| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Gastrointestinal Imaging |
1 From the Department of Radiology, University of Pittsburgh Medical Center, 200 Lothrop St, Pittsburgh, PA 15213. Received March 10, 2000; revision requested May 10; revision received June 30; accepted July 11. Address correspondence to M.P.F. (e-mail: federlemp@radserv.arad.upmc.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Retrospective review of 4,001 cases of OLT revealed 48 cases of bowel obstruction in 44 patients. Seventeen computed tomographic (CT) scans and seven barium-enhanced radiographs were reviewed to determine level and cause. Surgical proof was available in 45 cases, while three had characteristic clinical and radiographic features.
RESULTS: Adhesions caused the obstruction in 19 cases in 16 patients; three had bowel ischemia. Internal hernias caused obstruction in 18 patients; all obstructions were transmesenteric or retroanastomotic and occurred with choledochoenteric anastomosis. Seven patients had volvulus; four had bowel ischemia. CT signs of internal hernia included mesenteric vascular abnormalities and clusters of dilated small-bowel segments that displaced the colon away from the abdominal wall. Prospective diagnosis of internal hernia was made in only one case. Incisional or inguinal hernia caused obstruction in seven patients; CT was used just once. Four patients had neoplastic bowel obstruction, three due to posttransplantation lymphoproliferative disorder and one due to colon carcinoma.
CONCLUSION: Liver transplantation was complicated by bowel obstruction in 48 (1.2%) of 4,001 cases. While adhesions and incisional hernias are common and well recognized, other causes are more challenging to diagnose. The CT findings reported here may allow more accurate diagnosis of internal hernia.
Index terms: Computed tomography (CT), contrast enhancement, 74.12112 • Computed tomography (CT), helical, 74.12115 • Hernia, 74.158, 74.458 • Intestines, CT, 74.12111, 74.12112, 74.12115 • Intestines, hernia, 74.157, 74.158, 74.458 • Liver transplantation
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
We found computed tomography (CT) to play an important role in helping to distinguish among the myriad of causes of posttransplantation abdominal pain, and we have encountered many cases of bowel obstruction from unusual and previously unreported causes. The goals of this study were to review our experience with more than 4,000 recent cases of liver transplantation to determine the incidence and cause of bowel obstruction and to better understand the radiologic features in this setting.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Clinical and epidemiologic evaluation included assessment of the age and sex of the patients, cause of the obstruction, interval between transplantation and bowel obstruction, outcome of the obstruction, presence of bowel ischemia, and length of resected bowel if resection was performed. Signs and symptoms of bowel obstruction were also evaluated.
Radiologic evaluation included abdominal CT in 27 patients, and 17 had CT findings available for review. (For 10 patients, CT scans were not available for review because they were purged or lost.) Abdominal radiographs were obtained in most patients but were not reviewed in this investigation. We evaluated reports and available images obtained at seven barium small-bowel follow-through examinations.
We performed CT with a conventional scanner in 12 patients and with a helical scanner in 15 (HiLight Advantage and HiSpeed Advantage, respectively; GE Medical Systems, Milwaukee, Wis). All patients received contrast agents administered intravenously. Patients received 150 mL of 60% iodinated contrast medium (Conray 60 [iothalamate meglumine] or Optiray 350 [ioversol]; Mallinckrodt Medical, St Louis, Mo) administered at a rate of 2–3 mL/sec with a power injector (model OP 100; Medrad, Pittsburgh, Pa). Section thickness was 7 mm for both helical and conventional scans.
All patients had both clinical and radiologic signs of bowel obstruction. When the clinical and radiographic findings were considered unequivocal, patients generally underwent surgery without further evaluation. Barium studies and CT were performed at the sole discretion of the transplantation team caring for the patients, with no attempt to perform CT in patients clearly requiring surgery. All images were reviewed retrospectively by the authors. Because our purpose was to document the cause and radiologic findings of bowel obstruction in this study population rather than to test the accuracy of the diagnostic modality of readers, we did not test for interobserver variation but used consensus interpretation.
Bowel obstruction was diagnosed radiologically with the presence of air-fluid levels and dilated (>2.5-cm) lumen of proximal bowel with a transition to collapsed or normal-caliber distal bowel. Incisional hernias were identified by means of characteristic defects in the abdominal wall, and tumors were characterized by the presence of a soft-tissue mass at the point of obstruction. The CT criteria for diagnosis of internal hernia were derived from prior investigations (3) of a larger and more diverse patient population, which included the present group of transplant recipients and patients not undegoing transplantation who had proved internal hernias. The CT signs of internal hernia included abnormal clusters of dilated small bowel lying outside the confines of the colon and causing central displacement of the colon and associated abnormalities of the mesenteric vessels (crowding, displacement, twisting, stretching, and engorgement). Volvulus was diagnosed when bowel and mesenteric vessels were twisted—the whirl sign (4). Adhesions were diagnosed radiologically when small-bowel obstruction was noted and when no other specific cause was recognized. CT signs of ischemia were also noted, including bowel wall thickening, pneumatosis, and ascites.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Adhesions
There were 19 cases of obstruction due to adhesions that affected 16 patients (eight male, eight female; age range, 29–74 years; mean age, 54 years). (Three patients had two separate episodes of adhesions causing small-bowel obstruction.) All patients except one had small-bowel obstruction; the one patient had adhesions and a bascule that obstructed the ascending colon. The interval between transplantation and bowel obstruction was 1 week to 9 years (median, 2 years). The obstruction resolved spontaneously in two patients, while all other patients required surgery. Ten patients underwent surgery following clinical and only radiographic evaluation. Nine patients underwent CT evaluation, and five underwent barium studies of the intestine.
Barium small-bowel follow-through examination in four patients with adhesions demonstrated a dilated proximal bowel and delayed transit through the bowel. Dilution of the barium and the prolonged transit time prevented confident diagnosis of the level and cause of the obstruction. One patient underwent barium enema examination, which demonstrated a colon and distal small bowel of normal caliber.
In nine patients with adhesive obstruction, abdominal CT was performed and demonstrated a dilated proximal bowel with a transition to normal distal bowel, without an obstructing lesion identified. Bowel wall thickening and mesenteric infiltration were noted in three patients. All three patients had bowel ischemia confirmed at surgery, and segmental bowel resection of up to 30 cm was required in two patients. Bowel ischemia with necrosis and perforation was found in one of these two patients.
Internal Hernia
Internal hernia was the cause of bowel obstruction in 18 patients (11 male, seven female; age range, 12–62 years; mean age, 48 years). The characterization of hernia by the surgeons was inconsistent, with the terms "transmesenteric," "retroanastomotic," and "nonspecific" used to describe similar cases or even the same case. No patient had a paraduodenal internal hernia. Sixteen of the 18 patients had surgical proof, while one patient believed to have characteristic CT findings was lost to follow-up. The interval between transplantation and obstruction was 2 weeks to 10 years (mean, 14 months). All patients underwent biliary drainage with a Roux-en-Y biliary-enteric anastomosis. Three patients had received more than one liver transplant.
Signs and symptoms among the patients with internal hernia were indistinguishable from those of the other groups and included intermittent or constant acute or chronic abdominal pain, nausea, and distention. Seven patients had an acute abdomen with severe abdominal pain and tenderness. At surgery, seven patients had small-bowel volvulus of the herniated bowel segments, and four of these had ischemia of the bowel. One patient had extensive bowel ischemia and died despite two surgical interventions. In one patient with an internal hernia, the herniated bowel was twisted around the hepatic artery, which resulted in arterial thrombosis and graft revision.
Of the patients with internal hernia, only one underwent evaluation with barium small-bowel follow-through examination, which showed only nonspecific dilation of jejunal loops. Abdominal CT was performed in 13 patients, and eight had scans available for review. All had multiple CT signs (which included some combination of a cluster of dilated small-bowel segments pressed against the abdominal wall and displacing the colon centrally) that were suggestive of internal hernia (Fig 1).
|
|
|
|
|
Neoplasm
Neoplasm was the cause of obstruction in four patients (all male; age range, 33–59 years; mean age, 49 years). The time from transplantation to obstruction was 1–9 years (mean, 6 years). One patient developed an obstructing colon carcinoma, while three patients had lymphoma (posttransplantation lymphoproliferative disorder) that involved the small bowel alone, colon alone, or colon and small bowel (one patient each). CT demonstrated bowel wall thickening and soft-tissue masses (up to 5 cm in diameter) that caused narrowing of the bowel lumen, partial obstruction, mesenteric infiltration, and mesenteric and retroperitoneal adenopathy (Fig 3). In all four patients (including the patient with carcinoma), the prospective CT interpretation was posttransplantation lymphoproliferative disorder; retrospective review revealed no substantial differences between colon carcinoma and posttransplantation lymphoproliferative disorder.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
We found 48 cases of bowel obstruction among 4,001 patients who underwent OLT, or an incidence of 1.2%, which is similar to the incidence of 1.8% among 397 patients examined by Lebeau et al (1). In another recent review of gastrointestinal complications of liver transplantation, Caraceni and colleagues (2) described mechanical bowel obstruction as a rare event and further stated, "Other causes of mechanical obstruction occurring in the general population apply equally to liver transplant recipients." Lebeau and colleagues (1) reported only one case of internal hernia among their cases with complications of OLT. Our experience is different and may serve to remind radiologists and their clinical colleagues of other unique and relatively common causes of bowel obstruction in this patient population.
It is not surprising that adhesions and incisional hernias occur with some frequency as a result of this complex surgical procedure. Imaging, including CT, plays a relatively minor role in many of these patients, other than to help exclude other potential causes of symptoms (eg, hemorrhage, infection, pancreatitis) or complications of obstruction, particularly ischemia, that may influence therapy.
In our series, patients with adhesions and incisional hernias were usually evaluated at clinical examination and radiography. Some patients, particularly those who are obtunded or obese, may have incisional hernias that require CT diagnosis, which is straightforward and well described in prior articles (7). Similarly, the indications and criteria for CT diagnosis of bowel strangulation and ischemia are thoroughly documented (8–11) and require no elaboration. The clinical implications of bowel ischemia, however, may warrant reemphasis. Because of the multitude of potential causes of abdominal pain following OLT, bowel ischemia is likely to escape diagnosis with physical examination and history taking alone. We had an unusually large percentage of cases in which ischemia was first suggested on the basis of CT findings. Radiologists can play an important role in diagnosing this potentially devastating result of bowel obstruction.
We found that many clinical and demographic features were of little help in distinguishing the causes of bowel obstruction in patients who have undergone OLT. Age, sex, symptoms, and time from transplantation to obstruction were similar among all patients with different causes of obstruction.
Among the unique features of bowel obstruction in transplant recipients is the occurrence of unusually aggressive neoplasms, especially lymphoma (5,11–14). The incidence of colorectal cancer (and other common neoplasms such as breast, prostate, and pancreatic cancer) is not increased among transplant recipients, although immunosuppression may favor the growth of existing cancer cells (15). This observation is consistent with the case of an aggressive and rapidly growing colon cancer in one of our patients.
Approximately 2.0%–2.5% of patients who undergo OLT develop posttransplantation lymphoproliferative disorder, a spectrum of uncontrolled B-cell proliferative responses to Epstein-Barr virus infection. At the stage of monoclonal proliferation, posttransplantation lymphoproliferative disorder is functionally and histologically indistinguishable from other non–Hodgkin lymphomas (16). The abdomen is frequently involved in posttransplantation lymphoproliferative disorder, and among patients with abdominal involvement, the small intestine and colon are involved in 25% and 6% of patients, respectively (12). While gastrointestinal bleeding and intussusception are recognized complications of bowel lymphoma, obstruction is thought to be rare. We agree that posttransplantation lymphoproliferative disorder (lymphoma) usually causes wall thickening and, occasionally, aneurysmal dilatation of the lumen, but we noted three cases of surgically proved intestinal obstruction in our series.
The most surprising and unique feature of our experience was the frequent occurrence of internal hernia as a cause of bowel obstruction in our patients. Internal hernia was the proved cause in 18 (38%) of our 48 cases and was almost equal to adhesions, which caused 19 cases of obstruction. While internal hernias seem to be familiar to our experienced transplant surgeons, knowledge of these is not widespread nor adequately reflected in existing publications.
The most common type of biliary anastomosis performed during OLT is a choledochocholedochal (duct-to-duct) anastomosis, which is used in approximately 70% of OLTs at this institution. While choledochoenteric anastomosis was performed in only 30% of our recipients, it was performed in 100% of the patients with internal hernia following OLT. The internal hernias that develop after OLT occur through the mesenteric defect created with the formation of the Roux-en-Y loop at choledochojejunostomy. The defect can occur in the small-bowel mesentery or transverse mesocolon, if the loop is made retrocolic. Placement of the Roux-en-Y loop in an antecolic position creates a potential space between the loop and the transverse colon, the most common site of internal hernia in our experience (4). Experienced transplant surgeons attempt to close these defects surgically, but incomplete closure or breakdown of the suture line may occur.
Diagnosis of internal hernia is often difficult. It is likely that these begin as small and intermittent hernias that spontaneously reduce. Clinical or radiologic diagnosis at this stage, particularly during asymptomatic periods, is probably impossible. With time and repeated episodes of herniation, the mesenteric defects enlarge. Long segments of small bowel may herniate, twist, and become obstructed and ischemic. At the time of surgery for internal hernia, most of the patients in this series had major portions of small intestine within the hernia. Among them, seven had small-bowel volvulus, four had bowel ischemia, one died, and one had hepatic arterial thrombosis and liver ischemia due to the volvulus. Clearly, we need a better way to diagnose internal hernia.
We were able to diagnose only one case of internal hernia and two cases of small-bowel volvulus at original interpretation of the CT scans in these patients. Only two articles (17,18) have mentioned a few cases of internal hernia following OLT, and the reported radiographic findings were limited to those of volvulus. This prompted us to collect and review these cases, as well as other cases of internal hernia from congenital or non–transplantation-related causes (3). On the basis of CT findings that we noted consistently among this larger group of patients, we believe that we have developed a reliable set of CT criteria that should allow diagnosis of internal hernia in most cases (Table). To test this hypothesis, we are conducting both prospective trials and appropriate retrospective multiple-reader studies of test populations.
In conclusion, bowel obstruction occurs more frequently (1.2%) and occurs as a result of unique causes in liver transplant recipients compared with the general population. While adhesions and incisional hernias remain among the most common and readily diagnosed causes, bowel neoplasms, especially lymphoma, should be considered. Internal hernia is almost as frequent a cause of obstruction as adhesions in patients with liver transplants, and we report observations that may allow more accurate and timely diagnosis of internal hernia.
|
FOOTNOTES |
|---|
Author contributions: Guarantors of integrity of entire study, A.B., M.P.F.; study concepts and design, A.B., M.P.F.; definition of intellectual content, A.B., M.P.F.; literature research, A.B., M.P.F.; clinical studies, A.B., M.P.F.; data acquisition and analysis, A.B., M.P.F.; manuscript preparation, editing, review, and final version approval, A.B., M.P.F.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  P. S. Sandhu, B. N. Joe, F. V. Coakley, A. Qayyum, E. M. Webb, and B. M. Yeh Bowel Transition Points: Multiplicity and Posterior Location at CT are Associated with Small-Bowel Volvulus Radiology, October 1, 2007; 245(1): 160 - 167. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. C. Martin, E. M. Merkle, and W. M. Thompson Review of internal hernias: radiographic and clinical findings. Am. J. Roentgenol., March 1, 2006; 186(3): 703 - 717. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Mathieu and A. Luciani Internal Abdominal Herniations Am. J. Roentgenol., August 1, 2004; 183(2): 397 - 404. [Full Text] [PDF]
|
|
|
|
|
|
J. Yu, M. A. Turner, S.-R. Cho, A. S. Fulcher, E. J. DeMaria, J. M. Kellum, and H. J. Sugerman Normal Anatomy and Complications after Gastric Bypass Surgery: Helical CT Findings Radiology, June 1, 2004; 231(3): 753 - 760. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R Williamson Multiple-choice questionnaire: general Imaging, December 1, 2002; 14(6): 488 - 497. [Full Text] [PDF]
|
|
|
|
|
|
A. Blachar, M. P. Federle, K. M. Pealer, S. Ikramuddin, and P. R. Schauer Gastrointestinal Complications of Laparoscopic Roux-en-Y Gastric Bypass Surgery: Clinical and Imaging Findings Radiology, June 1, 2002; 223(3): 625 - 632. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
