A

K

crico

e

SFR

1d

Original article

Spectrum and treatment of small bowel obstruction afterRoux-en-Y gastric bypass

Lana G. Nelson, D.O., M.S.P.H., Rodrigo Gonzalez, M.D., Krista Haines, B.A.,Scott F. Gallagher, M.D., Michel M. Murr, M.D., F.A.C.S.*

Department of Surgery, Interdisciplinary Obesity Treatment Group, University of South Florida Health Sciences Center, Tampa, Florida

Received July 28, 2005; revised February 14, 2006; accepted February 23, 2006

bstract Background: Small bowel obstruction (SBO) is a well-recognized complication of bariatric sur-gery. Many factors that play a role in the etiology of SBO affect the presentation, timing, andtreatment after Roux-en-Y gastric bypass (RYGB). We reviewed our experience with SBO afteropen and laparoscopic RYGB.Methods: We reviewed prospectively collected data from 784 consecutive patients who hadundergone RYGB (458 open and 326 laparoscopic) from July 1998 to March 2005. The operativetechniques were standardized, including closure of the mesenteric defects. The follow-up data weretaken from clinic visit records and follow-up questionnaires. The mean follow-up period was 16 �1 months (range 1–75). The data presented are the mean � SEM.Results: The overall incidence of SBO after RYGB was 3.2%. Thirteen patients developed SBOafter laparoscopic RYGB (4%) and 12 patients did so after open RYGB (2.6%, P � NS).Obstruction at the jejunojejunostomy was more common after laparoscopic RYGB (77%, P �.05),and adhesive SBO was more common after open RYGB (50%, P �.05). The incidence of SBO frominternal hernia was low, regardless of the operative approach (open 0.7% versus laparoscopic 0.3%).Early SBO resolved with nonoperative treatment in 30% of patients.Conclusion: Understanding the anatomic considerations of RYGB in the development of SBOafter open and laparoscopic approach is essential to timely and effective treatment. © 2006American Society for Bariatric Surgery. All rights reserved.

eywords: Clinically significant obesity; Roux-en-Y gastric bypass; Small bowel obstruction; Laparoscopy; Internal hernia;

Surgery for Obesity and Related Diseases 2 (2006) 377–383

Adhesions

[3mt

toep

M

Small bowel obstruction (SBO) is a well-known compli-ation of abdominal surgery, including both open and lapa-oscopic procedures. Although it is difficult to estimate itsncidence after abdominal surgery, SBO results in signifi-ant morbidity and cost because of hospitalization and re-peration [1].

The reported incidence of SBO after laparoscopic Roux-n-Y gastric bypass (RYGB) ranges from 1.9% to 7.3%

*Reprint requests: Michel M. Murr, M.D., F.A.C.S., Department ofurgery, Interdisciplinary Obesity Treatment Group, University of Southlorida College of Medicine, c/o Tampa General Hospital, P.O. Box 1289,oom F-145, Tampa, FL 33601.

gE-mail: mmurr@hsc.usf.edu

550-7289/06/$ – see front matter © 2006 American Society for Bariatric Surgeroi:10.1016/j.soard.2006.02.013

2,3], and the incidence of internal hernias ranges from.1% to 5% [4,5]. It has been suggested that the rate of SBOay be greater after laparoscopic RYGB than after the open

echnique as a result of internal hernias [5,6].This study evaluated the incidence, timing, etiology, and

reatment of SBO after RYGB. A thorough understandingf the common and unique etiologies of SBO after RYGB isssential for any surgeon charged with the care of bariatricatients.

ethods

Data on all patients undergoing primary and revision

astric bypass by a single surgeon were prospectively main-

y. All rights reserved.

tTipg

wtliakrl�

bndgnjteiaacfi

otrtc

fnF

R

I

fottS

T

ilwtttS(peR

I

Spadulcd

F�

FS

378 L.G. Nelson et al. / Surgery for Obesity and Related Diseases 2 (2006) 377–383

ained in an electronic database (ZBMI Data Solutions,ampa, FL). The patient demographics, clinical character-

stics, operative details, and outcomes were reviewed. Allatients qualified for weight loss surgery according to theuidelines developed by the National Institutes of Health.

Although both laparoscopic and open RYGB patientsere included, open RYGB was done exclusively during

he first 14 months of the study. After the introduction ofaparoscopic RYGB, the decision to use open RYGB tooknto consideration a history of bariatric surgery or intra-bdominal surgery, body habitus, body mass index �60g/m2, and patient preference. As our experience with lapa-oscopic RYGB has increased, these criteria have becomeess restrictive and are not considered absolute. Currently,

90% of RYGB procedures are done laparoscopically.The operative technique for laparoscopic RYGB has

een previously reported [7]. In brief, the operative tech-ique is similar, regardless of the mode of access. A 30-cm3

ivided gastric pouch using the lesser curvature and the leftastric artery is made with linear staplers. The cardiojeju-ostomy is formed using a circular stapler. The side-to-sideejunojejunostomy is completed with the double-staplingechnique, using a 60-mm linear stapler and closing thenterotomies with a second application of the stapler. Earlyn our experience, RYGB was undertaken using a retrocolicntegastric Roux limb; currently, our preferred approach isn antecolic antegastric Roux limb. We reserve the retro-olic approach for patients with a shortened mesentery toacilitate a tension-free cardiojejunostomy and currently uset in �1% of patients.

The evaluation and management of SBO was dependentn individual patient presentation and severity of symp-oms. Patients underwent radiographic evaluation with plainadiographs, contrast studies, or computed tomography. Pa-ients who required operative intervention underwent either

ig. 1. Etiology of SBO in open and laparoscopic RYGB patients (*P.05, open versus laparoscopic). JJ � jejunojejunostomy.

eliotomy or celioscopy. The follow-up data were taken d

rom clinic visit records or periodic follow-up question-aires. Statistical analysis and comparison were done usingisher’s exact test; P �.05 was considered significant.

esults

ncidence of SBO

A total of 784 consecutive patients underwent RYGBrom July 1998 to March 2005. Of these, 458 underwentpen RYGB and 326 underwent laparoscopic RYGB. Ofhe 326 patients, 18 required conversion from a laparoscopico an open approach for technical reasons. The etiology ofBO in the 25 patients (3.2%) is summarized in Fig. 1.

iming of SBO

Early SBO occurred in 13 patients within 30 days of thendex RYGB. The remaining 12 patients developed SBOate in the postoperative period, after 30 days. All patientsho developed early SBO had an acute onset of symptoms;

hose who developed late SBO presented with a combina-ion of acute onset, intermittent symptoms, or chronic symp-oms of insidious onset. The most common etiology of earlyBO after RYGB was obstruction at the jejunojejunostomyn � 11). The most common etiology of late SBO in ouratient population was adhesions (n � 7). Fig. 2 shows thetiology of SBO in relationship to the timing of the indexYGB.

ncisional hernias

An incisional hernia through a port site precipitated earlyBO after laparoscopic RYGB in 1 patient who developedersistent nausea and vomiting in the immediate postoper-tive period. Plain radiographs of the abdomen revealedilated bowel loops and residual contrast from a routinepper gastrointestinal contrast study done 1 day earlier. Ataparoscopic exploration, a port site hernia with an incar-erated loop of small bowel was reduced (Fig. 3). Theefect was closed using interrupted sutures as was done

ig. 2. Obstruction at jejunojejunostomy (JJ) most common form of earlyBO (�30 days) and adhesions most common form of late SBO (�30

ays) after RYGB. *P �.05, open versus laparoscopic).

df

I

hcsi

oodanwr

ttpmltdta

Pbctw

Rhfistw

FcfiLa

Fah

379L.G. Nelson et al. / Surgery for Obesity and Related Diseases 2 (2006) 377–383

uring the index RYGB. No episodes of SBO developedrom an incisional hernia after open RYGB in this cohort.

nternal hernias

Three patients developed SBO secondary to an internalernia. One additional internal hernia was found duringeliotomy for acute abdominal pain in 1 patient withoutymptoms of bowel obstruction, contributing to an overall

ig. 3. (A) Routine upper gastrointestinal contrast study demonstratingontrast-filled dilated small bowel loops. (B) Corresponding operativendings of port site hernia with incarcerated small bowel in fascial defect.arge arrow indicates dilated small bowel trapped in fascial defect; smallrrow indicates broken fascial suture.

nternal hernia rate of 0.5%. b

Three internal hernias occurred after open RYGB, andne occurred after laparoscopic RYGB. One patient devel-ped an internal hernia through the small bowel mesentericefect between the alimentary and biliary limbs 26 monthsfter laparoscopic RYGB that required celiotomy. The her-iated small bowel was pulled out of the mesenteric defectith ease, and the mesenteric defect was closed with inter-

upted nonabsorbable sutures.Another patient developed an internal hernia through the

ransverse mesocolon, resulting in complete obstruction ofhe Roux limb 5 days after open retrocolic RYGB. Thisatient underwent emergent celiotomy and had an anasto-otic leak at the gastrojejunostomy as a result of a closed

oop obstruction. The pinpoint defect in the gastrojejunos-omy was repaired, and the herniated small bowel waselivered to the lower abdomen and secured to the mesen-eric defect in the mesocolon using interrupted nonabsorb-ble sutures.

A third patient developed an internal hernia througheterson’s space 28 months after open RYGB. The smallowel was rotated in a counterclockwise manner duringeliotomy to relieve the obstruction, and the defect betweenhe mesentery of the Roux limb and the colon was closedith sutures.The fourth patient presented 48 months after open

YGB. Although she had no obstructive symptoms, she didave persistent pain and abnormal contrast abdominal studyndings. Fig. 4 demonstrates the internal hernia through themall bowel mesenteric defect seen on celioscopic explora-ion. The herniated bowel was reduced and the defect closedith interrupted nonabsorbable sutures.

ig. 4. Internal hernia demonstrated between mesentery of alimentary limbnd biliopancreatic limb. Laparoscopic grasper demonstrates defect witherniated bowel. Small arrow points to twist in mesentery of herniated

owel.

O

(ajeaote

o2arcsrtntgpebtojpojrtotatlowtlw

joRjebviob

awlst

A

RpRdota

D

Fjtsf

380 L.G. Nelson et al. / Surgery for Obesity and Related Diseases 2 (2006) 377–383

bstruction at jejunojejunostomy

SBO at the jejunojejunostomy occurred in 14 patients10 laparoscopic, 4 open) and was more likely to present asn early obstruction (Fig. 2, P �.05). Obstruction at theejunojejunostomy was evaluated in relationship to surgeonxperience (first 100 procedures versus subsequent casesnd open versus laparoscopic RYGB). The incidence ofbstruction at the jejunojejunostomy was evenly distributedhroughout the study period, with no clustering in the earlyxperience of open or laparoscopic RYGB.

Overall, 11 patients developed early SBO because ofbstruction at the jejunojejunostomy after RYGB. Of those,presented with massive dilation of the excluded stomach

nd were treated nonoperatively after radiographically di-ected percutaneous gastrostomy tube placement. After de-ompression of the excluded stomach, the obstruction re-olved and patency of the jejunojejunostomy was confirmedadiographically by contrast perorally and contrast studieshrough the gastrostomy tube. No further treatment waseeded. Similarly, 3 additional patients developed obstruc-ion at the jejunojejunostomy according to routine upperastrointestinal contrast study findings in the immediateostoperative period. These 3 patients responded to nonop-rative management with nasogastric decompression untilowel function resumed as documented by flatus. No furtherreatment was required. Six patients had persistent dilationf the Roux limb from a fixed obstruction at the jejunoje-unostomy and required operative intervention. Of these 6atients, 1 was fully anticoagulated with vitamin K antag-nists and was found to have a hemobezoar occluding theejunojejunostomy during celiotomy. The obstruction waselieved by milking the hemobezoar past the jejunojejunos-omy. The remaining 5 patients required operative revisionf the jejunojejunostomy secondary to critical narrowing ofhe bowel lumen where the enterotomies for the staplednastomosis were closed longitudinally. The obstructionypically affected the Roux limb, and the biliopancreaticimb communicated freely with the alimentary limb by wayf a patent anastomosis. In those patients, the obstructionas relieved by a side-to-side jejunojejunostomy between

he obstructed Roux limb and the nonobstructed commonimb in proximity of the obstruction point. The enterotomiesere stapled closed transversely (Fig. 5).Three patients developed late SBO from a strictured

ejunojejunostomy. Fig. 6 demonstrates the massive dilationf the excluded stomach 21 months after laparoscopicYGB. This patient had had an adhesive band placed at the

ejunojejunostomy that allowed reflux of swallowed air andnteric fluid into the excluded stomach. The obstructingand was released, and the jejunojejunostomy was not re-ised; however, a gastrostomy decompression tube wasnserted. The patient recovered without incident. The sec-nd patient developed an obstruction from a dense adhesive

and at the jejunojejunostomy 6 weeks after open RYGB l

nd required open revision of the anastomosis. One patientho was pregnant developed an obstruction 11 months after

aparoscopic RYGB that responded to one attempt of endo-copic balloon dilation of a stricture at the jejunojejunos-omy. No further treatment was needed.

dhesions

Adhesions were more likely to cause SBO after openYGB (Fig. 1, P �.05). Adhesive SBO occurred in 1atient after laparoscopic RYGB and in 6 patients after openYGB. Three patients who responded to nasogastric tubeecompression had no demonstrable fixed obstruction seenn radiographic contrast studies and required no furtherreatment. Four patients underwent exploratory celiotomynd lysis of adhesions.

iscussion

The advent of laparoscopic bariatric surgery has chal-

ig. 5. Stricture (2) developed immediately proximal to original jejunoje-unostomy (1). Side-to-side jejunojejunostomy (3) was constructed be-ween obstructed Roux limb (light shade) and common channel (darkerhade) to bypass obstructing point. Parabolic arrow depicts passage ofood, and sigmoid arrow, passage of biliopancreatic secretions.

enged many dictums about the feasibility and safety of

lcrj

msdosaPmuhn

wgudlol

pimstalaoorot

gdofip1dctRrd

p

Fjcsd

TR

PFSNHPFGHF

381L.G. Nelson et al. / Surgery for Obesity and Related Diseases 2 (2006) 377–383

aparoscopic procedures in bariatric patients. Of the manyauses of SBO after RYGB, two are modifiable and may beelated to operator experience: internal hernias and jejunoje-unostomy strictures.

The anatomy of RYGB results in mesenteric defects thatay lead to internal herniation and subsequent bowel ob-

truction. One potential site of herniation is between theivided mesentery of the alimentary limb and the mesenteryf the biliopancreatic limb at the jejunojejunostomy. Theecond site is between the mesentery of the alimentary limbnd the transverse mesocolon, frequently referred to asetersen’s space. The third is located in the transverseesocolon when a retrocolic passage of the Roux limb is

sed. Antecolic anastomosis eliminates the possibility oferniation through the transverse mesocolon only, but doesot eliminate hernias in Peterson’s space.

ig. 6. Massive dilation of excluded stomach from obstruction distal toejunojejunostomy after RYGB required immediate drainage to preventatastrophic gastric perforation. Air-fluid level in stomach indicated ob-truction was distal to jejunojejunostomy; contrast-filled loops anterior toilated excluded stomach indicate Roux limb.

able 1eported incidence of small bowel obstruction and internal hernias after R

Study Year n

resent study 2005 784 2.6 (nernandez et al. [2] 2004 2111 3.3 (nmith et al. [10] 2004 779 3.3guyen et al. [11] 2004 225wang [12] 2004 1715apasavas et al. [13] 2003 246elsher et al. [13] 2003 115arza et al. [8] 2004 1000iga et al. [4] 2003 2000ilip et al. [6] 2002 100

SBO � small bowel obstruction.

Although some surgeons do not close mesenteric defects,e believe that routinely closing mesenteric defects hasreatly reduced the internal hernia rate in our patient pop-lation. As with the open procedure, we close mesentericefects during laparoscopic RYGB and have maintained aow incidence of internal hernias, irrespective of the modef abdominal access and during our early experience inaparoscopic RYGB.

The rate of SBO in our cohort was similar to that of otherublished studies; however, we had an exceedingly lowncidence of SBO due to internal hernias (Table 1). This

ay have been because of our meticulous closure of allmall bowel and mesocolic mesenteric defects when placinghe Roux limb in a retrocolic position. Our initial preferredpproach was a retrocolic antegastric placement of the Rouximb in both open and laparoscopic RYGB. We switched ton antecolic position, not because of an increased incidencef internal hernias or stenosis, but to significantly reduce theperative time. Nevertheless, others have reported that aetrocolic position is associated with an increased incidencef obstruction due to internal hernias and strictures at theransverse mesocolon [8,9,12,13].

It has been suggested that the internal hernia rate may bereater after laparoscopic RYGB and that these may notevelop until years after the index procedure. The trend inur practice has been to use a laparoscopic approach morerequently in recent years, and we recognize that follow-ups shorter for our laparoscopic patients than for our openatients. The mean follow-up was 16 � 1 months (range–75). On the basis of these short- to intermediate-termata, we anticipate that the rate of internal hernias willontinue to be low because we have adopted the sameechnical principles of open surgery for laparoscopicYGB. We continue to close mesenteric defects with inter-

upted nonabsorbable sutures as we don during open ab-ominal procedures.

Stenosis or obstruction at the jejunojejunostomy is aotential source of morbidity after RYGB. Evidence of

SBO (%) Internal Hernia (%)

Laparoscopic Open Laparoscopic

) 4 (n � 326) 0.7 0.31) 1.9 (n � 580)

2.743 17.35.2

4.51.75

YGB

Open

� 458� 143

obaslj

nsrgmjaccaOmsdd

ttwdijmdinv

cscrahjjctiOa

onpto

lraoNi

C

siiiasno

R

[

[

[

[

[

382 L.G. Nelson et al. / Surgery for Obesity and Related Diseases 2 (2006) 377–383

bstruction of the jejunojejunostomy may be subtle at first,ecause many patients may experience nausea and vomitingfter RYGB. Upper gastrointestinal contrast studies thathow a dilated Roux limb or air-fluid levels in the Rouximb should raise suspicion of obstruction at the jejunoje-unostomy and should be followed up with delayed films.

The exact location of the obstruction at the jejunojeju-ostomy results in different clinical manifestations. Ob-truction proximal to the common lumen of the anastomosisesults in nausea, vomiting, and abdominal cramps as in-ested food decompress through the Roux limb into theouth. Obstruction distal to the common lumen of the

ejunojejunostomy may decompress into the excluded stom-ch and will manifest as fullness, tachycardia, nausea, hic-oughs, and shoulder pain. Subsequent dilation of the ex-luded stomach requires emergent operative intervention tovoid the disastrous consequences of gastric perforation.bstruction in the biliopancreatic limb proximal to the com-on lumen will result in isolated dilation of the excluded

tomach with fluid (but not air) and, therefore, is bestiagnosed with computed tomography and requires imme-iate intervention and decompression.

Early anastomotic strictures at the jejunojejunostomy areechnical [6]. Nevertheless, we could not establish a rela-ionship with surgeon experience, because those patientsere not clustered in the early experience (first 100 proce-ures) of the operative surgeon. We found that operativentervention was not always necessary; one half of our earlyejunojejunostomy obstructions resolved with nonoperativeanagement consisting of bowel rest and either nasogastric

ecompression or drainage of the excluded stomach whenndicated. However, fixed obstruction due to stricturing orarrowing of the intestinal lumen requires operative inter-ention.

Fixed obstruction at the jejunojejunostomy is related tolosure of the enterotomies used to introduce the lineartapler for the side-to-side jejunojejunostomy and is moreommon after laparoscopic RYGB. Other surgeons haveeported obstruction at the jejunojejunostomy after RYGBnd several solutions have been suggested. One option is toand sew the enterotomy after undertaking the stapled je-unojejunostomy [15,16] or extending the stapled jejunoje-unostomy both proximally and distally with separate appli-ations of the linear stapler and then closing the enterotomyransversely [17,18]. We close the enterotomies transverselyn open RYGB and longitudinally in laparoscopic RYGB.ur study and other studies were underpowered to detect

ny significant differences between these two techniques.We continue to assess the role of nonoperative treatment

f early SBO, as for the few patients who were treated withasogastric tube or percutaneous gastrostomy tube decom-ression. In those instances, radiographic documentation ofhe obstruction and its resolution was affirmative. We cannly speculate regarding the etiology of those episodes.

Although adhesive SBO is expected to be lower after

aparoscopic surgery, studies have demonstrated that lapa-oscopic surgery does not completely eliminate the risk ofdhesion formation [14]. Adhesions were an unlikely causef SBO after laparoscopic RYGB in our patient population.evertheless, longer follow-up is needed to assess the true

ncidence of adhesive SBO.

onclusion

SBO after RYGB may develop because of adhesions,tricture or stenosis, bezoar, abdominal wall hernias, ornternal hernias. Although we may be able to identify trendsn the timing and etiology of SBO, a thorough understand-ng of all possibilities is necessary to assist in the evaluationnd treatment of patients. A heightened awareness of theubtle clues, along with the use of various diagnostic tech-iques, will assist in the diagnosis and subsequent treatmentf patients who develop SBO after RYGB.

eferences

[1] Beck D, Opelka F, Bailey R, Rauh S, Pashos C. Incidence of small-bowel obstruction and adhesiolysis after open colorectal and generalsurgery. Dis Colon Rectum 1999;42:241–8.

[2] Fernandez A, DeMaria E, Tichansky D, et al. Multivariate analysis ofrisk factors for death following gastric bypass for the treatment ofmorbid obesity. Ann Surg 2004;239:698–703.

[3] Papasavas P, Caushaj P, McCormick J, et al. Laparoscopic manage-ment of complications following laparoscopic Roux-en-Y gastric by-pass for morbid obesity. Surg Endosc 2003;17:610–4.

[4] Higa K, Ho T, Boone K. Internal hernias after laparoscopic Roux-en-Y gastric bypass: incidence, treatment, and prevention. Obes Surg2003;13:350–4.

[5] Marema RT, Perez M, Buffington CK. Comparison of the benefitsand complications between laparoscopic and open Roux-en-Y gastricbypass surgeries. Surg Endosc 2005;19:525–30.

[6] Filip J, Mattar S, Bowers S, Smith D. Internal hernia formation afterlaparoscopic Roux-en-Y gastric bypass for morbid obesity. Am Surg2002;68:640–3.

[7] Murr MM, Galagher SF. Technical considerations for transabdominalloading of the circular stapler in laparoscopic Roux-en-Y gastricbypass. Am J Surg 2003;185:585–8.

[8] Garza E, Kuhn J, Arnold D, Nicholson W, Reddy S, McCarty T, et al.Internal hernias after laparoscopic Roux-en-Y gastric bypass. Am JSurg 2004;188:796–800.

[9] Champion J, Williams M. Small bowel obstruction and internal her-nias after laparoscopic Roux-en-Y gastric bypass. Obes Surg 2003;13:596–600.

10] Smith S, Edwards C, Goodman G, Halversen C, Simper S. Open vslaparoscopic Roux-en-Y gastric bypass: comparison of operativemorbidity and mortality. Obes Surg 2004;14:73–6.

11] Nguyen N, Huerta S, Gelfand D, Stevens C, Jim J. Bowel obstructionafter Roux-en-Y gastric bypass. Obes Surg 2004;14:190–6.

12] Hwang RF, Swartz DE, Felix EL. Causes of small bowel obstructionafter laparoscopic gastric bypass. Surg Endosc 2004;18:1631–5.

13] Felsher J, Brodsky J, Brody F. Small bowel obstruction after laparo-scopic Roux-en-Y gastric bypass. Surgery 2003;134:501–5.

14] Duron J, Hay J, Msika S, et al. Prevalence and mechanisms of smallintestinal obstruction following laparoscopic abdominal surgery.

Arch Surg 2000;135:208–12.

[

[

[

[

ttgaapbTgc

jhnaiapsgcpb“tcapiFas(tsmtadpoTcds

383Carrasquilla / Surgery for Obesity and Related Diseases 2 (2006) 377–383

15] Nguyen N, Neuhaus AM, Ho HS, Palmer LS, Furdui GG, Wolfe, BM.A prospective evaluation of intracorporeal laparoscopic small bowelanastomosis during gastric bypass. Obes Surg 2001;11:196–9.

16] Higa KD, Ho T, Boone KB. Laparoscopic RYGB: technique and

Editorial co

tomach.

aathfiaAnbioebamidoncpsoops

R

[

[

[

[

17] Madan AK, Frantzides CT. Triple-stapling technique for jejunojejunos-tomy in laparoscopic gastric bypass. Arch Surg 2003;138:1029–32.

18] Ahmad A, Cho K, Brathwaite C. A technique of enteroenterostomy toprevent alimentary limb obstruction in laparoscopic Roux-en-Y gas-

3-year follow-up. J Laparoendosc Adv Surg Tech A 2001;11:377–82. tric bypass. J Am Coll Surg 2004;198:159–62.

mment

This is a very good analysis by the authors of the data fromheir own center regarding the incidence, timing, etiology, andreatment of obstruction of the small bowel after surgery forastric bypass. These reviews provide essential information forll surgeons. Without the guidance of a well experienced bari-tric surgeon, when a patient floats into any emergency roomresenting a certain history and symptoms, a tragic event coulde the unfortunate result of an otherwise preventable incident.he main differential of the small bowel obstruction afterastric bypass is the presentation: early or late. The causes andomplications vary between the two.

Early presentation is mostly found around the jejuno-ejunostomy. The causes can be multiple, including edema,ematoma, bending, foreign bodies, abdominal wall her-ias, adhesions, and/or technical problems. Internal hernias,lthough rare, can also occur at this stage [1]. Due to theirmportance as a potential factor in producing leaks as wells how the diagnosis is obtained, it is important to includeartial obstructions, especially when found in the earlytages. Many of these patients are asymptomatic and want too home. The surprise diagnosis is found at the routine UGIontrast study 24 hours after surgery. Even if the pouch androximal Roux limb are apparently normal, this could alsoe missed unless there is a trained radiologist performing anextended” UGI radiograph, including the jejuno-jejunos-omy [2]. The obstruction does not have to be a completelylosed loop. It only needs to be restricted enough to producen elevation of the intraluminal intestinal pressure. Thehysiopathology of this condition is the increase of thentraluminal intestinal pressure with two aggravating points.irst, there is, in fact, a weakened area at the site of a recentnastomosis. Second, the anastomoses are precisely at theite of the largest diameter, producing higher pressuresLaPlace’s Law) [3]. The emergent treatment of these pa-ients is decompression, usually performed with a soft na-ogastric tube. If the patient does not improve, then oneust look for a more permanent cause of the obstruction. In

hose cases in which the obstruction is at the bilio-pancre-tic segment or in which a distal obstruction is attempting toecompress itself going into the diverted stomach, a decom-ression of the bypassed stomach, usually with the insertionf a gastrostomy tube, is also required as soon as possible.his could avoid the very serious (and potentially fatal)omplication of gastric dilatation, and possible perforation,ue to the increased intraluminal pressure of the diverted

In cases of late presentation of the obstruction (usuallyfter 6 weeks), the causes are internal hernias, adhesions, orbdominal wall hernias. The most common site is distal, andhe influence on the production of leaks is gone. The ex-ibited behavior is similar to most cases in general surgeryor the adhesions and the abdominal wall hernias. However,n the example of an internal hernia, the elusive symptom-tology and findings in the obese patient are very intriguing.lthough a positive work-up is nice to have, in cases of aegative work-up with clinical suspicions, the patient muste explored to avoid a tragic event with a compromisedntestinal circulation. It is important to remember that somef the bariatric patients may hide or confuse the statisticalxpectancy. The site of the internal hernias is well describedy the authors. The amount of small bowel compromised instrangulated internal hernia could vary from a small seg-ent to a great majority, with secondary consequences

ncluding death. The bowel forming the ring or edge of theefect may also be compromised [4]. The presentation ofbstruction or symptoms for internal hernias is usually (butot always) later: a few years after the gastric bypass pro-edure. One last cause of an obstruction that is free from theatient’s influence, and perhaps is an iatrogenic effect, isometimes caused by glue or sealant used to decrease leaksr bleeding. Because of the possible early presentation of anbstruction, extreme caution must be taken whenever theseroducts are used to avoid the run down or spillage of theealant onto different areas of the abdominal cavity.

Carlos Carrasquilla, M.D., F.A.C.S.Florida Center For Surgical Weight Control, P.A.

Fort Lauderdale, Florida

eferences

1] Carmody B, DeMaria EJ, Jamal M, Johnson J, Carbonell A, Kellum J,et al. Internal hernia after laparoscopic Roux-en-Y gastric bypass. SurgObes Relat Dis 2005;1:543–8.

2] Carrasquilla C, English W, Esposito P, Gianos J, Sanchez-Torres M,Weiss M. Contrast study following gastric bypass could prevent seri-ous complications. Presentation, International Federation for Surgeryof Obesity (IFSO). European Symposium, May 2004 (Prague, CzechRepublic). Obes Surg 2004;14:468.

3] Carrasquilla C, Thomson SE, Cummings JF, Kolata RJ. Anterior vs.retro approach for gastric bypass: do luminal pressures influence thedevelopment of leaks? Clinical and laboratory outcomes – preliminaryreport. Obes Surg 2003;13:517.

4] Felsher J, Brodsky J, Brody F. Small bowel obstruction after laparo-

scopic Roux-en-Y gastric bypass. Surgery 2003;134:501–5.