Selecting appropriate management for the fetus with bilateral congenital hydronephrosis depends on our ability to accurately assess the severity of existing renal damage and to predict the potential for recovery of renal and pulmonary function if the obstruction is relieved. We reviewed our experience with 20 fetuses with congenital bilateral hydronephrosis to determine the prognostic value of various criteria used to assess functional potential, including temporary catheter exteriorization to measure fetal urine output and composition. Based on autopsy, biopsy, or clinical outcome, ten fetuses were classified retrospectively as "poor function," and ten fetuses as "good function." The good function group could be distinguished from the poor function group by the following criteria: Amniotic fluid (AF) status at presentation (P less than .001), ultrasound appearance of the fetal kidneys (P less than .05), fetal urine sodium and chloride concentration and osmolarity (P less than .001), and hourly urine output (P less than .02), but not by fetal urine iothalamate excretion or potassium and creatinine concentrations (P greater than .05). Based on these results, we have identified prognostic criteria that accurately identify the fetus with "good function" from the fetus with "poor function." We also reviewed the clinical management of our last 12 unreported cases. Ten fetuses had undergone diagnostic catheter placement and in utero renal function testing. This led to placement of a therapeutic indwelling catheter-shunt in seven fetuses (three required multiple shunts) and a suprapubic vesicostomy in another. Catheter related complications, including three cases of chorioamnionitis, emphasize the need for better methods of in utero decompression in selected cases. Our ability to select appropriate management has improved markedly.