| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Gastrointestinal Imaging |
1 From the Department of Radiology, University of North Carolina School of Medicine, Campus Box 7510, Rm 2016 Old Clinic Bldg, Manning Dr, Chapel Hill, NC 27599-7510. Received September 9, 1998; revision requested November 3; revision received December 9; accepted February 2, 1999. Address reprint requests to D.M.W. (e-mail: dmw@pop.unc.edu).
 |
Abstract |
|---|
 TOP Abstract IntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
MATERIALS AND METHODS: Retrospective review of CT and MR images and clinical records of all patients with an intussusception demonstrated on CT or MR images from January 1, 1991, through April 30, 1998.
RESULTS: Thirty-three patients had one or more intussusceptions demonstrated on CT (n = 30) or MR (n = 3) images. Twenty-nine patients had enteroenteric intussusceptions, and four had intussusceptions involving the colon. Ten patients (30%) had a neoplastic lead point, including all four of the intussusceptions involving the colon (benign mass, n = 3; malignant mass, n = 7). In 23 cases (70%), no neoplastic lead point was identified. A variety of causes were implicated in these cases, with 16 cases (48%) classified as idiopathic. Enteric intussusceptions in the nonneoplastic group were shorter in length (median, 4 vs 10.8 cm; P = .002), smaller in diameter (median, 3 vs 4 cm; P = .002), and less likely to be associated with obstruction (4.3% vs 50%; P = .02).
CONCLUSION: Less than one-third of adult intussusceptions demonstrated at CT or MR imaging were caused by a neoplastic lead point. Almost half of adult cases in this series were idiopathic.
Index terms: Celiac disease, 74.7611 • Enteritis, 74.291 • Fibrosis, cystic, 74.1496 • Intestines, CT, 70.12112, 70.12115 • Intestines, diseases, 70.31, 70.32, 70.33, 70.73, 70.7611 • Intestines, MR, 70.12143 • Intussusception, 70.73 • Lymphatic system, hyperplasia, 99.329 • Surgery, complications, 74.458
|
Introduction |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
|
|
MATERIALS AND METHODS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
To identify additional patients who may have undergone CT or MR imaging that showed intussusception but in whom the intussusception was missed, a search of patient records with a discharge diagnosis of intussusception was also performed. This revealed one patient in whom an intussusception was visible on MR images but had not been noted in the initial report. This patient met the previously listed criteria and was then added to our sample.
One additional patient who met the same criteria was included in our study group after a review of teaching files for the dates noted. In this case, a transient intussusception was noted at CT but inadvertently had not been mentioned in the initial report.
The 37 patients thus identified composed 0.2% (37 of 16,399) of all patients undergoing abdominal CT or abdominal MR imaging during the time interval studied. Four of the 37 identified patients were in the pediatric or adolescent age group (ages 1, 4, 10, and 14 years) and were not considered further in this analysis. The remaining 33 patients (24 male patients, nine female patients; age range, 18–84 years; median age, 41 years) constituted our study population.
Clinical Data and Pathogenesis
Demographic data and clinical information were obtained by reviewing the patients' charts (D.M.W.). The presence of abdominal symptoms was recorded along with findings at physical examination. The presence of guaiac-positive stools and the patient's preimaging hematocrit values were also recorded if obtained within 2 weeks of imaging. The pathogenesis of the intussusception was established by means of clinical follow-up, chart review, and review of surgical and histologic findings. If no local or systemic explanation for the intussusception was noted, the diagnosis was classified as idiopathic.
CT and MR Imaging
CT studies were performed with use of a Somatom Plus 24, Somatom Plus S, or Somatom Plus 4 CT Scanner (Siemens Medical Systems, Iselin, NJ). In 22 patients, the scan was obtained by using helical mode. Twenty-nine of the 30 CT examinations were performed with 100–150 mL of intravenously injected iohexol or diatrizoate meglumine (Omnipaque 300 or Hypaque; Nycomed Amersham, Princeton, NJ). Orally administered contrast material (280–1,080 mL) was used for all of the CT examinations and was either dilute barium (Cheetah Barium Sulfate Suspension, Lafayette Pharmaceuticals, Lafayette, Ind or E-Z-Cat, E-Z-Em, Westbury, NY) or dilute Gastrografin (Bracco Diagnostics, Princeton, NJ).
MR imaging studies were performed with use of a 1.5-T Magnetom SP or 1.5-T Vision system (Siemens Medical Systems). In all three of the MR imaging studies, contrast material (Magnevist; Berlex Laboratories, Wayne, NJ) was injected intravenously. No contrast material was administered orally for any of the MR imaging studies.
CT and MR images were evaluated by consensus reading of two experienced abdominal radiologists (D.M.W., J.K.T.L.). The appearance and number of intussusceptions and the type of bowel involved (enteroenteric, enterocolic, colocolic) were recorded as were the length and diameter of the intussuscipiens. Small-bowel obstruction was deemed present if multiple bowel loops proximal to the intussusception were dilated to a diameter of 3 cm or greater and bowel loops distal to the intussusception were decompressed to 1.5 cm or less in diameter.
Data Analysis
Basic descriptive statistics—median, range, and 25th and 75th percentiles—were calculated for patient demographic and clinical data, as well as for radiologic findings. Comparison of continuous data—age, hematocrit value, and length and diameter of the intussusception—between groups with a neoplastic lead point and those without a neoplastic lead point was performed by using the Mann-Whitney U test. This nonparametric statistic was chosen rather than a standard t test owing to the non-Gaussian distribution of the data discovered during analysis. Comparison of categorical data (sex, race, presence of small-bowel obstruction, presence of adenopathy, etc) between groups was performed by using the 
2 test. The Fisher exact test was applied in 2 x 2 analysis tables when any expected values in a cell were less than 5. A P value less than .05 was required for rejection of the null hypothesis. STATVIEW 5 software (SAS Institute, Cary, NC) was used for the statistical analysis.
|
RESULTS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Clinical Data
Twenty-nine of the patients had abdominal symptoms at the time of imaging. Abdominal pain was the most common complaint and was noted in 24 (73%) patients. Abdominal physical examination results were unremarkable in 11 patients. Tenderness or guarding were noted in the right or left upper quadrant (n = 4), in the right or left lower quadrant (n = 3), in the left costovertebral angle region (n = 1), and in a diffuse or unlocalized pattern (n = 7). An abdominal mass was palpated in seven patients. Two patients demonstrated abdominal distention. Stool was checked for occult blood within 2 weeks after CT or MR imaging in 19 of the 33 patients and results were positive in eight patients. Hematocrit values were evaluated within 2 weeks after imaging in 30 patients and showed a median value of 35.5 (25th percentile = 28.9, 75th percentile = 40.5).
Appearance at CT or MR Imaging
Forty-one intussusceptions were observed in the 33 patients. The diagnosis of intussusception was made by observing an axial bowel-within-bowel appearance in 22 intussusceptions, by noting a longitudinal bowel-within-bowel appearance in four intussusceptions, and by seeing the intussusception on both axial and longitudinal sections in 15 intussusceptions. In addition to the bowel-within-bowel appearance, 28 intussusceptions demonstrated fat within the bowel lumen adjacent to the intussusceptum. The appearance of mesenteric vessels being drawn into the intussusception was noted in 22 intussusceptions.
There were 37 cases of enteroenteric intussusception and four cases of either colocolic or ileocolic intussusception. The median length of the intussusceptions was 4.8 cm (25th percentile = 3.2, 75th percentile = 9.1). The median diameter of the intussusceptions was 3.4 cm (25th percentile = 2.8, 75th percentile = 4.0). Obstruction was present in only four patients.
Mesenteric adenopathy was noted in two cases in which there was a malignant neoplasm acting as a lead point for intussusception: in one case of reactive lymphoid hyperplasia and in one case of Crohn disease that was distant from the site of intussusception.
Pathogenesis
A neoplastic lead point was responsible for 12 intussusceptions in 10 patients. In seven patients, the lead point was a malignant neoplasm (Fig 1). Three patients had metastatic disease (large cell lung cancer and colon cancer metastatic to small bowel and squamous cell lung cancer metastatic to large bowel) and three patients had a primary bowel malignancy (two cases of adenocarcinoma of the colon and one of adenosquamous colon cancer). Immunoblastic lymphoma was diagnosed at resection in one patient (Fig 2). Four of the seven patients with a malignant lead point had either a colocolic or ileocolic intussusception.
|
|
|
|
Twenty-three patients had 29 intussusceptions with no identifiable neoplastic lead point. All nonneoplastic intussusceptions were enteroenteric. In seven of these patients, a possible cause for the intussusception could be found. One patient had cystic fibrosis. One patient with three intussusceptions seen at CT had celiac disease diagnosed at small-bowel biopsy; this patient has previously been reported on (13). One patient with two intussusceptions had reactive lymphoid hyperplasia producing mesenteric adenopathy. In three patients, there was local inflammation from pancreatitis (n = 2) or mesenteric panniculitis (n = 1) (Fig 3). One of the patients with pancreatitis had submucosal hemorrhage at endoscopic biopsy. In one patient, the intussusception occurred 3 months following ileocecectomy with ileoascending colon anastomosis (Fig 4).
|
|
|
|
In one patient, intussusception was observed after blunt abdominal trauma. At surgery, a small liver laceration was seen with bile staining of intraperitoneal structures; however, no bowel injury was noted.
In the 10 remaining patients with idiopathic intussusception, no intraabdominal process was noted nor was other cause suggested (Figs 5, 6). One patient in this group underwent exploratory pelvic laparoscopy with lysis of adhesions. No mention, however, was made of any small-bowel disease. One patient underwent repeat CT immediately following the initial CT examination, but this showed no intussusception or mass lesion. This patient had also undergone two prior CT examinations 5 and 7 months earlier that did not show intussusception, although a CT scan obtained at another institution 1 year earlier had shown a transient small intussusception. Three additional patients underwent follow-up CT a mean of 144 days following initial CT, but no intussusception was demonstrated at repeat imaging. Three patients underwent small-bowel follow-through within 7 days of initial CT, but in none of these patients was either an intussusception or a small-bowel mass identified.
|
|
|
|
|
Clinical-Pathologic Correlation
There was no significant difference in sex in patients with a neoplastic lead point and those without a neoplastic lead point. Patients with a neoplastic lead point were, however, significantly older (median, 47.5 years [25th percentile = 40.0, 75th percentile = 67] for neoplastic vs 39.0 years [25th percentile = 23.0, 75th percentile = 47.0] for nonneoplastic; P = .036). There was no significant difference in the frequency of abdominal symptoms between groups with a neoplastic lead point and those without a neoplastic lead point. Although the hematocrit value was significantly lower in patients in the neoplastic group (median hematocrit value = 27.6 [25th percentile = 24.3, 75th percentile = 33.8] for neoplastic vs 39.0 [25th percentile = 33.1, 75th percentile = 42.3] for nonneoplastic; P = .004), there was no significant difference in frequency of guaiac-positive stools (Table 2).
|
Similarly, enteroenteric intussusceptions with a neoplastic lead point had a significantly larger diameter than that of nonneoplastic intussusceptions (median, 4.0 cm [25th percentile = 3.7, 75th percentile = 4.5] for neoplastic vs 3.0 cm [25th percentile = 2.7, 75th percentile = 3.7] for nonneoplastic; P = .002). No neoplastic intussusceptions were smaller than 3.4 cm in diameter, whereas 66% of nonneoplastic intussusceptions were. Bowel obstruction was significantly more common in the enteroenteric intussusceptions secondary to a neoplastic lead point and occurred in three of six patients with neoplastic lead points versus one of 23 patients with nonneoplastic intussusception (Fisher exact test; P = .02) (Table 3; Figs 7, 8).
|
|
|
|
DISCUSSION |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
In contradistinction to these studies, our data suggest that intussusceptions discovered in the adult patient at CT or MR imaging are predominantly enteroenteric in type and more often nonneoplastic in cause. Only 30% of the patients in this study had a neoplastic lead point, although it is important to note that this included all four patients with intussusceptions involving the colon. Over two-thirds of the patients in our study had intussusceptions without a neoplastic lead point. Although a cause can be suggested in many of these cases, almost half (48%) of all cases were idiopathic.
The increased prevalence of transient nonobstructing enteric intussusception noted in this study probably more closely reflects the true physiologic situation. Previous studies (1–8) that required either surgical confirmation or substantial enough clinical effect for coding as a discharge diagnosis are clearly biased toward more marked and fixed lesions. Previous studies in the radiology literature (9,17) may have suffered from a similar bias, although the case ascertainment methodology is not spelled out. In studies by Merine et al (9) and Gayer et al (17), 20 of 25 patients underwent surgery, and 19 of 25 had neoplastic causes of the intussusception.
Although intussusception is most commonly thought of as secondary to mechanical factors (ie, a mass attached or in the bowel wall or a kink in the bowel wall is pulled forward by normal peristalsis with resultant invagination of the involved wall), functional disturbances also can cause intussusception without gross mural abnormality. Celiac disease is a prime example of this effect, in which gluten sensitivity causes muscular flaccidity followed by transient intussusception (13,18). Diarrheal diseases with resultant abnormal bowel motility also have been cited as the cause for the increased prevalence of nonneoplastic adult intussusception in tropical regions (19–21). In one study from Papua, New Guinea, 66% (21 of 32) of cases had no detectable associated disease (20).
We suspect that the increased number of nonneoplastic cases in our study reflects the presence of minor transient disturbances in bowel motility without clinical importance. Most of these cases would not have come to attention were it not for CT being performed to evaluate unrelated disease or symptoms. An analogy can be drawn to swallowing. Although substantial disturbances in pharyngeal function can result in clinically important aspiration, it is not uncommon for each of us to aspirate slightly. Such minor and infrequent aspiration is of no clinical importance.
Our finding of an increased prevalence of nonneoplastic intussusception is important in that it suggests that not all adult enteric intussusceptions diagnosed at CT or MR imaging require aggressive work-up. Although in previous studies (9,17) CT has been unreliable in directly distinguishing a neoplastic lead point from thickened or edematous bowel wall, several differences between patient groups with and without a neoplastic lead point were noted in this study. Patients with a nonneoplastic cause were significantly younger than those with small-bowel neoplasia and had significantly higher hematocrit values. Nonneoplastic enteroenteric intussusceptions were also significantly shorter, smaller in diameter, and less likely to be associated with obstruction than those with a neoplastic lead point. Furthermore, no neoplastic enteric intussusceptions were shorter than 3.6 cm or smaller than 3.4 cm in diameter, whereas 41% and 66% of nonneoplastic intussusceptions were shorter and smaller than these cutoff values. Obstruction was uncommon in the nonneoplastic group and occurred in only one patient. Nonneoplastic intussusceptions were also transient. Repeat CT sections through the involved bowel that were obtained either immediately or after a longer interval usually failed to demonstrate an intussusception.
Taken together, these findings suggest that aggressive work-up need not be undertaken in younger patients with transient, short, small, nonobstructing enteroenteric intussusception. This recommendation is supported by a recent description of five cases of transient nonobstructing intussusception seen at CT (22). In that study, the lesions were all enteroenteric intussusceptions unassociated with obstruction or substantial dilatation in the involved loop. The length of the intussusceptions was not discussed.
It is interesting that there was not a difference in abdominal symptoms between neoplastic and nonneoplastic groups in our study. We suspect that this is probably because of the ubiquitous and nonspecific nature of abdominal complaints.
Our study results suggest that the overall diagnosis of intussusception at CT and MR imaging may be becoming more common. From 1991 through 1996, 16 patients were examined in our study. From 1997 through April of 1998, an additional 17 patients were examined. This increase may in part be artifactual, reflecting our increased scrutiny of bowel. Part, however, may be due to improvements in technology, with spiral scans now able to cover the entire abdominal cavity in a single breath hold. This faster scanning allows transient events to be imaged in their entirety before they change position or resolve.
Several concerns should be noted regarding this study. There is the question of whether the diagnosis of intussusception at CT or MR imaging is valid without surgical proof. Because the diagnosis depends on the physical relationship of bowel loops and not a histologic diagnosis, we believe that the CT diagnosis is accurate. This view is supported in the literature (9–11). For increased assurance, we took only cases in which the intussusception could be identified clearly on at least two separate sections. The frequent finding of fat within the intussusception (seen in 76% of patients [68% of intussusceptions] in this study) and of mesenteric vessels entering the involved loop (seen in 61% of patients [54% of intussusceptions]) also supports the validity of our diagnosis.
We were also concerned in this study that we could be missing cases, that is, that our results may underestimate the frequency of incidental nonneoplastic intussusception, as these may not have been observed and noted on the report or some may not have shown a classic bowel-in-bowel configuration. This is a potential difficulty and cannot entirely be discounted. Our data, therefore, should be viewed as showing minimum frequency that, if anything, underestimates the actual occurrence rate.
Another potential problem was the possibility of misallocating diagnoses. For example, might some of our idiopathic group actually have mild adhesions, enteritis, or dysmotility? More important, was our follow-up adequate to exclude the presence of small tumors? Although misallocation is possible as the small bowel is difficult to evaluate, it should be noted that for the 16 nonneoplastic cases without clear local explanation (idiopathic plus distant inflammation and trauma groups), seven patients underwent surgery, seven had negative findings on immediate or delayed repeat CT scans, and six had negative small-bowel follow-through results. Clinical follow-up in those patients who had not undergone surgery disclosed no alternative diagnoses, and in only one case was recurrent abdominal pain noted, which was thought to be secondary to irritable bowel syndrome. Although perhaps less than ideal, clinical and radiologic follow-up is probably the best that can be achieved, as it is difficult to justify surgery or invasive testing in these patients with relatively few symptoms.
A final potential difficulty lies in the relatively small number of patients in this study. Although we found significant differences between neoplastic and nonneoplastic groups, it will be important to confirm these findings in larger or multiple series.
In summary, less than one-third of adult intussusceptions demonstrated at CT or MR imaging were caused by a neoplastic lead point. A variety of other causes were implicated in the remaining patients, with almost half of adult cases in this series being idiopathic. Enteroenteric intussusceptions with a nonneoplastic cause were shorter, smaller in diameter, and less likely to be associated with obstruction or anemia than intussusceptions secondary to neoplasm. Our study results suggest that extensive work-up in cases of transient, short, small enteric intussusceptions may be unnecessary in the patient without symptoms.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  K. M. Horton and E. K. Fishman MDCT and 3D Imaging in Transient Enteroenteric Intussusception: Clinical Observations and Review of the Literature Am. J. Roentgenol., September 1, 2008; 191(3): 736 - 742. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. T. Albright, S. N. Grief, R. E. Carroll, and J. Xu Ileoileal intussusception in an adult patient Can Fam Physician, February 1, 2007; 53(2): 241 - 243. [Full Text] [PDF]
|
|
|
|
 |
|
 Y. H. Kim, M. A. Blake, M. G. Harisinghani, K. Archer-Arroyo, P. F. Hahn, M. B. Pitman, and P. R. Mueller Adult intestinal intussusception: CT appearances and identification of a causative lead point. RadioGraphics, May 1, 2006; 26(3): 733 - 744. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S Yalamarthi and R C Smith Adult intussusception: case reports and review of literature Postgrad. Med. J., March 1, 2005; 81(953): 174 - 177. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. H. Choi, J. K. Han, S. H. Kim, J. M. Lee, K. H. Lee, Y. J. Kim, S. K. An, and B. I. Choi Intussusception in Adults: From Stomach to Rectum Am. J. Roentgenol., September 1, 2004; 183(3): 691 - 698. [Full Text] [PDF]
|
|
|
|
 |
|
 N. Lvoff, R. S. Breiman, F. V. Coakley, Y. Lu, and R. S. Warren Distinguishing Features of Self-limiting Adult Small-Bowel Intussusception Identified at CT Radiology, April 1, 2003; 227(1): 68 - 72. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G Gayer, R Zissin, S Apter, M Papa, and M Hertz Adult intussusception--a CT diagnosis Br. J. Radiol., February 1, 2002; 75(890): 185 - 190. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. A. Jabra, J. Eng, C. G. Zaleski, G. E. Abdenour Jr., H. V. Vuong, U. O. Aideyan, and E. K. Fishman CT of Small-Bowel Obstruction in Children: Sensitivity and Specificity Am. J. Roentgenol., August 1, 2001; 177(2): 431 - 436. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Fujimoto, T. Fukuda, M. Uetani, Y. Matsuoka, K. Nagaoki, N. Asoh, I. Isomoto, T. Okimoto, H. Ohtani, N. Matsunaga, et al. Unenhanced CT Findings of Vascular Compromise in Association with Intussusceptions in Adults Am. J. Roentgenol., May 1, 2001; 176(5): 1167 - 1171. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
