The authors collected data from some 55 hospitals of all types and sizes over a six-year period for more than 19,300 patients older than age 15; 91% of patients underwent one or more imaging studies. There were 16,852 (87.2%) CT scans, 1160 (6.0%) ultrasounds and 108 MRIs (0.6%) with 1677 (8.7%) patients having no imaging.
If a patient had an imaging study, the rate of normal appendix removal was only 4.5% compared to 15.4% for those who were not imaged, p < 0.001. Stated another way, on multivariate analysis a patient who had no imaging was over 3 times more likely to have a normal appendix at surgery. Only 4.1% of those undergoing CT scan had a normal appendix compared to 10.4% of those who had only an ultrasound, p < 0.001.
The difference in the rate of normal appendix removal was true regardless of gender—men 3% with imaging vs. 10% without and women 6.9% vs. 24.7%, both statistically significant differences. In other words, nearly 1/4 of all women who were not imaged had a normal appendix at surgery.
Perforated appendicitis occurred in 15% of all patients, a figure that has remained constant over many years and has not been influenced by imaging.
The Washington appendicitis experience is probably generalizable because in most hospitals, CT scan is the preferred imaging modality over ultrasound, which is more operator-dependent, less accurate and less available during off hours than CT scan.
My personal experience is similar to that of the authors of the study. My rate of removal of a normal appendix when operating for the presumed diagnosis of appendicitis is 4.3% for my last 200 cases. Only 18 (9%) of my patients, all young males, had no imaging and just 1 had a normal appendix. CT scans were the only studies performed in 99% of my patients who were imaged.
I have previously written that patients and their families seem to prefer an accurate diagnosis over a theoretical slightly higher cancer risk 20 or 30 years later. The paper from Washington also confirms that a high accuracy rate of CT scanning for appendicitis is achievable outside of academic centers.
For many reasons, I can’t imagine the rate of CT scans for the diagnosis of patients with right lower quadrant pain decreasing. The accuracy of CT is no longer an issue. Most ED MDs are scanning everyone with right lower abdominal pain. They may tell you it’s because surgeons insist upon it, but whatever the reason, a lot of CT scans are being done. Another advantage of CT scanning is if the diagnosis is something other than appendicitis, the CT will very often reveal what is wrong with the patient. In a previous blog, I have mentioned before that it would be impractical to ask surgeons to examine every patient with abdominal pain, one of the more frequent complaints of patients presenting to emergency departments.
Face it, CT scanning for the diagnosis of appendicitis is here to stay. Whether implementation of protocols to lower the radiation dose of the devices will calm all the doomsday prophets remains to be seen.
[Thanks to Dr. Frederick Thurston Drake, lead author of the paper from Washington, for providing me with information not published.]