Retrograde Intussusception (RINT): One Group’s Experience and Ideas

Introduction

Retrograde intussusception, known in our bariatric office as RINT, is also called reverse intussusception or antiperistaltic intussusception. Retrograde describes the direction the bowel intussuscepts—from distalto proximal (Figure 1). The much more common is operistaltic or antegrade intussusception, where the bowel intussuscepts from proximal to distal, is seen in children and adults. Antegrade intussusception is usually associated with a lead point, such as a pyerspatch or small bowel tumor that gets dragged in the direction of normal peristalsis. RINT does not appear to have an anatomic lead point. RINT is unique to Roux-en-Y anatomy and almost exclusively involves the jejunojejunostomy or near vicinity regardless ofthe use of the Roux-en-Y, such as with gastric bypass or biliary reconstruction. The diagnosisis made on computed tomography (CT) scan (Figure 2) with the classic “target” sign at the jejunojejunostomy, along with symptoms of nausea, vomiting, and epigastric and periumbilical or abdominal pain on the left side made worse with eating.

RINT is reported as being very rare with an unknown frequency of occurrence. Prior to our report at the 2007 ASMBS meeting in San Diego, we found only 21 patients reported in the literature at that time, and some cases may have been missed. We reported on 23 patients from our practice at that time, which was subsequently published in March 2008.[1] At the time of this writing, we have now seen and treated 42 patients with this problem and have read in the literature a number of other reports[2-8] from other institutions. The reason for this increase is unclear, but we believe that it has to do with the fact that there are now hundreds of thousands of patients that have had a Roux-en-Y (mostly for gastric bypass) over the past decade, placing a larger patient population at risk for developing this problem. At our office, we follow over 16,000 gastric bypass patients on whom we have performed the primary procedure and we are referred patients from throughout the Intermountain area who have gastric bypass-related problems. This may explain why we have such a large experience with this unusual problem.

Personal Experience
Our first experience with RINT was in 2001 when Sherman Smith, MD, came into the operating room where Dr. Simper was operating and asked for assistance on another case to follow. The patient had a small bowel obstruction several years after her gastric bypass surgery. After Smith, our senior partner, Charles Edwards, MD, who was assisting, said, “I’ll bet she has a reverse intussusception.” Although I (Dr. Simper) had just joined the group, I had been a very busy surgeon for over 15 years and had never heard of such a thing, much less seen a RINT. Sure enough, that is what the patient had, and it required resection as the bowel had already infarcted due to edema and ischemia. This case was fresh in the surgeons’ minds when one author attended one of Dr. Phil Schauer’s laparoscopic gastric bypass courses at the University of Pittsburgh days later. During one of the lolls in action in the live case presentation, the author asked the panel of experts if they had any experience with RINT. They looked puzzled and quickly stated that they had no experience and changed the subject. Since then, all of the experts present that day now have had experience with RINT or have at least heard of it. This experience puzzled me and raised my interest because RINT appeared to be so rare that the experts knew nothing about it, but my senior partner could diagnosis it upon just hearing a patient’s history.
Over the next few years I kept track of our new RINT patients and started to note the increasing number in our practice. My partners and I often discussed (and sometimes argued) what in fact was the best treatment for a patient with RINT. We looked up all the patients we have had with RINT and reviewed their charts to determine the outcomes of their treatment in hopes of resolving this conflict, and presented our findings at the 2007 ASMBS meeting. The outcomes review created as many questions as it answered and we have continued to collect data. We now have an Institutional Review Board (IRB)-approved protocol to treat these patients. While I have not completely reviewed all 42 patients at this time, on preliminary review the data appear similar to that seen in the first 23 patients we published in 2008.[1]

To review that data, all of our patients were female; this is in keeping with other reports that have been almost exclusively female. In our 2008 report, 83 percent had open surgery and 17 percent had laparoscopic surgery for their gastric bypass. This ratio has changed as we now are seeing mostly patients who have had primarily laparoscopic surgery. Of interest is that when we do an open gastric bypass, we reconstruct the jejunojejunostomy in a side-to-side, functional, end-to-side anastomosis in an isoperistaltic direction, but with laparoscopic surgery, we reconstruct it in an antiperistaltic direction. Since we have seen RINT with both open and laparoscopic procedures, we do not think this makes a difference.

Findings
The median time from gastric bypass to developing RINT was 51 months, indicating this may be a problem with a significant latency period; this would also explain why the number of cases is increasing since the boom in bariatric surgery began in 2000. We continue to follow patients who had surgery as far back as 1976, which may explain our large experience.

Of the total, 35 percent of our patients were found to have a gangrenous or perforated bowel, or had such extensive edema that the intussusception could not be manually reduced. Another 26 percent of patients were found to have obstructed bowels, but the intussusception could be reduced successfully for a total of 61 percent of patients who presumably would not have survived without surgery. These numbers should discount some criticism that I have heard regarding RINT as being transient and self-limiting, therefore rarely requiring surgical intervention. However, 39 percent (9/23) of our patients were found to have spontaneously reduced their intussusception, although we think some of those patients had their intussusception unsuspectingly reduced while we were running the bowel to find the problem, as almost all had evidence of edema and some hyperemia, indicating a recent problem at the jejunojejunostomy. This last group does raise the question about whether or not everyone with a RINT on CT scan needs surgery. Our collection of patients demonstrates that some patients will spontaneously reduce their RINT, but all nine of these initial patients continued to have pain while in the hospital, thus necessitating our intervention. Since our first report, we have seen patients who will have a RINT on CT at one time and not at other times. Almost all the patients we have seen continue to have pain, but we have been very careful to be conservative about operating; most have shown that the RINT will appear on subsequent CT scans if the scan is done while symptomatic.

Treatment
Treatment of these patients varied. Of our initial 23 patients, two patients were treated by simple reduction of the intussusception without any further therapy and were of followed postoperatively for evidence of recurrence. Both patients developed recurrence of their intussusception, one within the year and the other three years later. Although these numbers are small, we quickly abandoned this approach in favor of some form of intervention. We treated five patients (22%) with jejunopexy, plicating the common limb to the alimentary limb for a distance of at least 15cm; two of these patients had documented recurrence of RINT with virtually all of the others continuing to have the same RINT pain but without intussusception on CT scan. Most patients have required subsequent surgery to try to relieve their pain. Dr. Smith has continued to perform jejunopexyplication of the common limb, whereas Drs. McKinlay and I now perform a resection and revision of the jejunojejunostomy. Since review of the first 23 RINT patients, Dr. Smith has performed his jejunopexy on three subsequent RINT patients and has had to re-operate on one due to intractable pain and vomiting within six weeks postoperatively. He continues to follow the other two patients who continue to complain of intermittent intussusception-type pain.

The remaining patients of our original 23 (70%) were treated with resection and revision of the jejunojejunostomy. These patients appear to have fared better, but two patients had recurrent RINT and a third patient continued to complain of RINT-type pain. Given a longer followup, we are now starting to see more of the original 23 patients return with recurrence of their pain and some with documented recurrence of RINT, but we have not completed our data collection at this time to know what percent of patients continue or subsequently develop recurrent problems.

Nonetheless, resection and revision of the jejunojejunostomy has emerged as our mainstay treatment for this problem.
Given our initial data and our review of the literature, we developed an IRB-approved protocol for treating these patients that basically calls for the initial treatment of patients with persistent or documented recurring RINT to undergo revision of their jejunojejunostomy. If they have documented recurrent RINT or have persistent, severe, disabling symptoms of RINT after revision surgery, we have offered these patients reversal of their gastric bypass with or without sleeve gastrectomy. To date, we have reversed eight patients who have met these criteria, and all were accompanied with a sleeve gastrectomy. Using this approach, we have had no mortality in the patients whom we have treated or are following. Results on these patients were presented in August at the International Federation for the Surgery of Obesity (IFSO) meeting in Paris, France. Our findings were that all 8 patients had complete resolution of their symptoms of RINT at least in the short and midterm with no significant weight regain during this same period.  With these encouraging results we will continue to treat patients following this protocol.

Discussion
The real question is: Why does RINT occur in the first place? It seems logical if one is to have an intussusception moving in the reverse direction of normal bowel peristalsis than one must have reverse peristalsis in the area of the jejunojejunostomy in order for a RINT to occur. Goverman[9] was the first person to suggest that RINT was possibly a manifestation of Roux stasis syndrome. There have been a large number of studies, both animal and human, on Roux stasis syndrome around the world, but all of this research concentrated on the effect of Roux-en-Y anatomy after partial gastrectomy for ulcers or tumor disease. As an example, Schirmer’s 1994 review of Roux stasis syndrome10 listed 60 references, most of which were reporting studies directly about Roux stasis syndrome. The most recent report we found on laboratory studies was by Zhang in 2006.[11] What Zhang and others describe is congruent with what we have observed regarding RINT.

Roux stasis syndrome appears to be a dysmotility disorder of the Roux limb characterized by abnormal peristalsis that brought on by ectopic pacemakers in the small bowel, leading to a variety of motility changes. These changes include reverse peristalsis in the alimentary, hepatobiliary, and common limb just distal to the jejunojejunostomy. While most patients probably exhibited the physiology on motility studies, only a minority exhibit symptoms. In some studies, approximately 30 percent of gastric resection patients exhibited symptoms but most did not have to undergo surgery to treat the symptoms.[12] Interestingly, the preponderance of female to male is approximately 5 to 1. Converting patients to a subtotal or near total gastrectomy improved most patients but did not relieve all of the patients’ symptoms entirely.[13,14] Those symptoms are nausea, vomiting, and epigastric and periumbilical pain particularly aggravated by eating. Usually symptoms start several years after initial surgery.

These are the exact same symptoms and physiologic conditions we are seeing in our RINT patients, and what is a gastric bypass if it is not a modification of a subtotal gastrectomy? This may explain why only a small percentage of our patients have symptoms rather than the 30 percent of patients who have had less of a gastric resection. Steeg,[15] in his brief report on RINT, goes so far as to suggest that one of the reasons gastric bypass works so well for weight loss is that most of our patients probably have a low grade subclinical Roux stasis syndrome as one mechanism that controls hunger. We would add to that by proposing that Roux stasis syndrome may not only be the cause of RINT, but may also be the cause of chronic abdominal pain and bloating that is so troublesome in a small percentage of our patients. The pain and bloating often occur years after surgery[16] and may even be causal to the late, intractable, gastrojejunal ulcers that are such a mystery in patients who do not smoke or use nonsteroidal anti-inflammatory drugs (NSAIDS).[12] After all, Roux abandoned his operation at least as a primary surgery due to an unacceptable high incidence of gastrojejunal ulcers.[17] But that is another subject altogether.

The rationale for our treatment is simply that most of the patients on whom we were forced to do a resection of the jejunojejunostomy because of ischemia or perforation have seemed to do well. Also, the resection and revision of the jejunojejunostomy may, at least temporarily, interrupt or reduce the effect of Roux limb ectopic pacemakers, thereby limiting the effect on the larger part of the bowel.[18] Given the reasonable results thus far, it appears safe as a first-line treatment. Finally, if that fails, then reversal of the gastric bypass, an option not available in most other procedures using a Roux-en-Y, would presumably return the bowel motility to a more normal state by removing the site of the ectopic pacemakers. This appears to be a more drastic and higher-risk procedure and one that most patients will not accept owing to their risk of regaining weight, especially since most have done very well with their weight loss up to this point. This is a challenging laparoscopic operation but can be done with a reasonable level of safety with experience. When combined with a sleeve gastrectomy, it is more likely to be accepted by our patients as a reasonable solution to their predicament.[19]

Conclusion
In summary, RINT is a more common problem than previously realized, but still only occurs in a very small percentage of our patients. That percentage is as yet unknown. With an increasing number of patients with Roux-en-Y anatomy in our population both from gastric bypass and other uses of Roux-en-Y for other procedures, every bariatric surgeon and most general surgeons either already have or will in the future face this problem. The best treatment is truly unknown, but our experience would suggest that revision of the jejunojejunostomy is a reasonable first approach for these patients and that most will require urgent or emergency surgery to prevent or treat gangrenous bowel or perforation. Some patients can be selectively treated either with observation or a more elective approach to surgery, but when in doubt, I would recommend urgent surgery to avoid an intra-abdominal catastrophe as RINT can be a rapidly progressive process. If patients continue to have problems, then reversal of the Roux-en-Y may be indicated, but the authors would suggest that this only be done in high-volume centers with considerable experience with this type of surgery.

While Roux stasis syndrome appears to be the most likely cause of RINT and possibly other problems, more research is needed to confirm this observation and hopefully help define the best treatment of these patients. In the meantime, surgeons should continue to share their experiences with RINT with each other to increase awareness and stimulate new ideas of how to deal with this increasing problem. The authors would certainly like to hear about other surgeons’ experiences and ideas of how to deal with this problem.

References
1.    Simper SC, Erzinger JM, McKinlay RD, Smith SC. Retrograde (reverse) jejunal intussusception might not be such a rare problem: A single group’s experience of 23 cases. Surg Obes Relat Dis. 2008;4:77–83.
2.    Coster DD, Sundberg SM, Kermode DS, et al. Small bowel obstruction due to antegrade and     retrograde intussusception after gastric bypass: three case reports in two patients, literature     review, and recommendations for diagnosis and treatment. Surg Obes Relat Dis. 2008;4:69–72.
3.    Kasotakis G, Sudan R. Retrograde intussusception after Roux-en-Y gastric bypass for morbid obesity. Obes Surg. 2009;19:381–384.
4.    Efthimiou E, Court O, Christou N. Small bowel obstruction due to retrograde intussusception after laparoscopic Roux-en-Y gastric bypass. Obes Surg. 2009;19:378–380.
5.    McAllister MS, Donoway T, Lucktong TA. Synchronous intussusceptions following Roux-    en-Y gastric bypass: case report and review of the literature. Obes Surg. 2009 Jan 31. [Epub ahead of print].
6.    Chen LE, Bhalla S, Warner BW, Strasberg SM. Retrograde jejunoduodenal intussusception: A rare cause of acute pancreatitis after surgery for duodenal atresia. J Ped Surg. 2008;43:E31–E33.
7.    Humbyrd CJ, Baril DT, Dolgin SE. Postoperative retrograde intussusception in an infant: A rare occurrence. J Ped Surg. 2006;41:E13–E15.
8.    Pauli EM, Haluck RS. Antiperistaltic (retrograde) intussusception after laparoscopic Roux-    En-Y gastric bypass procedure. Surg Obes Relat Dis. 2008;4:567–568.
9.    Governman J, Greenwald M, Gellman L, Gadaleta D. Antiperistaltic (retrograde) intussusception after Roux-en-Y gastric bypass. Am Surg. 2004;70:67–70.
10.    Schirmer BD. Gastric atony and the Roux syndrome. Gastroent Clin N Am. 1994;23:327–343.
11.    Zhang YM, Liu XL, Xue DB, et al. Myoelectric activity and motility of the Roux limb after cut or uncut Roux-en-Y gastrojejunostomy. World J Gastroenterol. 2006;12:7699–7704.
12.    Gustavsson S, Ilstrup DM, Morrison P, Kelly KA. Roux-Y stasis syndrome after gastrectomy. Am J Surg. 1988;155:490–494.
13.    Karlstrom L, Kelly KA. Roux-en-Y gastrectomy for chronic gastric atony. Am J Surg.     1989;157:44–49.
14.    Forstner-Barthell AW, Murr MM, Nitecki S, et al. Near-total completion gastrectomy for severe postvagotomy gastric stasis: analysis of early and long-term results in 62 patients. J Gastrointest Surg. 1999;3:15–21.
15.    Steeg KV. Retrograde intussusception following Roux-en-Y gastric bypass. Obes Surg. 2006;16:1101–1103.
16.    Decker GA, DiBaise JK, Leighton JA, et al. Nausea, bloating and abdominal pain in the Roux-en-Y gastric bypass patient: More questions than answers. Obes Surg. 2007:17:1529–533.
17.    Ikard RW. The Y Anastomosis of Cesar Roux. Surg Gynecol Obstet. 1989;169:559–567.
18.    Zonca S, Rizzo P. Alteration of the Roux stasis syndrome by an isolated Roux limb: Correlation of slow waves and clinical course. Am Surg. 1999;65:666–672.
19.    Parikh M, Pomp A, Gagner M. Laparoscopic conversion of failed gastric bypass to duodenal switch: technical considerations and preliminary outcomes. Surg Obes Relat Dis. 2007;3:611–618.

Tags: Retrograde Intussusception

Category: Commentary, Past Articles