External validation of the MiGUTS nomogram for the prediction of bleeding control intervention after renal trauma.

Abstract

Introduction: The American Association for the Surgery of Trauma (AAST) renal trauma grading leads to a variable management of patients with high-grade renal injuries. For a better prediction of the risk for bleeding interventions, Keihani et al. introduced the multi-institutional genito-urinary trauma study (MiGUTS) renal trauma nomogram in 2019. The aim of this study was to conduct an external validation and generalization for all kidney trauma cases of the nomogram with a European cohort of a Swiss level 1 trauma center.

Methods: We collected data from the clinical information system of the Kantonsspital Graubünden, Chur, Switzerland. All patients ≥ 18 years of age from 01.01.2008 to 01.12.2020 with a renal trauma who underwent computed tomography imaging of the abdomen were included. The descriptive analysis was performed by a t-test/Wilcoxon signed-rank test and a Chi-square test. The predictions of the nomogram were analysed by the Pearson correlation coefficient. The threshold of prediction of a bleeding intervention was optimized by a ROC analysis.

Results: Overall, 166 patients were included. Most patients were male (80.7%) with a median age of 44 years. Using the prediction from the MiGUTS nomogram developed by Keihani et al. we were able to identify a threshold with a sensitivity of 1.00, specificity of 0.87, positive predictive value of 0.44, negative predictive value of 1.00 and accuracy of 0.88.

Conclusion: The MiGUTS nomogram by Keihani et al. demonstrated to be reliable in the prediction of an intervention for bleeding control in our validation study in a European cohort.