ABSTRACT
BACKGROUND: Patients undergoing radical cystectomy are predominantly elderly with many comorbidities and high risk of complications. Studies on comorbidity and complications following cancer surgery are often based on data collected retrospectively from records. However, prospective registration is often considered a more valid source of information. Therefore, it is relevant to investigate if the amount and severity of complications and comorbidities is valid when using retrospective registration compared to a more meticulous prospective registration. OBJECTIVE: To investigate the difference in registered comorbidities and complications between prospective and retrospective data collection in patients with bladder cancer undergoing radical cystectomy. METHOD: Seventy-three bladder cancer patients undergoing radical cystectomy were randomized to receive prospective or retrospective collection of data regarding comorbidities and complications. Data in the prospective arm was collected daily during hospitalization, 14-days after discharge and 90-days postoperatively. In the retrospective arm, medical records were reviewed retrospectively at 90-days. Comorbidities were compared using the Charlson Comorbidity Index (CCI) and complications were reported as overall, minor and major dependent on Clavien Dindo Classification (CDC). The primary endpoint was the difference in overall complication rate. RESULTS: No statistically significant difference in CCI was observed with median [IQR] 2[0;3] and 1[0;2] (p = 0.21). No statistically significant difference was found regarding all, minor (CDC I-II) or major (CDC III-V) complications at all three time points. CONCLUSION: No statistically significant difference in comorbidity and complications between retrospectively and prospectively collected data was observed. We find that retrospective collected data is reliable when strict reporting guidelines are used in this single-centre study.
