Methods: We retrospectively evaluated children diagnosed with pyelonephritis between 2-24 months using the Synthetic Derivative resource. The Synthetic Derivative includes de-identified clinical data extracted from our medical center electronic medical record. Patients were selected by ICD-9 codes and confirmed by chart review. Data collected included: age at 1st UTI, sex, race, need for hospitalization, use of intravenous antibiotics, history of abnormal prenatal ultrasound, renal ultrasound results, VCUG results, presence of recurrent UTI, and incidence of urologic surgical intervention. Statistical analyses were performed with univariate logistic regression and chi square test.
Results: A total of 174 patients were included. Most patients were Caucasian (56%) and female (77%). The mean age at diagnosis was 5.6 months. 83 (48%) required hospitalization and 105 (60%) received intravenous antibiotics for at least part of their treatment. 154 patients had renal ultrasound after pyelonephritis; 59 (38%) had abnormal findings. There were no identifiable factors associated with an abnormal ultrasound after pyelonephritis except for abnormal prenatal ultrasound (p=0.0001). However, many patients with reported normal prenatal ultrasounds went on to have abnormal postnatal ultrasounds.
We evaluated the subset of 95 patients with normal renal ultrasounds. Of these 95 patients, 84 had a VCUG. If a patient had bilateral vesicoureteral reflux (VUR), we grouped this patient by the higher severity. Of the 84, there were no abnormalities in 46, grade 1 in 1, grade 2 in 14, grade 3 in 15, grade 4 in 4, unspecified grade in 2, and other abnormality in 2. VUR was more likely in Caucasians (p=0.002) and females (0.04). Surprisingly, VUR was more likely in older patients (p=0.03). Of the 95 patients with normal renal ultrasound, 14 had a 2nd episode of pyelonephritis and 7 went on to have surgical intervention. Only abnormal VCUG was predictive of recurrent pyelonephritis (p=0.03) and need for surgical intervention (p=0.002).
Conclusion: We agree that renal ultrasound should be performed after 1st UTI in this age range as over one-third had abnormal ultrasound findings. More importantly, despite a normal ultrasound, a child may still have dilating reflux and recurrent pyelonephritis which requires surgical intervention. At this point we do not have a way to predict these patients other than abnormal VCUG. If VCUG is deferred, parents should be counseled regarding the risk of recurrent pyelonephritis and potential need for urologic intervention.