Correction represents a potential source of error in estimating split renal function and camera-based clearances. The purpose of this study was to determine which of five background options and four time intervals was associated with the least error for 99mTc-mercaptoacetyltriglycine (MAG3).
Methods: Fifteen single-kidney patients were imaged supine after 111-370 MBq (3-10 mCi) 99mTc-MAG3 injection. A phantom kidney was drawn on the 2-3-min images, approximately equal in size to the solitary kidney and used for all time intervals. Counts in the phantom and native kidneys were calculated using manual inferior and lateral regions of interest (ROIs), automated elliptical and perirenal background ROIs and no background correction at various time intervals (1-2, 1-2.5, 1.5-2.5 and 2-3 min) postinjection. With optimal background correction, counts and the relative function in the phantom kidney should be zero. The error was measuring by estimating both the relative function and absolute function expressed as the percent injected dose in the phantom kidney.
Results: The percent injected dose in the phantom kidney as well as the error in measuring relative function were significantly greater than zero for the inferior background correction and the no background correction options at all time intervals, p < 0.05. The percent dose in the kidney and the error associated with the lateral, elliptical and perirenal ROIs were not significantly different from zero.
Conclusion: Regardless of time interval, the greatest error was associated with no background correction. The inferior ROI consistently underestimated the background correction and probably should not be used for 99mTc-MAG3. There was no significant difference between errors generated using the lateral and automated ROIs, although automated ROIs are probably more reproducible for sequential studies.