AIDE: Decision Support for Anastomosis or Colostomy in Emergency Surgery for Complicated Acute Diverticulitis

AIDE: a Multicenter Observational Study to Support the Decision Between Anastomosis and Colostomy in Emergency Surgery for Complicated Acute Diverticulitis

The AIDE/OBS study is a multicenter, observational, non-interventional study designed to collect standardized clinical, radiological, intraoperative visual, and surgical reasoning data from adult patients undergoing non-elective surgery for complicated acute diverticulitis.

The study focuses on patients requiring urgent or emergency operative management, including cases following failure of non-operative management. The main intraoperative decision of interest is the choice between sigmoid resection with primary anastomosis, with or without diverting stoma, and Hartmann's procedure.

The current phase aims to build a structured multimodal dataset and to validate and refine a preliminary expert-informed decision-support tool. The study does not modify standard clinical practice, surgical indication, operative strategy, or postoperative management. All treatment decisions remain at the discretion of the treating surgical team according to local practice.

Complicated acute diverticulitis requiring non-elective surgery remains a challenging condition in emergency colorectal surgery. In patients with diffuse peritonitis or severe complicated disease, the intraoperative decision between primary anastomosis and Hartmann's procedure is complex and influenced by patient-related factors, disease severity, contamination, tissue quality, bowel perfusion, technical feasibility, and surgeon judgment.

The AIDE/OBS study aims to collect retrospectively and prospectively standardized real-world data from multiple centers to better understand the factors influencing this decision. The study will collect clinical and demographic characteristics, comorbidities, radiological findings including WSES CT-driven classification, intraoperative findings, operative strategy, postoperative outcomes, anonymized intraoperative images or videos when available, and surgeon-reported decision-making factors.

The collected data will be used to validate and refine a preliminary expert-informed AI-based decision-support framework. The long-term objective is to support the development of future artificial intelligence and computer vision tools that may assist emergency surgeons during intraoperative decision-making in complicated diverticulitis.

This is an observational, non-interventional study. No experimental treatment, device, or AI-guided recommendation is applied to patients during the study. Surgical management is performed according to standard clinical practice and local protocols.

Inclusion Criteria:

• Age 18 years or older
• Diagnosis of acute complicated diverticulitis requiring non-elective operative management
• Emergency or urgent surgery for complicated diverticulitis, including surgery after failure of non-operative management
• Minimally invasive approach: laparoscopic/robotic surgery
• Patients undergoing sigmoid resection with primary anastomosis, Hartmann's procedure, or other operative strategy according to local practice
• Availability of clinical and operative data
• Local ethical approval, authorization, or waiver according to national and institutional regulations

Exclusion Criteria:

• Age younger than 18 years
• Elective surgery for diverticular disease
• Uncomplicated diverticulitis managed non-operatively
• Surgery performed for conditions not related to diverticulitis
• Surgical procedures performed via primary open approach (i.e., no attempt at minimally invasive surgery), unless converted intraoperatively.
• Lack of required consent or authorization for data use, when required by local regulations