| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Editorials |
1 From the Department of Radiology, Children’s Hospital Boston and Harvard Medical School, 300 Longwood Ave, Boston, MA 02115. Received December 16, 2003; accepted December 18. Address correspondence to the author (e-mail: george.taylor@childrens.harvard.edu).
Index terms: Appendicitis, 751.291 • Appendix, CT, 751.12115 • Computed tomography (CT), in infants and children, 751.12117 • Editorials
In this issue of Radiology, Kaiser et al (1) compare the accuracy of two helical computed tomographic (CT) techniques in the diagnosis of appendicitis in the pediatric age group. They report improved sensitivity for helical CT of the entire abdomen with intravenous contrast material enhancement compared with nonenhanced limited-area CT of the pelvis. This is important work, given the challenges in diagnosis of appendicitis in children and the higher frequency of delayed diagnoses and complications in this age group. The authors should be commended for their careful study design and thoughtful analysis.
I would like to first discuss the few important points in which Dr Kaiser’s group and our group differ with respect to imaging technique and use the remainder of the space allotted to discuss the current issues in imaging of pediatric appendicitis.
The first point in which we differ is the practice of scanning the entire abdomen. With this technique, Dr Kaiser and colleagues were able to identify 12 alternative diagnoses in which the findings were located primarily above the pelvis. However, only eight of these findings (three in patients with hydronephrosis, two in patients with pyelonephritis, and one each in patients with pancreatitis, pneumonia, or urinary calculi) are important and might have been missed with limited-area CT alone. The eight CT scans with these findings represent approximately 2.5% of the 317 scans obtained in this study. In other words, more than 97% of the scans could have been limited to the pelvis without clinical consequence. At my institution, we limit CT scanning to the pelvis in the great majority of cases. If an abnormality is found on the uppermost image, then the study is expanded to include the upper abdomen (2). We have needed to use this option in fewer than 10% of studies.
The second point in which we differ is in the use of contrast material administered rectally. Kaiser and colleagues suggest that the typical location of the cecum and its contents makes rectally administered contrast material unnecessary. We find that opacification of the appendix is helpful in a child with little intraperitoneal fat. By filling the appendix with contrast material, we can more definitively identify a normal appendix. However, I must concede that partial opacification of the appendix occurs frequently and does not exclude the presence of distal or "tip" appendicitis. Gas contained in the appendiceal lumen can be equally helpful in identification of this structure.
Next, I would like to review the current state of imaging in pediatric appendicitis. Each year in the United States, approximately 60,000–80,000 children are treated for appendicitis (3). Although this is a considerable population, the number of children with suspected appendicitis referred for imaging evaluation is likely to be two to three times higher. Younger children often present with nonspecific signs and symptoms and are unable to verbalize their symptoms or localize pain. As a result, delayed or incorrect diagnoses and complications such as appendiceal perforation are common (3,4). Because of these challenges, one might advocate use of a lower threshold for surgical intervention in children. However, removal of a normal appendix is not necessarily a benign procedure. In a study of 39,901 adult and pediatric patients, Flum and Koepsell (5) report that the nonincidental removal of a normal appendix is associated with a significantly longer hospital stay (5.8 vs 3.6 days), higher hospital costs ($18,780 vs $10,584), higher patient fatality rate (1.5% vs 0.2%), and higher rate of infectious complications (2.6% vs 1.8%) compared with surgical results in patients with appendicitis.
It is clear that imaging plays a key role in the modern evaluation of suspected appendicitis in children. What is not so clear is which modality should be used under what circumstances. The ideal imaging test for appendicitis would be readily available, fast, inexpensive, reproducible, and safe, and it would accurately distinguish children with the disease from those with a normal appendix. Ultrasonography (US) has been a mainstay in the evaluation of children with suspected appendicitis for the past decade. In that time, 14 studies have been published about its use in pediatric appendicitis, with sensitivity, specificity, and accuracy reported. Nearly 10,000 patients were included in these studies. While the ranges of reported specificity (88%–99%) and accuracy (82%–99%) for US have been acceptable, sensitivity (50%–100%) has varied considerably (6–19). While US is generally available and is relatively inexpensive and safe, its biggest drawback is that negative findings at US do not exclude appendicitis unless a normal appendix is visualized with a high degree of confidence. Visualization rates vary widely in the published literature, from a high of 98% (14) to a low of 22% (9).
Several techniques have been proposed to improve appendiceal visualization with US and have yielded modest to impressive results. These techniques include the use of posterior compression (appendiceal visualization improved from 85% to 95%) (20), scanning through a full urinary bladder followed by posterolateral scanning to identify a retrocecal appendix (98% of appendixes visualized) (14), and scanning after a saline enema (appendiceal visualization improved from 50% to 75%) (13). In addition, radiologists’ confidence in their interpretations appears to be influenced by the choice of imaging technique and by the results of the examination. Findings in a recent study from our institution showed that radiologists, regardless of level of training, were more confident about their interpretations at CT than about those at US (21).
After initial reports by Rao et al in 1997 (22), CT quickly became the preferred technique for the imaging of suspected appendicitis in adults. Pediatric applications soon followed. Since 1999, nine studies performed to evaluate the role of CT in the diagnosis of pediatric appendicitis included a total of 1,425 patients (6,9,13,15,17,19,23–25). The range of reported test performance characteristics for CT are quite good (sensitivity, 84%–100%; specificity, 89%–100%; and accuracy, 93%–99%) despite wide variability in techniques used. As with US, CT is also generally available, reproducible, accurate, and safe. However, ionizing radiation and the use of intravenous or gastrointestinal contrast material make it a relatively more invasive test.
Which is a better imaging test for suspected pediatric appendicitis—US or CT? In my opinion, this question has a two-part answer. The first part of the answer is that an effective imaging protocol should include both US and CT. If findings at US are inconclusive or fail to show the appendix, then intravenous contrast material–enhanced CT should be performed. At our institution, this sequence of imaging had a combined sensitivity of 94%, specificity of 94%, positive predictive value of 90%, negative predictive value of 97%, and accuracy of 94% (9). We were able to forgo the radiation dose and additional stresses of a CT examination in 22% of patients after only one US examination. After we implemented this imaging protocol into our practice, the perforation rate decreased from 35.4% to 15.5% and the negative appendectomy rate decreased from 14.7% to 4.1% (26). This protocol also resulted in a total cost savings of $565 per patient (27).
The second part of the answer about the better imaging test is, "It depends." Since the frequency of appendiceal visualization appears to be dependent on the child’s body habitus, perhaps some consideration should be given to not performing an initial US examination in obese children (28,29). As in politics, all imaging is local. By that, I mean to say that the choice of imaging must depend on the resources available in each institution rather than on the diagnostic accuracy of the test. Generalization of imaging protocols that work in one clinical setting may not be efficacious in other settings with a different array of available resources. For example, at our hospital, we have pediatric radiology fellows and faculty available around the clock. In this environment, US followed by CT with intravenously and rectally administered contrast material is a reasonable approach. However, what should the approach be if the facility staff members have little experience with pediatric imaging, if there is limited after-hours US coverage, or if no CT scanner is available? Is a technically limited but readily available US examination performed by a resident radiologist preferable to focused, radiation bearing, and relatively more invasive intravenous contrast-enhanced CT performed by experienced personnel? Although the definite answer is beyond my grasp, one solution is for the staff of each institution to review their local experience with imaging and clinical outcomes to determine if there is room for improvement.
The authors of two studies (30,31) suggest that the use of a clinical scoring scheme can be helpful in the stratification of children into high-, intermediate-, and low-risk categories for the presence of appendicitis. I believe that these scoring schemes are a fruitful area of clinical research that needs to be pursued with the goal of more rational utilization of imaging resources and radiation dose reduction in the pediatric population.
FOOTNOTES
See also the article by Kaiser et al in this issue.
REFERENCES
Related Article
This article has been cited by other articles:
|
|
 |
|
  M. Baldisserotto, S. G. Valduga, and C. F. J. S. da Cunha MR Imaging Evaluation of the Normal Appendix in Children and Adolescents Radiology, October 1, 2008; 249(1): 278 - 284. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
