Complications associated with loop ileostomy: analysis of risk factors

Purpose: Loop ileostomy is a common surgical procedure but is associated with complications such as outlet obstruction (OO), parastomal hernia (PH), and high-output stoma (HOS). This study aimed to identify risk factors for these complications and their causal relationships. Methods: The study included 188 consecutive patients who underwent loop ileostomy between April 2016 and September 2021. Clinical factors and postoperative stoma-related complications (OO, HOS, and PH) were analyzed retrospectively. Stoma-related factors were evaluated by computed tomography (CT). The incidence, clinical course, and risk factors for the stoma-related complications were investigated. Results:


Introduction
Loop ileostomy is an important surgical procedure used in several situations, including diversion after anterior resection for rectal cancer and treatment of anastomotic leakage, preoperative treatment for advanced colorectal cancer with obstruction, and palliative decompression in unresectable cases.
However, with the increasing number of ileostomies being performed, complications such as outlet obstruction (OO), parastomal hernia (PH), and high-output stoma (HOS) are becoming a problem (1,2).
OO is a characteristic complication of loop ileostomy in which the small intestine becomes obstructed at the stoma site and is alleviated by inserting a catheter on the ileus opening side (3).OO is associated with multiple problems such as a need for fasting, prolonged hospital stay, delayed treatment, and sometimes a need for reoperation.Reports of the incidence of OO range from 5.4-27.3%(4)(5)(6)(7)(8).PH is de ned as herniation around the loop ileostomy and has a reported incidence of around 50% (9)(10)(11), but there is limited information in the literature on the relationship between stoma herniation and OO.HOS is a recently identi ed condition in which a large amount of uid is drained from the stoma and is thought to cause dehydration, electrolyte imbalance, and renal dysfunction.The reported incidence of HOS ranges from 16.4-31.0%(12)(13)(14)(15).In recent years, there have been some reports suggesting a relationship between OO and HOS, but the cause of HOS has not been fully investigated (16).
Knowledge of risk factors is important for prevention of these complications.Although there have been some studies of the risk factors for each complication separately and independently, there are few reports on the relationship between these risk factors.Therefore, the purpose of this study was to clarify the clinical characteristics of OO, PH, and HOS in patients with loop ileostomy and to identify risk factors for these complications and any causal relationships.

Patients and Methods
Consecutive patients who underwent loop ileostomy in the Department of Gastroenterological Surgery at Fukuoka University Hospital between April 2016 and September 2021 were reviewed in the study.The indications for loop ileostomy included bridge for surgery, diverting ileostomy when sphincter preservation surgery was performed, diversion for postoperative anastomotic leakage, and palliation in unresectable cases.
Information on demographic and clinical characteristics (sex, age, body mass index, primary disease, and reason for stoma construction) and on postoperative stoma-related complications (OO, HOS, and PH) was extracted from the medical records and analyzed retrospectively.Stoma-related factors, including thickness of subcutaneous fat and the rectus abdominis (RA) muscle, site where the stoma penetrates the RA muscle (medial, middle, lateral), and the long and short diameters of the stoma, were evaluated by thin-slice CT several months after surgery when the patient's condition had stabilized (Fig. 1).
OO was de ned as the presence of any two of the following: 1) symptoms of intestinal obstruction such as abdominal distension and vomiting, 2) CT scan showing small bowel dilatation shifting to nondilatation at the site where the stoma penetrates the abdominal wall, and 3) symptoms relieved by decompression via insertion of a tube through the stoma.HOS was de ned as drainage of at least 1500 ml per day from the stoma.PH was de ned as protrusion of abdominal contents through an abdominal wall defect in the vicinity of the stoma.
We investigated the incidence, clinical course, and risk factors for the three stoma-related complications (OO, HOS, and PH).

Creation of the loop ileostomy
First, the stoma site was marked preoperatively based on the Cleveland Clinic standards (17) by the wound ostomy care nurse assigned to the patient.The part of the ileum to be lifted was selected 20-30 cm orally from the ileocecal valve laparoscopically.After insu ation was stopped, a vertical skin incision was made at the marked site.The subcutaneous fat and the anterior sheath of the rectus abdominis muscle were then incised and the rectus abdominis was split.The intestine was exteriorized through the incision, and the antimesenteric border of the ileal loop was incised longitudinally to create an opening in the bowel.The bowel edges were then everted to expose the mucosa.The everted edges were sutured to the skin using interrupted absorbable sutures but not xed to the RA muscle or fascia.

Statistical analysis
Continuous variables are expressed as means ± standard deviations (SD) or medians with interquartile ranges (IQR) depending on the normality of the data distribution.Categorical data are shown as frequencies and percentages.
For group comparisons, continuous variables were compared using independent t-tests if normally distributed and with Mann-Whitney U tests if not.Categorical data were compared using the Chi-squared test or Fisher's exact test as appropriate.A p-value of less than 0.05 was considered statistically signi cant.
To identify factors associated with the complications, multivariate analysis was performed using a logistic regression model.Variables with a p-value less than 0.05 in the univariate analysis were included in the multivariate model.To compare stoma output during the rst 14 postoperative days between the groups, multivariate analysis of variance (MANOVA) was conducted.To investigate the relationship between size of the stoma and the stoma-related complication rate, cumulative sum analysis was used.
All statistical analysis was performed using JMP Pro version 15.0 (SAS Institute Inc, Cary, NC).

Ethical considerations
The study was approved by the Institutional Review Board of the Faculty of Medicine, Fukuoka University (approval number: 2018M022) and was conducted in accordance with the ethical principles outlined in the Declaration of Helsinki.Informed consent was obtained from all participants included in the study.

Data availability
The datasets used and analysed during the current study available from the corresponding author on reasonable request.

Patient background characteristics
The background characteristics of the 188 participants are shown in Table 1.The indication for stoma construction was initial decompression of stenosis at the primary site in 61 cases (32%), intraoperative diverting stoma in 112 (60%), and diverting stoma for anastomotic leakage in 15 (8.4%).The primary disease was colorectal cancer in most cases (96%).The scheduled stoma closure rate was 82.4% and the median time to stoma closure was 3.2 months (IQR 2.2, 4.4).

Outlet obstruction
OO was diagnosed in 28 cases, giving an incidence rate of 14.9%.The median time from creation of the ileostomy to diagnosis of OO was 4 days, and 85.7% (24/28) of the cases occurred within 10 days.Major symptoms at onset were abdominal distention (11 cases), nausea/vomiting (9 cases), and abdominal pain (8 cases).
Treatment involved insertion of a decompression tube via the stoma in 27 cases (96%), and most cases (24 cases) showed rapid improvement.However, in 5 cases, the stoma was closed earlier than planned.
Interestingly, in 24 cases (86%), signi cant stoma output (≥ 100 ml) was observed even on the day of diagnosis of OO, with a median output of 810 ml (IQR 180, 2100 ml).Moreover, 18 (67%) of the 24 patients with signi cant stoma output (≥ 100 ml) had stoma drainage that was more like a thin exudate than that from a normal ileostomy.
Univariate and multivariate analyses indicated that small stoma diameter at the RA level on CT and right sided stoma were signi cantly associated with OO (Table 2).

Stoma output
Stoma output was evaluated in 178 patients.Among these patients, the amount of stoma drainage continued to increase until 3 days after construction, at which time it reached a constant level of approximately 1000 ml (Fig. 2a).The stoma output within the rst 14 postoperative days was compared with various factors using multivariate analysis of variance (MANOVA).The results showed that OO and the reason for stoma creation were signi cantly associated with stoma output.Figure 2b and 2c shows the output of the ileostomy according to whether or not OO was present and the reason for stoma creation.
When de ned as output exceeding 1500 ml/day, HOS was observed in 57 of the 178 evaluable cases, giving an incidence rate of 32%.HOS developed at a median of 3 (IQR 2, 6) days after ileostomy creation.Among the 57 patients with HOS, the condition persisted for a median duration of 2 (IQR 1, 5) days.
Univariate and multivariate analyses that were performed to identify factors associated with HOS identi ed ileostomy creation (stoma for anastomotic leakage) as an independently associated factor (Table 3).

Parastomal hernia
PH occurred in 57 of the 188 patients, giving an incidence rate of 32.0%.Multivariate analysis revealed that high body weight and a large stoma diameter at the RA level on CT were signi cant related factors (Table 4).

Relationship between outlet obstruction, high-output stoma, and parastomal hernia
There was some correlation between the three types of complications (Table 5).There was a signi cant association of OO with HOS (p < 0.001).HOS was observed in 75% of patients with OO (21/28), whereas OO was observed in 36.8% of patients with HOS (21/57).As previously mentioned, the presence or absence of OO signi cantly affected postoperative stoma output according to the MANOVA analysis (Fig. 2b).The median maximum stoma output was 2295 ml for patients with OO and 1022 ml for those without OO (P < 0.001).
The association between occurrence of OO and PH was marginal (p = 0.071).As described previously, stoma diameter at the RA level on CT was signi cantly associated with OO and PH. Figure 3 shows the results of cumulative sum analysis, which was performed to assess the relationship between size of the stoma and the stoma-related complication rate.The incidence of OO decreased and the incidence of PH increased when the diameter of the stoma at the level of the RA muscle was larger than around 40 mm.

Discussion
In this study, we examined the clinical course, incidence, and risk factors associated with the important complications of loop ileostomy, with an emphasis on the relationship between these complications.We believe that our ndings will contribute to prevention and management of the complications of loop ileostomy.
Several studies have examined the risk factors for OO, including RA muscle thickness, sex, body mass index, and age, but a consensus has not yet been reached (5,8,15,16,18).In our study, the small diameter of the stoma at the fascial level was found to be involved in development of OO.To the best of our knowledge, no previous studies have explored the relationship between size of the fascial opening and OO.A previous study examined the association between skin incision length and OO but did not nd these factors to be signi cant risks (19).Importantly, that study focused on patients with ulcerative colitis undergoing total colorectal resection, a different population from ours.In addition, although there are reports indicating that RA muscle thickness ≥ 1 cm is a risk factor for OO (8,16,18,20,21), Although the thickness of RA muscle, both at the thickest part and at the stoma penetration site, was found to be greater in cases with OO, it was not identi ed as a signi cant risk factor in our study.If increasing the size of the fascial incision, a factor under the surgeon's control, can prevent OO, this could be a highly effective strategy for reducing the incidence of OO.
The observation that the incidence of PH increased in parallel with an increase in the stoma incision length identi ed the size of the stoma site as a signi cant factor in the development of hernia.While increasing the size of the stoma site is an effective way of preventing OO, it also increases the risk of hernia.According to our receiver-operating characteristic curve analysis, the optimal incision size for minimizing both OO and the risk of hernia is approximately 40 mm.However, considering the potential complications associated with OO, such as a need for fasting, extended hospital stay, and delay in postoperative recovery, a stoma that is larger than normal may be a viable option.Given that complications arising from OO can be considered more critical, this approach, which allows for some degree of herniation, might be acceptable, particularly if there is a plan to close the stoma in the future.
High stoma output in uences quality of life in patients with an ileostomy and their ability to perform activities of daily living (22).Although there have been several studies of risk factors for HO (13,14,23,24), our study has identi ed anastomotic leakage and OO as risk factors for high stoma output.MANOVA revealed a signi cant increase in stoma output in patients with both anastomotic leakage and OO, which is consistent with recent reports identifying organ/space, surgical site infection, and anastomotic leakage as factors associated with HO (13).Although OO is a type of bowel obstruction with dilatation of the small bowel, a certain amount of output was observed from the stoma at the time of diagnosis (median 810 ml [IQR 180, 2100]), and it is noteworthy that this output appeared to be more like exudate than intestinal uid.In our series, the onset of high output could occur before or after diagnosis of OO (Supplemental Table 1), and the causal relationship remains unclear.Further investigation is warranted.A considerable amount of stoma output may potentially delay the diagnosis and treatment of bowel obstruction, so it is clinically important to keep the characteristics of this condition in mind when symptoms of abdominal distension and nausea are encountered in patients with a loop ileostomy.Early diagnosis and intervention, including obtaining a CT scan, is crucial for effective clinical management.
This study has several limitations.First, it had a retrospective design and the sample size was small.Therefore, the possibility of bias cannot be excluded.Second, the fascial incision length was measured on postoperative CT scans rather than intraoperatively, which meant that we could not assess the impact of contractures resulting from scarring.Moving forward, a prospective study is needed to measure the incision lengths of the skin and fascia intraoperatively in order to ascertain the degree of concordance with CT measurements and whether incision length is a relevant factor.

Conclusions
This study has demonstrated associations between OO, PH, and HOS in patients with loop ileostomy and identi ed key factors contributing to their occurrence.Consideration of these complications together could improve our understanding of the complications associated with loop ileostomy and lead to the development of more effective strategies for their prevention and management.

Tables
Tables 1 to 5 are available in the Supplementary Files section.Measurement of stoma size when passing through the rectus abdominis muscle on a computed tomography scan several months after stoma construction when the patient's condition had stabilized.Cumulative sum analysis of the relationship between size of the stoma and the stoma-related complication rate.Horizontal axis represents each case arranged in ascending order of the long diameter of the stoma at the rectus abdominis level.Vertical axis denotes the complication rates of OO and PH (with reference rates set at 20% and 30%, respectively).For each stoma size, the cumulative sum was calculated by adding the difference between the observed complication rate and the reference value to the previous cumulative sum: CUSUM(t) = CUSUM(t-1) + (X(t) -Reference Value), where t represents the size of the stoma, X(t) denotes the observed complication rate at stoma size t, and CUSUM(t) is the cumulative sum at stoma size t.

Figure 2 Results
Figure 2