ADVERTISEMENT

Weigh Image Options in Diagnosing Appendicitis : Decide between computed tomography and ultrasound based on the imaging tools' strengths and weaknesses.

Author and Disclosure Information

CHICAGO — One way to avoid diagnostic pitfalls in pediatric appendicitis is by using imaging studies effectively, Dr. George Taylor said at the annual meeting of the Radiological Society of North America.

The decision to use diagnostic computed tomography (CT) versus ultrasound should be based on a thorough understanding of the institution's strengths and weaknesses. “As in politics, all imaging is local,” said Dr. Taylor, chief of radiology at Children's Hospital Boston.

The trade-off with ultrasound is that it is radiation free, but the miss rate for ultrasound is highly dependant on the prevalence of appendicitis in the area. A local prevalence rate of 33% at Children's Hospital Boston meant that ultrasound imaging was missing 40% of appendicitis cases, Dr. Taylor said.

This finding is consistent with a recent meta-analysis that shows CT had a significantly higher sensitivity than did ultrasound in diagnostic studies of children and adults (Radiology 2006;241:83–94).

Dr. Taylor recommends using ultrasound first, followed by CT if the ultrasound findings are negative. When CT is used, efforts should be made to minimize radiation exposure by reducing the area scanned and optimizing the dose according to the child's weight and age.

The classic study on radiation risk (Am. J. Roentgenol. 2001;176:289–96) was based on the assumption that pediatric abdominal CT images were acquired at 404 milliampere-seconds (mAs) and scanned from the diaphragm to the symphysis pubis. However, Children's Hospital acquires CT images from the iliac crest to the symphysis pubis at 230 mAs. The scan results in a whole-body dose of 18–30 mCi.

At this dose, the increased risk of all cancers is 0.009%–0.15%. This is a small but significant increased risk that is acceptable when viewed against the background lifetime risk of cancer at 16%–20%, and the risk involved in having a perforated appendix, Dr. Taylor said.

Another decision to be made is the type of contrast agent used for CT scans. Among 589 children seen at Children's Hospital for suspected appendicitis, the overall visualization of the appendix was 92% for all contrast agents, 93% for rectal contrast only, 94% for rectal plus IV, and 78% for oral plus IV.

Rectal agents would appear to be the best approach, and indeed, oral and IV contrast agents were initially reserved for younger children, those with developmental delays, or those who couldn't tolerate rectal contrast. But a review of the records revealed that when only rectal contrast was used, there was so much diagnostic uncertainty that an additional CT scan was ordered in a stunning 46% of cases—a number that Dr. Taylor called unacceptable.

Because many of the additional scans required IV contrast, the hospital began using rectal and IV agents.

During the first 6 months, there was no change in the number of additional CT scans being ordered.

But after this transitional period, the percentage fell to about 3%, and after adopting an IV-only approach, there has not been a single repeat CT ordered in the last 3 years, he said.

Not only is the radiation exposure reduced, but so is the number of complaints voiced about the invasiveness of rectal contrast.

Other diagnostic challenges with CT include reduced visualization of the appendix because of a lack of mesenteric fat, especially in children younger than 6 years; a tendency to overdiagnose appendicitis; lower sensitivity for ovarian pathology; the risk of perforation with rectal contrast; and false-positive findings, most commonly associated with an appendix greater than 6 mm in size, Dr. Taylor said.

A local prevalence rate of 33% meant that ultrasound was missing 40% of appendicitis cases. DR. TAYLOR

Although this CT was suggestive of appendicitis, the patient was eventually diagnosed with a normal appendix and a case of viral gastroenteritis. Courtesy Dr. George Taylor