| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Pediatric Imaging |
1 From the Division of Emergency Medicine (A.B.K., R.G.B.) and Department of Radiology (G.A.T.), Children's Hospital Boston, Harvard Medical School, Boston, Mass. Received January 30, 2006; revision requested March 28; revision received April 21; accepted May 25; final version accepted September 5. Address correspondence to A.B.K., Division of Pediatric Emergency Medicine, Morgan Stanley Children's Hospital of New York, Columbia University, 622 W 168th St, New York, NY 10069 (e-mail: ak493{at}columbia.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCE IN KNOWLEDGEReferences |
|---|
Materials and Methods: This HIPAA-compliant study was approved by the committee on clinical investigations. As part of a larger study, informed consent was obtained from all parents and from all children older than 7 years. Consecutive patients aged 5–21 years who presented to the emergency department and were suspected of having appendicitis were studied with CT. From April 2003 until February 2004, patients underwent intravenous and rectal contrast-enhanced CT. From March 2004 until December 2004, patients underwent intravenous contrast-enhanced CT. Demographic data, clinical outcomes, and test performance characteristics—including sensitivity, specificity, accuracy, and negative and positive predictive values—were compared.
Results: Of the 416 patients who met inclusion criteria, 223 underwent intravenous and rectal contrast-enhanced CT and 193 underwent intravenous contrast-enhanced CT. There were no differences in sex distribution (55% vs 52% male patients), frequency of appendicitis (36% vs 32%), or frequency of equivocal CT findings (4%) between the groups. Intravenous and rectal contrast-enhanced CT had a sensitivity of 92% (95% confidence interval [CI]: 85%, 97%), a specificity of 87% (95% CI: 79%, 92%), a negative predictive value of 94% (95% CI: 90%, 98%), and an accuracy of 89% (95% CI: 85%, 93%). Intravenous contrast-enhanced CT had a sensitivity of 93% (95% CI: 84%, 97%), a specificity of 92% (95% CI: 85%, 96%), a negative predictive value of 95% (95% CI: 90%, 99%), and an accuracy of 92% (95% CI: 88%, 96%) (P > .2 for all comparisons).
Conclusion: There was no significant difference between the performance of intravenous contrast-enhanced CT and that of rectal and intravenous contrast-enhanced CT in children suspected of having appendicitis.
© RSNA, 2007
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCE IN KNOWLEDGEReferences |
|---|
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCE IN KNOWLEDGEReferences |
|---|
Study Setting and Patients
Our urban, tertiary care, pediatric emergency department has approximately 52 000 visits per year. As part of a larger prospective study, we identified all children aged 5–21 years who visited our emergency department between April 2003 and December 2004 and for whom a pediatric surgeon was consulted by a pediatric emergency medicine attending physician to evaluate for appendicitis (17). Two authors (A.B.K. and R.G.B) reviewed the medical records of these patients to identify children who had undergone CT as part of their evaluation in the emergency department. Imaging was considered after initial evaluation by a pediatric emergency medicine physician and subsequent examination by a 4th- or 5th-year surgical resident. All eligible patients were given intravenous contrast material. Bowel opacification was achieved via administration of rectal contrast material.
Exclusion criteria for administration of rectal contrast material were age younger than 4 years, developmental delay, inability to tolerate or retain rectal contrast material, presence of bloody stools, or preexisting conditions that placed the patient at high risk for intestinal perforation (such as inflammatory bowel disease, organ transplants, oncologic disorders, or collagen vascular disorders). These children received oral contrast material 1 or 2 hours before imaging and were excluded from the study population. Children younger than 5 years were excluded from the study population because of their high rates of perforation and our experience that they often do not tolerate rectal contrast material. In addition, patients were excluded from this study if they were pregnant, had previously undergone abdominal surgery, or had a chronic medical condition (eg, cystic fibrosis, inflammatory bowel disease, or sickle cell anemia). The CT techniques used were clinically approved protocols at our institution at the time of the study, and all imaging studies were clinically indicated.
CT Imaging Techniques
In this study, two helical CT techniques were used to evaluate patients with clinically suspected appendicitis: CT with rectal and intravenous contrast material or CT with intravenous contrast material alone. Before February 2004, our institutional CT protocol for evaluation of patients suspected of having appendicitis was to administer rectal and intravenous contrast material. In February 2004, the institutional imaging protocol was amended to involve administration of only intravenous contrast material on the basis of the literature and changing standards of care. All patients underwent CT performed with a single–detector row helical GE 9800 HiLite scanner (GE Medical Systems, Milwaukee, Wis). Images were acquired from the third lumbar vertebral body through the pubic ramus at a pitch of 1.5:1.0, with 5-mm collimation and images reconstructed at 3-mm intervals. Rectal contrast material was administered as described elsewhere (16,18). Between 500 and 1500 mL of 17% iothalamate meglumine (Cysto-conray II; Mallinckrodt, St Louis, Mo) was administered per rectum via an enema catheter with a flexible tip (Junior Flex-tip; E-Z-Em Canada, Montreal, Quebec, Canada) Nonionic intravenous contrast material (Optiray 320; Mallinckrodt) was administered to all eligible patients at a dose of 2 mL per kilogram of body weight (maximum dose, 150 mL) by using rapid bolus infusion. Peak kilovoltage was held constant, and milliamperage settings were adjusted according to patient age (19,20).
Imaging Diagnostic Criteria
CT scans obtained between 8 am and 10 pm were immediately interpreted by a board-certified staff pediatric radiologist. Our department of radiology has 19 pediatric radiologists on staff, all of whom hold certificates of added qualification in pediatric radiology. One of them (G.A.T.) is a coauthor of this article. Departmental radiologists had 6–25 years of experience in the interpretation of pediatric abdominal CT scans. Between 10 pm and 8 am, a 1st-year pediatric radiology fellow immediately interpreted all CT scans. All pediatric radiology fellows were board certified in diagnostic radiology and had previous experience with pediatric abdominal CT before they started their fellowship. Experience with pediatric abdominal CT among fellows ranged from 4 to 5 years. All interpretations were recorded before final diagnoses were known. The department of radiology tracks discrepancies between pediatric radiology fellows and subsequent staff readings. None occurred in this study.
The official radiology reports were abstracted from the medical records and recorded as positive, negative, or equivocal for appendicitis on the basis of the report of the attending radiologist (A.B.K. and R.G.B.). Criteria for the diagnosis of appendicitis with CT included the presence of direct signs, such as an enlarged appendix (>6 mm in diameter), a nonopacified appendiceal lumen, and substantial wall enhancement with intravenous contrast material. Indirect CT signs included periappendiceal fat stranding, appendicolith, free fluid or abscess in the right lower quadrant or pelvis, or thickening of the focal cecal wall (irregular thickening of 2 mm or more) (16,21). A CT scan was considered positive if the preceding findings were detailed in the radiology report or if the radiologist explicitly commented that the findings were consistent with appendicitis. A CT scan was considered negative if the radiologist stated that there was no evidence of appendicitis. A CT examination in which the appendix was not depicted and no additional secondary signs of appendicitis (fat stranding, abscess, phlegmon) were depicted was also recorded as a negative study. An equivocal CT examination was defined as a study in which the radiologist specified that appendicitis could not be excluded, that the study findings were "equivocal," or that the study findings were "not conclusive" or "inconclusive." All authors, including a senior radiologist (G.A.T.), agreed on the key phrases with which to codify the final reading. Conflicts in coding the final CT reading were resolved by two authors (A.B.K. and R.G.B.). The senior radiologist involved with this study (G.A.T.) has more than 25 years of experience in pediatric abdominal imaging.
Outcome Measures, Final Diagnosis, and Follow-up
Final diagnosis was determined with pathologic examination for patients who had undergone appendectomy. A perforated appendix was determined according to the attending surgeon's postoperative diagnosis in the patient's medical record (A.B.K.). In patients who did not undergo surgery, the outcome was confirmed with a follow-up telephone call 2–4 weeks after the emergency department visit (A.B.K.). If the family could not be reached, the patient's pediatrician was contacted to determine the final diagnosis. Patients or primary physicians were asked whether the patient had visited any emergency department after the initial visit to our emergency department, whether the abdominal pain had persisted or improved, and whether the patient had undergone appendectomy at another institution.
Statistical Analysis
We compared the performance of CT with intravenous and rectal contrast material with that of CT with intravenous contrast material alone. We also compared the performance of these different imaging strategies in a subset of children younger than 10 years to test performance in prepubescent patients. Sensitivity, specificity, accuracy, and negative and positive predictive values were calculated (SPSS, version 11.5; SPSS, Chicago, Ill). In addition, we tracked the percentage of cases that were negative, positive, or equivocal according to CT modality. Baseline demographic and clinical characteristics of the two groups were compared by using a two-tailed t test for continuous variables and 
2 analysis or Fisher exact test for categorical variables and comparisons of test performance. A P value 
.05 was considered to indicate a significant difference. Confidence intervals were calculated for proportions by using the Newcombe-Wilson method (22). Positive and negative predictive values were determined by using Bayes theorem, assuming an appendicitis prevalence of 42%. Confidence intervals for negative and positive predictive values were constructed with the Wald formulation (23). We estimated that a clinically significant difference in accuracy between intravenous and rectal contrast-enhanced CT and intravenous contrast-enhanced CT would be between 5% and 10%. To detect a 7% difference in accuracy with 80% power and an 
value of .05, 200 patients were required in each group (PS: Power and Sample Size Calculations, version 2.1.30; Department of Biostatistics, Vanderbilt University, Nashville, Tenn). The study was powered to determine superiority of one CT technique over another.
A CT scan interpreted by the radiologist as positive for appendicitis was categorized as a true-positive finding when appendicitis was proved according to either pathologic findings or clinical course (in patients who underwent an interval appendectomy) and as a false-positive finding when pathologic findings were negative for appendicitis. A CT scan interpreted as negative for appendicitis by the radiologist was categorized as a true-negative finding when the pathologic findings were negative or if the patient did not undergo an appendectomy and was clinically well at 2-week follow-up; it was categorized as a false-negative finding when the pathologic findings were positive for appendicitis. Additionally, for the most conservative estimate of test performance, equivocal radiologic interpretations for appendicitis were recategorized as false-negative findings in patients in whom appendicitis was diagnosed on the basis of pathologic findings and as false-positive findings in patients who did not have appendicitis at pathologic examination or did not undergo surgical care during the follow-up period.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCE IN KNOWLEDGEReferences |
|---|
|
|
Test Performance Characteristics for the Two CT Techniques
The percentage of scans interpreted as equivocal was approximately 4% for both imaging modalities (eight of 223 for intravenous and rectal contrast-enhanced CT and seven of 193 for intravenous contrast-enhanced CT, P > .99) (Table 2). Sensitivity, specificity, and overall accuracy of both scanning techniques (Table 3) were not significantly different (P > .2 for all comparisons).
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCE IN KNOWLEDGEReferences |
|---|
In 2001, Lowe et al (14) examined the utility of unenhanced CT in the diagnosis of appendicitis in children. In their cohort of 72 children (median age, 13 years), they reported a sensitivity of 97% (95% CI: 91%, 100%), a specificity of 100% (95% CI: 96%, 100%), and an accuracy of 99% (95% CI: 96%, 100%). The authors commented that the colon and cecum were readily identifiable in children and that enteric contrast material was unnecessary (14). In a larger more recent study, Kaiser et al (12) examined 306 children (mean age, 10.6 years) to elucidate the benefit of intravenous contrast-enhanced CT versus unenhanced CT in the diagnosis of appendicitis. Children first underwent focal unenhanced imaging of the lower abdomen followed by imaging of the entire abdomen after injection of intravenous contrast material. Kaiser et al (12) reported a dramatic improvement in test performance with the addition of intravenous contrast material, with a sensitivity of 90% and a specificity of 94%. The authors recommended intravenous contrast-enhanced CT be used to diagnose appendicitis in children because of the smaller amount of intraperitoneal fat found in young children (12).
At our institution, the use of rectal contrast material without intravenous contrast material was the standard CT protocol used until May 2000 to examine patients suspected of having appendicitis (16). Although this technique was effective (sensitivity, 94%; specificity, 94%; negative predictive value, 97%; positive predictive value, 90%; accuracy, 94%) (9), the appendix was not easily seen in many small or slender patients because of the relative paucity of peritoneal fat in children. This resulted in reimaging a limited area (at the level of the ileocecal valve) after administration of intravenous contrast material or with the patient in the left lateral decubitus position.
Because 38% of our patients underwent additional imaging after administration of intravenous contrast material, we changed our clinical CT protocol in May 2000 to include the standard use of intravenous contrast material with rectal contrast material in all appropriate patients. Rectal contrast material allows identification of the cecum and patency of the base of the appendix. In many patients who underwent CT for other clinical indications, we noted that the position of the cecum was easily identifiable by the presence of intraluminal stool and gas. In addition, failure of contrast material to enter the appendix in patients examined with rectal contrast-enhanced CT can be an indirect sign of luminal obstruction and resultant appendicitis. However, the presence of an appendix that does not admit enteric contrast material or only partially fills adds to the diagnostic uncertainty for the radiologist, since these findings can be seen in patients with a normal or diseased appendix. In addition, in a small number of patients, enteric contrast material may mask the presence of an appendicolith with similar attenuation as CT. As a result of this experience and a review of the literature, we modified our protocol to eliminate rectal contrast material from our standard CT protocol in February 2004.
As in previous studies in pediatric and adult populations, we found a high sensitivity, specificity, and accuracy of CT with intravenous contrast material only. In addition, in our subgroup analysis of children younger than 10 years, intravenous contrast-enhanced CT performed similarly to intravenous and rectal contrast-enhanced CT. However, this study was underpowered to enable detection of a difference in this subgroup. With the limited sample size, we would have been able to detect a difference of 18% or more in test performance with 80% power.
Approximately 3% of the CT examinations obtained in our study had an equivocal result. Prior articles have not reported the role of equivocal studies. We used real-time radiology reports to determine CT performance because we believe that the radiologist's reading at the actual moment of medical decision making better reflects actual imaging performance than does the reading by a panel of radiologists blinded to clinical information. Of note, we found no statistical difference in the rate of equivocal studies between the two different imaging techniques. Interestingly, the majority of equivocal studies were in children younger than 10 years (four of eight patients who underwent intravenous and rectal contrast-enhanced CT and five of seven patients who underwent intravenous contrast-enhanced CT).
We believe that our study has several strengths. To our knowledge, our study represents the largest cohort of pediatric patients evaluated for appendicitis with two imaging techniques. Unlike previous studies, we directly compared two imaging techniques while keeping other aspects of our imaging protocol constant. The study groups had statistically similar ages and rates of perforated and nonperforated appendicitis. Another strength is that we determined the final diagnosis of our discharged patients through follow-up information rather than at the time of emergency department disposition.
Several limitations should be noted. First, this was an observational study; thus, we were not able to randomize patients to the different imaging techniques. Second, our study was limited to one institution; therefore, our results may not be applicable to other clinical settings. Third, our study was powered to detect only a 7% difference in accuracy between the two imaging techniques. While we could not detect small differences between the two imaging techniques, we felt that a difference of less than 7% would not be clinically relevant. Fourth, CT images were not reviewed by a panel of radiologists. We relied on the report dictated by the attending radiologist on duty during the emergency department visit of each patient. Thus, over the course of our study, several different attending radiologists read the individual images. The variability in the individual skill or confidence in reading of a particular study could affect the reported performance of CT. However, with a group of radiologists, the actual performance might be more realistic than relying on a select group of reviewers.
We conclude that the performance of CT with only intravenous contrast material is not significantly different from that of CT with both rectal and intravenous contrast material in the evaluation of children suspected of having appendicitis.
|
ADVANCE IN KNOWLEDGE |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCE IN KNOWLEDGEReferences |
|---|
|
FOOTNOTES |
|---|
Abbreviations: CI = confidence interval
Authors stated no financial relationship to disclose.
Author contributions: Guarantors of integrity of entire study, all authors; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; approval of final version of submitted manuscript, all authors; literature research, A.B.K., R.G.B.; clinical studies, all authors; statistical analysis, all authors; and manuscript editing, all authors
|
References |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCE IN KNOWLEDGEReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  K. Garcia, M. Hernanz-Schulman, D. L. Bennett, S. E. Morrow, C. Yu, and J. H. Kan Suspected Appendicitis in Children: Diagnostic Importance of Normal Abdominopelvic CT Findings with Nonvisualized Appendix Radiology, February 1, 2009; 250(2): 531 - 537. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
