Trial Information

INtracorporeal Versus EXTracorpoREal anastoMOsis After Laparoscopic Right Colectomy for Cancer: a Randomized Clinical Trial. (IN EXTREMO Study)

Background. Only a few retrospective publications comparing these anastomotic techniques and
no randomized clinical trial has been performed. We conducted a meta-analysis of the
existing non randomized comparative studies on the argument: the results suggest that the IA
results in better postoperative recovery outcomes, such as shorter hospital stay, faster
bowel movement recovery, faster first flatus, faster time to solid diet and lesser analgesic
usage.

Design: Two-armed, double-blinded, randomized, controlled study.

The patients will be randomized into two groups:

A. After laparoscopic right colectomy for cancer, anastomosis will be carried out
intracorporeally (IA) B. After laparoscopic right colectomy for cancer, anastomosis will be
carried out extracorporeally (EA)

Primary endpoint: Surgical morbidity rate defined as any diagnosed morbidity related to
surgical technique (anastomotic leakage, anastomotic bleeding, wound infection, ileus)
within 60 days from surgery.

Secondary endpoints:

- Intra-operative: operative time, largest incision length (mm), number of nodes
harvested, intraoperative complications.

- Early postoperative: mortality, non-surgical site complications.

- Postoperative recovery: bowel movement, first flatus, time to solid diet, nasogastric
tube (NGT) reintroduction, day of analgesic usage, length of stay, readmission.

Randomization. The randomization will be obtained through a computer generated program
(GraphPad Software, Inc, La Jolla, California, USA). The randomization, will be done before
the beginning of the study and it will consist with an ordered series of numbers with a
corresponding letter: A for intracorporeal anastomosis and B for extracorporeal anastomosis.
This series of numbers will be kept in a central dedicated web database; after right colon
cancer diagnosis, if the patient fulfills the inclusion criteria and signs the consent a
number, progressive as in order of inclusion, will be assigned to him or her. The letter
corresponding to the number, and hence the type of procedure to perform, will be notified to
the operating surgeon, who will proceed accordingly.

Blinding. A second surgeon team (different from the one who participate in the
intervention), aware of the operative findings but not the management of the anastomosis,
will then assume the care of the patient. Postoperative care and ability to be discharged
from the hospital will be determined by the second surgical team. This second surgical team
will be blinded to the kind of anastomosis. The primary operative team will be in every
moment available for emergent consultation.

Power Calculation. The reported prevalence of surgical morbidity after laparoscopic
colectomy varies widely from 5-15%16-39, usually depending on the definition of
complications and the type of resection performed. The most recent Cochrane review on the
argument indicates a 8.6% of local morbidity (Wound infection, anastomotic insufficiency,
postoperative ileus, postoperative bleeding)40. Therefore the number of patients required
was based on the hope of improving the rate of surgical morbidity from 8.6% to 4.3%;
considering a drop-out rate of 5%, 384 patients (192 in the IA group and 192 in the EA
group) are necessary to ensure an 80% power with an alpha of 5%, when using a two-sided
log-rank test. The statistical power sample calculation was carried out using PASS 2005
(Power Analysis and Sample Size, within the statistical package "NCSS 2004 and PASS 2005").

Sixty-four patients will be included by each of the six centers of the consortium. Only
surgeons with at least 30 right colectomies performed with intracorporeal anastomosis will
be allowed to recruit patients as first operator.

Statistics. To evaluate significance of differences between the two groups, chi-squared and
Fisher's exact test will be used as appropriated for categorical variables, and the
non-parametric Mann-Whitney U-test for continuous variables. Significance level will be set
at 5%. The statistical analysis will be carried out using the Statistical Package for the
Social Sciences (SPSS version 13; SPSS Inc. Chicago, Illinois, USA).

Monitoring. as commonly requested for Randomized Controlled Trials on patients, a strict
control of outcomes has been planned during the study. Three experts in laparoscopic
colorectal surgery will be designed as member of IN EXTREMO Trial Monitoring Committee. They
would have access to the data during the whole course of the study and indicate cessation of
the trial if one arm is providing manifestly inferior results.