CT outperforms ultrasound for diagnosing appendicitis in children, especially those over 10 years old, according to a study in the October American Journal of Roentgenology. But is the higher sensitivity worth the radiation dose? In comments posted to the European newsgroup Eufora, radiologists question the routine use of CT in children, and wonder if operator experience and other factors may have influenced the outcome.
According to a retrospective study by Drs. Carlos Sivit, Kimberly Applegate, and colleagues at the Rainbow Babies and Children's Hospital and Case Western Reserve School of Medicine in Cleveland, spiral CT is more sensitive than graded compression sonography for diagnosing appendicitis in children and young adults -- particularly those over 10 years old (AJR, Oct. 2000; Vol. 175: 4, pp. 977-980).
Between June 1996 and April 1999, 386 patients aged 1-21 with suspected appendicitis underwent sonography, CT, or both at the hospital. The total included 71 who underwent CT only and 233 who underwent sonography only. Eighty-two patients underwent both CT and US, according to the authors.
The imaging modality was determined at the discretion of the pediatric surgeon or emergency department physician, and the pediatric surgeon decided whether to operate, according to the article. Patients with unequivocal diagnoses underwent laparectomy without imaging, those with equivocal findings underwent sonography, and still-equivocal cases proceeded to CT imaging within 24 hours of sonographic evaluation, the authors wrote.
All ultrasound exams were performed using a 5 MHz or 7.5 MHz linear-array transducer (XP10 Acuson, Mountain View, CA), and oblique, transverse, and longitudinal scans of the right lower quadrant were obtained with the graded compression technique.
CT was performed on a spiral CT scanner (PQ 5000, Marconi Medical Systems, Cleveland) with intravenous contrast medium and/or opacification of the gastrointestinal tract. Collimation varied from 4 mm to 10 mm based on patient age, with thinner collimation employed in younger patients. However, the protocol was changed during the course of the study.
"The collimation and image reconstruction through the lower abdomen and pelvis were changed from 8-10 mm to 4 mm during the study period because we found the thinner collimation allows better visualization of anatomic structures in the right lower quadrant," the researchers wrote, acknowledging the study limitation implied by the change.
As confirmed at surgery and histology, appendicitis was found in 26% (83/315) of patients undergoing sonography, and in 40% (61/153) of those undergoing CT, the authors wrote. For the diagnosis of appendicitis, graded compression sonography had a sensitivity of 78%, specificity of 93%, and accuracy of 89%, whereas CT had sensitivity of 95%, specificity of 93%, and accuracy of 94%.
While the findings for patients 10 years and younger were not statistically significant between the two modalities, CT performed significantly better in patients more than 10 years old. In this subset of patients, CT had a sensitivity of 97% vs. 71% for sonography, and somewhat higher accuracy (97% vs. 91% for sonography), while differences in specificity were statistically insignificant, according to the authors.
"This finding is not surprising given that graded compression sonography is typically easier to perform in individuals with a smaller body habitus," the authors wrote.
CT was particularly useful for discordant examinations, according to the authors. When CT results were at odds with sonographic findings, the CT diagnosis was correct in 85% (17) of these cases, while the sonographic diagnosis was correct in only 3 (15%), they stated.
"It is clear ... that because of the significantly higher sensitivity of CT when compared with sonography ... a negative CT examination instills greater diagnostic confidence for the exclusion of appendicitis than do negative sonographic findings."
The findings underscore the authors' opinion that CT should have a greater role in diagnosing appendicitis in children, despite "valid concerns" regarding CT's higher cost, and exposure to radiation and iodinated contrast agents.
"These concerns may be better addressed by placing more emphasis on the use of clinical variables for stratifying risk of disease and allowing more selective use of imaging," the authors wrote.
The day after the study appeared online, two European radiologists expressed their concerns in messages posted on the Eufora newsgroup.
Eufora founder Dr. Erik Ranschaert had questions about the study's methodology, operator experience, and criteria for diagnosing appendicitis with US.
"I will not argue the fact that CT is more accurate in detecting acute appendicitis in adult and especially obese patients. It seems quite ridiculous to me however that children should receive a CT scan including oral and rectal opacification of the GI tract, instead of doing a US first," he wrote. "I believe it's better to start with US and really to keep CT for those cases with discordant clinical, laboratory, and US findings."
Another Eufora member, Dr. Gerald Pärtan, also questioned the study's methodology.
"It is not at all surprising that CT yielded better diagnostic performance than US, given the conditions under which both were compared, especially the fact that some of the CT [exams] were performed after US on the same patients," Pärtan wrote.
For their part, the authors noted that the methodology wasn't perfect, and that sonographic findings are highly operator-dependent. They acknowledged that sonography was always performed before CT in the study, which could have influenced the results. And in some cases the radiologist who monitored the CT exam may have also performed the sonographic exam, so radiologists were not blinded to the sonographic findings, they noted.
"This procedure reflects how these two imaging examinations are used in our clinical practice," they wrote.
According to Pärtan, even if CT does prove superior to US under European imaging standards, its sensitivity and specificity are less important than the overall clinical outcome. Once the study's shortcomings are accounted for, any remaining advantages of CT would have to be weighed against the chance of lethal cancer in children -- estimated by the International Commission on Radiological Protection at 10% per Sievert.
"I do not understand why especially in the U.S., doctors have to worry about every possible complication, however radiation dose never seems to be included [in] this kind of thinking, obviously not even in pediatric patients," Pärtan wrote.
By Eric Barnes
AuntMinnie.com staff writer
September 27, 2000
Europeans tout CT to check appendix in routine practice, April 12, 2000
Pass on the Rolaids: CT scans for appendicitis stump radiologists, May 11, 2000
Studies measure appendix size, shape for better US diagnosis, March 23, 2000
Copyright © 2000 AuntMinnie.com