[Interobserver variation in region-of-interest drawing in computerized gammagraphy]

Nuclear medicine uses computerized scintillation cameras for scan processing. For the radio-renogram, the main objective is to observe the passage of radionuclides through the kidneys during a given time. The new software automatically provides data of the dynamic studies, but there is one step (the drawing of regions of interest, ROI) handled by an operator. With the ROI drawings the computer integrates their radioactivity and displays it as time/activity curves, and calculates the Tmax (time to achieve maximal activity) and the T1/2 (time to eliminate half of the Tmax) of each kidney.

To evaluate inter-observer variability in drawing the ROI on renal scans.

Four observers with at least seven years of experience in the procedure drew independently the ROI of 38 renograms of 20 patients (two transplants) to obtain the Tmax and T1/2. The interobserver CV was calculated for the Tmax and the T1/2 of the 38 scans.

Globally the interobserver variability was larger for T1/2 than for Tmax. There were four scans with small differences in Tmax and/or T1/2 but which lead to inter-observer discrepancies in the classification (normal/abnormal). The partition of the scans in three groups (1 = Tmax and T1/2 normal; 2 = only Tmax normal; 3 = both abnormal) showed significant intergroup differences in the interobserver variability (Kruskal-Wallis p = 0.032) which were caused by the larger variability in group 2 (6 of 11 scans with CV > 4%) than in the other groups (none with CV > 4%).

1. The interobserver discrepancies in classification were observed only in cases with parameters slightly abnormal (8-9 min in Tmax, 15-20 min in T1/2). 2. We have no explanation for the larger interobserver variability of the Tmax in group 2 study of intraobserver variability in these same four observers may help in gaining insight to some of the observations in this study.