With the development of laparoscopic technology and staplers, the difficulty of totally laparoscopic resection and intracorporeal anastomosis (IA) is greatly reduced (1), and laparoscopic colectomy has been widely used in the treatment of colon cancer. Nonetheless, the complication rate after laparoscopic surgery is still high, which in part is related to the failure of anastomosis and reconstruction of the digestive tract (2). In addition to the anastomosis methods (side-to-side anastomosis, end-to-side anastomosis, etc.), it has been suggested that different anastomotic approaches may affect the recovery after laparoscopic right hemicolectomy in the previous studies, which can be divided into IA and extracorporeal anastomosis (EA) (3). IA is a critical step in total laparoscopic right hemicolectomy, in which all resection and anastomosis of bowel are performed intracorporeally, with a small incision to dislodge the intraoperative specimen (3); while EA refers to the laparoscopic-assisted resection and anastomosis after the externalization of the bowel through an abdominal incision (4). The theoretical advantages of laparoscopic IA include but not limit that it would avoid having to extract the bowel and mesentery through the thick abdominal wall in obese patients; it would reduce the hazard of intestinal and mesenteric twisting and tension during EA (5, 6). However, intestinal bacteria cannot be completely eliminated even by adequate preoperative bowel preparation and the bowel requires to be opened during IA, hence it is difficult to prevent the abdominal environment from bacterial contamination due to the leakage of intestinal contents and the contact of linear staplers with intestinal lumen. In addition, the technical difficulty of IA has discouraged some young doctors. Therefore, there are still some disputes about the impact of IA and EA on the perioperative period. Our hospital is a high-volume public tertiary hospital with a dedicated colorectal cancer team which has developed laparoscopic colectomy since 2013 and performed stapled anastomosis through IA or EA. This study reviewed laparoscopic right hemicolectomy through IA or EA in recent years and compared their intraoperative and postoperative outcomes.

From September 2019 to April 2022, 44 males and 34 females were included in this retrospective study with a mean age of 61.97 years (range: 40 - 88 years) and a mean BMI of 22.74 kg/m2 (range: 15.33 - 32.95 kg/m2). Among them, 28 patients underwent EA and 50 patients underwent IA. All patients are ASA II-III and survival within 30 days of surgery. There was no difference in demographics and clinical characteristics between the two groups (P>0.05) ( Table 1 ).

Both groups completed laparoscopic right hemicolectomy and lymph node dissection. Intraoperative conditions of two anastomotic approaches are shown in Table 2 . None of selected patients were converted to laparotomy. No statistical difference was observed with respect to lymph node dissection and operation time (P>0.05), yet anastomotic time between the two groups were significantly different (P<0.001); the anastomosis time of EA group was significantly longer than that of IA group. Compared with EA, the IA group had less intraoperative blood loss (median: 30 mL, P=0.010) and smaller incision (median: 3 cm, P<0.001) ( Table 2 ).

Postoperative outcomes of two groups are manifested in Table 3 . No statistical difference between the two groups was found in terms of postoperative hospital stay (P=0.294), totally hospital stay (P=0.366) and NLR during POD1 and POD3 (P>0.05), but the postoperative pain scores of IA group significantly decreased (P<0.001) and the first fart time of EA group significantly delayed (P=0.006). 27 patients in the two groups had at least one postoperative complication, including anastomotic leakage, anastomotic bleeding, abdominal infection, surgical site infection and postoperative ileus, and thereof the incidence of abdominal infection between groups existed statistical difference (P=0.044), while there was no significant difference in other complications ( Table 3 ).

Patients in the IA group were divided into three groups based on timeline: from September 2019 to August 2020 (group A), from September 2020 to June 2021 (group B), and from July 2021 to April 2022 (group C). 6, 3, and 0 cases of abdominal infection were observed respectively among the 3 groups (P=0.040), without difference between each group (P>0.05) ( Table 4 ).

The feasibility and safety of laparoscopic colectomy in terms of radical resection have been demonstrated by some trials with high-level evidence (8, 9). However, different types of anastomotic approaches have their own drawbacks. Theoretically, EA requires a larger abdominal incision compared to IA to achieve ileocolic exteriorization, and undue traction on the mesentery may lead to bleeding during EA (10). The result of our study demonstrated that intraoperative blood loss and incision length were reduced in IA different from EA, and a smaller incision achieved by the IA technique can decrease the hazard of incisional hernias and postoperative pain (11, 12). In our study, the postoperative pain complained basically came from the surgical incision. The incision of TLC with IA is mainly used for the removal of surgical specimens, including trans-umbilical incision which our team selected, Pfannenstiel incision and trans- McBurney incision, natural orifice. Although some studies have pointed to a lower incidence of incisional hernias using the Pfannenstiel incision compared with using the umbilical incision (13–15), none of incisional hernia was observed in our patients, so this view has not been confirmed by us yet. The umbilicus was usually employed for a Trocar and slightly enlarged to extract the specimen to avoid adding an incision. We predicted this would alleviate postoperative pain, less adhesions and offer decent cosmesis. Then our study verified this hypothesis, which tallied with the findings of Fabozzi et al. (16).

Our study showed that the first postoperative fart time of patients in IA group was faster than EA; previous studies have also found that patients receiving IA recover their diet and defecate more quicker than EA (17, 18). These evidences suggested that bowel function recovery after IA was faster, and less bowel manipulation and reduction of traction were thought to contribute to it (10, 19). In some research of IA, the bowel peristalsis recovery after overlap anastomosis (OLA) was found faster than after FEEA, which may be due to isoperistaltic pattern after OLA creating more natural digestive reconstruction over antiperistaltic pattern after FEEA (20, 21). However, these studies also showed that the difference between two anastomotic mode was not significant in the incidence of post-complications; On the other hand, the OLA is more complicated than FEEA, which will prolong the anastomotic and operative time (20–22). Our chief surgeon was adept at FEEA and we opined that the theoretical advantages of OLA may be counterbalanced by our extensive experience of FEEA.

IA reduces tissue trauma by better visualization of anastomosis and abdominal manipulation under the laparoscopic lens, which may also be one of explanations for less intraoperative bleeding or one of reasons for less postoperative inflammatory reaction. NLR is a validated prognostic scoring system, which uses two different leukocyte count components to predict short-term systemic inflammatory response; previous literatures suggested that IA had less inflammatory reaction than EA (17, 23). In contrast, there was no significant difference in NLR between the two anastomoses in our study, which was speculated to be related to the postoperative complications of the two anastomotic approaches.

Some scholars opined that EA needed more time to arrange the bowel and mesentery to avoid mesenteric torsion (10, 24). Although our study discovered that the anastomotic time of IA was shorter than EA, there was no significant difference respect to the operative time, probably because the extra time was taken in separating the mesentery and ileocolic vessels intracorporeally (25). Some studies (18, 26, 27) even found that TLC required longer operative time than laparoscopic-assisted colectomy (LAC), which may be related to the difficulty of performing IA. We conjectured that the mastery of IA and tacit understanding of the team cooperation would curtail the anastomotic time and alleviate the difficulty of IA operation. Hanna et al. (6) also found the operative time was improving steadily by analyzing the learning curve in IA, believing the overall operative time of TLC by surgeons well versed in IA should be comparable to that of LAC (5, 17). In addition, it is difficult to judge whether the bowel or mesentery twisted owing to the limitation of the assisted incision during EA; While IA reduce the time of avoiding intestinal and mesenteric twisting and tension depending on clear laparoscopic observation. Different types of sutures and materials may also impact the effect of anastomosis. It has been pointed out that the post-complications of double-layer anastomosis is less than single-layer anastomosis, and the use of barbed sutures in double-layer suturing can shorten the anastomotic time in TLC (28, 29). Although barbed sutures have not been frequently used by our team hitherto in TLC, it will be considered to utilize for further reducing the anastomotic time and post-complications of IA, which is significant for the development of totally laparoscopic surgery.

In addition to intraoperative-related factors, the occurrence of postoperative complications is often discussed when comparing the anastomotic approaches, and the reports are conflicting in the previous systematic reviews. Our research found that the incidences of postoperative complications of IA and EA were equivalent, and there was no difference between them when stratifying postoperative complications according to Clavien-Dindo classification in detail.

Theoretically it is redundant to drag out the bowel from coelom for stapling during IA, hence fewer bowel operations are required for IA and the incisions are less likely to be contaminated (30). However, no significant difference was observed in the postoperative ileus and surgical site infection (SSI) between the two anastomotic techniques, which may be induced by the low morbidity seemly due to limited sample size. Anastomotic leakage is one of the most concerned postoperative complications. It has previously been shown that exteriorization of the bowel, hand-sewn anastomosis and compromised vascular supply by lengthening of the mesentery bring about a higher incidence of anastomotic leaks in EA (17, 31). Our team was used to performing manual suture after stapled anastomosis regardless of whether IA or EA, which may be the reason for the low incidence rate and statistical similarity of anastomotic leakage and anastomotic bleeding between the two anastomotic methods in our study. It was worth noting that half of the complications in the IA were abdominal infections, which was significantly different from EA. Although many scholars indeed speculated that opening the bowel during IA increased the hazard of abdominal infection, their findings did not definitely support it (2, 17). We had found several years ago that the postoperative abdominal infection rate in IA was higher than EA, so we had tried a variety of methods to prevent it, including adequate preoperative bowel preparation, sufficient intraoperative peritoneal lavage and improving the postoperative nutrition and immunity. Further research revealed that the abdominal infection rate after IA had been descending in recent years, which may be related to the preventive measure or IA technical strides of our surgeons. We infer that with the comfort of our surgeons for IA technique, the abdominal infection after IA will further decline in the next few years (6); and as the incidence of postoperative complications such as abdominal infection decreases, the superiority of IA in postoperative recovery may be gradually revealed, including shortening the length of hospitalization and reducing the inflammatory response, which is relevant for the realization of “enhanced recovery protocol (ERAS)”.

In our opinion, IA is a safe and reliable technology and even has an advantage over EA in terms of anastomotic time, incision length, during surgery, intraoperative blood loss, postoperative pain and bowel function recovery. Along with the advance of IA technique, smaller hazard of postoperative complications will be obtained and the potential of IA in ERAS will be gradually realized. In addition, due to the limited patient number in this study, more cases will be required to verify these views in the future.

The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding authors.

YHZ, YCZ, CW and RY analyzed the data and wrote the manuscript. XL, ST and WC searched and collected data. YM and CY: modify the draft. SL, WL: designed the study. All authors contributed to the article and approved the submitted version.