Presentation Authors: JONATHAN MENDOZA*, ROGERSON TIANGCO, MICHAEL CHUA, JUN DY, ODINA GOMEZ, QUEZON CITY, Philippines
Introduction: Vesicoureteral Reflux (VUR) is a common condition affecting 34.8 to 37.4% of all pediatric populations worldwide. Prompt diagnosis and adequate management is required to avoid serious long-term sequelae like renal failure. We aim to assess the diagnostic accuracy and safety of contrast enhanced urosonography using second generation contrast with harmonic imaging (CEVUS-HI) based on standard fluoroscopic voiding cystourethrogram (VCUG) in detecting VUR among children.
Methods: Literature search was performed on several medical databases on March 8, 2018. Comparative studies were independently screened and assessed by three reviewers. Study quality was appraised using QUADAS-2 tool. Diagnostic accuracy of sensitivity (Sn), Specificity (Sp), positive likelihood ratio (pLR), negative likelihood ratio (nLR) and diagnostic odds ratio (dOR) were determined with VCUG as the reference standard. Adverse effects were summarized. The inter study variability, and heterogeneity and its possible source were determined and investigated via bivariate model metaregression and configuration of summarizing receiver operator characteristics (SROC). Assessment of the likelihood of publication bias was also done.
Results: A total of 12 studies with low to moderate risk of bias, including 1,917 ureteral units from 953 patients were assessed. The studies reported no associated serious adverse events. The pooled diagnostic accuracy parameters of CEVUS-HI are considered excellent in detecting VUR among children with Sn of 90.43 (95%CI 90.36 to 90.50), Sp of 92.82 (95%CI 92.76 to 92.87), pLR of 12.59 (95%CI 12.49 to 12.68), nLR of 0.103 (95%CI 0.102 to 0.104) and extrapolated pooled dOR of 122.12 (95%CI 120.75 to 123.49). Heterogeneity with interstudy variability was noted (p < 0.0001, I-squared >70%). Using QUADAS-2 domains and threshold effect as variables, no significant statistical difference was observed using univariate metaregression. A significant plot asymmetry suggesting likelihood of the presence of publication bias (Fixed effect summary estimate: 3.6946; t = -3.1275, df=10, p = 0.0107)was observed.
Conclusions: Based on VCUG as reference standard, the pooled diagnostic accuracy parameters have illustrated that CEVUS-HI is highly accurate in detecting VUR amongst children. Its excellent safety profile and absence of radiation exposure allows for its use as an alternative for screening and follow-up of VUR especially in highly specialized centers.