Lack of assessment of 90-d perioperative morbidity in elderly patients after radical cystectomy and pelvic lymph node dissection (PLND) using a standard reporting methodology, and the Clavien–Dindo classification (CDC) does not accurately reflect the burden of complications. We aim to report the 90-d complications of elderly patients after radical cystectomy, and to compare the validity of the Comprehensive Complication Index (CCI) and CDC.
Retrospective review of 280 patients aged ≥75 years who received radical cystectomy between 2006 and 2021. The 90-d complications of elderly patients after radical cystectomy were reported by implementing the EAU criteria. The CDC and CCI were both used for grading complications. The Spearman rank correlation coefficient was used to estimate the correlation between postoperative stay and CDC/CCI. Logistic regression was used to identify the risk factors for major complications. The sample size for a fictive superiority trial was calculated for different endpoints.
A total of 225 (80.36%) patients suffered from 528 complications. The cumulative CCI had a more accurate prediction of postoperative stay than the CDC (r = 0.378, p < 0.001 vs. r = 0.349, p < 0.001). The need for sample size could decrease when CCI was used for the primary endpoint. More risk factors for major complications were identified when CCI ≥33.7 was defined as the endpoint of major complications.
CCI is better than CDC for grading the severity of complications in elderly patients after radical cystectomy and PLND.