323 Summary, general discussion and future perspectives is confident in the practicality of effectively incorporating the evidence-based reporting elements into local practices and anticipates their acceptance by stakeholders in the end. During the CoReAL project, despite thoroughly reviewing all available literature on leaks, we encountered limitations in formulating evidence-based recommendations for all pertinent topics. Some of the evidence was too low to draw clear conclusions related to AL. For preoperative measures this included the comparison of preoperative selective decontamination to broad-spectrum antibiotics, the role of anemia correction and effect of oral nutritional supplements. Additionally, intraoperative evidence was scarce on potential human factors that influenced leak rates, the exact effect of anastomotic configurations and the role of intraoperative risk scoring systems. Lastly, the evidence on postoperative scoring systems, peritoneal biomarkers, postoperative laxatives or low fiber diet, incidence of chronic sequelae, financial consequences, and impact on Quality of Life (QoL) were too scarce to draw strong statements. We aimed to solve the knowledge gap of this last topic (QoL) in Chapter 9, but all other topics still need to be further addressed into systematic reviews with meta-analyses if possible, or large trials to be able to create additional recommendations. While we acknowledge that accurate reporting of leaks using our proposed framework can potentially lead to prevention (through preoperative risk assessment and intraoperative measures), as well as earlier diagnosis (detailed reporting during the index admission), the utilization of algorithms for predicting AL is another intriguing aspect that may enhance patient outcomes. We are cognizant of the REVEAL study, a prospective observational investigation aimed at developing algorithms for assessing the risk of developing AL 6. The two main goals of this study are to develop and validate an algorithm for predicting the preoperative risk of AL by incorporating various risk factors along with inflammatory, immunerelated, and genetic parameters, and to develop an algorithm for the post-operative diagnosis of AL at an earlier stage. If these algorithms work well, it would be from great value to additionally include them within our reporting framework. Final outcomes from the REVEAL study are expected in the short term and can hopefully help to predict AL and enhance early recognition and fast diagnosis. Also, research has shown that machine learning techniques have high predictive value for forecasting postoperative complications following CRC surgery in general 7-9. As summarized in chapter 3, the risk of AL is influenced by modifiable and nonmodifiable risk factors. It is therefore important to include both while developing predictive systems, and to gain insight into potential interactions. Additionally, the gut microbiome has recently emerged as playing a significant role in the pathophysiology of AL 10. Certain pathogens, like Enterococcus faecalis (E. faecalis) and Pseudomonas aeruginosa, have been implicated in causing AL, but the mechanisms behind their proliferation remain yet unclear 11. It is hypothesized that pathogens like E. faecalis can contribute to the development of AL due to their elevated collagenase activity and the activation of matrix metalloproteinase 9 (MMP9), which are crucial factors in tissue degradation and intestinal inflammation 12. It is also conceivable that a decrease in microbial diversity could prompt a shift towards a pathogenic 14
RkJQdWJsaXNoZXIy MTk4NDMw