Large study confirms CT’s accuracy for acute appendicitis
Thu, 05/20/2010 - 23:53 by Ron Ragan
Large study confirms CT’s accuracy for acute appendicitis
By Eric Barnes
AuntMinnie.com staff writer
May 20, 2010
SAN FRANCISCO – One of the largest CT studies to date of acute appendicitis has confirmed the modality’s sky-high diagnostic accuracy. The results, unveiled on Wednesday at the International Society for Computed Tomography (ISCT) meeting, also showed that CT is relatively insensitive to imaging technique, though the researchers said they have come to favor the use of oral contrast.
“I don’t think I have to convince you that acute appendicitis is an important clinical problem. The lifetime risk is about 7% to 8%, and it’s as much an adult disease as it is children only, with a median age of 22 years and a mean age that is considerably higher,” said Dr. Perry Pickhardt, professor of Radiology at the University of Wisconsin in Madison.
For its part, MDCT, the imaging method of choice for acute appendicitis, delivers diagnostic performance that is generally well established, with pooled sensitivity and specificity exceeding 95% in older studies.
In the more recent multislice era, there are less data on accuracy and more on utilization, Pickhardt said; other trends include the increasing use of MRI and ultrasound in certain groups such as pregnant and pediatric patients.
As for the older CT studies, Rao et al in 1998 had only 100 patients, but their study showed a remarkable rate of management change in nearly half the cases after CT, Pickhardt said. Flum and colleagues countered in 2001 with conflicting data showing no change in the negative appendectomy rate with CT, but the imaging utilization rate was unknown and probably quite low in that single-slice CT series, he said.
And at least three other studies covering the period between 2000 and 2006 (including Raman et al, 2008, and Coursey et al, 2010) with a prevalence of at least 20% confirmed that the negative appendectomy rate dropped from more than 20% to less than 10% with the use of CT, Pickhardt said.
“One thing that’s lacking is the actual diagnostic performance or accuracy with MDCT … so we looked at what I think is the largest CT-based cohort for appendicitis,” a total of 1,666 adults with suspected acute appendicitis between 2000 and 2006, Pickhardt said.
The prevalence of disease was 24.4% (20.8% women, 32.6% men), Pickhardt said, but other than that no diagnostic differences between the men and women were apparent. MDCT’s diagnostic performance in the cohort was as follows:
Sensitivity = 98.0%
Specificity = 98.2%
Negative predictive value = 99.4%
Positive predictive value = 94.5%
Accuracy = 98.1%
Area under the receiver operator characteristics (ROC) curve = 0.995 ± 0.0016
“We were very pleased to see that sensitivity, specificity, and accuracy were all over 98%, with high predictive values as well” Pickhardt said.
“Our negative appendectomy rate was under 10% as the other studies have shown, but, interestingly, it would have been under 4% had 21 cases that were negative for CT not gone on to appendectomy,” he said. “I think it goes to show that in some cases we should really stick to our guns and trust our instincts.”