Introduction
Colorectal cancer (CRC) ranks third in the global incidence of malignant tumors, affecting human health seriously [
1]. According to the American Cancer Society statistics, the incidence of CRC in males is lower than that in prostate and lung cancer. In females, CRC incidence is lower than that in breast and lung cancer [
2]. CRC ranks second in males and third in females worldwide [
2].
Since the 1990s, laparoscopic surgery has attracted more and more attention as a new minimally invasive surgical method to treat many diseases [
3,
4]. Many multi-center researchers and a large number of randomized controlled trials have confirmed the safety and reliability of laparoscopic rectal cancer surgery [
5‐
7]. The laparoscopic operation can follow the total mesorectal excision (TME) principle with many advantages, including minor surgical trauma and bleeding, lower surgical complications and postoperative pain, and faster postoperative recovery to achieve R0 resection of the tumor [
8,
9]. As a result, laparoscopic low anterior resection (LAR) has become an essential operation for rectal cancer.
Many patients receiving rectal cancer surgery routinely undergo protective ileostomy to avoid clinically significant anastomotic leakage and other complications. However, whether it is necessary to perform the ileostomy is still controversial. At present, there are only a small number of randomized controlled trials and comparative studies assessing the effect of ileostomy during laparoscopic rectal cancer surgery. Therefore, the efficacy of ileostomy on laparoscopic rectal cancer surgery was analyzed using the meta-analysis.
Methods
Search strategies
We perform this study according to the stated guidelines of Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) [
10]. The protocol was registered with PROSPERO in advance (CRD42021253824) (
https://www.crd.york.ac.uk/PROSPERO/#recordDetails). Cochrane Library, EMBASE, Web of Science, and PubMed were applied for systematic search of all relevant literature, updated to May 07, 2021, using the search strategy including the terms for “Rectum tumor,” “Rectal Neoplasms,” “Ileostomy,” “Ileostomies,” “Laparoscopy,” “Celioscopy,” and their variants. The detailed search strategy was shown in Additional file
1. In addition, reference lists of all primary studies and systematic reviews published were searched manually for additional trials.
Inclusion and exclusion criteria
The inclusion criteria include [
1] patients who were diagnosed with rectal cancer by histopathology and were treated with laparoscopic rectal cancer surgery, [
2] clinical trials compared patients with and without protective ileostomy, [
3] patients were grouped by whether they received protective ileostomy [
4] at least one of the following outcomes were reported (incidence of postoperative anastomotic fistula, reoperation rate, length of hospital stay, incidence of postoperative peritonitis, and postoperative incision infection rate), and [
5] studies reported in English.
The exclusion criteria include [
1] studies included benign colorectal diseases [
2]; review, letter, case report, or meta-analysis [
3]; other surgical methods were used, such as open abdominal Dixon surgery, Miles, robot-assisted resection of the rectum, and natural orifice transluminal endoscopic surgery (NOTES) [
4]; single group studies; and [
5] complete extraction could not be obtained.
In the case of studies comprising the same patient cohort, we only selected the most recent or the complete study.
Data extraction and methodology quality assessment
MY, ZHM, ZLX, HHY, and LJN extracted and assessed the data independently. At least two researchers assessed each citation. The extracted data mainly included the first author, publication year, number of patients, age, sex, clinical characteristics, and outcomes. The primary outcome mainly included clinically significant anastomotic leakage rate, and the secondary outcomes include operation time, length of hospital stay, complication rate, and reoperation rate. Any uncertainties about the extracted data were carefully discussed until a consensus was reached.
The NOS was applied for the assessment of non-randomized controlled trials in this study, including selecting patients, comparability and controls on the study design, and outcome assessment. The study was considered moderate to high quality if the score was more than six. Randomized studies were assessed with the Jadad scale. Two researchers scored each study independently, and a discussion was made if there was a disagreement.
Statistical analysis
Review Manager 5.3 provided by the Cochrane Collaboration Network was used for statistical analysis. Weighted mean differences (MDs) and risk ratios (RRs) with 95% confidence intervals (CI) were calculated for analyzing continuous and dichotomous data, respectively. P < 0.05 was considered as statistical significance. I2 test was applied to assess the heterogeneity of the studies. There was significant heterogeneity when I2 > 50% or P < 0.05 and the random-effects model should be used. Otherwise, the fixed-effects model should be applied. We would employ a sensitivity analysis when there was significant heterogeneity of studies. By omitting each study at a time, the effect of an individual study on the overall heterogeneity could be analyzed. Funnel plots assessed publication bias.
Discussion
Colorectal malignancies are common digestive system tumors with increased morbidity and mortality every year [
1]. Surgical resection is still the main method to treat CRC. TME is the key to radical resection for rectal cancer [
4,
8,
18], and rectal washout has been integrated with the TME technique in recent years [
19]. According to the principles of TME, laparoscopic LAR for rectal cancer has become the mainstream treatment method for patients [
20,
21]. Mainly, laparoscopic tumor-specific mesorectal excision (TSME) has been utilized for upper rectal cancer [
22]. Many problems that cannot be solved by conventional laparotomy, such as anus-preserving surgery for rectal cancer with anal distance less than 5cm, can be solved by laparoscopic technique, which significantly improves the life quality of patients [
23,
24]. Laparoscopic surgery has many advantages, such as a clear vision of operation, small incision, mild postoperative pain, and short hospital stay [
25‐
27]. Especially, to pursue better treatment benefits, many laparoscopic surgery regimens have been improved, such as 3D laparoscopy, robotic surgery, single-incision laparoscopic surgery, two-port laparoscopic anterior resection, and LAR with natural orifice specimen extraction (NOSE-LAR) [
28‐
32]. Moreover, some researchers proposed that sphincter-saving rectal resections can be seen as a feasible alternative to extra levator abdominoperineal resection or transanal total mesorectal excision [
33].
Clinically significant anastomotic leakage is the main postoperative complication that threatens the life of patients with rectal cancer. The CT scan was the most common diagnostic method, and contrast enema, endoscopic examination, reoperation, and other postoperative complications also can assist in the diagnosis [
34]. Clinically significant anastomotic leakage for laparoscopic rectal cancer surgery usually occurs on the 7th day post-surgery [
35]. Generally, the small anastomotic leakage can be treated through conservative treatment (fasting water, nutritional support, anti-inflammation, fluid replacement), while the severe clinically significant anastomotic leakage requires surgical intervention [
36]. We mainly focused on whether protective ileostomy can decrease the rate of clinically significant anastomotic leak, and tiny leaks are not evaluated in any studies.
Protective colostomy or ileostomy has been performed to reduce the incidence of clinically significant anastomotic leakage. However, it is still controversial that whether it is beneficial to perform the protective stoma. Some surgeons believe that preventive intestinal stoma is unnecessary because most clinically significant anastomotic leakage can be cured conservatively, and intestinal stoma increases the cost of hospitalization and requires secondary surgery [
37‐
39]. The second surgery will undoubtedly delay the subsequent treatment of patients with rectal cancer surgery, especially those old and have many underlying diseases (such as postoperative chemoradiotherapy), and even accelerate the deterioration and increase mortality [
40‐
43]. On the other hand, frail patients often require ileostomy closure surgery before adjuvant therapy (due to frequent nutritional deficiencies), which delays adjuvant therapy.
Moreover, researchers have not agreed on the concrete interval time from ileostomy closure surgery to adjuvant therapy. Some researchers believe that protective intestinal stoma can reduce intestinal edema, promote local intestinal function recovery, facilitate anastomotic healing, and reduce clinically significant anastomotic leakage and other complications [
11]. This meta-analysis was based on seven comparative studies with more than 1000 patients. The results showed that protective ileostomy could reduce clinically significant anastomotic leakage in patients who underwent laparoscopic rectal cancer surgery. However, ileostomy has no influence on postoperative hospital stay, reoperation, wound infection, and operation time.
According to the principles of enhanced recovery after surgery (ERAS), postoperative recovery has been significantly increased [
44]. The postoperative hospital stay showed no difference in the analyzed groups. According to the combination of included studies and clinical experience, although preventive stoma could reduce clinically significant anastomotic leakage, most clinically significant anastomotic leakage could be cured through conservative treatment. There was no significant difference in the perioperative reoperation rate between the two groups of patients. Besides, there was no statistical difference in wound infection rate and operation time in the ileostomy and non-ileostomy groups.
There are some limitations to this meta-analysis. First of all, the funnel plot suggests that publication bias has existed in this study because the included studies are all retrospective studies without a random controlled trial. Second, the surgeons have a clinical bias in performing protective ileostomy for patients receiving laparoscopic rectal cancer surgery possibly. Many different factors can influence the clinically significant anastomotic leakage rate, such as anastomotic position, sex, tumor diameter, operation time, location of ligation IMA, and postoperative management [
36,
45,
46]. Therefore, there is bias in choosing ileostomy for patients receiving laparoscopic rectal cancer surgery. Third, the bias is caused by the length between the anastomosis and anal verge, not reported in all studies. It seems that clinically significant anastomotic leakage is tentative to occur when the anastomosis is lower [
47,
48]. Fourth, the types of performed rectal resection are not thoroughly evaluated. In four studies, the authors have reported them as LAR, while in the other three studies, it has been TME. The other two studies have not reported rectal resection type. However, concerning the relative publication date of all seven studies, LAR is performed according to the principles of TME in all the cases. In addition, because it is unclear whether these included patients have received neoadjuvant therapy or not, this article did not evaluate the influence of neoadjuvant therapy on the leakage rate. Interestingly, a meta-analysis has reported that neoadjuvant therapy did not increase the rate of postoperative anastomotic leakage after middle and low rectal anterior resection [
49], which guides future research. Finally, this study did not analyze whether ileostomy can affect the long-term survival rate of patients, bowel function, and postoperative quality of life.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.