Short-term outcomes following intracorporeal vs. extracorporeal anastomosis after laparoscopic right and left-sided colectomy: a propensity score-matched study

Abstract

Background:

To compare the short-term outcomes of patients undergoing intracorporeal anastomosis (IA) during laparoscopic colectomy to those undergoing extracorporeal anastomosis (EA).

Methods and materials:

The study was a single-centre retrospective propensity score-matched analysis conducted. Consecutive patients who underwent elective laparoscopic colectomy without the double stapling technique between January 2018 and June 2021 were investigated. The main outcome was overall postoperative complications within 30 days after the procedure. The authors also performed a sub-analysis of the postoperative results of ileocolic anastomosis and colocolic anastomosis, respectively.

Results:

A total of 283 patients were initially extracted; after propensity score matching, there were 113 patients in each of the IA and EA groups. There were no differences in patient characteristics between the two groups. The IA group had a significantly longer operative time than the EA group (208 vs. 183 min, P=0.001). The rate of overall postoperative complications was significantly lower in the IA group (n=18, 15.9%) than in the EA group (n=34, 30.1%; P=0.02), especially in colocolic anastomosis after left-sided colectomy (IA: 23.8% vs. EA: 59.1%; P=0.03). Postoperative inflammatory marker levels were significantly higher in the IA group on postoperative day 1 but not on postoperative day 7. There was no difference in the postoperative lengths of hospital stay between the two groups, and no deaths occurred.

Conclusion:

The data suggest that performing IA during laparoscopic colectomy can potentially reduce the risk of postoperative complications, especially in colocolic anastomosis after left-sided colectomy.

Introduction

Highlights

• Laparoscopic colectomy is preferred over laparotomy due to favourable short-term postoperative outcomes

• We determined the short-term outcomes of patients who undergo intracorporeal anastomosis during laparoscopic colectomy

• Data were compared with those undergoing extracorporeal anastomosis)

• Intracorporeal anastomosis during laparoscopic colectomy can reduce risk of postoperative complications in colocolic anastomosis after left-sided colectomy.

Laparoscopic colectomy is a widely performed procedure owing to its more favourable short-term postoperative outcomes compared with laparotomy^1–5. Anastomosis after laparoscopic colectomy can be performed either intracorporeally or extracorporeally. Extracorporeal anastomosis (EA) is the standard anastomosis after right-sided or splenic flexure colectomy. EA is a procedure that requires sufficient bowel extraction from the abdominal cavity, which has the potential to damage the intestinal tract and mesentery^6,7. Furthermore, EA is more difficult to perform in obese patients with thick abdominal walls^8–12. On the other hand, some advantages and challenges with respect to intracorporeal anastomosis (IA) have also been reported; for example, the advantages are that it requires less intestinal mobilization, has a lower risk of adhesion formation, and involves a smaller site of extraction¹³, while the challenges are technical difficulty and longer operative time^14,15.

IA has been gaining attention in recent years owing to advances in the surgical techniques and devices associated with it. Moreover, this method has been reported to provide many favourable short-term results, particularly among patients who undergo ileocolic anastomosis after right-sided colectomy^16–24, although IA is rarely used during colocolic anastomosis after colectomy for left-sided or splenic flexure tumours^25–29. Most of the reports on IA to date have focused on results after right-sided colectomy, and there have been few analyses, including left-sided colectomy or sub-analyses by sidedness.

The purpose of this study was to investigate the short-term outcomes of IA in patients who underwent laparoscopic right-sided and left-sided colectomy and thereby assess its efficacy compared to EA. We also performed a sub-analysis of the postoperative results of ileocolic anastomosis and colocolic anastomosis.

Materials and methods

Patient selection

This study was performed in accordance with the principles outlined in the Declaration of Helsinki. We applied an opt-out method to obtain consent for this study, which was approved by the Institutional Review Board. This study was registered with the UMIN Clinical Trials Registry (UMIN000050592). All data were obtained from electronic medical records and our prospectively populated database. Consecutive patients who underwent elective laparoscopic colectomy for benign or malignant tumours located from the appendix to the descending colon at the splenic flexure between January 2018 and June 2021 were investigated retrospectively. In our hospital, IA was first introduced in January 2019; thereafter, EA was performed only at the surgeon’s discretion. Following a transition period in the year 2019, IA was implemented as the standard from 2020. Patients who underwent simultaneous resection of other organs and those with no anastomosis were excluded. Sigmoid colectomy and rectal resections requiring IA with the double stapling technique were also excluded because it is difficult to perform EA and comparative analysis is not possible. This work has been reported in line with the STROCSS criteria³⁰.

Surgical technique

All patients underwent routine preoperative mechanical bowel preparation with magnesium citrate and sennoside and no chemical preparation. All surgeons performed both IA and EA procedures. A pneumoperitoneum was created by placing a 12 mm trocar at the umbilicus via a mini-laparotomy. Another 12 mm trocar and three 5 mm trocars were also inserted for a total of five ports. If the tumour was on the right colon, a 12 mm trocar was placed at the left upper abdomen to enable the insertion of linear staplers; if the tumour was on the left colon, the trocar was placed at the right lower abdomen. Complete mesocolic excision and central vascular ligation were performed to remove malignant tumours.

When performing EA, the umbilical wound was extended, and a small laparotomy was performed. While protecting the incision with a wound protector, the section of the colon containing the tumour was removed from the small laparotomy wound. The line of dissection was determined, and the mesentery was dissected. The intestines on the oral and anal sides of the tumour were transected with linear staplers, and the specimen was removed.

When performing IA, after dissection of the mesentery and determining the intestinal transection line, the intestine on the oral and anal sides of the tumour were transected with linear staplers intracorporeally; the specimen was then placed in a bag. Anastomosis was performed in all cases using the isoperistaltic side-to-side stapler method in which an enterotomy was created on the anti-mesenteric side 2 cm from the oral stump and 8 cm from the anal stump. The jaws of the linear stapler were inserted through the holes and anastomosis was performed. The common channel of the enterotomy was closed with two layers of 3-0 barbed sutures. After anastomosis, the specimens were removed through either a small median umbilical wound or Pfannenstiel incision per the surgeon’s preference.

Data collection

We collected preoperative variables, including age, sex, BMI, American Society of Anesthesiologists (ASA) score, previous abdominal surgery, tumour location, and tumour stage. Furthermore, we assessed intraoperative data, including operative time, resection time, anastomotic time, estimated blood loss, intraoperative complications, type of surgery, type of anastomotic intestine, type of anastomosis, and site of skin incision. Finally, we extracted postoperative data, including postoperative complications and their severity (using the Clavien–Dindo classification), length of hospital stay, and mortality. Preoperative and postoperative laboratory values, including white blood cell (WBC) counts and C-reactive protein (CRP) levels were also analyzed. Blood tests were performed preoperatively and on the first, third, and seventh postoperative days. Resection time was defined as the interval between the start of mesenteric dissection after mobilization and transection of the intestine in the IA group, and as that between the extraction of the intestine (including the tumour) and removal of the specimen in the EA group. Anastomotic time was defined as the interval between the creation of the enterotomy at the intestine and completion of common enterotomy closure in the IA group and that between the removal of the specimen and completion of anastomosis in the EA group. Resection and anastomotic times were determined by reviewing videos of the procedure in patients who underwent IA and were obtained from the surgical records of those who underwent EA.

Statistical analysis

Categorical variables were compared using Pearson’s χ² test or Fisher’s exact test as appropriate. Continuous variables were compared using the Mann–Whitney U test. We performed propensity score matching (PSM) because of the imbalance in the patient characteristics of IA and EA groups. A propensity score for each patient was calculated from a logistic regression model with the method of anastomosis (IA vs. EA) set as the dependent variable, which was used to match each patient. Patients were matched 1:1 using the neighbour matching method; covariates included BMI, ASA score, and previous abdominal surgery. After PSM, we confirmed that there was no imbalance in the characteristics of the two groups and analyzed the cohort³¹. A P value less than 0.05 was considered significant. Statistical analyses were performed using R version 2.8.1 (2008-12-22; The R Foundation for Statistical Computing).

Results

Of the 325 eligible patients, 283 were analyzed after the exclusion of 42 in this study (Figure 1). The patients’ characteristics, before and after PSM, are shown in Tables 1 and 2. Of the 283 patients analyzed, 131 underwent IA and 152 underwent EA. The median BMI of the IA group (23.5 kg/m²) was significantly higher than that of the EA group (22.3 kg/m²; P=0.01). After PSM, 113 patients were included in each group; there were no significant differences in their characteristics.

Figure 1.

Flowchart showing the patient selection process.

Table 1.

Patient characteristics before propensity score matching.

	IA (n=131)	EA (n=152)	P
Age, yearsa	71 (22–90)	72 (31–90)	0.18
Male, n (%)	70 (53.4)	76 (50.0)	0.63
BMI, kg/m2 a	23.5 (16.2–31.5)	22.3 (14.8–33.9)	0.01
ASA score, n (%)			0.47
1	19 (14.5)	28 (18.4)
2	107 (81.7)	115 (75.7)
3	5 (3.8)	9 (5.9)
Previous abdominal surgery, n (%)	41 (31.3)	53 (34.9)	0.53
Tumour site, n (%)			0.50
Appendix	1 (0.8)	5 (3.3)
Caecum	28 (21.4)	36 (23.7)
Ascending colon	58 (44.3)	62 (40.8)
Transverse colon	32 (24.4)	40 (26.3)
Descending colon	12 (9.2)	9 (5.9)
Preoperative UICC Stage, 8th ed, n (%)			0.30
In situ/benign	5 (3.8)	7 (4.6)
I	62 (47.3)	66 (43.4)
II	36 (27.5)	39 (25.7)
III	23 (17.6)	24 (15.8)
IV	5 (3.8)	16 (10.5)

^aMedian (range).

ASA, American Society of Anesthesiologists; EA, extracorporeal anastomosis; IA, intracorporeal anastomosis; UICC, Union for International Cancer Control.

Table 2.

Patient characteristics after propensity score matching.

	IA (n=113)	EA (n=113)	P
Age, yearsa	71 (22–89)	72 (31–90)	0.26
Male, n (%)	60 (53.1)	57 (50.4)	0.63
BMI, kg/m2 a	23.1 (16.2–31.5)	22.6 (16.2–33.7)	0.81
ASA score, n (%)			0.13
1	17 (15.0)	19 (16.8)
2	94 (83.2)	86 (76.1)
3	2 (1.8)	8 (7.1)
Previous abdominal surgery, n (%)	36 (31.9)	39 (34.5)	0.78
Tumour site, n (%)			0.91
Appendix	1 (0.9)	2 (1.8)
Caecum	25 (22.1)	29 (25.7)
Ascending colon	49 (43.4)	46 (40.7)
Transverse colon	29 (25.7)	29 (25.7)
Descending colon	9 (8.0)	7 (6.2)
Preoperative UICC Stage, 8th ed, n (%)			0.32
in situ/benign	3 (2.7)	6 (5.3)
I	55 (48.7)	46 (40.7)
II	30 (26.5)	30 (26.5)
III	20 (17.8)	19 (16.8)
IV	5 (4.4)	12 (10.6)

^aMedian (range).

ASA, American Society of Anesthesiologists; EA, extracorporeal anastomosis; IA, intracorporeal anastomosis; UICC, Union for International Cancer Control.

The patients’ intraoperative outcomes are shown in Table 3. Patients in the IA group had a significantly longer operative time than did those in the EA group (208 vs. 183 min, P=0.001), particularly resection time (24 vs. 17 min, P<0.001) and anastomosis time (32 vs. 17 min, P<0.001); patients in the IA group also had significantly less blood loss (9 vs. 34 mL, P<0.001). There were no significant differences in the types of surgery undergone by patients in the two groups; ileocolic anastomosis for right-sided colectomy was performed in 81.4% and 80.5% of the patients in the IA and EA groups, respectively, while colocolic anastomosis for left-sided colectomy was performed in 18.6% and 19.5% of the patients, respectively.

Table 3.

Intraoperative outcomes.

	IA (n=113)	EA (n=113)	P
Operative time, mina	208 (112–369)	183 (80–379)	0.001
Resection time, mina	24 (7–68)	17 (5–52)	<0.001
Anastomotic time, mina	32 (16–80)	17 (2–43)	<0.001
Estimated blood loss, mLa	9 (0–318)	34 (0–273)	<0.001
Intraoperative complication, n (%)	1 (100)	0	1
Type of surgery, n (%)			0.30
Ileocecal resection	57 (50.4)	67 (59.3)
Right hemicolectomy	32 (28.3)	19 (16.8)
Extended right hemicolectomy	3 (2.7)	5 (4.4)
Transverse colectomy	13 (11.5)	15 (13.3)
Descending colectomy	8 (7.1)	7 (6.2)
Type of anastomotic intestine, n (%)			1
Ileocolic	92 (81.4)	91 (80.5)
Colocolic	21 (18.6)	22 (19.5)
Type of anastomosis, n (%)			<0.001
Antiperistaltic side-to-side	0	105 (92.9)
Isoperistaltic side-to-side	113 (100)	8 (7.1)
Site of skin incision, n (%)			<0.001
Umbilical midline	75 (66.4)	113 (100)
Pfannenstiel	38 (33.6)	0

^aMedian (range).

EA, extracorporeal anastomosis; IA, intracorporeal anastomosis.

Postoperative outcomes are shown in Table 4. The rate of overall postoperative complications was significantly lower in the IA group than in the EA group (15.9% vs. 30.1%, P=0.02). There were no differences in the rates of individual or severe complications (grade ≥3 per the Clavien–Dindo classification). There was also no difference in the lengths of postoperative hospital stays between the two groups, and none of the patients died. Postoperative inflammatory markers, including WBC counts and CRP levels were significantly higher in the IA group than in the EA group on postoperative day 1, as was the latter on postoperative day 3; however, there were no differences by postoperative day 7 (Figures. 2A and B).

Table 4.

Postoperative outcomes by the type of anastomotic intestine.

	Overall			IC			CC
	IA (n=113)	EA (n=113)	P value	IA (n=92)	EA (n=91)	P value	IA (n=21)	EA (n=22)	P
Postoperative complications, n (%)
Overall	18 (15.9)	34 (30.1)	0.02	13 (14.1)	21 (23.1)	0.13	5 (23.8)	13 (59.1)	0.03
Wound infection	8 (7.1)	10 (8.8)	1.00	7 (7.6)	9 (9.9)	0.61	1 (4.8)	1 (4.5)	1.00
Intra-abdominal abscess	1 (0.9)	3 (2.7)	0.62	0	2 (2.2)	0.25	1 (4.8)	1 (4.5)	1.00
Anastomotic leakage	1 (0.9)	4 (3.5)	0.37	1 (1.1)	2 (2.2)	0.62	0	2 (9.1)	0.49
Ileus	3 (2.7)	6 (5.3)	0.50	3 (3.3)	4 (4.4)	0.72	0	2 (9.1)	0.49
Bowel obstruction	1 (0.9)	3 (2.7)	0.62	0	0	—	1 (4.8)	3 (13.6)	0.61
Anastomotic bleeding	1 (0.9)	0	1.00	0	0	—	1 (4.8)	0	1.00
Urinary tract infection	1 (0.9)	2 (1.8)	1.00	1 (1.1)	1 (1.1)	1.00	0	1 (4.5)	1.00
Chylous ascites	0	2 (1.8)	1.00	0	1 (1.1)	1.00	0	1 (4.5)	1.00
Others	2 (1.8)	4 (3.5)	0.22	1 (1.1)	2 (2.2)	0.62	1 (4.8)	2 (9.1)	1.00
Clavien–Dindo classification, n (%)
Grade 1	8 (7.1)	12 (10.6)	0.19	6 (6.5)	9 (9.9)	0.43	2 (9.5)	3 (13.6)	1.00
Grade 2	9 (8.0)	16 (14.2)	0.20	7 (7.6)	10 (11.0)	0.46	2 (9.5)	6 (27.3)	0.24
Grade 3	1 (0.9)	6 (5.3)	0.12	0	2 (2.2)	0.25	1 (4.8)	4 (18.2)	0.35
Grade 4	0	0	—	0	0	—	0	0	—
Grade 5	0	0	—	0	0	—	0	0	—
Length of hospital stay, daysa	8 (5–23)	8 (5–52)	0.46	8 (5–19)	8 (6–33)	0.37	9 (7–23)	10 (5–52)	0.14
Mortality, n (%)	0	0	—	0	0	—	0	0	—

^aMedian (range).

CC, colocolic anastomosis; EA, extracorporeal anastomosis; IA, intracorporeal anastomosis; IC, ileocolic anastomosis.

Figure 2.

Perioperative changes in WBC count (A) and CRP (B). CRP, C-reactive protein; EA, extracorporeal anastomosis; IA, intracorporeal anastomosis; POD, postoperative day; Pre, preoperative; WBC, white blood cell.

The postoperative outcomes of patients who underwent IA and EA, and sub-analysis based on sidedness are shown in Table 4. Among patients who underwent ileocolic anastomosis for right-sided colectomy, there was no significant difference in the incidence of postoperative complications between the IA and EA groups; however, among patients who underwent colocolic anastomosis for left-sided colectomy, the incidence of postoperative complications was significantly lower in the IA group than in the EA group (23.8% vs. 59.1%, P=0.03).

Discussion

Our study demonstrated that, although IA required a longer operative time than EA, especially in terms of resection and anastomosis, IA resulted in less blood loss and fewer postoperative complications. Moreover, patients in the colocolic anastomosis group who underwent IA experienced significantly fewer complications than those who underwent EA. These results indicate that IA during laparoscopic colectomy is considered a safer and more promising anastomosis method than EA, especially for patients undergoing left-sided colectomies.

There have been multiple retrospective studies on the outcomes of patients who underwent IA during right-sided colectomy. Many of these have demonstrated its advantages in terms of recovery of postoperative bowel movement, shortening the length of the incision, and reducing complications^18,20,24. The overall complication rates associated with IA and EA, as reported in major randomized controlled trials, were 17–30% and 13–70%, respectively^{16,17,19,21,22}; it remained unclear whether IA contributes to a lower overall complication rate. Moreover, there have been only a few retrospective studies of patients who underwent left-sided colectomy; these showed overall complication rates of 0–18% and 9–28% for IA and EA, respectively^25–29. Two of these were propensity score-matched analyses that found IA to be associated with potentially fewer complications^27,28. In the present study, the complication rates in all cohorts, including right-sided and left-sided colectomy, were significantly lower in IA than in EA (15.9% and 30.1%, respectively), which was comparable to those previously reported. In addition, the complication rate after colocolic anastomosis during left-sided colectomy was significantly lower in the IA group than in the EA group.

Although EA has been the predominant anastomosis method after colectomy, there are several drawbacks associated with it¹³. The first is that it requires sufficient bowel mobilization to allow extraction through the small laparotomy wound; moreover, it is very difficult to determine the extent of mobilization that is necessary. Insufficient mobilization leads to difficulty in completing the anastomosis, and may cause complications such as leakage and stenosis³². In the case of ileocolic anastomosis, the mesentery of the ileum is long and can easily be mobilized. However, when performing colocolic anastomosis, it is difficult to extract the colon from the abdominal cavity, particularly the transverse colon with its short mesentery. Therefore, the risk of anastomotic complications is speculated to be high. Second, it is difficult to determine whether the mesentery is twisted during EA, whereas any twisting can easily be confirmed under laparoscopic view when performing IA³³. These drawbacks can explain the high complication rate associated with colocolic anastomosis in our EA group, especially anastomotic leakage, ileus, and bowel obstruction.

Regarding postoperative inflammatory response, Mari et al. ²¹ reported a significantly lower inflammatory response in IA compared with EA. On the other hand, our results showed that WBC and CRP levels on postoperative day 1 and CRP levels on postoperative day 3 were significantly higher in the IA group than in the EA group. However, postoperative complication rates such as wound infection, intra-abdominal abscess, and anastomotic leakage were comparable in both groups. This suggests that IA, in which the intestinal tract is opened intracorporeally, does not increase surgical site infection and other complications. Liao et al. ²⁰ also reported a similar association between inflammatory response and postoperative complications after IA. Although there are few studies of long-term outcomes after IA, there is no evidence to suggest that IA is oncologically unfavourable or that it can cause peritoneal dissemination^20,34–36. As such, the long-term prognoses of patients such as those in our study remain to be clarified.

There were several limitations in our study. First, it is a single-centre retrospective investigation; although PSM was performed to adjust for background factors in both groups, selection bias cannot be completely ruled out. One such example is that the decision to choose between EA and IA was left to the surgeon, creating some selection bias. Randomized controlled trials would be desirable to further clarify the outcomes of IA during laparoscopic colectomy. Second, the number of patients was small; a larger cohort may have increased the statistical power of our data. Third, a feature of our cohort was that all anastomoses in the IA group were isoperistaltic, whereas most of those in the EA group were antiperistaltic; this may have affected the postoperative outcomes. Some investigators have reported no difference in complication rates between the two anastomosis methods^32,37. Therefore, the imbalance between the isoperistaltic and antiperistaltic subgroups in our study is unlikely to be a major confounding factor. Fourth, no data on incision length were available in this study. In addition, the incidence of incisional hernia was not available due to the short follow-up period of the IA group.

Conclusions

Our data showed that IA during laparoscopic colectomy reduces postoperative complication rates in all cohorts, including right-sided and left-sided colectomy. This advantage appears to be prominently true for colocolic anastomoses. The long-term prognosis of the patients is not yet clear, and it is necessary to compare both the short-term and long-term outcomes of subjects undergoing both types of anastomoses in a large prospective study.

Ethical approval

The study protocol was approved by the Institutional Review Board of the National Cancer Center (2018-100). We utilized the Opt-out method to obtain consent in this study, which was approved by the Institutional Review Board.

Source of funding

There are no sources of funding for this manuscript.

Author contribution

K.T. contributed to drafting the manuscript. H.M. contributed to data acquisition. D.K. and M.I. were involved in the drafting and revision of the manuscript. All authors read and approved the final manuscript for submission.

Conflicts of interest disclosure

The authors declare that they have no competing interests for this manuscript.

Research registration unique identifying number (UIN)

1. Name of the registry: University Hospital Medical Information Network.

2. Unique Identifying number or registration ID: UMIN000050592.

3. Hyperlink to your specific registration (must be publicly accessible and will be checked): https://center6.umin.ac.jp/cgi-openbin/ctr/ctr_view.cgi?recptno=R000057620.

Guarantor

Masaaki Ito.

Data availability

The data that support the findings of this study are available from the corresponding author (M.I.) upon reasonable request.

Provenance and peer review

Not commissioned, externally peer-reviewed.