Is intraoperative surgeon's opinion an accurate tool to assess the outcome of endoscopic treatment for vesicoureteral reflux?

OBJECTIVE

Our experience in the endoscopic treatment of vesicoureteral reflux (VUR) has significantly increased during the last decade. To help develop diagnostic tests to check the success of this procedure, we evaluated the accuracy of surgeons' intraoperative observations as a predictor of treatment results.

METHOD

We performed a prospective study of patients with VUR who were endoscopically treated during 1 year (106 renal units). Patients' age and gender, laterality, material used, grade of reflux, presence of ureteral duplication or associated pathology, and morphology of ureteral orifice were recorded as predictive factors related to the success rate. Surgeon and assistant indicated at the end of the endoscopic procedure whether the VUR was cured or not for each renal unit. These estimations were compared with postoperative voiding cystourethrogram results.

RESULTS

Overall cure rate was 75.5%. Positive predictive value (PPV) for surgeon's opinion was 0.79 and negative predictive value (NPV) was 0.40. Statistical analysis demonstrated that the association between the surgeon's opinion and the cure rate was low with a Kappa value of 0.171 (p = 0.30). PPV of assistant's opinion was 0.80 and NPV was 0.40, with a Kappa value of 0.2 (p = 0.13). Concordance of surgeon and assistant's opinion resulted in PPV of 0.79 and NPV of 0.53 (Kappa = 0.261). Kappa value did not improve when surgeon's opinion was related to other factors such as the material employed, grade of reflux, presence of ureteral duplication or associated pathology and morphology of the ureteral orifice.

CONCLUSION

In our experience, surgeon's opinion is not an accurate tool to predict the outcome of endoscopic treatment of VUR.