EVT is a therapeutic option for patients with gastrointestinal perforations and post-surgical leaks. Typically, EVT is used in the upper GI tract and its use is limited due to the need for prolonged hospitalizations. In this video we present a case of a patient with a subacute colorectal anastomotic deficiency who we were able to manage outside of the hospital using EVT. A 73-year-old female initially presented to an outside hospital for vagal or abdominal pain and rectal bleeding. A stigmoidoscopy identified an obstructing rectal mass that was diagnosed as stage 3b rectal adenocarcinoma. Following six months of neoadjuvant chemoradiation therapy, the patient was admitted to an outside hospital with a large bowel obstruction. The patient underwent surgery for management of her large bowel obstructions and rectal tumor. While the initial surgery was planned to be laparoscopic, she required open surgical intervention due to the presence of pelvic adhesions. The patient underwent an open lower anterior section with primary end-to-end colorectal anastomosis with a diverting transverse colostomy. Following the procedure, the patient developed extreme pelvic pain, significant drainage from her midline incision, and the passage of surgical staples per rectum. Approximately two months following her initial surgery, the patient was referred to our hospital for further evaluation and management. Following her evaluation, a CT scan was ordered. This demonstrated the presence of a 7 cm peripherally enhancing presacral fluid collection concerning for an abscess. After this, a gastrographic enema was then obtained. The gastrographic enema demonstrated immediate filling of the abscess cavity with no passage of contrast into the sigmoid colon concerning for complete dehistance of the colorectal anastomosis. Following the completion of these imaging studies, a multidisciplinary team evaluated therapeutic options. Surgical management was thought to be high risk, thus a non-invasive option was needed. IR guided therapy would have involved placement of transgluteal percutaneous drains. While these may be technically feasible, there was concern that such management would be painful and uncomfortable for the patient and would not manage the underlying issue. Endoscopically, we considered options that would manage the sepsis, heal the area of dehistance, and decrease the midline drainage through her fistula. Finally, we wanted to do whatever we could to manage this issue in the outpatient setting. To achieve our goals, we believed the best management strategy would involve endoscopic vacuum therapy. Provided that we wanted to minimize the patient's discomfort and avoid having to keep the patient in the hospital, we looked for alternative ways to perform EVT without having to place the sponge transrectally. As such, we attempted to set up the EVT system using the patient's ostomy. First, we planned to perform a sigmoidoscopy and advance the endoscope to the ostomy. Then using forceps, the endoscope would pull standard suction tubing transrectally using the ostomy. Once the suction tubing was beyond the anus, we would then plan on attaching a surgical sponge to the distal end of the tubing. We would then retract the suction tubing and sponge into the abscess cavity. With this plan in place, we performed the patient's first sigmoidoscopy. Almost immediately upon traversal of the anal canal, the endoscope advances into a massive abscess cavity. Here we could see very little normal tissue, significant friability, and areas of purulence. It is clear from this point that the colorectal anastomosis has completely fallen apart. Initially, it was difficult to identify the opening of the remaining sigmoid colon. However, it becomes visible at the 4 o'clock position on your screen. The endoscope was then advanced into the sigmoid colon until we were able to exit the ostomy itself so that the suction tubing could then be grasped and then be pulled transrectally. To ensure that the tubing was not kinked and was functioning properly, we attached the ostomy end of the tubing to suction and put the distal end in a water bath. Once we confirmed that the suction tubing was functioning, we then worked on shaping a surgical sponge. Once the sponge was modified adequately, it was attached and sutured to the distal end of the suction tubing. The suture was then wrapped around the sponge to make it compact so that it would fit in the cavity. The sponge was then retracted transanally into the abscess. Provided that the suction tubing was coming through the ostomy, we then consulted our wound care nurse and worked to modify the patient's ostomy bags. To do this, we used a universal catheter access port to puncture the ostomy bag from the inside and create a conduit that the suction tubing could then pass through while also creating a tight seal to prevent leakage of stool and other contents. So that the patient could be managed outside the hospital, we obtained a portable wound vacuum that could then be connected to the suction tubing. The vacuum was set at a continuous suction at 125 mmHg. We then brought the patient back for a repeat sigmoidoscopy and sponge exchange 5 days later as an outpatient. Over the course of the patient's care, we utilized numerous tools for sponge removal, including snares and forceps. Once the sponge was removed, we closely evaluated the mucosa to assess for healing before attaching a new sponge to the suction tubing and placing the system back into the abscess. Over the next several weeks, we continued to perform these examinations and the patient did continue to report to us improvement in her pain as well as the output coming into the portable vacuum. Over time, we began to see small improvements in the overall mucosal appearance of the cavity and this did seem to correlate well with the patient's symptoms. On day 35, you can appreciate that there has been some significant difference in the overall size and appearance of the cavity. By day number 60, the patient had noted that her pain had completely gone away, her fistulas drainage had stopped, and on mucosal assessment, the cavity appeared to be healing. Thus, we decided to now place a new sponge and we wanted to see how the area would heal on its own. On day number 90, the abscess cavity appeared greatly improved and the fistulas tract appeared to be healed over. Looking back so that we can compare the side-by-side mucosal appearance from day 90 to day 1, you can see the significant impact of the EVT therapy. During the patient's sigmoidoscopy on day 90, we performed a contrast examination under fluoroscopy. As you can see, we are now filling the distal sigmoid colon. And again, when we compare the patient's examination prior to EVT to her current examination after EVT, we can again appreciate the impact that has happened with our therapy as we are now able to fill the distal sigmoid colon, whereas previously we were only filling the abscess cavity during the gastrograph and enema. In the four months since the patient's last sponge removal, she has required no further interventions for her anastomotic dehiscence. The patient has been able to return to work and her pain is substantially improved and she no longer requires narcotics for pain control. Her surgical team continues to follow her and is now considering further plans regarding reversal of her ostomy. EVT is a well-established strategy for the management of GI perforations, but often is limited to the upper GI tract and requires hospital admissions. In this case, we demonstrated the use of EVT for a subacute colorectal anastomotic dehiscence. By using the patient's transverse ostomy as a conduit, we were able to keep the patient out of the hospital for the duration of her EVT treatment and maintain her comfort as best as possible. EVT should be a minimally invasive option for patients with colorectal dehiscence and rectal perforation.

endoscopic vacuum therapy

colorectal anastomotic deficiency

minimally invasive

rectal adenocarcinoma

fistula drainage