Objective To determine the diagnostic value and necessity of excretory phase of CT urogrphy (CTU). Methods One hundred and one cases with 197 sides of cohort CTU studies were enrolled from PACS system retrospectively, and 44 cases had the records of radiation dosage. Three different seniority senior radiologists reviewed the images and diagnosed 2 times blindly with the clinical information. At the first time, plane scan, contrast enhanced renal cortical phase and renal parenchymal phase images (Diag. A) were reviewed. At the second time, with 3 kinds of images mentioned above plus excretive phase images (Diag. B) were reviewed, with interval of 3 months between the two reviews. With Diag. B as reference standard, accordance rate of diagnoses of Diag. A was calculated. The difference of the 3 rediologists's accordance rates were analyzed with Chi-square test, and the lesions with accordance rate ≤80% were recorded. The cases of 197 urinary tracts were divided into 4 groups: lesions of renal pelvic and ureter, lesions of bladder, lesions of kidney and extra-urinary tract, without lesion. The filling of contrast media in the urinary tract was recorded with 5-scale scoring system. The differences of the urinary tract filling among the 4 groups were analyzed with Kruskal-Wallis test. The effective radiation dosages of CTU with and without excretory phase scan were recorded and calculated, and the difference was analyzed with t test. Result The diagnostic accordance rate of CTU with and without excretory phase of the 3 radiologists was 95.4%(188/197), 93.9%(185/197), 92.4%(182/197)respectively(χ2=1.60, P>0.05). Lesions with diagnostic rate≤80%were parapelvic cyst and caliceal diverticulum. The median score of urinary tracts filling in the lesions of renal pelvic and ureter, lesions of bladder, lesions of kidney and extra-urinary tract, without lesion group were 2, 3, 3, 4, and the differences were significant (HC=31.7, P<0.05). The effective radiation dosages of CTU with and without excretory phase scan were (32.36±11.04)mSv and (41.47±14.27) mSv respectively (t=3.35, P<0.01). Conclusions No significant effect of excretive phase imaging is found on the diagnosis of most urinary system diseases. Poor contrast filling and high rate of failure of filling of the urinary tract with intra luminal pathologies, and increase of the radiation dosage make the excretory phase scan in CTU less useful and it should be used carefully and selectively.