Purpose: There is no universally accepted definition of obstruction in the use of dynamic renography for the evaluation of high-grade hydronephrosis. Prolonged, but insignificant pelvicalyceal tracer retention is often difficult to differentiate from significant obstruction in cases with neither clear obstruction nor unimpaired outflow. We hypothesized that an additional measurement of the residual activity 120 mins after tracer application might reduce the probability of equivocal findings.
Methods: In 37 consecutive MAG3 scans performed for the evaluation of isolated unilateral highgrade hydronephroses ≥SFU III in 28 patients a late static image after 120 mins was additionally acquired in the case of a high residual activity of >30% after 45 mins. Nine board-certified specialists (6 in nuclear medicine, 3 in pediatric urology) independently reviewed all examinations with and without the information on the additional late imaging 120 mins post injection (p.i.) using a predefined ordinal scale from 1 (=normal) to 5 (= obstructive). We measured the extent of change in reporting associated with the addition of late imaging and controlled for the level of interobserver agreement. After follow-up the impact on surgical indications was reviewed. Ranks were compared using Wilcoxon-Mann tests, interobserver agreement was calculated via evaluation of individual findings with ANOVA. P-values <0.05 were considered significant.
Results: The addition of a late image resulted in a "less obstructive" interpretation (decreased rank, -0.31, p = 0.000048) compared to standard protocol and led to a reduction of equivocal findings (39 % vs. 28%, p=0.003). There was a low variation in the assessments with no significant differences in rank distribution (p = 0.137) and no difference between nuclear medicine specialists and pediatric urologists. After a median 66 months of follow-up, 25/28 (89 %) patients were eligible for reevaluation, of whom 18/25 (72 %) proceeded to pyeloplasty. Of those with a significantly less obstructive assessment due to late imaging, 6/9 underwent pyeloplasty.
Conclusion: In preselected patients with high residual activity after 45 mins, measurement of the residual activity 120 min after tracer application influenced interpretation of the diuretic renography towards a less frequent diagnosis of obstruction and reduced the number of equivocal assessments. In this study, the clinical impact of the additional late imaging proved small and relevant only for few patients. Until further studies have found different results, the acquisition of additional late imaging should remain a very individual decision.

