|
|
||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Pediatric Imaging |
1 From the Departments of Pediatric Radiology (S.K., H.K.J.) and Pediatric Surgery (B.F.), Astrid Lindgren Childrens Hospital, Karolinska Institute, SE-171 76 Stockholm, Sweden. Received June 20, 2001; revision requested July 9; revision received September 24; accepted November 12. Supported by grants from the Crown Princess Lovisas Association for Childrens Health Care and the Axel Tielman Foundation. Address correspondence to S.K. (e-mail: sylvie.kaiser@ks.se).
| ABSTRACT |
|---|
|
|
|---|
MATERIALS AND METHODS: Six hundred children with suspected appendicitis were included in a prospective randomized study. After clinical examination, the patients were randomly assigned to undergo US only (283 patients) or US with abdominal CT (317 patients). Radiologic findings were correlated with surgical, histopathologic, and clinical follow-up findings.
RESULTS: Two hundred forty-four (40.7%) of all patients had appendicitis. In the US only group, US had a sensitivity of 86%, specificity of 95%, positive predictive value of 91%, negative predictive value of 92%, and diagnostic accuracy of 92%. The combination of US and CT performed in the other group yielded a sensitivity of 99%, specificity of 89%, positive predictive value of 87%, negative predictive value of 99%, and diagnostic accuracy of 93%. Analysis of data on US performed in all 600 patients and on CT performed in 317 patients revealed, respectively, sensitivities of 80% and 97%, specificities of 94% and 93%, positive predictive values of 91% and 92%, negative predictive values of 88% and 98%, and diagnostic accuracies of 89% and 95%. The overall negative appendectomy rate was 3.7%; and the perforation rate, 21%.
CONCLUSION: US is valuable in the diagnosis of appendicitis in children. In inconclusive cases, performing additional abdominal CT can improve diagnostic accuracy and thereby decrease the negative appendectomy rate without an increase in the perforation rate.
© RSNA, 2002
Index terms: Appendicitis, 751.291 Appendix, CT, 751.12111, 751.12112, 751.12115 Appendix, US, 751.12989 Children, gastrointestinal tract, 751.291
| INTRODUCTION |
|---|
|
|
|---|
The graded-compression ultrasonographic (US) technique described by Puylaert (6) has proved to be useful in the evaluation of suspected acute appendicitis and is now performed routinely in our diagnostic center. Helical computed tomography (CT) with use of a variety of techniques has been shown to be highly sensitive and specific for enabling the diagnosis of acute appendicitis (1,2,4,710), but few studies in which the accuracies of US and CT or the influence of these modalities on the negative appendectomy rate are compared are prospective and focused exclusively on children.
Astrid Lindgren Childrens Hospital, Stockholm, Sweden, is a university hospital that provides care in the majority of pediatric surgical cases in the greater Stockholm area (total population about 1.8 million people, corresponding to a referral area of approximately 350,000 children). About 300 children with appendicitis undergo surgery at this hospital annually. The aim of this study was to evaluate the accuracy of US and of abdominal CT performed in addition to US in the diagnosis of childhood appendicitis.
| MATERIALS AND METHODS |
|---|
|
|
|---|
|
Helical CT scanning (CT HiSpeed Advantage; GE Medical Systems, Milwaukee, Wis) was performed with 5-mm collimation, 10-mm/sec table speed (pitch, 2.0), and 5-mm reconstruction in patients younger than 6 years and with 7-mm collimation, 14-mm/sec table speed (pitch, 2.0), and 7-mm reconstruction in patients older than 6 years. No oral or rectal contrast medium was administered.
The lower part of the abdomen initially was scanned without contrast medium administration. Thereafter, the entire abdomen was scanned after the intravenous administration of nonionic contrast medium: iohexol in 29 patients (Omnipaque, 300 mg of iodine per milliliter; Nycomed Amersham, Oslo, Norway) and iodixanol (Visipaque, 270 mg of iodine per milliliter; Nycomed Amersham) in 288 patients. The injected dose was 2 mL per kilogram of body weight, with an upper limit of 100 mL. Patients with a history of asthma or possible previous reactions to contrast medium were excluded. There were no severe adverse drug reactions to the intravenous administration of contrast medium. One patient had a minor reaction: a mild skin rash limited to the chest that resolved spontaneously within 10 minutes.
The CT study was always performed after the US study and was interpreted by one of 12 pediatric radiologists (including S.K. and H.K.J.) or by one of nine senior residents. The interpreter had access to the results of the US study. The CT diagnosis of appendicitis was based on the visualization of an appendix larger than 6 mm in maximal diameter, with contrast enhancement in the thickened appendiceal wall and/or pericecal inflammatory changes, or on the visualization of an abscess, with or without an appendicolith (Figs 2, 3).
|
|
Both the US scan interpreter and the CT scan interpreter stated whether appendicitis was present and estimated their level of confidence in this finding on a scale from 0% to 100%. In the following analysis, only "yes" or "no" answers were considered, and the data are being subjected to further studies.
Final Diagnosis and Follow-up
The final clinical outcomes were determined at surgery and histopathologic analysis in the patients who underwent laparotomy (n = 252). The nonsurgically treated patients (n = 348) were followed up with a questionnaire 6 months after their emergency department admission. The questions about their state of health and possible treatment at other facilities, if any, after discharge from our hospital were designed to track any false-negative diagnoses.
Statistical Analysis Methods
Measures of imaging examination validitynamely, sensitivity, specificity, diagnostic accuracy, and positive and negative predictive valuesin the diagnosis of appendicitis were calculated in each randomly selected group. The same measures were also calculated for the diagnosis of appendicitis with US only in all 600 patients and for the diagnosis of appendicitis with CT only in 317 patients. The standard
2 test, which is equivalent to a two-sided two-sample binomial test, was performed to compare the calculated values of US and CT in each group. A P value of .05 or lower was considered to indicate a significant difference.
| RESULTS |
|---|
|
|
|---|
|
|
|
|
The results of US and CT in the 317 patients who underwent both of these examinations were concordant in 267 patients (84%) and discordant in 50 (16%). Of the patients with discordant results, 39 had correct CT findings and 11 had correct US findings (Table 4).
|
|
|
|
The follow-up questionnaire was completed by 327 (94%) of the 348 patients who were treated nonsurgically. One patient, who had negative US and CT findings, initially improved after being discharged but did not recover completely and was treated with drainage for an appendiceal abscess 2 weeks later. CT performed immediately before this treatment demonstrated the appendiceal abscess. The patient underwent appendectomy for recurrent gangrenous appendicitis a year later. This patients case was considered to be false-negative when statistical calculations were performed.
Another patient, who had negative US and CT findings and a diagnosis of nonspecific abdominal pain, recovered completely but was readmitted to the hospital 3 months later because of recurrent abdominal pain. Both US and CT performed at the second admission depicted appendicitis, and appendectomy of a phlegmonous appendix was performed. Because of the complete recovery between the two admissions, this patients case was considered to be true-negative when statistical calculations were performed.
All remaining patients recovered spontaneously or were treated for diseases other than appendicitis. Review of the medical records, including those of the patients who did not answer the questionnaire, did not reveal any other cases of appendicitis.
At admission, the surgeon estimated the likelihood of having appendicitis to be high (>75%) in 173 patients. However, only 119 of these patients had a final diagnosis of appendicitis. In 78 of these 173 patients, the clinical signs were considered to be convincing for a diagnosis of appendicitis, and the surgeon stated that laparotomy would be performed if imaging studies could not be performed. Only 58 of these 78 patients had appendicitis.
On the other hand, the likelihood of having appendicitis was estimated to be low (<50%) in 130 patients. In 25 of these patients, a correct diagnosis of appendicitis was made by means of US and/or CT.
| DISCUSSION |
|---|
|
|
|---|
US generally is the primary imaging modality performed in children suspected of having appendicitis because it is relatively quick to perform and does not involve the use of ionizing radiation. US is highly operator dependent, however, and there may be difficulty in identifying appendicitis when there is pain, obesity, overlying gas, and perforation.
The overall sensitivity of US in our study, 80%, is in accordance with sensitivities reported in previous studies (10), although higher (11,12) and considerably lower (1) values have been reported. Higher sensitivity would have been expected if the imaging examinations had been performed by a limited number of experienced specialists, as they have been in other studies (11,13). The fact that senior residents are included among the staff members who perform US at our institution may be a reason for the relatively low sensitivity, but this is the clinical reality because both our pediatric surgery and our radiology department offer 24-hour service.
The results of several studies (2,13,14) have demonstrated that a normal appendix can be visualized in a large number of patients who have symptoms that are suggestive of appendicitis. This has not been a common finding at routine imaging in our institution, and although a normal appendix has been seen occasionally, the possibility of mistaking a bowel loop for a normal appendix, and thus giving the surgeon the false impression that appendicitis has been ruled out, has led to some underreporting of this finding. The smaller volume of intraabdominal fat in children compared with that in adults may contribute to the relatively low rate of normal appendix detection at CT.
A variety of techniques have been used to perform appendiceal CT. Excellent results have been reported with CT performed with rectal contrast agents only, especially in adults (4,8). However, it has also been recommended that in pediatric patients, appendiceal CT be performed with intravenous contrast agents (7,15,16). Nonionic intravenous contrast agents are well tolerated in children, and the frequency of adverse reactions to contrast agents is lower than that in adults (17,18). Our technique, which includes scanning the entire abdomen by means of intravenous contrast agent administration, enables the acquisition of valuable information on the spatial extension of perforated appendicitis or an appendiceal abscess and for differential diagnoses. We have found this second scanning to be especially valuable, and limited nonenhanced scanning seems to be of less value. Our study data are being evaluated further for comparison of the techniques.
Appendicoliths appeared to be more frequent in our series than in previous studies (7,19). In addition, our study results support previous reports that appendicoliths are more frequently associated with perforated appendicitis and hence may represent a possible risk for perforation (7,19).
The sensitivity of 97% achieved with our CT technique in this study is in accordance with the reported sensitivity of the CT technique performed with rectal contrast agents only (4,8).
A possible limitation of our study is that the results of the previous US examinations were known when CT was performed. However, this is the clinical practice at our hospital. As mentioned earlier, our study data are being analyzed further, with a review of all CT studies without access to the US results. The specificities of both US and CT were in accordance with those in previously performed studies (4,9,10).
The patients who had US and/or CT findings that were suggestive of appendicitis but mild and spontaneously resolving clinical findings received a diagnosis of no appendicitis; hence, their cases were regarded as false-positive in this series. Some such cases may represent spontaneously resolving appendicitis, as has been described before (12,20). Lymphoid hyperplasia has been reported as a possible cause of a sonographically abnormal appendix (11). This abnormality was present in three of the patients in whom appendectomy was negative in our study and may have been present in some of the patients with false-positive findings who were not treated with surgery.
A negative appendectomy rate of about 20% generally has been considered acceptable (5,7). Avoiding unnecessary laparotomy, however, is desirable because the complication rate is not negligible and is even higher with negative laparotomy (2123). The negative appendectomy rate of 3.7% in our series is very low compared with that in most studies, but rates of 4% (9) and 6% (1) have been reported. If laparotomy had been performed in all 78 of the "clinically convincing" cases described herein earlier, it would have yielded a negative appendectomy rate of 11%. If all 173 of the patients with highly suspicious findings had undergone laparotomy, this rate would have risen to 21%. These values are similar to previously reported negative appendectomy rates in patients who did not undergo preoperative imaging (1,9,24,25).
The generally accepted, relatively high negative appendectomy rate has often been considered to be preferable to the complications of perforation. A range of appendiceal perforation rates in the pediatric population, from 23% to 73%, has been reported (2). The perforation rate of 21% in our series does not support the hypothesis (5) that preoperative imaging might contribute to an increase in perforation rate. It might be that the early diagnosis of appendicitis achieved by means of imaging in the equivocal cases could have contributed to the relatively low perforation rate, as has been previously suggested (4).
In conclusion, US is valuable in the diagnosis of appendicitis in children and should remain the method of first choice because it does not involve the use of ionizing radiation. Diagnostic accuracy can be increased significantly by performing CT, which we believe should be added to the imaging protocol for patients who have negative US results but convincing clinical presentations, in inconclusive cases, and/or when the radiologist lacks experience with US. When the findings of both US and CT are negative, close observation of the patient is recommended. As false-negative results may occur, the final decision is still the responsibility of the surgeon; if the clinical findings are convincing for a diagnosis of appendicitis or peritonitis, laparoscopy or laparotomy should be considered. With preoperative use of US and additional abdominal CT, a low negative appendectomy rate can be achieved without an increase in the perforation rate.
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
| REFERENCES |
|---|
|
|
|---|
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] |
||||
![]() |
M. Tsuboi, K. Takase, I. Kaneda, T. Ishibashi, T. Yamada, M. Kitami, S. Higano, and S. Takahashi Perforated and Nonperforated Appendicitis: Defect in Enhancing Appendiceal Wall Depiction with Multi Detector Row CT Radiology, January 1, 2008; 246(1): 142 - 147. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Yabunaka, T. Katsuda, S. Sanada, and T. Fukutomi Sonographic Appearance of the Normal Appendix in Adults J. Ultrasound Med., January 1, 2007; 26(1): 37 - 43. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. S. Doria, R. Moineddin, C. J. Kellenberger, M. Epelman, J. Beyene, S. Schuh, P. S. Babyn, and P. T. Dick US or CT for Diagnosis of Appendicitis in Children and Adults? A Meta-Analysis Radiology, October 1, 2006; 241(1): 83 - 94. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. Pinto Leite, J. M. Pereira, R. Cunha, P. Pinto, and C. Sirlin CT Evaluation of Appendicitis and Its Complications: Imaging Techniques and Key Diagnostic Findings Am. J. Roentgenol., August 1, 2005; 185(2): 406 - 417. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Keyzer, M. Zalcman, V. De Maertelaer, E. Coppens, M.-A. Bali, P. A. Gevenois, and D. Van Gansbeke Comparison of US and Unenhanced Multi-Detector Row CT in Patients Suspected of having Acute Appendicitis Radiology, August 1, 2005; 236(2): 527 - 534. [Abstract] [Full Text] [PDF] |
||||
![]() |
I. Dalal, E. Somekh, A. Bilker-Reich, M. Boaz, A. Gorenstein, and F. Serour Serum and Peritoneal Inflammatory Mediators in Children With Suspected Acute Appendicitis Arch Surg, February 1, 2005; 140(2): 169 - 173. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. O. Stolberg, G. Norman, and I. Trop Randomized Controlled Trials Am. J. Roentgenol., December 1, 2004; 183(6): 1539 - 1544. [Full Text] [PDF] |
||||
![]() |
T. A. Ponsky, Z. J. Huang, K. Kittle, M. R. Eichelberger, J. C. Gilbert, F. Brody, and K. D. Newman Hospital- and Patient-Level Characteristics and the Risk of Appendiceal Rupture and Negative Appendectomy in Children JAMA, October 27, 2004; 292(16): 1977 - 1982. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. A. Taylor Suspected Appendicitis in Children: In Search of the Single Best Diagnostic Test Radiology, May 1, 2004; 231(2): 293 - 295. [Full Text] [PDF] |
||||
![]() |
S. Kaiser, T. Finnbogason, H. K. Jorulf, E. Soderman, and B. Frenckner Suspected Appendicitis in Children: Diagnosis with Contrast-enhanced versus Nonenhanced Helical CT Radiology, May 1, 2004; 231(2): 427 - 433. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Poortman, P. N. M. Lohle, C. M. C. Schoemaker, H. J. M. Oostvogel, H. J. L. J. M. Teepen, K. A. H. Zwinderman, and J. F. Hamming Comparison of CT and Sonography in the Diagnosis of Acute Appendicitis: A Blinded Prospective Study Am. J. Roentgenol., November 1, 2003; 181(5): 1355 - 1359. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. P. Frush, L. F. Donnelly, and N. S. Rosen Computed Tomography and Radiation Risks: What Pediatric Health Care Providers Should Know Pediatrics, October 1, 2003; 112(4): 951 - 957. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. Tarantino, A. Giorgio, G. de Stefano, V. Scala, F. Esposito, G. Liorre, N. Farella, and G. Ferraioli Acute Appendicitis Mimicking Infectious Enteritis: Diagnostic Value of Sonography J. Ultrasound Med., September 1, 2003; 22(9): 945 - 950. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |