Screening for fetal urological abnormalities: how effective?

To analyse urological cases registered with the Northern Congenital Abnormality Survey (NorCAS) and thus assess antenatal diagnostic sensitivity, measure regional interhospital variation and determine whether antenatal ultrasonography has contributed to the prevention of urinary tract disease in childhood.

All notification documents forwarded to NorCAS from its inception in 1984 until 31 December 1996 were inspected and those relating to urological abnormalities were abstracted. Information about postnatal progress was often missing. Twenty-three hospitals delivering maternity services were identified and randomly allocated a code letter a-w. Data were analysed in two periods, 1984-90 and 1991-96. To quantify hospital performance an 'expected number' for each hospital was calculated from the overall incidence per 1000 births and the number of births in each hospital.

There was a clear improvement in antenatal diagnostic sensitivity for the major urological abnormalities in the second period, but there was considerable variation among hospitals in the number of cases notified; this applied particularly to antenatal renal dilatation. Two hospitals notified significantly more cases in this category than all the others. The difference in notification rates among hospitals was caused by diagnoses such as vesico-ureteric reflux (VUR), pelvi-ureteric junction obstruction and megaureter, which often produce antenatal renal dilatation. Hospitals which notified more of these cases performed more micturating cystograms and identified more children with VUR. Measurements of the fetal renal pelvis were regularly made by only seven hospitals; one of these had difficulties with postnatal follow-up. There was evidence that moderate degrees of antenatal renal dilatation (4-10 mm) were associated with VUR; 8.5% of the children in whom VUR was diagnosed had fetal renal dilatation of this order.

Antenatal renal dilatation and its significance are either not being recognized or not appreciated in some hospitals. The high incidence of VUR in cases with moderate antenatal renal dilatation indicates that children with this condition may be escaping diagnosis.