Wednesday, February 20, 2013


This retrospective review from the Archives sheds some light on every surgeon's worst nightmare when it comes to rectal surgery: the anastomotic leak.    This study is a retrospective review of over 70,000 low anterior resections, nationwide, from the years 2006-2009. 
Results The AL rate was 13.68%. The AL group had higher mortality vs the non-AL group (1.78% vs 0.74%). Hospital length of stay and cost were significantly higher in the AL group. Laparoscopic and open resections with a diverting stoma had a higher incidence of AL than those without a stoma (15.97% vs 13.25%). Multivariate analysis revealed that weight loss and malnutrition, fluid and electrolyte disorders, male sex, and stoma placement were associated with a higher risk of AL. The use of laparoscopy was associated with a lower risk of AL. Postoperative ileus, wound infection, respiratory/renal failure, urinary tract infection, pneumonia, deep vein thrombosis, and myocardial infarction were independently associated with AL.

An anastomotic leak is truly a malevolent, disruptive event.  Foremost, the patient suffers.  Sometimes you can get away with treating a small leak non-operatively, with drains, bowel rest and antibiotics.  But more often than not, the patient has to go back to the OR emergently, under less than ideal circumstances.  Most commonly, they end up with a colostomy or some form of intestinal diversion.  The abdominal sepsis that results can set off a chain reaction of pathophysiology leading to multiple organ failure and even death.  And if the unfortunate soul survives the acute leak, then, if they want to get put back together again, they would have to contemplate undergoing a third surgery to reverse the stoma, and with that, all the accompanying risks of another leak, another take back, etc etc ad infinitum. 

When a patient has a leak, it also is psychologically traumatic to the operating surgeon.  All the statistical and professional reassurances in the world (i.e. that leaks happen, no one is immune, technically you wouldn't have done anything differently) are little solace when you have to walk into an ICU room and see a patient, previously healthy and vibrant, reduced to an ashen gray shell of vitality, tubes coming out from everywhere and stool filling up an unanticipated colostomy.  Every day on rounds is like re-enacting your own personal failure.  Every day you have to look the patient in the eye, a wife, his kids, and tell them things will get better, that you are sorry events turned out the way they did.  You go home and feel guilty that it's not you trapped in that hospital bed, that you are the one who can play with your kids, take your wife out to dinner, and get up and go to work in the morning.  Everyday you see his name on your patient list and its like a scab getting scraped off all over again.   

Papers like this ought not to serve simply as a salve to apply to our wounded sense of professional competence.  They ought to instead prompt us to ask why leaks happen so often.  This paper quotes a 14% figure.  When you review the literature, you'll find leak rates ranging from 5-18%.  Either way, that's just a shit load of pelvic anastomotic problems.

Pelvic leaks generally can be attributed to three factors:  too much tension at the connection, poor blood supply, and poor protoplasm.  Tension is something the surgeon has to be able to recognize at the time of the operation.  When you perform a colo-rectal anastomosis, you have to mobilize the descending colon/splenic flexure enough to allow for a tension-free drop down of proximal stump to the rectum.  If you find yourself yanking on the proximal end or putting the patient severe steep reverse Trendelenburg to get it to reach, the patient is trying to tell you to "go get some more length".

  Poor blood supply is a trickier one.  Certainly, frank ischemia of the proximal descending stump is hard to miss.  No one I know would try to plug a bluish colon into a rectal stump.  But sub-clinical ischemia, where the collateralization from the marginal artery is perhaps just enough to maintain serosal perfusion, is dicier to predict and determine.  I sometimes wonder, especially in elderly patients, or patients with known atherosclerosis, if would be a good idea to study the arterial arcades feeding the left colon with a preoperative CT angiogram of the mesenteric vessels.  In oncologic resections, we generally take the IMA at its origin and so the entire descending colon must rely on collaterals coming over from the middle colic pedicle. 

The third factor requires a minimum of functioning brain cells in the operating surgeon.  If you make a habit of trying to re-anastomose brittle, malnourished patients, or patients presenting with frank pelvic sepsis and gross contamination, then you're going to find yourself presenting rather frequently at hospital M&M meetings.    


Anonymous said...

This is an excellent post and brings up every general surgeon's worst fear (or what should be their worst fear!). I myself have seen young, healthy patients go south, and even die, after a leak from a "simple" colostomy take-down. One of my attendings stressed to me that whenever there is a leak, it is the surgeon's fault, either in clinical judgement or in technique. He directed me to this helpful article for staples vs. hand-sewn for various situations that I think you may appreciate:

Also, I agree with you about the difficulty of assessing intestinal blood supply in certain situations. At my hospital, we recently started experimenting with using Laser Flourescense Angiography to evaluate the bowel, and my colorectal attendings are really harping on its merits. Do you have any opinion on that? There is a good paragraph devoted to that technique here

Thanks, and I look forward to your next post!

Jeffrey Parks MD FACS said...

thank you for the links. very interesting. I have thought about woods lamp evals, but worried about the lack of reproducibility of results from one patient to the next. I wonder what the cost of the deviec that quantifies the intensity of fluoroscein perfusion would be...