Computed tomography without oral contrast solution for blunt diaphragmatic injuries in abdominal trauma

American Journal of Emergency Medicine (2005) 23, 253 – 258

www.elsevier.com/locate/ajem

Computed tomography without oral contrast solution for blunt diaphragmatic injuries in abdominal trauma Todd L. Allen MDa,*, Brendan F. Cummins MDa, R. Thomas Bonk MDb, Colleen P. Harker MDb, Diana L. Handrahan BSc, Mark H. Stevens MDd a

Department of Emergency Medicine, LDS Hospital, Salt Lake City, UT 84143, USA Department of Radiology, LDS Hospital, Salt Lake City, UT 84143, USA c Statistical Data Center, LDS Hospital, Salt Lake City, UT 84143, USA d Department of Surgery, LDS Hospital, Salt Lake City, UT 84143, USA b

Received 22 April 2004; accepted 28 June 2004

Abstract Objective: The aim of this study was to estimate the sensitivity, specificity, and positive predictive value (PPV) of computed tomography (CT) without oral contrast for diaphragm injuries (DIs) in blunt abdominal trauma. Methods: We prospectively enrolled 500 consecutive btrauma-oneQ patients who received CT imaging and interpretation (CT-Read1) of the abdomen within 45 minutes of their arrival from July 2000 to December 2001. All patients were imaged without oral contrast but with intravenous contrast. Computed tomographic images were reviewed within 24 hours of admission by research radiologists (CT-Read2) blinded to CT-Read1. True DIs were determined hierarchically by either laparotomy or autopsy. Results: There were 9 patients with laparotomy or autopsy-proven blunt DIs; 8 of these injuries involved the left hemidiaphragm. The CT-Read1 correctly detected only 6 of 9 blunt DIs, thus missing 3 DIs. One of these involved the right hemidiaphragm, whereas the other 2 were left sided. There were no false-positive findings with CT-Read1 for blunt DI. The sensitivity and specificity of CT imaging with respect to DI were 66.7% (95% CI, 29.9%-92.5%) and 100% (95% CI, 99.2%-100%), respectively. The PPV for the test was 1.00 (95% CI, 0.65-1.00). Conclusion: Although the low number of blunt DIs in this study limits its general applicability, CT imaging of the diaphragm without oral contrast appears to perform within the range of reported imaging techniques using oral contrast. Still, CT scanning appears to have an unsatisfactorily low sensitivity to be reliably used in eliminating the diagnosis of blunt DI. D 2005 Elsevier Inc. All rights reserved.

1. Introduction T Corresponding author. Tel.: +1 801 408 8542; fax: +1 801 408 3185. E-mail address: [email protected] (T.L. Allen). 0735-6757/$ – see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.ajem.2005.02.020

Computed tomography (CT) with oral and intravenous contrast began to be used in the early 1980s to evaluate victims of blunt abdominal trauma (BAT). Discussion of the

254 utility of CT imaging with respect to specific abdominal injuries began soon thereafter [1-5]. Both oral and intravenous (and sometimes rectal) contrast materials were used in early scanning protocols. Computed tomography using oral contrast material has historically been advocated for the detection of bowel injuries and DIs in BAT. However, the recent medical literature has begun to question the theoretical advantage provided by oral contrast media in conjunction with CT scanning for these abdominal injuries [6-8]. As of yet, there is no specific discussion in the literature of the utility of oral contrast material in aiding the diagnosis of blunt DI. Diaphragm injuries (DIs) are thought to occur in approximately 1% to 8% of patients with BAT [3,9-14]. The detection of DIs can be difficult as the clinical and radiographic findings may be obscure, especially if the injury occurs in isolation. A prompt diagnosis is important, however, as increased morbidity and mortality are likely to be associated with a delay in the diagnosis and treatment of DIs [15]. An accurate diagnosis is arguably even more critical as the radiographic identification of DI mandates laparotomy, and thus a low positive predictive value (PPV) for DI using CT without oral contrast would lead to an unacceptably high rate of negative laparotomy for the injury. The present study was therefore initiated to better estimate the sensitivity, specificity, and PPV of CT imaging of the abdomen and pelvis without oral contrast solution with respect to DI. A study of the accuracy of this technique with respect to blunt bowel and mesenteric injuries was also part of our investigation and is reported elsewhere [16].

2. Materials and methods This prospective, nonblinded cohort study was conducted at a 520-bed tertiary and teaching referral center that serves a major metropolitan area and is a referral center for a 3-state region. The facility is an American College of Surgeons–verified level I trauma facility with full medical and surgical specialties, subspecialties, and ancillary services available 24 hours per day. The emergency department has a census of nearly 38 000 patients per year. Five hundred consecutive victims of BAT who were determined to require CT scanning of the abdomen and pelvis as part of a clinical assessment between July 2000 and December 2001 were enrolled in the study. Only those patients with blunt injury designated as btrauma one Q (those requiring the highest level of trauma team activation) based on specific physiological and/or anatomical findings were included (Table 1). Patients who were younger than 18 years, pregnant, or in the custody of police or state services were excluded from the study. In addition, patients were excluded if they received oral contrast material at a referring facility or if

T.L. Allen et al. Table 1 Physiological and anatomical standards required for designation as a btrauma-oneQ patient at our institution Physiological criteria Glasgow Coma Score V12 Systolic blood pressure b90 mm Hg at any time Respiratory rate b8 or N30 Revised Trauma Score V11 Intubated or question of airway security Transferred from outside facility receiving blood products Anatomical criteria All penetrating injuries to the head, neck, chest, abdomen (including back) Penetrating injuries to the extremities proximal to the elbow or knee Amputation or degloving injury proximal to the ankle or wrist Flail chest Suspected spinal cord injury with paralysis Open or depressed skull fracture Combination of trauma with burns

diagnostic peritoneal lavage (DPL) was performed before CT. The study was approved by the hospital’s institutional review board. All patients underwent CT scanning without the use of oral contrast material. Nasogastric tubes were placed at the discretion of the trauma surgeon but were not a routine protocol element. Computed tomographic images were acquired using a General Electric CTI helical scanner (GE Medical Systems, Milwaukee, Wis) using 7-mm section thickness at a 1.5:1 pitch and reconstructed at 7 mm in standard algorithm from the dome of the diaphragm to the symphysis pubis. All patients received 150 mL of iopamidol Isovue-300 intravenous contrast material (Bracco Diagnostics Inc, Princeton, NJ) by a mechanical injector at 2.0 mL/s. Examinations were reviewed at Impax digital workstations (AGFA Corporation, version 4.1, Ridgefield, NJ). The CT images were interpreted immediately in consensus by the attending radiologist, trauma surgeon, and the chief surgical resident (CT-Read1). The chief surgical resident completed a research worksheet for all study patients at the time of imaging. Recorded demographic data included age, sex, mechanism of injury, and Glasgow Coma Score. All intraabdominal injuries apparent by CT scan were detailed on the research worksheet for each enrolled patient. The worksheet for CT-Read1 specifically noted findings germane to DI including direct discontinuity of the diaphragm, intrathoracic herniation of abdominal contents, the collar sign, and the dependent viscera sign [12,14,17-19]. Nonspecific thickening of the diaphragm visualized on CT was not recorded [20]. Initial patient management decisions were based on CT-Read1. All scans were reread by 1 of 2 study radiologists (CTRead2), both of whom are fellowship trained in CT body imaging, within 24 hours. The study radiologist was blinded

CT without OC for Diaphragm Injuries Table 2 patients

255 recommendations for study continuance to the institutional review board. The sensitivity, specificity, PPV, and NPV along with 95% CIs of CT without oral contrast were calculated with respect to blunt DIs (SPSS version 12.0, SPSS Inc, Chicago, Ill) [21].

Demographic and injury information on enrolled

n Mean age (y) Men Women Mechanism [n (%)] MVC MCA Fall Assault Bike Pedestrian Other GCS (mean) SD GCS median ISS (mean) SD ISS median

With DI

Without DI

9 47.4 3 6

491 34.5 335 156

7 (77.8) 0 1 (11.1) 0 0 1 (11.9) 0 10.7 5.1 14 34.4 12.2 34

318 (64.8) 35 (7.1) 35 (7.1) 16 (3.3) 6 (1.2) 52 (10.6) 29 (5.9) 11.9 4.7 14 18.3 12.6 17

3. Results

MVC indicates motor vehicle collision; MCA, motorcycle accident; GCS, Glasgow Coma Score; ISS, Injury Severity Score.

to CT-Read1 and the patient’s clinical course. We were able to obtain data on all 500 consecutive patients. The presence of DI at laparotomy or autopsy served as the gold standard for the purpose of calculating the sensitivity, specificity, PPV, and negative predictive value (NPV) of CT without oral contrast for DI. The prior probability for DI in BAT was set at .05. The recorded presence of any CT sign for DI on CT-Read1 was considered indicative for DI and used to calculate sensitivity and specificity with regard to truly or falsely identified injuries. In addition, a 3-month telephone follow-up was attempted to identify any missed injury up to that time. An independent data monitoring committee provided an interim study safety evaluation after enrollment of 100, 200, and 400 patients. It reported its findings and

Table 3

A total of 500 patients was enrolled in the study. A summary of the demographic and injury data of patients with DI compared with those without DI is presented in Table 2. Motor vehicle collisions were the mechanism of injury for 77.8% of patients with and 64.8% of those without DI. The presenting Glasgow Coma Score was lower and the Injury Severity Score was higher for patients with DI compared with enrolled patients without DI. The mean age of the 2 patient populations was also somewhat different. By our study criteria, DI was determined to be present if laparotomy or autopsy identified DI. Using these criteria, 9 study patients, therefore, had true DI. Of these, 6 were identified by CT-Read1 for a sensitivity of 66.7% (95% CI, 29.9%-92.5%). Each of these 6 patients had intrathoracic herniation of abdominal contents noted by CT images. There were 3 false-negative interpretations of CT-Read1. Four hundred ninety-one study patients were thought to have no injury to the diaphragm, and there were no false-positive interpretations of CT-Read1 for a specificity of 100% (95% CI, 99.2%-100%). The PPV for the test was thus 1.00 (95% CI, 0.65-1.00), whereas the NPV was 0.99 (95% CI, 0.98-0.99). Table 3 details the 9 patients with laparotomy or autopsy-proven DI and describes the CT signs, location, size, and outcome for each patient. Note again that the CT sign bnonspecific thickening of the diaphragmQ was not used to identify DI on CT-Read1. Forty-two patients in our series of 500 required emergent laparotomy for any reason (rate, 8.4%). The overall incidence of blunt DI in this series was 1.8%. Excluding those patients who died or received emergent laparotomy, 71.8% (308/429) were successfully contacted by phone for

Injury, CT, and laparotomy findings in the study patients proven to have DI

Age (y)

Sex

Mechanism

GCS

ISS

21 57 24 36 38 25 79 73 74

M F M F F M F F F

Fall MVC MVC MVC MVC MVC MVC PED MVC

14 3 15 14 10 7 3 15 15

15 43 34 24 57 34 26 34 42

CT-Read1 signs of DI DDD

ITHAC

CS

CT-Read2 DVS

X X X X X X

X

X

X X X X

Grade

Outcome

Left Left Left Left Left Left Left Left Right

II III III III III IV V IV III

Alive Alive Alive Alive Alive Alive Dead Alive Alive

TD X

X X X X X X

Side

X X X X

DDD indicates direct discontinuity of the diaphragm; ITHAC, intrathoracic herniation of abdominal contents; CS, collar sign; DVS, dependent viscera sign; TD, nonspecific thickening of the diaphragm; PED, pedestrian auto accident.

256

Fig. 1 Computed tomographic image demonstrating left hemidiaphragm injury and intrathoracic herniation of bowel. Intrathoracic herniation of bowel (thin arrow) noted by proximity to the heart opacified by intravenous contrast material (large arrow).

follow-up. None of these patients was thought to have missed DIs at 3-month telephone follow-up.

4. Limitations There are several limitations to our study. Most importantly, this is not an equivalence study between CT imaging with and CT imaging without oral contrast. In the population of the traumatically injured, the overall prevalence of DIs is low, which makes the demonstration of true equivalence using a randomized study exceptionally difficult. In addition, our study has no institutional retrospective control group with which to compare our study population. We have depended on the published literature for comparative sensitivities and specificities of CT imaging for DI. Against that standard, CT without oral contrast for DI appears to perform within the range described for CT with oral contrast. However, the absence of a true control group imposes limitations inherent to such efforts. Clearly, the low number (n = 9) of blunt DIs in our study patients imposes additional inference limitations as evidenced by the wide CIs for sensitivity and PPV particularly. The possibility of missed or eventual delayed presentations of DI also limits the conclusions of this study. Only 42 of the 500 study patients received laparotomy for any reason. Any of the remaining 438 patients, therefore, may be at risk for missed DIs. Even those contacted by phone 3 months after their injury may have DIs and remain asymptomatic. In addition, there were 121 patients who could not be contacted for any follow-up.

5. Discussion Delayed recognition and repair of blunt DIs may lead to diaphragmatic fibrosis and late herniation or strangulation of

T.L. Allen et al. bowel [11,12,15,22,23]. As more and more victims of blunt trauma receive nonoperative care, it is easy to believe that more DIs may be missed during initial trauma evaluation and hospitalization. The initial chest radiograph is diagnostic of DI in only 27% to 60% of left-sided injuries and in a mere 17% of rightsided injuries [11,17,24]. The sensitivities reported in the literature for CT imaging and diagnosis of traumatic DIs do not fare much better and range from 0% to 84% [23,25-29]. Chen and Wilson [25] reported in a retrospective study in 1991 on 62 patients with surgically proven DIs. Seventeen of those injuries were caused by blunt mechanisms. They reported that CT scans of the thorax and abdomen were done on 11 of the 62 total patients and were diagnostic for DI in none. This reported sensitivity of 0% should probably be considered an anomaly but certainly indicates the potential difficulty when applying this particular type of CT technology to the evaluation of DI. The sensitivity of CT imaging for DI in the other referenced studies ranges from 42% to 84% [23,26-29]. Each of these studies is retrospective, and each study (including that by Chen and Wilson) used oral contrast material. The seminal study by Worthy et al [29] in 1995 reports on the retrospective review of CT images for 11 consecutive cases of surgically proven DI. Two of these were imaged using helical technology with reformatted images, whereas the others were imaged with a single-detector scanner. The CT images were reported to be diagnostic in 9 of 11 patients for a sensitivity of 82%. They remark, however, that the reformatted images bobtained on 2 patients did not add any information.Q The most recent published studies are more consistent in reporting higher sensitivities for CT and DI. Killeen et al [23] in 1999 and Larici et al [27] in 2002 used singledetector helical scanners and reformatted the images of the diaphragm in the sagittal and coronal planes to obtain sensitivities of 78% and 84%, respectively. Killeen et al [23] reported on 23 patients with surgically proven DI from blunt mechanisms that received CT imaging before operative repair. Computed tomography was diagnostic in 14 of 18 left-sided DIs (sensitivity, 78%) and in 3 of 6 right DIs (sensitivity, 50%). The overall sensitivity for DI in their study was 74% [23]. The study by Larici et al [27] included patients with both blunt and penetrating mechanisms. Of 25 total patients with DIs, 11 were injured by a blunt mechanism. The sensitivity of CT for DI in those with blunt injury was 82% (9/11 diagnostic of DI). Computed tomography with oral and intravenous contrast was used in the early 1980s as an alternative to DPL or exploratory laparotomy in traumatically injured patients [2,4]. Compared with DPL, CT scanning is a noninvasive test that allows identification of the specific organ injured, permits visualization of the retroperitoneum, and, in some cases, can quantify the extent of injury. Over the years, as scanning protocols have developed and CT technology has

CT without OC for Diaphragm Injuries advanced in a quantum fashion, CT imaging has become the standard of care in circulatory-stable patients with BAT whose mechanism of injury or clinical examination is suspicious for intraabdominal injury. In particular, the introduction of spiral CT technology and multidetector arrays (with reformatted images) may allow increased CT accuracies in multiple-trauma patients and thus may eliminate the need for oral contrast material [14,27]. Oral contrast may be associated with adverse events. These include a time delay to CT imaging while awaiting contrast transit to the proximal small bowel and the risk of vomiting and aspiration with or without a nasogastric tube [7,30-34]. The actual risk of vomiting or aspiration associated with administration of oral contrast material has been reported to occur in 1% to 23% of patients [7,8,32]. Traumatically injured patients are at risk for vomiting and aspiration in any case, and the contribution of oral contrast material to this risk is not well quantified. All of the 6 patients in this study identified by CT-Read1 to have DIs were noted to have unequivocal intrathoracic herniation of abdominal contents on CT images (Fig. 1). Naturally, each of those 6 patients was noted to have direct diaphragm discontinuity, whereas 5 patients had the dependent viscera sign by CT images. Only 1 patient manifested a collar sign. Although the frequency of these radiographic signs is generally in line with previous reports, the rate of intrathoracic herniation of abdominal contents in this series of patients is slightly higher than might have been expected using historic literature controls [12]. Of 3 false-negative DI cases, 2 had nonspecific thickening of the diaphragm noted on CT-Read2. One of these was a grade 3 right-sided DI identified at laparotomy. The other two were left-sided DIs also identified at laparotomy. Laparotomy was initiated for suspicion of blunt bowel injury in each of these 3 false-negative cases, and bowel injury was confirmed at laparotomy. Without the concomitant presence of bowel injury in these patients, it is easy to believe that the DI may have gone undetected initially. None of these 3 false-negative cases of DI was identified conclusively on CT-Read2. We would caution, however, against the routine use of nonspecific thickening of the diaphragm as indicative of true DI thus prompting laparotomy. Although including this sign would undoubtedly improve the sensitivity of the test (CT without oral contrast), this improvement may come at the expense of the PPV leading to an increased rate of negative laparotomy. Table 3 details the demographics and injury patterns in the 9 patients in this series with DI. In retrospect, the study would have been strengthened as a randomized, nonblinded comparison between CT imaging with and CT imaging without oral contrast. The study period would have had to be lengthened considerably, but an increased study time would allow for the enrollment of more patients with potential DIs, thus narrowing the test performance CIs. We might also have extended the follow-up time to provide additional evidence that there were no

257 missed injuries as some traumatic DIs may present with diagnostic delays measured in years. The number of patients with identified DIs in this study, however, is not far different from that in other published studies [23,26,29] and certainly represents the largest cohort of patients imaged without oral contrast for DI. As far as we can tell, this is the only prospective analysis of such patients. In summary, our study of 500 trauma patients evaluated by CT scan without oral contrast for blunt DI demonstrates an accuracy that is similar to studies in the literature that used oral contrast material, and it does not appear that withholding oral contrast will compromise the sensitivity of CT for DI. It seems reasonable to hypothesize that oral contrast would only rarely benefit the CT diagnosis of DI in blunt trauma because none of the primary CT signs of DI clearly requires opacification of the stomach or bowel. In fact, the elimination of oral contrast from trauma imaging protocols may confer some benefit. It should be clear, however, that CT imaging continues to demonstrate an insufficient sensitivity to allow emergency physicians and trauma surgeons to rely on this technology alone to rule out the diagnosis of DI. Importantly, however, the PPV of CT without oral contrast appears to be very good for DI, and thus the presence of the aforementioned signs for DI on CT without oral contrast should indicate the need for operative repair. Protocols should continue to be developed by emergency physicians, trauma surgeons, and radiologists who use CT without oral contrast for the detection of DI and other thoracoabdominal injuries. As always, serial clinical examination remains an integral component of trauma care in association with CT evaluation. We therefore conclude that the routine addition of oral contrast may not be necessary to make the critical diagnosis of DI in BAT. Further and larger studies, perhaps with newer CT imaging techniques including multidetector scanners with reformatted sagittal and coronal images, will be necessary to definitively answer this difficult question.

Acknowledgments This project was generously supported by a grant from the Deseret Foundation. The Deseret Foundation is a local not-for-profit foundation that serves as an independent funding agency supporting the medical research and education needs of the LDS Hospital. We are grateful for its assistance. We also thank Sue Day, RN, for her help with data management and the staff of the Statistical Data Center of the LDS Hospital for their expertise.

References [1] Marx JA, Moore EE, Jorden RC, Eule Jr J. Limitations of computed tomography in the evaluation of acute abdominal trauma: a

258

[2] [3] [4] [5]

[6]

[7]

[8]

[9]

[10]

[11]

[12]

[13]

[14] [15] [16]

[17] [18]

T.L. Allen et al. prospective comparison with diagnostic peritoneal lavage. J Trauma 1985;25(10):933 - 7. Sherck JP, McCort JJ, Oakes DD. Computed tomography in thoracoabdominal trauma. J Trauma 1984;24(12):1015 - 21. Cox EF. Blunt abdominal trauma. A 5-year analysis of 870 patients requiring celiotomy. Ann Surg 1984;199(4):467 - 74. Federle MP. Abdominal trauma: the role and impact of computed tomography. Invest Radiol 1981;16(4):260 - 8. Heiberg E, Wolverson MK, Hurd RN, Jagannadharao B, Sundaram M. CT recognition of traumatic rupture of the diaphragm. AJR Am J Roentgenol 1980;135(2):369 - 72. Clancy TV, Ragozzino MW, Ramshaw D, Churchill MP, Covington DL, Maxwell JG. Oral contrast is not necessary in the evaluation of blunt abdominal trauma by computed tomography. Am J Surg 1993;166(6):680 - 4 [discussion 684-5]. Tsang BD, Panacek EA, Brant WE, Wisner DH. Effect of oral contrast administration for abdominal computed tomography in the evaluation of acute blunt trauma. Ann Emerg Med 1997;30(1):7 - 13. Stafford RE, McGonigal MD, Weigelt JA, Johnson TJ. Oral contrast solution and computed tomography for blunt abdominal trauma: a randomized study. Arch Surg 1999;134(6):622 - 6 [discussion 626-7]. Buck III GC, Dalton ML, Neely WA. Diagnostic laparotomy for abdominal trauma. A university hospital experience. Am Surg 1986;52(1):41 - 3. Kearney PA, Rouhana SW, Burney RE. Blunt rupture of the diaphragm: mechanism, diagnosis, and treatment. Ann Emerg Med 1989;18(12):1326 - 30. Boulanger BR, Milzman DP, Rosati C, Rodriguez A. A comparison of right and left blunt traumatic diaphragmatic rupture. J Trauma 1993; 35(2):255 - 60. Iochum S, Ludig T, Walter F, Sebbag H, Grosdidier G, Blum AG. Imaging of diaphragmatic injury: a diagnostic challenge? Radiographics 2002;22(Spec No):S103 - 16 [discussion S116-8]. Voeller GR, Reisser JR, Fabian TC, Kudsk K, Mangiante EC. Blunt diaphragm injuries. A five-year experience. Am Surg 1990;56(1): 28 - 31. Shanmuganathan K, Killeen K, Mirvis SE, White CS. Imaging of diaphragmatic injuries. J Thorac Imaging 2000;15(2):104 - 11. Patselas TN, Gallagher EG. The diagnostic dilemma of diaphragm injury. Am Surg 2002;68(7):633 - 9. Allen TL, Mueller MT, Bonk RT, Harker CP, Duffy OH, Stevens MH. Computed tomographic scanning without oral contrast solution for blunt bowel and mesenteric injuries in abdominal trauma. J Trauma 2004;56(2):314 - 22. Shanmuganathan K, Mirvis SE. Imaging diagnosis of nonaortic thoracic injury. Radiol Clin North Am 1999;37(3):533-51, vi. Bergin D, Ennis R, Keogh C, Fenlon HM, Murray JG. The bdependent visceraQ sign in CT diagnosis of blunt traumatic diaphragmatic rupture. AJR Am J Roentgenol 2001;177(5):1137 - 40.

[19] Zinck SE, Primack SL. Radiographic and CT findings in blunt chest trauma. J Thorac Imaging 2000;15(2):87 - 96. [20] Leung JC, Nance ML, Schwab CW, Miller Jr WT. Thickening of the diaphragm: a new computed tomography sign of diaphragm injury. J Thorac Imaging 1999;14(2):126 - 9. [21] Rosner B. Fundamentals of biostatistics. 4th ed. Belmont (CA)7 Duxbury Press; 1995. p. 177. [22] Nau T, Seitz H, Mousavi M, Vecsei V. The diagnostic dilemma of traumatic rupture of the diaphragm. Surg Endosc 2001;15(9):992 - 6. [23] Killeen KL, Mirvis SE, Shanmuganathan K. Helical CT of diaphragmatic rupture caused by blunt trauma. AJR Am J Roentgenol 1999;173(6):1611 - 6. [24] Gelman R, Mirvis SE, Gens D. Diaphragmatic rupture due to blunt trauma: sensitivity of plain chest radiographs. AJR Am J Roentgenol 1991;156(1):51 - 7. [25] Chen JC, Wilson SE. Diaphragmatic injuries: recognition and management in sixty-two patients. Am Surg 1991;57(12):810 - 5. [26] Murray JG, Caoili E, Gruden JF, Evans SJ, Halvorsen Jr RA, Mackersie RC. Acute rupture of the diaphragm due to blunt trauma: diagnostic sensitivity and specificity of CT. AJR Am J Roentgenol 1996;166(5):1035 - 9. [27] Larici AR, Gotway MB, Litt HI, Reddy GP, Webb WR, Gotway CA, et al. Helical CT with sagittal and coronal reconstructions: accuracy for detection of diaphragmatic injury. AJR Am J Roentgenol 2002;179 (2):451 - 7. [28] Shapiro MJ, Heiberg E, Durham RM, Luchtefeld W, Mazuski JE. The unreliability of CT scans and initial chest radiographs in evaluating blunt trauma induced diaphragmatic rupture. Clin Radiol 1996;51(1):27 - 30. [29] Worthy SA, Kang EY, Hartman TE, Kwong JS, Mayo JR, Muller NL. Diaphragmatic rupture: CT findings in 11 patients. Radiology 1995; 194(3):885 - 8. [30] Federle MP. Diagnosis of intestinal injuries by computed tomography and the use of oral contrast medium. Ann Emerg Med 1998; 31(6):769 - 71. [31] Federle MP, Yagan N, Peitzman AB, Krugh J. Abdominal trauma: use of oral contrast material for CT is safe. Radiology 1997;205(1):91 - 3. [32] Federle MP, Peitzman A, Krugh J. Use of oral contrast material in abdominal trauma CT scans: is it dangerous? J Trauma 1995;38(1): 51 - 3. [33] Lim-Dunham JE, Narra J, Benya EC, Donaldson JS. Aspiration after administration of oral contrast material in children undergoing abdominal CT for trauma. AJR Am J Roentgenol 1997;169(4): 1015 - 8. [34] Trulzsch DV, Penmetsa A, Karim A, Evans DA. Gastrografin-induced aspiration pneumonia: a lethal complication of computed tomography. South Med J 1992;85(12):1255 - 6.