The essential aids in the evaluation of suspected blunt urinary tract injury are urinalysis and IVP. In 78 consecutive children who had IVPs because of trauma from January 1982 to March 1986, the following were evaluated: (1) the yield of IVP; (2) the correlations between IVP and hematuria, mechanism of injury, and associated clinical findings; and (3) the effect of IVP on patient management. Of the 26 children (33%) with abnormal IVP findings, 13 had congenital urinary tract anomalies only and 13 had urinary tract injuries (eight renal contusions, four renal lacerations with extravasation, and one bladder rupture). The number of RBCs per high-power field correlated with IVP evidence of injury (P less than .05). If only those patients with greater than or equal to 20 RBCs per high-power field had received IVPs, 42% of IVPs would have been avoided and no injuries or surgically correctable anomalies would have been overlooked. Urinary tract injury occurred significantly more often in patients with extremity fractures (P less than .05) and pelvic fractures (P less than .05). Mechanism of injury, admission to the hospital, and flank tenderness or hematoma were not associated with IVP evidence of trauma (P greater than .05), however. In four patients with trauma, results of IVP led to lengthened hospitalization or further diagnostic studies but did not result in surgery. Two patients in whom ureteropelvic junction obstruction was discovered incidentally had delayed corrective surgery.(ABSTRACT TRUNCATED AT 250 WORDS)