Colorectal anastomotic leaks (AL) are associated with high morbidity and mortality. Management of AL and its intra-operative decision making is often difficult. The aim of this multi-centre study is to explore different management strategies, including different surgical options, and analyse rates and patterns of failure of initial management.

All consecutive patients who had a confirmed AL after elective colorectal resections from 1st January 2014 to 31st December 2019 were included at seven hospitals across the East of England Region. Morbidity (length of stay, and failures) and mortality were compared across the different management strategies, and survival analyses were performed.

The International Study Group of Rectal Cancer proposed a grading system for the management of colorectal anastomotic leaks (Rahbari et al., 2010). Grade A anastomotic leaks are identified by radiographic findings of a perianastomotic fluid collection, leakage of contrast through the anastomosis, or observation of new drainage of enteric contents through either a drain or through a fistula but without accompanying clinical complaints. These may be managed expectantly. These may become apparent during the preoperative work-up prior to closure of a diverting ostomy and will at least delay reversal. Grade B leakage requires therapeutic intervention but does not necessarily require reoperation. Antibiotics and percutaneous drainage of fluid collections are the most common nonoperative interventions. Grade C anastomotic leakage requires relaparotomy. Surgical treatment is performed with the goal of controlling life-threatening sepsis. The traditional operation with takedown of the anastomosis and end colostomy may be appropriate, but washout with drain placement and diverting loop ileostomy may also be appropriate. 1 Elective colorectal resection surgeries have an over 30-day mortality rate of 5.8%. While emergency resections have a 7.8% mortality (Morris, 2011). Experiencing an anastomotic leaks should not have a higher mortality than the above rates.

Aim The main interest of this study is to firstly assess our practice and management of AL, evaluate whether hospitals in the region are identifying AL as per The International Study Group of Rectal Cancer, grading the AL as Grade A, B, and C above, and analyse the natural evolution of leaks and their subsequent treatment(s).

Outcomes of Interest

Management of anastomotic leaks and associated outcome in terms of:

Primary objectives:

1. 2-year mortality after anastomotic leaks of different grades.

   Standards of reference for audit:

   Elective colorectal resection surgeries have an over 30-day mortality rate of 5.8%. While emergency resections have a 7.8% mortality (Morris, 2011). Experiencing an anastomotic leaks should not have a higher mortality than the above rates.

2. Outcomes of different treatment strategies of Colorectal anastomotic leaks.

3. Grading of AL as described above.

Secondary Objectives include:

• Need for subsequent management strategies after failure of initial strategy.
• Proportion requiring laparotomy, bowel resection, defunctioning.
• Length of ITU stay, overall length of hospital stay
• Non-surgical complications (pneumonia, urine infection, DVT, PE, MI etc)
• Delay / change in oncological management Methods Data will be collected retrospectively going back to 2014 (five years), at multiple East of England sites. The study will initially commence across trusts in the East of England Deanery however efforts will be made to recruit other research collaboratives through existing and established ties with the National Research Collaborative.

Population Identification

Patients can be identified and recruited to the study through a variety of ways:

• Multi-Disciplinary Teams (MDTs): Most MDTs will record data as to who has had a leak and what treatment they have had or are to have
• Colorectal Clinical Nurse Specialists (CNS): In most trusts, the CNS keep records of patients who have had complications
• National Bowel Cancer Audit (NBOCAP): Although this will not capture patients who have had benign disease, it will provide data on patients with cancer
• Consultants: Many consultants collect their own data, and this can be an excellent source of information An estimated of 10-30 leaks occur every year at each site. We expect to have between 30-60 leaks with 3 year follow up per hospital site.

Study duration Data collection 3-6 months from date of project launch (October 2019). Reminders for date of study closure will be sent closer to the time.

Variables

It must be stated that the aim of this study is not to investigate 'leak-rates' or comparatively analyse data of individual surgeons or trusts. No patient identifiable or surgeon identifiable information will be requested. Data and outcomes from individual trusts will not be compared against each other. The name of the trust will be recorded as one of the variables only for data validation purposes and to identify the lead investigators at each trust for the purposes of communication only. Names of Trusts or Surgeon level details will not be published. The variables that will be recorded are as follows:

• Trust
• Local Patient ID
• Gender (M/F)
• Age At Time (of Surgery) (Whole number)
• Indication for surgery (where is cancer, IBD?)
• Comorbidity 1 (in any order)
• Comorbidity 2 (in any order)
• Comorbidity 3 (in any order)
• Comorbidity 4 (in any order)
• Date of Index Surgery
• Type of Index Surgery (right, left, segmental sigmoid colectomy, AR)
• Elective / emergency
• Date Leak Detected
• Leak size (description on CT (size, localized, small, free)
• Symptomatic/Non-Symptomatic Leak
• Treatment 1 (IV antibiotics are always given once a leak is diagnosed. So it should not be inserted automatically as treatment 1. So I would reserve inserting "IV antibiotics" or "conservative" as treatment 1 unless the patients actually had a trial of conservative management.)
• date treatment 1 (chronological order)
• Treatment 2
• date treatment 2 (chronological order)
• Treatment 3
• Date treatment 3 (chronological order)
• Type of Rescue Surgery Performed (?laparotomy, bowel resection, defunctioning)
• Length of Stay in ITU Post Leak (in days)
• Length of stay in Leak episode (total in days, if the leak occurred during the index operation episode, it is the total length of stay in that whole episode.)
• Date CT confirmed leak resolved
• Number of days of drainage (if drain inserted. This column is for radiologically inserted drains; or EUA inserted drains in low pelvic leaks. It is the total number of days a drain was kept in until removal of the drain (including changes).)
• Date of Death
• Date last seen alive (censored)
• Other Comments

Inclusion:

• Age ≥18 years old
• Patients undergoing colorectal anastomosis for any pathology (benign or malignant), during an elective or emergency admission, who had an AL.
• Laparoscopic or open surgery Ethical approval & Registrations This is a retrospective audit, aiming to analyse the region trust's management of colorectal anastomotic leaks, and whether this is in accordance with published literature. This study therefore needs to be registered as a retrospective audit in each trust. Please keep a record of those that were subsequently excluded and the reason why (most probably was not a leak, or not managed as leak).

Inclusion Criteria:

• Age ≥18 years old
• Patients undergoing colorectal anastomosis for any pathology (benign or malignant), during an elective or emergency admission, who had an AL.
• Laparoscopic or open surgery

Exclusion Criteria:

• IBD, Small bowel anastomosis,

Patients can be identified and recruited to the study through a variety of ways:

• Multi-Disciplinary Teams (MDTs): Most MDTs will record data as to who has had a leak and what treatment they have had or are to have
• Colorectal Clinical Nurse Specialists (CNS): In most trusts, the CNS keep records of patients who have had complications
• National Bowel Cancer Audit (NBOCAP): Although this will not capture patients who have had benign disease, it will provide data on patients with cancer
• Consultants: Many consultants collect their own data, and this can be an excellent source of information