We conducted a retrospective cohort study at King AbdulAziz University Hospital (KAUH), a tertiary center in Jeddah, Saudi Arabia. Data of 134 adult patients who underwent colectomy for colorectal cancer from January 2016 to June 2019 were collected from the electronic medical database and reviewed. Patients were divided into two groups, laparoscopy, and laparotomy approach. A total of 14 patients were excluded due to their conversion from laparoscopy to open surgery. The final population size was 120.

This study was approved by the Institutional Review Board of KAUH in 2019 (ref. 145-19). The need for obtaining patient consent was waived by the IRB.

All adult patients aged > 18 years that were diagnosed with non-metastatic colon or rectal cancer and who had undergone operation were extracted from the hospital’s data base and included. Key search terms were laparoscopic/open right or left colectomy, laparoscopic/open low anterior resection, laparoscopic/open total colectomy, laparoscopic/open subtotal colectomy, laparoscopic/open abdominoperineal resection and laparoscopic/open total coloproctectomy. Patients were classified into two groups according to the surgical approach used: either laparoscopic colectomy; or open colectomy. Patients who had laparoscopic conversion to open, i.e. surgeries which started with laparoscopy that ended up being laparotomy were excluded. In addition, patients with disseminated cancer, concurrent procedure, or sepsis (in the form of peritonitis) were excluded from the study to avoid confounders. Patients with incomplete records were also excluded from the analysis of that particular entity.

The data collected included demographic information (age, sex, Body Mass Index, comorbidities) as well as preoperative variables including diagnosis, neoadjuvant therapy, comorbidities, and previous abdominal surgeries. Intraoperative variables were also documented, including the American Society of Anesthesiologists (ASA) score, use of epidural anesthesia, colorectal procedure, approach (laparoscopy or open), stoma formation, estimated blood loss, blood transfusion, and procedure duration. All patients received a clear liquid diet 24 h after the operation and subsequently progressed to a regular diet. After the regular diet was tolerated and bowel movements returned to normal, the patient was discharged.

Postoperative variables included the type of analgesia, vomiting, abdominal distention, time of mobilization, time until clear and regular diets were tolerated, and the day of the first bowel movement. The length of hospital stay, from the time of operation until patient discharge, was also recorded. Readmission within 30 days and postoperative complications were assessed based on the Clavien–Dindo score. ⁷ POI was defined as the absence of first flatus and intolerance of an oral intake within three days, postoperatively, and PPOI was defined as the delayed passage of flatus with intolerance of an oral intake for more than three days, postoperative. ⁸

Statistical analysis was performed using IBM SPSS Statistics Version 21.0.Mean ± standard deviation was used to describe the numerical data. Patient treatment and outcome related variables where represented as percentages and frequencies of the total population. Comparisons and the relationships between the groups were made using independent t-test, chi-square test, and bivariate correlation. The results were considered significant where the two-tailed P value ≤ 0.05.

A total of 120 patients that were diagnosed with colorectal cancer who underwent different colonic procedures were included in this study (see Underlying data ²³ ). The patients’ baseline characteristics are presented in Table 1. Patients were divided into two groups: the laparoscopic approach group, 55 patients (45.8%); and the open approach group, 65 patients (54.2%). The age of the patients ranged from 22 years to 90 years, with a mean (SD) age of 61.36 ± 13.85 years. The patients were predominantly male (60%), and most of the cohort had colon cancer (67.5%). Over half of the patients were either overweight or obese (65%), and a third had diabetes mellitus (33.3%). Most patients were of ASA class 2 or 3 (92.5%). Table 1.

Variable	No. (%)
Age (years) (mean ± SD)	61.36 ± 13.85
Operative approach
Laparoscopy	55 (45.8)
Open	65 (54.2)
Gender
Male	72 (60.0)
Female	48 (40.0)
Smoking
Current smoker	9 (5.7)
Non-smoker	93 (77.5)
Ex-smoker	13 (10.8)
Unknown	5 (4.2)
BMI categories
Underweight	5 (4.2)
Normal weight	37 (30.8)
Overweight	47 (39.2)
Obese	31 (25.8)
Mean BMI (mean ± SD) kg/m 2	27.21 ± 5.5
Comorbidity
DM	40 (33.3)
HTN	45 (37.5)
Hyperlipidemia	16 (13.3)
Heart disease	9 (7.5)
CLD	1 (0.8)
COPD	1 (0.8)
CKD	2 (1.7)
Thyroid disorder	8 (6.7)
Asthma	1 (0.8)
CVA	5 (4.2)
Previous abdominal surgery
Colorectal surgeries	8 (6.7)
Cholecystectomy	6 (5.0)
Appendectomy	5 (5.0)
Hernia repair	6 (4.2)
OBS\GY	5 (5.0)
None	89 (74.2)
Diagnosis
Colon cancer	81 (67.5)
Rectal cancer	21 (17.5)
Colorectal cancer	18 (15.0)

BMI, body mass index; SD, standard deviation; DM, diabetes mellitus; HTN, hypertension; CLD, chronic liver disease; COPD, chronic obstructive pulmonary disease; CKD, chronic kidney disease; CVA, cerebrovascular accident; OBS/GY, obstetrics and gynecology.

Regarding operative variables, the majority were elective procedures (77.5%) and most had anastomoses (74.2%) ( Table 2). Most patients did not require blood transfusion, and among those who did, the average number of units transfused was 1.75 (SD, 0.46). Table 2.

Variable	No. (%)
Surgery timing
Elective	93 (77.5)
Emergency	27 (22.5)
Type of colonic surgery
Total colectomy	1 (0.8)
Right hemi-colectomy	26 (21.7)
Left hemi-colectomy	42 (35.0)
Subtotal colectomy	5 (4.2)
Abdominoperineal resection	9 (7.5)
Low anterior resection	34 (28.3)
Total proctocolectomy	3 (2.5)
Anastomoses
Yes	89 (74.2)
No	31 (25.8)
Stoma formation
Yes	58 (48.3)
No	62 (51.7)
Type of stoma
End colostomy	21 (17.5)
End ileostomy	10 (8.3)
Diverting colostomy	3 (2.5)
Diverting ileostomy	24 (20.0)
None	62 (51.7)
Epidural anesthesia
Yes	98 (81.7)
No	22 (18.3)
ASA score
I	6 (5.0)
II	63 (52.5)
III	48 (40.0)
IV	2 (1.7)
V	1 (0.8)
Required blood transfusion
Yes	27 (22.5)
No	93 (77.5)
Mean number of required units (mean ± SD)	0.46 ± 1.75

ASA, American Society of Anesthesiologists; SD, standard deviation.

Table 3 presents the postoperative outcomes including postoperative complications, POI, and PPOI in both the laparoscopic and the open approaches. The incidence of POI and PPOI in the study group was 4.2% and 15%, respectively. There was a higher incidence of POI in the laparoscopic approach (7.2% vs. 1.5%, P = 0.03); however, the incidence of PPOI was higher in the open approach (20% vs. 9.1%, P = 0.03). The mean time of first bowel movement in the laparoscopic and open approaches were 3.13 ± 2.13 vs. 3.94 ± 2.65 days; P = 0.09). Patients with previous abdominal surgeries showed no difference in PPOI compared to those with no previous surgeries ( P = 0.33). The operative duration was greater in the laparoscopy approach group (403.16 ± 133.73 vs. 277.34 ± 123.39 min; P < 0.001) along with less bleeding compared to the open approach group (139.09 ± 145.83 vs. 343.85 ± 307.78 mL; P < 0.001). Regular diet was tolerated earlier in the laparoscopy approach group than in the open approach group (5.44 ± 2.7 vs. 8.94 ± 6.64 days; P < 0.001). Mobilization was also observed earlier in the laparoscopic approach (3.12 ± 1.77 vs. 5.39 ± 3.48 days; P < 0.001). Table 3.

Variable	Laparoscopic (n = 55)	Open (n = 65)	P value
Age (years)	59.38 ± 14.35	63.03 ± 13.29	-
BMI	27.58 ± 5.14	26.82 ± 5.70	0.57
ASA score
I	3 (5.5%)	3 (4.6%)	0.35
II	33 (60.0%)	30 (46.2%)
III	19 (34.5%)	29 (44.6%)
IV	0	2 (3.1%)
V	0	1 (1.5%)
Procedure time (min)	403.16 ± 133.73	277.34 ± 123.39	<0.001
Estimated blood loss (ml)	139.09 ± 145.83	343.85 ± 307.77	<0.001
Stoma formation
Yes	28 (48.3%)	30 (51.7%)	0.60
No	27 (43.5%)	35 (56.5%)
Passing of first flatus (postoperative day)	3.13 ± 2.13	3.94 ± 2.65	0.09
Tolerance of clear liquid (days)	2.38 ± 1.52	3.77 ± 2.76	0.001
Tolerance of regular diet (days)	5.44 ± 2.70	8.94 ± 6.65	<0.001
Mobilization (days) b	3.12 ± 1.70	5.39 ± 3.48	<0.001
Length of hospital stay (days)	8.24 ± 4.29	14.52 ± 10.47	<0.001
POI	4 (7.2%)	1 (1.5%)	0.03
Prolonged POI	5 (9.1%)	13 (20.0%)	0.03
Post-OP complications a
SSI	3 (12.5%)	21 (87.5%)	<0.001
Anastomosis leak	2 (28.6%)	5 (71.4%)	0.34
Pneumonia	0 (0.0%)	2 (100%)	0.19
UTI	1 (20.0%)	4 (80.0%)	0.23
DVT	0 (0.0%)	1 (100%)	0.35
Atelectasis	1 (25.0%)	3 (75.0%)	0.39
Dindo classifications
No complications	44 (61.1%)	28 (38.9%)	0.001
Grade I	3 (37.5%)	5 (62.5%)
Grade II	2 (12.5%)	14 (87.5%)
Grade III	1 (11.1%)	8 (88.9%)
Grade IV	5 (38.5%)	8 (61.5%)
Grade VI	0 (0.0%)	2 (100%)

^a

Some patients had more than one complication.

BMI, body mass index; ASA, American Society of Anesthesiologists; POI, postoperative ileus; SSI, surgical site infection; UTI, urinary tract infection; DVT, deep vein thrombosis; PPOI, prolonged postoperative ileus.

Regarding the incidence of postoperative complications, 59% of patients had no complications, while 41% had varying degrees of complications. The open approach showed a higher rate of complications—37 patients (56.9%), compared to 11 patients (20%) in the laparoscopic group. According to the Clavien–Dindo score, this difference was statistically significant ( Table 3).

Furthermore, the LOS was longer in the open approach compared to the laparoscopic approach (14.52 ± 10.47 vs. 8.24 ± 4.29 days; P < 0.001). Patients with anastomotic leak had a longer LOS (23.86 ± 9.5 vs. 10.85 ± 8.17 days; P < 0.001). General and epidural anesthesia was used for most patients (98 [81.7%]), while the remaining patients received general anesthesia only (22 [18.3%]). No significant difference was found between patients with POI and PPOI in the use of epidural or general anesthesia ( P = 0.24).

Addressing the incidence of POI and PPOI is essential for determining the risk factors associated with these complications and can therefore help in decreasing their occurrence. In our study, the incidences of POI and PPOI following colorectal surgery were 4.2% and 15%, respectively, while the literature demonstrates that the incidence of POI is 10-30% and the incidence of PPOI is 2-54%. ^{4, 5, 9}

Previous studies have demonstrated that the risk factors for POI and PPOI include more invasive approaches, operative difficulty, older age, perioperative blood loss, and delayed mobilization. ^{10, 11} Our analysis showed that the open approach was associated with a significantly higher incidence of PPOI than the laparoscopic approach. In the literature, the open approach was identified as an independent risk factor for the development of an ileus. ⁹ At the same time, most of the patients with POI, which is a less severe form of ileus, were in the laparoscopic group. These results are more likely explained by the inflammatory and neurological responses to surgical trauma and increased requirements for narcotics, which are more pronounced in the open approach group. ^{12, 13} It has been suggested that adherence to protocols of enhanced recovery after surgery could potentially decrease the incidence of POI. These protocols that recommend minimally invasive procedures, early ambulation, and early feeding. ¹⁴

Previous studies, ^{2, 10, 14} including a cross-sectional study conducted in 2013 in Isfahan, Iran, ⁶ have found that the occurrence of ileus is directly associated with the duration of the surgical procedure. However, the Iranian study solely focused on the duration of the operation, rather than the type of surgical approach used. ⁶ In our analysis, the duration of laparoscopic colectomy was significantly longer than that of the open procedure; however, this did not lead to an increase in the incidence of PPOI. Another study that observed 295 patients who underwent laparoscopic colectomy for colorectal cancer, also found that the duration of the operation has little impact on POI and PPOI. ¹⁵ Some studies speculate that the longer duration of the laparoscopic approach is due to the limited range of motion and flexibility of instruments, in addition to the two-dimensional view and the dependence on surgeon skill. ¹⁶ However, we do not agree with that statement in view of ongoing innovations, and the early exposure of trainees to laparoscopic surgery. There is the potential risk of bias, however, as it may be the case that more advanced disease that requires open surgery and multi-organ resection could contribute to prolonged operative time, POI, and PPOI.

In terms of intraoperative complications, studies have shown that the higher the estimated blood loss, the higher the risk of developing POI. This is explained by the stimulation of the sympathetic stress response and its inhibition of bowel motility. ¹⁷ Many studies state that the open approach results in more blood loss and blood transfusions than the laparoscopic approach. ^{18, 19} Our results have also shown a significant difference in blood loss between open surgery and laparoscopy (343.85 vs. 152.73 mL). The significant increase in bleeding in the open approach may have occurred due to the larger surgical incision. However, this could also be influenced by the fact that more advanced disease that requires major resection would be associated with more blood loss and a higher rate of transfusion.

Regarding the postoperative hospital course, our results have shown a significant earlier tolerance of regular diet in the laparoscopy group compared to the open group (5.44 vs. 8.94 days). This may be due to less trauma during the operation and the release of less amounts of stress hormones, allowing for early bowel recovery. ¹⁸ Patients who underwent laparoscopy showed earlier ambulation (3.12 vs. 5.39 days) because of the small surgical incision and less pain. Early mobilization in the laparoscopy group also helped to stimulate bowel function. Similar results on early regular diet tolerance and early mobilization in the laparoscopy group were observed in other studies. ¹²

The laparoscopic approach was associated with a significantly lower risk of overall complications and this supports our point of view. ¹³ Furthermore, our study has shown that the Clavien–Dindo score was lower in the laparoscopy group than in the open group. A recent study showed that the most common surgical complication is surgical site infection (SSI), while POI is the third most common complication. ²⁰ In our study, the most frequent complications were POI and PPOI, followed by SSI. With regard to LOS, we found that it was shorter in the laparoscopy group compared to the open group. This was similar to the findings of a recent retrospective cohort study performed in the United States, which investigated both surgical approaches and robotic surgery, and showed that robotic and laparoscopic surgeries had shorter LOS than the open approach. ²¹ In addition, a better quality of life was observed in patients that underwent laparoscopic colectomy in a randomized controlled trial. ²²

The study’s main limitations were its retrospective nature, small number of patients, and the lack of standardized definitions of POI and PPOI. These results are limited to a small population in a single tertiary center, therefore not necessarily applicable to the general population and definitive conclusions cannot be made. Further studies with a larger sample size, involving multiple colonic diseases, are required to address this issue.

The incidence of both POI and PPOI were significantly influenced by the surgical approach used. The laparoscopic approach showed better postoperative mobilization and tolerance to regular diet. However, overall complications were more frequent in the open approach. Our suggestion for future studies is to address more extensive pre-and postoperative measures aimed at decreasing the incidence of POI, PPOI, and other complications of colorectal surgeries.

Zenodo: Complications and postoperative ileus in laparoscopic versus open colectomy: A retrospective cohort study. http://doi.org/10.5281/zenodo.4730346. ²³

This project contains the following underlying data: -

raw data.xlsx

Data are available under the terms of the Creative Commons Attribution 4.0 International license (CC-BY 4.0).