CT colonography (CTC) can increase colorectal cancer screening capacity in rural underserved communities by pairing local CT scanning with remote interpretation, according to a study in the November issue of the American Journal of Roentgenology.
Although the incidence of colorectal cancer among Native Americans is rising, screening rates remain far below those of the general U.S. population. Rural Arizona medical centers that serve Native American populations report poor compliance with fecal occult blood testing (due to a lack of modern plumbing and inadequate access to the postal service), a lack of flexible sigmoidoscopy and limited availability of optical colonoscopy. The University of Arizona radiology department remotely interprets many local imaging procedures.
The university sought to expand teleradiology service to included CTC screening and designed the study “to determine whether adequate examinations could be obtained with remote supervision after brief on-site instruction,” according to Arnold C. Friedman, MD, department of radiology at the University of Arizona in Tucson, and colleagues.
The researchers provided a one hour CTC presentation and demonstrated a CTC exam for technologists employed at two rural medical centers serving the primarily Native American population.
After training, primary care providers could order screening or diagnostic CTC. The study included 321 patients undergoing CTC between May 21, 2008 and June 8, 2009. Mean age was 60.9 years (range 35 to 88 years), 66 percent of participants were women, 34 percent were men and 87 percent of studies were screening exams.
The authors assessed image quality, grading images on a five-point scale and tabulating CTC Reporting and Data System (C-RADS) colon (C) and extracolonic (E) scores for clinical reports.
Friedman and colleagues reported that approximately 92 percent of patients had diagnostic quality exams according to measured parameters: stool, fluid and distention. Retrospective review of reports revealed similar results with 8 percent of studies nondiagnostic.
The authors reviewed potentially correctable incidents with CT technologists whenever possible, and techs referred to radiology reports for ongoing feedback.
Twenty-seven patients had C-RADS C2 exams; 24 of the 27 patients had polyps between 6 and 9 mm. Optical colonoscopy findings were available for 12 of the patients and confirmed polyps as true positive in 41.6 percent of these cases. The authors calculated an overall true-positive rate of 46 percent after one patient had a 6 mm polyp confirmed with follow-up CTC after one year.
The authors reported 17 patients with C3 CTC studies with at least one polyp 10 mm or larger or three or more 6- to 9-mm polyps. Fifteen patients underwent optical colonoscopy or surgical followup, which confirmed CTC results in 40 percent of patients.
Four percent of patients had E scores of E3 (“likely unimportant finding but incompletely characterized”), after retrospective review of the CTC reports. Another 4 percent had E4 exams, (“potentially clinically important”).
“Our results show that CTC can be introduced with minimal effort to rural underserved communities, adequately performed locally, and then interpreted remotely,” concluded Friedman, who pointed out that the seemingly low positive predictive value between 40 and 46 percent is higher than the 23 to 40 percent reported in the American College of Radiology Imaging Network trial for polyps 6 mm or larger. The authors also recommended optical colonoscopy followup for low-confidence CTC results, which may have contributed to the false positive rate.
Friedman and colleagues shared several obstacles: patient resistance to optical colonoscopy after a positive CTC, the need to repeatedly share indications for CTC with referring physicians and a perceived higher incidence of suboptimal studies with novice CT techs.
The authors also referred to several limits to the study. Specifically, many patients did not undergo optical colonoscopy, which may translate into false-negative colorectal polyps or CTC-missed cancers. In addition, the small study did not employ the randomized clinical trial model and it involved only one reader.
Ultimately, the authors stated that the data suggest the feasibility of CTC in underserved rural areas, noting that the model adheres to the NIH Colorectal Cancer Screening Conference Panel’s recommendation of