Short-Term Outcome Analysis of Temporary Ostomy in Low Anterior Resection for Rectal Cancer-A Retrospective Study

Chuanduo Zhao Department of Colorectal Surgery, National Cancer Center/ National Clinical Research Center for Cancer/ Cancer Hospital, Chinese Academy of Medical Science and Peking Union Medical College, Beijing Sicheng Zhou Department of Colorectal Surgery, National Cancer Center/ National Clinical Research Center for Cancer/ Cancer Hospital, Chinese Academy of Medical Science and Peking Union Medical College, Beijing Hao Su Department of Colorectal Surgery, National Cancer Center/ National Clinical Research Center for Cancer/ Cancer Hospital, Chinese Academy of Medical Science and Peking Union Medical College, Beijing Jianwei Liang Department of Colorectal Surgery, National Cancer Center/ National Clinical Research Center for Cancer/ Cancer Hospital, Chinese Academy of Medical Science and Peking Union Medical College, Beijing Zhixiang Zhou (  drzhixiangzhou@126.com ) Department of Colorectal Surgery, National Cancer Center/ National Clinical Research Center for Cancer/ Cancer Hospital, Chinese Academy of Medical Science and Peking Union Medical College, Beijing

and a lower site of anastomosis (6.0 cm vs. 9.0 cm, P < 0.001) than the ostomy group. The ostomy group had a lower rate of anastomotic leakage (0.4% vs 3.6%, P = 0.039), reduced reoperation rate (0.4% vs 2.9%, P = 0.044), longer operating time (214.9 min vs. 151.4 min, P = 0.009) and shorter time to rst atus (3.4 days vs. 4.1 days, P = 0.005) than the non-ostomy group. The safety of reversal surgery is acceptable; however, 21.1% of ostomates did not complete the reversal operation for various reasons within 1 year after the rst operation. Age (P = 0.029) and AJCC stage (P = 0.043) may be important factors affecting the closure of ostomy.
Conclusions Although time-consuming, temporary ostomy is a good option for high-risk patients to reduce the rates of anastomotic leakage and reoperation in low anterior resection. Considering the ostomy-related inconveniences in daily living, other related complications and rather high proportion of non-reversal, the scope of application should be more for patients with low anastomosis and neoadjuvant treatment.

Background
As one of the ve leading causes of cancer-related deaths in China (1), colorectal cancer is gaining increasing attention among doctors and the general public. With a special disease spectrum, China has a much higher proportion of rectal cancer, especially low rectal cancer, than that in western countries (2).
Anastomotic leakage is a severe complication and can greatly prolong the hospitalization period and increase medical costs in low anterior resection (LAR). Several studies have con rmed that construction of a temporary ostomy will bene t patients who have undergone low-rectal anus preserving surgery and are at high risk of anastomotic leakage (3)(4)(5)(6)(7). A multivariate analysis (8) showed that gender and level of anastomosis were risk factors for anastomotic leakage, which suggests that temporary ostomy may be a good option to protect the anastomosis.
Although the safety and advantages of temporary ostomy in LAR have been con rmed in previous studies, the necessity of ostomy was not strictly de ned in LAR. In this study, we retrospectively analyzed the short-term postoperative recovery parameters of two groups-ostomy and non-ostomy-to evaluate the impact of this approach. The clinical parameters included hospitalization situation, surgical procedure, recovery of gastrointestinal function, postoperative pain, and short-term postoperative complications. To assess the impact of the ostomy, we collected one-year follow-up data of the ostomy group after surgery.

Patients And Methods
The study was approved by the ethics committee at the Cancer Hospital, Chinese Academy of Medical Sciences, and the protocol conformed to the ethical guidelines of the 1975 Declaration of Helsinki. We retrospectively analyzed 2321 consecutive patients from May 2014 to August 2018 who were diagnosed with rectal cancer and underwent laparoscopic LAR, with or without temporary ostomy. Patients ranged from American Joint Committee on Cancer (AJCC) stage I to IV. All included patients underwent colonoscopy and pathological biopsy before surgery to con rm the diagnosis.
All patients in this study underwent cur curative-intent or palliative (For AJCC stage IV) LAR. Based on whether the intraoperative temporary ostomy was performed, the patients were divided into two groups. Because the number of patients in the non-ostomy group was much higher than those in the temporary ostomy group, we randomly selected the same number of patients as in the temporary ostomy group for control statistics. Patients with temporary ostomy were matched in a 1:1 ratio with patients without ostomy. The matching criteria included age, body mass index (BMI), American Society of Anesthesiology (ASA) score, and tumor stage.
At our institution, the majority of LARs are performed laparoscopically. Normally four trocars are used for bowel mobilization, division, ligation of vessel, and lymphadenectomy. Transection of the bowel and anastomosis were performed either intracorporeally or extracorporeally according to the situation. When the situation permitted, the natural ori ce specimen extraction surgery technique was also used in surgery. Regardless of the method used, the total mesorectal excision principle was strictly followed during surgery. If the patient required a protective stoma, loop ileostomy was the routine procedure. All patients underwent a preoperative colonoscopy, along with chest, abdominal, and pelvic enhanced computed tomography. However, for some patients, pelvic MRI and endoscopic ultrasonography were prescribed to further evaluate the local lesion to determine the surgical procedure.
We collected demographic and clinical data of the patients for this study, including age, gender, BMI, ASA score, and tumor location and size. Meanwhile, operative details including operating time and estimated blood loss were recorded. To comprehensively evaluate recovery, hospitalization days, postoperative hospitalization days, time to rst atus, time to rst oral intake, and postoperative pain score were also recorded. To assess the impact of the temporary ostomy, we recorded ostomy-related complications and patients' reasons for not undergoing reversal surgery within one year Quantitative variables are expressed as median and range, and comparisons were analyzed using the Student's t-test according to data distribution. Categorical variables are presented as number and percentage and were analyzed by chi-square test or Fisher's exact test. All tests were two-sided, and P < 0.05 was considered to indicate as statistical signi cance. The Statistical Package for the Social Sciences (SPSS, version 25.0; IBM Corp., Armonk, NY, USA) was used for all analyses.

Results
A total of 2321 consecutive patients with rectal cancer who underwent LAR at our institution from May 2014 to August 2018 were included. We selected laparoscopic surgery patients as the study subjects, among them, 279 patients underwent temporary ostomy. Patients with temporary ostomy were matched 1:1 with patients that did not undergo temporary ostomy (Fig. 1). Table 1 shows the demographic and clinical characteristics of the patients. The proportion of patients who received neoadjuvant chemoradiation in the ostomy group was signi cantly higher than that in the non-ostomy group (39.4% vs. 10.0%, P < 0.001). The tumor position in the ostomy group was lower than that in the non-ostomy group (6.0 cm vs. 9.0 cm, P < 0.001). In other aspects, gender, age, BMI, ASA score, tumor size, and tumor stage were essentially similar between both groups. The operative outcomes are detailed in Table 2. The mean operating time in the ostomy group was signi cantly longer than that in the non-ostomy group (214.9 min vs. 151.4 min, P = 0.009). The estimated blood loss and open surgery rate were not statistically signi cant between the two groups. There was no statistically signi cant difference between the two groups in the number of regional lymph nodes harvested and the distal resection margin, which were used to evaluate the thoroughness of surgical resection. For postoperative recovery, the mean time to rst atus in the ostomy group was shorter than that in the non-ostomy group (3.4 days vs. 4.1 days, P = 0.005). The postoperative pain score in the two groups showed no difference in the rst, second, and third day after surgery. The ostomy group showed a lower anastomotic leakage rate than the non-ostomy group (0.4% vs. 3.6%, P = 0.039). Therefore, the reoperation rate difference between the two groups was also statistically signi cant (0.4% vs. 2.9%, P = 0.044). Other postoperative complication rates including bleeding, intestinal obstruction, abdominal infection, wound infection, cardiopulmonary events, deep-vein thrombosis, and urinary infection were similar between the two groups. The hospitalization and cost outcomes are detailed in Table 3. The mean postoperative hospital stay was 8.3 days in the ostomy group versus 9.3 days in the non-ostomy group, with no statistically signi cant difference (P = 0.498). The mean total hospitalization costs for the two groups were $12087.2 and $11783.3, respectively, and there was no statistical difference (P = 0.097).  Table 4 describes the complications related to the stoma procedure after the rst discharge. It should be noted that many of these complications disappeared with the reversal of the stoma. For patients whose ostomy was reversed, complications were mainly investigated between the two operations. For patients whose ostomy was not reversed, complications were those that occurred within one year of the rst operation.  1%) had not yet received a reversal operation one year after the operation. The mean postoperative hospitalization duration was 4.8 days, and the mean interval between two operations was 127.5 days. The most common three complications for reversal surgery were wound infection (4.1%), cardiopulmonary disease (2.7%) and intestinal obstruction (1.4%).   Unwilling to undergo surgery 4 6.8 Table 7 presents the factors that may in uence whether the reversal surgery can be performed. Comparison of patients who did undergo reversal of temporary ostomy and those who did not yielded two factors effected the decision on closure, including age over 65 years old (P = 0.029) and advanced AJCC stage (III and IV) (P = 0.043).

Discussion
With increased life expectancy, the incidence and mortality of colorectal cancer is also increasing worldwide and especially in China (1,9,10). Unlike in western countries, rectal cancer is more common in China, and the majority of patients have lower rectal cancer (2). Because surgical resection is the main form of treatment for rectal cancer, the choice of surgical method is particularly important. Anastomotic leakage is the most serious complication after rectal resection and is associated with increased morbidity, mortality, and prolonged hospitalization (11)(12)(13). Mirnezami and McArdle reported that anastomotic leakage can lead to an increase in local recurrence of rectal cancer, reducing the overall survival (14,15). Rullier(8) conducted a multivariate analysis and showed that the level of anastomosis was an independent factor for development of anastomotic leakage. It is worth noting that surgeons are more inclined to perform temporary ostomy in male patients to avoid anastomotic leakage; this is consistent with Rullier's (8) and Lipska's (16) conclusions that male sex is a risk factor of anastomotic leakage. This may be due to the combination of anatomical differences in the narrower male pelvis (16,17) and the recently shown hormonal differences that in uence the intestinal microcirculation (18). Low anterior resection with intraoperative temporary ostomy is a type of anus-preserving surgery that is used by surgeons to minimize the consequences of postoperative anastomotic dehiscence (4). Several studies(4-7) have con rmed that temporary ostomy is effective and bene cial for patients at high-risk of anastomotic leakage, such as those with previous irradiation of the pelvis, those who are hemodynamically unstable (e.g., trauma or sepsis), those undergoing chemotherapy or treatment with other biological agents that can impair wound healing, or those who have an intestinal anastomosis < 5-7 cm from the anal verge. A meta-analysis and systematic review showed that a temporary ostomy reduces the rate of clinically relevant anastomotic leakages and is thus recommended in surgery for low rectal cancers (19).
Normally, patients with a temporary ostomy reported di culty in exercise, sleep, social activities, sexual activities, and wearing certain types of clothing. Additionally, some may suffer peristomal in ammation and dehydration resulting from high output. A systematic review (20) concluded that living with an ostomy negatively in uences the overall quality of life. However, studies have identi ed ostomates' two major phases: stoma-related di culties and perceived response shift. Patients' perception of quality of life with a temporary ostomy appears to have undergone a response shift through recalibration of their standards for measuring the quality of life and reconceptualization of what "good quality of life" is (21). In other words, patients' perceptions of the inconvenience of disease can be reshaped, in a positive direction. Florian investigated the impact of a diverting stoma on the quality of life in patients undergoing rectal cancer resection before and after stoma closure and found out that the negative impact on social functioning and GI symptoms had no clinically relevant in uence on the global quality of life (22).
However, Taylor reported that some patients will experience altered bowel function after temporary ostomy reversal (23). A randomized multicenter trial showed that the severity of LAR syndrome was comparable in the ostomy and the non-ostomy groups, but incontinence for atus and liquid stools were more commonly reported in the ostomy group (24). For the elderly (25), especially patients aged ≥ 70 years, they undergo proportionately more permanent fecal ostomy procedures than younger patients, with longer hospital stays, more postoperative complications, and higher mortality rates after undergoing the reversal operation.
In our study, we found that even with a higher proportion of neoadjuvant chemoradiotherapy and lower anastomosis, the ostomy group had a lower rate of grade C anastomotic leakage than the non-ostomy group. Cases of grade C anastomotic leakage are often fatal if operative reintervention with control of the septic source is delayed or not performed (26). Although the operation was more time-consuming, the rst hospitalization expenses and postoperative recovery were similar between the two groups, which means that ostomy does not place too much extra burden on the patient or the health care system.
Orit Kaidar-Person summarized a series of complications of construction and closure of temporary loop ileostomy and reported a complication rate between 5% and 100% and incidence of non-closure rates between 0% and 19% (27). In our study, the complications associated with the stoma were mild, did not cause the patient great pain, and were completely curable with the reversal operation. Den Dulk's (28) study showed that 97% of patients chose to have a reversal operation within 1 year after prophylactic ostomy, while most of the remaining ostomy tend to be permanent. Studies have shown that postoperative metastasis (29) and local recurrence(30) of rectal cancer are risk factors for stoma to turn permanent, which was also consistent with the conclusion that advanced AJCC stage affected the rate of reversal in this study. It should be noted that anastomotic stenosis has often been neglected in the past, and its incidence has been reported as high as 30% in the previous research (31). Although endoscopic dilatation (32) is effective in the treatment of anastomotic stenosis, there are still many patients who cannot close their ostomies for this reason. Considering the high percentage of patients who can not reverse their ostomies in our study, preoperative consent for surgical patients is extremely important, especially given the possibility of the ostomy becoming permanent.
Even so, temporary stomas have a good preventive effect on grade C anastomotic leakage and they should be considered as an important means to avoid serious postoperative complications, especially in patients who cannot tolerate severe infection and have other risk factors and complications such as older age, primary underlying diseases, and immunocompromised status. To improve the quality of life of patients, reduce medical costs, and avoid permanent ostomy some studies have suggested temporary percutaneous ileostomy as a safe, e cient, and cost-effective option (11,33). However, because this method of fecal diversion may not be su cient, it has not yet been used on a large scale.
Our study has some limitations. First, the study is retrospective in nature and hence, biases may exist. Second, because the patients were not randomized, the surgeons decided which surgical approach was used, leading to a likely selection bias. Third, the end point of our study was the one-year follow-up after rst discharge from hospital, which means that long-term survival data and complications could not be completely observed and assessed. The lack of above-mentioned data may affect the objectivity and comprehensiveness of the conclusion. Last, we did not evaluate other techniques involved in reducing anastomotic leakage and the corresponding reoperation rates. Despite these limitations, a large number of patients were included from multiple aspects in our study and relevant clinical data were complete, detailed, and reliable. Prospective randomized controlled trials and more observation time are needed to further validate our ndings.

Conclusion
At the expense of everyday inconvenience, but reasonable operating time, temporary ostomy should be considered in low anterior resection as it reduces the rate of anastomotic leakage and reoperation rate in patients with a higher proportion of chemoradiotherapy and lower site of anastomosis. Overall, a signi cant proportion of patients' prophylactic ostomies become permanent, and the risk factors are age over 65 and advanced AJCC stage. Considering the possibility of the inconvenience associated with permanent ostomy, the application scope should be applied more to patients with low anastomosis and neoadjuvant treatment. The study was examined and approved by the local ethics committee and the protocol conformed to the ethical guidelines of the 1975 Declaration of Helsinki and its later amendments and regarding the involvement of human subjects for research. Written informed consent was obtained from individual or guardian participants.

Consent for publication
Not applicable Availability of data and material The datasets generated during and analysed during the current study are not publicly available due to patients' private information need to be protected but are available from the corresponding author on reasonable request

Competing interests
The authors declare that they have no competing interests.

Figure 1
Patient ow in the study. Laparoscopic surgery requires bowel mobilization, division, ligation of vessel and lymphadenectomy were done intracorporeally.