Risk factors for postoperative stoma outlet obstruction in ulcerative colitis

Tomoaki Kitahara,Yu Sato,Takashi Oshiro,Rie Matsunaga,Makoto Nagashima,Shinichi Okazumi,Department of Surgery,Toho University Sakura Medical Center,Sakura 285-8741,Chiba,Japan

Abstract BACKGROUND Current medical treatments can achieve remission of ulcerative colitis (UC).Surgery is required when potent drug treatment is ineffective or when colon cancer or high-grade dysplasia develops.The standard procedure is restorative proctocolectomy (RPC) with ileal pouch-anal anastomosis,commonly performed as two- or three-stage RPC with diverting ileostomy.Postoperative stoma outlet obstruction (SOO) is frequent,but the causes are not well known.AIM To identify the risk factors for SOO after stoma surgery in patients with UC.METHODS We retrospectively reviewed the files of 148 consecutive UC patients who underwent surgery with stoma construction.SOO was defined as small bowel obstruction symptoms and intestinal dilatation just below the penetrating part of the stoma on computed tomography.Patients were divided into two groups:Those who developed SOO within 30 d after surgery and those who did not.Patient characteristics,intraoperative parameters,the stoma site,and rectus abdominis muscle thickness were collected.Moreover,we identified the patients who repeatedly developed SOO.Univariate and multivariate analyses were performed to identify risk factors for SOO and recurring SOO.RESULTS Eighty-nine patients who underwent two-stage RPC were included between January 2008 and March 2020.Postoperatively,SOO occurred in 25 (16.9%)patients after a median time of 9 d (range 2-26).Compared to patients without SOO,patients with SOO had a significantly higher rate of malignant tumors or dysplasia (36.0% vs 17.1%,P = 0.032),lower total glucocorticoid dose one month before surgery (0 mg vs 0 mg,P = 0.026),higher preoperative total protein level(6.8 g/dL vs 6.3 g/dL,P = 0.048),higher rate of loop ileostomy (88.0% vs 55.3%,P= 0.002),and higher maximum stoma drainage volume (2300 mL vs 1690 mL,P =0.004).Loop ileostomy (OR = 6.361;95%CI 1.322–30.611;P = 0.021) and maximum stoma drainage volume (OR = 1.000;95%CI 1.000–1.001;P = 0.015) were confirmed as independent risk factors for SOO.Eighteen patients with SOO were treated conservatively without recurrence (sSOO group).Seven (28.0%) patients repeatedly developed SOO (rSOO group) during the observation period.A significant difference was observed in the rectus abdominis muscle thickness between the two groups (sSOO 9.3 mm,rSOO 12.7 mm,P = 0.006).Muscle thickness was confirmed as an independent risk factor for recurring SOO (OR =2.676;95%CI 1.176-4.300;P = 0.008).CONCLUSION In this study,high maximum stoma drainage volume and loop ileostomy are independent risk factors for SOO.Additionally,among patients with a thick rectus abdominis muscle,the risk of SOO recurrence is high.

Key Words:Ileal pouch anal anastomosis;Ileostomy;Loop ileostomy;Proctocolectomy and restorative;Surgical stomas;Total proctocolectomy;Ulcerative colitis

INTRODUCTION

Ulcerative colitis (UC) is a chronic inflammatory bowel disease of unknown cause.The number of patients with UC is increasing in Japan,similar to other countries globally[1,2].Recent advancements in medical treatments have allowed UC patients to enter remission.However,surgery is required when potent drug treatment is ineffective,or when colon cancer or high-grade dysplasia occurs after long-term disease.Previous reports have indicated that 25%-30% of patients with UC require surgery[3-5].The standard procedure for UC is restorative proctocolectomy (RPC) with ileal pouch-anal anastomosis (IPAA)[6].One-stage RPC without an ileostomy may be performed in some cases[7,8],but pouch complications,especially anastomotic leakage,can be life-threatening for UC patients whose immune system is weakened by poor nutrition or the use of glucocorticoids and immunosuppressants.Weston-Petrideset al[9]established that diverting ileostomy reduces the frequency of pouch-related leaks.Therefore,it is common to perform two- or three-stage RPC with diverting ileostomy.Depending on the individual patient’s characteristics,such as age and preoperative activities of daily living,they may be fitted with a permanent stoma without anastomosis[10].

The construction of a stoma may cause various complications.Among them,stoma outlet obstruction (SOO) is often seen after surgery for UC.SOO does not only decrease a patient’s quality of life,but also,if it cannot be treated conservatively,the stoma has to be reversed.However,if there are pouch-related complications,stoma reversal cannot be performed.Therefore,it is clinically important to prevent SOO.However,the causes of SOO are not well understood.The aim of this study was to identify the risk factors for the development of SOO after stoma surgery in patients with UC.

MATERIALS AND METHODS

Patients

We retrospectively reviewed the medical files of 148 consecutive UC patients who underwent surgery with stoma construction at Toho University Sakura Medical Center,Chiba,Japan,between January 2008 and March 2020.

Definition of SOO

SOO was identified based on the following criteria:Small bowel obstruction (SBO)symptoms such as abdominal distension,abdominal pain,or vomiting,and computed tomography (CT) showing intestinal dilatation just below the penetrating part of the stoma site.

In the non-SOO group,17 cases presented with SBO symptoms.CT was not performed,and it was not possible to distinguish between SBO,ileus,and SOO in nine of these patients.Therefore,these cases were included in the non-SOO group based on the definition of SOO in this study.

Based on this definition,the patients were divided into two groups as follows:Those who developed SOO within 30 d after surgery,and those who did not.

Stoma construction methods

First,a stoma site was marked preoperatively based on the Cleveland Clinic standards[11]by the responsible wound ostomy care nurse of the Department of Surgery.An end ileostomy was created at the marked site of the lower-right quadrant from 2008 to September 2012.From October 2012 through to March 2020,a loop ileostomy was created at the marked site of the lower-left quadrant to reduce mesenteric torsion.For loop ileostomy,the part of the ileum to be lifted was selected 30-50 cm orally from the anastomosis.

The marked skin was cut circularly,and the subcutaneous fat around the stoma was removed.Thereafter,the anterior and posterior sheath of the rectus abdominis were incised longitudinally over a length of approximately 3 cm,and the rectus abdominis was split.Finally,the two sheaths of the rectus abdominis and the peritoneum were sutured to reduce the thickness of the muscle layer.The sheath and the peritoneum margins were then fixed to the serosa and muscle layer of the intestine using four stitches.

Variables

Patients’ age,sex,body mass index (BMI),the indication for surgery,Mayo endoscopic score,the extent of disease,disease duration,disease severity,total glucocorticoid dose from onset of UC to surgery,and total glucocorticoid dose one month before surgery were recorded.We also noted the preoperative total protein,albumin,and C-reactive protein levels,white blood cell count,and Onodera-Prognostic Nutritional Index.With regard to the intervention,we recorded the surgical approach,site and type of the stoma,operative time,amount of intraoperative bleeding,and maximum stoma drainage volume per day reported in the inpatient chart.

In this study,the stoma position was changed from right to left during the observation period.Therefore,we examined whether stoma sidedness contributed to the development of SOO.

Furthermore,we retrospectively calculated the thickness of the rectus abdominis muscle and the subcutaneous fat from the skin to the rectus abdominis muscle surface at the umbilical level using the axial view of the preoperative CT scan.

Moreover,we identified the patients who repeatedly developed SOO within the SOO group to identify the risk factors for repeated SOO.

Statistical analysis

Statistical analyses were performed using IBM SPSS Statistics for Windows,version 24(IBM Corp.,Armonk,NY,United States).Chi-square or Fisher's exact tests were used to compare categorical variables,while the Mann-WhitneyUtest was used to compare continuous variables.ORs and 95%CIs were calculated in a multivariate logistic regression analysis.Variables with aPvalue <0.05 in the univariate analysis were included in the multivariate analysis.Pvalues <0.05 in the multivariate analysis were considered statistically significant.

RESULTS

Eighty-nine patients who underwent RPC with IPAA (i.e.,two-stage RPC) were included;one patient underwent colectomy with ileorectal anastomosis,10 patients underwent colectomy with end ileostomy (i.e.,the first stage of three-stage RPC),46 patients underwent proctocolectomy with permanent end ileostomy,and two patients underwent proctectomy with end ileostomy (i.e.,patients in whom the anus could not be preserved in the second stage of two-stage surgery).The patient characteristics,operative details and outcomes,and CT measurements are shown in Tables 1 and 2.

SOO occurred in 25 (16.9%) patients.The median time to primary SOO was 9 d(range:2-26) after surgery.In the non-SOO group,17 cases presented with SBO symptoms.Of these,two were diagnosed with SBO,and six were diagnosed with ileus by CT only.CT was not performed,and it was not possible to distinguish between SBO,ileus,and SOO in nine of these patients.Therefore,these nine cases were included in the non-SOO group based on the definition of SOO in this study.

In 92 of the 148 cases,we were able to inspect the intraperitoneal cavity (such as during second-stage surgery,including stoma reversal) during the follow-up period.In the SOO group,adhesions below the ileostomy were observed in seven cases,but no significant difference was observed compared with the non-SOO group.In addition,there were no cases of ileum torsion around the ileostomy in either group.

Compared to patients without SOO,patients with SOO had a significantly higher rate of malignant tumors or dysplasia (36.0%vs17.1%,P= 0.032),a lower total glucocorticoid dose one month before surgery [0 mg (range 0-765 mg)vs0 mg (range 0-1720 mg),P= 0.026],a higher preoperative total protein level (6.8 g/dLvs6.3 g/dL,P= 0.048),a higher rate of loop ileostomy (88.0%vs55.3%,P= 0.002),and a higher maximum stoma drainage volume (2300 mLvs1690 mL,P= 0.004) in the univariate analysis (Tables 1 and 2).In the multivariate analysis,loop ileostomy (OR = 6.361;95%CI 1.322-30.611;P= 0.021) and maximum stoma drainage volume (OR = 1.000;95%CI 1.000-1.001;P= 0.015) were detected as independent risk factors for SOO(Table3).

Patient characteristics are compared between the two stoma sides with IPAA in Tables 4 and 5.There were significant differences in the BMI,disease duration,preoperative white blood cell counts,and distance from the pouch to the stoma,but no significant differences in SOO development and other complications.

Two different patterns in the clinical course were observed in the 25 patients in the SOO group.Among them,18 did not suffer a recurrence after their obstructive symptoms had been relieved by either insertion of a decompression tube through the stoma or nasogastric tubing and intravenous fluid resuscitation (sSOO group).However,seven (28.0%) had recurring SOO during the observation period (rSOO group).In the univariate analysis,a significant difference in the rectus abdominis muscle thickness was observed between the two groups (sSOO 9.3 mm,rSOO 12.7 mm,P= 0.006) (Tables 6 and 7).Rectus abdominis muscle thickness was an independent factor for recurring SOO in the multivariate analysis (OR = 2.676;95%CI 1.176-4.300;P= 0.008) (Table8).

DISCUSSION

In this retrospective study,we investigated the risk factors for SOO after stoma surgery in patients with UC.High maximum stoma drainage volume and loop ileostomy were independent risk factors for the development of SOO.Moreover,we found that SOO may recur in patients with thick rectus abdominis muscles.

Stoma-related complications include parastomal dermatitis,SBO,stoma prolapse,parastomal hernia,and stoma retraction[12].The incidence of stoma-related complications varies from 39% to 76% in studies[13-15].It has further been reported that inflammatory bowel disease has many stoma-related complications[16].

The procedure for creating a stoma in our study is not a globally standardized method.We found five cases (0.03%) of ileostomy prolapse and five cases (0.03%) of an incisional peristomal hernia during the observation period.There was no impression that there were more stoma-related complications compared to the standardized procedure.In addition,no particular complications were observed during stoma reversal in these patients.

With regard to stoma-related complications,SBO requires special attention as it makes it impossible for the patient to eat,and it sometimes requires re-operation if relief is not obtained with conservative treatment.In SBO,SOO with stenosis justbelow the stoma site can occur.Although this has previously been reported as SBO caused by an ileal fistula after total colectomy for UC[17],the number of non-UC patients with SOO is increasing due to the higher frequency of performing diverting ileostomy in surgery for rectal cancer[18,19].However,there are still no clear diagnostic criteria for SOO,and it is difficult to distinguish SOO from SBO based on the clinical evaluation and symptoms alone.Therefore,CT imaging is required for diagnosis to confirm SBO with stenosis just below the stoma.In this study,we used these signs on CT images as diagnostic criteria.

Table1 Patient characteristics in the compared groups

A few studies have implied that a diverting stoma reduced the leakage risk in surgery for rectal cancer[20,21].Although a similar report has been published in UC[9],other studies found that the leakage risk is not influenced by a stoma in surgery for UC[22,23].Moreover,it has been shown that one-stage RPC is possible in selected patients[7]and that a diverting ileostomy may not be necessary in UC.However,many patients who need surgery for UC receive high-dose glucocorticoids and immunosuppressants and are undernourished.Therefore,they have a high risk ofleakage.Anastomotic leakage may cause a pelvic abscess and pouch failure and can be fatal.For these reasons,we consider a diverting ileostomy necessary in surgery for UC and,thus,perform it routinely at our institution.

Table2 Operative details and outcomes and computed tomography measurements in the compared groups

Table3 Multivariate analysis of risk factors associated with stoma outlet obstruction

The causes of SOO have been reported to include torsion,adhesions around the abdominal wall,or stenosis of the ileum where it penetrates the abdominal wall[18,24].In this study,it was difficult to assess these aspects because CT imaging was not performed for patients who did not have SBO,and we could not compare between the two groups.However,there was no significant difference in the occurrence of adhesions or torsion between the two groups in the cases in which intraperitoneal inspection or CT could be performed.It is unlikely that SOO developed due to stenosis since all our patients had an approximately 3 cm incision at the initiation of stoma surgery.However,if an ileostomy is constructed under the condition of a pneumoperitoneum,it may actually be installed obliquely even if the surgeon intends to create it perpendicular to the abdominal wall due to the bowel loop[25,26].As a result,the gap between the skin and the abdominal wall may cause a shutter mechanism,resulting in stenosis of the ileum within the abdominal wall.Therefore,it has beenrecommended that the pneumoperitoneum should be deflated before constructing the ileostomy[27].Laparoscopic surgery has been reported to be a risk factor for SOO[28],and it is hypothesized that this results from the described technical difficulties during the ileostomy.

Table4 Patients characteristics in relation to stoma sidedness with ileal pouch-anal anastomosis

Previous studies have also found that subcutaneous fat and thickness of the rectus abdominis muscle are risk factors for SOO[19,29].In this study,there was no significant difference in these factors at the umbilicus level between patients with and without SOO.We assume this is because the ileostomy construction technique at our institution entails removing as much subcutaneous fat as possible,and the fascia to the peritoneum is ligated and then sutured to the ileum.Our procedure reduces both the subcutaneous fat and muscle thickness,thus preventing SOO.

Some authors described a cruciate incision of the rectus fascia and a distance from the anastomosis to the stoma of 30 cm or less as risk factors for SOO[28,30].In the patients included in this study,the fascia was incised longitudinally,and the distance from the anastomosis to the stoma was at least 30 cm.Thus,we could not determine theinfluence of these factors.Hisamitsuet al[31]reported that a high-output stoma is a risk factor for SOO and distinguished two types of SOOs:Those with large and those with small stoma drainage volumes.The author describes that high output causes relative stenosis at the stoma site.The large intestine is responsible for 5%-10% of water reabsorption within the intestinal tract.In an ileostomy,the stoma drainage volume increases because this reabsorption from the large intestine is eliminated.Depending on the individual,electrolyte abnormalities or dehydration may occur due to the loss of water.Additionally,the small intestine has an estimated internal pressure of approximately 9-10 cmH2O,which is lower than that of the large intestine[32].Therefore,it is expected that external pressure may easily affect the internal pressure in the ileum,particularly where it enters the abdominal wall.In this situation,because of the large drainage volume,high-output stoma cause passage obstructions in the abdominal wall,and this can result in SOO,even without apparent stenosis.This is believed to be one of the reasons why ileostomy results more frequently in SOOs than colostomy[33].

Table5 Operative details and outcomes and computed tomography measurements in relation to stoma sidedness with ileal pouch-anal anastomosis

A high-output stoma was also an independent risk factor for SOO in our study.Since there was no significant difference in the surgical procedures and rectus abdominis muscle/subcutaneous fat thickness,we consider a relative obstruction of the stoma caused by the high output as one of the causes of SOO.Thus,determining how to control a high-output stoma is important in preventing SOO.Many patients with UC have a history of malnutrition,and their body fluid tends to shift extracellularly and extravascularly after surgical interventions.This results in edema of the mucosa of the small intestine,and further suppression of water reabsorption is thought to cause high-output stoma more easily.To prevent this,it is important to limit postoperative fluid and to administer albumin and diuretics.In some cases,temporary glucocorticoid administration may be used to reduce edema of the intestinal mucosa.

Table6 Patient characteristics in the stoma outlet obstruction group

There were two types of SOO in this study:One in which SOO improved under conservative treatment without subsequent problems,and the other in which SOO recurred.The rectus abdominis was significantly thicker in the rSOO group than in the sSOO group,and this was an independent risk factor for SOO recurrence.In both types,a high-output stoma was a risk factor for SOO,and fluid management was important.However,in patients with a thick rectus abdominis,SOO recurred,even if the high output of the stoma had been controlled.As a result,re-operation was required in six patients at our institution because the patients’ quality of life was markedly reduced by frequent SOO recurrence.Based on the findings from a report,the rectus abdominis thickness is a risk factor for SOO.Constructing the stoma at the lateral edge of the rectus abdominis muscle when the muscle was more than 10 mm thick led to an improved condition and was thus recognized as a preventative method[27].In patients with a rectus abdominis thickness of more than 1 cm,the stoma construction method should be adopted accordingly.

A loop ileostomy was also an independent risk factor for SOO in our patients,similar to the findings from a previous study.Although we could not confirm this,a previous study[30]found that the twisting of the mesentery in a loop ileostomy is greater than in an end ileostomy.Therefore,ensuring correct positioning of the ileum when elevating it is important.An alternative option may be to avoid loop ileostomies altogether.

Table7 Operative details and outcomes and computed tomography measurements in the stoma outlet obstruction group

Table8 Multivariate analysis of the risk factors for repeated stoma outlet obstruction

Recently,an increasing number of studies compared modified two-stage RPC with traditional two-stage RPC[34,35].One of these studies concluded that modified RPC did not influence the risk of an anastomotic leak,which was associated with more severe UC[34].Modified two-stage RPC is a surgical procedure in which subtotal colectomy and end ileostomy are performed in the first stage,and IPAA and stoma reversal are performed in the second stage after a patient’s general condition has improved.With this method,a loop ileostomy,which is a risk factor for SOO,can be avoided.

Our results should be interpreted within the limitations of this study.This was a retrospective study in a small number of patients at a single institution.Eighty-nine patients were included,which may not be sufficient to generalize our findings.Therefore,other risk factors for SOO may exist.The diagnosis of SOO required CT imaging.Thus,clinically suspicious patients who did not undergo CT may have been included in the non-SOO group.Finally,the standard stoma site changed during the study period,which may have affected SOO occurrence due to different twisting of the mesentery.Future prospective studies with a larger sample size are required to clarify the risk factors for SOO.

CONCLUSION

In conclusion,high stoma drainage volume and loop ileostomy were independent risk factors for SOO in our UC patients.In patients with thick rectus abdominis muscles,SOO may recur regardless of initial improvement.Surgeons should be aware of the importance of fluid management and carefully choose the stoma position in patients with thick rectus abdominis muscles to prevent SOO recurrence.Finally,there is a need to develop alternative surgical procedures to avoid loop ileostomy.

ARTICLE HIGHLIGHTS

Research background

The standard procedure for ulcerative colitis (UC) is restorative proctocolectomy with ileal pouch-anal anastomosis,and it is common to perform two- or three-stage restorative proctocolectomy with diverting ileostomy.

Research motivation

Stoma outlet obstruction (SOO) often occurs after surgery for UC but its causes are not well known.

Research objectives

To identify the risk factors for SOO after stoma surgery in patients with UC.

Research methods

A retrospective study of 148 UC patients.Univariate and multivariate analyses were performed to identify risk factors for SOO and recurring SOO.

Research results

SOO occurred in 25 (16.9%) patients.In the multivariate analysis,loop ileostomy (OR =6.361;95%CI 1.322-30.611;P = 0.021) and maximum stoma drainage volume (OR =1.000;95%CI 1.000-1.001;P = 0.015) were found to be independent risk factors for SOO.Among the 25 patients with SOO,seven (28.0%) patients repeatedly developed SOO during the period of observation.Rectus abdominis muscle thickness was an independent risk factor for recurring SOO (OR = 2.676;95%CI 1.176-4.300;P = 0.008).

Research conclusions

High maximum stoma drainage volume and loop ileostomy were independent risk factors for SOO in this study.In patients with a thick rectus abdominis muscle,the risk of SOO recurrence is high.

Research perspectives

Surgeons should be aware of the importance of fluid management and careful selection of the stoma position in patients with thick rectus abdominis muscles to prevent SOO recurrence.Alternative surgical procedures that can avoid loop ileostomy are required.