Abstract
Purpose
To identify a cyclical pattern in short-term postoperative outcomes after colorectal cancer (CRC) surgery across the year.
Methods
Observational study including all consecutive patients diagnosed with CRC who underwent oncological surgery between January 2012 and December 2023. A time series study was performed to identify a possible cyclic pattern of postoperative complications. Non-supervised learning techniques were used to identify months of surgery with similar outcome variables in the time series. Multivariable analysis with logistic binary regression was conducted to identify independent risk factors for postoperative complications.
Results
A total of 1576 patients met the inclusion criteria. The postoperative morbidity and mortality rates were 41.1% and 3.7%, respectively. A cyclical pattern was identified, suggesting that postoperative complications were periodically higher in some months across the year. Three different clusters were identified. Group 1: January, April, July, and August; Group 2: February, May, September, November, and December; and Group 3: March, June, and October. No differences in baseline characteristics were found between clusters. Group 3 presented the highest rate of anastomotic leak (p ≤ 0.001;OR = 1.61, 95%CI = 1.30–2.00), unnoticed perforation p = 0.027;OR = 1.86, 95%CI = 1.18–2.93), diffuse postoperative peritonitis (p = 0.018;OR = 1.50, 95%CI = 1.10–2.04) and needed more postoperative reoperations (p = 0.013;OR = 1.33, 95%CI = 1.07–1.65). Multivariate analysis revealed male sex (p = 0.002), duration of operation (p = 0.017) and month grouping (p ≤ 0.001) as independent risk factors for anastomotic leak.
Conclusion
Postoperative complications after CRC surgery follow a cyclical pattern, but without seasonal distribution. Three well-defined clusters with different postoperative outcomes have been identified. Month cluster was one of the independent risk factors for anastomotic leak.