| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Gastrointestinal Imaging |
1 From the Department of Radiology, Duke University Medical Center, Box 3808, Durham, NC 27710. Received July 14, 2004; revision requested September 21; revision received September 29; accepted October 26. Address correspondence to E.K.P. (e-mail: pauls003@mc.duke.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: This study was approved by the institutional review board, and informed consent was waived. One hundred consecutive patients (21 men, 79 women; mean age, 38 years) with suspected appendicitis underwent 16-section multi–detector row CT (section thickness, 0.625 mm; pitch, 1.75; table speed, 35 mm/sec [17.5 mm per rotation, two rotations]; and gantry speed, 0.5 second per rotation), with coronal reformations. Twenty-four patients had appendicitis; 76 did not. Protocol included 150 mL oral iopamidol administered at 3 mL/sec. Transverse scans were reconstructed with 5-mm-thick sections at 5-mm intervals and 0.625-mm-thick sections at 0.625-mm intervals. The second data set was reformatted coronally, with 3-mm-thick sections at 5-mm intervals. Three independent blinded readers interpreted transverse scans alone and then coronal scans; confidence in visualization of any portion of appendix, entire appendix, wall thickening, distention, inflammation, fluid, and appendicitis was scored with 1–5 scale. Sensitivity and specificity were determined for each reader and compared by means of signed rank test. Agreement between readers was determined with 
statistic. Differences in mean confidence ratings for each finding were determined with Wilcoxon signed rank test.
RESULTS: Mean sensitivity and specificity for all three readers together were 96% and 95% for transverse reformations alone and 95% and 94% for combined transverse and coronal reformations (not significant), respectively. Visualization rates for portion or all of appendix were higher for combined transverse and coronal reformations than for transverse reformations alone (higher mean confidence scores: 0.23 higher [P < .009] and 0.51 higher [P < .001], respectively). In patients without appendicitis, transverse and coronal reformations together enhanced confidence in exclusion of wall thickening, distention, and fluid (lower confidence scores: 0.21 lower [P < .001], 0.17 lower [P < .01], 1.00 lower [P < .001], respectively). Combined transverse and coronal reformations enhanced confidence in identification of appendix in mean of 57 patients. Combined transverse and coronal scans helped exclude appendicitis in mean of 38 patients and aided diagnosis of it in 15.
CONCLUSION: Sixteen-section multi–detector row CT transverse and coronal reformations are equally sensitive and specific for diagnosis of appendicitis. Coronal reformations improve confidence in visualization of appendix (whether diseased or normal) and in diagnosis or exclusion of appendicitis.
© RSNA, 2005
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
With 16-section multi–detector row CT, it is now possible to scan the entire abdomen and pelvis within a single and comfortable breath hold at a resolution less than 1 mm in the x-axis, y-axis, and z-axis (11,12). These data sets result in voxels that are both less than 1 mm and nearly isotropic, which suggests that reformations in any desired plane will be similar in spatial resolution to those in the transverse plane (13). Thus, the purpose of our study was to assess retrospectively the added value of coronal reformations from isotropic voxels obtained with 16-section multi–detector row CT of the abdomen and pelvis in patients suspected of having acute appendicitis.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
From June 1, 2003, to February 1, 2004, a total of 100 consecutive patients who were suspected of having acute appendicitis underwent 16-section multi–detector row CT to rule out acute appendicitis. The patients included 21 men and 79 women, and the mean age was 38 years (range, 19–80 years). In 100 patients, acute appendicitis was confirmed in 24, and exclusion was confirmed in 76.
The medical records, surgical reports, and pathologic reports were reviewed by one author (J.P.H.) to determine the diagnosis. Patients were considered to have acute appendicitis if they underwent appendectomy and the pathologic specimen confirmed acute appendicitis (n = 24). Patients were considered not to have acute appendicitis if an alternative diagnosis was established (n = 33) and treated or if a normal appendix was removed at appendectomy. The alternative diagnoses in these patients included pelvic inflammatory disease or tubo-ovarian abscess (n = 11), colitis (n = 8), cholecystitis (n = 2), ureteral stone (n = 2), enteritis (n = 2), ileus (n = 2), pyelonephritis (n = 2), renal infarct (n = 1), mesenteric adenitis (n = 1), cystitis (n = 1), and pancreatitis (n = 1). One patient with enteritis had a normal appendix that was removed at surgery. Patients were also considered not to have acute appendicitis if they did not undergo appendectomy and had no evidence of persistent pain, abscess, or unexplained fever during follow-up (n = 43). In the patients without acute appendicitis, the mean follow-up interval was 9 months (range, 4–11 months).
Scanning
Scanning was performed from the dome of the diaphragm through the pubic symphysis with a CT scanner (LightSpeed 16; GE Medical Systems, Milwaukee, Wis). Patients ingested 450 mL of a 2% barium sulfate suspension (Readi-Cat 2; E-Z-Em, Westbury, NY) 1–2 hours before scanning. Iopamidol (Isovue; Bracco Diagnostics, Princeton, NJ) was injected (Empower CT; E-Z-Em) at a dose of 150 mL (300 mg of iodine per milliliter) and a rate of 3 mL/sec. Scanning was performed during the portal venous phase as determined with bolus tracking and automated triggering technology. The protocol was as follows: 140 kVp; 350 mA; sections, 16; section thickness, 0.625 mm; pitch, 1.75; table speed, 35 mm/sec (17.5 mm per rotation with two rotations); and gantry speed, 0.5 second per rotation. The transverse section data were reconstructed twice: first with 5-mm-thick sections at 5-mm intervals in the transverse plane and then with 0.625-mm-thick sections at 0.625-mm intervals. The second set of reconstructed transverse scans were then reformatted in the coronal plane with 3-mm sections at 5-mm intervals.
Reconstruction was performed with a commercially available console system devoted to rapid reconstruction (Xtream; GE Medical Systems); the system consists of dual 2.66-GHz processors (Xenon; Intel, Santa Clara, Calif) with a CT scan generator capable of reconstructing six to 10 scans per second. The scan generator required approximately 2 minutes to reconstruct both transverse and coronal scans. The entire process was performed by the technologist at the operator’s console. The 5-mm transverse and 3-mm coronal scans were transferred to a picture archiving and communication system workstation (Centricity 1.0; GE Medical Systems) as a separate series of scans for interpretation.
Scan Evaluation
After patient identifiers were removed from the CT scans, they were loaded onto a workstation (Advantage Windows; GE Medical Systems) for review. Three readers with subspecialty training in abdominal imaging (P.A.H., T.A.J., E.K.P.) served as independent readers who were blinded to the diagnosis; they had 1, 3, and 14 years of experience dedicated to abdominal imaging, respectively. Readers first assigned confidence scores to the transverse scans alone and then immediately assigned scores to the coronal scans obtained in the same patient. Impressions from the transverse scans were therefore fresh in the readers’ minds. Scores of 4 or 5 were considered affirmative. For both the transverse and coronal data sets, readers rated scans for findings indicative of acute appendicitis, and these findings included identification of a portion of the appendix on at least one scan, identification of the entire length of the appendix, distention (diameter, 
8 mm), wall thickening, periappendiceal inflammatory change, and periappendiceal fluid.
A confidence score was obtained for each finding, with a scale of 1–5 (1, definitely absent; 2, probably absent; 3, cannot determine; 4, probably present; and 5, definitely present). In addition, on the basis of all the findings from each set of scans, readers judged whether acute appendicitis was present (with the same scale of 1–5 as was used for the confidence score), and thus, confidence scores were provided for acute appendicitis. The judgment regarding acute appendicitis was based on the individual reader’s assessment. After assigning scores to the combined transverse and coronal scans, readers judged whether the coronal scans added value to the transverse scans for identification of the appendix (whether diseased or normal), the exclusion of appendicitis, or the diagnosis of acute appendicitis. This judgment was subjective.
Statistical Analysis
The sensitivity and specificity values of each reader were determined for both the transverse scans alone and the transverse and coronal scans combined. P values for comparisons of the mean sensitivity and specificity for all three readers between transverse and combined transverse and coronal scans were computed with the signed rank test. Differences with P values of .01 or less were considered statistically significant.
Agreement between readers for the diagnosis of acute appendicitis on the basis of transverse scans alone and on the basis of combined transverse and coronal scans was determined with the weighted statistic. Differences in the reader mean weighted values between the transverse scans alone and the combined transverse and coronal scans were assessed with the jackknife method. The mean confidence ratings for each finding of acute appendicitis were classified according to reader, diagnosis, and scan set. Differences in ratings were determined with the Wilcoxon signed rank test. Statistical software (SAS, version 8.2; SAS Institute, Cary, NC) was used.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
|
|
|
range, 0.69–0.73) among all three readers (Table 2). For the combined transverse and coronal scans, the statistic (range, 0.81–0.84) was even higher, and this finding indicated a significantly higher level of agreement for the combined transverse and coronal scans than for the transverse scans alone (P < .001).
|
|
|
|
|
|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
We explored whether coronal reformations would serve as a useful adjunct to transverse scans in the diagnosis of acute appendicitis. To our knowledge, the role of coronal reformations obtained with multi–detector row CT in the evaluation of acute appendicitis has not been previously reported. Our results indicate that a scan obtained in the coronal plane serves as a useful adjunct to a scan obtained in the transverse plane in patients who are suspected of having acute appendicitis. In this study, the coronal scans were interpreted immediately after the transverse scans were interpreted.
The sensitivity and specificity for the combined transverse and coronal scans were similar to those for the transverse scans alone and ranged from 92% to 96% for sensitivity and from 93% to 95% for specificity. The sensitivity and specificity reported in this work corroborate those in recent reports that indicate that multi–detector row CT performed with both intravenous and oral contrast material is an accurate examination in patients who are suspected of having acute appendicitis (2–10,19).
The value of the coronal reformations is apparent in measures of agreement and diagnostic confidence among independent observers. Specifically, the agreement for the diagnosis of acute appendicitis among the three readers was consistently higher for the interpretations of combined transverse and coronal scans than it was for those of transverse scans alone.
In the patients with acute appendicitis, the confidence scores of the three readers for the diagnosis of acute appendicitis were higher for the combined transverse and coronal scans than they were for the transverse scans alone. Similarly, in the patients without acute appendicitis, confidence scores for the findings indicative of acute appendicitis were lower for combined transverse and coronal scans; these results indicated that the readers were more confident in the absence of findings on the coronal scans.
In patients with or without appendicitis, confidence in identification of a portion of or the entire length of the appendix was higher with the combined transverse and coronal scans than it was with the transverse scans alone. Indeed, one of the primary tasks in the evaluation of the patient with suspected acute appendicitis is the identification of the appendix. Once it is found, the next step is to search for evidence of acute appendicitis or to confirm a normal appendix. Our study findings indicate that the combination of transverse and coronal scans enhances confidence in identification of the appendix, whether diseased or normal. Enhanced confidence in appendiceal identification has implications for the reporting of the findings of studies. Specifically, a report with findings that indicate that the appendix has been identified and is normal is more definitive than one that indicates that, although there was no evidence of acute appendicitis, the appendix itself was not clearly identified. The subjective impression of all three readers was that the coronal scans added value to the transverse scans for the identification of the appendix (whether diseased or normal), the diagnosis of acute appendicitis, and the exclusion of acute appendicitis.
Although multiplanar reformations of scans of the abdomen have been described for diseases in the liver, the urinary tract, the small bowel, and the pancreas, little has been written in regard to the role of multiplanar reformations in the setting of acute appendicitis (14–17,19). We restricted our reformations to the coronal plane because the processing was seamless, was semiautomatic, and required only a few strokes of the keyboard at the operator’s console by the CT technologist. In addition, the coronal view, which is analogous to a frontal view of an abdominal radiograph, may be more intuitive for surgeons and radiologists. Our protocol required no imaging manipulation by radiologists, either at the operator’s console or at a dedicated three-dimensional workstation. In our clinical practice, the reformations are sent directly to the picture archiving and communication system and appear as a separate series for interpretation. Although our study demonstrates the value of coronal reformations as an adjunct to transverse scans, reformations in the sagittal plane may be similarly helpful.
Even with coronal reformations, in some cases it is difficult to identify the entire length of the appendix. In this situation, a curved reformation along the length of the entire appendix may prove useful. Such a reformation, which is analogous to a curved reformation from a scan obtained along the pancreatic duct or ureter, would "straighten out" the tortuous appendix and allow visualization of it in entirety (17,18). Our rationale for not performing such reformations routinely is that radiologists would be required to spend additional time at the workstation.
Prior to the development of 16-section multi–detector row CT, it was difficult to scan the entire abdomen and pelvis at a section thickness of less than 1 mm during a single comfortable breath hold. While a volume acquisition of the entire abdomen and pelvis was possible with single–detector row CT or four- or eight-section multi–detector row CT, section thicknesses typically ranged from 1.25 to 5.00 mm. The multiplanar reformations created from such data sets provided an overview of gross anatomy and large diseased areas, but they were degraded by stair-step artifact, noise, motion artifact, and suboptimal spatial resolution caused by decreased resolution in the z-axis. Small structures and subtle findings, such as the appendix and associated signs of inflammation, were poorly visualized or obscured. With the 16-section multi–detector row CT protocol described in this study, coronal scan quality rivals that of transverse scans. Even the subtle signs of appendiceal inflammation could be identified with confidence.
We used a multi–detector row CT scanner with 16 sections, 0.625-mm section thickness, and a table speed of 35 mm/sec, which permits a single acquisition during a comfortable breath hold. Two sets of reconstructions were obtained. The first set was reconstructed in the transverse plane with 5-mm sections at 5-mm intervals. For a typical patient, this series resulted in approximately 70–100 scans. For the second set of reconstructions, the scans with 0.625-mm thickness were reconstructed at 0.625-mm intervals. This series resulted in approximately 560–800 scans (13), but the scans in this series were not used for diagnostic purposes because of the large number of scans and because noise caused problems that were due to the section thickness of only 0.625 mm. Rather, the second set of reconstructed transverse scans were reformatted in the coronal plane, with 3-mm sections at 5-mm intervals, which resulted in 50–75 scans per patient. In our practice, the 5-mm transverse and coronal scans are sent to the picture archiving and communication system as a series for interpretation. The 0.625-mm transverse scans are archived but are not interpreted or sent to the picture archiving and communication system.
The CT dose index for the 16-section multi–detector row CT protocol detailed previously in this study is equivalent to that of an analogous four-section multi–detector row CT (LightSpeed; GE Medical Systems) protocol with the following technical parameters: 140 kVp; 220 mA; section thickness, 2.5-mm; pitch, 1.75; table speed, 15 mm per rotation; gantry speed, 0.8 second per rotation; and reconstruction thickness, 5 mm (20).
This study had some limitations. Our intention was to demonstrate the value of coronal reformations as an adjunct to transverse scans alone. Accordingly, the coronal scans were interpreted after interpretation of the transverse scans, and impressions from the transverse scans were fresh in the minds of the interpreters. Such a design may have biased the results in favor of the subsequent interpretation, which in this case was based on the findings on the coronal scans. It was not our intent, however, to assess the value of the coronal reformations as stand-alone images, as we believe that in most clinical settings radiologists would be unlikely to abandon transverse scans in favor of coronal scans alone.
Another potential limitation was that we used a transverse diameter of 8 mm or more to identify appendiceal distention. While an 8-mm diameter is an optimal threshold measurement for distinguishing a normal from an abnormal appendix, a normal appendix filled with gas or contrast material may have a diameter as large as 10 mm (21,22). In this study, specific luminal contents were not considered. Nevertheless, readers identified diseased and normal appendices with a high level of sensitivity and specificity.
In conclusion, we found that combined transverse and coronal reformations from isotropic voxels obtained with 16-section multi–detector row CT rather than transverse scans alone are useful in the evaluation of patients who are suspected of having acute appendicitis. The addition of the coronal reformations adds confidence in the identification of the appendix and in the diagnosis or exclusion of acute appendicitis.
|
FOOTNOTES |
|---|
Author contributions: Guarantor of integrity of entire study, E.K.P.; study concepts/study design or data acquisition or data analysis/interpretation and manuscript drafting or manuscript revision for important intellectual content and manuscript final version approval, all authors; literature research, E.K.P.; clinical studies, E.K.P., J.P.H., T.A.J., P.A.H., R.C.N.; manuscript editing, all authors
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  H. Seo, K. H. Lee, H. J. Kim, K. Kim, S.-B. Kang, S. Y. Kim, and Y. H. Kim Diagnosis of Acute Appendicitis With Sliding Slab Ray-Sum Interpretation of Low-Dose Unenhanced CT and Standard-Dose IV Contrast-Enhanced CT Scans Am. J. Roentgenol., July 1, 2009; 193(1): 96 - 105. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S.-M. Joo, K. H. Lee, Y. H. Kim, S. Y. Kim, K. Kim, K. J. Kim, and B. Kim Detection of the Normal Appendix with Low-Dose Unenhanced CT: Use of the Sliding Slab Averaging Technique Radiology, June 1, 2009; 251(3): 780 - 787. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. T. Johnson, K. M. Horton, S. Kawamoto, J. Eng, M. J. Bean, S. J. Shan, and E. K. Fishman MDCT for Suspected Appendicitis: Effect of Reconstruction Section Thickness on Diagnostic Accuracy, Rate of Appendiceal Visualization, and Reader Confidence Using Axial Images Am. J. Roentgenol., April 1, 2009; 192(4): 893 - 901. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. J. Kim, J.-E. Kim, H. S. Kim, and H. Y. Hwang MDCT with Coronal Reconstruction: Clinical Benefit in Evaluation of Suspected Acute Appendicitis in Pediatric Patients Am. J. Roentgenol., January 1, 2009; 192(1): 150 - 152. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Marin, C. Catalano, G. De Filippis, M. Di Martino, A. Guerrisi, M. Rossi, and R. Passariello Detection of Hepatocellular Carcinoma in Patients with Cirrhosis: Added Value of Coronal Reformations from Isotropic Voxels with 64-MDCT Am. J. Roentgenol., January 1, 2009; 192(1): 180 - 187. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Keyzer, S. Pargov, D. Tack, V. Creteur, P. Bohy, V. De Maertelaer, and P. A. Gevenois Normal Appendix in Adults: Reproducibility of Detection with Unenhanced and Contrast-Enhanced MDCT Am. J. Roentgenol., August 1, 2008; 191(2): 507 - 514. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. T. Rhea Invited Commentary RadioGraphics, March 1, 2008; 28(2): 405 - 406. [Full Text] [PDF]
|
|
|
|
|
|
F. J. Parrish Volume CT: State-of-the-Art Reporting Am. J. Roentgenol., September 1, 2007; 189(3): 528 - 534. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Memarsadeghi, C. Schaefer-Prokop, M. Prokop, T. H. Helbich, C. C. Seitz, I. M. Noebauer-Huhmann, and G. Heinz-Peer Unenhanced MDCT in Patients with Suspected Urinary Stone Disease: Do Coronal Reformations Improve Diagnostic Performance? Am. J. Roentgenol., August 1, 2007; 189(2): W60 - W64. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. J. Hebert and P. J. Pickhardt MDCT Diagnosis of an Appendiceal Mucocele with Acute Torsion Am. J. Roentgenol., July 1, 2007; 189(1): W4 - W6. [Full Text] [PDF]
|
|
|
|
|
|
T. A. Jaffe, L. C. Martin, C. M. Miller, K. M. Franklin, E. M. Merkle, W. M. Thompson, R. C. Nelson, D. M. DeLong, and E. K. Paulson Abdominal Pain: Coronal Reformations from Isotropic Voxels with 16-Section CT--Reader Lesion Detection and Interpretation Time Radiology, January 1, 2007; 242(1): 175 - 181. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. C. Dalrymple, S. R. Prasad, F. M. El-Merhi, and K. N. Chintapalli Price of Isotropy in Multidetector CT RadioGraphics, January 1, 2007; 27(1): 49 - 62. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. C. Chan, B. N. Joe, F. V. Coakley, E. L. Prien Jr, R. G. Gould, S. Prevrhal, W. C. Barber, K. S. Kirkwood, A. Qayyum, and B. M. Yeh Gallstone Detection at CT in Vitro: Effect of Peak Voltage Setting. Radiology, November 1, 2006; 241(2): 546 - 553. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Oto, R. D. Ernst, W. J. Mileski, T. K. Nishino, O. Le, G. C. Wolfe, and G. Chaljub Localization of Appendix with MDCT and Influence of Findings on Choice of Appendectomy Incision Am. J. Roentgenol., October 1, 2006; 187(4): 987 - 990. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. A. Jaffe, R. C. Nelson, G. A. Johnson, E. R. Lee, T. T. Yoshizumi, C. R. Lowry, A. B. Bullard, D. M. DeLong, and E. K. Paulson Optimization of Multiplanar Reformations from Isotropic Data Sets Acquired with 16-Detector Row Helical CT Scanner Radiology, January 1, 2006; 238(1): 292 - 299. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. A. Jaffe, L. C. Martin, J. Thomas, A. R. Adamson, D. M. DeLong, and E. K. Paulson Small-Bowel Obstruction: Coronal Reformations from Isotropic Voxels at 16-Section Multi-Detector Row CT Radiology, December 1, 2005; 238(1): 135 - 142. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
