Background: The accuracy of dynamic renal scintigraphy in evaluating the function of hydronephrotic kidneys is controversial. This study sought to investigate the effects of ureteral obstruction on renal function and the role of the protein load test (PLT) in evaluating hydronephrotic kidneys' function. Methods: A ureteral obstruction model was generated with New Zealand white rabbits. The baseline glomerular filtration rate (GFR) and the GFR during the PLT were measured by dynamic renal scintigraphy at weeks 6, 9, and 12 after obstruction and the renal reserve GFRs were calculated. The renal GFR differences between different time points and between the PLT and baseline status were compared. Results: (I) Both the baseline GFRs (t=11.287, P=0.000) and the GFRs during the PLTs (t=16.235, P=0.000) of the hydronephrotic kidneys were significantly lower than those of the contralateral kidneys. (II) The baseline GFRs of the obstructed kidneys were not significantly different 6, 9, and 12 weeks after obstruction (F=0.122, P=0.887), but their net reserve GFRs (F=8.419, P=0.004) and relative reserve GFRs (F=3.770, P=0.049) gradually decreased over time. (III) There was no significant correlation between the net reserve GFRs (r=-0.006, P=0.977) or the relative reserve functions (r=-0.022, P=0.920) of the obstructed kidneys and their baseline GFRs. However, there was a significant negative correlation between the net reserve GFRs (r=-0.590, P=0.002) or the relative reserve functions (r=-0.546, P=0.006) of the obstructed kidneys and the duration of obstruction. (IV) The reserve GFRs of the obstructed kidneys were not proportional to their baseline GFRs. Conclusions: The functional changes in hydronephrotic kidneys are not sensitively reflected by baseline GFRs. GFRs during the PLTs and the reserve GFRs play important roles in the early and accurate evaluation of the function of obstructed kidneys.