Variability of diuresis renography interpretation due to method of post- diuretic renal pelvic clearance half-time determination

Purpose: We assessed variability in the interpretation of diuresis renography that may result from using different methods of clearance half- time determination. Materials and Methods: We reviewed 152 diuresis renography studies performed at diagnosis or during followup of 53 children enrolled in a prospective study assessing the natural history of unilateral neonatal hydronephrosis. Studies were classified as nonobstructive, indeterminate or obstructive using 4 methods of half-time determination. Intermethod correlation and agreement were evaluated. We compared the proportion of nonobstructive, indeterminate and obstructive classifications by each method, and the interpretation of individual studies based on each method. Results: Among methods we noted a high degree of correlation and fair to excellent agreement (Spearman ρ = 0.86 to 0.92 and κ = 0.57 to 0.86, respectively). However, in 27.8% of intermethod comparisons the proportion of studies classified as nonobstructive, indeterminate and obstructive differed significantly (p <0.05). The classification of pelvicaliceal drainage varied by method for all but the most severely dilated systems. In individual studies classification by 1 method was discordant with classification by another in 19% of comparisons. Of the discordant interpretations 97.7% involved nonobstructive versus indeterminate or indeterminate versus obstructive classifications. Conclusions: Variability in classifying drainage patterns based on half-time requires that practitioners be circumspect when applying this parameter for managing asymptomatic hydronephrosis. It also necessitates the description of quantitative methodology in published series of this important clinical problem.