Risk factors for developing high-output ileostomy: a retrospective study


 Background: Prophylactic ileostomy and colostomy have been widely used to reduce the risk and complications of anastomotic leakage with high-risk colorectal cancer after operation. However, prophylactic ileostomy itself has some complications, and ileostomy high out-put syndrome is one of them. This study was performed to explore the risk factors of HOS in ileostomy.Methods: A total of 114 patients with HOS were screened out from 494 eligible ileostomy patients in the last five years. The clinical and pathological data were analyzed. The relationship between HOS and clinicopathological data was analyzed. Multivariate analysis was performed by logistic regression.Results: There was no clear correlation between the occurrence of HOS with sex, age, gross typing, histological grade, location of tumors, lymph node metastasis and TNM stage (p > 0.05). Preoperative complications including inflammatory bowel disease, diabetes mellitus and neoadjuvant chemoradiotherapy were risk factors for HOS (p < 0.05). Total colectomy and abdominal infection were risk factors for HOS (p < 0.05) during operation.Conclusion: Inflammatory bowel disease, diabetes mellitus and neoadjuvant radiotherapy and chemotherapy in patients with colorectal cancer are the preoperative risk factors for HOS. Total colectomy and postoperative abdominal infection are the postoperative risk factors for HOS.

with obstruction, edema or others with high-risk factors, it may be necessary to adopt preventive ileostomy to reduce the risk and complications of anastomotic leakage when resecting the lesion and one-stage anastomosis. However, prophylactic ileostomy itself has some complications, and ileostomy high out-put syndrome is one of them.
The risk factors and predictors of high-output ileostomy have been reported in studies with a small sample size [5,6]. We conducted a single-institution retrospective review of CRC. Patients underwent ileostomy in our hospital, and were analyzed with clinical and pathological data to explore the risk factors of high-output ileostomy.

Methods
Ileostomy (both prophylactic and therapeutic) was performed in CRC patients between 2013 to 2018 in Anyang Tumor Hospital of Henan province China. HOS was defined as patients whose daily stoma output is more than 2000 ml and lasted more than two days as previous study reported [7]. Patients enrolled were between the age of 26 to 87 years old, with complete medical records and were followed up either with ileostomy reversal or more than one year after ileostomy. The others were excluded from the group whose stoma discharge was not detailed and did not meet the follow-up requirements. There were 114 patients (63 males and 51 females) with HOS were screened out from 494 eligible ileostomy patients in last five years.
Early HOS (EHOS) was defined as HOS developed within 3 weeks after operation, and late HOS (LHOS) was those over 3 weeks.
We collected clinical data using patient medical records, outpatient follow-ups, pathological stages, and complications like stoma discharge, electrolyte imbalance, dehydration, as well as disease treatments. Discharged patients were followed up to 1 year after ileostomy or ileostomy reversal. SPSS17.0 statistical software was used for the statistical analysis. Single factor analysis was performed using chi-square test and Fisher's exact probability method. Multifactor analysis was performed with logistic regression. P < 0.05 had statistical significance.

HOS and preoperative complications.
These data showed that preoperative complications of inflammatory bowel disease, diabetes mellitus and neoadjuvant chemoradiotherapy are associated with the occurrence of HOS in patients with colorectal cancer (Table 3). Table 3 Univariate analysis of preoperative complications in HOS and non-HOS groups.

HOS and surgical related indicators.
These figures showed that total colectomy and postoperative abdominal infection are correlated with the occurrence of HOS, but not with operation time, bleeding volume, application of diuretics and laparoscopy (Table 4).

EHOS/LHOS and related pathological factors.
The study demonstrated there is no significant difference between the occurrence of HOS and related pathological factors (inflammatory bowel disease, diabetes, mellitus, hypoproteinemia, neoadjuvant chemotherapy, anemia, etc.) in ileostomy patients (table 5). Table 5 The relationship between EHOS, LHOS and related pathological factors.

Discussion
HOSs or high-output syndrome is rarely studied and lacks attention. There is no consensus on the definition of HOSs (HOS) [5,7,8,9,10]. Dehydration, electrolyte disturbance (hyponatremia, hypochloremia, hypomagnesemia), renal failure and malnutrition (late stage) can occur in high-output ileostomy [5,7,8]. HOS increases the risk of readmission of CRC patients. Some scholars defined HOS as stoma output of more than 1500 ml per day for more than two consecutive days [8,9], while the others advocate that HOS should be defined as the output of more than 2,000 ml per day for more than two or three consecutive days [5,7]. As the complications are more likely to occur when the daily stoma output exceeds 2000 mL, we adopt the latter standard in our study. HOS can be classified as early (< 3 weeks after initial ostomy surgery) or late HOS (3 weeks after surgery) and this is widely accepted by previous studies.
In our cohort, the incidence of HOSs (114/494) was 23.07%, higher than earlier reports (17%) [7]. Dehydration was the most common symptom of HOS (37.7%), followed by electrolyte disturbance (28.1%), local dermatitis (21.9%), renal dysfunction (5.3%) and malnutrition (7.0%) in long-term HOS. It was reported that the incidence of HOS is 26% in 262 patients with ileostomy during hospitalization, 30% of patients were re-admitted within 30 days after discharge, and 37% of patients with re-hospitalization were due to dehydration [11]. The re-admission risk of ileostomy patients with inflammatory bowel disease was double that of other risk factors (OR 2.04) [12].
Our study found no clear correlation between the occurrence of HOS and gender, age, gross typing, histological grading, location of tumors, lymph node metastasis and TNM staging in patients with colorectal cancer.
Preoperative complications of inflammatory bowel disease, diabetes mellitus and neoadjuvant chemoradiotherapy are risk factors for HOS. Inflammation caused by abnormal reaction of intestinal mucosal immune system act as an important role in the pathogenesis of inflammatory bowel disease, and is also the main cause of HOS. Diabetes mellitus occurred in patients with colorectal cancer because of the disorder of glucose metabolism and utilization, such as inappropriate control, of blood sugar or intestinal movements, which may result in large amount of liquid discharge from ileostomy [5]. We found that neoadjuvant concurrent radiotherapy and chemotherapy (CRT) is a risk factor for HOS (HOS). Radiation enteritis caused by neoadjuvant radiotherapy might be responsible for HOS in preoperative CRC patients [13]. As neoadjuvant chemoradiotherapy is mostly carried out simultaneously before operation in our hospital, there is no stratified study of neoadjuvant chemotherapy, radiotherapy alone and neoadjuvant concurrent radiotherapy and chemotherapy. It has been reported that preoperative radiotherapy alone, the distance between the tumor and anal margin, could affect the intestinal function of patients after operation, while chemotherapy alone has no significant effect on intestinal function after operation [14].
Among the related factors of operation, total colectomy and abdominal infection are the risk factors of HOS, but not the operation time, bleeding volume, diuretic application and laparoscopic operation. Diabetes mellitus and total colectomy are high risk factors for HOS, which are consistent with previous study [5]. As it is reported bile acid deficiency is one of the mechanism of total colectomy caused HOS. Total proctocolectomy prevents the reabsorption of bile acids absorbed by the ileocecum, as a result decrease in bile acid pools inhibits lipid absorption. Consequently, unabsorbed long-chain fatty acids arehydroxylated or desaturated by anaerobic intestinal bacteria, triggering the secretion of fluid and electrolytes, which may lead to the development of HOS [5]. Bile acid deficiency may also cause changes in the intestinal flora that increase intestinal drainage [15]. The loss of water absorption by the colon in patients undergoing total colectomy can also lead to high ileostomy output. In our study, we did not find that the use of steroids and diuretics after ileostomy increase the high output of ileostomy, but the use of large doses of diuretics may increase the risk of dehydration.
Generally, half of the EHOS patients required no special drug intervention until recuperation, while the other half required drug intervention [7]. For the treatment and management of HOS, patients should be managed and guided throughout the hospital and after discharge [16]. In addition to basic treatments (such as restricting fluid intake, rehydration and correcting electrolyte disturbance) reducing the secretion of somatostatin and oral intake of loperamide is also effective in reducing discharge. Pieter-Jan Cuyle reported that 17% of patients with ileostomy had high excretion [17]. This complication would affect the implementation or completion of adjuvant therapy. Somatostatin analogues (ranitides, etc.) could reduce the excretion of ileostomy effectively. Routine dosage of loperamide is 20 mg twice daily by mouth, but dosage can be increased if not effective. Alicia Mackowski reported that increasing the dosage of loperamide to 300 mg per day in individual patients can reduce the amount of stoma excretion, without observed abnormality in renal function [18].
This paper is a retrospective study with some limitations. Prospective study should pay attention to the patients with high risk factors HOS, strengthen the supervision and treatment after operation and discharge for reducing the complications of stoma, and facilitate the smooth progress of comprehensive treatment as well as the readmission rate caused by HOS [19][20][21][22].
We followed up with patients who had ileostomy reversal or 1 year after ileostomy by inpatient medical records, outpatient review and telephone follow-up. These data mostly came from patients' observation records in the hospital (including outpatient review) belonging to on-site data collection. Using remote video conference to evaluate ileostomy output, and taking early intervention measures to improve prognosis is a method worth exploring [23]. Multicenter, large sample randomized controlled studies of HOS should be conducted in the future.

Declarations
Availability of data and materials The datasets used during the current study are available from the corresponding author on reasonable request.
Ethics approval and consent to participate The study protocol was approved by the ethical committee for scientific research and drug clinical trial of AnYang Tumor Hospital (NO. AZLL022019048190812).

Consent for publication
Not applicable

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

Funding
No resources or funding was obtained for this study.