Extracolonic findings (ECF) on CT colonography (CTC) in patients presenting with colorectal symptoms

Abstract

Background Computed tomographic colonography (CTC) is now an established method for imaging the colon and rectum in the screening and symptomatic setting. Additional benefit of CTC is the ability to assess for extracolonic findings especially in patients presenting with colorectal symptoms. Purpose To determine prevalence of extracolonic findings (ECF) in symptomatic patients undergoing CTC and determine accuracy of CTC for exclusion of significant abdominal disease and extracolonic malignancy (ECM). Material and Methods A total of 1359 unenhanced prone and postcontrast supine CTC studies were performed between March 2002 and December 2007. ECF were retrospectively classified according to C-RADS criteria into E1 to E4 findings. For ECM, a gold standard of clinical and/or radiological follow-up supplemented with data from the regional cancer registry with a median follow-up of 42 months was created. Sensitivity and negative predictive values for ECM was calculated. Results Following exclusions, 1177 CTCs were analyzed. Of 1423 extracolonic findings reported, 328/1423 (23%) E3 and 100/1423 (7%) E4 (including six eventual FP studies) findings were identified. Thirty-two ECMs were confirmed following further investigations. Seven further small ECMs were detected during the entire follow-up, of which two were potentially visible in retrospect (false-negative studies). Additional tests were generated from 55/1177 (4.7%) studies. Sensitivity and negative predictive value for ECM was 94.1% (95% CI 78.9–98.9%) and 99.8% (95% CI 99.3–99.9%), respectively. Conclusion One in 37 patients were found to have an ECM. Two potentially detectable cancers were missed. Only a small proportion of patients underwent additional work-up.