To report experience and technical refinements with the Koyanagi urethroplasty for proximal hypospadias. A retrospective study was conducted between 2004 and 2017. Medical records of patients treated by the Koyanagi technique were investigated. The penile ventral skin was closed either with a Byars flap or using an "Ombredanne's chasuble" (OC). The cohort was divided chronologically into two groups of the same number of patients (early -E- and late experience group-LEG-). Demographics, anatomical findings and surgical outcomes were compared between groups using univariate analysis. A logistic regression was performed to assess factors associated with the occurrence of a postoperative urethrocutaneous fistula or urethroplasty dehiscence (UD). The Koyanagi urethroplasty was performed in 67 patients, 33 in the EEG and 34 in the LEG. The overall surgical complication rate was 64.2% (n=43), including 42 fistula or UD, higher in the EEG (81.8%) than in the LEG (44.1%, P<0.01). However, in the LEG, patients underwent the surgery older and the use of OC more frequent. After multivariate analyses, factors associated with a postoperative fistula or UD were the year of surgery (OR=0.71 [0.53-0.96] P=0.02), the age at surgery (OR=1.11 [1.01-1.22], P=0.03); contrary to the skin coverage method or the stenting duration (P>0.05). An urethral stenosis occurred in 1 patient (1.5%). At last follow-up, 64.2% of patients required a further procedure and 80.6% of urethral meatus were glandular. In this study the complication rate, particularly the urethrocutaneous fistula, remained high. Urethral stenosis, were rare but the follow-up was too short to clearly identified them. Considering as a 2 stage procedure koyanagi uretroplasty allows to obtain finally good results. III.