Facebook Twitter YouTube

Utility of Initial Renal Ultrasound In Finding Vesicoureteral Reflux In Children Who Present with Febrile Urinary Tract Infection: Single Institutional Experience

Saturday, October 20, 2012: 11:39 AM
Grand Ballroom A/B (Hilton Riverside)
Eric Z. Massanyi, MD1, Janae Preece, MD2, Susan M. Lin3, Angela Gupta, MD1 and Ming-Hsien Wang, MD1, (1)Pediatric Urology, Johns Hopkins University School of Medicine, Baltimore, MD, (2)Division of Urology, University of Maryland Medical Center, Baltimore, MD, (3)Johns Hopkins School of Medicine, Baltimore, MD

Purpose: Classically, children with febrile urinary tract infection (UTI) were screened for vesicoureteral reflux (VUR) by voiding cystourethrogram (VCUG).  Changes to the AAP's guideline regarding the management of children less than 2 years of age who present with initial febrile UTI suggest that a normal screening renal ultrasound (RUS) precludes the need for VCUG.  While this practice reduces the need for invasive, radiation-based VCUG, it remains unclear how often patients with clinically significant VUR may be initially overlooked due to normal RUS.  The authors conducted a single institutional study of young children presenting for evaluation of febrile UTI and the utility of initial US for the diagnosis of VUR.   

Methods:   An IRB-approved retrospective review was performed of all children ages zero through two years who were evaluated at our clinic for febrile UTI between 2004 and 2011.  Initial RUS results were examined among those patients who were discovered to have renal scarring or required anti-reflux surgery.   Additionally, all children who underwent both RUS and VCUG were identified to evaluate the negative predictive value (NPV) and sensitivity of RUS for VUR.   

Results: Between 2004 and 2011, 144 patients with no other known urologic disease were evaluated at our institution for febrile UTI that occurred less than 2 years of age.  Initial RUS was normal in 12/19 (63%) patients who ultimately underwent anti-reflux procedures.  Eighteen patients were found to have renal scarring on DMSA scan, and initial RUS was normal in 11 (61%).  Among patients in whom both initial RUS and VCUG data were available, 217 kidneys in 110 patients were evaluated.  Of the 181 kidneys found to be normal on initial RUS, 136 had evidence of VUR, 31 of which were grades IV - V, yielding negative predictive values (NPV) of 25% (all grades) and 83% (grades IV - V).  Of the 36 kidneys with hydronephrosis, 22 were found to have VUR, and 14 were grades IV - V.  Among the 6 kidneys with grade 3 or 4 hydronephrosis, 3 (50%) were found to have grade V VUR.  Overall sensitivity of RUS for VUR was 14% (all grades) and 31% (grades IV - V).    

Conclusion: For children between 0 and 2 years of age who presented with febrile UTI at this institution, RUS had poor sensitivity and NPV for VUR.  Furthermore, in a subset of patients who eventually required anti-reflux procedures or developed renal scarring, the majority had a normal initial RUS.  These findings suggest that RUS alone may leave many children with potentially clinically significant VUR undiagnosed after initial evaluation.  Further studies will be needed to understand long-term clinical outcomes in light of the new AAP guideline.