Ghost Stoma Solving the Dilemma of Diversion After Anterior Resection: Surgical Technique Description and Experience from a Single Centre

Abstract

Anastomotic leak is the most dreaded complication after anterior resection (AR). To do prophylactic diversion stoma or not is a matter of constant dilemma that most surgeons face. In such a situation, ghost stoma (GS) technique offers a middle path, wherein unnecessary prophylactic stomas can be avoided and at the same time ensuring that a diversion stoma can be created if need arises without the need of any major surgery or anaesthesia. Retrospective data of patients who underwent anterior resection with ghost stoma at our institute from January 2015 to December 2019 was analysed for clinical characteristics, operative outcomes and postoperative complications. Ghost stoma is fashioned by pulling up a loop of intestine up to parietal peritoneum layer of anterior abdominal wall with the help of silicone or plastic tube after anterior resection. During the study period, this technique was used in 68 patients of which 7 patients required creation of stoma under local anaesthesia for suspected or confirmed post-operative leak. One of these patients developed congestion of stoma for which the stoma was refashioned under regional anaesthesia. There were no major complications like bowel obstruction, strangulation or tube migration. However, two patients had prolonged ileus and were managed conservatively. Thus, unnecessary stoma was avoided in 89% patients of anterior resection. This is a simple, easy to learn, less time-consuming and oncologically safe procedure which can prevent patients from morbidity as well as psychological and financial burden associated with unnecessary stoma.

Introduction

Anastomotic leak is one of the most dreaded complications after anterior resection (AR) especially in cases of distal rectal cancers with leakage rates ranging from 3 to >20% in various studies [1, 2]. Such anastomotic failures can lead to increased perioperative mortality and morbidity, poorer functional outcomes, delays in adjuvant treatment, and decreased oncologic outcomes [3].

To decrease the frequency and severity of such leaks, many surgeons routinely perform diversion stoma after anterior resection. However, the net benefit of diversion remains unclear. While diversion stomas are associated with decreased postoperative morbidity, their role in preventing such leaks remains unclear [4, 5]. Diversion is also associated with significant stoma-related complications, decreased quality of life, need for a second surgery and thus increased healthcare expenditure and utilisation. Thus, whether to do diversion in every case of anterior or low anterior resection (LAR) is a matter of constant debate. In view of the above, another approach has been adopted by us, which minimises the diversion stoma creation in such patients routinely, while simultaneously ensuring that in case of anastomotic leak, a diversion stoma can be safely created without subjecting patient to another major surgery or general anaesthesia. This approach has been described as “ghost stoma”, “ghost ileostomy” or “virtual ileostomy” in literature by authors like Miccini et al. and Sacchi et al. with slight variations in the technique of creation. Ghost stoma (GS) offers a simple yet effective alternative to avoid prophylactic diversion stomas. Hence, this study was undertaken with the aim of assessing the utility of this technique in patients who would usually require a defunctioning stoma and to assess associated complication profile [6, 7].

Material and Methods

Clinical Data

The technique of ghost stoma described below has been practised by the author for over 8 years. Retrospective data of the patients who underwent this procedure between January 2015 and December 2019 was analysed with the aim to assess its utility in avoiding prophylactic stoma and to study associated complication rate. The procedure of “ghost stoma” was performed by the first author or under his direct supervision in all the cases. This technique was used in 68 patients of carcinoma rectum who underwent partial or total mesorectal excision with anterior resection or low anterior resection and had no evidence of anastomosis dehiscence intraoperatively as indicated by triple test. Triple test post colorectal anastomosis being defined as intact suture line, complete doughnuts and no air leak. All cases of AR underwent stapler anastomosis using circular stapler followed by triple test. Patients in whom triple test came positive or in whom suture line was under tension underwent immediate diversion with or without revised anastomosis and have not been included in this analysis. Also, defunctioning stoma was fashioned in all cases who underwent ultra-low anterior resection and hence were not included in the analysis. In the rest of the cases, prophylactic stoma was avoided but ghost stoma was created by hitching a bowel loop (terminal ileum or transverse colon) to abdominal wall. The choice of ileum or colon for fashioning ghost stoma was at surgeon’s discretion and no specific criteria were followed. However, ileum was used for ghost stoma for most of the patients for ease of pulling out ileum loop. Transverse colon was used selectively for patients who had redundant transverse colon. This hitched bowel loop could be fashioned into functional stoma under local anaesthesia if need arises or hitch can be removed on post-operative day 14 once bowel is functioning well and risk of leak is minimal.

Surgical Technique

Ghost stoma technique is done by pulling a loop of intestine up to parietal peritoneal layer of anterior abdominal wall with the help of silicone/plastic tube (e.g. suction drain). If an anastomotic leak is detected in postoperative period, same loop of bowel can be pulled out and fashioned into stoma under local anaesthesia.

Steps are as follows: After performing colorectal anastomosis (Fig. 1)

• Step 1: Select a loop of terminal ileum 10 cm proximal to IC junction. Similarly, a loop of proximal transverse colon can also be selected.

• Step 2: Create a window of 1×1 cm in the mesentery, and pass a plastic tube (suction catheter/Ryles tube). A 12-Fr suction catheter has been used in the case depicted. It should be noted that mesenteric window should be created close to bowel wall so as to avoid any direct injury or compression to mesenteric vessels while passing plastic tube.

• Step 3: Two points are marked on anterior abdominal wall 2–3 cm apart in vertical orientation, at the site of potential stoma. The two ends of the plastic catheter are bought out through abdominal wall.

• Step 4: Now loop of intestine is inspected for orientation. There should not be any twist in the mesentery. Also, make sure that the suction catheter is held loose enough so as not to strangulate or obstruct bowel loop.

• Step 5: Two limbs of plastic catheter are now transfixed to anterior abdominal wall. Abdominal wall closure is done in layers.

Fig. 1.

Ghost stoma technique—stepwise description

The author also has a routine practice of closing pelvic peritoneum after placing drain in pelvic cavity post anterior resection and anastomosis, so even if contamination occurred it remained extraperitoneal and localised to pelvis.

In post-operative period, all patients were monitored for clinical signs and symptoms suggestive of anastomotic leak such as fever, tachycardia, abdominal pain, distention, tenderness, purulent or feculent discharge through the drain or the abdominal wound itself and increased total leucocyte counts. No Gastrografin study or CT scan was done routinely for detecting anastomotic leak. CECT abdomen was done only if leak was clinically suspected but drain was non-feculent to confirm the diagnosis. If an anastomotic leak was suspected or detected, without delay a diversion loop stoma was created under local anaesthesia by incising skin, sheath, muscles and pulling out bowel loop with the help of plastic catheter (Fig. 2). A thorough wash with warm saline through pelvic drain already placed during primary surgery was also given in case of faecal contamination and patients were started on intravenous antibiotics. If patient’s postoperative period remained uneventful, the plastic suction catheter was removed after 2 weeks on outpatient basis.

Fig. 2.

Creation of functional stoma under local anaesthesia

Results

There were a total of 87 cases of anterior resection performed during the period, of which 68 underwent ghost stoma formation and 19 had diversion stoma at the time of primary surgery and hence were excluded. The main reasons for diversion stoma in these 19 patients included ultralow anterior resection in 7 patients, and one had stapler failure following which hand-sewn anastomosis with diversion was done, while three others had positive air insufflation test. Owing to retrospective nature of the study, the exact reason for 8 remaining patients could not be retrieved from operative records. Hence, a total of 68 patients underwent anterior resection with “ghost stoma” technique at our institute from January 2015 to December 2019. Mean age of patients was 51 years with male:female ratio of 1:1. Seventy-two percent of patients had undergone neoadjuvant chemoradiation. The average distance of growth from anal verge was 6.8 cm with range from 5 to 13 cm. Selection of ileal or transverse colon loop was on surgeon’s choice and a terminal ileal loop was hitched in 56 patients while a transverse colonic loop in 12 patients (Table 1).

Table 1.

Demographic profile and other parameters

Age
Range	21–76 yrs
Mean	51 yrs
Sex
Male	35
Female	33
Location of tumour from anal verge
≤5cm	4
>5–10 cm	53
>10 cm	11
Neoadjuvant CTRT
Yes	49
No	19
Type of bowel loop used for hitch
Ileum	56
Colon	12

A total of 7 patients required creation of functional stoma post-operatively for suspected or confirmed leak. Anastomotic leak was detected on post-op days 3–10 (mean 5.5). Three of these patients had feculent content in drain while other 4 had suspicious clinical symptoms like fever, tachycardia and lower abdominal tenderness and underwent CECT abdomen for confirmation of diagnosis. Fashioning of stoma was done under local anaesthesia in all patients and warm saline wash was done through pelvic drain. Two patients had prolonged paralytic ileus requiring Ryles tube reinsertion and had extended hospital stay. One patient had congestion of stoma due to rotation of mesentery while creating diversion stoma under local anaesthesia and required refashioning of stoma under regional anaesthesia. The rest of the patients had the hitch tube removed at around 2 weeks (POD 13–22 days). However, none of the patients required re-laparotomy or peritoneal wash under general anaesthesia.

None of the patients had any major complications due to hitch tube like bowel rotation, obstruction, strangulation, adhesion, stricture or tube migration. One patient, however, had wound infection and later incisional hernia at site of stoma closure and was managed surgically after his adjuvant chemotherapy was over. Average hospital stay for patients who developed anastomotic leak and diversion stoma was 9.9 days (range 8–16) as against 6.7 days (range 5–9 days).

Discussion

Anterior resection (AR) is the standard operation for rectal cancer, and with the emphasis on functional preservation increasing in oncological procedures, rate of low AR and ultralow AR is increasing even more. Anastomotic leak remains the most dreaded complication after anterior resection. Many surgeons perform prophylactic diversion stoma routinely while others do in high-risk cases in the belief that diverting the faecal stream will prevent anastomotic dehiscence.

In a study by Bax et al., the authors have mentioned performance of routine diversion ileostomy in all patients with anastomosis less than 5 cm from anal verge, previous pelvic radiation therapy, obstruction and infection and had low clinically significant leak rate of 3% with asymptomatic, radiologically confirmed leak rate of 2%. However, they did not fail to mention that the diverting ileostomy and its closure has its own set of complications and morbidities including dehydration, difficulty with ileostomy care, stricture, wound infections and incisional hernias. In this study, 23 of 94 patients had difficulty managing stoma and skin irritation while another 12 patients had dehydration episodes requiring hospitalisation. Also, the fact that patients require a second surgery and anaesthesia for closure may question the favouritism for prophylactic diversion stoma [8]. Thus, it can be said that diversion stoma during anterior resection decreases morbidity in immediate post-operative period but it does come with a price. Through the technique mentioned in our study, we try to minimise severity of complications due to anastomotic leak in immediate post-operative period without paying this unnecessary price.

Whether prophylactic diversion stoma decreases leak rates post LAR has been a topic of continuous debate with studies available showing conflicting results. Most RCTs have failed to show significant difference in leak rates between the two groups [9–11]. Most of them concluded that a defunctioning ileostomy has no influence on whether a low rectal anastomosis will leak after surgery or not. Rather, it minimises sequelae of a leak in high-risk patients. It has significant morbidities and should therefore be used judiciously.

On the contrary, few of recent meta-analyses have shown decreased leak rates in patients who undergo diversion stoma but these are limited by the small number of RCTs available, poor methodology quality and low sample size. Moreover, these studies have also mentioned to take into account quality of life implications in the form of degraded physical and psychological wellbeing while making decision for diversion [12–14]. In a prospective longitudinal study by Tsunoda et al., a significant reduction in European Organization for Research and Treatment of Cancer QLQ-CR38 and QLQ-C30 scores after the LAR and diversion stoma procedure was found, indicating reductions in physical and role functioning [15].

Ghost stoma, though, has already been described in literature but is not being widely practised, and hence, data on technique, advantages and associated complications is limited.

In a systemic review by Baloyiannis et al., 11 studies and 554 patients on ghost or virtual stoma were included, among which 7 were retrospective studies, 3 prospective studies with sample size of less than 25 and 1 randomised controlled trial with total 55 patients (both arms combined). Most of these studies were from European countries. They showed anastomotic leak rate of 11.9% with virtual ileostomy conversion rate of 10.46% and virtual ileostomy-associated complication rate was 2.1%. Among the virtual ileostomy-associated complication, one case of twisting, 3 cases of hernia and 4 cases of surgical site infection were reported. Thus, available literature and also the present study confirm the low morbidity and complication rate associated with GS. Most of these studies have used vessel loop or elastic tape for looping the bowel while one had used Prolene [16]. In another retrospective comparative study by Zenger et al. from Turkey, only 3 of 42 patients in the ghost stoma group required conversion to diversion stoma in postoperative period. Thus, unnecessary stoma was avoided in 92% of the patients. Rehospitalisation rates, excluding those for stoma closure, were also significantly higher in the diversion stoma group compared to the ghost stoma group (p=0.01) [17].

Though limited by its retrospective nature and single-arm data, our study is an attempt to share our experience of this simple, cost-effective, oncologically safe technique in the Indian population. Our technique differs slightly from previously mentioned techniques in using a silastic or plastic tube for looping the bowel and keeping it little loose, probably the reason behind no case of post-operative obstruction or strangulation. We though had one patient who developed stoma congestion after the creation of diversion stoma through local site under local anaesthesia. Stoma creation under local anaesthesia has limitation of lack of proper exposure and working space, hence a risk of mesenteric twisting while pulling the bowel loop. This can be minimised by marking the proximal loop with suture while creating ghost stoma as described by Mori et al. [18].

Ghost stoma, though not a novel procedure, is a reliable and easy to learn and adopt technique which can effectively reduce the need for prophylactic stomas. In the present series, unnecessary stoma was avoided in 89% of AR patients. Though the number of patients studied is not huge, the results seem promising. A well-designed prospective randomised control trial with a greater number of patients will provide better analysis of the utility of this procedure and associated complications.

Conclusion

Bowel loop hitch technique is a simple, easy to learn and oncologically safe procedure which can decrease the need of prophylactic diversion stoma in most cases of anterior and low anterior resection. It decreases intraoperative time and also prevents patient from morbidity as well as psychological and financial burden of stoma.

Declarations

Conflict of Interest

The authors declare no competing interests.