Helical CT Signs in the Diagnosis of Intestinal Ischemia in Small-Bowel Obstruction

Transcription

1 Marc Zalcman 1 Marième Sy 1 Vincent Donckier 2 Jean Closset 2 Daniel Van Gansbeke 1 Received February 17, 2000; accepted after revision May 25, Department of Radiology, C. U.. Hôpital Erasme, 808, Route de Lennik, 1070 ruxelles, elgium. ddress correspondence to M. Zalcman. 2 Department of Gastrointestinal Surgery, C. U.. Hôpital Erasme, 1070 ruxelles, elgium. JR 2000;175: X/00/ merican Roentgen Ray Society Helical CT Signs in the Diagnosis of Intestinal Ischemia in Small-owel Obstruction OJECTIVE. The purpose of our study was to determine prospectively the value of helical CT and of various signs of ischemia in the context of acute bowel obstruction. SUJECTS ND METHODS. ll patients seen over a 3-year period with a CT diagnosis of small-bowel obstruction were included. There were 144 examinations in 142 patients. Images were interpreted prospectively with consensus by a fellow and an experienced gastrointestinal radiologist. ttention was focused on the presence of the following signs of strangulation and ischemia: reduced enhancement of the small-bowel wall, mural thickening, mesenteric fluid, congestion of small mesenteric veins, and ascites. diagnosis of ischemia was made if enhancement of the bowel wall was reduced or if at least two of the other signs were found. Results were correlated with surgical findings in 73 cases and clinical follow-up in 71 cases. RESULTS. diagnosis of ischemia was made at surgery in 24 patients. CT diagnosis was correct in 23 patients (96% sensitivity). There were nine false-positive diagnoses (93% specificity). The negative predictive value of CT was 99%. Reduced enhancement of the bowel wall had a sensitivity of 48% and specificity of 100%, mural thickening had a sensitivity of 38% and specificity of 78%, mesenteric fluid had a sensitivity of 88% and specificity of 90%, congestion of mesenteric veins had a sensitivity of 58% and specificity of 79%, and ascites had a sensitivity of 75% and specificity of 76%. CONCLUSION. Helical CT is a highly sensitive method to diagnose or rule out intestinal ischemia in the context of acute small-bowel obstruction. cute obstruction of the small bowel is a frequent reason for presentation to the emergency department. Causes of obstruction are multiple (tumorous, inflammatory, postoperative, iatrogenous, metabolic, or congenital), but the clinical presentation is usually nonspecific. Many articles in the radiology literature advocate the use of CT in the diagnostic management of this condition [1 5]. However, the attitude regarding the treatment is not yet clearly established; because of the imprecision of imaging techniques, many surgeons favor early surgical exploration for fear of misdiagnosing a strangulation [6 8]. significant percentage of laparotomies could be avoided if a reliable diagnosis of ischemia was possible preoperatively. Recent reports have shown the usefulness of CT in the diagnosis of mesenteric ischemia, especially in the context of acute bowel obstruction [9 15]. The purpose of our work was to determine the value of helical CT in the diagnosis of ischemia associated with small-bowel obstruction and to evaluate the various CT findings visible in that context. Subjects and Methods Over a 36-month period, between March 1, 1995, and February 28, 1998, all patients with clinical or radiologic findings suggestive of acute high-grade bowel obstruction underwent a CT study. Patients in the early postoperative period (1 month) were not included. Clinical suspicion was based on the known symptoms and signs of obstruction, including abdominal pain, nausea and vomiting, abdominal tenderness, abnormality of bowel sounds, and abdominal mass. Radiologic suspicion was based on a radiograph of the abdomen with the patient in the supine or upright position. Radiographs were interpreted by the patient s clinician. CT was performed immediately afterward. Patients who had no CT evidence of small-bowel obstruction were not included in the study. If small-bowel obstruction was present on CT, we tried to identify signs of ischemia, indicating the need of a rapid surgical treatment. The following signs were studied: reduced enhancement of the bowel wall after contrast material injection, thickening of the bowel wall, mesenteric fluid, congestion of small mesenteric veins, and ascites. Other signs reported in the literature such as increased enhancement of the bowel wall [12] and serrated beak [11, 13] were not included because JR:175, December

2 Zalcman et al. we failed to recognize them in our daily practice in the period preceding this study. ll CT examinations were interpreted prospectively with consensus by a radiologist-in-training and a senior abdominal radiologist, without knowledge of the laboratory data or results of other examinations. diagnosis of strangulation was made in the following situations: decreased enhancement of the bowel wall or association of at least two of the other four signs. In all cases, the final decision for surgery was made by the surgeon. Correlation between CT results and surgical findings was made by direct questioning of the surgeons. When the patient did not undergo surgery, the follow-up was made by clinical examination, laboratory data, and repeated radiographs of the abdomen in all patients. The gold standard for the diagnosis of ischemia was surgery, except in one patient with thrombosis of the superior mesenteric vein where visualization of the thrombus was considered sufficient proof. total of 73 patients underwent surgery, and 71 patients were treated conservatively. Patient Characteristics Our series included 144 consecutive CT examinations of 142 patients with CT evidence of smallbowel obstruction. Two patients underwent two examinations with an interval of more than 6 months. For the sake of clarity of results, these four cases can be considered as four different patients. There were 81 men and 61 women (age range, years; mean age, 60.9 years). history of previous laparotomy was present in 106 patients (75%). Fifty-two patients (37%) had a history of malignancy, including 10 patients (7%) who had undergone abdominal radiotherapy. Ten patients (7%) had a history of Crohn s disease, eight of whom had undergone an intestinal resection. Examination Technique ll the examinations were performed on a CT scanner (Somatom Plus 4 or 4C; Siemens, Erlangen, Germany). Oral contrast material was not administered. The acquisition of information was helical in all cases. The examinations started with a study of the abdomen before IV contrast material injection. volume of ml of a 35 mg/ 100 ml iodinated agent was then injected at a rate of 2 ml/sec for the 130 patients without medical contraindication. The delay between the start of injection and imaging varied between 40 and 70 sec. The slice thickness was 5 or 8 mm, the pitch was 1.4, and the reconstruction interval was 4 or 6 mm. Image nalysis Reduced enhancement of the bowel wall. This finding (Fig. 1) was based on visual comparison of the attenuation of the bowel wall at the site of the obstruction with the attenuation of bowel loops distant from this site, after contrast material injection. No attempt was made to quantify the enhancement abnormality or to differentiate reduced enhancement from absent or delayed enhancement. Thickening of the bowel wall. bowel-wall thickness of 2 mm or less was considered normal. owel-wall thickness was considered abnormal if it appeared to be thicker than 2 mm in a noncollapsed segment. Owing to the difficulty of accurately measuring the thickness of the small-bowel wall, this evaluation was partly subjective, and the values were therefore not recorded. Mesenteric fluid. The presence of localized fluid in the small-bowel mesentery attached to abnormal (dilated, thickened) small-bowel loops was considered abnormal. The amount of fluid was not evaluated, and the attenuation was not measured. Congestion of small mesenteric veins. Enlargement of the small mesenteric veins around the site of the obstruction as compared with mesenteric veins distant from this site was considered a positive finding. scites. This was defined as presence of fluid in the peritoneal cavity as opposed to fluid in the mesentery. The amount and attenuation value of ascites were not measured. Results There were 24 proven cases of ischemia. CT diagnosis of ischemia was made in 32 patients. Ischemia was confirmed at surgery in 22 of these patients. One patient with a thrombosis of the superior mesenteric vein did not undergo surgery and was treated with anticoagulant therapy. Ischemia was found in one patient who was considered free of strangulation on CT; this patient was an obese woman who had a substandard examination because of various artifacts related to obesity and who had no contrast material injection because of acute renal failure. CT revealed small-bowel obstruction related to a ventral abdominal wall hernia Fig year-old woman with strangulating small-bowel obstruction and infarction., Contrast-enhanced CT scan shows dilated and fluid-filled bowel loops in right flank (i). Small amount of fluid is present in mesentery (m)., Contrast-enhanced CT scan slightly more caudal reveals that wall of ischemic dilated loops is not clearly visible (white arrowheads), and proximal collapsed loops enhance normally (black arrowheads). Patient underwent surgery 24 hr later; she had complete volvulus of distal meter of small bowel caused by tight adhesive band. One meter of necrotic small bowel was resected JR:175, December 2000

3 CT of Small-owel Obstruction Fig year-old man with history of appendectomy, aortic surgery, and previous laparotomy for adhesive bands., Unenhanced CT scan shows dilated and fluid-filled jejunal loops (j). Veins in adjacent mesentery are enlarged (arrowheads)., Unenhanced CT scan reveals fluid in mesentery around congested veins (m). Note pseudothickening of bowel wall (arrowheads) caused by transverse scanning of Kerckring s folds. Gas bubbles in intestinal lumen (arrow) must not be confused with pneumatosis. lso note large amount of fluid in pelvis. Patient, who was immediately treated by nasogastric suction and fully recovered without surgery, was one of nine false-positive cases in our series. and the presence of ascites. ecause her clinical condition deteriorated, she underwent surgery in the following hours, but she died during the postoperative period. TLE 1 a IV contrast material used, n = 21. b IV contrast material used, n = 109. Only two of the nine patients with an unconfirmed CT diagnosis of ischemia underwent surgery. One had multiple ileal strictures caused by Crohn s disease, and one had an incarcerated CT Signs of Intestinal Ischemia in Small-owel Obstruction in 24 Patients CT Signs Ischemia (n = 24) a Nonischemia (n = 120) b Specificity (%) n % n % Reduced enhancement of bowel wall Thickening of bowel wall Thickening of bowel wall with target sign bnormal enhancement of bowel wall (reduced enhancement, target sign, or both) Mesenteric fluid Mesenteric congestion scites TLE 2 Value of Combination of Two Signs for the Diagnosis of Intestinal Ischemia hernia of the ventral abdominal wall without evidence of ischemia. The seven other patients did not undergo surgery because the discrepancy between clinical and radiologic diagnoses was considered too important by the patient s surgeon. These seven patients recovered with conservative treatment, and their diagnoses of ischemia were considered false-positive (Fig. 2). Eighteen of the 23 patients with a true-positive CT diagnosis of ischemia underwent surgery within 12 hr, three underwent surgery hr after CT, and one underwent surgery 3 days later. Ischemia was caused by an adhesive volvulus in 11 patients, by adhesive bands in seven patients, by a hernia in four patients (3 external, 1 internal), by mesenteric infarction in one patient, and by mesenteric vein thrombosis in one patient. owel resection was necessary in 14 of 23 patients with a true-positive CT diagnosis. The overall sensitivity of CT in the diagnosis of strangulation and ischemia in acute Sign Combination Sensitivity (%) Specificity (%) Positive Predictive Value (%) Negative Predictive Value (%) Mural thickening and mesenteric fluid Mural thickening and mesenteric vascular congestion Mural thickening and ascites Mesenteric fluid and mesenteric vascular congestion Mesenteric fluid and ascites Mesenteric vascular congestion and ascites JR:175, December

4 Zalcman et al. C Fig year-old woman with reversible ischemia. Contrast-enhanced CT scan shows signs of mechanical obstruction of small bowel with dilated and fluid-filled loops in left fossa iliaca (i). Note fluid in mesentery (m) and congestion of small mesenteric veins (arrowheads). Wall of segment of small bowel is barely visible (arrows). Small gas bubble is present in unenhanced loop. Patient underwent surgery same day and had surgical evidence of ischemic bowel due to secondary volvulus. fter devolvulation, ischemic bowel regained normal coloration and resection was not necessary. Patient left hospital 7 days after admission. Fig year-old woman with thrombosis of superior mesenteric vein who presented with clinical signs of intestinal obstruction., Contrast-enhanced CT scan reveals thrombus (arrowhead) in superior mesenteric vein. Second portion of duodenum (d) and proximal jejunum (i) are dilated and fluid-filled., Contrast-enhanced CT scan more caudal shows absence of enhancement (white arrowheads) of dilated jejunal loop, and more proximal loop enhances normally (black arrowheads). Note that nonenhancing loop has no evidence of wall thickening. C, Contrast-enhanced CT scan more caudal than. Jejunal loops have thickened wall with reduced enhancement (arrowheads). Mesentery (m) is hazy. Patient was treated medically with anticoagulant therapy and had uneventful evolution. Months later she developed fibrotic stricture but did not undergo surgery JR:175, December 2000

5 CT of Small-owel Obstruction small-bowel obstruction was thus 96% (23/24) with a specificity of 93% (111/120), a positive predictive value of 72% (23/32), and a negative predictive value of 99% (111/112). Findings on the 144 CT scans of patients with small-bowel obstruction are shown in Table 1. The diagnostic value of the various combinations of two signs of ischemia is shown in Table 2. Reduced Enhancement of the owel Wall Reduced enhancement of the ischemic bowel wall was observed in 10 of the 21 patients with confirmed ischemia who had a contrast material injection (sensitivity, 47.6%) (Figs. 1, 3, and 4). It was never present in noncomplicated obstruction (specificity, 100%). The positive and negative predictive values of this sign were 100% and 90.8% (109/120), respectively. Thickening of the owel Wall This sign, usually considered a cardinal feature of ischemia [16 19], was observed in only nine of the 24 patients with proven ischemia (sensitivity, 37.5%) (Fig. 4C). When only the patients with contrast material injection were considered, the sensitivity was 42.9% (9/21). owel-wall thickening was also present in 26 of 120 patients with noncomplicated obstruction (specificity, 78.3%). s a result, the positive (25.7%) and negative (86.2%) predictive values of this sign were insufficient. In four patients with ischemia, the thickened wall had a target configuration, constituted by the presence of rings of different attenuation, after contrast material injection. This form of wall thickening was more specific of ischemia; it was visible in only four of 109 patients with contrast material injection in the nonischemia group (specificity, 96%). Positive and negative predictive values for the target appearance were 50% and 86%. Twelve of 21 patients had an abnormal enhancement of the bowel wall (2 patients had both a reduced enhancement and target appearance) in the ischemia group (sensitivity, 57%). Mesenteric Fluid Mesenteric fluid was present in 21 of the 24 patients with ischemia (sensitivity, 87.5%) and was also visible in 12 of 120 patients in the noncomplicated group (specificity, 90%). Positive and negative predictive values were 63.6% (21/ 33) and 97.3% (108/111), respectively. Mesenteric Congestion Focal congestion of the small mesenteric veins around the site of obstruction was seen in 14 of 24 patients in the ischemia group (sensitivity, 58.3%) and in 25 of 120 patients in the nonischemia group (specificity, 79.2%). In the ischemia group, one patient only had this feature without evidence of mesenteric fluid. Positive and negative predictive values were 35.9% and 90.5%, respectively. scites scites was present in 18 of 24 patients in the ischemia group (sensitivity, 75%) and 29 of 120 patients in the nonischemia group (specificity, 75.8%). Positive and negative predictive values were 38.3% (18/47) and 93.8% (91/97), respectively. Discussion Recent years have seen a significant improvement in the radiologic diagnosis of acute obstruction of the small bowel [1 5] and its most severe complication, ischemia [9 15]. The advent of fast imaging techniques such as helical CT allows a more accurate assessment of the perfusion of various abdominal organs, including the small bowel. This development, coupled with a better understanding of the CT signs, has played a large part in the progress of the radiologic diagnosis of complicated smallbowel obstruction. ll studies reported until now have nevertheless been incomplete because they were retrospective [13] or the authors arbitrarily selected a subgroup of patients with adhesive bowel obstruction [12, 13]. In daily practice, no clinical or laboratory data differentiate bowel obstruction due to adhesions from obstruction due to anastomotic stricture, recurrent neoplasia, or reactivation of an inflammatory bowel disease. ll these conditions can be associated with CT abnormalities often encountered in ischemia, such as abnormal enhancement of the bowel wall, thickening of the bowel wall, or mesenteric changes, and may influence the specificity of these signs. The objective of our study was to determine prospectively the value of CT in obstruction-associated ischemia and the value of the CT signs we thought were the most reliable for the diagnosis. Unlike three other groups of researchers [12 14], we avoided opacification of the small bowel for several reasons. Obstruction prevents the progression of oral contrast material; ingestion of fluid by a patient with intestinal obstruction will invariably lead to vomiting; and, above all, the natural contrast offered by intestinal fluid allows an optimal study of the bowel wall after contrast material injection. Furthermore, it is illogical and potentially hazardous to delay CT examination up to 2 hr [12, 13] while waiting for the opacification of bowel loops in a patient with abdominal emergency. ll illustrations accompanying the latest reports on strangulation [12 14] show nonopacified ischemic bowel loops, which confirm that there is no indication for intestinal contrast medium in CT performed in the context of small-bowel obstruction. The overall sensitivity and negative predictive value of CT in the diagnosis of strangulation and ischemia were high, 96% and 99%, respectively. The discrepancy between our results and those of the latest extensive study by althazar et al. [14] is probably because more than two thirds of their patients underwent a sequential study of the abdomen instead of a helical study. s reported, reduced enhancement of the bowel wall can be a transient feature and can thus be inconspicuous on a nonhelical scan of the abdomen [20]. Furthermore, angiographic studies of the small-bowel volvulus [21] have shown the possibility of intense bowel-wall opacification in a later phase, which may contribute to the inadequate conspicuity of this sign on scans obtained during an intermediate phase. Reduced enhancement of the bowel wall was the most frequent mural abnormality in our series and had a 100% specificity. It is the only direct sign of vascular impairment of the small bowel and has been reported in predominantly arterial diseases such as mesenteric infarction [22, 23] as well as in predominantly venous diseases such as strangulation [20]. Reduced enhancement of the bowel wall was not associated with necrosis of the small bowel in three of our patients. The first had a small-bowel volvulus with discolored bowel at surgery, which returned to normal after devolvulation (Fig. 3). The second had a strangulated hernia with a cyanosed loop that returned to normal after the cure of the hernia. The third had a superior mesenteric vein thrombosis and did not undergo surgery (Fig. 4). The explanation for this sign is thus not totally clear, but it could at least partly be the result of an arterial spasm caused by the obstruction to the venous outflow; in one patient, it was visible in a nonischemic loop proximal to the diseased segment (Fig. 4). The most unexpected finding of our study was the limited value of mural thickening as an individual sign of ischemia. The target configuration of wall thickening had a high specificity (96%) but was visible in few patients (19%). This discrepancy between our experience and previously reported results is mainly related to the fact that there is no wide consensus regarding the normal thickness and appearance of the small-bowel wall in variable states of distention so that the initial observation of bowel wall JR:175, December

6 Zalcman et al. thickening on CT is usually made subjectively [24]. major cause of error, which has not been reported yet in the literature, is the confusion between true wall thickening and pseudowall thickening due to transverse scanning of Kerckring s folds in a distended fluid-filled bowel. The presence of a gas bubble in this pseudothickening can lead to a false-positive diagnosis of ischemia-induced pneumatosis (Fig. 2) and unnecessary surgery. The high number of patients with a thickened bowel wall in the noncomplicated group was related to two factors: the difficulty of assessing accurately the bowelwall thickness and the fact that virtually half of our patients had medical conditions that can be associated with thickening of the bowel wall (cancer and Crohn s disease). The low sensitivity of this sign is probably multifactorial. Most of our examinations were performed early in the evolution of the obstruction (i.e., at admission), dilatation of the bowel can reduce the conspicuity of moderate mural thickening and, as already mentioned, subtle CT findings are open to subjective interpretations. Fluid in the mesentery adjacent to abnormal (thickened, dilatated, or both) bowel loops was the individual sign most frequently associated with strangulation (88%) and had a good specificity (90%), despite the large proportion of patients with malignant or inflammatory disease in our series. The value of mesenteric changes as signs of small-bowel ischemia has already been reported in previous articles [12, 13, 21] and must be underscored for two reasons. These abnormalities are visible on unenhanced scans, and they have an extremely high negative predictive value. Only two of the 24 patients with ischemia had no mesenteric changes. One patient had two bowel wall abnormalities, including reduced enhancement, and the other one had the false-negative examination with artifacts. We did not find in our previous experience that bowel strangulation was associated with an increase in the attenuation value of ascites. lthough the volume of peritoneal fluid can be larger in patients with strangulation, it does not appear reasonable to consider ascites as a good predictor of ischemia. However, the absence of peritoneal fluid has a high negative predictive value, and thus the search for ascites is not without importance. The benefit of early surgery is underscored by the fact that all patients with small-bowel ischemia who did not undergo surgery the day of the CT diagnosis (n = 4) had a bowel resection and the eight patients with proven ischemia who had no bowel resection underwent surgery the day of the CT diagnosis. It must be added that seven of the nine patients with a false-positive diagnosis of ischemia did not undergo surgery and recovered fully with conservative treatment alone. It is impossible to tell if these cases represent in reality false-positive diagnoses of ischemia or cases of reversible ischemia. It will be interesting in future studies to evaluate whether CT can enable radiologists to differentiate cases of mild reversible ischemia from potentially life-threatening ischemia, when seen at an early stage. In the meantime it seems preferable to accept a somewhat imperfect specificity rather than decrease the high sensitivity of CT. nother issue that needs further evaluation is that of IV contrast material: Is it always necessary and what is the optimal delay between injection and imaging to visualize enhancement abnormalities? In two of 24 patients with confirmed ischemia, the diagnosis was made on an unenhanced examination (Fig. 5) in the presence of mesenteric changes and ascites. Ideally, contrast material injection should be limited to patients who have no clear evidence of ischemia on the initial CT scan of the abdomen. This would reduce the potential hazard of injecting a dehydrated patient, suppress the risk of an adverse reaction to iodine contrast material, and reduce the cost of the diagnostic procedure. If contrast material is necessary, choosing the adequate time-window to study the enhancement of the small bowel is another variable that needs further study. t the beginning of our series, we chose a relatively short delay of 40 sec that was gradually increased to sec to avoid potential diagnostic problems related to the variability of blood flow ve- Fig year-old man with history of allergy to iodine contrast agent., Unenhanced CT scan shows dilated and fluid-filled small-bowel loops (i) occupying left flank. Mesentery (m) is hazy because of presence of triangular-shaped fluid adjacent to dilated loops., Unenhanced CT scan 3 cm more cephalad than reveals presence of congested mesenteric veins (arrowheads). Surgery performed 6 hr later confirmed diagnosis of strangulation. It was caused by internal hernia, and 1.20 m of necrotic bowel was resected JR:175, December 2000

7 CT of Small-owel Obstruction locity between patients. 60- to 70-sec delay between the start of contrast material injection and the start of imaging seems a reasonable compromise between a purely arterial phase with the risk of overdiagnosing enhancement abnormalities and a delayed phase that can render inconspicuous the abnormal enhancement. In summary, we believe that CT should be the initial radiologic examination each time high-grade bowel obstruction is suspected. It should be performed early after admission with a helical acquisition. If the examination is correctly performed and of acceptable technical value, it can show signs of ischemia with a high sensitivity or can safely exclude the presence of ischemia. References 1. Megibow J, althazar EJ, Cho KC, Medwid SW, irnbaum, Noz ME. owel obstruction: evaluation with CT. Radiology 1991;180: Maglinte DD, Gage SN, Harmon H, et al. Obstruction of the small intestine: accuracy and role of CT in diagnosis. Radiology 1993;188: Fukuya T, Hawes DR, Lu CC, Chang PJ, arloon TJ. CT diagnosis of small-bowel obstruction: efficacy in 60 patients. JR 1992;158: Frager D, Medwid SW, aer JW, Mollinelli, Friedman M. CT of small-bowel obstruction: value in establishing the diagnosis and determining the degree and cause. JR 1994;162: Taourel PG, Fabre JM, Pradel J, Seneterre EJ, Megibow J, ruel JM. Value of CT in the diagnosis and management of patients with suspected acute smallbowel obstruction. JR 1995;165: Sarr MG, ulkley G, Zuidema GD. Preoperative recognition of intestinal strangulation obstruction: prospective evaluation of diagnostic capability. m J Surg 1983;145: Mucha P Jr. Small intestinal obstruction. Surg Clin North m 1987;67: Otamiri T, Sjodhal R, Ihse I. Intestinal obstruction with strangulation of the small bowel. cta Chirurgica Scandinavica 1987;153: althazar EJ, irnbaum, Megibow J, Gordon R, Whelan C, Hulnick DH. Closed-loop and strangulating intestinal obstruction: CT signs. Radiology 1992;185: althazar EJ. CT of small-bowel obstruction. JR 1994;162: Ha HK, Park CH, Kim SK, et al. CT analysis of intestinal obstruction due to adhesions: early detection of strangulation. J Comput ssist Tomogr 1993;17: Frager D, aer JW, Medwid SW, Rothpearl, ossart P. Detection of intestinal ischemia in patients with acute small-bowel obstruction due to adhesions or hernia: efficacy of CT. JR 1996;166: Ha HK, Kim JS, Lee MS, et al. Differentiation of simple and strangulated small-bowel obstructions: usefulness of known CT criteria. Radiology 1997;204: althazar EJ, Liebeskind ME, Macari M. Intestinal ischemia in patients in whom small bowel obstruction is suspected: evaluation of accuracy, limitations and clinical implications of CT in diagnosis. Radiology 1997;205: Donckier V, Closset J, Van Gansbeke D, et al. Contribution of computed tomography to decision making in the management of adhesive small bowel obstruction. r J Surg 1998;85: Federle MP, Chun G, Jeffrey R, Rayor R. Computed tomographic findings in bowel infarction. JR 1984;142: lpern M, Glazer GM, Francis IR. Ischemic or infarcted bowel: CT findings. Radiology 1988; 166: Smerud MJ, Johnson CD, Stephens DH. Diagnosis of bowel infarction: a comparison of plain films and CT scans in 23 cases. JR 1990;154: althazar EJ, Hulnick D, Megibow J, Opulencia JF. Computed tomography of intramural intestinal hemorrhage and bowel ischemia. J Comput ssist Tomogr 1987;11: Zalcman M, Van Gansbeke D, Lalmand, raude P, Closset J, Struyven J. Delayed enhancement of the bowel wall: a new sign of small-bowel strangulation. J Comput ssist Tomogr 1996;20: Kadir S, thanasoulis C, Greenfeld J. Intestinal volvulus: angiographic findings. Radiology 1978;128: Klein H-M, Lensing R, Klosterhalfen, Töns C, Günther RW. Diagnostic imaging of mesenteric infarction. Radiology 1995;197: Taourel PG, Deneuville M, Pradel J, Regent D, ruel JM. cute mesenteric ischemia: diagnosis with contrast-enhanced CT. Radiology 1996;199: Low VHS. owel wall thickening on CT. bdom Imaging 1998;23: JR:175, December