Now, we'll be looking at what we do for esophageal perforations and what we can offer endoscopically. So esophageal perforations, they are life-threatening conditions. They are emergencies. They can lead to sepsis, secondary to contamination of that mediastinum in the pleural spaces with esophageal contents. They can lead to mediastinitis, pleural effusion and empyemas, pneumonia, multi-organ failure, and fistula formation if they are chronic. Overall, mortality rate remains very high, but there is really a significant variation depending on the etiology of the perforation as well as the promptness of treatment. You look at etiology and risk factors, by far the most common cause of esophageal perforation is iatrogenic. They happen during endoscopy, unfortunately, as we all know, that's a risk factor, including during dilations, EMR, ESDs, as well as during endoscopically placed stents. It can also happen during surgeries, either surgeries within the chest cavity where they accidentally nick or cause trauma to the esophagus. And then it can also occur after esophagectomies and occur in anastomotic leaks. In regards to perforation rates for EGDs, they vary widely based on patient to patient, but generally speaking, we know they're higher for male sex that are greater than 70, when we're dilating radiation-induced strictures or malignant strictures, when we're dilating strictures caused by corrosive injury, and when we're performing pneumatic dilation for acupuncture. Other etiologies, there's the spontaneous perforation, otherwise called bore haves. This is secondary to a sudden increase in the endoesophageal pressure combined with a negative intrathoracic pressure during forceful vomiting, trauma, either foreign bodies, blunt penetrating injuries, food impaction, or injections of caustic substances. And then there may also be malignancy-related perforations. Others can cause either frank perforations of the esophagus, it can also frequently cause fistulas connections from the esophagus into the airway. So symptoms of esophageal perforation, patients will typically present with acute onset severe chest pain, they may have dysphagia, they may have fever, they may have subcutaneous emphysema, as well as hematemesis. Symptoms will depend on location and size of perforation. Red flag signs include signs of sepsis, including tachycardia, hypotension, leukocytosis, mediastinitis, or systemic compromise, and subcutaneous emphysema with rapid clinical deterioration warrants urgent intervention. Imaging options to evaluate for esophageal perforation, ideally we're getting some type of PO contrast. So a chest CT abdomen with IV and PO contrast, or as you might hear, a CT esophagram. These will have pertinent findings with pneumo-mediastinum, pleural fusion, pneumothorax, or extra-luminal oral contrast material. And then esophagram, you'll see an extravasation of contrast into the mediastinum typically, and this is considered the gold standard for diagnosis of esophageal perforation by the American College of Radiology. Some limiting rate factor in obtaining esophagrams is oftentimes they are not something that's available 24-7, because in most institutions there needs to be a radiologist there to perform the exam. So I know at my institution, we have it available Monday through Friday from like 8 to 5. So it's a great study, but it isn't necessarily something that's available 24-7. So if you don't have that available to you, a CT chest with PO contrast is also a great option. And I should add on, in some instances it may be most appropriate to get a CT neck, chest, and abdomen, depending on what you're evaluating. And initial management includes hemodynamic stabilization, if you want to make the patient strict in PO, initiate IV fluids, broad-spectrum antibiotics and antifungals to prevent mediastinitis, and then initiation of IV PPI. Depending on the patient's presentation, you may need to include the ICU. And early multidisciplinary involvement improves outcomes. So that typically includes thoracic surgery, GI, IR, as well as our critical care colleagues. So when we're talking about who should get endoscopic management, patients really need to be stable, right? We should not be ever bringing any critically unstable patient down to the endoscopy lab. And then we have to make sure that we have the appropriate availability of endoscopic expertise. Typically, we cannot manage esophageal perforations that are near or above the upper esophageal sphincter. That often requires an ENT consult. And then if we're also concerned that this perforation is spanning from the esophagus into the stomach, spanning that gastroesophageal junction, those are very difficult to, sometimes impossible to manage endoscopically. So if we know or have a very high concern that that perforation is extending from the esophagus into the stomach, we probably should be getting our surgical colleagues involved. If we're not managing these perforations endoscopically, you can always try conservative management, usually with chest tubes, NPO, prolonged NPO, and IV antibiotics, versus taking them to the OR with thoracic surgery. When we look at endoscopic management, we have primary closure options, and then we have options that heal more by secondary intention. Endoscopic clips are great closure options in the acute setting. I think we see these often deployed if there is a perforation noted at the time of endoscopy. Or through the scope clips, these are ideal for smaller perforations that are less than one centimeter, and then there's larger over-the-scope clips for defects that are closer to one to two centimeters. There's a high success rate of closing these, about 85%, when they're small in size, with these clips at the time of perforation. And you also have to think about when we're applying these clips, the perforation has to be amenable to that, and that the edges have to be fairly well-approximated and easy that you can clip them together. The other option for primary closure for perforations is endoscopic suturing. You may hear like overstitch or X-TAC. These are best for management of larger perforations. So if you're getting perforations over the two centimeters, these can be a great option. They can also be a good option with defects that have some irregular edges. Again, usually better in the acute setting, you want to consider the state of surrounding tissue. So for example, if someone has a large degree of ischemia or compromised tissue at an anastomotic defect site, maybe that's not going to be the best tissue to try to suture or close the defect. And this is a highly technical skill, so it's dependent on endoscopist availability and who is around to try to close that with suturing. Those are the primary ones. Other ways to heal esophageal perforations more by secondary intention. One is esophageal stents. So these may be used for benign or malignant esophageal perforations. They can help treat fistulas in some cases, and they can also treat post-surgical anastomotic leaks. You may also see esophageal stents be used in treatment of esophageal structures. These stents are metal stents, they're like a mesh metal, and they have covered plastic on the outside. They are self-expanding, which means over a period of 24 to 72 hours, they literally expand and settle into that esophageal tissue. In some cases, you may utilize a partially covered stent. Partially covered stents are the same concept, they're that mesh metal, but part of that metal is exposed and doesn't have the plastic covering and embeds in the tissue, so those are typically a little more permanent. These stents work by sealing off the area of the perforation. So the stent is placed proximally to the perforation, and it expands and settles in, and then it redirects any GI contents or PO intake to go through the stent rather than down and around into the perforation, so it allows for healing by secondary attention. They can be useful in large perforations when primary closure is not feasible. They're placed endoscopically, usually with the aid of fluoro, but not always, and they're most effective for perforations that do not transverse the gastroesophageal junction, and for perforations that are less than six centimeters in length. They have a very high success rate of placement. They're placed successfully about 91 to 96% of the time, and they have a high clinical success rate of about 81 to 87%, and the clinical success rate in that study was defined as just healing of that perforation. Typically once they're placed, we leave them in place for four to six weeks, and the patient does require another EGD for removal. They're fantastic because they allow for early return to enteral nutrition. Once we place them, I usually do have the patient remain in PO for at least 24 hours until we can get a follow-up esophagram. The reason I don't get an esophagram right away after the stent is placed is, like I mentioned earlier, it expands and settles in over 24 to 72 hours, so I want to make sure I'm not getting it too early before I give that stent the appropriate time span to really expand and get settled into the esophagus. It can significantly reduce hospital stays compared to surgical management. One common complication with stents is stent migration. Typically they migrate distally, about 16% migration rate with these stents. You can apply a stent-mix device or suture the stent into place to help mitigate that. We want to be very careful in placing these stents on proximal esophageal perforations. The higher up and the closer it approaches the UES, the patients can start having discomfort. It can cause a lot of coughing, it can cause a lot of pain in general, but pain with swallowing as well. There's also a risk of that stent that could migrate proximally and occlude the airway, which would be an absolute emergency. And then, same concept, but the other way, when we place that stent more distally and the distal aspect of the stent spans into the stomach, so it spans that gastroesophageal junction, these patients need to have strict reflux precautions. I'll leave them on PPI. I will make sure that they have their head of bud greater than 30 degrees at all times. And then after they eat, they need to be upright for at least three hours because that's basically just a superhighway right there. So fluid can go up and down. And one other aspect to consider with these stents is patients can have pain. I see it more when we're treating strictures with stents, but it can also happen in the setting of perforations. A lot of people can have a decent amount of pain, chest pain, especially within that first 24 hours. Usually, it resolves after 72 hours once that stent is fully expanded. Can we play this video, Sam? Thank you. So this video is looking at an esophageal perforation that occurred during an endoscopy. So this was an iatrogenic cause, as most are. And we're looking here at the perforation. Actually, he came out and then went back in is what happened. So he's going down, evaluating his stomach there. Now he first applies clips, trying to approximate the edges that close this, ultimately applied clips. And then he did find that there was still a contrast extravasation under flora when he placed a stent. And the concept here, too, is that sometimes both are applied. Sometimes we'll stitch, do endoscopic suturing, and then seal that with the stent to ensure we get a proper seal. Okay. And then moving on to the next form of management is this is a newer management technique. There is less data around it, but it's something called endoluminal vacuum-assisted closure. You might hear endoluminal wound vac, sponge therapy, and it takes the concept of wound vacs and it applies it to the esophagus. So the procedure involves placing this polyurethane sponge, it's connected to an NG tube, and you endoscopically place the NG tube right next to that esophageal defect within that endoscopic lumen. You're not placing it through the defect, you're placing it right next to it within the lumen of the esophagus. And then that NG tube is hooked up to suction, 125 suction, 24-7. So the patient wakes up, they have the NG tube in place, and we hook them up to suction. The patient is kept strict in PO during this time, as long as the sponge therapy is in place. And then the sponge is changed roughly every 72 hours. And the typical duration is 17 days is what I found in the literature. At my institution, I've seen this taking up to four to five weeks to heal these perforations. So mentally doing that math, that is something that's hard on patients because this is a lot of repeat EGDs, a lot of repeat anesthesia events. It's also pretty labor intensive for the GI team too, bringing someone back to the lab that frequently to change out these sponges. We like to use them in large or persistent esophageal perforations, occasionally fistulas where traditional stunting methods may not be effective. Oftentimes we will try, in our mind, we'll try to stent first. And if it doesn't look like it will be amenable to stenting due to multiple defects, due to the way the surrounding tissue looks, or if there's like a dilated upstream esophagus, then we will move on to placing the sponge. They are particularly effective in managing post-operative anastomotic leaks, especially in the setting of esophagectomies. Complications from the sponge therapy itself include spontaneous sponge migration or dislodgement. If that occurs, then typically we'll have to urgently repeat that EGD. And then bleeding in the esophagus, as well as the nares, and I've seen some pretty good nosebleeds on our patients from the NG tube itself, especially if they're on anticoagulation, if they're like being maintained on a hyperdrip. There's a wide range of support reported success rates where studies are required, but they range anywhere from 89 to 96% in the literature. And have your patient communication is key because this is a really difficult set of procedures. This is a really difficult way of healing these esophageal perforations on patients. They have to be strict in PO, so they're either on TPN or getting J tube feeds, but oftentimes that means no sips, chips, that means absolutely nothing going in their mouth. They can have pain in their nose from the tube and the throat from their tube. They're coming to lab every three to five days to get the sponge changed. They're typically hospitalized for this entire duration. So patients can get really down and they can get, it's tough on them. It's really tough on them. And when I'm consenting them, I don't paint any rosy pictures for them. I let them know that this is a long and arduous treatment method, but we see really, really, really incredible success rate with them. So it's keeping the patient from having a possible prolonged NPO conservative management. It's also keeping them from having to go back to the OR, which would have really significant morbidity and mortality. So, and oftentimes too, when we're talking about if we need to place a sponge, if we need to place a stent, this is also conversations that are having with our surgical team in tandem and everyone's kind of contributing to these conversations. All right. I have a couple of case studies for us to kind of work through what we just heard. So Jimmy is a 67 year old. He has past medical history of hypertension, diabetes, and esophageal adenocarcinoma. He's post-op day eight from a minimally invasive esophagectomy with J2 placement, who overnight developed acute tachycardia, hypotension. He was febrile to 38.5 and a new leukocytosis. They obtained a chest CT with IV and PO contrast and findings are notable for pneumomedia stinum, a large pleural effusion and extravasation of olfactory contrast. And then we have some polling questions. Patient is made strict NPO and a broad spectrum antibiotics are initiated. What is the next most appropriate step in the care of this patient? A would be to get a stat EGD for evaluation and management of an anastomotic perforation. B emergent return to OR for closure. C surgical ICU consult, IR consult and GI consult or D conservative management. No further intervention at this time. Fabulous. Yeah. So I'm, I'm actually really glad. So the answer is C, but I was trying to trick you. We do have a diagnosed perforation. There's extra luminal contrast, but as I mentioned earlier, he's hypotensive. He has this large pleural effusion and I am not ready to bring him down to our endoscopy lab. He would probably not do great with the anesthesia. I think we should try to stabilize this patient before we go go on to any, any more aggressive measures within our lab. So as I mentioned earlier, the early involvement of multi-disciplinary team is really important for these patients to improve clinical outcomes. So he is escalated to the ICU. IR puts in a chest tube into that large pleural effusion and he's clinically stabilizes. So we go and see him because we were consulted and he endorses no chest pain, no belly pain. He's not having any nausea or vomiting. He actually says, since this chest tube went in, I feel better. You know, there's no notable findings on exam. You don't appreciate any prepotence. And you start discussing endoscopic options for healing Jimmy's anastomotic perforation. What closure options are viable for Jimmy? Endoluminal vacuum closure, esophageal stent, esophageal suturing, or all of the above? Perfect. Yeah, all the above. And oftentimes we don't know what we are going to treat the patient with before we go in. So we might have an idea, but all we know right now is that there's a perforation, there's extravasation of contrast. But I don't know how big it is. I don't know what it'll look like. So when I go and they do these consults, I take the time, I sit with the patient and I say, these are all the options we have. We are going to do what we think is clinically in your best interest. You know, there are certain cases when I end up talking about the endoluminal wound back and they say, I don't want that no matter what. So of course we would respect that. But it's a longer consult when we're talking about bringing him down to the lab and going over exactly all of our treatment options. Okay. And then we perform the EGD. And we see this large amount of anastomotic dehiscence with multiple areas of perforation. So you can see the dehiscence in pictures one and two here at 40 in that lower part of your screen. This image number four, there was another defect over at 38 and then some pretty significant defects at 36 centimeters. So due to the extent of the defects, they made a decision to place the endoluminal sponge, the endoluminal wound vac. Picture six is the picture, what that looks like from the top of the sponge. And then the number seven, you can kind of appreciate them on x-ray, which can be helpful if you're concerned they're migrated. It's nice to have a floral picture right after they're placed because it helps in the future if there's a question of if the sponge has migrated proximally or distally. Jimmy has kept strict NPO. He utilizes his J tube for P feeds. And during the endoluminal vacuum therapy, he ultimately did require five weeks of a sponge. This gentleman, this case was on our service over the holidays. But ultimately, we were able to resolve the defect after removal of the last sponge and esophogram was obtained and that leak was negative for a leak and then PO intake was initiated for him. This is what the esophagus looks like after the sponges are removed. You can see that there's definitely these obvious mucosal changes. I will also note that after any type of treatment for esophageal perforation, stenting, sponging, suturing, patients are at a higher risk for esophageal stricture down the road. Okay. So Jimmy got better with some time. We'll do one more case study. Sarah is a 25-year-old. She has no significant past medical history, and she presents to a small rural ED with history of severe persistent nausea and vomiting for the last eight hours after a night of binge drinking. She reports her last episode of emesis contained bright red blood, and she subsequently developed severe chest and neck pain. Vitals are notable for tachycardia to 105. She's afebrile, on room air. Her blood pressure is within normal limits. Her BMP is normal. Her CBC has a mild leukocytosis to 15. The ED performs a chest CT without oral contrast, but it is remarkable for a pneumomedia sinum with concern for esophageal wall defect. They start Sarah on broad-spectrum antibiotics. She's emergently transferred to your center for higher level of care for management of esophageal perforation and bore haves. Upon arrival, Sarah remains hemodynamically stable. Her tachycardia has resolved, and she's on room air. Her leukocytosis has normalized. So you're consulted when she gets there. We go to see her, and she's still looking really good. On exam, she's well-appearing. You appreciate some mild crepitus on her interior chest wall, but otherwise the exam is benign. What would you like to do next for Sarah? Would you like to get an EGD? Would you obtain a surgical concept for operative management, obtain further imaging, or conservative management with NPO and continued antibiotics with close monitoring? Okay, pretty split. Great. So we don't have an oral contrast study yet. I want to emphasize how important it is for us to try to get some type of oral contrast study before we jump to anything. So I would recommend the CT at the outside hospital was without oral contrast. So I would recommend going forward and getting some type of imaging. I think that's my next question, so hold on. Yes. What do we want? So what type of imaging would you want? Now that I kind of gave it away. Would you get a chest x-ray, an esophogram, a CT chest with PO contrast, or B or C? Yeah, yes. So we want any type of PO contrast would be acceptable. Ideally an esophogram if they're able to, but it shouldn't delay care if we can't get an esophogram. So a CT chest with PO contrast is also perfectly acceptable. But PO contrasted studies, esophogram or chest CT are really important before we go in. So we have a good idea of what to expect and if there is an ongoing perforation. Okay, so the esophogram is obtained and it demonstrates in the read it says a two centimeter mucosal tear in the distal esophagus, but there is no contrast extravasation. What should we do next? EGD, surgical management, or conservative? Perfect. So the answer is C, conservative management. I included this case because it's a little bit tricky. So there was no contrast extravasation. Esophograms are considered to be our gold standard. So if there's no contrast extravasation, there's no ongoing leak. So if there's no ongoing leak, what would we go and treat with an EGD? That mucosal tear, we would expect to heal on its own. She's an otherwise healthy young lady. She's clinically stable. So we're also need to look at the overall picture too. So the esophogram has no leak. She is a young woman. She is clinically stable to possibly improving her tachycardia resolved by the time she got to us, her leukocytosis is improving. I think with her, we would expect that she probably did have a small spontaneous perforation because she did have that crepitus. There was the free air on the CT, but it likely spontaneously healed and requires no other further intervention from us. So for patients with this, typically we recommend antibiotics for seven to 14 days. Sometimes infectious disease will get involved to kind of help guide that therapy, initiate the patient on clear liquids, at least while they're admitted, usually like to initiate them and watch them for a day or so. And if they do okay with clears for a few days, you do a really slow advancement of diet with clears for a few days, full liquids for a few days, and then soft foods, and then slowly advanced to regular. And then with that mucosal tear, you'd probably want to consider putting her on a PPI therapy for at least four to six weeks. Okay. So esophageal perforation take-home points, early recognition and stabilization are critical. Endoscopic approaches are safe and effective and are increasingly becoming first-line therapy. Multidisciplinary approaches improve patient outcomes, and we have multiple endoscopic options available to patients, including primary closure options with endoscopic lip placement and endoscopic suturing. We also have esophageal stents and then endoluminal vacuum there. That's all I have. Thank you.

esophageal perforations

iatrogenic causes

endoscopic management

multidisciplinary treatment

imaging diagnosis

conservative management