TENTACLE Study

TreatmENT of AnastomotiC LeakagE after COLON cancer resection

SUMMARY

Anastomotic leakage (AL) is a common and severe complication after colon cancer resection. Recent literature reports leak rates up to 8.4% and substantial associations with the need for re-interventions, prolonged hospital stay, a higher likelihood of permanent stomas, and mortality. Despite these alarming numbers, comparative studies on treatment strategies are lacking.

In clinical practice, treatment decisions are frequently influenced by factors such as the clinical presentation, resource availability, and surgeon preferences. Several leakage characteristics (e.g. size of the defect, etc.) also play an important role, but there is limited understanding of all these factors and the relative influence of these factors on treatment decisions and subsequent mortality. Additionally, no previous studies have comprehensively examined the effectiveness of various treatment strategies for AL after colon cancer resection considering all these factors collectively.

Objectives:

1. To identify predictive factors associated with 90-day mortality and 90-day Clavien-Dindo grade 4-5 complications amongst patients who developed AL following colon cancer resection and to develop and validate a prediction model for predicting 90-day mortality as well as the co-primary composite endpoint Clavien-Dindo grade 4-5 complications.

2. To explore and compare the effectiveness of various treatment strategies for AL following colon cancer resection, considering patient, tumour, resection and leakage characteristics.

STUDY POPULATION

POPULATION

All consecutive adult patients who underwent colon cancer resection between the 1st of January 2018 and 31st of December 2022 will be retrospectively screened and evaluated for an anastomotic leakage. All patients with an anastomotic leakage diagnosed within 90 days from colon cancer resection are suitable for inclusion.

INCLUSION CRITERIA

To be eligible for inclusion in this study, a patient must satisfy all of the following conditions:

1. Aged 18 years or older;

2. Surgical resection for primary colon cancer (cT1-4b, N0-2, M0-1) with formation of a primary colonic anastomosis above the peritoneal reflection, with or without a diverting stoma;

3. Postoperative AL defined as: “any clinical, radiological or intraoperative signs of disrupted integrity of the anastomosis and/or blind loop. This also includes suspected leaks with any degree of extraluminal air or fluid on CT, perianastomotic abscess, purulent peritonitis without clear anastomotic defect, or any other suspicious condition in which there is no ultimate macroscopic proof of disrupted anastomosis.”

Regarding the type of colon cancer resection, the following patients will also fulfil the inclusion criteria: patients who underwent cytoreductive surgery (CRS) simultaneous with resection of the primary colon cancer with or without hyperthermic intraperitoneal chemotherapy (HIPEC), simultaneous ablations/resections of metastasis, multivisceral resection, emergency resection, patients diagnosed with perforated disease/peritumoral abscess or fistula, and acute obstructions.

EXCLUSION CRITERIA

A participant who meets any of the following conditions will be excluded:

• Surgical resection for benign colon disease;

• Recurrent colon cancer resection;

• Any primary colon malignancy other than adenocarcinoma (e.g. neuroendocrine tumour, gastrointestinal stromal tumour);

• Any clinical condition that does not fulfil the broad definition of AL as used in this study (e.g. only free air on CT that is considered to be compatible with an appropriate postoperative day in the absence of any other clinical signs related to a potential anastomotic leakage)

METHODS

PRIMARY OUTCOME

• 90-day mortality

COMPOSITE PRIMARY OUTCOME

• 90-day Clavien-Dindo grade 4-5 complications [29]

SECONDARY OUTCOMES

• Time from colon cancer resection to diagnosis of AL and to initial treatment of AL.

• Number of reinterventions and readmissions.

• Hospital length of stay and intensive care unit length of stay.

• Mortality (in-hospital, 30-day, one year).

• Stoma present at last possible date of follow-up; type of stoma present.

• Disease status at last possible date of follow-up.

LIST OF STUDY PARAMETERS

This is a retrospective study and it is expected that not all relevant data can be obtained from the patient files. For example, estimation of leak circumference will not be possible without surgery or an endoscopy and it is expected that not all patients underwent surgery or an endoscopy. However, it is expected that the targeted number of patients will provide sufficient data to analyse whether factors with many missing values will be of substantial influence. Therefore, these factors that are likely to have a lot of missing data will be collected, even though missing data may introduce bias.

Hospital characteristics

Upon study entry, each participating centre will be required to complete a questionnaire to gather comprehensive information regarding institutional characteristics. This questionnaire covers the following main subjects: hospital type, annual volume of CRC resections, annual volume of colon cancer resections, anastomotic leakage rates, number of hospital beds and availability of hospital resources, protocols and treatment modalities.

Patient and tumour characteristics

The following patient and tumour characteristics will be collected: sex, age, year of surgery, height, weight, body mass index, Charlson comorbidity index (CCI), American Society of Anaesthesiologists (ASA) classification, smoking status, baseline creatinine and hemoglobin level (i.e. within 3 months

TENTACLE - Colon study Version 2.3 (August 2024)

before colon cancer resection), and history of immunosuppressant medication. The following tumour characteristics will be collected: tumour histology, tumour location, pathological tumour node metastasis (TNM) stage according to the respective Union for International Cancer Control (UICC) classification, preoperative tumour-related complications (e.g. bleeding, obstruction) and corresponding interventions (e.g. ablation/resection of metastasis, diverting stoma) and neoadjuvant therapeutic regimens.

Surgical and intraoperative characteristics

The following characteristics are about preparation before colon cancer resection: emergency admittance before resection (and if so, at what urgency the resection is performed (e.g. acute, urgent, elective)), selective decontamination of digestive tract received, mechanical bowel preparation received. The following surgical characteristics will be collected: intent of resection (i.e. curative, palliative), abdominal approach, conversion and type of conversion, type of resection (e.g. ileocecal resection, left hemicolectomy), (specification of) multivisceral resection, simultaneous resections/interventions (e.g. metastatectomy, peritonectomies as part of cytoreductive surgery with or without hyperthermal intraperitoneal chemotherapy), number of anastomosis, type of anastomosis (e.g. ileo-colostomy, colo-colostomy), configuration of anastomosis (e.g. end-to-end/side-to-end stapled/hand-sewn), (specification of) problems during anastomosis reconstruction, indocyanine green assessment (and corresponding result), (specification of) intraoperative complications (e.g. bowel perforation), blood loss during resection, amount of blood transfusions and duration of resection.

Diagnostic characteristics

Collected diagnostic characteristics will be: date and time of diagnosis of AL, ward at leak diagnosis, antibiotics around leak diagnosis, nasogastric tube around diagnosis, application of various diagnostic modalities (type, timing and result), clinical parameters (i.e. Glasgow Coma Scale, respiratory rate, systolic blood pressure and modified early warnings sign), biochemical parameters (C-reactive protein, leukocyte count, hemoglobin and creatinine levels) and intensive care unit parameters (i.e. need for inotropic support, mechanical ventilation and/or dialysis) around diagnosis and up to leak treatment.

Anastomotic leakage characteristics at diagnosis

The following leakage characteristics at diagnosis will be collected: radiological characteristics (presence and amount of air/fluid, contrast extravasation, ileus), intraoperative characteristics (presence and extent of purulent peritonitis and faecal contamination, bowel ischemia, abscess), location of the leakage (i.e. anastomosis, blind loop, inconclusive) and estimated circumference of the

TENTACLE - Colon study Version 2.3 (August 2024)

leakage, and presence of other complications according to the Clavien-Dindo classification (pulmonary, cardiac, gastrointestinal, urologic, thromboembolic, infection, other infections requiring antibiotics, other complications) with corresponding grading. [29] The comprehensive complications index (CCI) will be calculated from all scored complications. [30]

Treatment of anastomotic leakage

All therapeutic strategies including primary, secondary and tertiary treatment strategies, as well as the modalities employed for managing AL. This comprehensive data collection includes: date and time of start treatment, need for re-admission, applied treatment modalities (i.e. antibiotic regimens, radiologic, endoscopic, (specification of) surgical procedures, and/or abstination/palliative care.

Follow-up of treatment

The following characteristics will be collected to assess outcomes of leakage treatment: leak healing and modality that confirmed leak healing, date of initial discharge, total intensive care unit length of stay, date of death if applicable, presence of disease recurrent (and if so, date and location of recurrence), (type of) stoma present at last moment of follow up, total number of radiologic and surgical interventions one year after colon cancer resection and (start date of) adjuvant chemotherapy.