Predefined and Standardized Complication Catalogs: An Essential Tool to Assess Reproducible Estimates of Overall Morbidity After Urologic Surgery

Dear Editors:

On behalf of the Urothelial Carcinoma Working Party of the European Association of Urology (EAU) Young Academic Urologists (YAU), Albisinni et al recently published a large multi-institutional series of almost 1000 patients undergoing robot-assisted radical cystectomy (RARC) with intracorporeal urinary diversion (ICUD). ¹ We have read the article with great interest and we would like to compliment the authors on putting together such large cohort from robotic referral centers with prospectively assessed short-term morbidity data. Thorough complication reporting using validated classification systems, such as the Clavien–Dindo classification (CDC) and the Comprehensive Complication Index (CCI^®), is paramount. ^2,3

Particularly the CCI has found its way into urologic surgery morbidity reporting for the past couple of years, ⁴ as it could be shown that it may confer a significant advantage in mirroring cumulative morbidity of a procedure compared with the CDC alone, which only reflects one single complication—the one with the highest grading. Despite this benefit, to date, unfortunately the CCI did not find its way into the EAU guidelines of standardized complication reporting. ² These elaborated guidelines include 14 quality criteria of reporting (Table 1), aiming at improving accuracy, comprehensiveness, and comparability of morbidity reporting between institutions.

In this context, two guideline criteria are unfortunately missing in this study. In our opinion, this should be kept in mind when interpreting the encouraging moderate overall complication rate of only 53%. ¹ Complications were grouped into subgroups such as bleeding, infectious, or gastrointestinal complications, but the authors did not provide definitions of complications or define procedure-specific complications (criteria nos. 6 and 7 of the EAU guidelines). ² Although the study was launched in 2020, patients were included from 2015 to 2020.

Thus, even if complication reporting was performed in some kind of prospective matter at each center, it is rather unlikely that all centers collected morbidity data according to the exact same complication catalog or data dictionary. There is huge variability in defining complications, ³ which holds particularly true for minor adverse events. Without such predefined data dictionary regarding the exact nature of the complication and the subsequent intervention(s), it is almost impossible to draw meaningful conclusions regarding the actual operative morbidity.

There is a huge bias toward the classical morbidity endpoint “overall complications,” as the latter may be significantly underestimated. Particularly for minor complications not requiring a surgical, endoscopic, or radiologic intervention (i.e., CDC grade ≤II) there is a huge variability in reporting. We believe that this is highly relevant for interpretation of raw complication rates, as we have recently performed a similar comprehensive reporting of short-term morbidity after RARC and ICUD from a high volume robotic referral center and found much higher overall complications of 92%, ⁵ which resembles similar 30-day morbidity from a proof-of-concept open RC series in >500 patients. ⁴ Both of these studies entail detailed procedure-specific catalogs of how complication were defined, grouped, and reported.

That said, we congratulate Albisinni et al for their important study, which is another vital component in raising awareness of standardized reporting and the implementation of the CCI in urologic oncology. We are looking forward to future guideline-adherent ² investigations integrating the CCI in morbidity reporting of complex urologic procedures. This entails exciting prospects even beyond an evolving procedural quality, patient counseling, and quality of life.