Who should record surgical complications? Results from a third-party assessment of complications after radical cystectomy

Objective: In Sweden complications after radical cystectomy have been reported to the nationwide population-based Swedish Cystectomy Registry since 2011. Here, validation of the reporting was assessed in two healthcare regions. Materials and methods: Complications were ascertained from patient records by a third party not involved in the care delivered to 429 randomly selected patients from 949 who had undergone radical cystectomy since 2011 in four hospitals. Without knowledge of the outcome in the primary registration, post-operative complications within 90 days post-operatively were assessed by an independent review of patient charts, and the results were compared with the primary reports in the Swedish Cystectomy Registry. Results: The third-party assessment identified post-operative complications in 310 patients (72%). Low-grade complications (Clavien-Dindo I–II) were noted in 110 (26%) of the patients in the primary registration, but increased to 182 (42%) in the validation (p < 0.00001). High-grade complications (Clavien-Dindo III–V) were reported in 113 (26%) patients in the primary registration, but in 128 (30%) of the patients in the validation (p = 0.02). According to the third-party assessment, 18 patients (4%) had Clavien-Dindo grade IV complications and 12 (3%) died within 90 days of surgery (Clavien-Dindo grade V); corresponding values in the primary registration were 15 (3%) and 9 (2%), respectively. The readmission rate within 90 days increased from 27 to 32% in the validation (p < 0.00001). Conclusions: Compared with registry data, third-party assessment revealed more complications and readmissions after radical cystectomy. Hence such evaluation may improve the validity of reported complication data.