Alexis Grucela, MD
Endometriosis from a colorectal perspective
Advancing the Science and Surgery of Endometriosis
Monday and Tuesday, April 18-19, 2016
The Union Club, New York
Good afternoon, thanks for being here to listen to me. We are going to talk about endometriosis from a colorectal perspective.
Just to give a brief overview I am going to give some background which you probably already know. I will talk about bowel endometriosis and deep infiltrating endometriosis and then the treatment options and what the outcomes data are. I will try to give a little bit of tips and tricks of what I do just to make things more helpful for you and I am happy to answer any questions later.
As you all are probably aware endometriosis is a disorder defined as a presence of endometrial tissue outside of the endometrial cavity. The prevalence is about ten percent and the symptoms can be pelvic pain and infertility. We will get more into the bowel symptoms in a little bit. There is peritoneal, ovarian and deeply infiltrating endometriosis. We are going to focus here today in this talk on bowel endometriosis.
This is really endometriosis on the surface of the bowel, which can involve the small bowel and/or the colon and rectum. The deep infiltrating endometriosis extends more than 5 mm beneath the serosa into the muscularis propria and it is usually multi-focal. This can involve the uterosacral ligaments, the vagina, the rectovaginal septum, the bladder and/or ureters and the bowel.
We have already heard a lot about managing the other aspects so I am going to be elaborating on the bowel aspects here. Obviously surgical excision and complete removal of all the deep infiltrating endometriosis in a single surgical procedure can improve the patient’s quality of life, may improve fertility, however, when it is this extreme there is a high risk of post-operative complications. We are going to focus on minimizing those.
The incidence is about four to 37 percent of women with deep infiltrating endometriosis the most common site is the rectum. This could be the rectovaginal septum and/or the rectosigmoid, which is about 70 to 93 percent of patients who have this. Next is the sigmoid colon, followed by the appendix and the terminal ileum, which is the last part of the small bowel and the cecum. The malignant transformation interestingly of the endometriotic lesions is estimated between .3 and 1 percent of cases.
How do these patients present? Symptoms can be related to constipation, diarrhea, abdominal cramping and bloating, tenesmus and some patients may actually have rectal bleeding with their menstrual cycles. This can, and oftentimes is misdiagnosed as IBS. And these patients can suffer a lot and be sort of taken down this path of IBS management before they actually have their true pathology identified and treated. I am sure by the time sometimes these patients make it to your office they have just been suffering and tearing their hair out dealing with IBS symptoms. Rarely these cause obstruction but in large lesions these can actually cause a bowel obstruction as well. This can be a very tricky situation.
Treatment of deep infiltrating endometriosis with colorectal involvement is difficult and challenging. The medical or conservative management includes NSAIDs, oral contraceptives and hormones, which can suppress the symptoms. However, it is not curative and it is unclear whether it really prevents the progression but we know that discontinuing these can lead to recurrence. Surgery is really the mainstay of treatment here. When it is a multifocal disease we really need a multidisciplinary team approach so this means that the friendly GYN surgeons with us colorectal surgeons and often the GU surgeons will all help each other in the OR to deal with these complex problems.
The previous talk already alluded to the pre-operative assessment but obviously what is really important is imaging. They need an exam, ultrasound which can be transrectal, transvaginal or abdominal and then an MRI of the pelvis I find very helpful. The other thing that you want to consider in these patients is either a flexible sigmoidoscopy or a full prepped colonoscopy, especially in patients who have cyclical bleeding and/or have constipation. I think that this could be valuable. What you may see on colonoscopy depending on how infiltrating the lesions are you actually see this outpouching into the lumen of the bowel. It is kind of subtle and it can be missed and the gastroenterologist if they do not know or do not suspect could think it is a polyp and biopsy it. So it is important that the clinical information and the suspicion be communicated as well.
Before operating on bowel endometriosis we really want to know the precise diagnosis about the presence, the location and the extent of the endometriosis to plan our surgical treatment. Operative considerations, which was mentioned before, really depends on how infiltrating the lesions are versus superficial serosal endometriosis so that you can decide whether the patient needs a resection versus a disc excision versus shaving or locally excising the lesion. Thirty-eight percent of bowel endometriosis has been described as multicentric. This is additional lesions beyond 2 cm from the main lesion and these are usually best treated with resection but it really depends on each clinical case. You have to evaluate each patient differently and come up with a treatment plan hopefully before you go to the operating room that way there are no surprises.
The goal of surgery here is to complete excision of all affected tissues and there is really no best consensus as to how best to achieve this. I think, like I said before, it is really a multidisciplinary approach with the goal of getting all the disease out. We can approach these surgically via minimally invasive surgery, that being robotic or laparoscopic resections or excisions. Or open laparotomy depending on the clinical situation. Your options include resection with primary anastomosis, full thickness disc excision or shaving like I mentioned.
I am just going to show you here – so in a robotic low anterior resection or rectal resection you want to divide distally once you have done your resection distal to the area of concern. This is just the robotic stapler which I think gives you a very nice staple line and you can get very, very low with it. Here you see basically stapling at the pelvic floor at the levators, which is a really nice, low resection and sometimes these lesions, because they are so low, you really need to get very low on the rectum. I think the robot gives you very nice access. I personally do my resections lateral to medial just because I think mobilizing the bowel up off the ureter is the safest way to do it. It is a little bit technically more challenging but I think that it is a little bit safer. If you have any concern of ureteral involvement or there being inflammation of the ureter you can always electively stunt the patient with a temporary ureteral catheters prior to.
This is just – once we divide the rectum distally we can bring the end out and divide the specimen and then this is us sewing in the EEA anvil, which is the end to end circular anastomotic anvil proximally. Then we are going to basically use the Alexis with the cap. We basically put the end back into the abdominal cavity and we are going to place the Alexis cap back on. We can get re-insufflation and then create our anastomosis. Here we are putting the cap back on the Alexis which was a really nice invention because we used to do all these jerry-rig things to put you know Penroses around the trocars but here you see we have re-insufflated the abdominal cavity. We put the EEA stapler up. I prefer trying not to cross my staple lines. There are two different theories; there is you want to get right in the middle across that staple line or I like it anterior because I think the blood supply is from me the rectum is posterior so I try to bring it out and not cross my staple lines but there are two camps way of thinking. You bring the proximal limb down, you marry the two anvils, and then you fire the stapler, withdraw and you check your donuts. Prior to firing you want to make sure – this is just showing that your mesentery is not twisted, that the bowel is straight. Obviously at this point all your disease should have been removed. Rather than doing the traditional leak test with some air I actually fill the pelvis with saline and then I do a flexible sigmoidoscopy at the time. I do think that there has been a little bit of literature out there that number one, you can actually look at the anastomosis or you are visualizing a circle. You are making sure that there is no bleeding. But then you are also if there is an area with a leak you can actually fix it either laparoscopically or robotically at the time of surgery and it can give you a better sense of whether or not you need to divert. I think scoping it just takes an extra second. They have a flexible sigmoidoscopy there in the OR for us and we just take a quick peek. I think it make a big difference.
You can either drain or not drain. I tend to drain when it is pretty low so anything that is lower than about 7 to 8 cm I will drain because you are doing much more of a dissection down below. Diversion wise you can make a hard and fast rule about 5 cm from the inner verge or less. We know that the leak rates are higher the lower you go but these patients are not radiated such as our cancer patients so I do not think there is a hard and fast rule that these patients have to be diverted but obviously a protective diverting ileostomy is a good option especially in a low anastomosis or one you are concerned about. The general rule of thumb though is if you are worried about the anastomosis either redo it or figure out why you are concerned about it and fix it but you never want to leave the OR stressed about an anastomosis so if there is some issue obviously divert the patient. Recurrence rates are about seven to eight percent and the subsequent pregnancy rates or fecundity is about 50 percent from what I saw reported.
Just to give you some data on outcomes, in a retrospective review of 750 patients, now you have to keep in mind this is a single surgeon, but the conversion rates were very low 1.6 percent. The temporary ileostomy rate was about 14.5 percent. The overall morbidity was nine percent with no mortality. The leak rate was quite low with three percent, which I thought was impressive. Fistula rate was two percent, bleeding was low 1.2 percent and about 40 patients required reoperation or 5.5 percent. They concluded that doing this minimally invasive was safe and feasible.
Then, to look at a meta-analysis there were about 49 studies or about 3900 patients. They looked at the procedures done, the histologic data, the complications, outcomes with regard to pain and quality of life, the post-op pregnancy rates and recurrence rates. About 71 percent were resections and anastomosis, about 20 percent did superficial shaving and a full thickness disc excision was done in ten percent of patients.
Complications: obviously the most concerning would be rectovaginal fistula. These were managed with fecal diversion. Three percent had resection and anastomosis and less than one percent had disc excision or shaving. The leak rate was overall 1.5 percent and this was true for resection and anastomosis and also less than one percent for disc excision or shaving. The abscess rate was also less than one percent. I think those are pretty good outcomes.
The histological data in patients undergoing bowel resection about 95 percent penetrated into the muscularis propria and you can see here to visualize the bowel wall. Thirty-seven percent penetrated into the submucosa, six percent into the mucosa and about 20 percent had positive margins.
What does that mean, what do the positive margins mean? They found in this study that, again, it is a meta-analysis; a 3 cm margin was insufficient to achieve an endo-free margin in one third of patients. Full thickness disc excision may result in an incomplete removal of bowel endometriosis. The question is do positive margins and disc excision result in a higher recurrence rate?
I think that you have got to just – this is obviously not a prospective randomized controlled trial so when you look at techniques and recurrence there was a study from Australia with 91 patients. About 71 percent of the patients had disc resections and 27 had segmental resections and one had both. Eighty-eight percent were done laparoscopically and three percent required diversion. They had minimal complications with one ureteral injury with no sequelae after re-implantation, no leaks and no fistulas. I think that is pretty good. About 11 percent required re-intervention for recurrent symptoms. Of these, eight patients or 8.8 percent have recurrent endometriosis. They found no correlation between involved margins on the pathology and need for re-operative surgery.
I think the data is very mixed across the board. The longest follow up that I was able to discern was from a study in Finland which was a retrospective review but nonetheless had 164 patients with 60 percent done laparoscopically and 32 percent done open or via laparotomy and these were all for deep infiltrative endometriosis of the bowel wall.
They did describe a learning curve here so their complication rates decreased from 2004 to 2006 from 27 percent to eight percent. They correlated the size of the nodules resected, so greater than or equal to 4 cm in size resected was significantly associated with major complications. They had a median follow up of 61 months ranging from 16 to 116 months. They found seven percent recurrence rates requiring re-operation, which is consistent with the data and a 47 percent fecundity rate, which is also consistent. They also concluded that minimally invasive surgery is safe and feasible with good long term outcomes and the patients were generally pain free with good fertility.
In summary, deep infiltrating endometriosis with colorectal involvement is a complex and challenging problem. It is usually multifocal and surgical management is usually the primary form of treatment. But complete excision at a single operation should be the goal and it really does require a multidisciplinary approach.
Thank you so much for your attention.