eighty-six nulliparae with singleton pregnancies who were delivered between April 1996 and November 1997 attended for antenatal assessment after 34 weeks of gestation and 161 who returned postpartum.

Setting Department of Obstetrics and Gynaecology in a London teaching hospital.

Results The mean urodynamic values both in pregnancy and postpartum lower than values defined in a non-pregnant population. The prevalence of genuine stress incontinence and detrusor instability were antenatally 9% and 8%, respectively, and postpartum 5% and 7%, respectively. Obstetric and neonatal factors were not related to urodynamic variables.

Conclusions Despite the reported high prevalence of urinary incontinence related to pregnancy and childbirth, neither pregnancy nor delivery resulted in any consistent effects on objective bladder function. Postpartum urodynamic measurements were not related to either obstetric or neonatal variables, but were dependent on antenatal values.