Assessment of deconvolution technique for diagnosis of hydronephrosis in infants (under 6 months) [persian]

Abstract

Introduction: Hydronephrosis is one of the major causes of referring infants for nuclear medicine investigation. Renal dynamic study is the most reliable investigation for such children. However direct evaluation of renogram leads to generation of parameters that have no physiological interpretation. The configuration of renogram is very much dependent upon the rate at which radiopharmaceutical enters the kidney. In order to decrease such effect, the radiopharmaceutical has to be injected in a small volume and very fast into the vein, which is not always possible, especially in the infants. Deconvolution is a method in which the injection quality does not have any effect upon renogram decreasing the diagnostic errors. The renal retention function (RRF) is independent of input function and derived parameters have physiological significance and clinical values. Materials & methods: In this study 54 patients data were evaluated. All the patients were under 6 month of age, suspicious for hydronephrosis referred to our imaging center for 99mTc-DTPA renal dynamic studies. All data were transformed into interfile format. Full renal package software was developed in visual basic 6 to read and process the data. Deconvolution was performed using matrix algorithm. Background subtraction was performed using conventional and Ruthland-Patlack method. All transit time parameters were calculated in 6 different filtering conditions. Results & discussion: Our results show that deconvolution is useful for evaluation of infant hydronephrosis. Mean transit time is the most useful parameter for estimation of renal function in infants.