Clinical differentiation of acute pyelonephritis from lower urinary tract infection in children.

BACKGROUND AND PURPOSE To evaluate clinical variables for diagnosing childhood acute pyelonephritis (APN) when technetium-99m dimercaptosuccinic acid (DMSA) scintigraphy is not available. METHODS We retrospectively reviewed the records of 590 children with febrile UTI seen from January 1999 to February 2004. On the basis of DMSA scintigraphy performed within 7 days after admission, they were divided into APN (n = 237) or non-APN (n = 353) groups. Gender, age, clinical presentation, absolute neutrophil count, C-reactive protein (CRP), urinalysis, culture, and sonographic findings were recorded from charts. RESULTS A CRP level of > or =66.4 mg/L, in patients with >2 days prior to admission had a sensitivity of 71.6% and a specificity of 72.5% for APN. Similarly, a CRP of >27.3 mg/L in patients with < or =2 days prior to admission and a white cell count of >14,990/mm3 had sensitivities of 68.6% and 62.0% and specificities of 66.1% and 63.0%, respectively. Combining two or more variables did not result in better discrimination. CONCLUSIONS If a DMSA scan is not available, it is reasonable to treat a febrile UTI as APN if the CRP is >66.4 mg/L in a patient with >2 days of fever or if the CRP is >27.3 mg/L in a patient febrile for < or =2 days.