Vesicoureteric reflux in infants with febrile urinary tract infection: Avoiding a cystourethrogram cannot be justified yet

Abstract

To the Editor: We read with interest the report by Ismaili et al. They concluded thatwhen renal ultrasound scanning examination (US) results were normal in young infants after a urinary tract infection (UTI), the risk ofmissing high-grade vesicoureteric reflux (VUR)was extremely low. This statement is based on the findings that the sensitivity rate of abnormal findingsof theurinary tract on US for detecting the presence of high-grade (grade IV-V) VUR was 100%, and the negative predictive value was 100%. All 3 infants with high-gradeVURdetectedwith a voiding cystourethrogram (VCUG) had abnormal US findings, and 29 infants with low-grade (grade IIII) VUR or normal urinary tract showed a no abnormal US results. Our retrospective analysis of equivalent subjects contradicts their results: 102 children with febrile UTI aged 0 to 3 months of age (male, 84; female, 18) received both US and VCUG. As shown in the Table, the sensitivity rate of abnormal US results for predicting the presence of highgrade VUR on VCUG was 64%, the specificity rate was 89%, the positive predictive value was 47%, and the negative predictive value was 94%. These results indicate that avoiding VCUG on the basis of the normal US findings would have resulted in missing 5 of 14 infants with high-grade VUR (35.7%). These variations in diagnostic accuracy may be explained not only by the sample sizes, but also methodologies. Results of both renal US and VCUG depend on diverse contributing factors, such as patient conditions (hydration, movement) or administered volume of contrast medium, although they did not describe them in detail. Until a study assessed with the standardized imaging techniques provides evidence that the risk of missing a significant renal abnormality is extremely low, avoiding VCUG in infants with febrile UTI cannot be justified.