Author: David S. Morris, M.D.
Small bowel obstruction (SBO) is a very common problem for patients; treated by many different types of medical providers, including primary care, emergency medicine, internal medicine, and surgery. In the United States alone, there are an estimated 300,000 laparotomies performed annually for SBO, and about one third of these obstructions are complicated by intestinal ischemia, with significantly higher morbidity and mortality resulting.
So every patient who shows up with a bowel obstruction should have surgery, to prevent ischemic complications, right? Actually, more than half of patients who present with SBO resolve with nonoperative therapy. Because the most common cause of SBO is post-operative adhesions, and each subsequent exploration carries a higher risk of unintentional damage to the intestine, surgical exploration therefore should be reserved for patients who absolutely need it.
So who needs an operation? Does it take years of surgical experience to know? What goes into the decision? First off, the traditional terminology used to classify bowel obstruction probably should be discussed. And then thrown out the window. Let me explain. During the past century or so, bowel obstructions have been dogmatically categorized as either partial or complete. But what does this mean, exactly? Partial bowel obstruction describes a patient who has dilated bowel on imaging, has nausea and vomiting, but continues to pass flatus or even stool intermittently. A complete obstruction has all the same signs and symptoms except for passage of flatus or stool. So the difference basically boils down to obstipation.
Now for the throwing out the window. Obstipation, which has been arbitrarily defined as no flatus or stool for more than 24 hours, is a very imprecise concept. When does the clock start on the 24 hours? And who really knows with absolute certainty when the last time flatus was passed (personally, I certainly don’t keep track that closely, do you). And if a patient has an obstruction associated with ischemia, why does it matter whether there is anything passing out the rectum?
A more useful way of thinking of SBO is operative vs. nonoperative. In the operative category, I would place the obstructions that are strangulated (by adhesions, hernia, or volvulus) as well as those that will not resolve with nonoperative therapy. Nonoperative obstructions are not strangulated, and will resolve without surgery.
Let me say a quick word about the pathophysiology of bowel ischemia in the setting of bowel obstruction. On a very basic level, and obstruction that results in, or is caused by, twisting of the mesenteric vasculature (volvulus, internal hernia, or adhesion-related mesenteric restriction) can, and often does, result in ischemic bowel. Patients with these types of problems are not difficult to identify clinically in most cases; these are the patients that present with excruciating pain, peritonitis, and possibly septic shock if perforation has already occurred. Diagnostically, they are the simplest patients – they need an operation or they will die.
Ischemia may also develop in a more insidious way. As an obstruction progresses, the bowel proximal to the point of obstruction becomes progressively more dilated over time as the succus and GI secretions accumulate. Healthy bowel can tolerate an impressive amount of dilation provided that such dilation occurs gradually. If the bowel is not healthy (history of IBD, prior surgery, radiation, etc.), or if the dilation happens quickly, the luminal pressures exceed the perfusion pressure of the intestinal capillary network, and blood flow stops. If effective decompression (either with surgery or with NG tube) does not occur promptly (typically around 6 hours or so), the ischemia may be irreversible, and the chance of intestinal perforation increases dramatically.
This latter type of bowel ischemia is the reason behind the traditional dictum in surgical training to “Never let the sun rise or set on a complete bowel obstruction.” If you wait too long, bad things can happen. But how reliably can a surgeon categorize bowel obstructions as complete or partial? It turns out, not very reliably, even when the surgeon has many years of experience. In one study, more than 50% of bowel obstructions with ischemia were misdiagnosed. Ouch!. An experienced surgeon is worse than a coin flip. That hurts on a deep level.
Fortunately, recent work done by my colleagues here at Mayo has improved our ability to identify ischemic obstructions. Dr. Martin Zielinski led a group of providers to create and prospectively validate a model to predict ischemic bowel obstruction. The team showed that three signs are associated with an ischemic obstruction: obstipation (no flatus or stool for 24 hours), mesenteric edema on CT scan, and absence of small bowel feces sign on CT.
Mesenteric edema (arrow) and small bowel feces sign (*)
If you don’t look at a lot of CT scans, you might be asking yourself, “What in the world is ‘small bowel feces sign?’” Well, trust me when I tell you that the appearance of the stuff in that loop of dilated bowel on that image looks exactly like the stuff inside the colon (i.e. poop), which you can see on the patient’s right side in the image above. Small bowel feces sign results from gradual trapping of fibrous material while allowing fluid to pass and be reabsorbed, and thus represents a more chronic process.
If a patient does not have any of the three Zielinski signs of ischemic obstruction, he or she might have a chance of resolving without surgery. Great! Problem solved! Admit them to the floor with an NG tube and IV fluids, and high fives all around, right?
The problem now becomes differentiating the patients who will resolve with nonoperative therapy from those who won’t. In many hospitals there are lots of patients sitting around in therapeutic limbo as the team takes a “wait and see” approach. Usually, after a few days or sometimes many days (this number is highly variable, depending on the surgeon, the patient, the day of the week, if it's a holiday, the phases of the moon, etc. etc.), the patient is either getting better or is declared a failure of nonoperative therapy and gets a laparotomy. Studies seem to indicate that if the patient doesn’t resolve in less than three days, he or she probably needs an operation.
Is this the best modern medicine can do? Guess what -- There is a better way. At the suggestion of a surgical resident, we investigated the use of soluble enteric contrast as a diagnostic adjunct in the management of small bowel obstruction. The “Gastrograffin® (GG) Challenge” became our standard of care for the patients who did not have ischemic SBO. Our protocol is as follows:
- Rule out ischemic obstruction (see “Zielinski signs” above)
- NG suction for at least 2 hours
- 100 ml of GG mixed in 50 ml of water and flushed down the NG tube.
- NG clamping x 8 hours
- Abdominal X ray at 8 hours.
If contrast is seen in the colon on X-ray, or if the patient passes flatus or stool while the NG is clamped, this is a “pass” and the NG is removed and diet is advanced. Hospital discharge typically happens sometime in the subsequent 24 hours. If no contrast is seen in the colon, or the patient does not tolerate 8 hours of NG clamping, or has recurrent nausea and vomiting with diet advancement, this is a “fail” and the patient gets an operation. We have had great success using this protocol, and have seen our length of stay improve significantly, because one way or another, the team knows which direction the patient is headed in only a few hours.
Let me finish with a few pearls for the non-surgeon, since I realize that most of the readers here are not my knife-wielding colleagues.
- If you think the patient has a bowel obstruction, go ahead and place a nasogastric tube. Placing a tube won’t make the patients like you, but by letting them stay bloated, and vomiting, and letting the ischemic cascade continue by not decompressing the bowel, you’re not doing the patients any favors. And don’t place the small-bore, soft silastic NG tube. They don’t work. The stuff that needs to come out is thick and nasty – like pond scum, in most cases – the small tube just will not do the job. The only thing worse than placing a large, stiff, NG tube is explaining to the patient that the soft “nice” tube that was placed initially is ineffective and that a large stiff tube is needed anyway.
- Patients with SBO can sequester liters and liters of fluid in the lumen of the bowel. Start some isotonic fluids. Run them in surgical doses, even for the dialysis patients and the heart failure patients. Intravascular depletion leads to poor perfusion, leads to bowel ischemia. We can deal with excess fluid when the obstruction is resolved.
- Make the patient NPO. Strict NPO. I’ve had patients drink gallons of ice chips while sitting in the ED. I’ve seen consults on the medicine floor where a patient is eating a cheeseburger around the NG tube. Remember the pond scum. Don’t add to it.
- Get a surgical opinion early on. My own (admittedly arrogant, heavily-biased) opinion is that SBO is a surgical disease, and should be managed by surgeons, even when operative therapy isn’t needed (cue the onslaught of hate mail, car bombs, and snarky comments from my surgical colleagues who disagree).
- Khasawneh MA, Ugarte ML, Srvantstian B, Dozois EJ, Bannon MP,Zielinski MD. Role of gastrografin challenge in early postoperative small bowel obstruction. J Gastrointest Surg. 2014 Feb; 18(2):363-8. Epub 2013 Oct 29.
- Zielinski MD, Eiken PW, Heller SF, Lohse CM, Huebner M, Sarr MG, Bannon MP. Prospective, observational validation of a multivariate small-bowel obstruction model to predict the need for operative intervention. J Am Coll Surg. 2011 Jun; 212(6):1068-76. Epub 2011 Mar 31.
- Bickell NA, Federman AD, Aufses AH Jr. Influence of time on risk of bowel resection in complete small bowel obstruction. J Am Coll Surg. 2005 Dec;201(6):847-54.
- Sarr MG, Bulkley GB, Zuidema GD. Preoperative recognition of intestinal strangulation obstruction. Prospective evaluation of diagnostic capability. Am J Surg. 1983 Jan;145(1):176-82.