complications from gastric bypass surgery

[womansurg]

Doc Ollie, Kim, anyone else with experience, share your thoughts of this:

Currently we have a 400 lb woman who blew her j-j anastamosis on day 3 after an open procedure. She went back urgently and underwent primary repair with drainage. Subsequently, she opened up what initially appeared to be a controlled leak, then spiraled into sepsis with multi organ system failure requiring re-exploration.

Enter Dr. womansurg at this point (I wasn't involved in initial 2 surgeries). On exploration, she had the most hostile abdomen I have encountered. She was blocked in with adhesions and inflammation, and it took a good hour to even define interloop planes and release the seared on omentum (naturally, she was hypotensive and intermittently unable to be ventilated as we worked...). Her j-j was again wide open with widespread peritonitis and countless interloop abscesses. The epigastric region harbored a large abscess, and the g-j was disrupted with nasogastric tube lying free in the abdomen. This was the most difficult exposure of my young career. Imagine: the GE junction is difficult access in a virgin belly of normal sized person. Now add 400+ pound morbid obesity, now add massive inflammation, mesenteric foreshortening and tissue which is hugely edematous, friable, and with nearly unrecognizable tissue planes. O my.

We set up the Thompson retractor, my attending reached in and, with his arms shaking violently from the effort, managed to retract just enough that I had about a 2 inch wide window that stretched down to reveal the gastric pouch stump at the bottom of a narrow, black hole. Wearing a headlamp and using 18 inch long instruments, I was just able to see enough to cut back to fresh edges on each stump, and perform a hand sewn, practically esophagojejunal anastamosis (given that so little gastric tissue remained). We then repaired the J-J, by dividing about 10cm on each limb to fresh tissue, and performing two separate anastamoses for the re-establishment of j-j continuity, and then the re-introduction of the hepatopancreatic limb. We brought out a gastric tube from the gastric remnent, placed drains, and got out.

So far, although she remains critically unstable, there is no sign of anastamotic breakdown after one week. Although I would never choose to re-anastamose in the setting of widespread intrabdominal sepsis if I could help it, I'm not sure what other options remain here, as the anastamoses are so very proximal. Someone mentioned treating it as a difficult duodenal stump, placing lateral duodenostomy tube, so forth. In fact, I believe that the GT does decompress this limb effectively (after revision, the jejunal segment was very close to ligament of Trietz), so I don't think that method of decompression would add anything other than additional risk.

I'm curious how you might have seen leak complications managed, whether diversion was ever used and, if so, by what method, and what the outcomes were. Our leak rate at the moment is 4%, consistent with other published series, and we've not had any deaths from any of them (we'll see how this one turns out...). Given that the published mortality of a j-j leak is 50%, I feel that we must be doing something right with our management.

thanks guys,
-ws

---

Members don't see this ad.

 

[Skylizard]

When I worked with Dr. C in academic surgery during my first internship, he routinely placed a drain in all GB patients before closure so to be able to ID leaks early. Your case sounds like a disaster by all accounts! I am not sure what else you could have done short of treating it like a disrupted esophagus and simultaneous bowl rupture and brought out a spit fistula and several stomas in the abdomen with alot of "BADs" (aka= Big Ass Drains).

The fact that she is alive suggests as much luck as skill. It sounds like you might end up with your first mortality🙁 Under these conditions and from my limited experience point of view, I do not think you, WS or your team did anything wrong in the "salvage" operation. It sounds like the surgical plan was drawn up by the emergency and pathology of the patient. She/he drew the approach via their anatomy and pathology in-vivo.

Just my humble opinions,
Loki Skylizard

PS: With the comment per spit fistula, I was not suggesting an esophagectomy, only suggesting the options seem to be
a) re-do anastomosis
or
b) divert everything out and cool her down and then come back 6 plus months later to restore her anatomy as much as possible. Given her disease, the return would suck...though, if she lived, she might be thinner but the adhesions would hurt you where the inflamation once had been.
"Basically pay now play later" vs "play now and pay later"

 

[womansurg]

Generally a g-j leak is treated with drainage, which she has in place. I couldn't see that performing a cervical esophagostomy would add much, as the minimal soilage from salivary secretions is effectively evacuated by the proximal NG tube. So while we considered esophageal exclusion, repair with prophylactic drain placement for the event of a leak seemed reasonable, given the easier management pathway of a g-j leak and the extreme morbidity and mortality which would be associated with future surgical approach.

As I mentioned, the proximal anastomosis was nearly at the ligament of Trietz. There is no way to bring out a five inch long piece of bowel through the foot-and-a-half thick pannus of an anasarcoid 400 lb person in order to create an ostomy. Hence the consideration of lateral duodenostomy tube, but as I pointed out, she decompresses very effectively via her gastric remnent and placement of a duodenal tube in that abdomen would have been technically inadvisable. In fact, we couldn't even approximate the gastic remnent to the anterior abdominal wall due to the inflammation and foreshortened tissues - we had to leave an intervening segment of intraperitoneal tube.

Anyone else? I'm especially interested in others experiences with and management of leaks.

 

[droliver]

I think that gastro-j you redid is going to bite you in the ass. With a delay in recognition with pretty ugly looking tissue, I might have been inclined to treat her like an esophageal rupture that was not identified early - spit fistula, take down her roux-limb, G-tube in the gastric remenant (which is now back in continiuity), and a Jejunostomy tube.


I'm not sure this G-J that you re-did will leak (and being amenable to control with a drain) as much as I'm concerned that it would completely disrupt again

 

[womansurg]

I'm glad to see I'm not the only one who can't spell 'remnant' 🙂 .

Thanks for the replies. I agree, the g-j hook up has been the one I've most fretted over. We're at day 11 now, and we tested the anastomosis a couple of days ago with nothing showing up in the drains, so god willing, maybe this poor girl will finally get a break.

I do like the idea of a J tube. Although the g-j is essentially already diverted from the GI track while the patient is npo (all hepatico and pancreatico drainage goes down the Y limb), it certainly wouldn't hurt to have additional decompression of that anastomosis during healing, and it could contribute to the decompression of the j-j's as well.

So, you would take down the roux, reconnect with the proximal jejunal limb at the ligament of Treitz, re-establish jejunal continuity at the site of the breakdown downstream, then bring a J-tube out in the intervening segment? Not a bad thought. I do think trying to re-approach her surgically to establish esophageal continuity would be a highly morbid procedure, and that factor is what swayed us to attempt primary repair with drainage. So far so good, but I won't rest until her GI track is functioning.

Has anyone actually had a patient they've dealt with by diversion? Looking at the national data, most of these critically ill anastomotic failures die, it seems, so I don't know how aggressive docs have had the opportunity to be in these circumstances. This is the first time we've had failure of primary repair and/or drainage in dealing with a leak.

 

[Winged Scapula]

Can't say I've seen too many of these who've survived for very long. Me thinks you've got a mess on your hands (as if you didn't already know).

However, recently involved in the take back and he was treated as a duo stump leak - as you mention in your first post - and had a duo tube and drain placed - which we watched like hawks (closest I ever came to strangling an attending when he asked, seemingly every hour, what the drain output looked like and then went to check for himself after I told him what it looked like).

I have seen a few G-J diversions done here when the patient was "stable" enough to tolerate such; however, as I noted above, most of them seem to give up the ghost before they reach that point. While I can't quote the standard of practice here amongst our bariatric surgeons, taking down the Roux and placing a feeding J (if not done at the time of surgery, which it often is HERE) and gastric drainage might be the preferred route.

Interesting...and just 1 more reason by GBPs scare me! 😱

 

[droliver]

Womansurg- my thoughts on what I suggested (taking down the bypass with esophagostomy) are influenced by a couple of cases like this I've been involved with. I've done what you did (redo the G-J, drain it, & pray) for most of the leaks after gastric bypass I've done when feasible. Your description of the gastric pouch tissue + (what I'm assuming is) a Jejunal limb under a good deal of tension after the redo is why I was inclined to treat it like GE junction perf that was identified late. They look exactly like what you described. I've done two of these that ended up getting reconstituted later on & did ok. You put the jejunum back in continuity so you decompress via a G-tube & not worry about blowing out your stump on your billiary limb & a j-tube to feed distal.

Did I forget to mention how much I hate taking care of bariatric patients😡

Kimberli- it sounds like you've seen the same thing done which I was proposing. I think the limiting factor of which one you do will depend on how much stomach looks ok on the gastric pouch to work with & the tension on your Jejunal limb for your GJ. Foreshortened mesentary might exclude revising the GJ (the easiest way to fix this)

 

[womansurg]

Very good. There simply isn't a body of literature accumulated yet on dealing with catastrophic anastomotic disruptions in these cases, so it's good to get some anecdotal evidence, at least.

Update: CT abd/pelvis today (done for persistent leukocytosis) showed intact g-j and j-j's, and no abscesses. Also, none of the contrast showed up in the drains, which was probably the best test anyway. She's still critically ill, but showing improvement in several organ systems: cardiovascular (down to 5 mcg/kg/h dopa), pulmonary (lower Peep and Fi02), and hepatic (declining LFTs). Her kidneys are basically kaput.

 

[Skylizard]

WS, how is your lady doing?

 

[womansurg]

Unfortunately, development of severe pancreatitis, accompanied by hepatic insufficiency, hypotension, sepsis...

Not looking very hopeful, I'm very sad to say.

 

[chicagosurgres]

womansurg-
Just wanted to share a little "case report" of mine that will hopefully brighten your spirits. After doing about my 30th RYGB without a leak or complication was very humbled by the following disaster. POD#1 s/p open RYGB on a 25yo F with persistent tachycardia in 130s and subsequent leak on gastrograffin swallow with early sepsis. We re-explored and found a blowout of the stapled G-J anastomosis. After debriding were left with about a 15 or 20 cc gastric pouch and managed to redo the G-J. Convinced attending to put in a lateral duodenostomy tube and a feeding j-tube, and left in two 19Fr JPs. Needless to say after 4 weeks of everything from ARF requiring dialysis to being maxed on Neo and Dopa at one point...she managed to improve. Decannulated her trach, took out the feeding j tube, and even managed to clamp the duodenostomy tube (not brave enough to pull it before being discharged) and then sent her home. My attending said she is doing fine now. Hope this encourages you that you did the right thing for your patient regardless of the outcome.

 

[womansurg]

Thanks so much for the feedback.

My patient actually looks better today - wbc down a bit, enzymes improved, actually making some urine (she's been on dialysis for weeks).

It's amazing what young people can recover from. I'm holding out hope yet.

-ws

 

[droliver]

Just a tip. Don't rely to heavily on those UGI's to diagnose a leak. The majority of the leaks I've seen have all had normal swallows without extravasation. A neat trick that is very sensitive is to give some water with methylene blue for them to drink & see if there's any staining on your JP effluent (if you have a drain in still). Tachycardia has always preceded everything else (pain,fever,WBC) on all the leaks I've done except one who had an unexplained bump in her creatinine but looked great clinically & went on to sick as hell on POD #6 despite doing everything but reexplore her to prove she was leaking (UGI,dye in the drain,CT scan for excess fluid)

 

[womansurg]

You're exactly right - imaging doesn't do justice for picking up the g-j leaks.

We had started elemental trickled tube feeds via NG in my young lady, which we had to stop when the fulminant pancreatits set in. Likely the pancreatitis was secondary to newly added primaxin. She also had the worst case of 'red man syndrome' I've ever seen from an earlier trial with vanco (as if her care was not complex enough - she has to have idiosyncratic reactions to medications...).

The pancreatitis caused severe retroperitoneal hemorrhage and swelling, blocking off her retrocolic g-j limb, which then was associated with identification of extravasation into her drains where none had been before: a leak 🙁 . With cessation of feedings she has stabilized, and actually looks better than she has for a while. We'll have to see what a course of nonoperative management will buy us for this very tenuous repair.

 

[droliver]

I thought a leak was inevitable unfortunately. Hopefully your drain can control your fistula until she gets stable enough to revise things. Just from your thumbnail description, it sounds like she will not survive this last insult. I imagine you'll be having some septic episodes from undrained collections intrabdominally. BTW is her abdomen open @ this point or is her fascia closed?

 

[womansurg]

Her fascia has been open since the first re-exploration. This has been ongoing for about a month now. She keeps experiencing these major setbacks, and everyone (including myself) says, 'well, this is it...surely this will be the end...'. Then she does better - until the next setback. Of course we've had the usual line infections, abscess drainages, ARDS...

Right now she is more stable than she has been since we first did the repairs. I do think the leak is a controlled one. If we have to go back, we would try esophageal exclusion, hopefully with jejunostomy if adequate mobilization is possible. I'm hopeful that we might progress to enteric feeds via her gastric tube after more complete resolution of the pancreatitis. The g-j seems walled off, and may not be a frank anastomotic dissolution, given that she tolerated tube feeds via that route for several days prior to the pancreatitis. She is four weeks s/p repair at this point and didn't show a problem with the anastomisis until the retroperitoneal hemorrhage event.

Well, here's hoping.

 

[pjr]

I'm not entirely sure from your description, but have you been doing off and on enteral feeds, even trickled? If so, I might suggest a long term course of strict NPO/no enteral feeds, TPN and octreotide, i.e. plan no enteral feeds for a month and just go with it. It is a little extreme, but in bariatric disasters like this one, I have unfortunately had the occasion to try it a few times. In the patients that survived, the results were actually pretty good, and they were able to heal their fistulae and go on to eventual po intake.

best of luck to you and your patient...

 

[ussdfiant]

I am having gastric bypass surgery on July 2nd. Even though I am only a first year student and understand a teeny, tiny fraction of this thread, you guys are scaring me!🙁

 

[womansurg]

I'm guessing you must be pretty young to be a first year med student. In your twenties, perhaps?

The vast majority of people undergoing bariatric surgery do fine. Most go home in two or three days. But as you've seen, the complications, when they occur, can be devastating. Surgical correction of obesity is sort of an admitting of defeat that other methods of weight loss have no hope. I'd be pretty reluctant to make that determination in a young person, but only you and your docs know what you've been through before coming to this decision.

I hope you'll keep us updated as to how you're doing. That would make a great educational opportunity for docs at all levels of training.

 

[ussdfiant]

Thanks for the age compliment, womansurg! I am a 32 year old nontraditional student. I will certainly keep you guys informed!

 

[droliver]

womansurg,

what happened to this patient after your last post?

 

[womansurg]

She continued with a progressive multi-organ system dysfunction, had multiple CTs, some perc drains with finding of infected hematoma, one further ill fated re-exploration with findings of concrete abdomen and uncontrollable coagulapathy. The family withdrew support after development of irreversible immune system failure with pancytopenia and coagulopathy requiring ongoing tranfusions and complicated by intracranial hemorrhage.

She died about two weeks ago.