Objective To study the diagnostic value of magnetic resonance urography (MRU) in the urinary tract obstruction(UTO) in children.Methods 65 children with UTO underwent intravenous pyelography(IVP),B-ultrasound(BUS) and MR urography(MRU) ,MRU using hearvily T_2-weighted fat suppression turbo echo sequence.MRU features were compared with that of operation.The accurate rate in diagnosis of UTO with MRU was analysed comparatively with that of IVP and BUS.Results The lesions diagnosed with MRU included stricture of ureteropelvic junction in 36 children,and duplication of renal pelvis in 3 children (diagnostic error in 1),primary megaloureter in 9,narrowing of cystic inlet of ureter in 4,cysts of cystic inlet of ureter in 3,solitary kidney in 2,nephrolith in 5,nervous bladder in 2 and congenital renal dysplasia in 1.The accuracy was 98.5% in defining the UTO with MRU(64/65),which was superior to that with IVP and BUS(P＜0.05).Conclusion MRU is of great value in the diagnosis of upper urinary tract obstruction in children.

Objective:To explore the relationship of risk factors and clinical features to assessments of children with cerebral palsy (CP ) using a magnetic resonance imaging classification system (MRICS).Methods:Medical records of CP patients under 18 years old were reviewed retrospectively. Data including high-risk factors, cranial MRI results and clinical characteristics were collected. The cranial MRI results were classified according to the MRICS.Results:Of 1357 patients studied, 1112 (82%) had received cranial MRI scans. Among them, 962 (86.5%) showed MRI-identified brain abnormalities, 489 in the periventricular white matter. Subjects with different weeks of gestation, birth weights, delivery times, neonatal hypoxic-ischemic encephalopathy, and neonatal cerebral hemorrhage had significantly different MRI classifications according to the system. Premature birth, low birth weight and multiple births correlated with the incidence of white matter brain injury. Only 4 of the subjects with neonatal cerebral hemorrhage were classified as having normal brain structures using the MRICS. However, gender, birth method, and pathological jaundice had no significant relationship with MRICS ratings. Significant differences in MRICS classifications were observed between patients with different CP subtypes, gross motor function scores, as well as with or without epilepsy, speech or language impairment. But degrees of mental retardation were not significantly related with MRICS classifications.Conclusion:MRICS classifications relate closely with risk factors and the clinical characteristics of CP patients. The system can play an important role in finding pathogenesis and predicting clinical outcomes. It is worthy of applying and promoting in the clinic.