Thompson et al.

Radiography of Small-Bowel Obstruction

Abdominal Imaging Original Research

Accuracy of Abdominal Radiography in Acute Small-Bowel Obstruction: Does Reviewer Experience Matter?
William M. Thompson1 Ramsey K. Kilani Benjamin B. Smith John Thomas Tracy A. Jaffe David M. Delong Erik K. Paulson
Thompson WM, Kilani RK, Smith BB, et al. OBJECTIVE. The purposes of this study were to determine the accuracy of abdominal radiography in the detection of acute small-bowel obstruction (SBO), to assess the role of reviewer experience, and to evaluate individual radiographic signs of SBO. MATERIALS AND METHODS. A retrospective study was performed in which the subjects were 90 patients with suspected SBO who underwent CT and abdominal radiography within 48 hours of each other. The patients were enrolled from June 1, 2003, to February 2004. Twentynine of the patients had proven SBO. Hard-copy radiographs were reviewed by three groups of radiologists: senior staff, junior staff, and second-year radiology residents. Each reviewer evaluated the quality of the radiographs, patient position for acquisition of the radiographs, and whether SBO was present. The reviewers rated their confidence on a five-point scale and recorded the presence or absence of specific radiographic signs of SBO. Chi-square tests were used to compare the three groups. A statistically significant finding was considered p < 0.05. Receiver operating characteristic (ROC) curves were fit with a 10-point confidence scale. RESULTS. The sensitivity for SBO among the six reviewers ranged from 59% to 93%. The senior staff members were significantly more accurate. The mean sensitivity, specificity, and accuracy for all six reviewers were 82%, 83%, and 83%, respectively. Three radiographic signs were highly significant (p < 0.001): two or more airfluid levels, airfluid levels wider than 2.5 cm, and airfluid levels differing more than 5 mm from one another in the same loop of small bowel. ROC analysis showed that senior staff is significantly more accurate than the other groups in the detection of acute SBO. CONCLUSION. Our results confirmed that abdominal radiographs are accurate in the detection of acute SBO, that more-experienced radiologists are more accurate than less-experienced reviewers in the evaluation of abdominal radiographs, and that three types of airfluid levels are highly predictive of the presence of SBO. T has become the preferred imaging technique for evaluation of patients with suspected small-bowel obstruction (SBO) [121]. Despite reports of success with CT, some authors [2225] find that conventional abdominal radiography is the preferred initial examination of patients with symptoms of acute SBO. There is concern, however, about the accuracy of abdominal radiography in patients with proven SBO, the reported accuracy ranging from 50% to 60% [11, 2629]. A recent article from our institution [21] reported that CT scout digital radiography alone had a sensitivity of 88% and a specificity of 86%, similar to the results reported for axial CT and axial plus coronal CT. These findings suggest that in the clinical setting of suspected acute SBO, abdominal radiography may be both highly sensitive and

Keywords: abdominal imaging, radiography, small bowel DOI:10.2214/AJR.06.0817 Received June 26, 2006; accepted after revision August 18, 2006. Presented at the 2006 annual meeting of the American Roentgen Ray Society, Vancouver, BC, Canada.
1All authors:

Department of Radiology, Duke University Medical Center, Box 3808, Durham, NC 27710. Address correspondence to W. M. Thompson (thomp132@mc.duke.edu). WEB This is a Web exclusive article.

AJR 2007; 188:W233W238 0361803X/07/1883W233 American Roentgen Ray Society

highly specific. To our knowledge, there are no reports in the literature on evaluations of the level of experience of the radiologists, particularly residents, interpreting abdominal radiographs of patients with suspected acute SBO. A number of signs on abdominal radiographs have been reported to be sensitive and specific for differentiating high-grade SBO from low-grade SBO and normal intestine [30]. The best radiographic signs are more than two airfluid levels, airfluid levels wider than 2.5 cm, and airfluid levels differing more than 5 mm from one another in the same small-bowel loop [30]. Other findings, such as gasless abdomen, the string-of-pearls sign, distended stomach, the stretch sign, and collapsed colon, have not been shown to be as predictive as the other three signs of the presence of high-grade SBO [30]. The purposes of this study were to deter-

AJR:188, March 2007

W233

Thompson et al.
nior staff members with 15 (S1) and 30 (S2) years of experience after radiology residency; group 2, two junior radiology staff members with 4 (J1) and 5 (J2) years of experience after training; and group 3, two second-year radiology residents (R1 and R2). The reviewers were asked to evaluate the following: quality of study (adequate or inadequate) and patient position for radiography (flat, upright, or left lateral decubitus). The reviewers then were asked whether there was radiographic evidence of SBO and to rate on a five-point scale (15, low to high) their confidence with the interpretation. Reviewers were asked to document the presence or absence of the following: more than two airfluid levels, airfluid levels measuring 2.5 cm or wider, airfluid levels of different heights in the same loop of small bowel (Fig. 1), string-of-pearls sign, gasless abdomen, dilated colon, dilated stomach or duodenum, nasogastric tube, free air, portal venous gas, biliary gas, ascites, and abdominal mass. The number of patients without either an upright or a decubitus radiograph was tabulated for the true-positive, true-negative, false-positive, and falsenegative findings for each of the six reviewers.

Fig. 138-year-old woman with closed-loop smallbowel obstruction found at surgery. Highly predictive airfluid levels are visible. Upright anteroposterior abdominal radiograph shows all three highly predictive types of airfluid levels: two or more levels (thick arrows), levels with a width of 2.5 cm or more (long line), and levels differing 5 mm or more from one another in same loop (thin arrow, short lines).

mine the sensitivity, specificity, and accuracy of abdominal radiography for acute SBO, to assess the role of reviewer experience, and to evaluate the individual radiographic signs reported to be sensitive in the diagnosis of SBO. Materials and Methods
This retrospective study, which was compliant with the Health Insurance Portability and Accountability Act, was approved by the institutional review board at our medical center, which waived informed consent. From June 1, 2003, to February 2004, 100 consecutively registered patients with suspected SBO underwent 16-MDCT and formed the study group. Ninety of these patients also underwent abdominal radiography within 48 hours of CT. The medical records and surgical and pathologic reports were reviewed to establish a diagnosis. A diagnosis of SBO (n = 29) was made if the patient had surgical proof (n = 13 [45%]) or had relief of obstructive symptoms after bed rest and nasogastric tube suction (n = 16 [55%]). SBO was excluded if an alternative diagnosis was established and the patient was treated (n = 40) or if the patient did not undergo exploratory laparotomy; the symptoms resolved without nasogastric tube suction; and there was no evidence of persistent pain, abscess, or unexplained fever during the hospital stay (n = 21). Of the 90 patients, 49 were women and 41 were men. The mean age was 55 years (range, 3094 years).

Statistical Analysis
Sensitivity, specificity, and combined sensitivity and specificity (a measure of accuracy) were tabulated for each reviewer and combined for comparisons among the three groups. The groups were compared with use of chi-square tests. Wilcoxons rank sum test was used to compare the six reviewers results between patients with and those without upright or decubitus radiographs. The association between the outcome variable proven SBO and the other variables was assessed by means of a Wilcoxon and Mann-Whitney test for continuous variables and Pearsons chi-square test for binary-value variables. Significance was considered p < 0.05 in all statistical tests. Comparison of sensitivity and specificity estimates among groups were based on across-reviewer average scores. Receiver operating characteristic (ROC) curves from Students t tests were fit in a semiparametric binormal model [31] with a 10-point confidence scale. Areas under the ROC curves were assessed by means of the Wilcoxon and Mann-Whitney test as applied with methods of Delong et al. [32].

Radiographic Evaluation
Hard-copy radiographs were loaded on several viewers and interpreted by three groups of radiologists with wide variation in training: group 1, two se-

Results A mean of 8.5 (10%) of the 90 studies (range, 1 senior staff member study to 20 resident studies) were rated inadequate, most because of the lack of an upright or decubitus radiograph and others mostly because of motion. However, only one case of proven SBO in this group was misclassified by all six reviewers. The patient had a gasless abdomen that could be visualized only on a supine radiograph with only minimal motion. Supine abdominal ra-

diographs were obtained for all 90 patients; either upright or decubitus images were not obtained in the cases of 23 (26%) of the 90 patients. Either upright (n = 15) or left lateral decubitus radiographs (n = 7) were obtained in the cases of 22 (76%) of the 29 patients with proven SBO (Figs. 14). Either upright (n = 31) or left lateral decubitus radiographs (n = 14) were obtained in the cases of 45 (74%) of the 61 patients without SBO (Table 1). Sensitivity for the detection of SBO ranged from 59% to 93% for the six radiologists (Table 2). In comparisons of the three groups of radiologists, senior staff had a statistically significantly higher rate of detection of SBO than either junior staff (p < 0.001) or residents (p < 0.004). Residents had statistically significantly better sensitivity than junior staff (p < 0.02). There was less variation of specificities and there were no statistically significant differences in specificity among the three groups of radiologists. For the combination of sensitivity and specificity, which is a measure of accuracy, senior staff were statistically significantly more accurate than either of the other two groups, but there was no difference between junior staff and residents (Table 2). The mean sensitivity, specificity, and combined sensitivity and specificity for all six reviewers were 82%, 83%, and 83%, respectively (Table 2). There were 174 possible positive responses for SBO (six reviewers 29 proven cases of SBO) and 366 possible true-negative responses (six reviewers 61 no SBO). The residents had more false-positive findings (mean, 12.5) (Fig. 5) than the senior (mean, 10) and junior staff (mean, nine). The six reviewers collectively had 33 (19%) false-negative findings, and 12 (36%) of the patients in these cases did not undergo upright or decubitus radiography (Fig. 6), whereas 29 (21%) of the patients with true-positive findings did (Figs. 1, 3, 4, and 7). Nine (14%) of the patients with false-positive findings and 86 (28%) of those with true-negative findings did not undergo either upright or decubitus radiography. There were no significant differences (p > 0.05) among these four groups for the six reviewers. The lack of upright or decubitus images did not influence the detection or exclusion of SBO. ROC analysis of confidence showed a statistically significant difference between senior staff and the other two groups (p < 0.005 and p < 0.05) but no difference between junior staff and residents (p > 0.52). Three radiographic signs were highly significant in prediction of the presence of SBO: more than two airfluid levels (Figs. 1, 3, and 4), airfluid levels wider than 2.5 cm (Figs. 1, 3,

W234

AJR:188, March 2007

Radiography of Small-Bowel Obstruction

Fig. 242-year-old woman with abdominal pain. Truenegative findings were reported. A and B, Supine (A) and upright (B) abdominal radiographs show multiple nondilated air-containing loops (arrows) of small bowel in left side of abdomen. B shows no significant airfluid levels. CT on same day as radiography (not shown) did not reveal small-bowel obstruction, and symptoms resolved.

B
Fig. 378-year-old woman with abdominal pain, nausea, and vomiting. True-positive findings were reported. A, Supine abdominal radiograph shows multiple dilated loops (arrows) of small bowel in right lower quadrant. B, Upright radiograph shows multiple airfluid levels in right lower quadrant, some of which are wider than 2.5 cm, and airfluid levels (arrows) of unequal height. Patient was treated conservatively, and obstruction resolved.

Fig. 483-year-old man with abdominal pain. Truepositive findings were reported. A, Upright radiograph shows dilated small bowel (arrows) in left lower quadrant. B, Left lateral decubitus radiograph shows multiple airfluid levels (arrows), some of which are wider than 2.5 cm and of unequal heights. Patient was treated conservatively, and obstruction resolved.

AJR:188, March 2007

W235

Thompson et al. TABLE 1: Position for Plain Abdominal Radiographs of 90 Patients with Suspected Small-Bowel Obstruction
Position Flat Upright Left lateral decubitus Neither upright nor decubitus Total No Obstruction 61 31 14 16 61 Proven Obstruction 29 15 7 7 29 Total 90 46 21 23 90

TABLE 2: Sensitivity and Specificity for Detection of Small-Bowel Obstruction

Reviewer Senior staff S2 S1 Mean Junior staff J1 J2 Mean Second-year residents R1 R2 Mean Mean for all three groups
a p < 0.001, senior staff b p > 0.08,

Sensitivity (%)a 93 90 92 79 59 69 86 79 83 82

Specificity (%)b 79 89 84 82 89 86 77 82 80 83

Combined Sensitivity and Specificityc 86 90 88 81 74 77 84 81 83 83

and 4), and airfluid levels differing more than 5 mm from one another in the same loop (Table 3) (Figs. 1, 3, and 4). Nineteen (42%) of the 45 patients without SBO had multiple airfluid levels compared with 21 (95%) of the 22 with SBO. Airfluid levels measuring 2.5 cm or wider and airfluid levels of unequal heights in the same loop were present in 15 (68%) of 22 and 14 (64%) of 22, respectively, of the patients with acute SBO but only in eight (18%) of 45 and two (4%) of 45, respectively, of the patients without SBO. Although dilated colon (p < 0.05) and the string-of-pearls sign (p < 0.001) were statistically significant discriminating signs in exclusion or detection of SBO, these signs rarely occurred in the patients. One (3%) of the 29 patients with SBO had dilated colon, and only eight (14%) of the 61 patients without SBO had this sign. The string-of-pearls sign occurred in only three (10%) of 29 SBO patients (Fig. 7) and one (1.7%) of 61 patients without SBO. None of the other findings was helpful in excluding or detecting SBO. Discussion The overall sensitivity, specificity, and accuracy of abdominal radiography in the detection of acute SBO were higher in our series than in most previous reports. Frager et al. [10] reported only 19% sensitivity of radiography, compared with 100% sensitivity of CT, in examinations of 36 patients with suspected SBO. Maglinte et al. [11] in 1996 reported a sensitivity of 67% for both abdominal radiography and CT in patients with low-grade and those with high-grade SBO. With classification into highgrade obstruction, the sensitivity increased to 86%, a finding similar to ours. Shrake et al. [22], however, described 117 consecutively examined patients with clinically suspected SBO and found the abdominal radiographic findings were misleading. Twenty-two percent of patients with normal findings had proven obstruction, and 42% of patients with abnormal findings did not have obstruction. The overall sensitivity was 66%. A number of authors have reported only 5060% accuracy of abdominal radiography in the detection of SBO [2629]. Our results suggest that abdominal radiography is accurate in the evaluation of patients with suspected acute SBO. The findings can be used to screen patients for whom a physician may consider ordering CT. Radiography also is a relatively inexpensive way of following patients who have been treated for SBO. This form of management has become more important with the increasing emphasis on nonsurgical treatment of

compared with junior staff; p < 0.02, junior staff compared with residents; p < 0.04 senior staff compared with residents. all three comparisons of specificity. c p > 0.003 senior staff compared with junior staff; p < 0.18 junior staff compared with residents; p < 0.019 senior staff compared with residents.

Fig. 574-year-old man with abdominal distention. False-positive findings were reported. A and B, Supine (A) and upright (B) abdominal radiographs show dilated small bowel out of proportion to colon; condition causes airfluid levels (arrows, B). CT (not shown) performed later same day revealed only ascites and no evidence of small-bowel obstruction.

W236

AJR:188, March 2007

Radiography of Small-Bowel Obstruction

A
Fig. 691-year-old woman with abdominal pain. Falsenegative findings were reported. Supine radiograph shows little small-bowel gas. Five of six reviewers rated study inadequate owing to motion. All reviewers interpreted findings as no obstruction. CT (not shown) and surgery on same day revealed incarcerated right femoral hernia, which produced marked small-bowel obstruction with fluid-filled loops of dilated small bowel.

Fig. 748-year-old woman with nausea and vomiting. True-positive findings were reported. Gasless abdomen and string-of-pearls sign were seen. A, Supine abdominal radiograph shows paucity of small-bowel gas and suggestion of stretch sign valvulae conniventes in dilated small-bowel loop outlined by air (arrow). Stretch sign is defined as abnormal distention of predominantly fluid-filled small-bowel loops in which luminal gas has striped appearance running perpendicular to long axis of bowel [26]. B, Upright radiograph shows multiple tiny airfluid levels (arrows) in small bowel (string-of-pearls sign). Smallbowel obstruction due to adhesions was found at surgery.

patients who have undergone abdominal surgery and are likely to have an adhesion causing obstruction. Abdominal radiography also is less expensive and requires less radiation than CT. To our knowledge, no studies have been conducted to examine the role of reviewer experience in the evaluation of abdominal radiographs of patients with suspected SBO. In 1989 Markus et al. [33] studied interobserver variation among four radiologists in the interpretation of abdominal radiographs but did not examine experience per se. Those authors found fair to good agreement for SBO. In our comparison of three groups of radiologists whose levels of experience differed, we found a statistically significant difference in sensitivity and accuracy among the most experienced radiologists compared with those with less than 5 years of experience. There was variation in the two senior staff: S1 had a slightly high sensitivity (93% vs 90%) but a lower specificity (77%) than S2 (90%). Although there was no statistically significant difference in specificity for the two junior staff members, there was a difference in sensitivity. J1 had a sensitivity (82%) similar to that of the residents, but J2 had a much lower sensitivity (59%). J2, however, had a specificity of 89%, which was higher than that of any of the reviewers, except for senior staff member S2. We have no good explanation for these variations. Most abdominal radiographs are interpreted by members of the

gastrointestinal fluoroscopic service, which is usually covered by senior staff. Our junior staff rarely work on this service and therefore may lack the experience of the senior staff. Additional evidence that experience plays a role in interpretation of abdominal radiographs of patients with suspected SBO was found in the ROC analysis. Senior staff had significantly more confidence than either junior staff or residents. There was no difference between junior staff and residents. Senior staff members were consistently more confident in detecting both the presence and the absence of SBO. Our findings corroborate those of Lappas et al. [30], who found that the two most significant

(p < 0.0003) radiographic signs of SBO were airfluid levels of different heights or an airfluid level width of 2.5 cm or more. Lappas et al. also found that when both of these findings were present, the degree of SBO was likely high grade or complete. The findings of Lappas et al. contrast to those of other investigators [3338], who have not found the presence of multiple airfluid levels and airfluid levels of differing heights helpful in the diagnosis of SBO. Mirvis et al. [38] found that findings on upright radiographs did not add to those on supine radiographs of patients with proven SBO. In our study, three patients who had SBO had the string-of-pearls sign, which has been shown to be highly sensitive in

TABLE 3: Evaluation of Plain Abdominal Radiographic Signs of Small-Bowel Obstruction on 46 Upright and 21 Decubitus Radiographs
Sign Multiple airfluid levels Position Upright Decubitus Total Airfluid levels wider than 25 mm Upright Decubitus Total Airfluid levels at different heights Upright Decubitus Total No Small-Bowel Obstruction 14/31 5/14 19/45 (42) 7/31 1/14 8/45 (18) 2/31 0/14 2/45 (4) Small-Bowel Obstruction 14/15 7/7 21/22 (95) 14/15 4/7 15/22 (68) 10/15 4/7 14/22 (64)

Notep < 0.001 for all three signs of small-bowel obstruction. Values in parentheses are percentages.

AJR:188, March 2007

W237

Thompson et al. the diagnosis of SBO [26, 31], on upright or decubitus images. Although the role of upright and decubitus radiography in the detection of highly predictive airfluid levels is still in question, we believe these images do add additional information. This point was emphasized by Mindelzun and McCort [39]. None of the other signs evaluated in our study were predictive of SBO. Dilated colon, however, was highly suggestive that a patient did not have SBO. There were a number of limitations to our study. First, it was retrospective, and the reviewers had only two choices: SBO or no SBO. Second, 26% of our patients did not undergo upright or decubitus radiography. Third, 16 (55%) of 29 patients with SBO did not undergo surgery. Their diagnosis was based on clinical improvement with nasogastric suction. Fourth, there was a wide range in what was considered an adequate or an inadequate study (one inadequate study for a senior staff member to 20 for a resident) because we did not provide instruction about what was considered an adequate examination. This factor, however, did not affect any of the six reviewers results because lack of upright or decubitus radiographs was the most common cause of inadequate studies. In addition, there were no differences among the six reviewers results in comparison of patients with and those without upright or decubitus radiographs. The results of this study confirm that abdominal radiography is sensitive and specific in the detection of acute SBO. Radiologists with a greater degree of experience are likely to be more accurate in the evaluation of abdominal radiographs. Three specific types of airfluid levels are highly predictive of SBO.
guishing small-bowel obstruction from other causes of small-bowel dilatation. AJR 1994; 162:4347 Frager D, Medwid SW, Baer JW, Mollinelli B, Friedman M. CT of small bowel obstruction: value in establishing the diagnosis and determining the degree and cause. AJR 1994; 162:3741 Balthazar EJ. CT of small-bowel obstruction. AJR 1994; 162:255261 Taourel PG, Fabre JM, Pradel JA, Seneterre EJ, Megibow AJ, Bruel JM. Value of CT in the diagnosis and management of patients with suspected acute small bowel obstruction. AJR 1995; 165:11871192 Frager DH, Baer JW, Rothpearl A, Bossart PA. Distinction between postoperative ileus and mechanical small-bowel obstruction: value of CT compared with clinical and other radiographic findings. AJR 1995; 164:891894 Maglinte DD, Reyes BL, Harmon BH, et al. Reliability and role of plain film radiography and CT in the diagnosis of small-bowel obstruction. AJR 1996; 167:14511455 Ha HK, Kim JS, Lee MS, et al. Differentiation of simple and strangulated small-bowel obstructions: usefulness of known CT criteria. Radiology 1997; 204:507512 Balthazar 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; 205:519522 Zalcman M, Sy M, Donckier V, Closset J, Gansbeke DV. Helical CT signs in the diagnosis of intestinal ischemia in small-bowel obstruction. AJR 2000; 175:16011607 Caoili EM, Paulson EK. CT of small-bowel obstruction: another perspective using multiplanar reformations. AJR 2000; 174:993998 Boudiaf M, Soyer P, Terem C, Pelage JP, Maissiat E, Rymer R. CT evaluation of small bowel obstruction. RadioGraphics 2001; 21:613624 Furukawa A, Yamasaki M, Furuichi K, et al. Helical CT in the diagnosis of small bowel obstruction. RadioGraphics 2001; 21:341355 Khurana B, Ledbetter S, McTavish J, Wiesner W, Ros PR. Bowel obstruction revealed by multidetector CT. AJR 2002; 178:11391144 Horton KM, Fishman EK. The current status of multidetector row CT and three-dimensional imaging of the small bowel. Radiol Clin North Am 2003; 41:199212 Lazarus DE, Slywotsky C, Bennett GL, Megibow AJ, Macari M. Frequency and relevance of the small-bowel feces sign on CT in patients with small-bowel obstruction. AJR 2004; 183:13611366 Jaffe TA, Martin LC, Thomas J, Adamson AR, Delong DM, Paulson EK. Small bowel obstruction: coronal reformations from isotropic voxels at 16section multidetector row CT. Radiology 2005; 238:135142 22. Shrake PD, Rex DK, Lappas JC, Maglinte DD. Radiographic evaluation of suspected small bowel obstruction. Am J Gastroenterol 1991; 86:175178 23. Maglinte DD, Balthazar EJ, Kelvin FM, Megibow AJ. The role of radiology in the diagnosis of smallbowel obstruction. AJR 1997; 168:11711180 24. Maglinte DD, Heitkamp DE, Howard TJ, Kelvin FM, Lappas JC. Current concepts in imaging of small-bowel obstruction. Radiol Clin North Am 2003; 41:263283 25. Maglinte DD, Kelvin FM, Sandrasegaran K, et al. Radiology of small bowel obstruction: contemporary approach and controversies. Abdom Imaging 2005; 30:160178 26. Mucha P Jr. Small intestinal obstruction. Surg Clin North Am 1987; 67:597620 27. Laws HL, Aldrete JS. Small bowel obstruction: a review of 465 cases. South Med J 1976; 69:733734 28. Barnett WO, Petro AB, Williamson JW. A current appraisal of problems with gangrenous bowel. Ann Surg 1976; 183:653659 29. Leffall LD, Syphax B. Clinical aids in strangulation intestinal obstruction. Am J Surg 1970; 120:756759 30. Lappas JC, Reyes BL, Maglinte DD. Abdominal radiography findings in small bowel obstruction: relevance to triage for additional diagnostic imaging. AJR 2001; 176:167174 31. Metz CE. Basic principles of ROC analysis. Semin Nucl Med 1978; 8:283298 32. Delong ER, Delong DM, Clarke-Pearson DL. Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics 1988; 44:837845 33. Markus JB, Somer SS, Franic SE, Moola C, Stevenson GW. Interobserver variation in the interpretation of abdominal radiographs. Radiology 1989; 171:6971 34. Gammill SL, Nice CM Jr. Air fluid levels: their occurrence in normal patients and their role in the analysis of ileus. Surgery 1972; 71:771780 35. Levin B. Mechanical small bowel obstruction. Semin Roentgenol 1973; 13:281297 36. Simpson A, Sandeman D, Nixon SJ, Goulbourne IA, Grieve DC, Macintyre IM. The value of an erect abdominal radiograph in the diagnosis of intestinal obstruction. Clin Radiol 1985; 36:4142 37. Harlow CL, Stears RL, Zeligman BE, Archer PG. Diagnosis of bowel obstruction on plain abdominal radiographs: significance of airfluid levels at different heights in the same loop of bowel. AJR 1993; 161:291295 38. Mirvis SE, Young JW, Keramati B, McCrea ES, Tarr R. Plain film evaluation of patients with abdominal pain: are three radiographs necessary? AJR 1986; 147:501503 39. Mindelzun RE, McCort JJ. Answer to question. Radiology 1996; 166:716719

7.

8. 9.

10.

11.

12.

13.

14.

15.

16.

References
1. Rubesin SE, Herlinger H. CT evaluation of bowel obstruction: a landmark articleimplications for the future. Radiology 1991; 180:307308 2. Megibow AJ, Balthazar EJ, Cho KC, Medwid SW, Birnbaum BA, Noz ME. Bowel obstruction: evaluation with CT. Radiology 1991; 180:313318 3. Fukuya T, Hawes DR, Lu CC, Chang PJ, Barloon TJ. CT diagnosis of small-bowel obstruction: efficacy in 60 patients. AJR 1992; 158:765769 4. Balthazar EJ, Birnbaum BA, Megibow AJ, Gordon RB, Whelan CA, Hulmich DH. Closed-loop strangulating intestinal obstruction: CT signs. Radiology 1992; 185:769775 5. Maglinte DD, Gage SN, Harmon BH, et al. Obstruction of the small intestine: accuracy and role of CT in diagnosis. Radiology 1993; 188:6164 6. Gazelle GS, Goldberg MA, Wittenberg J, Halpern EF, Pinkney L, Mueller PR. Efficacy of CT distin17.

18.

19.

20.

21.

W238

AJR:188, March 2007