Virtual endoscopy--comparison with colonoscopy in the detection of space-occupying lesions of the colon

Endoscopy. 2000 Mar;32(3):226-32. doi: 10.1055/s-2000-100.

Abstract

Background and study aims: A new technique has been described which combines abdominal helical computed tomography (CT) scanning and virtual reality computer technology, known as virtual colonoscopy (VC); the reconstructed images provide a simulation of the interior of the colon as viewed by endoscopy. We compared VC with conventional colonoscopy in patients with suspected or known colonic neoplasia.

Patients and methods: A total of 38 patients, in whom there was a high likelihood of colonic polyps or cancer, underwent a noncontrast helical CT scan of the abdomen and pelvis after regular colonoscopy bowel preparation. The images were reconstructed into a VC presentation and compared with the subsequent conventional colonoscopy in a blinded manner.

Results: Conventional colonoscopy identified a total of 24 polyps 5 mm or greater. VC correctly identified five of 13 polyps 5-9 mm in size, and ten of 11 lesions greater than or equal to 10 mm in diameter. The reasons for four missed lesions were identified as being secondary to a collapsed rectum in two patients and stool in the right colon in two patients. The sensitivity and specificity per patient of VC for lesions greater than or equal to 5 mm were 66.7% and 75.0% respectively, and for lesions greater than 1 cm were 90.0% and 82.1%, respectively.

Conclusions: Virtual colonoscopy is feasible, well tolerated, and capable of detecting most lesions greater than 10 mm in diameter. This technique is continuing to be developed and warrants further evaluation as a diagnostic and screening tool in colorectal neoplasia.

Publication types

  • Comparative Study

MeSH terms

  • Adult
  • Aged
  • Aged, 80 and over
  • Colonic Neoplasms / diagnosis*
  • Colonic Neoplasms / diagnostic imaging
  • Colonic Polyps / diagnosis*
  • Colonic Polyps / diagnostic imaging
  • Colonoscopy / methods*
  • Female
  • Humans
  • Image Processing, Computer-Assisted
  • Male
  • Middle Aged
  • Prospective Studies
  • Sensitivity and Specificity
  • Tomography, X-Ray Computed / methods*