Inter-observer Reliability in Reporting Complications of Transurethral Resection of the Prostate with Clavien-Dindo Classification: A Retrospective Observational Study

Abstract

Introduction: Transurethral Resection of the Prostate (TURP) is a gold standard treatment for Benign Prostatic Hyperplasia (BPH). The Clavien classification system was proposed in 1992 to grade complications of general surgery, which was then modified in 2004 by Dindo. In 2017, the European Association of Urology (EAU) endorsed its use for reporting complications in urological procedures. However, the system relies on observers to grade complications, and different observers may assign different grades to the same negative outcome, highlighting the need for uniformity. Aim: To assess the inter-observer reliability and uniformity among urologists in reporting complications of TURP using the modified Clavien-Dindo classification system. Materials and Methods: A retrospective observational study was conducted at Department of Urology, Shyam Shah Medical College, Rewa, Madhya Pradesh, India, from January 2018 to May 2023. The study included symptomatic patients with BPH and prostate gland sizes greater than 40 cc. Negative outcomes were evaluated and complications were rated by nine different raters. Inter-observer reliability was tested using weighted kappa with Statistical Package for Social Sciences (SPSS) version 21.0 (Fleiss’s kappa test). Results: A total of 523 men with BPH who underwent TURP were evaluated, with a mean age of 66.7±5.2 years and a mean prostate volume of 62 cc. Adverse events occurred in 96 patients (18.36%), and these complications were classified according to the Clavien-Dindo System (CDS) by nine different blinded observers. The rating of these complications was then assessed for concordance. A concordance rate using weighted kappa of 0.847 was observed, indicating high inter-observer reliability for TURP. Conclusion: Clavien-Dindo’s system demonstrated good interobserver concordance and reliability. Some modifications to the grading system are needed, taking into account the impact of pre-existing co-morbidities and auxiliary procedures. Largescale prospective studies are recommended to further validate these findings.