Appendectomy ranks among the most common surgical procedures in the U.S. But in a surprising number of casesan estimated 40,000 a yearthe removed appendix is not actually infected. In fact, according to a report appearing today in the Journal of the American Medical Association, the rate of misdiagnosis of appendicitis is increasingdespite the availability of improved tests. Such misdiagnoses translate to more than $700 million dollars a year in hospitalization costs and can leave patients with unnecessary post-operative complications.

Inflammation of the appendix is understandably difficult to identify. The tiny organ itself lies hidden behind the colon, and the characteristic pain, loss of appetite and nausea can stem from other maladies. For example, in men an inflamed lymph node or intestinal tract infection can cause these symptoms, whereas in women the source may be infection of the uterus or ovary. In recent years, however, researchers have developed a number of diagnostic tests based on computed tomography, ultrasound, laparoscopy and other techniques that are purportedly effective detectors of appendicitis. Yet the new study, led by David Flum of the University of Washington, indicates that between 1987 and 1998a time when the improved tests were becoming more widely availablethe rate of misdiagnosis among patients older than 65 shot up by eight percent a year and among women of childbearing age, it climbed by one percent a year. (Overall, appendicitis misdiagnosis has remained at 15 percent.) Today, nearly a quarter of all appendectomies performed on women of childbearing age remove healthy appendices.

"It's very striking to have such a contrast between published reports of how good these tests are and this new population-based evidence that they are not making a big impact," Flum observes. One possible explanation is that the results may not come quickly enough. "When you are suspicious that someone has appendicitis, you don't want to be wrong and leave the appendix in. Untreated appendicitis can lead to rupture and even more complications, so many surgeons prefer to take it out even if it looks normal at the time of the operation," he notes. Whatever the reason, Flum remarks, "the promise of this new diagnostic technology has not been realized. We need to conduct more studies to determine what is going on when these new technologies are applied around the country