Predicting operation time and creating a difficulty scoring system in donor nephrectomy.

Background To determine predictive formulas for operation time and surgical difficulty in laparoscopic living-donor kidney transplantation. Methods We retrospectively analyzed data for 222 living donors aged > 20 years and recorded factors affecting operation time from patients' computed tomography images and medical records. We used the factors significantly affecting operation time to create a formula to predict operation time and designed a model to predict surgical difficulty based on the standardized partial regression coefficient, β. We also analyzed the relationship between surgical difficulty (high vs low) and operation time. Results This study involved 111 pure retroperitoneal donor nephrectomies (PRDN) and 111 hand-assisted laparoscopic donor nephrectomies (HALDN). Patients' mean age was 55.7 years, and 59.5% were women; 5.0% underwent right nephrectomy, and 77.0% vs. 23.0% had single- vs. multiple renal arteries. The average estimated kidney graft weight was 160.0 g; actual average graft weight was 155.3 g. The following factors were significantly correlated with operation time in the regression analysis: number of renal arteries, Mayo adhesive probability (MAP) score, estimated kidney graft weight, right nephrectomy, and operation type (PRDN). These five factors were used to create the operation time prediction equation and difficulty scoring system. The multiple r2 value was 0.40 for the operation time prediction equation. Receiver operating characteristic curve analysis of the difficulty scoring system revealed the following: sensitivity: 78.0%, specificity: 64.9%, and c-statistic: 0.76 (95% confidence interval: 0.70-0.83). Conclusions The equation to predict operation time and the surgical difficulty prediction model created in this study are easy to calculate and are accurate. Both may help in selecting an appropriately-skilled surgeon and in improving safety in living-donor kidney transplantation.