Objective: To explore the effect of enterostomy on analgesic pattern in advanced digestive tract cancer. Methods: A retrospective cohort study was carried out, which was approved by the Ethics Committee of the Sixth Affiliated Hospital of Sun Yat-sen University (E2018026). Inclusion criteria were as follows: (1)age and gender were not limited; (2) all the gastrointestinal malignancies were confirmed histologically, and local recurrence or metastasis were confirmed by CT or MR; (3) numerical rating scale (NRS) ≥4 points, opioid analgesic drugs were required; (4) informed consents were signed by patients of their own. Exclusion criteria were as follows: (1) malignancies of early stage; (2) suspicious adverse mental states which might lead to poor administration compliance; (3) hypersensitivity or allergic reactions to opioids. Clinical data of patients with advanced gastrointestinal cancer receiving comprehensive treatment at the Medical Oncology Department of the Sixth Affiliated Hospital of Sun Yat-sen University from September 2016 to April 2017 were retrospectively collected. The patients were divided into the stoma group and the non-stoma group. The clinical findings of two groups were analyzed, including age, sex, ostomy status, pain location, presence or absence of intestinal obstruction, pain characteristics, selection of opioid analgesic agents, treatment of side effects of analgesics. Pain was assessed using brief pain inventory(BPI) table and NRS score. Strong opioids were prescribed for patients of NRS ≥4. Patients who were intolerant to opioids required opioid titration. The titration drugs included oral or IV morphine and oxycodone. After achievement of adequate pain control, long-acting opioids were administered, which included sustained-release morphine tablets, controlled release oxycodone and transdermal fentanyl. Criteria for pain relief included NRS≤3, breakthrough pain <3 times/day and duration of adequate pain control >3 days. The χ(2) test and the Wilcoxon signed rank sum test (nonparametric test) were used to analyze the clinical features of patients in the stoma and non-stomach groups. In order to find the factors associated with maintenance therapy and the use of laxatives, the variables were compared as well as in multivariate analysis with multiple regression models. For all the statistical tests, a value of P<0.05 in a two-tailed test was established as the alpha significance level. Result: A total of 123 patients were enrolled in this study, including 79 males (64.2%) and 44 females (35.8%) with a median age of 51 years. Fifty-two patients were in stoma group, including 30 (24.4%) of ileostomy and 22 (17.9%) of colostomy, and 71 patients were in non-stoma group. Pain of 40 (76.9%) patients in stoma group located in abdomenopelvic site while the pain of 44 (62.0%) patients in non-stoma group located in other sites. Compared with non-stoma group, cases in stoma group complained more abdominopelvic pain (73% vs. 62.0%, P<0.001).The median NRS pain score before treatment in the stoma group and the non-stoma group was 5.7 and 5.6, respectively, without statistically significant difference (P=0.741). After analgesic management, the above scores reduced to 2.1 and 2.3, respectively, without statistically significant difference as well (P=0.092). Analgesic treatment was effective in 111 cases (90.2%), including 49 cases (94.2%) in the stoma group, and 62 cases (87.3%) in the non-stoma group, and there was no statistically significant difference between the two groups (P=0.202). There was more application of fentanyl transdermal patch [34.6%(18/52) vs. 9.8%(7/71)] in the stoma group, while more application of lactulose laxative [78.9%(56/71) vs. 61.5%(32/52)](χ(2)=10.023, P=0.002) in the non-stoma group. Multivariate analysis revealed that ostomy (OR=0.290, 95%CI: 0.102-0.824, P=0.009) and pain site (OR=5.691, 95%CI:1.709-18.948, P=0.005) were independent factors affecting the choice of the first line opioid sustained release agent. Of the 123 patients with maintaining analgesia, 98 had available data of laxative use, of whom 46 used laxatives to prevent or treat constipation, and the proportion of laxatives in stoma group (21.2%, 11/52) was significantly lower than that in non-stoma group (49.3%, 35/71) (χ(2)=6.957, P=0.008). Multivariate analysis of the application of laxative use showed that age (OR=0.281, 95% CI: 0.123-0.684, P=0.010) and ostomy (OR=2.621, 95% CI: 1.033-6.687, P=0.045) were independent factors affecting the use of lactulose laxatives. Conclusions: Enterostomy may affect the analgesic pattern in advanced digestive tract cancer. Patients with stoma are more likely to use fentanyl transdermal patches and younger patients with stoma do not need prophylactic use of laxatives.