Intracorporeal versus extracorporeal anastomosis in right colectomy – a protocol for a randomised trial

The gold standard in the treatment of right-sided colonic cancer is minimally invasive right colectomy [1], but at present, no standardised technique exists for ileocolic anastomosis. Surgeons, therefore, perform the procedure according to local practices and experience. The procedure is performed as a “minimally invasive assisted” technique – a combination of open and minimally invasive surgery with an extracorporeal anastomosis (ECA) or a “totally minimally invasive” technique with intracorporeal anastomosis (ICA).

The possible advantages of ICA are 1) a suprapubic incision associated with a reduction in pain, fewer pulmonary and wound complications [2], 2) no externalisation of the intestines, and 3) less extensive mobilisation of the transverse colon. The combined effect is thought to mitigate the surgical stress response and improve short-term outcomes compared with ECA. The disadvantage of ICA is that it is perceived to be technically more challenging [3], possibly because it requires proficiency with intracorporeal suturing.

The prevailing technique is ECA. However, there is no conclusive evidence on the superiority of one technique over the other. Systematic reviews on the subject conclude that ICA results in faster post-operative recovery with a faster return of bowel function and a shorter length of stay and improved short-term outcomes [4-7]. However, the literature predominately consists of studies with either a retrospective or a prospective, non-randomised design and are thus subjected to the inherent biases of these approaches. Only three recently published randomised controlled trials [8-10] exist. No current studies have reported on patient-reported recovery outcome parameters.

The objective of this study was to compare ICA with ECA in robotic right colectomy. We hypothesise that ICA leads to an improved post-operative recovery compared with ECA.

METHODS

Study setting and population

This is a multicentre, triple-blind, randomised controlled trial comparing robotic right colectomy with ECA with robotic right colectomy with ICA. We plan to include 100 patients undergoing elective minimally invasive right colectomies in two colorectal centres in Denmark. Both centres are highly specialised in the treatment of colorectal cancer and competent in both techniques investigated in the study.

Inclusion criteria

Eligible are patients above 18 years of age, American Society of Anesthesiologists classification system (ASA) I-III, with verified or clinical suspicion of malignant epithelial colonic lesion proximal to mid-transverse colon scheduled for minimally invasive right colectomy or extended right colectomy with curative intent. Laparoscopic and robotic approaches are accessible in both centres. If patients are enrolled in the study, they are scheduled for a robotic approach. No other selection criteria for a robotic or laparoscopic approach apply. If logistic circumstances result in the robotic approach not being available on the day of surgery, the patient is not eligible but is registered in the screening log.

Exclusion criteria

Not eligible are patients undergoing an emergency procedure or major concomitant procedures and patients who are unable to comply with study requirements, provide informed consent, or have lesions located in the distal transverse colon or for whom a splenic flexure is planned for extended right colectomy. In case of protocol violations or withdrawal of consent, patients are excluded and a detailed description is published as part of the results of the study.

Randomisation

Patients are randomised in a 1:1 ratio to either ICA or ECA in robotic right colectomy.Randomisation is performed with a random sequence generator with random block sizes of 4-10. The randomised sequences are concealed in sequentially numbered, opaque envelopes by an independent member of staff with no further involvement in the study. When patients are recruited, they are assigned a unique study ID and the envelope with the corresponding number is opened at the time of randomisation.

Randomisation is done intraoperatively before mobilisation of the transverse colon. Randomisation will be completed only if both techniques are possible. A thorough description of the patients who are not randomised will be included in the final publication. After randomisation, data will be analysed according to the intention-to-treat principle.

Requirements to the surgeon

Only surgeons who are proficient in both techniques may contribute to the study. Proficiency is self-assessed and with no number of cases needed. Even so, surgeons are required to have performed both techniques routinely before contributing to the study. The number of cases for whom the surgeon has performed surgery is registered for each patient in order to assess any residual learning curve effect.

Surgical technique

Pneumoperitoneum is achieved by the method of choice of the operating surgeon. A camera port is placed close to the umbilicus, and three additional robotic ports are placed, one in the suprapubic area, one in the right iliac fossa, and one in the upper left quadrant. Finally, a 12-mm trocar for the assistant is placed in the left side of the abdomen. Additional assisting trocars may be placed in any number of locations as needed. An initial assessment is made by laparoscopy and the robot is docked.

A medial to lateral approach is applied. The ileocolic pedicle is isolated and ligated and the dissection is continued to the lateral-side wall. The right branch of the middle colic vessels is isolated and ligated. If an extended right colectomy is performed, the middle colic artery is isolated and ligated centrally. The gastrocolic ligament is taken down and the hepatic flexure is mobilised. Randomisation will be completed before mobilisation of the transverse colon. The rest of the procedure follows the description of either ICA or ECA.

Intracorporeal anastomosis

No further mobilisation is performed, and an appropriate site of transection of the transverse colon is identified in pursuance of the normal principles of oncological resection, visual inspection of colour change of the bowe, and a non-quantitative ICG perfusion test. A laparoscopic or robotic stapler is used to transect the transverse colon and terminal ileum, and the resected specimen remains inside the abdomen until the end of the procedure. The ileum and transverse colon are brought together and a stay suture may be applied at the surgeon’s preference. An isoperistaltic side-to-side anastomosis is created through two small enterotomies on the antimesenteric sides of the transverse colon and ileum. A laparoscopic or robotic 45/60 mm linear stapler is applied with a jaw inserted in each enterotomy, and the residual defect is closed in a single layer by laparoscopically continuous suturing with a barbed suture. The specimen is removed with a wound protector in place through a suprapubic transverse incision – if suitable, by enlargement of the suprapubic port.

Extracorporeal anastomosis

Mobilisation of the transverse colon is completed, and the intestines are brought extracorporeally by a right-sided transverse subcostal incision with a wound protector in place and, if possible, with muscle preservation. An appropriate site of transection of the transverse colon is identified in pursuance of the normal principles of oncological resections, visual inspection of colour change of the bowel, and test for bleeding/pulsation of the marginal artery. The ileum and transverse colon are transected extracorporeally by sharp incision, and the specimen is removed from the operative field. The two ends of the bowel are aligned for ECA and an end-to-end hand-sewn anastomosis with continuous one-layer suturing is performed.

Outcomes

The primary outcome is patient-reported post-operative recovery measured by the “Quality of Recovery – 15 Danish” (QoR-15D) questionnaire (Table 1 and Table 2). The QoR-15 is a validated, patient-reported outcome measure of post-operative quality of recovery on a scale ranging from 0 to 150 [11-13].

Secondary outcomes are additional measures of post-operative recovery, intraoperative outcomes, post-operative outcomes, and pathophysiological tests (Table 1 and Table 2).

Timing of the assessments and blinding of the outcomes are listed in Table 2. Patients will be followed until 90 days after surgery, during which period medical records will be evaluated for morbidity, readmittance, reinterventions, and mortality.

Blinding

Only the operating surgeons and other staff members present at the operating theatre are knowledgeable about the actual intervention performed.The surgeons will describe all procedure-related details in a separate “surgeon’s case report form” document, which is stored in an opaque, sealed envelope, and no information related to the anastomosis is available in the patient chart. Treatment assignment will be revealed after completion of all study-related questionnaires and tests on post-operative day (POD) 14. If it is not possible to maintain blinding until completion of all study-related questionnaires or tests on POD 14, this is reported.

Research staff are blinded to the intervention. Research staff assist patients with questionnaires, assess post-operative recovery (gastrointestinal function, post-operative pain, post-operative nausea and vomiting, and mobilisation), and perform pathophysiological tests. As the specimen extraction sites in ICA and ECA are different, blinding of these groups is maintained by applying large abdominal wound patches covering both potential specimen extractions sites immediately after surgery. The operating surgeons do the daily rounds including an assessment for post-operative complications and of the wounds with replacement of the abdominal wound patch in case of permeation.

Patients are blinded to the intervention. Obviously, patients only feel discomfort from one site, but specifically, they are not informed about which incision relates to ICA and ECA.

A blinded data analyst with no relation to any other parts of the study will perform the statistical analyses.

Perioperative care

The operation is performed under general anaesthesia and all patients are managed according to the anaesthetic protocol regardless of the intervention (Table 3). All doses of administered medicine and violations of the anaesthetic protocol are registered.

Irrespective of the intervention performed, patients are managed with an enhanced recovery after surgery (ERAS) protocol following the “ERAS society colorectal guidelines” [14] implemented at both centres.

Discharge criteria are pre-specified as 1) adequate pain relief on oral analgesia regimen, 2) tolerance of oral intake, 3) return of bowel function/flatus and 4) absence of unmanaged complications.

Timeline

Enrolment was initiated in the first centre in May 2018 and in the second centre in September 2019. Completion is expected in January 2021.

Study size

Following our power calculation, we plan to enrol 100 patients in the study. The power calculation was based on QoR-15D data from patients who underwent minimally invasive colonic resections with a standard deviation (SD) of 23 at baseline and an assumption that a decline of 10% (15 units) from baseline to POD 2 is clinically significant. At a power of 0.80 and a two-sided alpha significance level of 0.05, 42 patients are required in each study arm. Based on a dropout rate (conversions, exclusions, etc.) of 20%, we plan to include 50 patients in each arm.

Statistical analysis plans

All statistical analyses will be conducted by a blinded biostatistician after data collection has been completed in accordance with the current statistical analysis plan.

Primary analysis

For the primary outcome of QoR-15D, we plan to conduct a mixed analysis of variance (ANOVA) with a between-subjects factor with two levels (treatment: ICA or ECA) and a within-subjects factor with six levels (time: baseline, six hours post-operatively, POD 1, POD 2, POD 3, and POD 14) provided the assumptions of sphericity and approx. normal distribution are met.

With a median length of stay of 2-3 days in patients undergoing minimally invasive right colectomy in centres with a high adherence to ERAS protocols, the post-operative recovery on POD 2 is of particular clinical interest. Hence, we will conduct a separate analysis of covariance (ANCOVA) on the specific difference in QoR-15D from baseline to POD 2 with the baseline value of QoR-15D as an independent variable.

A post-hoc sensitivity analysis with consideration of “as-treated” analysis, subgrouping, definitions, statistical assumptions, missing data strategy, etc., will be applied with guidance from the results of the primary analysis. A two-sided alpha value of < 0.05 will be considered significant.

Secondary analyses

Continuous variables will be expressed as means ± SD and categorical data as frequencies. Distributional assumptions of normality of continuous data will be assessed through visual inspection of histograms and Q-Q plots or Shapiro-Wilk tests. Normally distributed and not-normally distributed data will be analysed with parametric and non-parametric tests, respectively. Dichotomous and categorical data will be analysed with Fisher’s exact test and χ² test, respectively. If applicable, we plan to conduct a mixed ANOVA on clinical outcomes with repeated measurements and an ANCOVA on the specific difference from baseline to POD2 (e.g., post-operative pain, blood test analysis, etc.) in accordance with the clinical justification in the primary analysis.

Data sharing statement

Researchers who provide a methodologically sound proposal will be able to attain the de-identified individual participant data that underlie the results reported in the final article (text and tables) for any purpose.

Trial registration: Prior to commencement, this study was registered with Clinicaltrials.gov (ID NCT03130166).

Discussion

The continuous striving to minimise the impact of surgery on patients together with the increased adoption of robotic surgical systems have revived the interest in the ICA technique. The present study will contribute with valuable knowledge on the best surgical management of right-sided colonic cancer.

Correspondence Niclas Dohrn. E-mail: niclas.dohrn@regionh.dk
Accepted 13 April 2021
Conflicts of interest Potential conflicts of interest have been declared. Disclosure forms provided by the authors are available with the article at ugeskriftet.dk/dmj