High false-positive diagnoses tolerated with CT colonography

Patients and healthcare professionals are willing to tolerate high rates of false-positive diagnoses with CT colonography in exchange for diagnosis of extracolonic malignancy, according to a study published online May 22 by Radiology.

Colorectal cancer screening with CT colonography often inadvertently images extracolonic abdominal and pelvic tissues, potentially detecting disease in organs that are not the primary target. Extracolonic screening is widely debated, as clinicians and policy makers have yet to reach a consensus about whether or not it is beneficial for population screening.

“It is unclear how individual patients and health care professionals balance the possibility of detecting life-threatening extracolonic pathologic findings against the larger chance of fruitless (or even harmful) testing precipitated by extracolonic findings,” wrote lead author Andrew A. Plumb, MA, MRCP, FRCR, of University College Hospital in London, and colleagues.

The researchers sought to determine the maximum rate of false-positive diagnoses that patients and healthcare professionals are willing to accept in exchange for detection of extracolonic malignancy by using CT colonography for colorectal cancer screening.

The study included 52 patients and 50 health care professionals who were involved with two discrete choice experiments. When faced with different scenarios, the participants had to choose between unrestricted CT colonography that examined intra- and extracolonic organs and CT colonography restricted to the colon.

The first of the two experiments detected one extracolonic malignancy per 600 cases with a false-positive rate that varied across scenarios from 0 percent to 99.8 percent. One experiment examined radiologic follow-up generated by false-positive diagnoses and the other examined invasive follow-up. Intracolonic performance was the same for both tests.

After calculation, it was found that the median tipping point for radiologic follow-up occurred at a false-positive rate greater than 99.8 percent. Participants would tolerate at least a 99.8 percent rate of unnecessary radiologic tests to detect an additional extracolonic malignancy.

The median tipping point for invasive follow-up occurred at a false-positive rate of ten percent. Results indicated that tipping points were significantly higher for patients than healthcare professionals in both experiments. At population prevalence of one in 600, this finding translates to 60 additional invasive tests per extracolonic malignancy.

Moreover, the median number of false-positive diagnoses tolerated per extracolonic malignancy was significantly higher for the radiologic testing than the invasive testing experiment, illustrating that additional imaging tests were perceived as more acceptable than additional invasive tests.

“The false-positive rate of screening CT colonography for extracolonic findings in current clinical practice is likely to be highly acceptable to both patients and health care professionals,” wrote Plumb and colleagues. “Our data, while framed in the context of CRC screening, potentially have wider implications for incidental findings discovered with other imaging modalities.”